Search

Sotomayor Cases

Loading...

Flash Player 9 (or above) is needed to view presentations.
We have detected that you do not have it on your computer. To install it, go here.

0 comments

Post a comment

    Post a comment
    Embed Video
    Edit your comment Cancel

    Notes on slide 1

    Favorites, Groups & Events

    Sotomayor Cases - Presentation Transcript

    1. 07-0581-cv Maloney v. Cuomo UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT ______________ August Term, 2008 (Argued: December 15, 2008 Decided: January 28, 2009) Docket No. 07-0581-cv ______________ JAMES M. MALONEY , Plaintiff-Appellant, —v.— ANDREW CUOMO , in his official capacity as Attorney General of the State of New York, DAVID PATERSON , in his official capacity as Governor of the State of New York, KATHLEEN A. RICE , in her official capacity as District Attorney of the County of Nassau, and their successors,* Defendants-Appellees. ______________ B e f o r e: POOLER, SOTOMAYOR, and KATZMANN , Circuit Judges. ______________ Appeal from a judgment of the United States District Court for the Eastern District of New York (Spatt, J.) dated January 17, 2007, granting defendants-appellees Andrew Cuomo and David Paterson’s motion to dismiss and defendant-appellee Kathleen A. Rice’s motion for judgment on the pleadings, and from an order dated May 14, 2007, denying plaintiff-appellant’s motion for reconsideration. Affirmed. * Pursuant to Federal Rule of Appellate Procedure 43(c)(2), Governor David Paterson is automatically substituted for former Governor Eliot Spitzer as a defendant in this case.
    2. ______________ JAMES M. MALONEY , appearing pro se, for Plaintiff- Appellant. KAREN HUTSON , Deputy County Attorney (Lorna B. Goodman, County Attorney, on the brief) for Defendant- Appellee Kathleen A. Rice, Nassau County District Attorney, Mineola, N.Y. ______________ PER CURIAM : Plaintiff-appellant James Maloney was arrested at his home on August 24, 2000, and charged with possessing a chuka stick in violation of N.Y. Penal Law § 265.01(1). A “chuka stick” (or “nunchaku”) is defined as any device designed primarily as a weapon, consisting of two or more lengths of a rigid material joined together by a thong, rope or chain in such a manner as to allow free movement of a portion of the device while held in the hand and capable of being rotated in such a manner as to inflict serious injury upon a person by striking or choking. Id. § 265.00(14).1 This charge was dismissed on January 28, 2003, and Appellant pleaded guilty to one count of disorderly conduct. As part of the plea, he agreed to the destruction of the nunchaku seized from his home. Appellant filed the initial complaint in this action on February 18, 2003, and then an amended complaint on September 3, 2005, seeking a declaration that N.Y. Penal Law §§ 265.00 through 265.02 are unconstitutional insofar as they punish possession of nunchakus in one’s home. The district court dismissed the amended complaint as against the New York State 1 There are two sections of the New York Penal Law numbered 265.00(14). 2
    3. Attorney General and the Governor for lack of standing, concluding that neither official is responsible for enforcing the statutes at issue. The district court granted defendant Nassau County District Attorney Kathleen Rice’s motion for judgment on the pleadings in relevant part because the Second Amendment does not apply to the States and therefore imposed no limitations on New York’s ability to prohibit the possession of nunchakus. Appellant moved for reconsideration on the ground that the district court had failed to consider certain other claims raised in his amended complaint; the district court denied that motion. On appeal, Appellant challenges only the district court’s dismissal of his claims against Rice.2 He argues, inter alia, that New York’s statutory ban on the possession of nunchakus violates (1) the Second Amendment because it infringes on his right to keep and bear arms, and (2) the Fourteenth Amendment because it lacks a rational basis. Neither of these arguments has any merit. The Second Amendment provides: “A well regulated Militia, being necessary to the security of a free State, the right of the people to keep and bear Arms, shall not be infringed.” U.S. Const. amend. II. The Supreme Court recently held that this confers an individual right on citizens to keep and bear arms. See District of Columbia v. Heller, 128 S. Ct. 2783, 2799 (2008). It is settled law, however, that the Second Amendment applies only to limitations the federal government seeks to impose on this right. See, e.g., Presser v. Illinois, 116 U.S. 252, 265 (1886) (stating that the Second Amendment “is a limitation only upon the power of congress and the 2 Appellant makes no argument in his brief concerning the district court’s dismissal of his claims against the Attorney General and the Governor. We therefore deem any challenges to that aspect of the district court’s judgment waived. See Yueqing Zhang v. Gonzales, 426 F.3d 540, 541 n.1 (2d Cir. 2005). 3
    4. national government, and not upon that of the state”); Bach v. Pataki, 408 F.3d 75, 84, 86 (2d Cir. 2005) (holding “that the Second Amendment’s ‘right to keep and bear arms’ imposes a limitation on only federal, not state, legislative efforts” and noting that this outcome was compelled by Presser), cert. denied, 546 U.S. 1174 (2006). Heller, a case involving a challenge to the District of Columbia’s general prohibition on handguns, does not invalidate this longstanding principle. See Heller, 128 S. Ct. at 2813 n.23 (noting that the case did not present the question of whether the Second Amendment applies to the states). And to the extent that Heller might be read to question the continuing validity of this principle, we “must follow Presser” because “[w]here, as here, a Supreme Court precedent ‘has direct application in a case, yet appears to rest on reasons rejected in some other line of decisions, the Court of Appeals should follow the case which directly controls, leaving to the Supreme Court the prerogative of overruling its own decisions.’” Bach, 408 F.3d at 86 (quoting Rodriguez de Quijas v. Shearson/Am. Express, Inc., 490 U.S. 477, 484 (1989)) (alteration marks omitted); see also State Oil Co. v. Khan, 522 U.S. 3, 20 (1997). Thus, N.Y. Penal Law §§ 265.00 through 265.02 do not violate the Second Amendment. The Fourteenth Amendment similarly provides no relief for Appellant. “Legislative acts that do not interfere with fundamental rights or single out suspect classifications carry with them a strong presumption of constitutionality and must be upheld if ‘rationally related to a legitimate state interest.’” Beatie v. City of New York, 123 F.3d 707, 711 (2d Cir. 1997) (quoting City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 440 (1985)). We will uphold legislation if we can identify “some reasonably conceivable state of facts that could provide a rational basis for the legislative action. In other words, to escape invalidation by being declared irrational, the 4
    5. legislation under scrutiny merely must find some footing in the realities of the subject addressed by the law.” Id. at 712 (internal quotation marks and citations omitted). The legislative history of section 265.00 makes plain that the ban on possession of nunchakus imposed by section 265.01(1) is supported by a rational basis. Indeed, as Appellant concedes, when the statute was under consideration, various parties submitted statements noting the highly dangerous nature of nunchakus. For example, New York’s Attorney General, Louis J. Lefkowitz, asserted that nunchakus “ha[ve] apparently been widely used by muggers and street gangs and ha[ve] been the cause of many serious injuries.” Mem. from Attorney Gen. Louis J. Lefkowitz to the Governor (Apr. 8, 1974). And the sponsor of the bill, Richard Ross, stated that “[w]ith a minimum amount of practice, [the nunchaku] may be effectively used as a garrote, bludgeon, thrusting or striking device. The [nunchaku] is designed primarily as a weapon and has no purpose other than to maim or, in some instances, kill.” See N.Y. Penal Law § 265.00, practice commentary, definitions (“Chuka stick”) (quoting Letter of Assemblyman Richard C. Ross to the Counsel to the Governor (1974)). Appellant does not dispute that nunchakus can be highly dangerous weapons. Rather, his principal argument is that section 265.01(1) prevents martial artists from using nunchakus as part of a training program. But the fact that nunchakus might be used as part of a martial-arts training program cannot alter our analysis. Where, as here, a statute neither interferes with a fundamental right nor singles out a suspect classification, “we will invalidate [that statute] on substantive due process grounds only when a plaintiff can demonstrate that there is no rational relationship between the legislation and a legitimate legislative purpose.” Beatie, 123 F.3d at 711. Appellant has not carried this burden. Consequently, in light of the legislature’s view of the danger posed 5
    6. by nunchakus, we find that the prohibition against the possession of nunchakus created by N.Y. Penal Law § 265.01(1) is supported by a rational basis. We have considered Appellant’s remaining arguments and find them to be without merit. Accordingly, for the foregoing reasons, the judgment of the district court is hereby AFFIRMED. Appellant’s pending motions to strike defendant Kathleen Rice’s brief and material in her July 28, 2008 Rule 28(j) letter are hereby DENIED. 6
    7. 06-4996-cv Ricci v. DeStafano UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT August Term, 2007 (Argued: December 10, 2007 Decided: June 9, 2008) Docket No. 06-4996-cv FRANK RICCI, MICHAEL BLATCHLEY, GREG BOIVIN, GARY CARBONE, MICHAEL CHRISTOFORO, RYAN DIVITO, STEVEN DURAND, WILLIAM GAMBARDELLA, BRIAN JOOSS, JAMES KOTTAGE, MATTHEW MARCARELLI, THOMAS J. MICHAELS, SEAN PATTON, CHRISTOPHER PARKER, EDWARD RIORDAN, KEVIN ROXBEE, TIMOTHY SCANLON, BENJAMIN VARGAS, JOHN VENDETTO AND MARK VENDETTO, Plaintiffs-Appellants, -v- JOHN DESTEFANO, KAREN DUBOIS-WALTON, THOMAS UDE JR., TINA BURGETT, BOISE KIMBER, MALCOM WEBER, ZELMA TIRADO AND CITY OF NEW HAVEN, Defendants-Appellees. Before: POOLER, SACK and SOTOMAYOR, Circuit Judges. AFFIRMED.
    8. KAREN LEE TORRE, Attorney, New Haven, CT, for Plaintiffs-Appellants. RICHARD A. ROBERTS, Attorney, Cheshire, CT (Nicole C. Chomiak, Stacey L. Pitcher, and Todd J. Richardson, on the brief), for Defendants-Appellees. PER CURIAM: We withdraw our Summary Order of February 15, 2008. Ricci v. DeStefano, 2008 U.S. App. LEXIS 3293, 2008 WL 410436 (2d Cir. Feb. 15, 2008). Plaintiffs appeal from a judgment of the United States District Court for the District of Connecticut (Arterton, J.) granting the defendants’ motion for summary judgment on all counts. We affirm, for the reasons stated in the thorough, thoughtful, and well-reasoned opinion of the court below. Ricci v. DeStefano, 2006 U.S. Dist. LEXIS 73277, 2006 WL 2828419 (D. Conn., Sept. 28, 2006). In this case, the Civil Service Board found itself in the unfortunate position of having no good alternatives. We are not unsympathetic to the plaintiffs’ expression of frustration. Mr. Ricci, for example, who is dyslexic, made intensive efforts that appear to have resulted in his scoring highly on one of the exams, only to have it invalidated. But it simply does not follow that he has a viable Title VII claim. To the contrary, because the Board, in refusing to validate the exams, was simply trying to fulfill its obligations under Title VII when confronted with test results that had a disproportionate racial impact, its actions were protected. CONCLUSION The judgment of the district court is AFFIRMED. 2
    9. 06-4996-cv Ricci v. DeStefano 1 UNITED STATES COURT OF APPEALS 2 3 FOR THE SECOND CIRCUIT 4 5 August Term, 2007 6 7 (Argued: December 10, 2007 Decided: June 9, 2008 8 Order issued: June 12, 2008) 9 10 Docket No. 06-4996-cv 11 12 - - - - - - - - - - - - - - - - - - - -X 13 14 FRANK RICCI, MICHAEL BLATCHLEY, GREG 15 BOIVIN, GARY CARBONE, MICHAEL 16 CHRISTOFORO, RYAN DIVITO, STEVEN 17 DURAND, WILLIAM GAMBARDELLA, BRIAN 18 JOOSS, JAMES KOTTAGE, MATTHEW 19 MARCARELLI, THOMAS J. MICHAELS, SEAN 20 PATTON, CHRISTOPHER PARKER, EDWARD 21 RIORDAN, KEVIN ROXBEE, TIMOTHY SCANLON, 22 BENJAMIN VARGAS, JOHN VENDETTO AND MARK 23 VENDETTO, 24 25 Plaintiffs-Appellants, 26 27 - v.- 28 29 JOHN DESTEFANO, KAREN DUBOIS-WALTON, 30 THOMAS UDE JR., TINA BURGETT, BOI 31 KIMBER, MALCOM WEBER, ZELMA TIRADO AND 32 CITY OF NEW HAVEN, 33 34 Defendants-Appellees 35 36 - - - - - - - - - - -X 37 38 FOR PLAINTIFFS-APPELLANTS: KAREN LEE TORRE, New Haven, CT 39 40 FOR DEFENDANTS-APPELLEES: RICHARD A. ROBERTS {Nicole C.
    10. 1 Chomiak, Stacey L. Pitcher, and 2 Todd J. Richardson, 3 ~~~), Cheshire, CT 4 5 6 ORDER 7 After disposition of this appeal by summary order dated 8 February 15, 2008, an active judge of Court requested a 9 poll on whether to rehear the case in banco A poll on 10 whether to rehear case in banc was conducted among the 11 active judges of the Cou After the poll was concluded, 12 on June 9, 2008, the original three-judge panel withdrew the 13 summary order and filed a per curiam opinion; no subsequent 14 poll has been requested. Because a majority of the 15 court's act judges voted to deny rehearing in banc, 16 rehearing is hereby DENIED. 17 Judges Calabresi, Straub, Pooler, Sack, Sotomayor, 18 Katzmann, and B.D. ker concur in the denial of rehearing 19 in banco Chief Judge Jacobs and Judges Cabranes, Raggi, 20 Wesley, Hall and vingston dissent from t denial of 21 rehearing in banco 22 With this order, Judge Katzmann is filing a 23 concurring opinion, in which Judges Pooler, Sack, Sotomayor 24 and B.D. Parker joins; Judge B.D. Parker is filing a 25 concurring opinion, in which Judges Calabresi, Pooler, Sack 2
    11. 1 and Sotomayor join; and Judge Cabranes is filing a 2 dissenting opinion, in which Chief Judge Jacobs and Judges 3 Raggi, Wesley, Hall and Judge vingston join. 4 Other opinions may be filed with respect to this case, 5 concurring or dissenting in the denial of in bane review. 3
    12. KATZMANN, Circuit Judge, with whom Judge POOLER, Judge SACK, Judge SOTOMAYOR, and Judge B.D. PARKERjoin, concurring in the denial of rehearing en banc: I concur in the denial ofrehearing en bane, consistent with our Circuit's longstanding tradition of general deference to panel adjudication a tradition which holds whether or not the judges of the Court agree with the panel's disposition of the matter before it. Throughout our history, we have proceeded to a full hearing en banc only in rare and exceptional circumstances. See Wilfred Feinberg, Unique Customs and Practices o/The Second Circuit, 14 Hofstra L. Rev. 297, 311-12 (1986). The Supreme Court now has before it a petition for certiorari in this case, which I recognize presents difficult issues. As the Supreme Court de~ides whether to grant certiorari, it has for its review the district court's opinion, the panel's per curiam opinion, and opinions concurring with and dissenting from the decision denying rehearing en bane. The issues are therefore sharply defined for the Supreme Court's consideration of whether to grant certiorari.
    13. BARRINGTON D. PARKER. Circuit Judge. with whom Judge CALABRESI. Judge POOLER. Judge SACK, and Judge SOTOMAYOR join. concurring in the denial of rehearing en bane: At the heart of the dissent from the denial of rehearing en bane is the assertion that there was no Supreme Court or circuit law to guide this district court. or future district courts faced with similar claims. I disagree. The district court correctly observed that this case was unusual. Nonetheless. the district court also recognized that there was controlling authority in our decisions-among them. Hayden v. County ofNassau. 180 F.3d 42 (2d Cir. 1999) and Bushey v. N. Y. State Civil Servo Comm 'n, 733 F.2d 220 (2d Cir. 1984). cert. denied, 469 U.S. 1117 (1985). These cases clearly establish for the circuit that a public employer, faced with a prima facie case of disparate-impact liability under Title VII, does not violate Title VII or the Equal Protection Clause by taking facially neutral, albeit race-conscious, actions to avoid such liability. Insofar as the dissent suggests that the plaintiffs produced evidence of a racial classification or the imposition of a quota, I think it entirely mistaken. Although the City acted out of a concern that certifying the exam results would have an adverse impact on minority candidates-and although, as the panel noted in its decision, the result was understandably frustrating for applicants who passed the test-the City's response, to decline to certify any of the exams, was facially race-neutral. The City did not classify or confer any actual benefit on applicants on the basis of race. The dissent's citations to Adarand Constructors, Inc. v. Pena, 515 U.S. 200(1995), and CityofRichmondv. J.A. Croson Co., 488 U.S. 469 (1989), are 1
    14. therefore inapposite.! See Hayden, 180 F.3d at 49 (distinguishing those cases as "concerned with select affinnative action tools, such as quota systems, set-aside programs, and differential scoring cutoffs, which utilize express racial classifications and which prevent non-minorities from competing for specific slots or contracts."). Because there was no racial classification, the plaintiffs bore the burden of persuasion on the issue of discriminatory purpose. Jana-Rock Constr., Inc. v. N. Y State Dep 't ofEcon. Dev., 438 F.3d 195,204 (2d Cir. 2006). Here, however, there was no evidence of a discriminatory purpose; according to the record evidence, the City was motivated by a desire to comply with, and avoid liability under, Title VII and its implementing regulations. See Bushey, 733 F.2d at 226 ("It is settled that voluntary compliance is a preferred means of achieving Title VII's goal of eliminating employment discrimination." (internal quotation marks and alteration omitted)); see also Hayden, 180 F.3d at 51 ("A desire to reduce the adverse impact on [minority] applicants ... is not analogous to an intent to discriminate against non-minority candidates."). I think the dissent also quite unfairly caricatures the district court's evaluation of the plaintiffs' Title VII claim: "Under the District Court's rationale, it appears that any race-based employment decision undertaken to avoid a threatened or perceived Title VII lawsuit is itself I It may be worth noting that the Croson Court based its decision partly on the fact that "[t]here [was] nothing approaching a prima facie case of a constitutional or statutory violation by anyone in the Richmond construction industry." Croson, 488 U.S. at 500. Here, by contrast, the City was faced with a prima facie case of a violation of Title VII. See Gulino v. N. Y State Educ. Dep't, 460 F.3d 361,382 (2d Cir. 2006) (defining prima facie case of disparate-impact liability under Title VII); Robinson v. Metro-North Commuter R.R. Co., 267 F.3d 147, 160 (2d Cir. 2001) (same); see also 42 U.S.c. § 2000e-2(k) (codifying the disparate-impact theory ofliability and legislatively overruling Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 659 (1989)). 2
    15. immune from scrutiny under Title VII." This is simply not the case. Prior to reaching its conclusion, the district court assessed whether the examination results demonstrated a statistically disproportionate adverse racial impact under the EEOC Guidelines and whether the City had presented evidence to support its belief that less discriminatory alternatives to this particular test existed. This analysis shows that, contrary to the dissent's suggestion, the district court did not rubber stamp the City's proffered non-discriminatory reason for not certifying the exam results. Moreover, I hardly think that in order to decline to certify the exam results, the City was required to prove, through a validation study or some other means, that its own tests were not "job related for the position in question and consistent with business necessity," 42 U.S.C. § 2000e-2(k)(I)(A)(i) (defining affirmative defense to prima/acie case of disparate impact violation). In fact, our case law explicitly rejects that proposition. See Bushey, 733 F.2d at 226 (disagreeing with the assertion that "before adopting remedial measures" the employer must "prove that [the] prima face case [of a disparate-impact Title VII violation] was not rebuttable through job-related explanations"). I also disagree with the dissent's view that en banc review is warranted because the district court analyzed the plaintiffs claims using the McDonnell Douglas pretext test rather than the Price Waterhouse mixed-motive test. See Price Waterhouse v. Hopkins, 490 U.S. 228 (1989); McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973). As the dissent is well aware, the plaintiffs did not argue the mixed-motive theory; a non-party raised it in an amicus brief. "Although an amicus brief can be helpful in elaborating issues properly presented by the parties, 3
    16. it is normally not a method for injecting new issues into an appeal, at least in cases where the parties are competently represented by counsel." Universal City Studios, Inc. v. Corley, 273 F.3d 429,445 (2d Cir. 2001); see also Bano v. Union Carbide Corp., 273 F.3d 120, 127 n.5 (2d Cir. 2001) (same). Finally, the dissent suggests that the panel's per curiam opinion inappropriately adopted the reasoning set forth in the district court's opinion, one that the panel recognized was "thorough, thoughtful, and well-reasoned." The adherence of a Court of Appeals to the decision and reasoning of a district court is anything but novel. In fact, the practice pre-dates the formal establishment of this Court in 1891 by at least fifty years. United States v. Libellants & Claimants ofthe Schooner Amistad, 40 U.S. 518,590 (1841) ("The Circuit Court, by a mere pro forma decree, affirmed the decree of the [Connecticut] District Court .... And from that decree the present appeal has been brought to this Court."). This Court has followed this practice on numerous occasions in appeals covering myriad issues. See, e.g., In re Bankers Trust Co., 450 F.3d 121, 123 (2d Cir. 2006) (per curiam); Murphy ex rei. Estate ofPayne v. United States, 427 F.3d 158,159 (2d Cir. 2005) (per curiam); In re Red Dot Scenic, Inc., 351 F.3d 57, 58 (2d Cir. 2003) (per curiam); United States v. Gluzman, 154 F.3d 49,50 (2d Cir. 1998); Trans World Airlines, Inc. v. Sinicropi, 84 F.3d 116, 116 (2d Cir.) (per curiam), cert. denied, 519 U.S. 949 (1996). The plaintiffs were entitled to a careful and thoughtful review of their claims. The panel decided that the district court had given them just that, and thus adopted the district court's reasoning in its per curiam opinion. Nothing more is required. 4
    17. JOSE A. CABRANES, CircuitJudge, with whom ChiefJudge JACOBS, Judge RAGGI, Judge WESLEY, Judge HALL, and Judge LIVINGSTON join, dissenting: This appeal raises important questions of fIrst impression in our Circuit-and indeed, in the nation-regarding the application of the Fourteenth Amendment's Equal Protection Clause and Title VII's prohibition on discriminatory employment practices. At its core, this case presents a straight- forward question: Maya municipal employer disregard the results of a qualifying examination, which was carefully constructed to ensure race-neutrality, on the ground that the results of that examination yielded too many qualified applicants of one race and not enough of another? In a path-breaking opinion, which is nevertheless unpublished, the District Court answered this question in the affirmative, dismissing the case on summary judgment. A panel of this Court affirmed in a summary order containing a single substantive paragraph. Ricci v. DeStifano, No. 06-4996-cv (2d Cir. Feb. 15, 2008).1 Three days prior to the filing of this opinion, the panel withdrew its summary order and filed a per CIInam opinion adopting hI toto the reasoning of the District Court, thereby making the District Court's opinion the law of the Circuit. See Ricci v. DeStefimo, _ F.3d _ (2d Cir. 2008). The use of per CIInam opinions of this sort, adopting in full the reasoning of a district court without further elaboration, is normally reserved for cases that present straight-forward questions that do not require explanation or elaboration by the Court of Appeals. The questions raised in this appeal cannot be classified as such, as they are indisputably complex and far from well-settled. These questions include: Does the Equal Protection Clause prohibit a municipal employer from discarding examination results on the ground that "too many" applicants of one race received high scores and in the hope that a future test would yield more high-scoring applicants of other races? Does such a 1 Reproduced as .-ppendix A. 1
    18. practice constitute an unconstitutional racial quota or set-aside? Should the burden-shifting framework applicable to claims of pretextual discrimination ever apply to a claim of explicit race-based discrimination in violation of Title VII? If a municipal employer claims that a race-based action was undertaken in order to comply with Title VII, what showing must the employer make to substantiate that claim? Presented with an opportunity to address en bane questions of such "exceptional importance," Fed. R. App. P. 35(a)(2), a majority of this Court voted to avoid doing so. I respectfully dissent from that decision, without expressing a view on the merits of the questions presented by this appeal, in the hope that the Supreme Court will resolve the issues of great significance raised by this case. BACKGROUND In late 2003, 118 applicants took a written and oral examination administered by the New Haven Fire Department ("NHFD") for promotion to the ranks of Captain and Lieutenant. Forty-one applicants took the Captain examination, of whom twenty-five were white, eight black, and eight Hispanic. Based on the examination results and New Haven's protocol for civil service promotions, it appeared, at the time that the tests were scored, that "no blacks and at most two Hispanics would be eligible for promotion" to Captain. Ricci v. DeStefano, No. 04cvll09, at 3 (D. Conn. Sept. 28,2006).2 Vith respect to the Lieutenant examination, the racial composition of the seventy-seven applicants was as follows: forty-three whites, nineteen blacks, and fifteen Hispanics. The examination results indicated that no blacks or Hispanics would be promoted to the rank of Lieutenant. Between January and March 2004, the New Haven Civil Service Board ("CSB") held hearings to determine whether to certify the examination results and confer promotions according to those results. Despite the substantial efforts undertaken by the examination designer to ensure that it would be race-neutral, the City of New Haven 2 Reproduced as .,-ppendix B. 2
    19. (the "City") frankly stated its fear that, if the results were certified, it would face an employment discrimination lawsuit from non-white applicants who were not promoted. The CSB did not certify the examination results, and no promotions were made. Eighteen candidates-seventeen whites and one Hispanic-brought an action in the U.S. District Court for the District of Connecticut. They alleged in their complaint that the City and several municipal officials-acting in violation of Ticle VII of the Civil Rights Act of 1964,42 U.S.c. § 2000e et seq., the Equal Protection Clause of the Fourteenth Amendment to the U.S. Constitution, and other provisions of federal and state law-disregarded the results of two promotional examinations that produced "too many" eligible white candidates and "too few" eligible non-white candidates. On cross- motions for summary judgment, the District Court (Janet Bond Arterton, Judge) granted defendants' motion for summary judgment, denied plaintiffs' motion, and directed the Clerk of Court to close the case. In a forty-eight page opinion, the District Court observed that (1) "[P]laintiffs' evidence-and defendants' own arguments-show that the City's reasons for advocating non-certification [of the examination results] were related to the racial distribution of the results" and (2) "[a] jury could infer that the defendants were motivated by a concern that too many whites and not enough minorities would be promoted were the [eligibility] lists to be certified." Ricci, No. 04cv1109, at 20-21. The District Court recognized the exceptional circumstances presented by the case, noting that it "presents the opposite scenario of the usual challenge to an employment or promotional examination, as plaintiffs attack not the use of allegedly racially discriminatory exam results, but defendants' reason for their reftsal to use the results." ld at 22. Applying the burden-shifting framework set forth in McDonnell Douglas Corp. v. Gree1l, 411 U.S. 792 (1973), the District Court held that "[d]efendants' motivation to avoid making promotions based on a test with a racially disparate impact, even in a 3
    20. political context, does not, as a matter of law, constitute discriminatory intent, and therefore such evidence is insufficient for plaintiffs to prevail on their Title VII claim." Ricci, No. 04cvll 09, at 39-40 (footnote omitted). The District Court further concluded that defendants had not violated plaintiffs' rights under the Equal Protection Clause by, as plaintiffs alleged, "employing a race-based classification system for promotion or, alternatively, by applying facially neutral promotion criteria in a racially discriminatory manner." Id. at 40, 44. Although it is not disputed that the decision to discard the examination results was based on racial considerations, the District Court determined as a matter of law that no racial discrimination had occurred "because [all ofj the test results were discarded and nobody was promoted," id. at 42, and because "[n]othing in the record in this case suggests that the City defendants or CSB acted 'because of discriminatory animus toward plaintiffs or other non-minority applicants for promotion," 11. at 43. The District Court also rejected plaintiffs' civil rights conspiracy and First Amendment claims and declined supplemental jurisdiction over a state law tort claim. On appeal, the parties submitted briefs of eighty-six pages each and a six-volume joint appendi.x of over 1,800 pages; plaintiffs' reply brief was thirty-two pages long. Two amici briefs were ftled and oral argument, on December 10,2007, lasted over an hour (an unusually long argument in the practice of our Circuit). More than two months after oral argument, on February 15, 2008, the panel affirmed the District Court's ruling in a summary order containing a single substantive paragraph. The operative portion of the summary order read as follows: We affirm, substantially for the reasons stated in the thorough, thoughtful, and well-reasoned opinion of the court below. In this case, the Civil Service Board found itself in the unfortunate position of having no good alternatives. We are not unsympathetic to the plaintiffs' expression of frustration. Mr. Ricci, for example, who is dyslexic, made intensive efforts that appear to have resulted in his scoring highly on one of the exams, only to have it invalidated. But it simply does not follow that he has a viable Title VII claim. To the contrary, because the Board, in refusing to validate the exams, was simply trying to fulfill its obligations under Title VII 4
    21. when confronted with test results that had a disproportionate racial impact, its actions were protected. The judgment of the district court is AFFIRl1ED. See App. A. Four months later, and three days prior to the publication of this opinion, the panel withdrew its summary order and published a per curiam opinion that contained the same operative text as the summary order, with the addition of a citation to the District Court's opinion in the Xlestlaw and LexisNexis databases. This per cunam opinion adopted in toto the reasoning of the District Court, without further elaboration or substantive comment, and thereby converted a lengthy, unpublished district court opinion, grappling with significant constitutional and statutory claims of first impression, into the law of this Circuit. It did so, moreover, in an opinion that lacks a clear statement of either the claims raised by the plaintiffs or the issues on appeal. Indeed, the opinion contains no reference whatsoever to the constitutional claims at the core of this case, and a casual reader of the opinion could be excused for wondering whether a learning disability played at least as much a role in this case as the alleged racial discrimination. This perfunctory disposition rests uneasily vith the weighty issues presented by this appeal. 3 Prior to the entry of the per curiam opinion and in light of the "question[s] of exceptional importance," Fed. R. App. P. 35(a)(2), raised in this appeal, the Court considered a motion for en bam: reVIew. A majority of this Court declined to take up the appeal. DISCUSSION 3 Judge Parker's observation that "[t]he adherence of a Court of Appeals to the decision and reasoning of a district court is anything but novel" cannot be gainsaid. In appropriate cases, such a disposition is entirely unobjectionable. X'here significant questions of unsettled law are raised on appeal, however, a failure to address those questions--or even recognize their existence-should not be the approved modus operandi of the U.S. Court of Appeals. 5
    22. The core issue presented by this case-the scope of a municipal employer's authority to disregard examination results based solelY on the race of the successful applicants-is not addressed by any precedent of the Supreme Court or our Circuit. Plaintiffs alleged that the City's actions violated, inter alia, their rights under the Equal Protection Clause and Title VII. The District Court disagreed, but did so without the benefit of pertinent guidance from a higher court. The questions raised by the instant appeal clearly merit further review. 4 A. The Equal Protection Clause Plaintiffs claim that the City's decision to discard the examination results was race-based discrimination in violation of the Equal Protection Clause because it was undertaken solely to reduce the number of high-scoring white applicants and increase the number of eligible non-white candidates. Defendants contend that their decision, though race-based, was necessary because compliance with federal anti-discrimination laws required them to reduce the number of eligible white candidates. See Ricci, No. 04cv1109, at 20-21; Appellee Br. at 15-20, 30-31. The Supreme Court has addressed a government entity's claim that race-based decisions were necessary to redress a racial imbalance in the closely analogous context of government contracts. In Ciry qfRichmond v. J. A. Croson Co., the Supreme Court held that: "[w]hile there is no doubt that the sorry history of both private and public discrimination in this country has contributed to a lack of opportunities for black entrepreneurs, this observation, standing alone, cannot justify a rigid racial quota in the awarding of public contracts ...." 488 U.S. 469, 499 (1989). The Court further observed that: [W]hen a legislative body chooses to employ a suspect classification, it cannot rest upon a generalized assertion as to the classification's relevance to its goals. A governmental actor cannot render race a legitimate proxy for a particular condition merely by declaring that the condition exists. The history of racial classifications in this country suggests that blind judicial 4 Indeed, in his opinion concurring in the denial of en bane review, Judge Katzmann recognizes as much, observing that this appeal "presents difficult issues." 6
    23. deference to legislative or executive pronouncements of necessity has no place in equal protection analysis. Id. at 500-01 (internal citations omitted). More recently, the Supreme Court has identified "three general propositions with respect to governmental racial classifications." Adarand Constmdors, Inc. v. Pena, 515 U.S. 200, 223 (1995). They are: First, skepticism: Any preference based on racial or ethnic criteria must necessarily receive a most searching examination. Second, consistency: The standard of review under the Equal Protection Clause is not dependent on the race of those burdened or benefited by a particular classification, i.e., all racial classifications reviewable under the Equal Protection Clause must be strictly scrutinized. And third, congruence: Equal protection analysis in the Fifth Amendment area is the same as that under the Fourteenth Amendment. Taken together, these three propositions lead to the conclusion that at!} person, of whatever race, has the right to demand that aI!} governmental actor su,?ject to the Constitutionjustifj af!) racial classification su,?jecting that person to unequal treatment under the strictestjudicial scmtif!). Id. at 223-24 (quotation marks, internal citations, and brackets omitted) (emphasis added). Whether the District Court's judgment comports with these propositions is a question of immense importance that is not addressed in the panel's per curiam opinion. The District Court's ruling rested in part on the premise that "where a test is administered and scored in the same manner for all applicants, plaintiffs cannot make out a claim that the exam was a facially neutral test used in a discriminatory manner." Ricci, No. 04cv1109, at 42. Neutral administration and scoring--even against the backdrop of race-conscious design of an employment examination, see Hqyden v. Coun!} ofNassau, 180 F.3d 42, 50 (2d Cir. 1999)-is one thing. But neutral administration and scoring that is followed by race-based treatment of examination results is surely something else entirely. Where, as here, examination results are disregarded on the ground that too many candidates of one race qualified for promotion on the basis of those results, the fact of neutral administration and scoring may not necessarily immunize defendants from the claims of civil rights violations brought by plaintiffs. If it did, municipal employers could reject the results of an employment examination whenever those results 7
    24. failed to yield a desired racial outcome-i.e., failed to satisfy a racial quota. Croson and Adarand est.'lblish that racial quotas are impermissible under the Constitution absent specific findings of past discrimination that are not in the record here. Whether Croson and Adaraltd preclude the actions challenged in this case, or whether Hqydm can fairly be read to compel judgment in defendants' favor as a matter of law, are questions that admit no easy answer. As such, they require the careful analysis of a full opinion of an appellate court, not abbreviated disposition. The District Court held that the test was administered in the same manner for all applicants because the City discarded the scores of all exam-takers. Insofar as the decision to not certify the results was based on the race of the high-scoring applicants, however, it is arguable that the deck was stacked against applicants of that race: If too many white applicants obtained high scores, the City stood ready to nullify the results in the hope that non-white applicants would score relatively higher on a subsequent examination. 5 Whether such action amounts to an impermissible racial quota was not addressed in the District Court's opinion or in the decisions issued by the panel, which do not even note that this action arises under the Equal Protection Clause of the Fourteenth Amendment. See App. A (summary order of Feb. 15, 2008); per curiam opinion filed on June 9, 2008. The District Court also held as a matter oflaw that none of the City's reasons for disregarding the examination results amounted to intentional discrimination because the City had acted based on the following concerns: that the test had a statistically adverse impact on African-American and Hispanic examinees; that promoting off of this list would undermine their goal of diversity in the Fire Department and would fail to develop managerial role models for aspiring firefighters; that it would sulyect the City to public criticism; and that it vould likely subject the City to Title VII lawsuitsfrom minority applicants that, for political reasons, the City did 110t want to difend. 5 In his opinion concurring in the denial of en bane review, Judge Parker concludes that the City "did nor ... confer any actual benefit on applicants on the basis of race." It is, at the very least, an open question whether discarding examination results on the basis of race so that members of certain races could have a "second chance" to compete constitutes the conferral or denial of a benefit on the basis of race. 8
    25. Ricci, No. 04cv1109, at 43 (emphasis added). Leaving aside the propriety of the District Court's evaluation, on summary judgment, of the City's motives-a quintessential question of fact, see, e.g., Hunt v. Cromartie, 526 U.S. 541, 552-53 (1999)-it is at least arguable that the District Court failed to subject the City's justifications to the "most searching examination" prescribed by the Adarand Court. See 515 U.S. at 223. The record suggests that the District Court took the City's justifications at face value, as it appears Judge Parker has done in his opinion concurring in the denial of en banc review. An appellate court ought to consider whether this level of scrutiny is consistent with Justice O'Connor's observation, in Croson, that "[a]bsent searching judicial inquiry into the justification for such race-based measures, there is simply no way of determining what classifications are 'benign' or 'remedial' and what classifications are in fact motivated by illegitimate notions of racial inferiority or simple racial politics." 488 U.S. at 493 (plurality opinion). Justice O'Connor's cautionary note on "racial politics" is particularly relevant in light of the District Court's observation that fear of "public criticism" and other "political reasons" factored into the City's decision. Whether the District Court subjected the City's claims to sufficient scrutiny-and whether the City's claims could have withstood such scrutiny-are vital "question[s] of exceptional importance," Fed. R. App. P. 35(a)(2), that warrant further review, both for the proper resolution of this case and for the guidance of other courts and municipalities in future cases. B. Title VII Plaintiffs urge that the City's race-based action also violated Title VII's prohibition of employment discrimination. See 42 U.S.c. § 2000e-2. The District Court dismissed plaintiffs' Title VII claim by applying the three-step burden-shifting framework for adjudicating claims of pretextual discrimination established by McDonnell Douglas. The dismissal of the Title VII claim on this basis raises two significant questions: (1) whether the McDonnell Douglas test for prete-"'dual discrimination should be 9
    26. applied to claims of discrimination that is overt, and (2) whether a race-based decision allegedly made to avoid perceived liability for racial discrimination is exempt from scrutiny under Title VII and, if not, what quantum of proof is required to substantiate such a defense. Courts generally apply McDonnell Douglas in cases where plaintiffs "presentD no direct evidence of discriminatory treatment." Woodman v. WWOR-TV, Inc., 411 F.3d 69, 76 (2d Cir. 2005); see also Graves tJ. Finch Pruyn & Co. Inc, 457 F.3d 181, 187 (2d Cir. 2006). "If a plaintiff can convince the trier of fact that an impermissible criterion in fact entered into the employment decision, [however,] a somewhat different analysis takes place." 7)ler 11. Bethlehem Steel Corp., 958 F.2d 1176, 1181 (2d Cir. 1992). In that kind of "mixed-motive" case, the burden-shifting analysis set forth in Price Waterhouse v. Hopkins, 490 u.s. 228 (1989), governs the claim. Under this framework, the plaintiff ... must focus his proof directly at the question of discrimination and prove that an illegitimate factor had a motivating or substantial role in the employment decision. If the plaintiff convinces the factfinder that the illegitimate factor played such a role, the employee has proved that the decision was made at least in part because of the illegitimate factor. At this point the employee is entitled to succeed subject only to the employer's opportunity to prove its affIrmative defense; that is, that it would have reached the same decision as to the employee's employment even in the absence of the impermissible factor. 7)ler, 958 F.2d at 1181 (internal citations, quotation marks, and brackets omitted); see also Sista v. CDC IxiJ N. Am., I,IC., 445 F.3d 161, 173-74 (2d Cir. 2006); Raskin v. u::jatt Co., 125 F.3d 55,60-61 (2d Cir. 1997) ("Evidence potentially warranting a Price Waterhouse burden shift includes, inter alia, policy documents and evidence of statements or actions by decisionmakers that may be viewed as directly reflecting the alleged discriminatory attitude." (internal quotation marks omitted)). The Ricci plaintiffs offered evidence that an impermissible factor-their race-motivated defendants to discard the results of the employment examination. As the District Court itself candidly observed: "[P]laintiffs' evidence-and defendants' own arguments-show that the City's reasons for advocating non-certification [of the examination results] were related to the racial distribution of the 10
    27. results" and "[a] jury could infer that the defendants were motivated by a concern that too many whites and not enough minorities would be promoted were the [eligibility] lists to be certified." Rim, No. 04cvl109, at 20-21. The District Court's application of the McDonnell Douglas test for pretextual discrimination, its conclusion that plaintiffs cannot pass that test as a matter of law, and its failure to consider the possibility that defendants themselves might bear a burden of proof under the analysis set forth in Price Waterhouse, all raise serious concerns left unaddressed by the panel in its per a/riam opinion and by the full Court, which declined en banc review of the appeaL Assuming arguendo that a claim of overt racial discrimination is ever appropriately evaluated under the McDonnell Douglas framework for pretextual discrimination, the application of that framework to this case required a "reversal" of the usual roles assigned to plaintiffs and defendants in such cases. As the District Court observed: [T]his case presents the opposite scenario of the usual challenge to an employment or promotional examination. . .. Ordinarily, as contemplated by the statute, the "complaining party" bears the burden of proving a disparate impact, and the "respondent" bears the burden of "demonstrat[ing] that the challenged practice is job related for the position in question and consistent with business necessity," or, alternatively, the "complaining party" may prevail by showing that an alternative employment practice with less disparate impact existed and that the respondent failed to utilize it. Here, the roles of the parties are in essence reversed, with the defendants, normally reflecting a "respondent" role in the Title VII disparate impact analysis, contending that use of the promotional exams, if they had been certified, would have had an adverse impact, and the plaintiffs, normally the "complaining party," arguing that the test results were sufficiently job-related to be defensible under the law. Ricci, No. 04cv11 09, at 22 (alteration in original) (internal citations omitted). Unlike the Court of Appeals, the District Court answered the exceptional, and difficult, questions presented, concluding that the City's expressed desire to comply with "the letter and the spirit of Title VII," id. at 22, constituted a non-pretextual reason for its action, id. at 39-40, and therefore no employment discrimination occurred. Under the District Court's rationale, it appears that any race-based employment decision undertaken to avoid a threatened or perceived Title VII lawsuit is immune from 11
    28. scrutiny under Title VII." This appears to be so, moreover, regardless of whether the employer has made any efforts to verify that a valid basis exists for the putative Title VII suit. Applying this rationale, the District Court concluded that the City, which had not conducted any study to determine whether latent racial bias had tainted the results of the promotion examination, could discard the results of the examination, id. at 25-26, in the hope that a future test would yield a preferable racial distribution, id. at 36. Regardless of how one may decide the matter, there can be little doubt that a decision of this Court thus sanctioning race-based employment decisions in the name of compliance with Title VII raises novel questions that are indisputably of "exceptional importance." CONCLUSION It is arguable that when an appeal raising novel questions of constitutional and statutory law is resolved by an opinion that tersely adopts the reasoning of a lower court-and does so without further legal analysis or even a full statement of the questions raised on appeal-those questions are insulated from further judicial review. It is arguable also that the decision of this Court to deny C11 banc review of this appeal supports that view. What is not arguable, however, is the fact that this Court has failed to grapple with the questions of exceptional importance raised in this appeal. If the Ricci plaintiffs are to obtain such an opinion from a reviewing court, they must now look to the Supreme Court. Their claims are worthy of that review. 6 Despite Judge Parker's assertion to the contrary, I do not charge the District Court with applying a "rubber stamp" to the Ciry's race-based decisions. I simply question whether the Court of A.ppeals has set forth a standard for determining when such action is acceptable and when it violates the constitutional and statutory rights of citizens. If any fault is to be levied in this regard, it falls on our Court for failing to provide guidance, and not on the District Court which endeavored to confront this question of exceptional importance. 12
    29. Appendix A
    30. 06-4996-cv Ricci v. DeStefano 1 2 UNITED STATES COURT OF APPEALS 3 FOR THE SECOND CIRCUIT 4 5 SUMMARY ORDER 6 7 RULINGS BY SUMMARY ORDER DO NOT HAVE PRECEDENTIAL EFFECT. CITATION 8 TO SUMMARY ORDERS FILED AFTER JANUARY 1,2007, IS PERMITTED AND IS 9 GOVERNED BY THIS COURT'S LOCAL RULE 32.1 AND FEDERAL RULE OF APPELLATE 10 PROCEDURE 32.1. IN A BRIEF OR OTHER PAPER IN WHICH A LITIGANT CITES A 11 SUMMARY ORDER, IN EACH PARAGRAPH IN WHICH A CITATION APPEARS, AT LEAST 12 ONE CITATION MUST EITHER BE TO THE FEDERAL APPENDIX OR BE ACCOMPANIED 13 BY THE NOTATION: "(SUMMARY ORDER)." A PARTY CITING A SUMMARY ORDER 14 MUST SERVE A COpy OF THAT SUMMARY ORDER TOGETHER WITH THE PAPER IN 15 WHICH THE SUMMARY ORDER IS CITED ON ANY PARTY NOT REPRESENTED BY 16 COUNSEL UNLESS THE SUMMARY ORDER IS AVAILABLE IN AN ELECTRONIC 17 DATABASE WHICH IS PUBLICLY ACCESSIBLE WITHOUT PA YMENT OF FEE (SUCH AS 18 THE DATABASE AVAILABLE AT HTTP://WWW.CA2.USCOURTS.GOV I). IF NO COpy IS 19 SERVED BY REASON OF THE AVAILABILITY OF THE ORDER ON SUCH A DATABASE, 20 THE CITATION MUST INCLUDE REFERENCE TO THAT DATABASE AND THE DOCKET 21 NUMBER OF THE CASE IN WHICH THE ORDER WAS ENTERED. 22 23 24 At a stated Term of the United States Court of Appeals for the Second Circuit, held at the 25 Daniel Patrick Moynihan United States Courthouse, 500 Pearl Street, in the City of New York, 26 on the 15 th day of February, two thousand and eight. 27 28 Present: ROSEMARY S. POOLER, 29 ROBERT D. SACK, 30 SONIA SOTOMAYOR, 31 Circuit Judges, 32 33 34 FRANK RICCI, MICHAEL BLATCHLEY, GREG BOIVIN, GARY CARBONE, MICHAEL 35 CHRISTOFORO, RYAN DIVITO, STEVEN DURAND, WILLIAM GAMBARDELLA, 36 BRIAN JOOSS, JAMES KOTT AGE, MATTHEW MARCARELLI, THOMAS J. MICHAELS, 37 SEAN PATTON, CHRISTOPHER PARKER, EDWARD RIORDAN, KEVIN ROXBEE, 38 TIMOTHY SCANLON, BENJAMIN VARGAS, JOHN VENDETTO AND MARK 39 VENDETTO, 40 41 Plaintiffs-Appellants, 42 43 -v- (06-4996-cv) 44 45 JOHN DESTEFANO, KAREN DUBOIS-WALTON, THOMAS UDE JR., TINA BURGETT, 46 BOISE KIMBER, MALCOM WEBER, ZELMA TIRADO AND CITY OF NEW HAVEN, 47 48 Defendants-Appellees. 49
    31. I Appearing for Plaintiffs-Appellants: Karen Lee Torre, New Haven, CT. 2 3 Appearing for Defendants-Appellants: Richard A. Roberts (Nicole C. Chomiak, 4 Stacey L. Pitcher, and Todd 1. Richardson 5 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _~on:.:..:::.th.:.::e:...:b::.:.n.:.::·e::Jf), Cheshire, CT. 6 7 8 UPON CONSIDERATION WHEREOF, IT IS HEREBY ORDERED, ADJUDGED, 9 AND DECREED that the judgment of the distri~t court is AFFIRMED. 10 11 Plaintiffs appeal from a judgment of the United States District Court for the District of . 12 Connecticut (Arterton, J.) granting the defendants' motion for summary judgment on all counts. 13 14 We affirm, substantially for the reasons stated in the thorough, thoughtful, and well-reasoned 15 opinion of the court below. In this case, the Civil Service Board found itself in the unfortunate 16 position of having no good alternatives. We are not unsympathetic to the plaintiffs' expression 17 of frustration. Mr. Ricci, for example, who is dyslexic, made intensive efforts that appear to 18 have resulted in his scoring highly on one of the exams, only to have it invalidated. But it 19 simply does not follow that he has a viable Title VII claim. To the contrary, because the Board, 20 in refusing to validate the exams, was simply trying to fulfill its obligations under Title VII when 21 confronted with test results that had a disproportionate racial impact, its actions were protected. 22 23 The judgment of the district court is AFFIRMED. 24 25 26 FOR THE COURT: 27 Catherine O'Hagan Wolfe, Clerk 28 29 By:. _ 2
    32. Appendix B
    33. UNITED STATES DISTRICT COURT DISTRICT OF CONNECTICUT FRANK RICCI, et al., Plaintiffs, v. Civil No. 3:04cvll09 (JBA) JOHN DESTEFANO, et al., Defendants. RULING ON CROSS-MOTIONS FOR SUMMARY JUDGMENT [DOCS. ii 52, 60] In March 2004 the New Haven Civil Service Board (UCSB") refused to certify the results of two promotional exams for the positions of Lieutenant and Captain in the New Haven Fire Department. This lawsuit arises from the circumstances leading to that decision and its consequences. Plaintiffs are seventeen white candidates and one Hispanic candidate who took the promotional exams, on which they fared very well, but received no promotion because without the CSB's certification of the test results, the promotional process could not proceed. Defendants are the City of New Haven, Mayor John DeStefano, Chief Administrative Officer Karen Dubois-Walton, Corporation Counsel Thomas Ude, Director of Personnel Tina Burgett, and the two members of the CSB, Malcolm Weber and Zelma Tirado, who voted against certification. Plaintiffs assert that defendants' actions in voting or arguing against certification of the examination results violated their rights under Title VII of 1
    34. the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq., the Equal Protection Clause, the First Amendment, and 42 U.S.C. § 1985; plaintiffs also allege a common law claim of intentional infliction of emotional distress. The parties have cross-moved for summary judgment on the Title VII and Equal Protection claims, and defendants additionally move for summary judgment on plaintiffs' other claims. For the reasons that follow, defendants' motion for summary judgment [Doc. # 52] will be granted as to plaintiffs' federal claims; plaintiffs' cross-motion for summary judgment [Doc. # 60] will be denied; and the Court will decline jurisdiction over plaintiffs' state law claim. 1 I. Factual Background While the parties strenuously dispute the relevance and legal import of, and inferences to be drawn from, many aspects of this case, the underlying facts are largely undisputed. In November and December 2003, the New Haven Fire Department administered written and oral examinations for promotion to Lieutenant and Captain. The City's Department of Human Resources issued a Request for Proposal for these examinations; as a result of which I/O Solutions (IOS"), a seven-year-old Illinois company that specializes in entry-level and promotional examinations for lDefendants also moved to strike portions of plaintiffs' Local Rule 56(a)2 Statement, which motion was denied. See Ruling Denying Motion to Strike [Doc. # 130]. 2
    35. public safety (police and fire) departments, designed the examinations. Pl. Ex. IV(C) at 8. Under the contract between the City and the New Haven firefighters' union, the written exam result counted for 60% of an applicant's score and the oral exam for 40%. Those with a total score above 70% on the exam would pass. Forty-one applicants took the Captain exam, of whom 25 were white, 8 black, and 8 Hispanic. Twenty-two of those applicants passed, of whom 16 were white, 3 black, and 3 Hispanic. Pl. Ex. Vol. I, at 43. Given that there were 7 Captain vacancies in the department when the tests were administered, and that the "Rule of Three" in the City Charter mandates that a civil service position be filled from among the three individuals with the highest scores on the exam, it appeared at that time that no blacks and at most two Hispanics would be eligible for promotion, as the top 9 scorers included 7 whites and 2 Hispanics. 2 Seventy-seven applicants took the Lieutenant exam, of whom 43 were white, 19 black, and 15 Hispanic. Thirty-four passed, of whom 25 were white, 6,black and 3 Hispanic. Id. There were 8 vacancies, but because of the top 10 scorers were white, it appeared that no blacks or Hispanics would be promoted. 3 2Hispanics ranked 7, 8 and 13; blacks ranked 16, 19 and 22. Pl. Ex. Vol. I, at 43. 3Hispanics ranked 27, 28 and 31; blacks ranked 14, 15, 16, 20, 22, and 24. PI. Ex. Vol. I, 43. 3
    36. Certified promotional lists remain valid for two years. The CSB held five hearings between January and March 2004 on the issue of whether to certify the test results. The issue appears to have been raised by New Haven's Corporation Counsel, Thomas Ude. At the initial hearing on January 22, 2004, Mr. Ude characterized the exam results as ~a very significant disparate impact . . . that caused us to think this was something we should bring to you, the Civil Service Board, to evaluate and to be part of and to ultimately make a decision about the process." Pl. Ex. Vol. IV(A) at 32. While it is disputed whether Mr. Ude already had made up his mind to advise the CSB against certifying the results, his comments "emphasize [dl . that the case law does not require that the City find that the test is indefensible in order to take action that it believes is appropriate to remedy . . disparate impact from examination." Id. at 34-35. He advised that "federal law does not require that you [the CSB] make a finding that this test . . . was not job-related, which is another way of saying it wasn't fair. A test can be job-related and have a disparate impact on an ethnic group and still be rejected because there are less discriminatory alternatives for the selection process." Id. at 36. During the hearings, the tests results were not released by name, and therefore none of the firefighters knew where they had placed. The only information provided to the CSB and the public, 4
    37. including plaintiffs, was the scores by race and gender. Nonetheless, several firefighters, although they did not know where they had placed, spoke in favor of certifying the results. Plaintiff Frank Ricci stated that the questions on the test were drawn from "nationally recognized" books and New Haven's own Rules and Regulations and Standard Operating Procedures. Pl. Ex. Vol. 1V(A) at 88. He stated that he "studied 8 to 13 hours a day to prepare for this test and incurred over $1,000 in funds [sic] to study for this test," including purchasing the books and paying an acquaintance to read them on tape because he is dyslexic and learns better by listening. Other firefighters, who believed the tests were fair, ·also spoke in support of the certifying the results. See,~, Testimony of Michael Blatchley, id. at 75 ("[N]one of those questions were not in that study material. Everyone of those questions came from the material.") . During the first hearing, the CSB also took statements from several New Haven firefighters who complained that some of the questions were not relevant to knowledge or skills necessary for the positions (see, ~, Statement of James Watson, at 85 ("I think this test was unfair. We don't use ~ lot of things that were on that test" such as whether to park a firetruck facing "uptown" or "downtown"», or that the study materials were difficult to obtain (see Testimony of Gary Kinney, id. at 77 5
    38. (~The only books that most of us had in front of us in the fire houses were Essentials of Fire Fighting . . . . [T]hese books [on the syllabus] were never in the fire houses."». At the second hearing on February 5, Patrick Egan, president of the firefighters' union, urged the CSB to conduct a validation study to determine the job-relatedness of the test, referring generally, although not by name, to the EEOC's ~Uniform Guidelines of Employee Selection Procedures." Pl. Ex. Vol. IV(B) at 11-12. Plaintiffs' counsel in the present case also ~poke and urged certification. On the other side, Donald Day, a representative of the Northeast Region of the International Association of Black Professional Firefighters, argued against certification, stating that previous promotional examinations in 1996 and 1999 had black and Latino firefighters ranked sufficiently high to have a realistic opportunity for promotion, whereas ~there was something inherently wrong with this test" because minorities did not score as highly. Id. at 33-34. He suggested that the CSB speak with the director of the civil service in Bridgeport ~to find out what Bridgeport is doing different [sic] than New Haven," as they have more diversity in their firefighter ranks. rd. at 35. In particular, he stated that Bridgeport had "changed the relative weights" among the portions of the exam, such that the written test counts for 30% of the total score, the oral test for 65%, 6
    39. and seniority 5%. Id. at 36-37. Ronald Mackey, the Internal Affairs Officer for the Northeast Region of the International Association of Black Professional Firefighters, supported Patrick Egan's suggestion of obtaining a validation study, and also suggested that New Haven could "adjust the test" as Bridgeport had done, in order to "meet the criteria of having a certain amount of minorities get elevated to the rank of Lieutenant and Captain." Id. at 43-45. On February 11, 2004, the CSB heard from Chad Legel, Vice President of lOS, who was the "project manager" in charge of developing the exams at issue. He stated that lOS had prepared "both an entry-level exam and a physical ability test for the firefighter position" in New Haven, but had not previously prepared a New Haven promotional exam. Id. at 10. However, in recent years his company had worked with similarly-sized public safety departments with demographics similar to New Haven, including Lansing, Michigan, Orange County, Florida, and the North Miami Police Department, among others. Id. at 9. Legel described the way in which the test was developed. First, the company interviewed a random sample of current New Haven Fire Department Lieutenants, Captains and Battalion Chiefs to determine basic information concerning the structure of the department, the tasks required of individuals at each rank, and the materials the department generally utilizes for training. 7
    40. Based on the interviews, lOS developed a written job analysis questionnaire ("JAQ") that asked all incumbents in the positions of Lieutenant and Captain "to provide information about how important they feel a specific task, knowledge area, skill or ability is. " I d . at 17. The JAQ asked how important each task was to successful performance on the job and how frequently it was necessary to perform it. Importance and frequency were merged into a metric called "criticality or essentiality." Id. at 19. Tasks above a certain criticality threshold were designated for testing on the written and oral portions of the exam. In response to the question of whether he has generally found a difference between information tested in various departments "based on the racial content of the city and the force," Legel stated, "definitely no." rd. at 21. The one difference among the New Haven firefighters of similar rank that Legel noted was different levels of training in certain specialized fields such as hazardous materials; such variation "throws up a red flag" indicating that lOS should not ask "high- level questions about hazardous materials. " I d . at 22. Legal further stated that all the questions were firmly rooted in the study materials on the syllabus, which was distributed with the promotion applications. See Def. Ex. 16 ("Written Examination Reference List"). Once the test was completed, an "independent reviewer;" a Battalion Chief from the 8
    41. Cobb County, Georgia, Fire Department, "reviewed the written exam for content and fidelity to the source material." Pl. Ex. Vol. IV(B) at 24-25. Another independent reviewer, a retired Fire Chief from outside Connecticut, reviewed the oral exam questions. Id. at 26. lOS refrained from utilizing reviewers from Connecticut because the RFP had specified that examiners must come from outside Connecticut, due to concerns that utilizing internal personnel could potentially facilitate cheating on the test. Likewise, lOS selected the panelists for the oral examination panels from departments outside Connecticut, making an effort "to gain maximum diversity." Id. at 32. All but one panel had one African-American, one Hispanic and one white assessor, and a standby panel had two African-Americans and one white. Id. The assessors were trained on how to grade the oral exam scenarios consistently, utilizing checklists of desired criteria. Each panelist also held at least an equal rank (if not superior) to the position being tested, in order to be able to identify an answer that was good but not quite the best answer outlined in the checklist. at 33-34, 37. Legel concluded by "implor[ing] anyone that had . . . concerns [about disparate impact] to review the content of the exam. In my professional opinion, it's facially neutral." Id. at 49. 9
    42. Noelia Marcano, Chief Examiner for the City of-New Haven and Secretary to the CSB, explained the process by which the RFP was developed and lOS chosen. She further explained that the job applications for the Lieutenant and Captain positions contained a job description, employment application, and "the actual study list in final form," and that when questions arose concerning conflicting information in some of the study books, lOS sent a letter to all applicants that they would not be asked questions on material where the sources differed. Id. at 78. At the next hearing on March 11, 2004, the CSB heard from Christopher Hornick, Ph.D., an industrial/organizational psychologist from Texas who runs a consulting business in competition with IOS. 4 See Pl. Ex. Vol. IV(D} at 7, 12. Dr. Hornick stated that he had "not had time to study the test at length or in detail." at 13. However, he reviewed stati~tics provided by the City and concluded that "we're seeing relatively high adverse impact" from the lOS tests. Id. at 11. He opined that his company finds "significantly and dramatically less adverse impact in most of the test procedures that we 4Plaintiffs argue that Dr. Hornick's non-sworn, hearsay statement at the CSB hearing is inadmissible as non-disclosed expert evidence. Plaintiffs' argument is rejected because defendants proffer Dr. Hornick's not for the truth of his conclusion that the tests had a racially disparate impact, but to show that defendants had a good faith belief, based in part on Dr. Hornick's testimony, that such a disparate impact existed and justified the decision not to certify the exams. 10
    43. design." Id. at 12. He stated: Normally, whites outperform ethnic minorities on the majority of standardized testing procedures. That is, in , the case with the data that we've seen in New Haven. I'm a little surprised at how much adverse impact there is in these tests. And I hope at some point here we'll be talking in detail about that. But my conclusion is that we did have significant adverse impact. Some of it is fairly typical of what we've seen in other areas of the countries (sic) and other tests that people have developed. But in other ways it is somewhat worse than what we're typically seeing in the profession practiced by others. Id. at 11-12. Dr. Hornick acknowledged that he had not looked at specific statistics from previous promotional examinations in New Haven to compare their results with the 2003 exam results. Id. at 14. When asked about the reasons behind any possible disparate impact, Dr. Hornick answered, "I'm not sure that r can explain it," but suggested that perhaps the 60%/40% breakdown mandated by the collective bargaining agreement could be responsible, and further suggested that there were "perhaps different types of testing procedures that are much more valid in terms of identifying the best potential supervisors in your fire department." rd. at 15. He stated that based on his interviews with firefighters, "we know that" a written test is "not as valid as other procedures that exist." rd. at 16. He also suggested that "[b]y not having anyone from within the department review the items [on the test] you inevitably get things in there" that 11
    44. are not relevant to the specific department. rd. at 17-18. Finally, Dr. Hornick identified as an alternative to traditional written and oral testing processes ~an assessment center process, which is essentially an opportunity for candidates to demonstrate their knowledge of the . . . standard operating procedures, to demonstrate how they would address a particular problem as opposed to just verbally saying it or identifying the correct option on a written test. For example, there's concepts of situation judgment tests that can be developed and designed, customized within organizations that demonstrate dramatically less adverse impacts. . . " Id. at 22-23. At the same hearing, Vincent M. Lewis, a Fire Program Specialist for the u.S. Department of Homeland Security, and a retired career firefighter from Michigan, testified that he believed the test was appropriate. He stated that he had looked ~extensively at the Lieutenant's exam and a little less at the Captain's exam," and believed that the candidates ~should know that material." Id. at 34-35. His one comment was that ~a number of questions in the Lieutenant's exam dealt with issues that an apparatus driver needed to know," and a candidate who had not had such training would be disadvantaged on those questions. rd. at 34, 41. However, he generally ~felt the questions were relevant for both exams," and believed that the New Haven applicants were advantaged over examinees in other locations 12
    45. because they were instructed exactly which chapters from the study materials would .be on the tests. Id. at 36. He stated that he would not have changed anything about the way in which the tests were developed, and opined that any disparate impact could be due to a general pattern that ~usually whites outperform some of the minorities on testing,H or that ~more whites . . . take the exam. H Id. at 37-38. The last expert witness was Dr. Janet Helms, a professor of counseling psychology and the Director of the Institute for the Study and Promotion of Race and Culture at Boston College. Her area of expertise "is not with firefighters per se but is more in the general area of how race and culture influence test performance more generally.H Id. at 43. She did not examine the specific tests at issue. Id. at 55. However, she offered several potential explanations for racially disparate impact on the tests. First, "[w]e know for a fact that regardless of what kind of written test we give in this country that-we can just about predict how many people will pass who are members of under- represented groups. And your data are not that inconsistent with what predictions would say were the case. H Id. at 44 (emphasis supplied). Second, Dr. Helms suggested that because 67% of the respondents in the JAQ survey were white, the questions may have been skewed toward their job knowledge~ as "most of the literature on firefighters show that the different 13
    46. [racial. and gender] groups perform the job differently." Id. at 46. Relying on information she had read in newspaper accounts of the situation in New Haven, she stated that the difference in performance may have been due to differences in opportunities for training and "informal mentoring" available to mirtorities. Id. at 48. With respect to the oral exam, Dr. Helms suggested that people who are bilingual or "speak acc~nted speech" may elicit more negative reactions from evaluators. Id. at 49-50. As general concerns, Dr. Helms mentioned that test takers may score lower if they are expected not to perform well, or if tests focus on "traditional ways of doing the job and the test-taker, in fact, uses innovative approaches." rd. at 51. Additionally, she suggested that "removing" "socioeconomic status" from test scores "reduces the disparate impact to some extent." Id. at 57. At the final hearing on March 18, 2004, defendant Ude, the Corporation Counsel, strongly advocated against certifying the exam results. He concluded: "You have a choice. It is my opinion that promotions under our rules as a result of these tests would not be consistent with federal law, would not be consistent with the purposes of our.Civil Service Rules or our Charter, nor is it in the best interests of the firefighters and Lieutenants who took the exams." Pl. Ex. Vol. IV(E) at 15-16. As a primary reason not to certify the results, Ude argued that the "results of previous exams in this department and in other 14
    47. departments have not had this kind of a result, which is one of the reasons why these results were so startling when they came down. These results were different." Id. at 19. He argued that Dr. Hornick's statements to the CSB, standing alone, were "sufficient" reason not to certify, and advised the board "that it is the employer's burden to justify the use of the examination" if a Title VII suit were brought. Id. at 21. Defendant Walton spoke "on behalf of the Mayor" and also advocated discarding the test results, primarily because the eligibility list, when combined with the Rule of Three and the number of vacancies then available, would "create a situation in which African-Americans are excluded from promotional opportunity on both the Captain and Lieutenant positions and Latinos are excluded from promotional opportunity on the Lieutenant examination." Id. at 30. She questioned whether there were "other ways of making the selection," that would be l~s~ "discriminatory." Id. at 31-32. The board split two to twoS on the question of certifying each exam, see id. at 70-73, as a result of which the promotional lists were not certified. Plaintiffs allege that the non-certification vote was due to SThe fifth member of the CSB, Barbara Tinney Jennings, was recused because her brother, Lt. Gary Tinney, was a candidate for promotion on the Captain's examination. She did not attend the hearings concerning these promotional exams. 15
    48. political pressure, particularly by defendant Rev. Boise Kimber, a vocal African-American minister who, it is acknowledged by all parties, is a political supporter and vote-getter for Mayor DeStefano. Plaintiffs' theory is that the defendants urged the CSB not to certify the results in the interest of pleasing minority voters and other constituents in New Haven whose priority was increasing racial diversity in the ranks of the Fire Department. Plaintiffs further argue that this pattern of political manipulation is in keeping with prior actions by the City of New Haven disregarding the Charter-mandated Rule of-Three in promotional decisions in the City's police and re departments. In support of this argument, plaintiffs proffer evidence regarding prior litigation in the Connecticut Superior and Appellate Court, the substance and outcome of which is largely admitted by defendants,6 and which resulted in sharp rebukes against the City for violating the civil service rules. See Pl. L.R. 56(a)1 Stmt. ~~ 64-90; Def. Am. L.R. 56(a)2 Stmt. ~~ 64-90. Plaintiffs argue that the apparent racial disparity in the results of the Lieutenant and Captain exams was due to the fact that hiring into, and promotion within, the Fire Department historically has been based on political patronage and promotion 6Defendants challenge the relevance of this evidence; however, as the Court held in its ruling on defendants' motion to strike, such evidence is relevant as background information to the present case. 16
    49. of racial diversity rather than merit; and they argue that the higher-scoring firefighters simply studied harder. In addition, they argue that the evident disparity was not appreciably worse on the 2003 examinations than other past promotional examinations. Defendants argue that "the decision not to certify [the test] results was mandated by anti-discrimination federal," state and local laws." Def. Mem. in Support of Mot. for Summary Judgment [Doc. # 52] at 4. Alternatively, they argue that they had a good faith belief that Title VII mandated non-certification of the examinations, and they cannot be liable under Title VII for attempting to comply with that very statute. Defendants additionally argue that plaintiffs lack standing to bring their Equal Protection claim, or, if they do have standing, the claim lacks merit because all firefighters were treated the same, regardless of race, as no orie was promoted as ~ result of the contested exams. Plaintiffs counter that a "good faith belief" that certifying the test results would violate Title VII does not constitute a defense, as a matter of law, to an allegation of Title VII or Equal Protection violations against the plaintiffs. II. Standard Summary judgment is appropriate where "there is no genuine issue as to any material fact and ... the moving party is 17
    50. entitled to a judgment as a matter of law." Fed. R. Civ. P. 56(c). An issue of fact is "material" if it "might affect the outcome of the suit under the governing law," and is "genuine" if "the evidence is such that a reasonable jury could return a verdict for the nonmoving party." Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986). On cross-motions for summary judgment "neither side is barred from asserting that there are issues of fact, sufficient to prevent the entry of judgment, as a matter of law, against it. When faced with cross-motions for summary judgment, a district court is not required to grant judgment as a matter of law for one side or the other." Heublein, Inc. v. United States, 966 F.2d 1455, 1461 (2d Cir. 1993) (citing Schwabenbauer v. Board of Educ. of Olean, 667 F.2d 305, 313 (2d Cir. 1981». "Rather, the court must evaluate each party's motion on its own merits, taking care in each instance to draw all reasonable inferences against the party whose motion is under consideration." Schwabembauer, 667 F.2d at 314. III. Discussion A. Title VII Plaintiffs argue that defendants' decision and/or advocacy against certifying the exam results amounted to intentional discrimination against plaintiffs, 17 of whom are white and one of whom is Hispanic, in favor of Hispanic and African-American 18
    51. examinees who were favored due to their race and their alleged political support of Mayor DeStefano, via the Rev. Boise Kimber. Plaintiffs essentially argue that defendants' professed desire to comply with Title VII's anti-disparate-impact requirements was in fact a pretext for intentional discrimination against white candidates. Because plaintiffs allege intentional discrimination, the familiar McDonnell Douglas three-prong burden-shifting test applies. 1. Burden-Shifting Framework Under that framework, plaintiffs first must establish a prima facie case of discrimination on account of race. Weinstock v. Columbia Univ., 224 F.3d 33, 42 (2d Cir. 2000). To do so, they must prove: (1) membership in a protected class; (2) qualification for the position; (3) an adverse employment action; and (4) circumstances giving"rise to an inference of discrimination on the basis of membership in the protected class. "See, ~, McDonnell Douglas Corp. v.Green, 411 U.S. 792, 802 (1973), Graham v. Long Island R.R., 230 F.3d 34, 38 (2d Cir. 2000). "A plaintiff's burden of establishing a prima facie case is de minimis." Abdu-Brisson v. Delta Airlines, Inc., 239 F.3d 456, 467 (2d Cir. 2001). Defendants do not dispute the first three prongs of the test, but argue that plaintiffs cannot establish an inference of discrimination because all applicants were treated the same, as nobody was promoted off the examination 19
    52. lists. Proof of a prima facie case shifts the burden to defendant "to produce evidence that the plaintiff was [terminated] for a legitimate, nondiscriminatory reason. This burden is one of production, not persuasion; it can involve no credibility assessment." Reeves v. Sanderson Plumbing, 530 U.S. 133, 142 (2000). (internal citations, quotations, and alterations omitted). Defendant's burden is satisfied if the proffered evidence "'taken as true, would permit the conclusion that there was a nondiscriminatory reason for the adverse action.'" Schnabel v. Abramson, 232 F.3d 83, 88 (2d Cir. 2000) (quoting St. Mary's Honor Ctr. v. Hicks, 509 U.S. 502, 509 .(1993». In this case, defendants proffer a good faith attempt to comply with Title VII as their legitimate nondiscriminatory reason for refusing to certify the exams. If the employer articulates a neutral reason for the plaintiff's termination, the burden shifts back to the plaintiff to show pretext. That is, the plaintiff "may attempt.to establish that he was the victim of intentional discrimination by showing that the employer's proffered explanation is unworthy of credence." Reeves, 530 U.S. at 143. 2. Prima Facie Case Plaintiffs' evidence- and defendants' own arguments - show that the City's reasons for advocating non-certification were 20
    53. related to the racial distribution of the results. As the transcripts show, a number of witnesses at the CSB hearings, including Kimber, mentioned "diversityH as a compelling goal of the promotional process. Ude, Marcano, and Burgett specifically' urged the CSB not to certify the results because, given the number of vacancies at that time, no African-Americans would be eligible for promotion to either Lieutenant Dr Captain, and no Latinos would be eligible for promotion to Captain. They believed this to be an undesirable outcome that could subject the City to Title VII litigation by minority firefighters,- and the City's leadership to political consequences. Had the tests not yielded what defendants perceived as racially disparate results, defendants would not have advocated rejecting the tests, and plaintiffs would have had an opportunity to be promoted .. A jury could infer that the defendants were motivated by a concern that too many whites and not enough minorities would be promoted were the lists to be certified. Given their minimal prima facie burden, the Court will assume arguendo that plaintiffs have proffered sufficient evidence to satisfy the fourth prong of the prima facie case, given defendants' acknowledgment that racial concerns, i.e. the disparate impact of the test results on minority firefighters, provided the impetus for their actions. 21
    54. 3. Pretext/Discriminatory Intent Defendants proffer as their legitimate non-discriminatory reason that they desired to comply with the letter and the spirit of Title VII. Plaintiffs deride this ~feigned desire to 'comply' with Title VII," Pl. Mem. of Law [Doc. I 81] at 3, arguing that defendants in fact violated that statute, and their actions were a mere pretext for promoting the interests of African-American firefighters and political supporters of the mayor. As plaintiffs point out, this case presents the opposite scenario of the usual challenge to an employment or promotional examination, as plaintiffs attack not the use of allegedly racially discriminatory exam results, but defendants' reason for their refusal to use the results. See Pl. Mem. of Law at 32, 34 35. Ordinarily, as contemplated by the statute, the "complaining party" bears the burden of proving a disparate impact, and the "respondent" bears the burden" of "demonstrat[ing] that the challenged practi~e is job related fbr" the positio~ in question and consistent with business necessity," or, alternatively, the "complaining party" may prevail by showing that an alternative employment practice with less disparate impact existed and that the respondent failed to utilize"it. See 42 U.S.C. § 2000e-2(k). Here, the roles of the parties are in essence reversed, with the defendants, normally reflecting a "respondent" role in the Title VII disparate impact analysis, contending that use of the 22
    55. promotional exams, if they had been certified, would have had an adverse impact, and the plaintiffs, normally the "complaining party," arguing that the test results were suff iently job- related to be defensible under the law. a. Existence of Racially Disparate Impact Although the parties dispute the exact racial breakdown of candidates passing the Captain's test,' plaintiffs do not dispute that the results showed a racially adverse impact on African- American candidates for both the Lieutenant and Captain positions, as judged by the EEOC Guidelines. Pl. L.R. 56 Stmt. ~ 246; Def. L.R. 56 Stmt. ~ 246. Thus, it is necessarily undisputed that, had minority firefighters challenged the results the examinations, the City would have been in a position of defending tests that, under applicable Guidelines, presumptively had a disparate racial impact. Specifically, the EEOC "four-fifths rule" provides that a selection tool that yields "[al selection rate for any race, sex, or ethnic group which is less than· four-fifths (4/5) (or eighty percent) of the rate for the group with the highest rate will generally be regarded by the Federal enforcement agencies as evidence of adverse impact, while a greater than four-fifths rate 'Plaintiffs assert that 32% of African-American examinees passed the Captain's examination, while defendants assert the figure is 37.5%. See Marcano Aff., Def. Ex. 4, ~ 21; Pl. L.R. 56(a) Stmt. ~~ 244-47. 23
    56. will generally not be regarded by Federal enforcement agencies as evidence of adverse impacL" 29 C.F.R. § 1607.4(D). Here, the evidence shows that on the 2003 Lieutenant's exam the pass rate for whites was 60.5%, for African-Americans 31.6% and Hispanics 20%. The four-fifths score would be 48%. In other words, African-Americans had a pass rate that was about half the pass rate for Caucasians, yielding an adverse impact ratio ("AIR") of 0.59, significantly below the AIR of 0.80 that is presumed to not evidence adverse impact under the'EEOC Guidelines. See Pl. L.R. 56(a) Stmt. ~ 246; Def. L.R. 56(a) Stmt. ~ 246. While- the parties dispute the Captain's exam pass rate for African-Americans and Hispanics (see supra note 7), the pass rate far Caucasians was 88%, which is more than double that of minorities and thus by either party's statistic an AIR far below the four-fifths guideline is yielded. Plaintiffs argue that these AIRs were not appreciably different from those on past promotional exams, and therefore defendants' stated concern with avoiding adverse impact must be pretextual. The parties agree that the AIRs on the 1999 promotional examinations would have failed the four-fifths rule as well. The AIR for African~Americans on the 1999 Lieutenant's exam was 0.58, compared to 0.59 on the 2003 test. See Pl. L.R. 56(a) Stmt. ~ 246; Def. L.R. 56(a) Stmt. ~ 246. The 1999 Captain examination had an AIR of 0.45 on African-American test-takers. 24
    57. See Pl. Ex. Vol. I, 40 (1999 scores). However, it is also undisputed that, because of the Rule of Three, the pass rate is not synonymous with the promotion rate, because only the top three scorers may be considered for each vacant position. Thus, the rank of the minority applicants is also a key factor. In 2003, given the number of vacancies, it appeared that at most two Hispanics and no African-Americans would have the opportunity to be promoted to Captain, and no minor~ties would have the opportunity to be promoted to Lieutenant. Although the record lacks specification, witnesses at the CSB hearings testified to the effect that in 1999 more minority candidates had scored toward the top of the lists, and therefore had more promotional opportunities. In any event, in 2003 defendants' concern was with the absence of minority candidates potentially eligible to be promoted, and with the diversity of the Fire Department's management in general. Thus, the fact that the 1999 exams also had a statistically adverse impact yet were certified, while the 2003 results were not, is insufficient in itself to show that defendants' concerns about complying with Titl~ VII were pretextual. b. Validation Study and Less Discriminatory Alternatives Plaintiffs additionally argue that defendants' decision was pretextual because they failed to complete a validation study to 25
    58. test whether the 2003 exams could be defended as adequately job- related. Going further, plaintiffs argue that defendants were legally required to conduct such a validation study before rendering a decision on cert cation of the results. Title VII provides: nNotwithstanding any other provision . it shall not be an unlawful employment practice for . . . an employer to give and to act upon the results of any professionally developed 'ability test prov~ded,that such test~ its administration or action upon the results is not designed, intended or used to discriminate because of race, color, religion, sex or national ~rigin." 42 U.S.C. § 2000e-2(h). As plaintiffs concede, this section nprovides that professionally developed and properly validated tests ~re a defense to a claim of disparate impact." Def. Mem. of Law at 32 (emphasis supplied). The statute itself does not require employers to implement or continue to use any test simply because it is professionally developed, nor does it provide a defense to an employer who "use[s]" a test with a discriminatory impact where other less-discriminatory, equally effective, alternatives are available. 42 U.S.C. § 2000e-21h). Although plaintiffs argue that EEOC guidelines mandated that defendants conduct a validation study before deciding not to certify the exams, the language of the guidelines does not support such a requirement. A validation .study is a method for 26
    59. determining whether a test is sufficiently related to the position for which the test or other criterion is administered. The EEOC's Uniform Guidelines for Employee Selection Procedures create a presumption that "[t]he use of any selection procedure which has an adverse impact on the hiring, promotion, or other employment or membership opportunities of members of any race, sex, or ethnic group will be considered to be discriminatory and inconsistent with these guidelines, unless the procedure has been validated in accordance with these guidelines." 29 C.F.R. § l607.3(A). The Guidelines further state: Where two or more selection procedures are available which serve the user's legitimate interest in efficient and trustworthy workmanship, and which are substantially equally valid for a given purpose, the user should use the procedure which has been demonstrated to have the lesser adverse impact. Accordingly, whenever a validity study is called for by these guidelines, the user should include, as a part of the validity study, an investigation of suitable alternative selection procedures and suitable alternative methods of using the selection procedure which have as little adverse impact as possible, to determine the appropriateness of using or validating them in accord with these guidelines. Id. at § l607.3(B) . . Where a selection procedure results in an adverse impact on a race, sex, or ethnic group . . . and that group is a significant factor in the relevant labor market, the user generally should investigate the possible existence of unfairness for that group if it is technically feasible to do so. The greater the severity of the adverse impact on a group, the greater the need to investigate the possible existence of unfairness. 29 C.F.R. § l607.l4(B) (8) (b) The Guidelines provide technical 27
    60. guidance for three types of studies: criterion-related validity studies, content validity studies, and construct validity studies. See generally 29 C.F.R. § 1607.14. The Guidelines are written on the assumption that the employer would be defending a certain test and seeking to validate such test in response to a disparate impact challenge from protected group employees. They do not address the tuation in the present case of an employer rejecting a test without conducting a validation study. Nonetheless, it is evident from the language of the guidelines that a validation study is contemplated as one method by which an employer can defend its use of a test or other selection method it desires to utilize by demonstration that it is sufficiently job-related -to pass muster under the statute, despite a racially adverse impact. The guidelines do not require or mandate a validity study where an employer decides using a certain selection procedure that manifests this impact and plaintiff's argument that defendants violated Title VII by refusing to conduct a validity study before rejecting testing results is thus unpersuasive. Plaintiffs argue that the CSB did not have extensive evidence of the existence of other, less-discriminatory, and equally-effective selection measures. Dr. Hornick telephonically testified that other tests, particularly ones he had developed, generally yield less adverse impact, and mentioned that an 28
    61. ~assessment center approach" might benefit New Haven, without specifically explaining what that approach entailed. As plaintiffs argue, there was no testimony that an ~assessment center" approach has a demonstrably less adverse impact, and there is some evidence in the record in this case, including from Dr. Hornick's website, that such an approach may still have some adverse impact. Dr. Hornick acknowledged that he had not had time to review the exams carefully, and his comments illustrated lack of famil rity with the methods lOS utilized to develop the tests. He suggested that lack of internal review by members of the New Haven Fire Department could have yielded questions that were less relevant to the particular department, but offered no explanation of why such a circumstance would have an adverse impact on minority candidates in particular. Dr. Helms from Boston College testified that the racial disparity 6n the exams at issue were not significantly different from the statistical disparities apparent on standardized tests nationwide. Mr. Lewis, the arson specialist from the Department of Homeland Security, stated that he believed the tests were fair and focused on material that a Lieutenant or Captain should kno~. On the other hand, Dr. Hornick and representatives of the black firefighters' union suggested that the 60/40 weighting system for the oral and written examinations could have produced an adverse impact. The testimony suggested that changing the 29
    62. weighting system yielded increased minority pass rates and diversity in the ranks of Bridgeport fLrefighters and officers. Dr. Helms suggested that because different employees have different ways of doing the same job, the fact that approximately 2/3 of those interviewed for the JAQwere white could have unintentionally introduced a bias into the test instrument. She and Mr. Lewis also suggested that differences in the availability of formal training and informal mentoring may have created the disparate effect apparent in the results. Plaintiffs purport to counter this argument with affidavits emphasizing how much they studied and sacrificed to perform well on the exams, compared to their observations of the efforts of some other examinees, and point to the availability of study groups and informal mentoring in the department. It appears that the reasons for testing disparities remain elusive. Dr. Helms testified that many theories exist, but experts on standardized testing nationwide have been unable to satisfactorily fully explain the reasons for the disparity in . performance observed on many tests. Plaintiffs' argument boils down to the assertion that if defendants cannot prove that the disparities on the Lieutenant and Captain exams were due to a particular flaw inherent in those exams, then they should have certified the results because there was no other alternative in place. Notwithstanding the 30
    63. shortcomings in the evidence on existing, effective alternatives, it is not the case that defendants must certify a test where they cannot pinpoint its deficiency explaining its disparate impact under the four-fifths rule simply because they have not yet formulated a better selection method. c. Diversity Rationale The real crux of plaintiffs' argument is that defendants refused to explore alternatives or conduct a validity study because they had already decided that they did not like the inevitable promotional results if the process continued to its expected conclusion,8 and that their udiversityH rationale is prohibited as reverse discrimination under Title VII. In Haydenv. County of Nassau, 180 F.3d 42 (2d Cir. 1999), the Second Circuit held that race-conscious configuration of an entry-level police department exam did not violate Title VII or the Equal Protection Clause. In that case, the Nassau County Police Department was operating under several cons~nt decrees prohibiting it from engaging in discrimination in its selection of police officers, and particularly from utilizing examinations with disparate impact on minority applicants. Following development of a test by the county and Department of Justice 8Plaintiffs present evidence in the form of emails from the Mayor's staff suggesting they desired to convince the CSB not to certify, and further suggesting that if the CSB had certified, the Mayor would have announced his intention to refuse to forward the lists to the Fire Department for promotion. 31
    64. advisors, a validity analysis was conducted to determine which configuration of the test was sufficiently job-related "yet minimized the adverse impact on minority applicants. Of the twenty-five sections administered to the applicants, the [technical report] recommended that Nassau County use nine sections as the . test." Id. at 47. A class of White and Latino officers challenged use of the adjusted test under Title VII and the Fourteenth Amendment, inter alia, contending that the deliberate design of the test to reduce adverse impact on African~American candidates necessarily discriminated against them on the basis of race." The Court of Appeals rejected the plaintiffs' contentions, finding plaintiffs were "mistaken in treating racial motive.as a synonym for a constitutional .violation" and observing that "[e]veryantidiscrimination statute aimed at racial discrimination, and every enforcement measure taken under such a statute, reflect a concern with race. That does not make such enactments or actions unla~ftil or automatically suspect . . . " Id. at 48-49 (quoting Raso v. Lago, 135 F.3d II, 16 (1st Cir.») (internal quotation marks omitted). The Hayden court further held that the construction of the Nassau County test for the purpose of minimizing adverse impact on minorities was not intentional "reverse discrimination" against whites because the same nine test sections were used for all applicants, so it was "simply not analogous to a quota system or 32
    65. 33
    66. claim of discrimination to serve as a predicate for a voluntary compromise containing race-conscious remedies." Id. at 1130. The Second Circuit expanded Kirkland in Bushey v. New York State Civil Service Commission, 733 F. 2d 220 (2d Cir. 1984). There, the civil service commission had administer.ed a promotional examination that had a significant adverse impact, with non-minority applicants passing at almost twice the rate of minority ap.plicants. The defendants race-normed the scores for each group, increasing the pass rate of the minority group to the equivalent of the non~minority group, and effectively making an additional 8 minority individuals eligible for promotion, without taking any non-minorities off the list. The Court of Appeals held that the initial results, particularly ~the score distributions of minority and nonminority candidates, were sufficient to establish a prima facie showing of adverse impact," id. at 225, and, consistent with Kirkland, ~a showing of a prima facie case of employment discrimination through a statistical demonstration of disproportional racial impact constitutes a sufficiently serious claim of discrimination to serve as a predicate for employer-initiated, voluntary race-conscious remedies," id. at 228. In other words, a prima facie case is one way that a race-conscious remedy is justified, but it is not required: all that is required is ~a sufficiently serious claim of discrimination" to warrant such a remedy. Id. at 228; see 34
    67. also id. at 226 n. 7. In this case, the parties agree that the adverse impact ratios for African-American and Hispanic test-takers on both the Lieutenant and Captain exams were too low to pass muster under the EEOC's "four-fifths rule." As Kirkland and Bushey held, a statistical showing of discrimination, and particularly a pass rate below the "four-fifths rule," is sufficient to make out a prima facie case of discrimination, and therefore sufficient to justify voluntary race-conscious remedies. 9 Here, defendants' remedy is "race conscious" at most because their actions reflected their intent not to implement a promotional process based on testing results that had an adverse impact on African- Americans and Hispanics. The remedy chosen here was decidedly less "race conscious" than the remedies in Kirkland and Bushey, 9Plaintiffs denigrate reliance on Kirkland and Bushev on the grounds that the "race-norming" procedures utilized in those cases would be unlawful under the 1991 amendments to the Civil Rights Act. 42 U.S.C. § 2000e-2(1) ("It shall be an unlawful employment practice for a respondent, in connection with the selection or referral of applicants or candidates for employment or promotion, to adjust the scores of, use different cutoff scores for, or otherwise alter the .results of, employment related tests on the basis of race, color, religion, sex, or national origin."). See also Hayden, 180 F.3d at 53 (this provision was "intended to prohibit 'race norming' and other methods of using different cut-offs for different races or altering scores based on race.") (emphasis in original). While plaintiffs are correct that Title VII now prohibits race-norming, none is alleged to have happened here and the 1991 amendments do not affect the reasoning and holding of either case, namely, that a showing of a "sufficiently serious claim of discrimination" is adequate to justify race conscious, remedial measures. 35
    68. because New Haven did not race-norm the scores, they simply decided to start over, to develop some new assessment mechanism with less disparate impact. Thus, while the evidence shows that race was taken into account in the decision not to certify the test results, the result was race-neutral: all the test results were discarded, no one was promoted, and firefighters of every race will have to participate in another selection process to be considered for promotion. Indeed, there is a total absence of any evidence of discriminatory animus towards plaintiffs - under the reasoning of Hayden, 180 F.3d at 51, "nothing in our jurisprudence precludes the use of race-neutral means to improve racial and gender representation. [T]he intent to remedy the disparate impact of the prior exams is not equivalent to an intent to discriminate against non-minority applicants. H1o lOTaxman v. Bd. of Educ. of T'wp of Piscataway, 91 F.3d 1547, 1558 (3d Cir. 1996) (en banc) , cert. dismissed, 522 U.S. 1010 (1997), relied on by plaintiffs, is readily distinguishable. There, the board of education relied on an affirmative action plan to defend its decision to layoff a white teacher instead of a black teacher with equal seniority, and the Third Circuit held that promoting racial diversity on the faculty, absent a history of past discrimination, was insufficient justification for laying off the white teacher because of her race and violated Title VII. Here, defendants had ample statistical evidence that the tests had an adverse impact on minority candidates and importantly did not opt to select black applicants over white applicants for promotion, but rather decided to select nobody at all. Williams v. Consolo City of Jacksonville, No. 00cv469, 2002 U.S. Dist. LEXIS 27066 (D. Fla. July 5, 2002), can similarly be distinguished as that case did not concern a decision not to certify test results, but rather a post-certification decision not to create the positions which would result in plaintiffs' promotions because plaintiffs were not African-American. 36
    69. Plainti contend that Hayden is distinguishable by the fact that the remedy approved there was pursuant to previous consent decrees; they do not explain why they view this distinction as significant. As Bushey held, it would contravene the remedial purpose of Title VII if an employer were required to await a lawsuit before voluntarily implementing measures with less discriminatory impact. Bushey, 733 F.2d at 227 (rejecting the plaintiffs' argument that the remedial measures in Kirkland were only permissible as part of a settlement in that case, because that "would create an anomalous situation. It would require an employer. ' . . ' to issue a presumptively discriminatory eligibility list, wait to be sued by minority candidates, and only then seek a settlement .... Such an approach would serve no purpose other than to impede the process of voluntary compliance with Title VII and cause the proliferation of litigation in all such cases . . . . ff). Plaintiffs also attempt to distinguish Hayden on the grounds that the challengers to that test, which was constructed from the nine most job-related sections with the least disparate impact, were not injured or disadvantaged, whereas "the instant plaintiffs have been both injured, as they were deprived of promotions, and disadvantaged as they will now be forced to compete once again." Pl. Mem. of Law at 58. Plaintiffs take this language from Hayden out of context. In holding that the 37
    70. Hayden plaintiffs did not prove disparate impact on nonminority applicants, the Court of Appeals held that because "appellants continued to score higher than black candidates, on average, the exam did not impair or disadvantage these appellants in favor of African-American applicants. Thus, appellants are unable to set forth a claim that they endured any disparate impact as a result of the design and administration of the . examination." Hayden, 180 F:3d at 52. Here, plaintiffs. allege disparate .treatment, not disparate impact. Nor do they have a viable claim of disparate impact because the decision to disregard the test results affected all applicants equally, regardless of race - all applicants will have to participate in a new test or selection procedure. 11 Furthermore, plaintiffs were not "deprived of promotions." As the parties agree, under New Haven's civil service ru~es, no applicant is entitled .to promotion unless and until the CSB certifies the results. Even then, application of the Rule of Three would give top scorers an opportunity for promotion, depending on the number of vacancies, but no guarantee of llWhile plaintiffs, who describe their considerable efforts to perform well on this infrequently given promotion exam, are understandably disappointed and frustrated that their successful study efforts have come to naught this time, this result is not evidence of being disadvantaged because of their race nor evidence of disparate impact because it does not show injury or disadvantage, only uncertainty as to their performance in the City's next promotion selection process. 38
    71. promotion; it is even conceivable that the applicant with the highest score never would be promoted. See United States v. City of Chicago, 869 F.2d 1033, 1038 (7th Cir. 1989) (where state law permitted promotion from among five highest-ranked individuals on eligibility list, challenger had no property right to promotion: ft a roster ranking may create an expectation of promotion, but an officer has no entitlement to a particular roster position or to promotion."); Bridgeport Firebird Society v. City of Bridgeport, 686 F. Supp. 53, 58 (ftAt best, the provisions of the City Charter [mandating a Rule of One for promotions] provide the refighters ranked on the eligibility list only with a mere expectation of promotion, which does not rise to the level of a legally protected interest, especially in the face of 'presumptively discriminatory employment practices.'") (quoting Kirkland, 711 F. 2d at 1126». Thus, while the facts of Hayden were slightly different than those here, the Court finds the holding quite relevant and instructive. Defendants' motivation to avoid making promotions based on a test with a racially disparate impact, even in a political context,12 does not, as a matter of law, constitute 12Assuming arguendo that political favoritism or motivations may be shown to have been intertwined with the race concern, that does not suffice to establish a Title VII violation. See,~, EEOC v. Flasher Co., Inc., 986 F.2d 1312, 1321 (10th Cir. 1992) (pretext is not shown merely because ft some less seemly reason personal or political favoritism, a grudge, random conduct, an error in the administration of neutral rules actually accounts 39
    72. discriminatory intent, and therefore such evidence is insufficient for plaintiffs to prevail on their Title VII claim. Accordingly, the Court will grant defendants' motion and deny plaintiffs' motion for summary judgment on this claim. B. Equal Protection Claim Plaintiffs argue that defendants violated the Equal Protection Clause either by employing a race-based classification system for promotion or, alternatively, by applying facially neutral promotion criteria in a racially discriminatory manner. Defendants counter that they did not employ any racial classi cations because every applicant was treated the same when the CSB decided that nobody would be promoted off the lists, and there was no discriminatory intent against whites motivating their non-certification decision. Additionally, defendants argue that plaintiffs lack standing to bring an Equal Protection claim. 1. Standing Defendants acknowledge, as they must, that non-minorities have been found to be in a protected group for purposes of standing under the Equal Protection Clause. See,~, Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 210 (1995) (holding that non-minority-owned business' ~allegation that it has lost a contract in the past because of a [minority set~aside] subcontractor compensation clause of course entitles it to seek for the decision") . 40
    73. damages for the loss of that contract."). However, defendants argue that because plaintiffs have not suffered any harm, and specifically because plaintiffs were not "passed over for a benefit that was given to an allegedly less deserving minority," Def. Reply Mem. at 37, they lack standing. Defendants confuse standing with the merits of the case. The constitutional injury plaintiffs claim here is not failure to be promoted, but failure to be treated equally on the basis of race. Plaintiffs have standing to bring such a claim. Comer v. Cisneros, 37 F.3d 775~' 791 (2d Cir. 1994) (plaintiff had standing to bring equal protection claim where she alleged that the defendant's Section 8 housing subsidy program "rules and regulations, in their administration, violate the Constitution because they erect a barrier that makes it more difficult for economically disadvantaged blacks to obtain a housing benefit than it was for rion-minorities") . 2. Racial Classification/Discriminatory Intent Plaintiffs' Equal Protection claim, however, lacks merit, with respect to both the racial classification and disparate treatment arguments. As the Second Circuit held in Hayden when rejecting plaintiffs' classification argument, if an exam is "administered and scored in an identical fashion for all applicants," there is no racial classification. Hayden, 180 F.3d at 48. Further, a "desire" "to design an entrance exam which 41
    74. would diminish the adverse impact on black applicants does not constitute a 'racial classification.'" Id. Here, all applicants took the same test, and the result was the same for all because the test results were discarded and nobody was promoted. This does not amount to a facial classification based on race. I3 Likewise, where a test is administered and scored in the same manner for all applicants, plaintiffs cannot make out a claim that the exam was a facially neutral test used in a discriminatory manner. at 50. Plaintiffs argue that their equal protection rights were violated because they passed the tests and therefore were not similarly-situated to minority applicants who failed~Plainti argue that if a black employee "shows up for work and works a full day" and a white employee does,not, and the black employee 13Therefore, plaintiffs' reliance on Berkley v. United '287 F.3d 1076 (Fed. Cir. 2002), is unavailing. In that case, the Air Force employed facially different criteria for selecting women and minority employees for layoff compared to white male employees, and the Federal Circuit held that such a program should be subjected to strict scrutiny (without ruling on the merits). Likewise, in Dallas Fire Fighters Assoc. v City of Dallas, 150 F.3d 438 (5th Cir. 1998), also relied on by plaintiffs, the city followed an affirmative action plan that specifically called for promoting African-American, Hispanic and female firefighters out of rank, ahead of white and Native American male fighters with higher test scores. Here, no classification system was employed, as the test results were discarded for every examinee regardless of race. While defendants clearly were concerned with achieving diversity in the department by enhancing minority promotional opportunity, plaintiffs offer no evidence that defendants employed an actual race-based affirmative action plan that advantaged minority over white applicants for promotion. 42
    75. complains "that he was due his wages,H the employer cannot be heard to defend the complaint on the ground that the employees were treated the same because neither was paid. Pl. Mem. in Opp. at 64. Plaintiffs' analogy is faulty because performing well on the exam does not create an entitlement to promotion, whereas working entitles an employee to be paid. Second, a presumptively . flawed test result may not be a proper measure for determining whether anyone should be promoted. Finally, plaintiffs cannot show that defendants acted out of an intentionally discriminatory purpose. "Discriminatory purpose 'implies that the decisionmaker ... selected or reaffirmed a particular course of action at least in part 'because of,' not merely 'in spite of,' its adverse effects upon an identifiable group.'H (quoting Personnel Administrator v. Feeney, 442 U.S. 256, 279 (1979». Nothing in the record in this case suggests that the City defendants or CSB acted "because of~' discriminatory animus toward plaintiffs or other non-minority applicants for promotion. Rather, they acted based on the following concerns: that the test had a statistically adverse impact on African- American and Hispanic examinees; that promoting off of this list would undermine their goal of diversity in the Fire Department and would fail to develop managerial role models for aspiring firefighters; that it would subject the City to public criticism; and that it would likely subject the City to Title VII lawsuits 43
    76. from minority applicants that, for political reasons" the City did not want to defend. "[T]he intent to remedy the disparate impact of [the tests] is not equivalent to an intent to discriminate against non-minority applicants." Hayden, 180 F.3d at 51. None of the defendants' expressed motives could suggest to a reasonable juror that defendants acted,"because of" animus again~t non-minority firefighters who took the Lieutenant and Captain exams. Accordingly, defendants' motion for summary judgment on this claim will be granted and plaintiffs' motion will be denied. 14 C. Civil Rights Conspiracy Title 42 U.S.C. § 1985(3) permits recovery of damages if a plaintiff can prove a conspiracy "for the purpose of depriving, either directly or indirectly, any person or class of persons of the equal protection of the laws, or of equal privileges and immunities under the laws." Because the Court has found that plaintiffs fail to present sufficient evidence that their equal protection rights were violated, their § 1985 conspiracy claim must fail as well. See Mian v. Donaldson, Lufkin & Jenrette Securities Corp., 7 F.3d 1085, 1088 (2d Cir. 1993) (evidence of "racial or perhaps otherwise class-based, invidious discriminatory animus" required to prevail on § 1985 claim). 14For this reason the Court need not reach defendants' arguments that they are entitled to qualified immunity on the Equal Protection claim. 44
    77. Accordingly, defendants' motion for summary judgment on this claim will be granted. D. First Amendment Defendants additionally move for summary judgment on plaintiffs' First Amendment freedom-of-association claim, which motion will also be granted. Plaintiffs do not attempt to rebut defendants' contentions that plaintiffs have not identified a free speech activity in which they participated nor claimed that any chilling of speech resulted. Rather, plaintiffs argue that th~ CSB's non- certification decision, and the City defendants' advocacy of that decision, resulted from political pressure by defendant Kimber, who threatened the CSB with ~political ramifications H if they voted to certify the results. Plaintiffs argue that ~a jury could rationally infer that city officials worked behind the scenes to sabotage the promotional examinations because they knew that, were the exams certified, the Mayor would incur the wrath of Kimber and other influential leaders of New Haven's African- American community." PI. Mem. in Opp. at 73. While a jury could make such an inference, it would not lead to the conclusion that plaintiffs' First Amendment right to freedom of association was violated as. a matter of law. The evidence shows that Kimber spoke at the first CSB hearing and strenuously argued against certification, and the City defendants 45
    78. do not dispute that Kimber is a close political ally of the Mayor. However, there is no evidence in the record to suggest that the non-certification decision was made in retaliation for plaintiffs' refusal to "associate with," or their expression of disagreement with, Kimber. As with the Equal Protection claim, the fact that defendants desired to avoid the wrath of one group (in this case African-American firefighters and other political supporters of Kimber and DeStefano) does not logically lead to the conclusion that defendants intended to discriminate or retaliate against plaintiffs because they were not members of that group. More importantly, there is no evidence in the record even to. suggest that defendants· knew plaintiffs' political affiliations, i.e., whether they supported Kimber and/or DeStefano on any issue other than the certification of these particular exam re?ults. In sum, in plaintiffs' terms, the record shows that defendants acted to head off.the potential adverse impact of the promotion tests on African-American and Hispanic firefighters in order to curry favor with minority voters and political leaders in the City, but it does not contain any evidence of an intent or purpose to target plaintiffs for not supporting that political coalition or its interests. Thus, defendants' motion for summary judgment on the First Amendment claim must be granted. 46
    79. 47
    80. distress. The Clerk is directed to close this case. IT IS SO ORDERED. lsi Janet Bond Arterton United States Dis Judge Dated at New Haven, Connecticut this 28th day of September, 2006. 48
    81. 06-4996-cv Ricci v. DeStefano UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT ____________________________________ August Term, 2007 (Argued: December 10, 2007 Decided: June 9, 2008 En Banc Concurrence Decided: June 13, 2008) Docket No. 06-4996-cv ____________________________________ FRANK RICCI, MICHAEL BLATCHLEY, GREG BOIVIN, GARY CARBONE, MICHAEL CHRISTOFORO, RYAN DIVITO, STEVEN DURAND, WILLIAM GAMBARDELLA, BRIAN JOOSS, JAMES KOTTAGE, MATTHEW MARCARELLI, THOMAS J. MICHAELS, SEAN PATTON, CHRISTOPHER PARKER, EDWARD RIORDON, KEVIN ROXBEE, TIMOTHY SCANLON, BENJAMIN VARGAS, JOHN VENDETTO and MARK VENDETTO, Plaintiffs-Appellants, v. JOHN DESTEFANO, KAREN DUBOIS-WALTON, THOMAS UDE JR., TINA BURGETT, BOISE KIMBER, MALCOLM WEBER, ZELMA TIRADO and CITY OF NEW HAVEN, Defendant-Appellees. ____________________________________ CALABRESI, Circuit Judge, concurring in the denial of rehearing en banc: 1 I join entirely Judge Parker’s opinion concurring in the denial of a rehearing en banc. I also 2 join fully Judge Katzmann’s opinion because, as he points out, going en banc is unnecessary as all 3 that is involved in this case has already been described in the filed opinions. I write today to -1-
    82. 1 emphasize one reason that, I believe, makes it particularly inappropriate for us to exercise our purely 2 discretionary power to review this case en banc. 3 The question of whether a municipality incurs liability when, motivated only by a desire to 4 comply with federal anti-discrimination law, it takes race-neutral actions that have racially 5 significant consequences, is undoubtedly an interesting one.1 To reach that question one must, 6 however, first examine whether the municipality’s proffered desire to comply with federal law is in 7 good faith and not a pretext. After that, we would normally ask whether that asserted desire, 8 although in good faith, is not also in part motivated by other, racial, considerations. In this case, the 9 municipality claimed that its actions were grounded solely in the desire to comply with federal law. 10 The plaintiffs alleged instead that this was not the real reason for the city’s actions, and asserted that 11 the city had other less salubrious, and directly racial-political, reasons for what it did. 12 The district court and the panel readily rejected the notion that the city’s stated reason was 13 just a pretext. But neither court went on to consider whether the city was influenced by mixed 14 motives. And that is why Judge Cabranes, in his dissent from the denial of en banc review, suggests 15 that, since the plaintiffs alleged that their race motivated the defendants’ decision, the district court 16 should have undertaken such a mixed motive analysis. He contends, that is, that the courts should 17 have examined the situation as one in which a legitimate motive may have combined with an 18 improper motive to bring about the challenged action. See Price Waterhouse v. Hopkins, 490 U.S. 19 228 (1989). He would be precisely right . . . except for the fact that that type of analysis is not 1 For an exceptionally thoughtful and thorough discussion of this area, see Richard A. Primus, Equal Protection and Disparate Impact: Round 3, 117 Harv. L. Rev. 494 (2003). -2-
    83. 1 available to us in this case. It is not available for the most traditional of legal reasons. The parties 2 did not present a mixed motive argument to the district court or to the panel.2 3 It is the unavailability of mixed motive analysis that makes this case an especially undesirable 4 one for elective review. The interesting issue the case might present – concerning the obligations 5 of a municipality seeking only to comply with the relevant federal anti-discrimination law – is, in 6 the circumstances before us, clouded by the allegations that something more is going on. Given the 7 plaintiffs’ failure to argue mixed-motive analysis, those allegations cannot be adequately evaluated. 8 But they nevertheless cannot help but affect how we look at the city’s actions. And they may even 9 influence, inappropriately, how we are inclined to rule on the underlying, “interesting” issue. 10 Difficult issues should be decided only when they must be decided, or when they are truly 11 well presented. When they need not be decided – and rehearing en banc is always a matter of choice, 12 not necessity – it is wise to wait until they come up in a manner that helps, rather than hinders, clarity 13 of thought. That is not so in this case. 14 For this reason too, I concur in the denial of rehearing en banc. 15 2 It is unavailable, that is, unless we reach out and consider a legal theory that the parties have eschewed. Sometimes – for example, in matters of life and death – such a reaching out may be appropriate. But generally, and specifically in this case, it is not. -3-
    84. 06-4996-cv Ricci v. DeStefano 1 DENNIS JACOBS, Chief Judge, dissenting from the denial of 2 rehearing in banc: 3 4 Along with almost half of the members of this Court, I 5 join Judge Cabranes’s dissent, which does the heavy lifting 6 on the procedural merits of in banc review. I write 7 separately to answer respectfully the concurring opinions of 8 Judge Calabresi and Judge Katzmann. 1 9 Judge Katzmann and those of my colleagues who signed 10 his opinion “recognize” that this case “presents difficult 11 issues,” but would leave further review and consideration to 12 the Supreme Court, citing a Circuit “tradition” of deference 13 to panel adjudication. In effect, this has become a Circuit 14 tradition of hearing virtually no cases in banc. 15 The grant or denial of in banc review is governed by 16 Fed. R. App. P. 35, which says that in banc rehearing is 17 disfavored--unless such review is needed for coherence of 1 I have not solicited concurrences for my opinion. 1
    85. 1 the Court’s decisions or “the proceeding involves a question 2 of exceptional importance.” Fed. R. App. P. 35(a). 3 Accordingly, the next subdivision of Rule 35 requires the 4 petition to explain why the case falls within one or both of 5 these categories. See Fed. R. App. P. 35(b). 6 This weighing calls for an exercise of discretion. 7 Judge Calabresi’s concurring opinion deprecates this 8 standard as a “purely discretionary power” that is “always a 9 matter of choice” (emphasis added). He nevertheless 10 “join[s] fully” in both Judge Parker’s opinion, which 11 counsels against in banc review as a matter of (plain 12 ordinary) discretion, and Judge Katzmann’s opinion, which 13 decides against in banc review as a matter of tradition. I 14 understand Judge Calabresi to be saying, in effect, that 15 when it comes to in banc review, discretion should be 16 leavened by caprice. As applied to this case, that means 17 that there might be discretionary grounds for denying in 18 banc review were it not already foreclosed by tradition.2 2 In the alternative, Judge Calabresi contends that we cannot consider whether the District Court applied the correct legal standard to plaintiffs’ Title VII claim because the “parties did not present [that] argument to the district court or the panel” and we can only consider a 2
    86. 1 This occluded view of our discretion to sit in banc 2 runs counter to the criteria set down for our guidance in 3 Rule 35. No doubt, the proper exercise of discretion 4 results in the denial of review in the overwhelming number 5 of cases. And the resulting pattern may resemble the 6 pattern of denial that would result from saying “no” by 7 tradition. But the decision to grant or deny in banc review 8 is like any other discretionary decision in the sense that 9 discretion should be exercised, not elided or stuck in a “legal theory that the parties have eschewed” in such circumstances as “matters of life and death.” Judge Calabresi provides no authority for this proposition for the good reason that it is unsound. Writing for a unanimous Supreme Court, Justice Thurgood Marshall explained that “[w]hen an issue or claim is properly before the court, the court is not limited to the particular legal theories advanced by the parties, but rather retains the independent power to identify and apply the proper construction of governing law.” Kamen v. Kemper Fin. Servs., Inc., 500 U.S. 90, 99 (1991) ; see also Hankins v. Lyght, 441 F.3d 96, 104 (2d Cir. 2006) (“We are required to interpret federal statutes as they are written . . . and we are not bound by parties’ stipulations of law.”); Neilson v. D'Angelis, 409 F.3d 100, 105 n.2 (2d Cir. 2005) (“The parties’ apparent agreement on the standard of ‘similarity’ for ‘class of one’ cases does not control our judgment, because this court is not bound by stipulations of law.”); United States v. Pabon-Cruz, 391 F.3d 86, 97 (2d Cir. 2004) (“It is clear that we have the authority to resolve this question despite its not having been raised in the District Court proceedings or in the parties’ initial briefs.”). 3
    87. 1 default position. See United States v. Campo, 140 F.3d 415, 2 419 (2d Cir. 1998) (holding that “refusal to exercise 3 discretion accorded [the court] by law . . . constitutes an 4 error of law”). 5 The exercise of discretion to hear cases in banc is 6 integral to the judicial process. The advisory notes 7 emphasize that “an en banc proceeding provides a safeguard 8 against unnecessary intercircuit conflicts.” See Fed. R. 9 App. P. 35, Advisory Committee Notes (1998 Amendments). In 10 other words, issues of exceptional importance that may 11 divide the circuits should be subject to in banc review lest 12 a three-judge panel adopt a rule of law that would not 13 command a majority vote of the appeals court as a whole, and 14 thereby provoke an avoidable circuit conflict that the 15 Supreme Court would have to resolve. 16 That is why I respectfully disagree with those of my 17 colleagues who are pleased to defer as a matter of tradition 18 to the ruling of the three-judge panel, and thereby leave 19 further consideration to the Supreme Court. Cf. Landell v. 20 Sorrell, 406 F.3d 159, 167 (2d Cir 2005) (Sack, J., and 4
    88. 1 Katzmann, J., concurring) (observing that in banc hearing 2 should be avoided where it “would only forestall resolution 3 of issues destined appropriately for Supreme Court 4 consideration”). 5 I do not think it is enough for us to dilate on 6 exceptionally important issues in a sheaf of concurrences 7 and dissents arguing over the denial of in banc review. If 8 issues are important enough to warrant Supreme Court review, 9 they are important enough for our full Court to consider and 10 decide on the merits. Of course, if an in banc poll 11 discloses broad-based agreement with the panel opinion, in 12 banc review may be a spinning of wheels. Under such 13 circumstances, it may very well be an appropriate exercise 14 of discretion to deny rehearing in banc. But to rely on 15 tradition to deny rehearing in banc starts to look very much 16 like abuse of discretion. 5
    89. UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT August Term, 2005 (Argued: June 8, 2006 Decided: January 25, 2007) Docket Nos. 04-6692-ag(L), 04-6693-ag(CON), 04-6694-ag(CON), 04-6695-ag(CON), 04-6696-ag(CON), 04-6697-ag(CON), 04-6698-ag(CON), 04-6699-ag(CON) _____________________________________________ RIVERKEEPER, INC., NATURAL RESOURCES DEFENSE COUNCIL, WATERKEEPER ALLIANCE, SOUNDKEEPER, INC., SCENIC HUDSON, INC., SAVE THE BAY-PEOPLE FOR NARRAGANSETT BAY, FRIENDS OF CASCO BAY, AMERICAN LITTORAL SOCIETY, DELAWARE RIVERKEEPER NETWORK, HACKENSACK RIVERKEEPER, INC., NEW YORK/NEW JERSEY BAYKEEPER, SANTA MONICA BAYKEEPER, SAN DIEGO BAYKEEPER, CALIFORNIA COASTKEEPER, COLUMBIA RIVERKEEPER, CONSERVATION LAW FOUNDATION, SURFRIDER FOUNDATION, STATE OF RHODE ISLAND, STATE OF CONNECTICUT, STATE OF DELAWARE, COMMONWEALTH OF MASSACHUSETTS, STATE OF NEW JERSEY, STATE OF NEW YORK, APPALACHIAN POWER COMPANY, ILLINOIS ENERGY ASSOCIATION, UTILITY WATER ACT GROUP, PSEG FOSSIL LLC, PSEG NUCLEAR LLC, ENTERGY CORPORATION, Petitioners, – v. – UNITED STATES ENVIRONMENTAL PROTECTION AGENCY, STEPHEN L. JOHNSON, in his official capacity as ADMINISTRATOR OF THE UNITED STATES ENVIRONMENTAL PROTECTION AGENCY, Respondents. ____________________________________________ Before: STRAUB, SOTOMAYOR, and HALL, Circuit Judges. ____________________________________________ Petitioners challenge a final rule promulgated by the Environmental Protection Agency pursuant to section 316(b) of the Clean Water Act, 33 U.S.C. § 1326(b), that is intended to protect aquatic organisms from being harmed or killed by cooling water intake structures at large, existing power-producing facilities. While we conclude that certain aspects of the rule are
    90. based on a reasonable interpretation of the Clean Water Act and supported by substantial evidence in the administrative record, several aspects of the rule are not consistent with the statute, are not supported by sufficient evidence, or were not properly subject to notice and comment. We therefore grant in part and deny in part the petitions for review and dismiss in part one aspect of the petitions for lack of jurisdiction because there is no final agency action to review. REED W. SUPER, Morningside Heights Legal Services, Inc., Environmental Law Clinic, Columbia University School of Law (Michelle Avallone, Julia Errea, Vivian Mills, Ian Dattner, Monique Mendez, Misti Duvall, Devon Knowles, Molly McOwan, Adam Orford, Scott Sneddon, on the brief; P. Kent Corell, of counsel), New York, New York, for Petitioners Riverkeeper, Inc., Natural Resources Defense Council, Waterkeeper Alliance, Soundkeeper, Inc., Scenic Hudson, Inc., Save the Bay–People for Narragansett Bay, Friends of Casco Bay, American Littoral Society, Delaware Riverkeeper Network, Hackensack Riverkeeper, Inc., New York/New Jersey Baykeeper, Santa Monica Baykeeper, San Diego Baykeeper, California Coastkeeper, Columbia Riverkeeper, Conservation Law Foundation, and Surfrider Foundation. TRICIA K. JEDELE, Special Assistant Attorney General of Rhode Island, Providence, Rhode Island (Patrick C. Lynch, Attorney General of Rhode Island; Michael Rubin, Special Assistant Attorney General, Providence, Rhode Island; Richard Blumenthal, Attorney General of Connecticut, Kimberly Massicotte and Matthew Levine, Assistant Attorneys General, Hartford, Connecticut; Carl C. Danberg, Attorney General of Delaware, Kevin Maloney, Deputy Attorney General, Wilmington, Delaware; Thomas F. Reilly, Attorney General of Massachusetts, Andrew Goldberg, 2
    91. Assistant Attorney General, Boston, Massachusetts; Zulima V. Farber, Attorney General of New Jersey, Ellen Barney Balint, Deputy Attorney General, Trenton, New Jersey; Eliot Spitzer, Attorney General of New York, Maureen F. Leary, Assistant Attorney General, Albany, New York, on the brief), for State Petitioners Rhode Island, Connecticut, Delaware, Massachusetts, New Jersey, and New York. KRISTY A.N. BULLEIT, Hunton & Williams, Washington, D.C. (James N. Christman, Elizabeth E. Aldridge, Hunton & Williams, Richmond, Virginia, on the brief), for Petitioners Appalachian Power Company, Illinois Energy Association, and Utility Water Act Group. KARL S. LYTZ, Latham & Watkins LLP, San Francisco, California (Christopher J. McAuliffe, PSEG Services Corporation, Newark, New Jersey; David J. Hayes, Cassandra Sturkie, Latham & Watkins LLP, Washington, D.C., on the brief), for Petitioners PSEG Fossil LLC and PSEG Nuclear LLC. CHUCK D. BARLOW, Entergy Services, Inc., Jackson, Mississippi and ELISE N. ZOLI, Goodwin Procter LLP, Boston, Massachusetts (Robert H. Fitzgerald, U. Gwyn Williams, Goodwin Procter LLP, Boston, Massachusetts, on the brief), for Petitioner Entergy Corp. DAVID S. GUALTIERI, CYNTHIA J. MORRIS, and JESSICA O’DONNELL, United States Department of Justice (Sue Ellen Woolridge, Assistant Attorney General, John C. Cruden, Deputy Assistant Attorney General, on the brief; Leslie J. Darman, United States Environmental Protection Agency, of counsel), Washington, D.C., for Respondents. Lisa Madigan, Attorney General of Illinois, Matthew Dunn, Chief, Environmental Enforcement/ 3
    92. Asbestos Litigation Division, Ann Alexander, Environmental Counsel and Assistant Attorney General, Chicago, Illinois, for Amicus Curiae State of Illinois. Jon Bruning, Attorney General of Nebraska, Lincoln, Nebraska (David D. Cookson, Special Counsel to the Attorney General, Lincoln, Nebraska; Troy King, Attorney General of Alabama, Montgomery, Alabama; Gregory D. Stumbo, Office of the Attorney General of the Commonwealth of Kentucky, Frankfort, Kentucky; Wayne Stenehjem, Attorney General of North Dakota, Bismarck, North Dakota; Paul G. Summers, Attorney General of Tennessee, Nashville, Tennessee; Steve Carter, Office of the Indiana Attorney General, Indianapolis, Indiana, on the brief), for State Amici Curiae Nebraska, Alabama, Kentucky, North Dakota, Tennessee, and Indiana. Nancy Elizabeth Olinger, Assistant Attorney General (Greg Abbott, Attorney General of Texas, Barry R. McBee, First Assistant Attorney General, Edward D. Burbach, Deputy Attorney General for Litigation, Karen W. Kornell, Chief, Natural Resources Division, on the brief), Austin, Texas, for Amicus Curiae Texas Commission on Environmental Quality. Russell S. Frye, FryeLaw PLLC, Washington, D.C., for Amicus Curiae American Petroleum Institute. Jonathan F. Lewis, Clean Air Task Force, Boston, Massachusetts, for Amici Curiae Healthlink, Kentucky Resources Council, New England Clean Water Action, The Ohio Environmental Council, and Ohio Valley Environmental Council. Lisa Heinzerling, Georgetown University Law Center, Washington, D.C., for Amicus Curiae OMB Watch. 4
    93. SOTOMAYOR, Circuit Judge: This is a case about fish and other aquatic organisms. Power plants and other industrial operations withdraw billions of gallons of water from the nation’s waterways each day to cool their facilities. The flow of water into these plants traps (or “impinges”) large aquatic organisms against grills or screens, which cover the intake structures, and draws (or “entrains”) small aquatic organisms into the cooling mechanism; the resulting impingement and entrainment from these operations kill or injure billions of aquatic organisms every year. Petitioners here challenge a rule promulgated by the Environmental Protection Agency (“the EPA” or “the Agency”) pursuant to section 316(b) of the Clean Water Act (“CWA” or “the Act”), 33 U.S.C. § 1326(b),1 that is intended to protect fish, shellfish, and other aquatic organisms from being harmed or killed by regulating “cooling water intake structures” at large, existing power- producing facilities. For the reasons that follow, we grant in part and deny in part the petitions for review, concluding that certain aspects of the EPA’s rule are based on a reasonable interpretation of the Act and supported by substantial evidence in the administrative record, but remanding several aspects of the rule because they are inadequately explained or inconsistent with the statute, or because the EPA failed to give adequate notice of its rulemaking. We also dismiss for lack of jurisdiction one aspect of the petitions because there is no final agency action to review. BACKGROUND Our decision in Riverkeeper, Inc. v. EPA, 358 F.3d 174 (2d. Cir. 2004) 1 We refer to statutory provisions mentioned in the text by their section in the CWA and provide in citations both the section of the Act and the parallel section of the United States Code. 5
    94. (“Riverkeeper I”), which addressed challenges to the EPA’s rule governing cooling water intake structures at new – as opposed to existing – facilities discusses at length the procedural and factual background of the rulemaking pursuant to section 316(b). We presume familiarity with Riverkeeper I and provide here only a brief overview of the statute and the various stages of the rulemaking. These consolidated petitions for review concern a final rule promulgated by the EPA regarding the water that large, existing power plants withdraw from rivers, lakes, and other waterways of the United States to cool their facilities. See 40 C.F.R. § 125.91(a). This cooling process requires power plants to extract billions of gallons of water per day from the nation’s waters, thereby impinging and entraining a huge number of aquatic organisms. Riverkeeper I, 358 F.3d at 181. Indeed, a single power plant can kill or injure billions of aquatic organisms in a single year. Id. Cognizant of this problem, Congress in 1972 amended the CWA, 33 U.S.C. §§ 1251-1387, to regulate cooling water intake structures. See Federal Water Pollution Control Act Amendments of 1972, Pub. L. No. 92-500, 86 Stat. 816 (1972). We have described Congress’s regulation of such structures as “something of an afterthought,” Riverkeeper I, 358 F.3d at 186 n.12, given that the directive appears in a section of the Act addressing the seemingly unrelated issue of thermal pollution, see CWA § 316(a), 33 U.S.C. § 1326(a). The Act, as amended, provides that “[a]ny standard established pursuant to section 1311 of this title [CWA section 301] or section 1316 of this title [CWA section 306] and applicable to a point source shall require that the location, design, construction, and capacity of cooling water intake structures reflect the best technology available for minimizing adverse environmental impact.” 6
    95. CWA § 316(b), 33 U.S.C. § 1326(b). The provisions of the Act cross-referenced in section 316(b) direct the EPA to issue rules regulating the discharge of pollution from existing point sources, CWA § 301, 33 U.S.C. § 1311, and new point sources, CWA § 306, 33 U.S.C. § 1316.2 As we noted in Riverkeeper I, “[w]hen the EPA established new source performance discharge standard[s] . . . it ought then to have regulated . . . intake structures . . . .” 358 F.3d at 185 (internal quotation marks omitted; emphasis in original). Put differently, section 316(b) required the EPA to promulgate regulations for cooling water intake structures at the same time that it established pollution discharge standards pursuant to sections 301 and 306. The EPA’s first attempt at regulation under section 316(b), however, was remanded by the Fourth Circuit in 1977 on procedural grounds, and years passed without the EPA issuing new rules. Id. at 181 (citing Appalachian Power Co. v. Train, 566 F.2d 451 (4th Cir. 1977)). Environmental groups ultimately sued the EPA and won a consent decree, pursuant to which the Agency established a timetable to issue rules pursuant to Section 316(b) in three “phases.” Id. & n.3. Phase I – addressed in Riverkeeper I – governs new facilities; Phase II – addressed here – covers large, existing power plants; and Phase III will regulate existing power plants not governed by Phase II, as well as other industrial facilities. See Riverkeeper, Inc. v. Whitman, No. 93 Civ. 0314 (AGS), 2001 WL 1505497, at *1 n.3 (S.D.N.Y. Nov. 27, 2001). Our interpretation of section 316(b) is informed by the two provisions it cross- references, CWA sections 301 and 306. Section 301 sets forth a framework under which 2 A “point source” is “any discernible, confined and discrete conveyance . . . from which pollutants are or may be discharged.” 33 U.S.C. § 1362(14). 7
    96. limitations on the discharge of pollutants from existing sources would become more stringent over time. CWA § 301(b), 33 U.S.C. § 1311(b); see Riverkeeper I, 358 F.3d at 185. Section 301(b)(1)(A) required the EPA, beginning in 1977, to set effluent limitations for existing sources based on “the best practicable control technology currently available,” or “BPT.” CWA § 301(b)(1)(A), 33 U.S.C. § 1311(b)(1)(A). By 1989, existing source effluent limitations were to be based on the more stringent “best available technology economically achievable,” or “BAT.” CWA § 301(b)(2)(A), 33 U.S.C. § 1311(b)(2)(A). Additionally, section 306 requires the EPA to establish “standards of performance” for the control of the discharge of pollutants from new sources based on “the best available demonstrated control technology,” a standard that “reflects the greatest degree of effluent reduction.” CWA § 306(a)(1), 33 U.S.C. § 1316(a)(1). In section 316(b), Congress established yet another standard to govern cooling water intake structures, which requires such structures to reflect the “best technology available for minimizing adverse environmental impact,” or “BTA.” CWA § 316(b), 33 U.S.C. § 1326(b). We noted in Riverkeeper I that “[a]lthough the EPA is permitted to consider a technology’s cost in determining whether it is ‘practicable,’ ‘economically achievable,’ or ‘available,’ it should give decreasing weight to expense as facilities have time to plan ahead to meet tougher restrictions.” 358 F.3d at 185 (citations omitted). Additionally, we observed that “[b]ecause section 316(b) refers to sections 301 and 306 but provides a different standard (‘best technology available for minimizing adverse environmental impact’ instead of, for example, ‘best available demonstrated control technology’) and does not explicitly provide that regulations pursuant to section 316(b) are subject to the requirements of sections 301 and 306, we think it is permissible for the EPA to look to those sections for guidance but to decide that not every statutory directive 8
    97. contained therein is applicable” to rulemaking under section 316(b). Id. at 187. With this general background in mind, we consider Phases I and II of the EPA’s rulemaking. I. The Phase I Rule On December 18, 2001, the EPA issued its first rule (“the Phase I Rule”) governing cooling water intake structures for new – as opposed to existing – facilities. Regulations Addressing Cooling Water Intake Structures for New Facilities; Final Rule, 66 Fed. Reg. 65,256 (Dec. 18, 2001) (codified at 40 C.F.R. pts. 9, 122-25). The Phase I Rule established a two-track approach to regulating cooling water intake systems at new facilities, under which a new facility could choose one of two “tracks” to comply with the statute. Track I created national intake capacity and velocity standards based on closed-cycle cooling technology,3 which the EPA deemed the best technology available for minimizing adverse environmental impacts. See Riverkeeper I, 358 F.3d at 182-83. Track II did not require the use of any specific technology so long as the facility “can show, in a demonstration study, ‘that the technologies employed will reduce the level of adverse environmental impact . . . to a comparable level to that which’ would be achieved applying Track I’s capacity and velocity requirements.” Id. at 183 (quoting 40 C.F.R. § 125.84(d)(1)). 3 As we noted in Riverkeeper I, Cooling water systems fall into three groups. “Once-through” systems take water in, use it to absorb heat, and return the water to its source at a higher temperature. “Closed-cycle” systems recirculate the water (after allowing it to cool off in a reservoir or tower before being reused) and add water to the system only to replace that which is lost through evaporation. Closed-cycle systems, therefore, withdraw far less water than once-through systems. Dry cooling systems . . . use air drafts to transfer heat, and, as their name implies, they use little or no water. 358 F.3d at 182 n.5 (internal citations omitted). 9
    98. Environmental and industry groups challenged certain aspects of the rule, including, inter alia, the part of the Track II procedure allowing power plants to comply with section 316(b) by undertaking so-called “restoration measures,” such as restocking the waterbody with fish, reclaiming abandoned mines to reduce drain-off, or removing barriers to fish migration, to maintain fish and shellfish in a waterbody at certain levels. In Riverkeeper I, we upheld most aspects of the Phase I Rule, but remanded the provisions relating to the Track II restoration option. We found that the restoration option was inconsistent with section 316(b)’s requirement that the EPA minimize adverse environmental impacts by regulating the “‘location, design, construction, and capacity of cooling water intake structures’” because this option has nothing to do with the location, design, construction, or capacity of such structures. Id. at 189 (quoting CWA § 316(b), 33 U.S.C. § 1326(b)). Given this, we held that the EPA had impermissibly exceeded its authority in allowing Phase I facilities to use these restoration measures to comply with regulations implementing the statute. Id. II. The Phase II Rule On July 9, 2004, the EPA issued a final rule, pursuant to the second phase of the consent decree (“the Phase II Rule” or “the Rule”), that governs cooling water intake structures at large, existing power plants. See Final Regulations to Establish Requirements for Cooling Water Intake Structures at Phase II Existing Facilities, 69 Fed. Reg. 41,576 (July 9, 2004) (codified at 40 C.F.R. pts. 9, 122-125). The Phase II Rule covers existing facilities that are “point sources” and that, as their primary activity, “both generate[] and transmit[] electric power, or generate[] electric power but sell[] it to another entity for transmission,” “use[] or propose[] to use cooling water intake structures with a total design intake flow of 50 million gallons per day (MGD) or 10
    99. more,” and “use[] at least 25 percent of water withdrawn exclusively for cooling purposes.” 40 C.F.R. § 125.91. Although we will discuss the specifics of the Rule with respect to each challenge, we provide here an overview of the Rule. The Phase II Rule sets forth five compliance alternatives. See 40 C.F.R. § 125.94(a). Section 125.94(a) requires that a facility select and implement one of the following “for establishing best technology available for minimizing adverse environmental impact”: (1)(i) You may demonstrate to the Director that you have reduced, or will reduce, your flow commensurate with a closed-cycle recirculating system. In this case, you are deemed to have met the applicable performance standards and will not be required to demonstrate further that your facility meets the impingement mortality and entrainment performance standards specified in paragraph (b) of this section. . . .; or (ii) You may demonstrate to the Director that you have reduced, or will reduce, your maximum through-screen design intake velocity to 0.5 ft/s or less. In this case, you are deemed to have met the impingement mortality performance standards and will not be required to demonstrate further that your facility meets the performance standards for impingement mortality specified in paragraph (b) of this section and you are not subject to the requirements in §§ 125.95, 125.96, 125.97, or 125.98 as they apply to impingement mortality. However, you are still subject to any applicable requirements for entrainment reduction . . . .[;] (2) You may demonstrate to the Director that your existing design and construction technologies, operational measures, and/or restoration measures meet the performance standards specified in paragraph (b) of this section and/or the restoration requirements in paragraph (c) of this section[;] (3) You may demonstrate to the Director that you have selected, and will install and properly operate and maintain, design and construction technologies, operational measures, and/or restoration measures that will, in combination with any existing design and construction technologies, operational measures, and/or restoration measures, meet the performance standards specified in paragraph (b) of this section and/or the restoration requirements in paragraph (c) of this section; (4) You may demonstrate to the Director that you have installed, or will install, and properly operate and maintain an approved design and construction technology in accordance with § 125.99(a) or (b); or 11
    100. (5) You may demonstrate to the Director that you have selected, installed, and are properly operating and maintaining, or will install and properly operate and maintain design and construction technologies, operational measures, and/or restoration measures that the Director has determined to be the best technology available to minimize adverse environmental impact for your facility in accordance with paragraphs (a)(5)(i) or (ii) of this section. . . . 40 C.F.R. § 125.94(a). The Phase II Rule does not require large, existing power plants to install closed- cycle cooling systems, although a facility with such a system (or one whose intake flow is commensurate with that of a closed-cycle system) will be considered in compliance with the Rule. 40 C.F.R. § 125.94(a)(1)(i). The Rule instead references national performance standards, discussed below, that “are based on consideration of a range of technologies that EPA has determined to be commercially available for the industries affected as a whole.” 69 Fed. Reg. at 41,598-99. And rather than limiting BTA to technologies based on closed-cycle cooling systems, the EPA designated a “suite” of technologies – including fine- and wide-mesh wedgewire screens, aquatic filter barrier systems, barrier nets, and fish return systems, among others, id. at 41,599; see also 40 C.F.R. § 125.99(a) – as BTA for large, existing power plants. Section 125.94(b) establishes national performance standards to be achieved through one of the compliance alternatives set forth in section 125.94(a). With respect to impingement mortality, it provides that facilities choosing “compliance alternatives in paragraphs (a)(2), (a)(3), or (a)(4) of this section . . . must reduce impingement mortality for all life stages of fish and shellfish by 80 to 95 percent from the calculation baseline.”4 Id. § 125.94(b)(1). With 4 The “calculation baseline” is “an estimate of impingement mortality and entrainment that would occur” at a specific site based on a number of quantitative assumptions regarding intake velocity and the location and design of the site’s intake structures. 40 C.F.R. § 125.93. The regulation also permits a site’s specific historical and current data on impingement and 12
    101. respect to entrainment, facilities that choose compliance alternatives in paragraphs (a)(1)(ii), (a)(2), (a)(3), or (a)(4) of this section . . . must also reduce entrainment of all life stages of fish and shellfish by 60 to 90 percent from the calculation baseline if: (i) [the] facility has a capacity utilization rate of 15 percent or greater, and (ii)(A) . . . uses cooling water withdrawn from a tidal river, estuary, ocean, or one of the Great Lakes; or (B) . . . uses cooling water withdrawn from a freshwater river or stream and the design intake flow of your cooling water intake structures is greater than five percent of the mean annual flow. Id. § 125.94(b)(2). Section 125.94(c) permits facilities to comply with the Rule by implementing restoration measures “in place of or as a supplement to installing design and control technologies and/or adopting operational measures that reduce impingement mortality and entrainment.” Id. § 125.94(c). In order to adopt restoration measures under the Rule, a facility must demonstrate that “meeting the applicable performance standards or site-specific requirements through the use of design and construction technologies and/or operational measures alone is less feasible, less cost-effective, or less environmentally desirable than meeting the standards . . . through the use of restoration measures.” Id. § 125.94(c)(1). Moreover, the restoration measures implemented by the facility must produce ecological benefits “at a level that is substantially similar” to what would be achieved by meeting the national performance standards of section 125.94(b). Id. § 125.94(c)(2). The compliance provision of section 125.94(a)(5) permits what is in effect a site- specific compliance alternative to the generally applicable performance standards in two entrainment to serve as a basis for this baseline. Id. 13
    102. circumstances. In the first circumstance (“the cost-cost compliance alternative” or “the cost-cost variance”), if a facility demonstrates that its compliance costs “would be significantly greater than the costs considered by the Administrator,” the permitting authority must make a site- specific determination of BTA that is “as close as practicable to the applicable performance standards . . . without resulting in costs that are significantly greater than the costs considered by the Administrator” in establishing those standards. 40 C.F.R. § 125.94(a)(5)(i). In the second circumstance (“the cost-benefit compliance alternative” or “the cost-benefit variance”), the permitting authority must make a site-specific determination of BTA that is “as close as practicable” to the national performance standards if a facility demonstrates that its compliance costs would be “significantly greater than the benefits of complying” with the performance standards at the facility. Id. § 125.94(a)(5)(ii). For those facilities installing technologies designated as BTA, section 125.94(d) allows the national performance standards set forth in section 125.94(b) to be satisfied by demonstrating compliance with a technology installation and operation plan (“TIOP”), which concerns, inter alia, a facility’s installation, operation and maintenance of BTA. As the Rule is enforced through the permitting process under the National Pollutant Discharge Elimination System (“NPDES”),5 section 125.94(d)(1) provides that a facility that uses one of the compliance methods other than closed-cycle cooling may request that compliance with the national performance standards during the first permit cycle be determined with respect to whether the facility has complied with the TIOP it submitted with its permit application. Section 5 The NPDES process is promulgated under CWA section 402(a)(1), 33 U.S.C. § 1342(a)(1). 14
    103. 125.94(d)(2) authorizes facilities to request that compliance during subsequent permit terms be determined based on whether a facility remains in compliance with its TIOP, and in accordance with any necessary revisions, “if applicable performance standards are not being met.” 40 C.F.R. § 125.94(d)(2). Finally, section 125.94(f) applies solely to nuclear power facilities. It provides that if a nuclear facility’s compliance with the Rule would conflict with a safety requirement established by the Nuclear Regulatory Commission, the EPA must make a site-specific determination of BTA that would not conflict with the Commission’s safety requirement. 40 C.F.R. § 125.94(f). For purposes of judicial review, the Phase II Rule was promulgated on July 23, 2004. See 69 Fed. Reg. at 41,576. Three sets of petitioners, discussed below, brought timely challenges to the Rule.6 DISCUSSION I. Standard of Review We have jurisdiction to review this Rule pursuant to CWA section 509(b)(1), 33 U.S.C. § 1369(b)(1). See Riverkeeper I, 358 F.3d at 183 (stating that the Phase I Rule is covered by the jurisdictional grant of § 1369(b)(1)). As we explained in Riverkeeper I, our substantive review is twofold. “First, we examine the regulation against the statute that contains the EPA’s charge.” Id. at 184. If Congress “has directly spoken to the precise question at issue” and its 6 The parties filed petitions for review here as well as in several of our sister circuits. The petitions were consolidated in the Ninth Circuit by order of the judicial panel on multi-district litigation pursuant to 28 U.S.C. §§ 1407 and 2112(a)(3). The Ninth Circuit thereafter transferred the case here pursuant to 28 U.S.C. § 2112(a)(5). 15
    104. intent is clear, we “must give effect to the unambiguously expressed intent of Congress.” Chevron U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 842-43 (1984). If, however, the statute is silent or ambiguous, we ask whether “the agency’s answer is based on a permissible construction of the statute.” Id. at 843. Second, if the agency has followed Congress’s unambiguously expressed intent or permissibly construed an ambiguous statute, “we measure the regulation against the record developed during the rulemaking, but we ‘hold unlawful’ the agency’s regulation only if it is ‘arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.’” Riverkeeper I, 358 F.3d at 184 (quoting 5 U.S.C. § 706(2)(A)). “Normally, we must deem arbitrary and capricious an agency rule where ‘the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.’” Waterkeeper Alliance, Inc. v. EPA, 399 F.3d 486, 498 (2d Cir. 2005) (quoting Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (internal quotation marks and citations omitted)). Finally, our review has a procedural dimension. The Administrative Procedure Act (“APA”) requires that notice of proposed rulemaking be published in the Federal Register, 5 U.S.C. § 553(b)(3), and that interested parties be allowed an opportunity to comment on proposed rules, id. § 553(c). Where an agency fails to comply with the APA’s notice and comment provisions, we remand to the agency for further proceedings. See Sprint Corp. v. FCC, 315 F.3d 369, 371 (D.C. Cir. 2003) (“Because the [agency] failed to provide adequate notice and 16
    105. opportunity to comment, we grant the petition and remand the case to the [agency].”). II. The Petitions for Review The state petitioners7 and the self-styled environmental petitioners8 challenge the Rule on similar grounds. Their petitions contain the following arguments: (1) the EPA exceeded its authority in rejecting closed-cycle cooling as BTA for existing facilities generally, and the Agency’s rejection of closed-cycle cooling as BTA for facilities on sensitive waterbodies is not entitled to deference because the decision was made at the direction of the Office of Management and Budget (“OMB”); (2) the EPA exceeded its authority by establishing ranges of acceptable performance rather than a single-numeric performance standard; (3) the CWA does not allow restoration measures as a means of compliance; (4) the EPA failed to give adequate notice that it would allow site-specific determinations of BTA based on cost-cost analysis, and the EPA impermissibly construed the statute to allow site-specific determinations of BTA based on cost- benefit analysis; (5) the provision allowing that compliance with the performance standards be determined by establishing compliance with a facility’s TIOP is unauthorized and violates the rulemaking requirement of notice and comment; and (6) the Agency has classified certain new constructions as “existing facilities” contrary to the definitions set forth in the Phase I Rule without providing adequate notice and opportunity for comment. 7 Rhode Island, Connecticut, Delaware, Massachusetts, New Jersey, and New York. 8 Riverkeeper, Inc., Natural Resources Defense Council, Waterkeeper Alliance, Soundkeeper, Inc., Scenic Hudson, Inc., Save the Bay–People for Narragansett Bay, Friends of Casco Bay, American Littoral Society, Delaware Riverkeeper Network, Hackensack Riverkeeper, Inc., New York/New Jersey Baykeeper, Santa Monica Baykeeper, San Diego Baykeeper, California Coastkeeper, Columbia Riverkeeper, Conservation Law Foundation, and Surfrider Foundation. 17
    106. Three groups of industry petitioners, which we will refer to collectively as the “industry petitioners” or individually as Entergy Corporation (“Entergy”), the Utility Water Act Group (“UWAG”),9 and PSEG Fossil LLC and PSEG Nuclear LLC (“PSEG”), advance various challenges to the Phase II Rule. Their challenges raise the following arguments: (1) section 316(b) of the CWA does not apply to existing facilities; (2) the Agency’s definition of “adverse environmental impact” is insufficiently supported by the record; (3) the EPA’s assumption of zero entrainment survival is insufficiently supported by the record; (4) the EPA improperly requires evaluation of qualitative non-use benefits in site-specific cost-benefit analyses; (5) the Agency failed to account for the Rule’s disproportionate impact on nuclear facilities; (6) the EPA gave inadequate notice of the independent-supplier provision; and (7) the Agency provided no notice of its post-rulemaking definition of “Great Lakes.” The industry petitioners also seek to preserve the right on this petition for review to raise new challenges to the Rule if we remand significant aspects of it. We consider first the challenges raised by the state and environmental petitioners and then will turn to the industry petitioners’ challenges. III. The State and Environmental Petitioners A. Determination of BTA Perhaps the most significant challenge to the Phase II Rule is the petitioners’ contention that the EPA exceeded its authority in rejecting closed-cycle cooling, and selecting instead the suite of technologies, as the “best technology available” as required by section 316(b), 33 U.S.C. § 1326(b), in large part because the Agency engaged in improper cost considerations. 9 UWAG petitions this court in conjunction with the Appalachian Power Company and the Illinois Energy Association. 18
    107. This challenge requires us at the outset to determine to what extent, if any, the EPA can consider cost when selecting “the best technology available for minimizing adverse environmental impact” under the statute. 1. Cost Analysis Pursuant to Sections 301 and 306 Section 316(b) does not itself set forth or cross-reference another statutory provision enumerating the specific factors that the EPA must consider in determining BTA. The statute, however, does make specific reference to CWA sections 301 and 306, which we have taken previously as “an invitation” to look to those sections for guidance in “discerning what factors Congress intended the EPA to consider in determining” BTA. Riverkeeper I, 358 F.3d at 186. We look to each of these statutes in turn. Section 301(b)(1)(A) established the BPT standard that governed the effluent limitations applicable to existing sources through 1989. Congress provided that, in determining BPT, the Agency could consider “the total cost of application of technology in relation to the effluent reduction benefits to be achieved from such application.” CWA § 304(b)(1)(B), 33 U.S.C. § 1314(b)(1)(B). As noted above, however, the CWA created standards that were to become increasingly stringent over time, and in 1989, the more lenient BPT standard for existing sources was replaced by the BAT standard of section 301(b)(2)(A), in which Congress provided that the EPA could consider only “the cost of achieving such effluent reduction.” CWA § 304(b)(2)(B), 33 U.S.C. § 1314(b)(2)(B). Notably omitted from the list of permissible factors to which the EPA could look in determining BAT was the cost of technology in relation to the benefits that technology could achieve. This shift from BPT to BAT fundamentally altered the way in which the EPA 19
    108. could factor cost into its CWA determinations. Indeed, in analyzing BTP and BAT, the Supreme Court stated that in “assessing BAT[,] total cost is no longer to be considered in comparison to effluent reduction benefits,” as it had been in assessing BPT. EPA v. Nat’l Crushed Stone Ass’n, 449 U.S. 64, 71 (1980). The Court indicated that the less stringent BPT standard had allowed for a “limited cost-benefit analysis” intended to “‘limit the application of technology only where the additional degree of effluent reduction is wholly out of proportion to the costs of achieving such marginal level of reduction.’” Id. at 71 n.10 (quoting Remarks of Senator Muskie reprinted in Legislative History of the Water Pollution Control Act Amendments of 1972 (Committee Print compiled for the Senate Committee on Public Works by the Library of Congress) Ser. No. 93-1, p. 170 (1973)). In determining BAT, by contrast, the EPA may consider cost as a factor to a limited degree, see id., but only as to whether the cost of a given technology could be reasonably borne by the industry and not the relation between that technology’s cost and the benefits it achieves, Riverkeeper I, 358 F.3d at 195. Section 306, which governs the effluent limitations that apply to new sources, provides that a “standard of performance” established by the EPA must reflect the “best available demonstrated control technology.” CWA § 306(a)(1), 33 U.S.C. § 1316(a)(1). In language identical to the text of § 304(b)(2)(B) governing BAT, Congress provided that in establishing standards of performance, the EPA “shall take into consideration the cost of achieving such effluent reduction,” CWA § 306(b)(1)(B), 33 U.S.C. § 1316(b)(1)(B), but did not require the EPA to conduct cost-benefit analysis. Nat’l Wildlife Fed’n v. EPA, 286 F.3d 554, 570 (D.C. Cir. 2002) (“[S]ection 306 requires that, when setting the [new source performance standards], the Administrator must take costs into consideration, but does not require that she conduct a 20
    109. cost-benefit analysis.”). Sections 301 and 306 of the CWA thus demonstrate that, after 1989, cost is a lesser, more ancillary consideration in determining what technology the EPA should require for compliance under those sections. The shift from the BPT standard to the more stringent BAT one clearly signaled Congress’s intent to move cost considerations under the CWA from a cost-benefit analysis to a cost-effectiveness one. We understand the difference between these two analyses to turn on the difference between means and ends. Cost-benefit analysis, like BPT, compares the costs and benefits of various ends, and chooses the end with the best net benefits. By contrast, cost- effectiveness considerations, like BAT, determine which means will be used to reach a specified level of benefit that has already been established.10 Given the above and considering the parallel language of sections 304(b)(2)(B) and 306(b)(1)(B), the reasoning of National Crushed Stone strongly suggests that cost-benefit analysis is no longer permitted under those sections of the CWA. 2. Cost Analysis Pursuant to Section 316(b) As already noted, section 316(b) does not itself set forth the factors that the Agency can consider in determining the “best technology available for minimizing adverse environmental impact.” The BTA standard of section 316(b), however, is linguistically similar to the BAT standard of section 301 and the standard that applies to new sources under 10 See, e.g., OMB Circular A-94, Guidelines and Discount Rates for Benefit-Cost Analysis of Federal Programs, Appendix A (1992) (defining “benefit-cost analysis” as “[a] systematic quantitative method of assessing the desirability of government projects or policies when it is important to take a long view of future effects and a broad view of possible side-effects” and “cost-effectiveness” as “[a] systematic quantitative method for comparing the costs of alternative means of achieving the same stream of benefits or a given objective”). 21
    110. section 306, and to the extent that cost-benefit analysis is precluded under those statutes, one might reasonably conclude that it is similarly not permitted under section 316(b). We conclude in any event that the language of section 316(b) itself plainly indicates that facilities must adopt the best technology available and that cost-benefit analysis cannot be justified in light of Congress’s directive. We stated in Riverkeeper I that the EPA can consider cost in establishing BTA, but only in a limited fashion and not as a primary consideration. Indeed, “[w]ith respect to costs, ‘the Administrator must inquire into the initial and annual costs of applying the technology and make an affirmative determination that those costs can be reasonably borne by the industry.’” Riverkeeper I, 358 F.3d at 195 (quoting Chem. Mfrs. Ass’n v. EPA, 870 F.2d 177, 262 (5th Cir. 1989)) (emphasis added). While the statutory language suggests that the EPA may consider costs in determining BTA, in that a technology that cannot not be reasonably borne by the industry is not “available” in any meaningful sense, cost-benefit analysis is not similarly supported by the language or purpose of the statute. Section 316(b) expressly requires a technology-driven result, cf. Natural Res. Def. Council, Inc. v. EPA, 822 F.2d 104, 123 (D.C. Cir. 1987) (“[T]he most salient characteristic of [the CWA’s] statutory scheme, articulated time and again by its architects and embedded in the statutory language, is that it is technology-forcing.”), not one driven by cost considerations or an assessment of the desirability of reducing adverse environmental impacts in light of the cost of doing so. A selection of BTA based on cost-benefit considerations is thus impermissibly cost-driven, but a selection based in part on cost-effectiveness considerations, while taking cost into account, remains technology-driven. The statute therefore precludes cost- benefit analysis because “Congress itself defined the basic relationship between costs and 22
    111. benefits.” Am. Textile Mfrs. Inst., Inc. v. Donovan, 452 U.S. 490, 509 (1981). Moreover, this conclusion is further supported by the fact that Congress in establishing BTA did not expressly permit the Agency to consider the relationship of a technology’s cost to the level of reduction of adverse environmental impact it produces.11 “When Congress has intended that an agency engage in cost-benefit analysis, it has clearly indicated such intent on the face of the statute.” Id. at 510. Given the above, the EPA may permissibly consider cost in two ways: (1) to determine what technology can be “reasonably borne” by the industry and (2) to engage in cost- effectiveness analysis in determining BTA. Thus, the EPA must first determine what is the most effective technology that may reasonably be borne by the industry. In making this initial determination, the most effective technology must be based not on the average Phase II facility but on the optimally best performing Phase II facilities, see, e.g., Kennecott v. United States EPA, 780 F.2d 445, 448 (4th Cir. 1985) (“In setting BAT, EPA uses not the average plant, but the 11 This conclusion accords with the analysis in Riverkeeper I that the EPA may consider cost in establishing BTA pursuant to section 316(b), but only in a limited way. In our discussion of the EPA’s choice of closed-cycle, rather than dry, cooling as BTA for Phase I facilities, see 358 F.3d at 194-95, 194 n.22, we noted that “dry cooling costs more than ten times as much per year as closed-cycle wet cooling,” but emphasized that “it is estimated to reduce water intake by only an additional 5 percent relative to once-through cooling.” Id. at 194 (internal footnotes omitted). We acknowledged that dry cooling is both much more effective and much more expensive than closed-cycle cooling in absolute terms, but stressed that, as compared to the baseline of once-through cooling systems, the marginal benefits of dry cooling were small: “it is undeniably relevant that that difference represents a relatively small improvement over closed-cycle cooling at a very significant cost.” Id. at 194 n.22. In dicta, we characterized this mode of analysis as “relevant” and stated that it “adds a useful perspective,” id., but did not treat it as the fulcrum of our analysis. Ultimately, we deferred to the EPA’s determination insofar as it was based on the grounds that dry cooling was too expensive for industry reasonably to bear and that dry cooling has negative environmental effects best left to the considered judgment of the Agency. Id. at 195-96. 23
    112. optimally operating plant, the pilot plant which acts as a beacon to show what is possible.”), although, of course, the EPA must still ascertain whether the industry as a whole can reasonably bear the cost of the adoption of the technology, bearing in mind the aspirational and technology- forcing character of the CWA. This technology constitutes the benchmark for performance. Once this determination has been made, the EPA may then consider other factors, including cost- effectiveness, to choose a less expensive technology that achieves essentially the same results as the benchmark.12 For example, assuming the EPA has determined that power plants governed by the Phase II Rule can reasonably bear the price of technology that saves between 100 - 105 fish, the EPA, given a choice between a technology that costs $100 to save 99 - 101 fish and one that costs $150 to save 100 - 103 fish (with all other considerations, like energy production or efficiency, being equal), could appropriately choose the cheaper technology on cost-effectiveness grounds. Cost-benefit analysis, however, is not permitted under the statute because, as noted, Congress has already specified the relationship between cost and benefits in requiring that the 12 We note that the EPA is by no means required to engage in cost-effectiveness analysis. Indeed, to require the Agency to conduct cost-effectiveness analysis would transform such analysis into a primary factor in choosing BTA, which clearly is contrary to the technology- forcing principle that animates the CWA. Equally important, we note that the Agency may also depart from this performance benchmark because of other permissible considerations aside from cost, for instance, energy efficiency or environmental impact. See Riverkeeper I, 358 F.3d at 195-96 (noting “the EPA was permitted to consider . . . energy efficiency in determining the ‘best technology available’” and could also factor in environmental impact). While the EPA has indicated throughout the record and its briefing before us that its determination of BTA here included such considerations as energy efficiency and production concerns, we nevertheless must remand, as we explain below, for further explanation from the Agency. Accordingly, we express no view on the merits of the EPA’s determination regarding the other factors it claims influenced its decision. 24
    113. technology designated by the EPA be the best available.13 Cf. Am. Textile Mfrs. Inst., 452 U.S. at 509-10. The Agency accordingly could not make the policy decision, in the face of Congress’s determination that facilities use the best technology available, that an economically feasible level of reduction of impingement mortality and entrainment is not desirable in light of its cost. Indeed, in the example above, the EPA could not choose the cheaper technology on cost considerations under section 316(b) if the EPA had first determined that the power plants could reasonably bear the cost of technology that could save at least 102 fish. We nevertheless acknowledge that the comparable technologies considered by the Agency need not be identically effective for the Agency to engage in cost-effectiveness analysis. Were that the case, all that would be required would be the simple determination of which among competing technologies that achieved the same degree of reduction of adverse environmental impacts is the cheapest. Instead, the specified level of benefit is more properly understood as a narrowly bounded range, within which the EPA may permissibly choose between two (or more) technologies that produce essentially the same benefits but have markedly different costs. With these considerations in mind, we turn to the Rule as promulgated. 3. The Determination of BTA under the Phase II Rule As noted previously, unlike the Phase I Rule, the Phase II Rule does not require facilities to reduce intake flow to a level commensurate with the intake of closed-cycle systems. Instead, the Rule requires facilities to meet the national performance standards associated with the suite of technologies the EPA identified as BTA. Petitioners’ challenge here has two 13 For this reason, we reject Entergy’s argument that the Rule is improper because its cost exceeds its benefits. 25
    114. components. First, the state petitioners contend that closed-cycle cooling is the best technology available and that the EPA has exceeded its authority by promulgating a rule that does not require closed-cycle cooling, or the use of technologies producing a commensurate reduction of water usage for existing facilities in the same manner as the Phase I Rule required for new facilities.14 Second, the environmental petitioners argue that the EPA improperly rejected closed-cycle cooling as BTA for the largest facilities on the most sensitive waterbodies at the direction of OMB because it sought to maximize net economic benefits rather than to minimize adverse environmental impact. They further argue that the BTA standard of section 316(b) requires a commitment of the maximum resources economically feasible to the goal of eliminating adverse environmental impacts and that the statute does not permit the EPA to select BTA on the basis of cost-benefit analysis. For the reasons that follow, we conclude that the statute’s “best technology available” standard permits cost-effectiveness considerations to influence the choice among technologies whose performance does not essentially differ from the performance of the best- performing technology whose cost the industry reasonably can bear, but that the statute does not permit the EPA to choose BTA on the basis of cost-benefit analysis. As we explain below, however, the record is unclear as to the basis for the EPA’s selection of the suite of technologies as BTA, and we therefore remand for clarification of the basis for the Agency’s decision and potentially for a reassessment of BTA. 14 While the state petitioners frame their argument as a challenge to the site-specific compliance alternatives, their brief presses the view that closed-cycle cooling is the best technology available and that the EPA exceeded its authority in failing to require Phase II facilities to adopt it. 26
    115. The EPA stated in the Rule’s preamble that the BTA standard should be interpreted as “best technology available commercially at an economically practicable cost,” and explained that “an important component of economic practicability” is “the relationship of costs to environmental benefits.” 69 Fed. Reg. at 41,604. The EPA further explained that this inquiry required that “there should be some reasonable relationship between the cost of cooling water intake structure control technology and the environmental benefits associated with its use.” Id. (emphasis added). The EPA took this “economically practicable” concept directly from the text of a floor speech of a single representative – the only specific reference to section 316(b) in the congressional debates. See Riverkeeper I, 358 F.3d at 186 n.12. We noted in Riverkeeper I that the “paucity” of legislative history “counsels against imputing much specific intent to Congress beyond the section’s words themselves.” Id. Moreover, we find the EPA’s interpretation of section 316(b) problematic because its construction significantly resembles the less stringent, and now obsolete, BPT standard of section 301(b)(1)(A). As noted earlier, in setting forth the factors for the EPA to consider in establishing BPT under section 301(b)(1)(A) and the more stringent BAT under section 301(b)(2)(A), Congress made only one distinction: while the Agency could consider the relationship between cost and benefits in establishing BPT, CWA § 304(b)(1)(B), 33 U.S.C. § 1314(b)(1)(B), it could consider cost insofar as it can be “reasonably borne” by the industry, but not the relationship between cost and benefits, in establishing BAT, CWA § 304(b)(2)(B), 33 U.S.C. § 1314(b)(2)(B). Riverkeeper I, 358 F.3d at 195. This difference in how the EPA can consider cost under section 304(b) in establishing BPT and BAT is directly mirrored by the most significant textual distinction 27
    116. between sections 301(b)(1)(A) and 301(b)(2)(A) – the requirement that a technology be “practicable” under only the less stringent BPT standard. Compare CWA § 301(b)(1)(A), 33 U.S.C. § 1311(b)(1)(A) (BPT is the “best practicable control technology”) with CWA § 301(b)(2)(A), 33 U.S.C. 1311(b)(2)(A) (BAT is the “best available technology”). The use of the word “practicable,” therefore, when coupled with the permissible cost considerations under section 304, signals that Congress intended the EPA to strike a balance between cost and benefits in determining BPT. But the word “practicable” is missing from the more stringent BAT standard, under which Congress prohibited the EPA from considering the relation of cost to benefits. This omission is thus significant. See Russello v. United States, 464 U.S. 16, 23 (1983) (“[W]here Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion.” (quotation marks and citation omitted; alteration in original)). Because Congress also omitted “practicable” from section 316(b), we are troubled by the Agency’s interpretation of the statute to require “practicability” analysis here and its implicit corollary that the Agency can undertake a cost-benefit analysis in establishing BTA under section 316(b). Our concern with the EPA's determination with section 316(b) is further deepened by the Agency’s rejection of closed-cycle cooling and selection of a suite of technologies as the basis for BTA for existing facilities because the suite of technologies were the most “cost effective” option. 69 Fed. Reg. at 41,667. The EPA explained this decision on several grounds. It first noted that it was rejecting closed-cycle cooling as BTA because of (1) “its generally high costs (due to conversions),” (2) “the fact that other technologies approach the performance of this 28
    117. option,” and (3) “concerns for energy impacts due to retrofitting existing facilities, and other considerations.” Id. at 41,605. The EPA emphasized that it selected BTA based on its determination that “a national requirement to retrofit existing systems is not the most cost- effective approach and at many existing facilities, retrofits may be impossible or not economically practicable.” Id. It further explained that its rejection of closed-cycle cooling as BTA was based on “total social costs” and “lack of cost-effectiveness,” as well as “concerns regarding potential energy impacts.” Id. at 41,606. Given the EPA’s discussion, noted above, of economically practicability, it is unclear whether the Agency improperly weighed the benefits and the costs of requiring closed- cycle cooling. Indeed, a comparison between the cost of closed-cycle cooling and the monetized benefits of this technology appears to have played some role in the EPA’s rejection of this option as BTA. In the preamble to the proposed Rule, for instance, the EPA examined whether to require closed-cycle cooling on specific large bodies of waters and stated that “the incremental costs of [this closed-cycle cooling] option relative to the proposed option ($413 million) significantly outweigh the incremental benefits ($146 million).” 67 Fed. Reg. at 17,158. Other record evidence on the EPA’s rejection of closed-cycle cooling as BTA is a terse EPA memorandum indicating that a requirement commensurate with closed-cycle cooling for facilities on sensitive waterbodies would cost three times as much as the option ultimately adopted by the EPA and reduce entrainment, at most, by 1.33 times that option. Given the above indications that the EPA engaged in cost-benefit analysis, we remand for the EPA to explain its conclusions. At the outset, it is difficult to discern from the record how the EPA determined that the cost of closed-cycle cooling could not be reasonably 29
    118. borne by the industry.15 Additionally, the EPA did not explain its statement that the suite of technologies “approach[es]” the performance of closed-cycle cooling. We see no adequate comparison in the Rule’s proposal, the final Rule or its preamble, or the EPA’s submissions to this Court of the effectiveness of closed-cycle cooling and the group of technologies whose effectiveness provided the basis for the Phase II Rule’s performance standards.16 In a technical 15 Indeed, at one point in the Rule’s proposal, the EPA mentions that requiring closed- cycle cooling at 539 existing power plants nationwide subject to the Phase II Rule would cost upwards of $2.26 billion and could close nine power plants (about 1.6% of all Phase II facilities). 67 Fed. Reg. at 17,155. There is little discussion about whether the industry could reasonably bear this burden. We note that in Riverkeeper I, it appears the EPA supplied the court with better data on this question, noting that of the 83 facilities subject to the Phase I Rule, dry cooling compliance costs would equal on average more than 4% of revenue for all 83 producers and more than 10% of the revenue at 12 facilities (nearly 15% of all Phase I facilities), presenting a possibly high barrier to entry for new facilities. See Riverkeeper I, 358 F.3d at 194. We have found no comparable data in the record, nor has the EPA cited any. 16 Some limited evidence, however, is available in the record in intelligible form. In the Rule’s proposal, the EPA noted that closed-cycle systems “generally reduce the water flow from 72 percent to 98 percent, thereby using only 2 percent to 28 percent of the water used by once- through systems.” Proposed Regulations to Establish Requirements for Cooling Water Intake Structures at Phase II Existing Facilities; Proposed Rule, 67 Fed. Reg. 17,122, 17,189 (Apr. 9, 2002). The Agency went on to state that “[i]t is generally assumed that this would result in a comparable reduction in impingement and entrainment,” id., indicating that closed-cycle systems reduce the adverse impacts of impingement and entrainment by 72 to 98 percent. In a technical development document (“TDD”) for the proposal, however, the EPA disaggregated these data, stating that closed-cycle cooling systems use 96 to 98 percent less fresh water and 70 to 96 percent less salt water than once-through systems. TDD for the Proposed § 316(b) Phase II Existing Facilities Rule 4-1; see also Riverkeeper I, 358 F.3d at 194 n.22 (“[C]losed-cycle wet cooling systems use 96 to 98 percent less fresh water (and 70 to 96 percent less salt water) than similarly situated once-through systems.” (citing 66 Fed. Reg. at 65,273)). The EPA’s analysis in its proposal suggests that the disaggregated data indicate that closed-cycle cooling would reduce impingement mortality and entrainment by 96 to 98 percent at facilities that use fresh water and by 70 to 96 percent at facilities that use salt water. A 72 to 98 percent reduction of impingement mortality and entrainment, and the corresponding disaggregated percent reductions for facilities using salt and fresh water, differs from the Phase II requirement that facilities reduce impingement mortality by 80 to 95 percent and entrainment by 60 to 90 percent. These differences seem potentially significant, especially in determining whether this suite of BTA technologies achieve essentially the same result as closed-cycle cooling, but are neither explained 30
    119. area of this sort, it is difficult for judges or interested parties to determine the propriety of the Agency’s action without a justification for the action supported by clearly identified substantial evidence whose import is explained. The record evidence alone here, which consists in large part of a voluminous database compilation of studies that assess the efficacy of various technologies at different locations, is oblique, complicated, and insufficient to permit us to determine what the EPA relied upon in reaching its conclusion. As the Supreme Court has emphasized, “[o]ur recognition of Congress’ need to vest administrative agencies with ample power to assist in the difficult task of governing a vast and complex industrial Nation carries with it the correlative responsibility of the agency to explain the rationale and factual basis for its decision, even though we show respect for the agency’s judgment in both.” Bowen v. Am. Hosp. Ass’n, 476 U.S. 610, 627 (1986) (plurality opinion). The EPA was required to explain its judgment and the basis for it. Because the EPA purported to base its decision in large part on cost-effectiveness considerations, it was required to identify and explain any evidence indicating a minimal performance difference between comparable technologies, but it did not do so here. It stated only that the performance of the technologies it identified as BTA “approach” the performance of closed-cycle cooling.17 nor adequately compared for purposes of our review here. For instance, there does not appear to be any discussion regarding the seemingly large differences in the rates of impingement and entrainment reduction between closed-cycle cooling and the Phase II national performance standards. 17 The Agency, however, did provide some indication of the relative costs of closed-cycle cooling and the suite of technologies identified as BTA. The EPA stated in the preamble to the Rule that the “total social cost” of closed-cycle cooling would be “$3.5 billion per year,” 69 Fed. Reg. at 41,605, and that the “final rule will have total annualized social (pre-tax) costs of $389 million,” id. at 41,650. 31
    120. We therefore find it impossible to judge whether the performance of these technologies is essentially the same as the performance of closed-cycle cooling, or whether they simply are cheaper per percentage point of reduction in entrainment and impingement mortality. That is, on the record before us, it is impossible to tell whether the EPA based its decision on permissible cost-effectiveness analysis or exceeded its authority by relying impermissibly upon a cost-benefit analysis. To the extent that the record does not indicate the EPA’s basis, however, its statement that “the relationship of costs to environmental benefits is an important component of economic practicability,” 69 Fed. Reg. at 41,604, indicates that cost-benefit analysis, under the cover of considerations of “practicability,” was central to the Agency’s decisionmaking. In short, the EPA’s failure to explain its decision frustrates effective judicial review. If the EPA construed the statute to permit cost-benefit analysis, its action was not “based on a permissible construction of the statute.” Chevron, 467 U.S. at 843. It may also be that the EPA misunderstood or misapplied cost-effectiveness analysis. If so, its decision was arbitrary and capricious because the Agency relied on factors Congress has not intended it to consider. See Waterkeeper Alliance, 399 F.3d at 498. Finally, the EPA may have simply failed either to perform the required analysis or to explain adequately a decision that was within its authority to make. We cannot opine on this subject, because we must consider only those justifications that the EPA offered at the time of the rulemaking. See SEC v. Chenery, 318 U.S. 80, 87-88 (1943). Moreover, while the EPA could rely on factors other than impingement and entrainment in establishing BTA, such as negative environmental impacts or concerns about energy production and efficiency, see Riverkeeper I, 358 F.3d at 195-96, we are unable to determine, on the record before us, whether the EPA gave paramount consideration to an improper factor in determining 32
    121. BTA. We therefore remand for clarification of the basis for the Agency’s action and possibly for a new determination of BTA.18 4. Performance Standards Expressed as Ranges The Phase II Rule establishes performance standards expressed as an 80 to 95 percent reduction in impingement mortality and a 60 to 90 percent reduction in entrainment, which existing power plants must achieve, subject to certain exceptions, in order to be considered in compliance with the Rule. 40 C.F.R. § 125.94(b)(1), (2). The environmental petitioners challenge the Rule’s “wide and indeterminate ranges” as failing to constitute “precise single-level limitations based on the best technology available for minimizing adverse environmental impact” and argue that these ranges are inconsistent with Congress’s intent that there be a national standard under section 316(b). We agree in part and, because the EPA in reconsidering its selection of BTA on remand may alter the suite of technologies it originally selected, thereby causing a coordinate alteration in the performance ranges, we provide some guidance to the EPA insofar as the petitioners’ challenge touches on the limits of the Agency’s authority. Although the EPA may, in the circumstances to be discussed, set performance standards as ranges, it must 18 As previously noted, the environmental petitioners, supported by amicus curiae OMB Watch, have challenged the EPA’s decision to reject closed-cycle cooling as BTA for 59 facilities on the most sensitive waterbodies in part because of OMB’s participation in the rulemaking process. They contend that the EPA’s action is not entitled to deference because the EPA was improperly influenced by OMB in promulgating this aspect of the Phase II Rule. The petitioners and the EPA have also sought either to strike from the record or to supplement the record with certain OMB review documents. Because we have granted the petitioners’ challenge to the EPA’s determination of BTA without considering OMB’s role in interagency review, we do not reach the petitioner’s arguments regarding OMB’s involvement in the rulemaking and deny the motions to strike and to supplement as moot. See Waterkeeper Alliance, 399 F.3d at 524 n.34 (denying petitioners’ motion to supplement the record with OMB review documents as moot where the Court granted the petitions without considering the OMB review documents). 33
    122. require facilities to minimize the adverse environmental impacts attributable to their cooling water intake structures to the best degree they can. The petitioners note that the EPA has found that certain screens and filter systems can reduce impingement mortality by up to 99 percent and that similar technologies can produce 80 to 90 percent reduction in entrainment. 69 Fed. Reg. at 41,599. They contend that the CWA therefore requires the EPA to set BTA standards reflecting these best performers, see Texas Oil & Gas Ass’n v. EPA, 161 F.3d 923, 928 (5th Cir. 1998) (“Congress intended these [BAT] limitations to be based on the performance of the single best-performing plant in an industrial field.” (citation and internal quotation marks omitted)), particularly given the EPA’s acknowledgment that “[t]he higher end of the range is a percent reduction that available data show many facilities can and have achieved with the available technologies upon which the performance standards are based.” 69 Fed. Reg. at 41,600. The petitioners emphasize that the Rule’s ranges impermissibly fail to require facilities even to attempt to achieve performance equal to the upper bound of the prescribed ranges. According to the EPA, section 316(b) does not require a single-numeric standard applicable to all Phase II existing facilities, and expressing the performance standards as ranges is necessary to account for the variables involved in reducing impingement mortality and entrainment under local conditions at particular facilities. The EPA contends that “[b]ecause the Phase II requirements are applied in a variety of settings and to existing facilities of different types and sizes, no single technology is most effective for all facilities subject to the Rule.” The Agency argues that the technologies do not provide a fixed level of performance at all facilities and that their performance is affected by the nature of the waterbody, facility intake 34
    123. requirements, climatic conditions, and the waterbody’s biology. The EPA argues also that the permit process requires facilities to reduce impingement mortality and entrainment commensurate with the efficacy of the installed technologies, which it claims ensures that the installed technologies will be maintained to ensure their utmost efficacy.19 The difficulty with the EPA’s arguments is that the Rule does not require facilities to choose technologies that produce the greatest reduction possible. Our decision in Riverkeeper I sheds some light on the parties’ arguments. In that case, we discussed the differences between the two tracks in the Phase I Rule: Track I set forth precise velocity and capacity requirements while Track II permitted compliance via technologies that would achieve at least 90 percent of the reduction in impingement mortality and entrainment that compliance with Track I would yield. See 358 F.3d at 182-83. The petitioners in that case challenged the Track II provision on the ground that it deviated from the statutory requirement that the EPA establish a single level of performance applicable to all facilities. Id. at 187. The EPA argued that Tracks I and II reflected the same standard and that 10 percent is an acceptable margin of error given that measurements of reduction of impingement mortality and entrainment are necessarily inexact and depend upon natural fluctuations in animal populations and sampling errors. Id. at 188. In assessing the parties’ arguments, we stated that “the EPA, consistent with Congress’s intention that there be a national standard governing the discharge of pollutants, must 19 The specific provision requires that when a facility’s permit expires, any reissued permit must contain a requirement that “the facility . . . reduce impingement mortality and entrainment . . . commensurate with the efficacy at the facility of the installed design and construction technologies, operational measures, and/or restoration measures.” 40 C.F.R. § 125.98(b)(1)(iii). 35
    124. promulgate precise effluent limitations under sections 301 and 306 . . . .” Id. (emphasis added). We went on to note, however, that while pollutant concentration and the velocity and volume of water withdrawn can be measured accurately, impingement mortality and entrainment “cannot always be measured directly and with mathematical precision.” Id. at 189. We concluded that the EPA acted reasonably in specifying “how much ambiguity it is willing to tolerate in measuring compliance and what it considers a reasonable margin of error in comparing the performance of different technologies.” Id. In short, we acknowledged that the Track II performance requirements, unlike the Track I requirements, could not be measured precisely and that it was therefore reasonable to consider a margin of error in comparing performance under the two standards. This case is not entirely similar to Riverkeeper I because of the rationales that animate the EPA’s creation of the performance ranges in Phases I and II. The Phase II Rule generally require facilities to reduce impingement mortality and entrainment by the specified percent ranges from the calculation baseline. 40 C.F.R. § 125.94(b). These ranges, as explained by the EPA, are based on the reductions achievable by using various technologies. See 69 Fed. Reg. at 41,599. The EPA explained that it expressed the performance standards “in the form of ranges rather than a single performance benchmark because of the uncertainty inherent in predicting the efficacy of any one of these technologies.” Id. at 41,600. It stated further that the lower end of the range is the percent reduction it “expects all facilities could eventually achieve if they were to implement and optimize available design and construction technologies and operational measures on which the performance standards are based” and that the higher end of the range “is a percent reduction that available data show many facilities can and have achieved 36
    125. with the available technologies upon which the performance standards are based.” Id. Unlike Riverkeeper I, therefore, a margin of error from a relatively precise benchmark that is tolerable given measurement difficulties is not at issue here. Instead, the performance standards reflect the range of performance associated with various technologies identified as BTA. That performance, in turn, depends in part on local conditions and natural fluctuations. Id. Record evidence supports the EPA’s conclusion that the percent reduction of impingement mortality and entrainment is not completely within the control of a facility and therefore may not be precisely achieved by a facility. See TDD for the Final § 316(b) Phase II Existing Facilities Rule 4-3. Reducing these adverse environmental impacts is not as easily measured and controlled as are the discharge of pollutants and the capacity and flow rate of water intake.20 We therefore acknowledge that in many cases it may be difficult, as a practical matter, for the EPA or other permitting authority to predict which plants will be able to achieve the upper, as opposed to the lower, end of the ranges. This uncertainty, however, does not justify a rule that permits even those facilities that could achieve the upper end of a range to be deemed in compliance if they reach only the lower end, particularly when the EPA has acknowledged that many facilities “can and have” achieved reductions at the high end of the range. 69 Fed. Reg. at 41,600. Congress’s use of the superlative “best” in the statute cannot be read to mean that a facility that achieves the lower end of the ranges, but could do better, has complied with the law. 20 Nothing in Hooker Chems. & Plastics Corp. v. Train, 537 F.2d 620 (2d Cir. 1976), suggests the contrary. While we did announce in that case that performance ranges did not comport with the effluent limitations established by the CWA, id. at 630, the discharge limitations at issue there were easily measured, unlike the reductions in impingement and entrainment at issue in both this case and Riverkeeper I, which “cannot always be measured directly and with mathematical precision,” Riverkeeper I, 358 F.3d at 189. 37
    126. The statutory directive requiring facilities to adopt the best technology cannot be construed to permit a facility to take measures that produce second-best results, see Chevron, 467 U.S. at 843, especially given the technology-forcing imperative behind the Act, Natural Res. Def. Council, 822 F.2d at 123. Insofar as the EPA establishes performance standards instead of requiring facilities to adopt particular technologies, it must require facilities to choose the technology that permits them to achieve as much reduction of adverse environmental impacts as is technologically possible.21 For this reason, the EPA on remand should address these concerns if in its BTA determination, it retains performance ranges. B. Restoration Measures The Phase II Rule allows a facility to meet the national performance standards set 21 Although it may be difficult to know ex ante which plants can achieve the upper end of the range, this is at least in part because the technologies preferred by the EPA are somewhat new and untested. In future permitting cycles, permitting authorities will be likely to have a clearer notion of which facilities can feasibly achieve the upper end of the range. Finally, the upper end of the range established by the EPA, should it retain this provision on remand, should not be set at a level that many facilities “have achieved” with the installation of one or more of the technologies determined to be BTA but, as we noted earlier, at the best possible level of impingement and entrainment reduction the EPA determines these technologies can achieve. See, e.g., Kennecott, 780 F.2d at 448 (“In setting BAT, EPA uses not the average plant, but the optimally operating plant, the pilot plant which acts as a beacon to show what is possible.”). If, at a particular Phase II facility, the adoption of BTA technologies can achieve a 95% reduction in entrainment and impingement, it is unclear why, under our jurisprudence and the clear dictates of the CWA, the EPA could establish a performance standard that has placed the ceiling at the 90% threshold which “many” Phase II facilities “can and have” achieved with the same technology. See Am. Iron & Steel Institute v. EPA, 526 F.2d 1027, 1051 (3d Cir. 1975) (“It will be sufficient, for the purpose of setting the level of control under available technology, that there be one operating facility which demonstrates that the level can be achieved.”) (internal quotation marks and citations omitted). This would not require every Phase II facility to meet the upper end of the ranges, but only that each Phase II facility achieve the highest reduction it can with the installation of technologies determined by the EPA to be BTA. The performance ranges, if retained on remand, should accordingly reflect this understanding. 38
    127. forth in 40 C.F.R. § 125.94(b) through the use of restoration measures such as restocking fish killed by a cooling water system and improving the habitat surrounding the intake structure in order, as the EPA explains, “to provide additional flexibility to facilities in complying with the rule by eliminating or significantly offsetting the adverse environmental impact caused by the operation of a cooling water intake structure.” 69 Fed. Reg. at 41,609; 40 C.F.R. § 125.94(c). The state and environmental petitioners contend that the EPA exceeded its authority by allowing compliance with section 316(b) through restoration measures because Riverkeeper I held that the statute’s meaning is plain and that restoration measures cannot substitute for the “best technology available for minimizing adverse environmental impact” in cooling water intake structures. The EPA contends that its interpretation of the statute to permit restoration measures as a means of compliance is entitled to deference because it defined certain statutory terms in the Phase II Rule that it had not defined in the Phase I Rule. The EPA also relies on the Supreme Court’s holding in National Cable & Telecommunications Association v. Brand X Internet Services, 545 U.S. 967, 125 S. Ct. 2688 (2005), and our statement in Riverkeeper I limiting the decision’s reach to the Phase I Rule, to argue that our prior interpretation of the statute does not trump the Agency’s construction. We agree with the petitioners that Riverkeeper I held that the Agency’s decision to permit restoration measures in the Phase I Rule was not “based on a permissible construction of the statute,” Chevron, 467 U.S. at 843, and that this holding applies equally here. We began Riverkeeper I by noting that we were remanding the Phase I restoration provision because it “contradicts Congress’s clearly expressed intent.” 358 F.3d at 181. We went on to state that “however beneficial to the environment, [restoration measures] have nothing to do with the location, the design, the construction, or the capacity of cooling water intake 39
    128. structures, because they are unrelated to the structures themselves.” Id. at 189. “Restoration measures correct for the adverse environmental impacts of impingement and entrainment,” we noted, but “they do not minimize those impacts in the first place.” Id. (emphasis added). For this reason and others, we concluded in Riverkeeper I that the EPA had exceeded its authority in promulgating the Phase I Rule by allowing compliance with section 316(b) through restoration measures because this Rule was “plainly inconsistent” with the statute’s text and Congress’s intent. Id. at 189, 191. The EPA’s argument that Riverkeeper I is not binding on this issue here has three components. First, the EPA contends that our rejection of the restoration measures at issue in Riverkeeper I did not turn on the statute’s text, but instead was based on various other indicators of Congressional intent. The Agency makes much of Riverkeeper I’s brief discussion of Congress’s rejection of a proposed amendment to section 316(b) that would have explicitly allowed restoration measures and of the EPA’s support of that amendment because in its opinion, the existing language did not authorize restoration measures. Id. at 190-91. Second, the EPA argues that its interpretation of section 316(b) in the Phase II Rule is entitled to deference because the Rule defined certain statutory terms it had not defined in the previous rulemaking phase and that its reasonable interpretation of these terms is entitled to deference. Specifically, the Agency “defined” three statutory terms in the preamble: it (1) read the phrase “minimiz[e] adverse environmental impact” to let facilities “minimize adverse environmental impact by reducing impingement and entrainment, or to minimize adverse environmental impact by compensating for those impacts after the fact,” 69 Fed. Reg. at 41,628; (2) interpreted “reflect” to authorize it to consider the full range of technologies, including restoration measures, that 40
    129. minimize adverse environmental impact; and (3) viewed “restoration measures as part of the ‘design’ of a cooling water intake structure,” and “one of several technologies that may be employed . . . to minimize adverse environmental impact,” 69 Fed. Reg. at 41,637. Finally, the EPA relies on our statement in Riverkeeper I that our ruling on the Phase I Rule was not meant to “predetermine the factors and standard[s] applicable to Phases II and III of the rulemaking,” 358 F.3d at 186 n.13, and on the Supreme Court’s holding in Brand X that “[a] court’s prior judicial construction of a statute trumps an agency construction otherwise entitled to Chevron deference only if the prior court decision holds that its construction follows from the unambiguous terms of the statute and thus leaves no room for agency discretion.” 545 U.S. at —, 125 S. Ct. at 2700. We reject each of the EPA’s contentions. First, our primary conclusion in Riverkeeper I was that restoration measures are “plainly inconsistent” with the statute’s text, 358 F.3d at 189, and our statements regarding the legislative history of a proposed amendment, which we offered as ancillary, but not dispositive, support for our construction of the statute, in no way diminish the force of our conclusion that Congress unambiguously expressed its intent in the statute. See Chevron, 467 U.S. at 842-43 (“If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress.”). Second, as to the EPA’s claim that its construction of the statute is entitled to deference because it has now interpreted certain statutory terms, our holding in Riverkeeper I was and remains clear: restoration measures contradict the unambiguous language of section 316(b). The EPA’s promulgation of the Phase II Rule obviously did nothing to alter the text of section 316(b), and the Agency cannot create ambiguity where none otherwise exists by defining statutory terms contrary to their plain meaning. Finally, as the foregoing analysis 41
    130. suggests, our decision in Riverkeeper I foreclosed the EPA from interpreting section 316(b) in the Phase II Rule to permit restoration measures as a means of complying with the statute, and, therefore, nothing in Brand X undermines the precedential value of our prior holding. Our statement in Riverkeeper I that we did not “mean to predetermine the factors and standard applicable to Phases II and III of the rulemaking,” was made in the narrow context of identifying “one reasonable reading” of particular statutory language relating to the standard for new and existing sources, not the restoration measures. 358 F.3d at 186 n.13. Where we held that the statutory language is unambiguous, Riverkeeper I is binding. Even assuming arguendo that we did not consider ourselves bound by Riverkeeper I, we are persuaded by its reasoning as applied here. Restoration measures are not part of the location, design, construction, or capacity of cooling water intake structures, Riverkeeper I, 358 F.3d at 189, and a rule permitting compliance with the statute through restoration measures allows facilities to avoid adopting any cooling water intake structure technology at all, in contravention of the Act’s clear language as well as its technology-forcing principle. As we noted in Riverkeeper I, restoration measures substitute after-the-fact compensation for adverse environmental impacts that have already occurred for the minimization of those impacts in the first instance. Id. The Agency’s attempt to define the word “minimize” to include “compensati[on] . . . after the fact,” 69 Fed. Reg. at 41,628, is simply inconsistent with that word’s dictionary definition: “to reduce to the smallest possible extent,” Webster’s Third New Int’l Dictionary 1438 (1986). Accordingly, the EPA impermissibly construed the statute by allowing compliance with section 316(b) via restoration measures, and we remand that aspect of the Rule. 42
    131. C. Site-Specific Compliance Alternatives As noted earlier, the Phase II Rule includes two site-specific compliance alternatives or variances from the generally applicable requirements. The cost-cost alternative authorizes a site-specific determination that “data specific to [a] facility demonstrate that the costs of compliance under . . . this section would be significantly greater than the costs considered by the Administrator . . . in establishing the applicable performance standards,” 40 C.F.R. § 125.94(a)(5)(i), while the cost-benefit alternative authorizes a site-specific determination that “data specific to [a] facility demonstrate that the costs of compliance under . . . this section would be significantly greater than the benefits of complying with the applicable performance standards.” Id. § 125.94(a)(5)(ii). If a facility makes either showing, the permitting authority “must make a site-specific determination of the best technology available” and impose “site-specific alternative requirements” that are “as close as practicable to the applicable performance standards.” Id. § 125.94(a)(5)(i), (ii). Petitioners challenge the cost-cost compliance alternative because, inter alia, they claim as a threshold matter that the Agency failed to comply with the APA’s notice and comment requirements by disclosing cost data for specific facilities that would be used in determining whether a facility qualifies for the cost-cost compliance alternative only at the time the final Rule was issued. Petitioners also challenge the cost-benefit compliance alternative on two substantive grounds. They contend that this alternative (1) impermissibly allows compliance with the statute to be based on cost-benefit analysis and (2) is analogous to a water-quality standard, which the Act permits only for thermal pollution. CWA § 316(a), 33 U.S.C. § 1326(a). We address each 43
    132. of these arguments in turn.22 1. Cost-Cost Compliance Alternative As already noted, a variance may be available to a facility pursuant to 40 C.F.R. § 125.94(a)(5)(i) if the facility’s compliance costs would be “significantly greater than” the costs considered by the Agency “in establishing the applicable performance standards.” This variance requires a calculation of compliance costs based on the suite of BTA technologies that the EPA has identified and promulgated in the final Rule. 69 Fed. Reg. at 41,644-46. We remand this provision because (1) the EPA did not give interested parties the requisite notice and opportunity to challenge the variance by failing to identify cost data for actual, named facilities, as opposed to model facilities, until after the notice and comment period had ended, Sprint Corp., 315 F.3d at 22 Section 316(b) does not specifically authorize the Agency to allow variances from generally applicable requirements. In Riverkeeper I, we accepted the argument that the EPA can allow variances even in the absence of explicit statutory authority. 358 F.3d at 193. Although the statutes to which section 316(b) refers contained variance provisions, we did not read them to require that section 316(b) permit (or by its silence prohibit) variances. Instead, we stated that [s]ection 316(b)’s silence with respect to variances does not . . . equal an unambiguous prohibition. In the absence of such a statutory bar, we think, consistent with precedent, that it is reasonable for the EPA to allow variances from regulations promulgated pursuant to section 316(b), for a regulatory system which allows flexibility, and a lessening of firm proscriptions in a proper case, can lend strength to the system as a whole. Id. at 193 (citation and internal quotation marks omitted). Riverkeeper I thus rejected the view that the EPA’s authority to grant a variance from the generally applicable requirements of a rule promulgated pursuant to section 316(b) derives from the statutory variance provisions in other sections of the Act. Instead, Riverkeeper I found that the variance provision would not be authorized if it left alternative requirements to the Agency’s “unfettered discretion” but would be authorized if it “guide[d] the permitting authority to consider appropriate factors and allow[ed] relaxation of the Rule’s uniform technology requirements only insofar as necessary to account for unusual circumstances not considered by the Agency during its rulemaking.” Id. at 193-94 (internal quotation marks omitted). 44
    133. 371, and (2) the variance is expressly premised on the validity of the BTA determination,23 which itself has been remanded for further explanation, see, e.g., Solite Corp. v. U.S. EPA, 952 F.2d 473, 494-95 (D.C. Cir. 1991) (remanding rule where the underlying grounds for its promulgation had been remanded to the EPA for procedural defects); cf. Chenery, 318 U.S. at 87-88 (a rule may only be upheld on the grounds that the agency proffers). In the Rule’s proposal, the EPA indicated that it had estimated compliance costs for 539 “model plants” based on factors such as “fuel source, mode of electricity generation, existing intake technologies, waterbody type, geographic location, and intake flow.” 67 Fed. Reg. at 17,144. An accompanying technical development document set forth the Agency’s cost calculation methodology for these model plants and listed the compliance cost estimates for each of the 539 model plants. The proposal indicated that a facility must “determine which model plant [it] most closely” resembles in order to identify the costs considered by the Agency in establishing the national performance standards. See id. The EPA subsequently published in the Federal Register a so-called Notice of Data Availability (“NODA”) in which it explained that it had changed its methodology for estimating the model plants’ compliance costs. Proposed Regulations to Establish Requirements for Cooling Water Intake Structures at Phase II Existing Facilities; Notice of Data Availability; Proposed Rule, 68 Fed. Reg. 13,522, 13,527 (Mar. 19, 2003). Accompanying documents explained in greater detail the costing methodology and cost data underlying the revised approach. The revised proposal, however, did not depart from the 23 This variance requires a comparison between the actual costs of compliance and those contemplated by EPA under the Rule. Because the Agency has calculated the costs it believes specific facilities will incur in adopting the appropriate BTA technologies (as currently defined) and then promulgated these costs in the final Rule, any change in the selection of BTA on remand will necessarily alter these costs. 45
    134. “model plant” approach. The final Rule, by contrast, assigned cost estimates to specific, named facilities rather than model facilities. 69 Fed. Reg. at 41,670-82. The Agency explained in the preamble to the final Rule that the EPA will adjust facility-specific costs pursuant to a multiple- step calculation formula to arrive at a final estimated cost the EPA considers a comparison for purposes of the cost-cost variance. Id. at 41,644-47. The EPA acknowledges that it did not disclose in the proposal or the NODA specific facility names in connection with cost data and explains that it failed to do so because it needed to protect certain confidential business information (“CBI”) and had not developed during the proposal stage a means to protect that information while still providing cost data to the public. We accept the EPA’s argument that masking the facility names did not prevent interested parties from commenting on the methodology and general cost data underlying the EPA’s approach because the NODA explained the costing methodology and because the general cost data, while not identified by the Agency as relating to actual, specific facilities, was made available to interested parties. Nat’l Wildlife Fed., 286 F.3d at 564-65 (holding that the EPA cannot be faulted for lack of notice in not releasing CBI data). We are persuaded, however, that the release of information and request for comments on the EPA’s new approach to developing compliance cost modules via the NODA did not afford adequate notice of the costs associated with specific facilities promulgated in the final Rule. We have previously stated that “[n]otice is said not only to improve the quality of rulemaking through exposure of a proposed rule to comment, but also to provide fairness to interested parties and to enhance judicial review by the development of a record through the commentary process.” Nat’l Black Media Coalition v. FCC, 791 F.2d 1016, 1022 (2d Cir. 1986). 46
    135. “While a final rule need not be an exact replica of the rule proposed in the Notice, the final rule must be a ‘logical outgrowth’ of the rule proposed.” Id. “The test that has been set forth is whether the agency’s notice would fairly apprise interested persons of the subjects and issues” of the rulemaking. Id. (citation and internal quotation marks omitted).24 Agencies accordingly are not permitted “to use the rulemaking process to pull a surprise switcheroo.” Envtl. Integrity Project, 425 F.3d at 996. Here, only the final Rule identified facilities by name in estimating compliance costs. Interested parties therefore could not comment on the basis for particular facilities’ cost figures that the EPA established. This is problematic because the availability of a variance turns on the relationship between the costs estimated in the Rule and those that a specific facility establishes in a permit proceeding. The EPA focuses on the notice it gave of its intended methodology for calculating the costs the Agency considered, but ignores the overriding importance of the cost estimates for a particular facility in determining whether a site-specific cost-cost variance is appropriate. Thus, the EPA should have afforded notice and an opportunity to challenge the cost estimates for specific facilities and not simply an opportunity to comment 24 The D.C. Circuit has phrased the test somewhat differently, stating that “[w]hether the ‘logical outgrowth’ test is satisfied depends on whether the affected party ‘should have anticipated’ the agency’s final course in light of the initial notice.” Covad Comms. Co. v. FCC, 450 F.3d 528, 548 (D.C. Cir. 2006) (citation omitted). It has also indicated that “[t]he ‘logical outgrowth’ doctrine does not extend to a final rule that finds no roots in the agency’s proposal because something is not a logical outgrowth of nothing, nor does it apply where interested parties would have had to divine the agency’s unspoken thoughts because the final rule was surprisingly distant from the Agency’s proposal.” Envtl. Integrity Project v. EPA, 425 F.3d 992, 996 (D.C. Cir. 2005) (citations, internal quotation marks, and alteration omitted). 47
    136. on the EPA’s methodology and general cost data.25 We remand this variance for inadequate notice and because of our remand of the BTA determination. 2. Cost-Benefit Compliance Alternative26 If a facility requests that it be permitted to demonstrate compliance with the Phase II Rule through the site-specific cost-benefit provision of 40 C.F.R. § 125.94(a)(5)(ii), the facility must submit with its application a Comprehensive Cost Evaluation Study, Benefits Valuation Study, and Site Specific Technology Plan. 40 C.F.R. § 125.95(b)(6). As part of the Benefits Valuation Study, the facility must indicate the monetized value of commercial, recreational, and ecological benefits of compliance with the generally applicable national performance standards 25 Because we remand on this procedural ground, we do not consider the provision on the merits. We presume that after comment on remand, however, the EPA will reevaluate the merits of the cost-cost variance provision, particularly in light of any reevaluation of BTA in which the Agency may engage. While we do not decide the issue here, we note our discomfort with the “significantly greater than” standard of 40 C.F.R. § 125.94(a)(5)(i), given the historical applicability of a “wholly disproportionate to” standard and the use of the latter standard in the Phase I Rule. We have not found entirely persuasive the EPA’s position that [t]his difference in standards for new and existing facilities is based on (1) the greater flexibility available to new facilities for selecting the location of their intakes and installing technologies at lower costs relative to the costs associated with retrofitting existing facilities and (2) the desire to avoid economically impracticable impacts on energy prices, production costs, and energy production that could occur if large numbers of Phase II existing facilities incurred costs that were more than “significantly greater” than but not “wholly out of proportion” to the costs in EPA’s record. 68 Fed. Reg. at 13,541. The EPA would presumably consider each of these two factors in establishing BTA for existing facilities, and need not further consider them in determining whether a particular facility warrants a variance from the generally applicable BTA. Because cost is not supposed to be a paramount consideration in determining BTA, see Riverkeeper I, 358 F.3d at 185, the “significantly greater than” standard poses substantial concerns. 26 In contrast to the cost-cost variance, the cost-benefit compliance alternative will not change on remand for the reconsideration of BTA. We thus reach the merits here. 48
    137. as well as a qualitative assessment of any so-called “non-use” benefits that cannot be monetized. 40 C.F.R. § 125.95(b)(6)(ii)(A), (E). Ultimately, the facility must demonstrate that its compliance costs are “significantly greater than” the benefits of compliance. The petitioners contend that this alternative impermissibly focuses on cost-benefit considerations, contrary to Congress’s directive, and is analogous to the kind of water-quality-based standard we found to be inconsistent with the statute in Riverkeeper I. 358 F.3d at 190. For both reasons, we are persuaded that the EPA exceeded its authority in permitting site-specific cost-benefit variances. In light of this conclusion, we do not reach the industry petitioners’ claim that the provision impermissibly requires consideration of qualitative non-use benefits in the cost-benefit analysis. As we discussed previously in analyzing the EPA’s determination of BTA, cost- benefit analysis is not consistent with the requirement of § 316(b) that cooling water intake structures “reflect the best technology available for minimizing adverse environmental impact.” Indeed, the statutory language requires that the EPA’s selection of BTA be driven by technology, not cost. The Agency is therefore precluded from undertaking such cost-benefit analysis because the BTA standard represents Congress’s conclusion that the costs imposed on industry in adopting the best cooling water intake structure technology available (i.e., the best-performing technology that can be reasonably borne by the industry) are worth the benefits in reducing adverse environmental impacts. Cf. Am. Textile Mfrs. Inst., 452 U.S. at 509 (noting that where Congress has defined the basic relationship between costs and benefits, a regulatory standard that strikes a different balance is inconsistent with the statute). Just as the Agency cannot determine BTA on the basis of cost-benefit analysis, it cannot authorize site-specific determinations of BTA based on cost-benefit analysis. 49
    138. The cost-benefit variance also impermissibly authorizes the EPA to consider the degraded quality of waterways in selecting a site-specific BTA. We stated in Riverkeeper I that in enacting the CWA, Congress rejected regulation by reference to water quality standards. 358 F.3d at 189-90. Before 1972, Congress “regulated point sources based on their effect on the surrounding water and allowed sources to discharge pollutants provided the discharge did not cause water quality to dip below an acceptable level.” Id. at 189. Congress changed its approach in 1972, in part because a plaintiff attempting to prove a violation of the law faced a nearly impossible burden of showing that a particular polluter had caused the water quality to dip below the regulatory standards. Id. at 189-90. The Act now regulates discharges from point sources rather than water quality. We thus concluded in Riverkeeper I that water-quality standards cannot be considered under section 316(b). Id. at 190. Of course, “water quality” in the context of the Act is generally understood to refer to pollutant concentration. As we noted in Riverkeeper I, however, for purposes of section 316(b), which regulates water intake rather than the discharge of pollutants, water quality is “measured by wildlife levels.” Id. at 189. This analysis in Riverkeeper I is, thus, equally applicable here.27 27 As the petitioners note, section 316(a) permits consideration of the quality of the receiving water for purposes of granting variances with respect to the rules concerning thermal pollution. Section 316(a) provides that the Administrator may issue a variance to the rules governing “thermal discharges” to a facility that establishes that those rules “require effluent limitations more stringent than necessary to assure the pro[t]ection and propagation of a balanced, indigenous population of shellfish, fish, and wildlife in and on the body of water into which the discharge is to be made.” CWA § 316(a), 33 U.S.C. § 1326(a). As the petitioners also note, section 316(b) does not similarly permit consideration of the quality of the receiving water, and this distinction is significant. See Russello, 464 U.S. at 23. Moreover, as we discussed in Riverkeeper I, the heat-pollution provision of section 316(a) is a “notable exception” to the CWA, which “otherwise relies on limitations on what a source can put into the water, not the ultimate effect of that discharge.” 358 F.3d at 190. Nothing in the statute, therefore, supports the view that the EPA can consider the quality of the receiving water in granting variances from 50
    139. The challenged provision of the Phase II Rule apparently would permit a facility to argue that, based on water quality (i.e., the level of aquatic wildlife in a particular body of water), the cost of complying with the national performance standards is not justified. The Agency explained in the preamble to the Rule that “in a waterbody that is already degraded, very few aquatic organisms may be subject to impingement or entrainment, and the costs of retrofitting an existing cooling water intake structure may be significantly greater than the benefits of doing so.” 69 Fed. Reg. at 41,604. This kind of water-quality-based regulation is not authorized by the CWA because it would exempt facilities from meeting the mandated performance standards simply because wildlife levels in the waterbody were already low, and as we held in Riverkeeper I, the CWA does permit the EPA to consider water quality in making BTA determinations. Finally, we note that to the extent that facilities on highly degraded waterbodies with relatively low wildlife levels face high compliance costs to achieve the national performance standards, those facilities may qualify for the cost-cost variance if such variance is retained on remand. Because the EPA exceeded its authority under section 316(b) by permitting (1) cost-benefit analysis and (2) assessment of the quality of the receiving water (i.e., the receiving water’s wildlife levels) in determining whether a variance is warranted, we do not need to defer to the Agency’s construction of the statute. We therefore remand this aspect of the Rule. D. TIOP Provision The Phase II Rule’s TIOP provision, 40 C.F.R. § 125.94(d), permits a facility to generally applicable rules promulgated pursuant to section 316(b). 51
    140. comply with the national performance standards determined on the basis of whether the facility has “complied with the construction, operational, maintenance, monitoring, and adaptive management requirements of a Technology Installation and Operation Plan.” Id. § 125.94(d)(1). The petitioners contend that this provision impermissibly allows a facility’s compliance to be determined not by reference to the performance standards themselves, but by evaluating whether a facility has complied with a plan to achieve the performance standards. In other words, they argue that the TIOP provision essentially allows for an unauthorized margin of error. The petitioners also argue that the EPA denied the public an opportunity to comment on the provision. Like the cost-cost compliance alternative, remand is appropriate here on two grounds: (1) the EPA did not give adequate notice regarding the provisions in section 125.94(d)(2); and (2) the record justification for the TIOP provision depends on the EPA’s selection of a suite of technologies as BTA, a selection which has been remanded for further explanation. Given this, we remand the TIOP provision without reaching the merits here. The Rule provides that during the first permit term, a facility may request that its compliance be determined based on whether it has complied with its TIOP, which must be designed to meet the performance standards, 40 C.F.R. § 125.94(d)(1), and submitted with a permit application, id. § 125.95(b)(4)(ii). During subsequent permit terms, if a facility has complied with its TIOP but is not meeting the performance standards, the facility may request that its compliance with the standards during the following term be based on whether it remains in compliance with its TIOP, revised in accordance with the facility’s adaptive management plan. 40 C.F.R. § 125.94(d)(2). The EPA explained in the Rule’s preamble that it is difficult to determine reductions in impingement mortality and entrainment relative to what would have 52
    141. occurred in the absence of control technologies given natural variability and the vagaries of sampling methods. 69 Fed. Reg. at 41,613. The EPA explained further that it established the TIOP compliance options to account for these variabilities on the ability of a technology to meet the performance standards consistently over time. Id. at 41,613-14. As previously noted, a “final rule must be a ‘logical outgrowth’ of the rule proposed.” Nat’l Black Media Coalition, 791 F.2d at 1022. The final rule must have roots in the proposal, Envtl. Integrity Project, 425 F.3d at 996, which must “fairly apprise interested persons of the subjects and issues” involved in the rulemaking, Nat’l Black Media Coalition, 791 F.2d at 1022. An agency cannot “pull a surprise switcheroo” on interested parties between a proposal and the issuance of a final rule. See Envtl. Integrity Project, 425 F.3d at 996. Although the Rule’s proposal notified interested parties that the Agency was considering a provision that would give facilities time to achieve the performance standards after implementing new technologies, the EPA gave inadequate notice of the potentially indefinite scope of this provision. Specifically, the EPA failed to provide notice of the Rule codified at 40 C.F.R. § 125.94(d)(2), which permits a facility to be deemed in compliance with the Phase II Rule in subsequent permit terms if it continues to adhere to its TIOP. In the Rule’s proposal, the EPA stated only that it was considering “the need for regulatory language that would allow facilities time to come into compliance [with the performance standards] if they choose to install technologies to meet the performance standards.” 68 Fed. Reg. at 13,586. The proposal indicated further that the EPA was “evaluating and considering allowing six months, one year, two years, or five years (one permit term) for a facility to come into compliance after issuance of its permit.” Id. The TIOP provision in the 53
    142. final Rule, by contrast, does not simply allow facilities additional time, up to one permit term, to come into compliance with the performance standards. Instead, it appears to permit a facility to satisfy the Rule’s requirements in subsequent permit terms, for an indefinite period, without ever demonstrating compliance with the performance standards, so long as the facility has adhered to its TIOP. 40 C.F.R. § 125.94(d)(2) (stating that “[d]uring subsequent permit terms” a facility “may request that compliance . . . be determined based on whether [it] remain[s] in compliance with” its TIOP). This aspect of the TIOP provision appears then not to be a “logical outgrowth” of the proposal, see Nat’l Black Media Coalition, 791 F.2d at 1022, because interested parties would not have divined from the proposal that facilities could be given an indefinite period to come into compliance with the national performance standards. We thus remand the rule for failure to provide notice and comment and because the record justification for the TIOP provision depends on the EPA’s selection of BTA, which has been remanded. E. Definitions of “New Facility” and “Existing Facility” The environmental petitioners challenge the reclassification in the Phase II Rule preamble of certain new constructions as “existing facilities,” thereby rendering them subject to the Phase II Rule rather than the more stringent Phase I requirements that apply to new facilities. We agree with the petitioners that the Agency interpretively modified a definition appearing in the Phase I Rule via statements in the preamble to the Phase II Rule without providing interested parties notice and an opportunity for comment. “An agency’s interpretation of its own . . . regulation must be given controlling 54
    143. weight unless it is plainly erroneous or inconsistent with the regulation.” Fowlkes v. Adamec, 432 F.3d 90, 97 (2d Cir. 2005) (citations and internal quotation marks omitted). Although we typically owe considerable deference to an agency’s construction of its own regulation, Udall v. Tallman, 380 U.S. 1, 16 (1965), “[u]nder settled principles of statutory and rule construction, a court may defer to administrative interpretations of a statute or regulation only when the plain meaning of the rule itself is doubtful or ambiguous,” Pfizer, Inc. v. Heckler, 735 F.2d 1502, 1509 (D.C. Cir. 1984) (emphasis in original). “Deference to agency interpretations is not in order if the rule’s meaning is clear on its face.” Id. Implicit in the rule that an agency cannot interpret a regulation contrary to its unambiguous meaning is the requirement that “an agency must adhere to its own rules and regulations.” Reuters Ltd. v. FCC, 781 F.2d 946, 950 (D.C. Cir. 1986). An agency may modify a regulation that has already been promulgated, therefore, only through the process of notice and comment rulemaking. See Alaska Prof’l Hunters Ass’n, Inc. v. FAA, 177 F.3d 1030, 1034 (D.C. Cir. 1999); see also 5 U.S.C. § 551(5) (defining “rule making,” which is governed by the notice and comment requirements of 5 U.S.C. § 553, as the “agency process for formulating, amending, or repealing a rule”); Shalala v. Guernsey Memorial Hosp., 514 U.S. 87, 100 (1995) (noting in dicta that APA rulemaking is required where an agency interpretation “adopt[s] a new position inconsistent with . . . existing regulations”). The Phase I Rule defined “new facility” as any structure whose construction commenced after January 17, 2002 that meets both the definition of “new source” in 40 C.F.R. § 122.29 (discussed below) and one of two other requirements: the structure must use either (1) “a newly constructed cooling water intake structure” or (2) “an existing cooling water intake structure whose design capacity is increased to accommodate the intake of additional cooling 55
    144. water.” See 40 C.F.R. § 125.83. A “new source” under section 122.29 is a facility that (1) “is constructed at a site at which no other source is located,” (2) “totally replaces the process or production equipment that causes the discharge of pollutants at an existing source,” or (3) undertakes “processes . . . substantially independent of an existing source at the same site.”28 40 C.F.R. § 122.29(b) (emphasis added). Section 122.29(b) provides further that, in determining whether a facility is “substantially independent” of an existing source, the director should consider “the extent to which the new facility is integrated with the existing plant; and the extent to which the new facility is engaged in the same general type of activity as the existing source.” 40 C.F.R. § 122.29(b)(iii). In determining whether a new construction qualifies as a “new facility” for purposes of the Phase I Rule, therefore, the permitting authority must perform a two-part, but not necessarily sequential, analysis. It must determine whether the construction uses a new cooling water intake structure or an existing structure whose capacity has been increased. The permitting authority must also determine whether the new construction qualifies as a “new source.” Failure to meet either part of this analysis precludes the new construction from qualifying as a “new facility,” and thus from falling under the Phase I Rule’s purview. The Phase I Rule stated that “new facilities” meeting the foregoing requirements 28 The Phase I Rule defined facilities meeting either of the first two tests of section 122.29(b) as “greenfield” facilities and facilities meeting the third test as “stand-alone” facilities. 40 C.F.R. § 125.83 (“A greenfield facility is a facility that is constructed at a site at which no other source is located, or that totally replaces the process or production equipment at an existing facility. A stand-alone facility . . . is constructed on property where an existing facility is located and whose processes are substantially independent of the existing facility at the same site.”). 56
    145. include a “stand-alone” facility, which the Phase I Rule defined as “a new, separate facility that is constructed on property where an existing facility is located and whose processes are substantially independent of the existing facility at the same site.” Id. § 125.83. This is because such a facility, by definition, essentially qualifies as a “new source.” Compare id. with § 122.29 (defining “new source” as including a facility that undertakes “processes . . . substantially independent of an existing source at the same site”). The Phase I Rule clarified, however, that “new facility” does not include “new units that are added to a facility for purposes of the same general industrial operation (for example, a new peaking unit at an electrical generating station).” Id. § 125.83. This is presumably because such units do not qualify as “new sources” in that they are not substantially independent of existing sources. See id. § 122.29(b)(1)(iii) (setting forth the factors to be considered in determining substantial independence, including “the extent to which the new facility is integrated with the existing plant; and the extent to which the new facility is engaged in the same general type of activity as the existing source”). The Phase I Rule thus appears to have left regulation over the following to a subsequent rulemaking phase: (1) new stand-alone facilities that use existing intake structures whose design capacity is not increased and (2) new units that are added to a facility for purposes of the same general industrial operation even if they require either an increase in the intake structure design capacity or the construction of a new cooling water intake structure altogether. Id.29 29 A facility would clearly find it advantageous to classify a new construction on its grounds as one added for purposes of the same general industrial operation rather than as a stand- alone facility. A stand-alone facility is treated as an existing facility only if it uses an existing intake structure whose capacity is not increased; a “new unit,” by contrast, is treated as part of an existing facility even if it requires the construction of a new intake structure or an increase in an existing intake structure’s capacity. 57
    146. The Phase II Rule defines “existing facility” as any facility whose construction commenced on or before January 17, 2002, “and any modification of, or any addition of a unit at such a facility that does not meet the definition of a new facility at § 125.83.” Id. § 125.93. Thus, from this definition, it appears that new stand-alone facilities that use existing, unmodified intake structures and new units added to a facility for purposes of the same industrial operation, regardless of their impact on the facility’s cooling water intake structure, (i.e., the two kinds of new constructions left unregulated by the Phase I Rule) are considered “existing facilities” and governed by the Phase II Rule. The parties’ dispute concerns statements in the preamble to the Phase II Rule that purportedly narrow, by way of interpretation, the Phase I Rule’s definition of “new facility”30 without the required procedures of notice and comment. In the preamble to the Phase II Rule, the EPA states that “the Phase I [R]ule treated almost all changes to existing facilities for purposes of the same industrial operation as existing facilities.” 69 Fed. Reg. at 41,579. The preamble then appears to distinguish stand-alone facilities from new units that are part of the same industrial operation, thereby defining the latter as existing facilities without reference to the definition of “new source” or the “substantial independence” test of 40 C.F.R. § 122.29. Id. at 41,579 n.2a. The preamble states that the “substantial independence” test does not apply where there is an addition to an existing facility for purposes of the same industrial operation, such as the “addition 30 We find no merit to the EPA’s argument that the state and environmental petitioners here are ostensibly challenging the Phase II Rule’s definition of “existing facility” but are actually seeking review of the meaning of “new facility” under the Phase I Rule – for which their claim would be time-barred. Though the EPA claims that “[n]othing in the Phase II Rule altered or amended the definition of ‘new facility,’” it is clear from the discussion above that the preamble to the Phase II Rule eliminates without notice or comment the analysis the Phase I Rule had required. 58
    147. of new generating units at the same site” as an existing facility, id. at 41,579, because such additions “are categorically treated as ‘existing facilities’ regardless of any other considerations unless they completely replace an existing facility and its cooling water design intake capacity is increased,” id. at 41,579 n.2a. These comments are contrary to the plain meaning of the relevant portion of the Phase I Rule. The Phase I Rule unambiguously stated that “new facility” means any structure that is a “new source,” as defined by 40 C.F.R. § 122.29, subject to certain other requirements. Under this provision, a source is considered “new” if, inter alia, “[i]ts processes are substantially independent of an existing source at the same site.” 40 C.F.R. § 122.29(b)(1)(iii). A permitting authority could not classify a source constructed at the site of an existing source as new or existing for purposes of the Phase I Rule, therefore, without reference to the “substantial independence” test. It is plain, then, that the Phase I Rule distinguished between “stand-alone” facilities and “new units,” where the new construction is not built at an empty site and does not totally replace an existing source, by reference to the definition of a “new source.” A stand-alone facility is “substantially independent” of an existing facility, and therefore a new source; a new unit that is part of the same industrial operation as an existing facility is not substantially independent of an existing facility, and therefore not a new source. It is impossible to determine which classification applies to a particular construction under the Phase I Rule without referring to the definition of “new source,” i.e., whether it satisfies the “substantial independence” test. Put differently, the touchstone of the definition of “new facility” in the Phase I Rule is whether a source is a “new source.” The Phase I Rule’s plain terms thus indicate that a unit that is “substantially independent” of an existing facility is not “part of the same general industrial 59
    148. operation” as the existing facility. Any elimination of the “substantial independence” inquiry, therefore, strikes at the heart of the Phase I Rule and its classification of what facilities are subject to its requirements. The EPA claims that the Phase II Rule has in no way eliminated the “substantial independence” test and that the Rule’s preamble merely makes clear that the fifth sentence in section 125.83 exempts “new units” from regulation under the Phase I Rule. This argument fails because the Phase I Rule provides no way to distinguish between stand-alone facilities and new units where the construction is built on a site where a source is already located and does not totally replace the existing source except by reference to the “substantial independence” test, i.e., without assessing the factors set forth at 40 C.F.R. § 122.29(b)(1)(iii) in order to determine whether the source is new or existing. Just as “stand-alone facility” has no intrinsic meaning, neither does “new unit.” The Phase I Rule defines each by reference to the “substantial independence” test of section 122.29(b). Thus, while an existing facility can be repowered with new generating units and remain an “existing facility” for regulatory purposes under section 316(b), the determination can only be made by reference to whether a particular new generating unit is a stand-alone facility or a new unit that is part of the same general industrial operation as an existing facility. In fact, a permitting authority must first determine whether a source is “new” within the meaning of 40 C.F.R. § 122.29(b) before it can conclude that the source is a stand-alone facility or a new unit added to an existing facility for purposes of the same general industrial operation. Because the Phase I Rule was not ambiguous, we do not owe deference to the Agency’s interpretation of the Phase I Rule in the preamble to the Phase II Rule. See Fowlkes, 60
    149. 432 F.3d at 97; Pfizer, 735 F.2d at 1509. By permitting the Agency to determine that a new construction is not subject to the Phase II Rule without any definitional guidance and in contravention of the Phase I Rule, the EPA has expanded the scope of what may be classified as a “new unit” while narrowing the Phase I definition of “stand-alone” facility. Moreover, by including a potentially expansive definition of “new unit” in the preamble to the Phase II Rule, the EPA has interpretively modified the definitions that appeared in the Phase I Rule without providing interested parties an opportunity for notice and comment. Accordingly, we direct the EPA on remand to adhere to the definitions set forth in the Phase I Rule, see Reuters, 781 F.2d at 950-52, or to amend those definitions following notice and comment, see Alaska Prof’l Hunters, 177 F.3d at 1034. IV. The Industry Petitioners A. Applicability of Section 316(b) to Existing Facilities Entergy argues that the EPA lacks authority to apply CWA section 316(b) to existing, as opposed to new, facilities. We disagree and conclude that, at the very least, the EPA permissibly interpreted the statute to cover existing facilities and that its interpretation is therefore entitled to deference under Chevron. Entergy’s argument turns primarily on the statutory language that the best technology available be reflected in the “location, design, construction, and capacity of cooling water intake structures” – a collection of words Entergy contends indicates Congress’s intent to 61
    150. regulate only new facilities.31 Entergy argues further that the EPA has authority to approve cooling water intake structures only before construction and cannot regulate these structures through the NPDES permits issued pursuant to CWA section 402(a)(1), 33 U.S.C. § 1342(a)(1), which allows the Administrator to “issue a permit for the discharge of any pollutant,” but not, in Entergy’s view, to cover existing intake structures. The EPA emphasizes that section 316(b) cross-references section 301, which applies to existing facilities, and that the Agency’s historical practice of applying section 316(b) to existing facilities effectuates Congress’s objectives in enacting the CWA. The EPA contends that Congress’s link between effluent limitations established pursuant to section 301 and BTA under section 316(b) indicates an intent to regulate cooling water intake structures at existing facilities. The EPA thus argues that the intake-structure standard is to be applied whenever an intake structure is present at a point source of pollutant discharge, whether that point source is new or existing. As to Entergy’s argument concerning the ability to assess intake structures during NPDES permitting proceedings, the EPA argues that its decision to implement the Phase II requirements through NPDES permits has a strong textual basis. It notes that section 402 31 Entergy cites a number of federal statutory and regulatory provisions using similar words and argues that each provision applies only in the pre-construction context. The environmental petitioners contend that the provisions upon which Entergy relies explicitly limit their application to new facilities and that, as a result, the words “location, design, construction and capacity,” standing alone, do not apply solely to new facilities; otherwise, the environmental petitioners’ argument goes, there would be no need to limit expressly the provisions’ applicability to new facilities. For example, 49 U.S.C. § 60103, which governs safety standards for liquefied natural gas pipelines, by its express terms does not apply to certain existing facilities. 49 U.S.C. § 60103(c) (“Except as provided in paragraph (2) of this subsection, a design, location, installation, construction, initial inspection, or initial testing standard prescribed under this chapter after March 1, 1978, does not apply to an existing liquefied natural gas pipeline facility . . . .”). These are additional reasons for the holding discussed in this opinion. 62
    151. provides for the issuance of “a permit for the discharge of any pollutant” so long as the discharge meets “all applicable requirements under sections 1311 . . . [and] 1316,” sections 301 and 306 of the CWA respectively. CWA § 402(a)(1), 33 U.S.C. § 1342(a)(1). EPA contends that the Phase II requirements are “requirements under sections 1311 . . . [and] 1316” and therefore that cooling water intake structures may be regulated via permits issued pursuant to section 402. The EPA has the better of both arguments. As to whether section 316(b) applies to existing facilities, we find Entergy’s textual argument, while superficially appealing, ultimately to be unavailing. Nothing in section 316(b) indicates that because it applies to the “location, design, construction, and capacity” of a facility’s cooling water intake structure, the section is therefore limited to new facilities and does not require existing facilities either to modify existing intake structures or to construct new intake structures in order to come into compliance with the EPA’s Rule.32 In fact, given the cross-references in section 316(b) to provisions governing both new and existing facilities, the EPA’s reading is far more reasonable than Entergy’s. See E.I. du Pont de Nemours & Co. v. Train, 430 U.S. 112, 121 (1977) (“Section 301(b) [to which section 316(b) expressly refers] defines the effluent limitations that shall be 32 We are not persuaded by Entergy’s selective definitions of the terms “location, design, construction, and capacity.” While we agree that words in a statute are to be accorded “their ordinary, contemporary, common meaning,” Harris v. Sullivan, 968 F.2d 263, 265 (2d Cir. 1992), it is clear that the EPA’s interpretation of the statute does not deviate from this requirement. See, e.g., Black’s Law Dictionary 958 (8th ed. 1999) (“[L]ocation” is “[t]he specific place or position of a person or thing.”); id. at 478 (“[D]esign” means “a plan or scheme” or “[t]he pattern or configuration of elements in something.”); id. at 332 (“[C]onstruction” is “[t]he act of building,” but also “the thing so built.”); Webster’s Third New Int’l Dictionary 330 (1986) (“[C]apacity” means “the power or ability to hold, receive or accommodate” and “a containing space: a measure of content for gas, liquid, or solid.”). Nothing in the above definitions suggests the EPA could not have reasonably interpreted such words to include existing facilities. 63
    152. achieved by existing point sources . . . .”); id. at 136 (holding that “301 does authorize the [Agency] to promulgate effluent limitations for classes and categories of existing point sources”). The cross-reference to section 301 is particularly significant given that Congress, having made this explicit cross-reference, did not then limit section 316(b)’s application to new facilities – which would have been a simple task to do. At the very least, the EPA’s view that section 316(b) applies to existing facilities is a reasonable interpretation of the statute, and we therefore accord it deference. See Chevron, 467 U.S. at 842-43. Entergy’s argument concerning the permitting process presents a closer question, but it is ultimately not persuasive and does not undermine our conclusion that section 316(b), on its face, applies to existing facilities. The textual basis for the EPA to regulate cooling water intake structures during the periodic permitting process applicable to the discharge of pollutants is not immediately apparent. Section 402 conditions the issuance of a permit on the circumstance that a “discharge will meet . . . all applicable requirements under sections 1311 . . [and] 1316.” CWA § 402(a)(1), 33 U.S.C. § 1342(a)(1) (emphasis added). While the Phase II requirements are “requirements” under sections 301 and 306, they do not apply to the discharge of pollutants, and section 402 says nothing about conditioning a permit on compliance with other requirements of sections 301 and 306, i.e., requirements not relating to the discharge of pollutants. Despite this textual hiccup, the EPA’s decision to use the NPDES process to enforce section 316(b) is not unreasonable. Insofar as the provision applies to existing facilities – and the cross-reference in section 316(b) to section 301 provides a clear textual basis for that conclusion – the EPA could enforce it only through some permit process following the issuance of an initial construction permit. Moreover, the structure of section 316(b) supports the view that its requirements are to 64
    153. be enforced through the same process used to enforce the effluent limitations of sections 301 and 306. As noted, the statute requires that any standard established by the EPA to govern the discharge of pollutants from existing facilities must also regulate cooling water intake structures. CWA § 316(b), 33 U.S.C. § 1326(b) (providing that “[a]ny standard established pursuant to section 1311 . . . or section 1316” shall also regulate the cooling water intake structures of point sources). In light of this language, it is at least reasonable to conclude that Congress intended the requirements of section 316(b) to be part and parcel of any regulation of, and therefore any permit issuance relating to, the discharge of pollutants. See Riverkeeper I, 358 F.3d at 185-86. It is a fair conclusion that section 402 implicitly requires permitting authorities to ensure compliance with section 316(b) as a permit condition. See U.S. Steel Corp. v. Train, 556 F.2d 822, 850 (7th Cir. 1977) (“[Section] 402(a)(1) implicitly requires the Administrator to insure compliance with § 316(b) as one of the permit conditions.”), overruled on other grounds by City of West Chicago, Ill. v. U.S. Nuclear Regulatory Comm’n, 701 F.2d 632, 644 (7th Cir. 1983) (abandoning the view that 5 U.S.C. § 558(c) independently provides that formal adjudicatory hearings must be held when requested by a license applicant under CWA § 402). Section 402 thus does not undermine the deference to which the Agency’s interpretation of section 316(b) is entitled under Chevron. Because section 316(b) plainly applies to existing facilities and Congress intended the requirements of section 316(b) to apply in tandem with the effluent limitations established pursuant to sections 301 and 306, we conclude that the EPA may regulate cooling water intake structures via the NPDES permit process. Otherwise, Congress’s intent to regulate the intake 65
    154. structures of existing facilities could not be effectuated.33 Accordingly, we reject this aspect of Entergy’s challenge. B. Definition of “Adverse Environmental Impact” In the Phase II Rule, as in the Phase I Rule, the EPA has interpreted the statutory directive of section 316(b) to minimize “adverse environmental impact” (“AEI”) to require the reduction of “the number of aquatic organisms lost as a result of water withdrawals associated” with cooling water intake structures. 69 Fed. Reg. at 41,586. This interpretation reflects the fact that section 316(b) is a somewhat unusual provision of the CWA in that it governs the environmental effects of large scale withdrawals from waters of the United States rather than the release of pollutants into receiving water. As did the industry petitioners in Riverkeeper I, PSEG argues that the EPA arbitrarily defined AEI to include any loss of or harm to aquatic organisms 33 Entergy also contends that section 316(b) cannot be enforced via a citizen suit pursuant to section 505(a)(1), 33 U.S.C. § 1365(a)(1), because that provision covers only violations of “an effluent standard or limitation” or “an order issued by the Administrator or a State with respect to such a standard or limitation,” and that the statute therefore applies only in the pre-construction context. Because the case before us is not a citizen suit, we are not directly presented with the question of whether a citizen can sue to enforce section 316(b) and do not decide the question. We are not, however, persuaded by Entergy’s argument. The citizen suit provision states that a citizen may commence a suit against any person who is alleged to be in violation of “an effluent standard or limitation under this chapter.” CWA § 505(a)(1), 33 U.S.C. § 1365(a)(1). Viewed in light of this language alone, it is difficult to characterize a violation of regulations promulgated pursuant to section 316(b) as a violation of an “effluent standard or limitation.” The statute, however, defines “effluent standard or limitation” to include “an effluent limitation or other limitation under section 1311” and a “standard of performance under section 1316.” CWA § 505(f), 33 U.S.C. § 1365(f). Arguably, therefore, a limitation established pursuant to section 316(b) is an “other limitation under section 1311” and a “standard of performance under section 1316,” given the cross-reference to sections 301 and 306 in section 316(b), and thus falls within the scope of the citizen-suit provision. While we do not decide the question here, we do not read the citizen suit provision to undermine the deference we accord the EPA’s reasonable interpretation pursuant to Chevron that section 316(b) applies to existing facilities. 66
    155. due to impingement mortality and entrainment rather than only more severe “population-level effects.” It contends that the EPA has historically focused on the “population dynamics” of aquatic organisms and the fact that many of these organisms are “density dependent,” i.e., they produce large numbers of offspring, only a few of which survive to adulthood.34 On the theory that the “vast majority” of entrained organisms would have died of natural causes in any event, PSEG argues that reduction of impingement mortality and entrainment will have only a marginal positive environmental impact.35 PSEG contends that because the Agency has changed course from its longstanding interpretation of AEI without giving a clear and reasoned justification for its decision, the EPA’s Rule is entitled to less deference than it otherwise would receive. PSEG also argues that the Phase II Rule’s exclusion from the entrainment-reduction requirement of facilities withdrawing small amounts of water, facilities withdrawing water from lakes other than the Great Lakes, and facilities that have a small capacity utilization rate undercuts the EPA’s 34 PSEG relies on the Draft Guidance for Evaluating the Adverse Impact of Cooling Water Intake Structures on the Aquatic Environment, prepared by the EPA in 1977. While the EPA indicated in this document that “[a]dverse environmental impacts occur whenever there will be entrainment or impingement damage as a result of the operation of a specific cooling water intake structure,” PSEG focuses on the Agency’s statement that the “critical question” is “the magnitude of any adverse impact.” 35 The environmental petitioners contend that PSEG has missed the mark here entirely. They argue that when organisms die of natural causes they remain available as food for top predators, but that when organisms are entrained, and potentially disintegrated, they are consumed by lower organisms. They argue also that entrained organisms are not available to consume organisms lower on the food chain. The environmental petitioners thus contend that one of the primary adverse environmental impacts of impingement mortality and entrainment is their effect on the ecosystem as a whole by disrupting the food chain. The EPA discussed these and other environmental impacts attributable to impingement mortality and entrainment in the preamble to the Phase II Rule, concluding that “[d]ecreased numbers of aquatic organisms can disrupt aquatic food webs and alter species composition and overall levels of biodiversity.” 69 Fed. Reg. at 41,586. 67
    156. determination that impingement mortality and entrainment are per se adverse environmental impacts. We are not persuaded by PSEG’s arguments. We agree with the EPA that the Phase II Rule is based on substantially the same record evidence of impingement mortality and entrainment relied upon in promulgating the Phase I Rule and that we rejected substantially the same arguments advanced here by the industry petitioners in Riverkeeper I, 358 F.3d at 197 (“The EPA considered all of the factors that UWAG now raises, and we are inclined to defer to the EPA’s judgment of how best to define and minimize ‘adverse environmental impact.’” (internal footnote omitted)). Riverkeeper I thus controls this issue. In Riverkeeper I, we rejected the arguments that some species are nuisances and require eradication, that other species respond to population losses by increasing their reproduction, and that removing large numbers of aquatic organisms from waterbodies is not in and of itself an adverse impact. 358 F.3d at 196. We specifically rejected the view that “the EPA should only have sought to regulate impingement and entrainment where they have deleterious effects on the overall fish and shellfish populations in the ecosystem, which can only be determined through a case-by-case, site-specific regulatory regime.” Id. We emphasized that “the EPA’s focus on the number of organisms killed or injured by cooling water intake structures is eminently reasonable.” Id. We reiterated that Congress had “rejected a regulatory approach that relies on water quality standards,” analogizing the argument pressed there as urging what is essentially a water quality standard that focuses on fish populations and consequential 68
    157. environmental harm.36 Id. at 196-97. Given that the record evidence on this issue has not changed in any meaningful way since the Phase I rulemaking, we are both persuaded and bound by our statements on this issue in Riverkeeper I. Were we considering the issue in the first instance, however, we would be inclined to defer to the EPA’s judgment in any event. The EPA explained that it has set “performance standards for minimizing adverse environmental impact based on a relatively easy to measure and certain metric – reduction of impingement mortality and entrainment.” 69 Fed. Reg. at 41,600. It explained further that it chose this approach “because impingement and entrainment are primary, harmful environmental effects that can be reduced through the use of specific technologies” and stated that “where other impacts at the population, community, and ecosystem levels exist, these will also be reduced by reducing impingement and mortality.” Id. We see no reason to second-guess this judgment, given the Agency’s consideration of the various environmental consequences of cooling water intake structures. See Nat’l Wildlife Fed’n, 286 F.3d at 570 (noting that courts afford the EPA considerable discretion to weigh and balance 36 We make an additional observation here, which we cited in Riverkeeper I as a reason for rejecting restoration measures as impermissible under the statute. It is significant that in section 316(a), which governs thermal discharges, Congress permits the EPA to vary the standard applicable to a point source “by considering the particular receiving waterbody’s capacity to dissipate the heat and preserve a ‘balanced, indigenous’ wildlife population.” 358 F.3d at 190. It is also significant that Congress “did not include that [water quality or population level] approach (or make any reference to it) in the very next subsection,” id., since “where Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion.” Id. (quoting Bates v. United States, 522 U.S. 23, 29-30 (1997) (internal quotation marks omitted)). The statutory structure thus indicates that Congress did not intend to limit “adverse environmental impact” in section 316(b) to population-level effects. 69
    158. various factors in determining how to establish performance standards).37 C. Zero Entrainment Survival Assumption The Phase II Rule requires a reduction of impingement mortality, but a reduction of entrainment generally. See 40 C.F.R. § 125.94(b). The industry petitioners contend that the EPA improperly presumed that all entrained organisms are killed. They argue that the Rule’s assumption that no organisms survive entrainment is contrary to the evidence which, in their view, indicates a survival rate of significantly more than zero. We conclude that in light of uncertain record evidence, the EPA acted within its discretion in assuming zero entrainment survival. The EPA explained in the preamble to the Phase II Rule that it assumed zero percent entrainment survival because it “does not have sufficient data to establish performance standards based on entrainment survival for the technologies used as the basis for today’s rule.” 69 Fed. Reg. at 41,620. It explained further that it “believes the current state of knowledge does not support reliable predictions of entrainment survival that would provide a defensible estimate for entrainment survival above zero at a national level.” Id. It also stated that the performance standard would likely have been higher had it incorporated entrainment survival into its 37 We also find no merit in PSEG’s claim that the final Rule gave no notice of the supposed change in the EPA’s view of “adverse environmental impact.” Not only did the proposed Rule seek comment on several competing definitions, one of which referred to gauging such impact in part by examination of the “absolute damage” wrought by impingement and entrainment, but in which the EPA noted that “there will be adverse environmental impact whenever there is entrainment or impingement ‘damage’ as a result of a cooling water intake structure.” 67 Fed. Reg. at 17,162. While the EPA did not ultimately adopt any of the definitions it noticed, it is clear that PSEG was “fairly apprise[d]” that the EPA was considering an approach to AEI that looked to entrainment and impingement at the individual level. Nat’l Black Media Coalition, 791 F.2d at 1022. 70
    159. conclusions. Id. While impingement mortality can be readily quantified, the EPA contends, entrainment mortality cannot because many entrained organisms are small, fragile, and prone to disintegrate during entrainment. The EPA also contends that death from entrainment sometimes occurs immediately but other times only after an organism is discharged back into the waterbody. In view of these factors, the EPA claims that it reasonably concluded that the available data did not support an estimate of entrainment survival at the national level. None of the peer reviewers accepted the EPA’s assumption of zero percent survival. For instance, peer reviewer Dr. Mark Bain was “not convinced by the arguments presented that fish do not survive entrainment in significant numbers” and concluded that there is “very strong evidence that entrainment survival is not zero.” Another peer reviewer, Dr. Charles Hocutt, concluded that the EPA’s assumption is based on inference and innuendo and does not statistically refute opposing views. Although peer reviewer Dr. Greg Garman stated that the study submitted by the power industry was “very clearly biased” and “too seriously flawed to provide a serious challenge to the EPA position,” he also noted that “EPA’s position is only slightly more defensible” given the lack of rigorous statistical analyses. No peer reviewer, however, expressed the belief that reliable national statistics on entrainment survival exist. Hocutt found “it difficult for the layman and professional alike to draw confident conclusions without a detailed analysis” of the evidence, and Garman suspected a lack of sufficient data “to conduct a definitive and statistically valid test of the EPA zero survival assumption.” While these comments do not, on balance, support the EPA’s assumption, neither do they reflect any meaningful agreement on the incidence of entrainment survival. Given the statutory directive to set national standards and the well-documented 71
    160. uncertainty in the entrainment data, the EPA was well within its authority to determine that it could not provide a reasonable estimate of entrainment survival on a national basis. Judicial review is considerably deferential when “the agency’s decision rests on an evaluation of complex scientific data within the agency’s technical expertise.” Texas Oil & Gas Ass’n, 161 F.3d at 934; City of Waukesha v. EPA, 320 F.3d 228, 247 (D.C. Cir. 2003) (stating that an agency is entitled to “an extreme degree of deference . . . when it is evaluating scientific data within its technical expertise” (citation and internal quotation marks omitted)). Moreover, “it is within EPA’s discretion to decide that in the wake of uncertainty, it would be better to give the values a conservative bent rather than err on the other side.” Am. Iron & Steel Inst. v. EPA, 115 F.3d 979, 993 (D.C. Cir. 1997). Indeed, one peer reviewer expressly noted that the EPA had adopted a “conservative approach” by its assumption of zero entrainment survival in the Rule. It is thus clear that the EPA acted well within its discretion in presuming zero entrainment survival after the Agency had reviewed a substantial body of complex scientific data, and acknowledging that the evidence is inconclusive, it adopted a conservative approach. For these reasons and those stated in Riverkeeper I, we therefore “defer to the EPA’s judgment of how best to define and minimize ‘adverse environmental impact.’” 358 F.3d at 197. D. Nuclear Plants Entergy contends that the Phase II Rule fails to account for its purportedly disproportionate impacts on nuclear power plants and is therefore arbitrary and capricious. Entergy argues that nuclear facilities face unique safety concerns associated with the stable flow of cooling water to ensure safe reactor operation and shutdown. Any change in water intake or 72
    161. obstruction of water intake systems due to, for example, the clogging of screens, it argues further, affects nuclear power facilities in specific and serious ways. Entergy takes the position that the EPA failed to account for these issues in the Phase II Rule. We disagree because the record demonstrates adequate consideration by the EPA of nuclear plants’ particular concerns. The EPA considered and responded to comments from nuclear facilities during the rulemaking process. Most importantly, the Agency considered whether the Rule’s requirements presented any concerns relating to the safety of nuclear facilities. 69 Fed. Reg. at 41,585 (noting that the EPA had coordinated with the Nuclear Regulatory Commission to ensure that there would not be a conflict between the EPA Rule and safety requirements applicable to nuclear facilities). The EPA ultimately included in the Phase II Rule a provision that accounts for this concern by providing for a site-specific compliance alternative for nuclear facilities. This provision states that if a nuclear facility “demonstrate[s] to the [EPA] based on consultation with the Nuclear Regulatory Commission that compliance with this subpart would result in a conflict with a safety requirement established by the Commission, the [EPA] must make a site-specific determination of best technology available for minimizing adverse environmental impact that would not result in a conflict.” 40 C.F.R. § 125.94(f). We defer to the EPA’s determination that this compliance alternative ensures that any safety concerns unique to nuclear facilities will prevail over application of the general Phase II requirements. See Nat’l Wildlife Fed’n, 286 F.3d at 570 (noting that appellate courts give an agency considerable discretion when it has weighed and balanced the appropriate factors); BP Exploration & Oil, Inc. v. EPA, 66 F.3d 784, 802 (6th Cir. 1995) (“The overriding principle in our review of the Final Rule is that the agency has broad discretion to weigh all relevant factors 73
    162. during rulemaking. The CWA does not state what weight should be accorded to the relevant factors; rather, the Act gives EPA the discretion to make those determinations.”). Moreover, we are persuaded that the generous cost-cost compliance alternative, which we remand for lack of notice but do not address on the merits, may further account for Entergy’s concerns. Accordingly, we deny the petition for review insofar as it challenges the Rule’s application to nuclear facilities. E. Independent Suppliers The Phase II Rule provides that a large, existing facility is subject to the Rule’s requirements even when it obtains cooling water from an independent supplier that is not itself a Phase II existing facility. 40 C.F.R. § 125.91(c). Put differently, a Phase II facility can purchase cooling water only from suppliers whose intake structures are in compliance with the Phase II Rule. The provision is intended to prevent circumvention of the Rule by purchasing cooling water from non-Phase II entities. UWAG argues that the EPA gave inadequate notice of the scope of this so-called third-party or independent-supplier Rule. We agree. UWAG contends that while the Phase I Rule included an independent-supplier provision similar to the one at issue here, the parallel provision of the Phase I Rule applies only to third-party suppliers that are not point sources, and thus do not need discharge permits and are not subject to any rule promulgated pursuant to section 316(b).38 UWAG argues that the Rule’s 38 The Phase I Rule provided that [u]se of a cooling water intake structure includes obtaining cooling water by any sort of contract or arrangement with an independent supplier (or multiple suppliers) of cooling water if the supplier or suppliers withdraw(s) water from waters of the United States. . . . [t]his provision is intended to prevent 74
    163. proposal indicated that the Phase II provision would track the language of the Phase I provision. UWAG contends, therefore, that the EPA gave inadequate notice of the more broadly phrased provision that was actually promulgated: a rule applying both to the intake structures of third- parties that are not point sources, and therefore not subject to direct regulation under section 316(b), as well as to facilities with intake structures that will be governed by the Phase III rule, which will encompass smaller power plants and other facilities. The issue is thus whether the EPA provided adequate notice that the Rule ultimately promulgated would impose Phase II requirements on Phase III facilities that supply cooling water to Phase II facilities. As we have noted, our inquiry into whether an agency has provided adequate notice of its rulemaking as required by the APA is guided by the “logical outgrowth” test. Nat’l Black Media Coalition, 791 F.2d at 1022. That is, we must determine whether the Agency’s notice fairly apprised interested parties of the rulemaking, id., or whether the final Rule was sufficiently remote or distant from the Agency’s proposal to constitute a “surprise switcheroo,” Envtl. Integrity Project, 425 F.3d at 996. The proposed Rule published in the Federal Register stated that “[u]se of a cooling water intake structure includes obtaining cooling water by any sort of contract or arrangement with an independent supplier (or multiple suppliers) of cooling water if the supplier or suppliers withdraw(s) water from waters of the United States.” 67 Fed. Reg. at 17,220. The notice then stated that the provision was intended “to prevent circumvention of these circumvention of these requirements by creating arrangements to receive cooling water from an entity that is not itself a point source. 40 C.F.R. § 125.81(b). 75
    164. requirements by creating arrangements to receive cooling water from an entity that is not itself a point source,” id. (emphasis added), indicating that the provision would track the Phase I Rule’s language. The initial language of the provision appearing in the proposal – defining “independent supplier” as any provider that withdraws water from waters of the United States – is virtually identical to the language in the final Rule.39 The final Rule, however, states that it “is intended to prevent circumvention of these requirements by creating arrangements to receive cooling water from an entity that is not itself a Phase II existing facility.” 40 C.F.R. § 125.91(c) (emphasis added). This difference, while small, is not insignificant. As noted, under the proposed Rule, Phase II facilities could purchase cooling water from facilities complying with the Phase I, II, or III requirements. Under the final Rule, by contrast, Phase II facilities can purchase cooling water only from facilities complying with the Phase II Rule. The proposal therefore apprised affected parties of the general subject, but not the scope, of the Rule ultimately promulgated. Interested parties were thus given notice that the EPA sought to prevent circumvention of its rules and an opportunity to comment on this general policy, but could not have anticipated the final Rule’s scope. Indeed, while the final provision has roots in the proposal, it clearly reaches further than the proposed provision and even the more stringent Phase I Rule. Affected parties would therefore have had no reason to anticipate the Agency’s final 39 The final Rule provides that “[u]se of a cooling water intake structure includes obtaining cooling water by any sort of contract or arrangement with one or more independent suppliers of cooling water if the supplier withdraws water from waters of the United States but is not itself a Phase II existing facility . . . . This provision is intended to prevent circumvention of the requirements by creating arrangements to receive cooling water from an entity that is not itself a Phase II existing facility.” 40 C.F.R. § 125.91(c). 76
    165. course in light of the initial notice. Because the EPA provided inadequate notice of the scope of the Phase II Rule’s independent-supplier provision, we conclude that the Rule ultimately promulgated is not a “logical outgrowth” of the proposed Rule. Nat’l Black Media Coalition, 791 F.2d at 1022. Our conclusion rests in large part on the similarity between the Phase I Rule’s independent-supplier provision and the proposed Phase II Rule’s parallel provision. Given that the requirements under the Phase I Rule are more stringent than those imposed under the Phase II Rule, the proposal provided no notice that the Phase II Rule’s independent-supplier provision would be more stringent than the Phase I Rule’s provision. We therefore remand this aspect of the Rule. F. Definition of “Great Lakes” The entrainment performance standard of 40 C.F.R. § 125.94(b)(2) applies to facilities that use cooling water “withdrawn from . . . one of the Great Lakes.” 40 C.F.R. § 125.49(b)(2)(ii)(A). “Great Lakes,” however, is not defined in the Rule. An affidavit of UWAG’s counsel states that UWAG members have been told by state regulators that they have been told by EPA Headquarters that EPA would apply the Great Lakes national performance standards regarding impingement and entrainment to the cooling water intake structures located in the Great Lakes connecting channels or in waterways with open fish passage to a Great Lake and within 30 miles from a Great Lake. UWAG argues that interested parties were given no notice of this interpretation of the Phase II Rule or opportunity to comment on it and that the interpretation has no record support. UWAG also remarks that the EPA mentioned in the preamble to the Rule that in assessing the national 77
    166. environmental benefits of its final Phase II Rule, it had evaluated the benefits in seven study regions. The EPA defined the Great Lakes region for that purpose as follows: The Great Lakes region includes all facilities in scope of the Phase II rule that withdraw water from Lakes Ontario, Erie, Michigan, Huron, and Superior or are located on a waterway with open fish passage to a Great Lake and within 30 miles of the lake. 69 Fed. Reg. at 41,655. UWAG contends that the EPA never indicated that this definition would apply to § 125.94(b)(2) and seeks either a ruling that “Great Lakes” has its ordinary meaning or a remand for an express definition of what the term means. Notwithstanding UWAG’s proffer that the EPA has informally interpreted “Great Lakes,” the Rule itself does not define what is encompassed by the term, and UWAG has provided no documentary evidence that the EPA has issued a formal and binding definition or even applied a particular definition in a permitting proceeding. There is, therefore, no final agency action for us to review. We agree with the holding of the Seventh Circuit in American Paper Institute, Inc. v. Environmental Protection Agency, that in the absence of a formal and binding rule or some other final agency action, judicial review is not available at this time. 882 F.2d 287, 289 (7th Cir. 1989). We therefore dismiss for lack of jurisdiction so much of the petition for review as challenges the purported definition of “Great Lakes.” G. Preemptive Preservation of Issues UWAG has purported to “preserve” the right to raise new challenges to the Phase II Rule if we remand significant aspects of it to the EPA because UWAG views the Rule as an integrated whole. If certain aspects of the Rule are remanded, UWAG suggests, previously unobjectionable provisions may become, in its view, unacceptable. We are not sure what 78
    167. challenges UWAG seeks to preserve. Of course, UWAG will have the right to challenge any rule the EPA may promulgate on remand through a new petition for review. See 33 U.S.C. § 1369(b). Insofar as UWAG seeks to raise further challenges following the issuance of our ruling but before the EPA acts on remand, however, it waived those challenges by failing to raise them in the briefs it has already submitted to this Court. See Norton v. Sam’s Club, 145 F.3d 114, 117 (2d Cir. 1998) (noting that issues not argued in a party’s briefs are considered waived). We thus reject UWAG’s claim to have preserved the right to raise new challenges to the Rule currently before us on this petition for review. CONCLUSION For the foregoing reasons, the state and environmental petitioners’ petitions are granted in part and denied in part, and the industry petitioners’ petitions granted in part, denied in part, and dismissed in part for lack of jurisdiction. We remand to the EPA the provision establishing BTA so that it may provide either a reasoned explanation of its decision or a new determination of BTA based on permissible considerations. We further remand the site-specific cost-cost variance and the TIOP provision because the cost-cost variance and subpart (d)(2) of the TIOP provision provided inadequate notice and both depend on the BTA determination, which we remand today. We remand as based on impermissible constructions of the statute those provisions that (1) set performance standards as ranges without requiring facilities to achieve the greatest reduction of adverse impacts they can; (2) allow compliance through restoration measures; and (3) authorize a site-specific cost-benefit variance as impermissible under the statute. We further remand for notice and comment the independent suppliers provision. We 79
    168. also direct the EPA on remand to adhere to the definition of “new facility” set forth in the Phase I Rule or to amend that definition by rulemaking subject to notice and comment. Finally, we dismiss for lack of jurisdiction so much of the petitions as challenges the purported definition of “Great Lakes” and deny as moot the motions to strike certain material from the record and to supplement the record with other material. 80
    169. UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT August Term, 2004 (Argued: January 5, 2005 Decided: February 16, 2006) Docket No. 04-0743-cv _____________________________________________ JOHN PAUL HANKINS, Plaintiff-Appellant, v. ERNEST S. LYGHT and NEW YORK ANNUAL CONFERENCE OF THE UNITED METHODIST CHURCH, Defendants-Appellees, STONY BROOK COMMUNITY CHURCH, Defendant. ____________________________________ SOTOMAYOR, Circuit Judge, dissenting: The Religious Freedom and Restoration Act (“RFRA”) is not relevant to this dispute. First, appellees have unambiguously indicated that they do not seek to raise a RFRA defense, and the statute’s protections, even if otherwise applicable, are thus waived. Second, the statute does not apply to disputes between private parties. Third, we should affirm the judgment of the district court without reaching the RFRA issue on the ground that Supreme Court and Second Circuit precedent compels a finding that the Age Discrimination in Employment Act (ADEA), 29 U.S.C. § 621 et seq., does not govern disputes between a religious entity and its spiritual leaders. 1
    170. The majority’s opinion thus violates a cardinal principle of judicial restraint by reaching unnecessarily the question of RFRA’s constitutionality. For these reasons, I respectfully dissent. A. Because the parties’ original submissions to this Court mentioned RFRA without providing a detailed analysis of either the Act’s constitutionality or its relevance to this case, we ordered supplemental briefing. The letter-briefs submitted in response to our order make clear that appellees have waived any RFRA defense. In several portions of appellees’ supplemental brief that the majority neglects to mention, appellees state plainly that they do not intend to raise a RFRA defense. Appellees’ supplemental brief explains that “the reference to RFRA in Appellees’ [original] brief was for the limited purpose of providing an example of how critically the question of ‘entanglement’ was viewed” by Congress. In other words, appellees’ aim was not to rely on the statute as a defense against appellant’s claims, but merely to illustrate Congress’s agreement with the proposition that “entanglement of the Government in church affairs [was] prohibited by the First Amendment.” (emphasis added). In fact, appellees explicitly reject the application of RFRA to their claims because they believe that the statute does not apply to suits between private parties, and “the case at bar is a matter relating to a private employment situation and does not involve actions by the government.” The letter-brief concludes: “We do not think this issue [RFRA] is determinative in the matters raised by this case.” While the majority might find appellees’ position unwise or “supris[ing],” Maj. Op. at 12, appellees’ letter-brief clearly waives any RFRA defense.1 1 In addition to ignoring most of the language in the appellees’ brief relating to waiver, the majority opinion makes two factually erroneous claims regarding the content of the supplemental letter-briefs. First, the majority writes that “appellant argues that the RFRA is 2
    171. The majority does not contest that RFRA’s protections are generally waivable. Maj. Op. at 13; see United States v. Amer, 110 F.3d 873, 879 n.1 (2d Cir. 1997); see also In re Watson, 403 F.3d 1, 7 (1st Cir. 2005) (holding that RFRA argument was forfeited); Bethesda Lutheran Homes & Servs., Inc. v. Leean, 122 F.3d 443, 449 (7th Cir. 1997) (holding that RFRA argument was waived); Cochran v. Morris, 73 F.3d 1310, 1317 n.3 (4th Cir. 1996) (holding that RFRA claim was waived). In the majority’s view, however, because appellees’ arguments relate to rights protected under RFRA—namely, First Amendment religious rights—appellees have “[i]n substance” relied on RFRA and thus have not, despite their explicit disclaimer, waived its protections. Maj. Op. at 14. The majority’s refusal to recognize appellees’ waiver in this case is at odds with RFRA’s text, which provides that individuals “may assert” a RFRA defense when challenging a substantial burden on their religious rights, not that they must assert a RFRA defense when religious rights are burdened. 42 U.S.C. § 2000bb-1(c) (emphasis added). Moreover, the majority’s insistence on the viability of a RFRA defense despite appellees’ waiver leads the Court to assess RFRA’s constitutionality unnecessarily. See Cutter v. Wilkinson, 125 S. Ct. 2113, 2118 n.2 (2005) (noting that the Supreme Court has “not had occasion to rule” whether RFRA “remains operative as to the Federal Government”); see also City of Boerne v. Flores, 521 inapplicable only because it is unconstitutional.” Maj. Op. at 12. On the contrary, appellant argues also that RFRA is inapplicable because “[t]here is no substantial burden to the free exercise of religion that could result from a ruling by this court” that appellees violated the ADEA. The majority also contends that “[t]he parties have not briefed the issue of how [RFRA] impacts the merits of this case.” Maj. Op. at 24. Both parties, however, have addressed the question of RFRA’s relevance to this suit. Appellant argues that RFRA is inapplicable because “[t]here is no substantial burden to the free exercise of religion” in this case, and that, in any event, “application of RFRA to federal law is unconstitutional.” Appellees, in turn, argue that RFRA is constitutional but should not affect the outcome of this case. 3
    172. U.S. 507, 532-36 (1997) (invalidating RFRA as applied to state law). By going out of its way to reach this constitutional question, the majority violates one of the “cardinal rules governing the federal courts,” namely, “never to anticipate a question of constitutional law in advance of the necessity of deciding it.” Brockett v. Spokane Arcades, Inc., 472 U.S. 491, 501 (1985) (citation and internal quotation marks omitted).2 The majority’s approach is also inconsistent with our case law, which has recognized waiver of statutory religious rights even where a litigant raises claims under the Free Exercise Clause. In Fifth Avenue Presbyterian Church v. City of New York, 293 F.3d 570 (2d Cir. 2002), for example, the plaintiff argued before this Court that its religious rights had been violated under both the First Amendment and the Religious Land Use and Institutionalized Persons Act (RLUIPA)—a statute virtually identical to RFRA in all aspects relevant to the issue of waiver in the instant case.3 Although we ruled on the merits of the plaintiff’s Free Exercise claim in Fifth Avenue Presbyterian Church, we refused to reach the RLUIPA issue because the plaintiff had raised it for the first time on appeal. See id. at 576. It is impossible to square our refusal to 2 The other “cardinal rule[]” cited in Brockett is that federal courts should “never . . . formulate a rule of constitutional law broader than is required by the precise facts to which it is to be applied.” 472 U.S. at 501 (citation and internal quotation marks omitted). 3 Like RFRA, RLUIPA prohibits the government from imposing substantial burdens on religion even where the burden results from a neutral law of general applicability. See 42 U.S.C. § 2000cc. RLUIPA’s remedial provision is virtually identical to RFRA’s. Compare 42 U.S.C. § 2000cc-2(a) (“A person may assert a violation of this Act as a claim or defense in a judicial proceeding and obtain appropriate relief against a government.”), with 42 U.S.C. § 2000bb-1(c) (“A person whose religious exercise has been burdened in violation of this section may assert that violation as a claim or defense in a judicial proceeding and obtain appropriate relief against a government.”). RLUIPA is simultaneously more broad and more narrow than RFRA, however. RLUIPA is more broad because it still reaches state law. See Cutter v. Wilkinson, 125 S. Ct. 2113 (2005). It is more narrow because it applies only to certain government actions involving land use regulations and correctional facilities. See 42 U.S.C. §§ 2000cc, 2000cc-1. 4
    173. consider plaintiff’s belated RLUIPA claim in Fifth Avenue Presbyterian Church with our refusal to recognize the defendant’s voluntary waiver of a RFRA defense in the instant case. There is no meaningful difference between RFRA and RLUIPA that could justify such inconsistent results. The most troublesome aspect of the majority’s ruling on waiver, however, is that it fundamentally misconstrues the nature of RFRA and First Amendment rights, and, in doing so, directly contradicts Supreme Court precedent. The majority holds that because appellees invoke the First-Amendment-based “ministerial exception” and allege interference with their rights under the Free Exercise and Establishment Clauses, they have effectively “ask[ed] us to apply the RFRA, but not to mention it.” Maj. Op. at 14. This is incorrect. RFRA and the First Amendment do not provide identical protections, and the invocation of First Amendment rights—whether under the Free Exercise or the Establishment Clause—does not necessarily implicate RFRA. As interpreted by the Supreme Court, for example, the Free Exercise Clause does not normally “inhibit enforcement of otherwise valid laws of general application that incidentally burden religious conduct,” Cutter, 125 S. Ct. at 2118 (citing Employment Div., Dep’t of Human Res. v. Smith, 494 U.S. 872, 878-82 (1990)), such as the ADEA. RFRA, in contrast, requires strict scrutiny of such laws where the incidental burden on religion is substantial. See 42 U.S.C. § 2000bb-1. Indeed, the fact that RFRA’s protections sweep more broadly than those of the Free Exercise Clause provided the principal basis for the Supreme Court’s holding in City of Boerne v. Flores that RFRA could not be considered “preventive” or “remedial” legislation under Section Five of the Fourteenth Amendment. 521 U.S. at 532. The Court found RFRA’s protections “so out of proportion to a supposed remedial or preventive object that [the statute] 5
    174. cannot be understood as responsive to, or designed to prevent, unconstitutional behavior.” Id. Because RFRA went so far beyond what the First Amendment required, the Boerne Court understood the statute as “attempt[ing] a substantive change in constitutional protections”—a change that Congress was not authorized to make. Id. Although Boerne does not resolve the issue of RFRA’s constitutionality as applied to federal law, as opposed to state law,4 the case does firmly establish that RFRA and the Free Exercise Clause create different standards for the protection of religion and that RFRA’s substantive protections extend far beyond what the Free Exercise Clause requires. Thus, the majority’s suggestion that a claim alleging unconstitutional interference with the free exercise of religion is “[i]n substance” a RFRA claim flies in the face of Boerne.5 See also Kaufman v. McCaughtry, 419 F.3d 678, 681 (7th Cir. 2005) (noting that RFRA provides protections beyond those guaranteed by the First Amendment); Brzonkala v. Va. 4 I express no view on whether RFRA is constitutional as applied to federal law because it is unnecessary for us to reach this question. 5 Before Boerne, a reasonable argument could have been made that all Free Exercise Clause claims required scrutiny under RFRA. The Tenth Circuit, for example, held in Werner v. McCotter, 49 F.3d 1476 (10th Cir. 1995), that RFRA applied to all Free Exercise claims, even where the parties had not raised a claim or defense under the statute. In a subsequent en banc opinion, however, the Tenth Circuit recognized that Boerne had undermined its earlier conclusion: [I]n Werner, decided prior to City of Boerne, we were laboring under the false understanding that RFRA “legislatively overturned a number of recent Supreme Court [free exercise] decisions” and that it created a new rule of constitutional law. Thus, we concluded that because the language of RFRA made it applicable to “all cases where free exercise of religion is substantially burdened,” its standard ought to control a Free Exercise Clause claim even when not raised. Because the Supreme Court has made clear that the Werner court’s assumptions about RFRA were faulty, its rationale is no longer convincing. United States v. Hardman, 297 F.3d 1116, 1125 n.15 (10th Cir. 2002) (en banc) (alteration in original) (citations omitted). 6
    175. Polytech. Inst. & State Univ., 169 F.3d 820, 881-82 (4th Cir. 1999) (“The [RFRA] created a right of religious exercise that was more generous than that right protected by the Constitution . . . .”), aff’d sub nom. United States v. Morrison, 529 U.S. 598 (2000). Nor can the majority plausibly argue that appellees’ Establishment Clause defense necessarily implicates RFRA. To satisfy the Establishment Clause: (1) the statute must have “a secular legislative purpose”; (2) the statue’s “principal or primary effect must be one that neither advances nor inhibits religion”; and (3) “the statute must not foster an excessive government entanglement with religion.” Lemon v. Kurtzman, 403 U.S. 602, 612-13 (1971) (citation and internal quotation marks omitted). Thus, like the Free Exercise Clause, the Establishment Clause imposes less stringent requirements on federal statutes than RFRA, which mandates strict scrutiny even of neutral, generally applicable laws that incidentally impose substantial burdens on religion.6 Furthermore, Congress made clear in enacting RFRA that the statute was not intended to have any effect on Establishment Clause claims. See 42 U.S.C. § 2000bb-4 (“Nothing in this chapter shall be construed to affect, interpret, or in any way address that portion of the First Amendment prohibiting laws respecting the establishment of religion.”). The majority’s assertion that appellees have presented a RFRA defense in “all but name” would be more plausible if something in appellees’ briefs indicated that they sought protection beyond that which the Constitution guarantees. Nothing in the briefs, however, supports such a conclusion. Appellees’ briefs rely heavily on the Free Exercise Clause, the Establishment 6 As recently emphasized by a plurality of Justices, the Supreme Court has not applied the Lemon test with much consistency. See Van Orden v. Perry, 125 S. Ct. 2854, 2860-61 (2005) (plurality opinion). I am unaware of any application of the Establishment Clause, however, that would invalidate a neutral, generally applicable law imposing an incidental but substantial burden on religion. 7
    176. Clause, and case law interpreting those provisions. Nowhere do they ask that the Court apply a standard stricter than what the First Amendment requires.7 On the contrary, appellees’ supplemental brief explicitly disclaims any intent to rely on RFRA. In sum, because appellees’ religious freedom argument relies only on the Free Exercise and Establishment Clauses, and because the substance of the protections afforded by these constitutional provisions differs considerably from the protections afforded by RFRA, as interpreted by the Supreme Court, I cannot agree with the majority’s conclusion that appellees have “[i]n substance” relied on RFRA. Maj. Op. at 14. The majority’s refusal to recognize appellees’ clear waiver of any RFRA defense appears to rest, in part, on its disagreement with the reasons underlying appellees’ decision not to pursue such a defense. Specifically, the majority takes issue with appellees’ conclusion that RFRA does not apply to suits between private parties. See Maj. Op. at 10-11. I am unaware of any other case in which this Court, after ordering supplemental briefing to allow a party to discuss a waivable statutory defense, refused to recognize the party’s subsequent waiver of that defense on the ground that the Court disagreed with counsel’s reasons for declining to rely on the statute. Cf. DeLuca v. Lord, 77 F.3d 578, 588 (2d Cir. 1996) (observing that where defense counsel in a 7 The closest appellees come to making a RFRA argument, as opposed to a First Amendment argument, is a statement in their original brief that application of the ADEA would “substantially burden the free exercise rights of the United Methodist Church.” This is the only occasion, however, in which appellees employ RFRA-like language by referring to the alleged intrusion on their rights as a “substantial[] burden,” and it is clear from context that the statement formed part of appellees’ Establishment Clause argument that application of the ADEA would foster an excessive entanglement with religion. The brief did not purport to raise a separate defense under RFRA. In any event, even if appellees’ mention of a “substantial[] burden” in their original brief could be generously construed as an attempt to present a RFRA defense, appellees’ subsequent letter-brief makes clear that this was not their intent and that they do not seek to invoke RFRA’s protections. 8
    177. criminal case has made “a considered decision, after investigation, not to pursue” a particular defense, this Court should be “extremely reluctant to second-guess that decision”). Even if such second-guessing of a party’s decision not to pursue a particular defense is appropriate in certain limited circumstances, the majority’s refusal to acknowledge the clear waiver in this case is improper, given that appellees are adequately represented by counsel and based their waiver on a reasonable interpretation of the law. Indeed, the majority concedes that it is unable to find a single holding that contradicts appellees’ view that RFRA does not apply to suits between private parties. See Maj. Op. at 25 n.4. Quoting the Supreme Court, the majority argues that “[w]hen an issue or claim is properly before the court, the court is not limited to the particular legal theories advanced by the parties, but rather retains the independent power to identify and apply the proper construction of governing law.” Maj. Op. at 14 (quoting Kamen v. Kemper Fin. Servs., Inc., 500 U.S. 90, 99 (1991)). This certainly is true, but it only begs the question of whether the “issue or claim is properly before the court.” Id. Given appellees’ clear indication that they do not seek to rely on RFRA, the applicability of that statute is not before us. The majority’s disagreement with appellees’ reasoning does not change that fact. B. Even assuming, arguendo, that appellees’ clear disclaimer of any RFRA defense does not suffice to waive such a defense, I would find it improper to remand the case to the district court for consideration of RFRA’s implications because I disagree with the majority’s conclusion regarding RFRA’s applicability. RFRA by its terms does not apply to suits between private parties. 9
    178. Two provisions of the statute implicitly limit its application to disputes in which the government is a party. Section 2000bb-1(c) states that “[a] person whose religious exercise has been burdened in violation of this section may assert that violation as a claim or defense in a judicial proceeding and obtain appropriate relief against a government” (emphasis added). In the majority’s view, we should read this provision as “broadening, rather than narrowing, the rights of a party asserting the RFRA.” Maj. Op. at 11. This interpretation would be questionable even if Section 2000bb-1(c) were the only provision of the statute affecting the question of whether RFRA applies to private suits. When read in conjunction with the rest of the statute, however, it becomes clear that this section reflects Congress’s understanding that RFRA claims and defenses would be raised only against the government. For instance, section 2000bb-1(b) of RFRA provides that where a law imposes a substantial burden on religion, the “government” must “demonstrate[] . . . that application of the burden” is the least restrictive means of furthering a compelling governmental interest (emphasis added). The statute defines “demonstrate” as “meet[ing] the burdens of going forward with the evidence and of persuasion.” 42 U.S.C. § 2000bb-2(3). Where, as here, the government is not a party, it cannot “go[] forward” with any evidence.8 In my view, this provision strongly suggests that Congress did not intend RFRA to 8 There are two plausible ways to reconcile section 2000bb-1(b) of RFRA with the majority opinion in this case. The first would be to require government intervention in every private suit where one of the parties asserts that a law has—even incidentally—imposed a substantial burden on religious freedom. Absent a clear statement that Congress intended such a result, it is not the role of this Court to mandate such widespread and automatic federal intervention in lawsuits between private parties. Moreover, were we to read the statute to require government intervention, this would surely underscore the wisdom in recognizing appellees’ explicit waiver of any RFRA defense. The second would be to force private parties to bear the burden RFRA places on the government. The statute gives no indication that Congress intended private parties to bear such a burden, nor would it be appropriate to require private parties to satisfy the stringent burden RFRA places on the government. 10
    179. apply in suits between private parties.9 I recognize that according to RFRA’s “applicability” section, the statute applies “to all Federal law.” 42 U.S.C. § 2000bb-3. This provision, however, is not inconsistent with a finding that the statute does not apply to suits between private parties. Read in conjunction with the rest of the statute, the provision simply requires courts to apply RFRA “to all Federal law” in any lawsuit to which the government is a party. The majority objects that this interpretation makes RFRA’s protections improperly dependent on whether a private party, as opposed to the EEOC, brings suit under the ADEA. “[T]he substance of the ADEA’s prohibitions,” the majority argues, “cannot change depending on whether it is enforced by the EEOC or an aggrieved private party.” Maj. Op. at 11. The majority does not explain, however, why this is so. If RFRA amends all federal statutes as they apply to suits in which the government is a party, then the substance of the ADEA’s prohibitions most certainly can change depending on who enforces it. Although the majority evidently finds this unsatisfactory from a policy perspective, there is no acceptable reading of the statute that would yield the kind of consistency the majority desires. Finally, as noted above, the majority concedes that it is unable to locate a single court holding that directly supports its novel application of RFRA to a suit between private parties. 9 All of the examples cited in the Senate and House Reports on RFRA involve actual or hypothetical lawsuits in which the government is a party. See S. Rep. No. 103-111 (1993); H.R. Rep. 103-88 (1993). The lack of even a single example of a RFRA claim or defense in a suit between private parties in these Reports tends to confirm what is evident from the plain language of the statute: It was not intended to apply to suits between private parties. 11
    180. See Maj. Op. at 25 n.4.10 This is telling, for Congress enacted RFRA over twelve years ago. The plain language of the statute, its legislative history, and its interpretation by courts over the past twelve years demonstrate that RFRA does not apply to suits between private parties. C. Even if appellees had not waived the RFRA defense, and even if RFRA applied to suits between private parties, I would still find it unnecessary to reach the RFRA issue, or to analyze the statute’s constitutionality, because Supreme Court and Second Circuit precedent compel the conclusion that the ADEA does not apply to this dispute. Because the ADEA does not apply, there is no “substantial burden” on religion, and RFRA, even if constitutional, is irrelevant. In analyzing the ADEA’s applicability to this case, we find guidance in the principles articulated by the Supreme Court in NLRB v. Catholic Bishop of Chicago, 440 U.S. 490 (1979). To determine whether the National Labor Relations Act (NLRA) authorized the National Labor Relations Board to regulate labor relations between a parochial school and its faculty, the Catholic Bishop Court considered two principal questions. See id. at 501. First, it considered whether this application of the NLRA raised First Amendment concerns. The Court concluded that it did, explaining that judicial oversight of labor relations at a parochial school would risk excessive entanglement between secular and religious authorities in violation of the Establishment Clause. Id. at 501-04. Second, the Court examined whether Congress expressed an intention to apply the statute to religious institutions despite these constitutional concerns. Because the Court discerned no such congressional intent, it construed the NLRA in a manner 10 The majority cites dicta from district court opinions in Indiana and Arizona but concedes that those courts “assumed” that RFRA could apply without analyzing the issue in any depth. See Maj. Op. at 25-26 n.4. 12
    181. that avoided the constitutional difficulty, holding that the statute did not apply to labor disputes between parochial schools and their employees.11 Id. at 504-07; see id. at 500 (citing the longstanding principle that acts of Congress “ought not be construed to violate the Constitution if any other possible construction remains available”) (citing Murray v. The Charming Betsy, 6 U.S. (2 Cranch) 64, 118 (1804)); see also Hsu By & Through Hsu v. Roslyn Union Free Sch. Dist. No. 3, 85 F.3d 839, 854 (2d Cir. 1996) (noting this Court’s “consistent . . . practice of avoiding constitutional questions where possible”). Distinguishing Catholic Bishop, we concluded in DeMarco v. Holy Cross High Sch., 4 F.3d 166 (2d Cir. 1993), that the ADEA, unlike the NRLA, generally applies to religious institutions. Id. at 172. Specifically, we held that a former lay teacher could bring an ADEA action against a parochial school even though the teacher performed some religious duties. Id. at 168-72. In so holding, we observed that the ADEA, unlike the NRLA, does not pose the risk of “extensive or continuous administrative or judicial intrusion into the functions of religious institutions.” Id. at 170. Instead, the ADEA involves “‘routine regulatory interaction’” and requires “‘no inquiries into religious doctrine, no delegation of state power to a religious body, and no detailed monitoring [or] close administrative contact between secular and religious bodies.’” Id. at 170 (quoting Hernandez v. Comm’r, 490 U.S. 680, 696-97 (1989) (internal quotation marks omitted)); see also id. (“In age discrimination cases, the EEOC’s authority extends only to the investigation and attempted conciliation or resolution of individual or group complaints; it is limited in time and scope.” (citation and internal quotation marks omitted)). 11 The Court reached this conclusion even though the NLRA did not expressly include religious institutions in its list of eight types of employers exempted from the act. See Catholic Bishop, 440 U.S. at 511 (Brennan, J., dissenting) (citing 29 U.S.C. § 152(2)). 13
    182. These factors distinguished the ADEA from the NLRA.12 As a general rule, federal courts may decide civil disputes, including employment discrimination disputes, between a religious institution and its employees without violating the First Amendment. See Merkos L’Inyonei Chinuch, Inc. v. Otsar Sifrei Lubavitch, Inc., 312 F.3d 94, 99-100 (2d Cir. 2002) (citing Martinelli v. Bridgeport Roman Catholic Diocesan Corp., 196 F.3d 409, 431 (2d Cir. 1999); Gargano v. Diocese of Rockville Ctr., 80 F.3d 87, 90 (2d Cir. 1996); DeMarco, 4 F.3d at 172; cf. Employment Div., Dep’t of Human Res. v. Smith, 494 U.S. 872, 879 (1990) (“[T]he right of free exercise does not relieve an individual of the obligation to comply with a valid and neutral law of general applicability on the ground that the law proscribes (or prescribes) conduct that his [or her] religion prescribes (or proscribes).” (internal quotation marks omitted)). The instant case, however, presents the more difficult question of whether this general rule applies in the narrow context of a forced-retirement dispute between a religious body and a member of its clergy. As we noted in DeMarco, the relationship between a religious institution and certain of its employees may be “so pervasively religious that it is impossible to engage in an age- discrimination inquiry without serious risk of offending the Establishment Clause.” Id. at 172. This risk is particularly serious in employment disputes between religious institutions and their spiritual leaders where the enforcement of statutes like the ADEA might threaten the “power of religious bodies to decide for themselves, free from state interference, matters of church government as well as those of faith and doctrine.” Serbian E. Orthodox Diocese for the U.S. & 12 As discussed below, DeMarco also found the ADEA distinguishable from the NLRA because Congress clearly intended the ADEA to apply to religious institutions. See 4 F.3d at 172. 14
    183. Can. v. Milivojevich, 426 U.S. 696, 722 (1976) (internal quotation marks and alteration omitted). “A church’s selection of its own clergy” is a “core matter of ecclesiastical self-governance” at the “heart” of the church’s religious mission. Bollard v. Cal. Province of the Soc’y of Jesus, 196 F.3d 940, 946 (9th Cir. 1999). Federal court entanglement in matters as fundamental as a religious institution’s selection or dismissal of its spiritual leaders risks an unconstitutional “trespass[] on the most spiritually intimate grounds of a religious community’s existence.” EEOC v. Roman Catholic Diocese of Raleigh, N.C., 213 F.3d 795, 800 (4th Cir. 2000). In light of these serious constitutional concerns, we must ask whether Congress intended to apply the ADEA to religious institutions in their selection of spiritual leaders. See Catholic Bishop, 440 U.S. at 504. We concluded in DeMarco that Congress “implicitly expressed an intention to apply the ADEA to religious institutions.” 4 F.3d at 172. We based this conclusion, in part, on the ADEA’s similarity to Title VII of the Civil Rights Act, 42 U.S.C. § 2000e et seq. “Given that Congress intended to apply Title VII to religious institutions, and that Congress modelled the ADEA’s coverage upon that of Title VII,” we were “convinced that [Congress] also intended to apply the ADEA to such institutions.” Id. at 173. DeMarco, however, involved an employment dispute between a religious institution and a math teacher who, despite having some religious duties, served primarily non-religious functions in a parochial school. Here, in contrast, the dispute is between a minister with primarily religious duties and a church that no longer wishes him to serve as pastor of a congregation. That Congress intended the ADEA and Title VII to apply under the circumstances described in DeMarco does not indicate an intention that those statutes should apply in all circumstances. Nothing in the text, structure, or legislative history of the ADEA indicates an intention to extend 15
    184. its provisions to a religious body’s selection or dismissal of its ministers. See Catholic Bishop, 440 U.S. at 504; DeMarco, 4 F.3d at 169, 172-73. Accordingly, I believe that the ADEA does not apply to the case at bar.13 Because the ADEA does not apply, there is no substantial burden on religion that could trigger RFRA. The majority suggests that reliance on Catholic Bishop (and DeMarco) is unwarranted, because “RFRA [is] the full expression of Congress’s intent with regard to the religion-related issues before us and displace[s] earlier judge-made doctrines that might have been used to ameliorate the ADEA’s impact on religious organizations and activities.” Maj. Op. at 8. Even if RFRA applied to private suits and had not been waived in this case, I would disagree with the majority’s suggestion that the statute completely displaces the Catholic Bishop analysis. Although the Catholic Bishop rule and RFRA serve similar purposes, they require courts to undertake different inquiries. See Univ. of Great Falls v. NLRB, 278 F.3d 1335, 1347 (D.C. Cir. 13 This conclusion is consistent with the holdings of at least seven of our sister Circuits, which have adopted a limited “ministerial exception” that exempts religious institutions on First Amendment grounds from employment discrimination suits brought by clergy members or other employees serving primarily religious roles. See Roman Catholic Diocese of Raleigh, N.C., 213 F.3d at 800, 805; Gellington v. Christian Methodist Episcopal Church, Inc., 203 F.3d 1299, 1304 (11th Cir. 2000); Bollard, 196 F.3d at 949; Combs v. Central Tex. Annual Conf. of the United Methodist Church, 173 F.3d 343, 351 (5th Cir. 1999); EEOC v. Catholic Univ. of Am., 83 F.3d 455, 463 (D.C. Cir. 1996); Young v. N. Ill. Conf. of United Methodist Church, 21 F.3d 184, 187 (7th Cir. 1994); Scharon v. St. Luke’s Episcopal Presbyterian Hosps., 929 F.2d 360, 363 (8th Cir. 1991). Most circuits have reached the constitutional question directly and have held that the First Amendment bars adjudication of ministerial employment disputes. See, e.g., Gellington, 203 F.3d at 1304; Combs; 173 F.3d at 351; Young, 21 F.3d at 187. Here, in contrast, I would apply Catholic Bishop’s principles of statutory construction so as to avoid making definitive pronouncements on the constitutional question. See 440 U.S. at 507; see also Scharon, 929 F.2d at 361-63 (applying the Catholic Bishop analysis to an employment discrimination action brought by a priest). Despite this difference, my conclusion is substantially the same as that of other Circuits: courts may not adjudicate employment discrimination lawsuits brought by clergy members challenging a religious body’s refusal to select or retain them as spiritual leaders. 16
    185. 2002) (holding that the court need not address a university’s RFRA argument because the university was entitled to an exemption under Catholic Bishop, and observing that “RFRA presents a separate inquiry from Catholic Bishop”). Catholic Bishop requires courts, where possible, to interpret statutes in ways that would avoid raising serious constitutional concerns. In some cases, no such interpretation will be reasonably available. In those cases, RFRA may provide an independent avenue both for protecting religious rights and for avoiding definitive resolution of constitutional questions. Thus, RFRA should not be read to supplant the Catholic Bishop inquiry, but to supplement it. Indeed, given that RFRA’s express purpose was to enhance protection for religion, see 42 U.S.C. § 2000bb, it makes little sense to read the statute as eliminating the protection afforded by the Catholic Bishop rule. D. I believe that a remand is a wasteful expenditure of judicial resources and an unnecessary and uninvited burden on the parties. The district court is in no better position than we are to decide either the statutory or constitutional questions presented in this case. In my view, the most appropriate disposition of this case would be to affirm the district court’s dismissal of appellant’s claims on the ground that the ADEA does not apply to employment suits brought against religious institutions by their spiritual leaders. Because the majority’s contrary approach disregards a clear and voluntary waiver, conflicts with RFRA’s text and with binding precedent, and unnecessarily resolves a contested constitutional question, I respectfully dissent.14 14 I take no issue, however, with the analysis of the ADEA’s procedural requirements in section (a) of the majority’s opinion. See Maj. Op. at 5-8. 17
    186. 1 UNITED STATES COURT OF APPEALS 2 3 FOR THE SECOND CIRCUIT 4 5 August Term, 2004 6 7 (Argued: January 5, 2005 Decided: February 16, 2006) 8 9 Docket No. 04-0743-cv 10 11 12 13 JOHN PAUL HANKINS, 14 15 Plaintiff-Appellant, 16 17 v. 18 19 ERNEST S. LYGHT and NEW YORK ANNUAL CONFERENCE OF THE UNITED 20 METHODIST CHURCH, 21 22 Defendants-Appellees, 23 24 STONY BROOK COMMUNITY CHURCH, 25 26 Defendant. 27 28 29 30 B e f o r e: WINTER, SOTOMAYOR, and B.D. PARKER, Circuit 31 Judges. 32 33 Appeal from the dismissal of a minister’s age discrimination 34 action against his church in the Eastern District of New York 35 (Denis R. Hurley, Judge). We hold that the Religious Freedom 36 Restoration Act of 1993 is constitutional as applied to federal 37 law. It therefore amended the ADEA and governs the merits of 38 this action. We vacate and remand for reconsideration in light 39 of the RFRA. 40 Judge Sotomayor dissents in a separate opinion. 41 1
    187. 1 BRUCE MILES SULLIVAN, Stony Brook, 2 New York, for Plaintiff-Appellant. 3 4 FREDERICK K. BREWINGTON, Hempstead, 5 New York, for Defendants-Appellees. 6 7 WINTER, Circuit Judge: 8 9 John Paul Hankins appeals from the dismissal by Judge Hurley 10 of his age discrimination action. Hankins was a clergy member 11 ordained by appellee New York Annual Conference of the United 12 Methodist Church ("NYAC"). He was forced into retirement when he 13 attained the age of 70. Appellee Ernest S. Lyght is the Bishop 14 of the NYAC and has the power to appoint clergy to NYAC churches. 15 Hankins claims that the NYAC’s mandatory retirement policy 16 violates the Age Discrimination in Employment Act of 1967 17 ("ADEA"), 29 U.S.C. § 621 et seq. We hold that the Religious 18 Freedom Restoration Act of 1993 ("RFRA"), 42 U.S.C. § 2000bb et 19 seq., is constitutional as applied to federal law; it therefore 20 amended the ADEA and governs the merits of the principal issue 21 raised by the parties. We vacate the dismissal of Hankins’ 22 complaint and remand for a determination of whether application 23 of the ADEA to Hankins’ relationship with the NYAC and Lyght 24 violates the RFRA. 25 BACKGROUND 26 We assume the existence of the facts as alleged in the 27 complaint. Hankins was ordained by the NYAC and served as a 28 clergy member from 1962 to July 1, 2003. He turned 70 on 2
    188. 1 November 5, 2002, and was forced into retirement on July 1, 2 2003, as prescribed by paragraph 356 of the Methodist Book of 3 Discipline. 4 According to a statement by the Methodist Church's Council 5 of Bishops, the Book of Discipline is neither "sacrosanct" nor 6 "infallible, but . . . is the most current statement of how 7 United Methodists agree to live together" as "an inclusive 8 society without regard to ethnic origin, economic condition, 9 gender, age, or the disabilities of its constituents." The 10 complaint alleges that the Book of Discipline contains "subject 11 matters that are sectarian and ecclesiastical in nature[,] being 12 related to the nature of the Deity and the Trinity, the 13 scriptures, the tenets of the United Methodist Church, the 14 theological grounding of biblical faith, the teachings of John 15 Wesley and/or other religious principles or values (. . . 16 ‘religious considerations')," as well as "subject matters that 17 are secular, temporal and/or civil in nature[,] not being 18 determined, controlled or influenced by any religious 19 considerations." The complaint further claims that paragraph 20 356, under which Hankins was mandatorily retired, "is a secular, 21 temporal, and/or civil subject matter, not being determined, 22 controlled or influenced by any religious considerations." 23 Bishop Lyght told Hankins and other members of the Church 24 that he had the authority to reappoint Hankins as pastor, despite 3
    189. 1 the fact that Hankins is over 70 years old. However, Bishop 2 Lyght also stated that it is his "personal policy (as 3 distinguished from the policy set forth in the Book of 4 Discipline) never to reappoint members of the clergy who have 5 attained age seventy to the church out of which they were 6 retired." 7 Appellant brought an age discrimination charge to the Equal 8 Employment Opportunity Commission ("EEOC") on March 19, 2003. 9 The EEOC issued a Notice of Right to Sue on April 11, 2003. 10 Appellant also filed a Verified Complaint with the New York 11 Division of Human Rights on June 11, 2003; that Complaint was 12 dismissed for administrative convenience on July 1, 2003. 13 Appellant filed the instant suit on July 3, 2003. 14 Appellant’s complaint claimed that the mandatory retirement 15 policy violated the ADEA, the New York Human Rights Law, and the 16 NYAC's covenant with him (Counts I, II, and IV); and that Bishop 17 Lyght's personal policy against reappointing retired clergy 18 violated the ADEA and Human Rights Law (Count III).1 19 Appellees moved to dismiss for lack of subject matter 20 jurisdiction and for failure to state a claim upon which relief 21 could be granted, under Rules 12(b)(1) and (6) respectively. The 22 district court, ruling orally, declined to decide the 12(b)(1) 23 motion, which was apparently based on deficiencies in the EEOC’s 24 review of appellant’s charge. Instead, the court granted the 4
    190. 1 12(b)(6) motion based on a “ministerial exception” to the ADEA -- 2 a rule adopted by several circuits that civil rights laws cannot 3 govern church employment relationships with ministers without 4 violating the free exercise clause because they substantially 5 burden religious freedom. See, e.g., McClure v. Salvation Army, 6 460 F.2d 553, 560 (5th Cir. 1972) (applying Title VII to church- 7 minister relationship "would result in an encroachment by the 8 State into an area of religious freedom into which it is 9 forbidden to enter" by the Free Exercise Clause). The court 10 dismissed the complaint under Rule 12(b)(6). 11 DISCUSSION 12 Appellant argues that the ministerial exception should not 13 insulate a church’s non-religious regulations that discriminate 14 against ministers on the basis of age. Appellees assert that 15 this action is barred by EEOC errors. Alternatively, they 16 continue to rely upon "the ministerial exception," the Free 17 Exercise clause, and the Establishment Clause, claiming that 18 applying the ADEA to the church-minister relationship would 19 substantially burden religion. In that regard, appellees note 20 that "for this very reason" Congress passed the RFRA. We address 21 the alleged EEOC errors before turning to the main issue: 22 whether the RFRA amended the ADEA. 23 a) Completion of Administrative Proceedings 24 Appellees argue that the district court lacked jurisdiction 5
    191. 1 because the EEOC issued appellant’s Notice of Right to Sue fewer 2 than sixty days after his charge was filed.2 We disagree. 3 Appellant satisfied all statutory requirements for bringing 4 this private action under the ADEA. He filed an age 5 discrimination charge with the EEOC on March 19, 2003; the EEOC 6 issued a Notice of Right to Sue on April 11, 2003. Under 29 7 U.S.C. § 626(d) and (e), appellant had to file the instant suit 8 more than sixty days after filing his EEOC complaint and within 9 ninety days of his receipt of the EEOC Notice. Hankins complied 10 with both requirements by filing suit on July 3, 2003 -- more 11 than 60 days after March 19, and 83 days after April 11. 12 Furthermore, contrary to appellees’ arguments, the instant suit 13 was not barred by appellant’s June 11, 2003 filing of a Complaint 14 with the New York Division of Human Rights because the Division 15 dismissed the complaint on July 1, 2003, before appellant filed 16 this suit. See 29 U.S.C. § 633(b) (ADEA prohibits bringing suit 17 before 60 days after commencement of state proceedings, "unless 18 such proceedings have been earlier terminated"). 19 Appellees rely for their jurisdictional contention on two 20 Title VII cases: Martini v. Fed. Nat’l Mortgage Ass’n, 178 F.3d 21 1336 (D.C. Cir. 1999), and Rodriguez v. Connection Tech. Inc., 65 22 F. Supp. 2d 107 (E.D.N.Y. 1999). These cases inferred from the 23 language of 42 U.S.C. § 2000e-5(f)(1)3 that the EEOC lacks 24 authority to issue right-to-sue notices based on Title VII claims 6
    192. 1 before 180 days after a charge is filed. E.g., Martini, 178 F.3d 2 at 1347 (“[T]he EEOC’s power to authorize private suits within 3 180 days undermines its express statutory duty to investigate 4 every charge filed, as well as Congress’s unambiguous policy of 5 encouraging informal resolution of charges up to the 180th 6 day.”). We have not decided whether the regulation allowing 7 early issuance of right-to-sue notices, 29 C.F.R. § 8 1601.28(a)(2), is a permissible construction of Section 2000e-5. 9 We express no opinion on the issue here, although we note that 10 two circuits and several district courts within this circuit have 11 disagreed with Martini and Rodriguez. Sims v. Trus Joist 12 MacMillan, 22 F.3d 1059, 1061-63 (11th Cir. 1994) (early issuance 13 of right-to-sue letter by EEOC does not bar a Title VII suit); 14 Saulsbury v. Wismer & Becker, Inc., 644 F.2d 1251, 1257 (9th Cir. 15 1980) (same); Commodari v. Long Island Univ., 89 F. Supp. 2d 353, 16 381-83 (E.D.N.Y. 2000) (same); Palumbo v. Lufthansa German 17 Airlines, 1999 U.S. Dist. LEXIS 11412, No. 98 Civ. 5005, 1999 WL 18 540446, at *2 (S.D.N.Y. July 26, 1999) (same); Figueira v. Black 19 Entm’t Television, Inc., 944 F. Supp. 299, 303-08 (S.D.N.Y. 1996) 20 (same). 21 The key fact in the present matter is that the language of 22 29 U.S.C. § 626, which authorizes suits under the ADEA, differs 23 significantly from that of Section 2000e-5(f)(1). Section 626 24 provides that “[n]o civil action may be commenced by an 7
    193. 1 individual under this section until 60 days after a charge 2 alleging unlawful discrimination has been filed with the [EEOC]." 3 Id. § 626(d). Appellant complied with this provision by waiting 4 sixty days after filing his EEOC charge before bringing the 5 instant suit. The fact that the EEOC terminated its proceedings 6 prior to the expiration of sixty days was irrelevant to the 7 district court’s authority to entertain the case. This is 8 especially so because Section 626, unlike Section 2000e-5, 9 explicitly contemplates early termination of EEOC investigations. 10 Id. § 626(e) (“If a charge filed with the [EEOC] under this 11 chapter is dismissed or the proceedings of the [EEOC] are 12 otherwise terminated by the [EEOC], the [EEOC] shall notify the 13 person aggrieved.”). This suit was therefore properly before the 14 district court. 15 b) The Religious Freedom Restoration Act 16 In our view, the dispositive issue in this matter concerns 17 the application of the RFRA. The statute's substantive 18 provisions state: 19 (a) In general. Government shall not substantially 20 burden a person's exercise of religion even if the 21 burden results from a rule of general applicability, 22 except as provided in subsection (b). 23 (b) Exception. Government may substantially burden a 24 person's exercise of religion only if it demonstrates 25 that application of the burden to the person-- 26 (1) is in furtherance of a compelling governmental 27 interest; and 28 (2) is the least restrictive means of furthering that 29 compelling governmental interest. 30 8
    194. 1 42 U.S.C. § 2000bb-1. 2 The test set out in Subsection (b)(1) and (2) “applies to 3 all Federal law, and the implementation of that law, whether 4 statutory or otherwise, and whether adopted before or after 5 November 16, 1993.” Id. § 2000bb-3(a). The RFRA's remedial 6 provision states that “[a] person whose religious exercise has 7 been burdened in violation of this section may assert that 8 violation as a claim or defense in a judicial proceeding and 9 obtain appropriate relief against a government.” Id. § 10 2000bb-1(c). "[G]overnment" is in turn defined to include any 11 "branch, department, agency, instrumentality, and official (or 12 other person acting under color of law) of the United States." 13 Id. § 2000bb-2(1). 14 The present action is a suit against a church and an 15 official of that church. The suit claims that the defendants 16 violated a federal statute, the ADEA, and seeks judicial 17 remedies; appellees claim that application of the statute would 18 substantially burden the exercise of their religion. If the 19 RFRA's test for evaluating burdens on religious activity -- 20 Subsections (b)(1) and (2) -- is not met, appellees can arguably 21 assert a violation of the RFRA as a complete defense. 22 The district court dismissed the case based on a 23 "ministerial exception" that some courts had read into various 24 anti-discrimination laws -- an unresolved issue in this circuit - 9
    195. 1 - including the ADEA. Whatever the merits of that exception as 2 statutory interpretation or policy, it has no basis in statutory 3 text, whereas the RFRA, if applicable, is explicit legislation 4 that could not be more on point. Given the absence of other 5 relevant statutory language, the RFRA must be deemed the full 6 expression of Congress’s intent with regard to the religion- 7 related issues before us and displace earlier judge-made 8 doctrines that might have been used to ameliorate the ADEA’s 9 impact on religious organizations and activities. City of 10 Milwaukee v. Illinois, 451 U.S. 304, 314 (1981) ("Federal common 11 law is a necessary expedient, and when Congress addresses a 12 question previously governed by a decision rested on federal 13 common law the need for such an unusual exercise of lawmaking by 14 federal courts disappears.") (internal quotation marks and 15 citations omitted). 16 There is little caselaw addressing the issue whether the 17 RFRA applies to an action by a private party seeking relief under 18 a federal statute against another private party who claims that 19 the federal statute substantially burdens his or her exercise of 20 religion.4 The RFRA's language surely seems broad enough to 21 encompass such a case. The statutory language states that it 22 "applies to all federal law, and the implementation of that law," 23 42 U.S.C. § 2000bb-3(a), and that a defendant arguing that such a 24 law substantially burdens the exercise of religion "may assert [a 10
    196. 1 violation of the RFRA] as a . . . defense in a judicial 2 proceeding." Id. § 2000bb-1(c). This language easily covers 3 the present action. The only conceivably narrowing language is 4 the phrase immediately following: "and obtain appropriate relief 5 against a government." Id. However, this language would seem 6 most reasonably read as broadening, rather than narrowing, the 7 rights of a party asserting the RFRA. The narrowing 8 interpretation -- permitting the assertion of the RFRA as a 9 defense only when relief is also sought against a governmental 10 party -- involves a convoluted drawing of a hardly inevitable 11 negative implication. If such a limitation was intended, 12 Congress chose a most awkward way of inserting it. The 13 legislative history is neither directly helpful nor harmful to 14 that view. 15 We need not, however, decide whether the RFRA applies to a 16 federal law enforceable only in private actions between private 17 parties. The ADEA is enforceable by the EEOC as well as private 18 plaintiffs, and the substance of the ADEA's prohibitions cannot 19 change depending on whether it is enforced by the EEOC or an 20 aggrieved private party. See United States v. Brown, 79 F.3d 21 1550, 1559 n.16 (11th Cir. 1996) ("The meaning of the statutory 22 words 'scheme to defraud' does not change depending on whether 23 the case is Civil RICO or criminal."). An action brought by an 24 agency such as the EEOC is clearly one in which the RFRA may be 11
    197. 1 asserted as a defense, and no policy of either the RFRA or the 2 ADEA should tempt a court to render a different decision on the 3 merits in a case such as the present one. Indeed, appellant 4 argues that the RFRA is inapplicable only because it is 5 unconstitutional. 6 1. Waiver 7 First, however, we must address whether appellees have 8 waived or forfeited reliance upon the RFRA. In their original 9 brief, as noted, appellees argued that the ADEA was an unlawful 10 burden on their religious activities and that Congress has 11 enacted the RFRA, a statute that applied to all federal laws, 12 "for this very reason." Appellant’s Brief at 28. Believing that 13 this reference to a seemingly dispositive but otherwise 14 unmentioned statute needed some elaboration and unconvinced that 15 appellant's claim that the Supreme Court had held the RFRA 16 unconstitutional in all circumstances was correct, we asked for 17 further briefing. 18 Somewhat to our surprise, appellees’ post-argument letter- 19 brief states that, although all pertinent portions of the RFRA 20 are constitutional, the statute is inapplicable because “the case 21 at bar is a matter relating to a private employment situation and 22 does not involve actions by the government.” Nevertheless, 23 appellees continue to rely upon the "ministerial exception" and 24 the Free Exercise and Establishment Clauses. 12
    198. 1 In our view, as discussed above, the RFRA’s provisions are 2 directly on point, and allow parties who, like appellees, claim 3 that a federal statute, like the ADEA, substantially burdens the 4 exercise of their religion to assert the RFRA as a defense to any 5 action asserting a claim based on the ADEA. The issue then is 6 whether their post-argument letter-brief constitutes a waiver or 7 forfeiture of that defense. 8 A party may certainly waive or forfeit a RFRA defense by 9 failing to argue that a law or action substantially burdens the 10 party’s religion. For example, in United States v. Amer, 11 appellant had forfeited the defense that his child kidnaping 12 conviction violated the RFRA, because “[a]t no point during the 13 pretrial, trial, or sentencing proceedings did [appellant] argue 14 that his act of removing and retaining the children was 15 religiously mandated or inspired.” 110 F.3d 873, 879 & n.1 (2d 16 Cir. 1997). Where a party fails to assert a substantial burden 17 on religious exercise before a district court, therefore, the 18 party may not raise that issue -- an inherently fact-based one -- 19 for the first time on appeal. 20 However, appellees argued in the district court and here -- 21 and continue to argue -- that application of the ADEA to the 22 relationship between their church and appellant substantially 23 burdens their religion. They continue to assert the "ministerial 24 exception," which in their view tracks the Free Exercise clause 13
    199. 1 of the Constitution and the Establishment Clause as well. 2 Appellees’ Brief at 4-15; see Elvig v. Calvin Presbyterian 3 Church, 397 F.3d 790, 790 (9th Cir. 2005) ("[T]he 'ministerial 4 exception' to Title VII is carved out from the statute based on 5 the commands of the Free Exercise and Establishment Clauses of 6 the First Amendment."). In substance, therefore, they ask us to 7 apply the RFRA, but not to mention it. 8 Appellees' position that the RFRA does not apply to suits 9 between private parties is not determinative of our analysis, 10 given that they have vigorously pursued and preserved the 11 substance of the issue. We are required to interpret federal 12 statutes as they are written -- in this case the ADEA as amended 13 by the RFRA -- and we are not bound by parties’ stipulations of 14 law. Becker v. Poling Transp. Corp., 356 F.3d 381, 390 (2d Cir. 15 2004); see also Kamen v. Kemper Fin. Servs. Inc., 500 U.S. 90, 99 16 ("When an issue or claim is properly before the court, the court 17 is not limited to the particular legal theories advanced by the 18 parties, but rather retains the independent power to identify and 19 apply the proper construction of governing law."). We are not in 20 the business of deciding cases according to hypothetical legal 21 schemes, particularly when the hypothetical scheme posed by a 22 party tracks the actual law in all but name. 23 2. Constitutionality 24 In addressing the constitutional issues raised by appellant 14
    200. 1 with regard to the RFRA, we first describe the statutory 2 background. 3 The RFRA was passed in response to Employment Div. v. Smith, 4 494 U.S. 872 (1990). The Supreme Court held there that "the 5 right of free exercise does not relieve an individual of the 6 obligation to comply with a valid and neutral law of general 7 applicability on the ground that the law proscribes (or 8 prescribes) conduct that his religion prescribes (or 9 proscribes)." Id. at 879 (internal quotation marks and citation 10 omitted). Smith limited the applicability of the "compelling 11 state interest" test the Court had previously applied to neutral 12 laws before allowing them to place a substantial burden on 13 religious practice. Id. at 883-84 (limiting test to mean that 14 "where the State has in place a system of individual exemptions, 15 it may not refuse to extend that system to cases of ‘religious 16 hardship' without compelling reason").5 17 Congress enacted the RFRA pursuant to two sources of 18 authority, Section 5 of the Fourteenth Amendment and the 19 Necessary and Proper Clause of the Constitution. See H.R. Rep. 20 No. 103-88, at 17 (1993) (“Finally, the Committee believes that 21 Congress has the constitutional authority to enact [the RFRA]. 22 Pursuant to Section 5 of the Fourteenth Amendment and the 23 Necessary and Proper Clause of the Constitution, the legislative 24 branch has been given the authority to provide statutory 15
    201. 1 protection for a constitutional value . . . .”). The Supreme 2 Court held that the RFRA could not be enacted under Section 5 of 3 the Fourteenth Amendment, which empowers Congress to enforce the 4 Amendment's other provisions against the states. City of Boerne 5 v. Flores, 521 U.S. 507, 519 (1997) (“Congress does not enforce a 6 constitutional right by changing what that right is.”). The RFRA 7 is therefore unconstitutional as applied to state law. 8 However, the RFRA applies by its terms not only to the 9 states but also to “all Federal law, and the implementation of 10 that law, whether statutory or otherwise, and whether adopted 11 before or after November 16, 1993.” 42 U.S.C. § 2000bb-3(a); see 12 also id. § 2000bb-2(1) (“‘[G]overnment’ includes a branch, 13 department, agency, instrumentality, and official (or other 14 person acting under color of law) of the United States.”). 15 Boerne could not have addressed whether the RFRA was validly 16 enacted under the Necessary and Proper Clause because the only 17 issue before the Court was the denial of a building permit to a 18 church by local zoning authorities. 521 U.S. at 512. Since 19 Boerne, “[e]very appellate court that has squarely addressed the 20 question has held that the RFRA governs the activities of federal 21 officers and agencies." O'Bryan v. Bureau of Prisons, 349 F.3d 22 399, 401 (7th Cir. 2003); Guam v. Guerrero, 290 F.3d 1210, 1221 23 (9th Cir. 2002); Henderson v. Kennedy, 265 F.3d 1072, 1073 (D.C. 24 Cir. 2001); Kikumura v. Hurley, 242 F.3d 950, 960 (10th Cir. 16
    202. 1 2001); Christians v. Crystal Evangelical Free Church (In re 2 Young), 141 F.3d 854, 856 (8th Cir. 1998); see also Madison v. 3 Riter, 355 F.3d 310, 315 (4th Cir. 2003). 4 We join the other circuits in holding that the RFRA is 5 constitutional as applied to federal law under the Necessary and 6 Proper Clause of the Constitution. As presented in this case, 7 the issue is simply whether Congress had the authority to amend 8 the ADEA to include the RFRA standard. See In re Young, 141 F.3d 9 at 861 (the RFRA “has effectively amended the Bankruptcy Code, 10 and has engrafted the additional clause to § 548(a)(2)(A) that a 11 recovery that places a substantial burden on a debtor's exercise 12 of religion will not be allowed unless it is the least 13 restrictive means to satisfy a compelling governmental 14 interest.”). 15 Congress enacted the ADEA pursuant to its Commerce Clause 16 powers under Article I. Kimel v. Fla. Bd. of Regents, 528 U.S. 17 62, 78 (2000) (“the ADEA constitutes a valid exercise of 18 Congress' power ‘[t]o regulate Commerce . . . among the several 19 States’”) (citing EEOC v. Wyoming, 460 U.S. 226, 243 (1983)) 20 (alterations in original); McGinty v. New York, 251 F.3d 84, 91 21 (2d Cir. 2001); see U.S. Const., Art. I, § 8, cl. 3 (“The 22 Congress shall have power . . . [t]o regulate commerce with 23 foreign Nations, and among the several States, and with the 24 Indian Tribes."). Furthermore, the Necessary and Proper Clause 17
    203. 1 authorizes Congress “[t]o make all Laws which shall be necessary 2 and Proper for carrying into Execution” its Article I powers, 3 including its Commerce Clause powers. U.S. Const. art. I, § 8, 4 cl. 18. The Clause allows all legitimate legislation “plainly 5 adapted” to a constitutional end. M’Culloch v. Maryland, 17 U.S. 6 (4 Wheat.) 316, 421 (1819) (“Let the end be legitimate, let it be 7 within the scope of the constitution, and all means which are 8 appropriate, which are plainly adapted to that end, which are not 9 prohibited, but consist with the letter and spirit of the 10 constitution, are constitutional.”). Finally, the "plainly 11 adapted" standard requires only “that the effectuating 12 legislation bear a rational relationship to a permissible 13 constitutional end.” United States v. Wang Kun Lue, 134 F.3d 79, 14 84 (2d Cir. 1998). 15 It is obvious to us that because Congress had the power to 16 enact the ADEA, it also had the power to amend that statute by 17 passing the RFRA. The RFRA was authorized by the Necessary and 18 Proper Clause because its purpose -- to protect First Amendment 19 rights as interpreted by the Congress, see S. Rep. No. 103-111, 20 at 14 (1993), reprinted in 1993 U.S.C.C.A.N. 1892, 1903 -- was 21 permissible. “When Congress acts within its sphere of power and 22 responsibilities, it has not just the right but the duty to make 23 its own informed judgment on the meaning and force of the 24 Constitution.” Boerne, 521 U.S. at 535. 18
    204. 1 The RFRA was also proper as applied to the ADEA in 2 particular because, as noted, Congress had authority to enact 3 that statute under the Commerce Clause. See INS v. Chadha, 462 4 U.S. 919, 941 (1983) ("’Congress has plenary authority in all 5 cases in which it has substantive legislative jurisdiction, so 6 long as the exercise of that authority does not offend some other 7 constitutional restriction.’”) (quoting Buckley v. Valeo, 424 8 U.S. 1, 132 (1976)) (internal citation omitted); Guerrero, 290 9 F.3d at 1220 (“Congress derives its ability to protect the free 10 exercise of religion from its plenary authority found in Article 11 I of the Constitution; it can carve out a religious exemption 12 from otherwise neutral, generally applicable laws based on its 13 power to enact the underlying statute in the first place.”); In 14 re Young, 141 F.3d at 861 (“[W]e can conceive of no argument to 15 support the contention[] that Congress is incapable of amending 16 the legislation that it has passed.”).6 17 In his post-argument letter-brief, appellant argues that 18 application of the RFRA to federal law violates separation of 19 powers principles and the Establishment Clause of the 20 Constitution.7 We address these issues in turn. 21 Appellant's separation of powers challenge is that because 22 the RFRA mandates evaluation of laws and actions that burden 23 religion by a standard different from that prescribed by the 24 Supreme Court, it is a Congressional usurpation of judicial 19
    205. 1 power. However, we agree with the Eighth Circuit that “[t]he key 2 to the separation of powers issue in this case is . . . not 3 whether Congress disagreed with the Supreme Court’s 4 constitutional analysis, but whether Congress acted beyond the 5 scope of its constitutional authority in applying RFRA to federal 6 law.” In re Young, 141 F.3d at 860; United States v. Marengo 7 County Comm'n, 731 F.2d 1546, 1562 (11th Cir. 1984) 8 (“[C]ongressional disapproval of a Supreme Court decision does 9 not impair the power of Congress to legislate a different result, 10 as long as Congress had that power in the first place.”). 11 Indeed, “Congress has often provided statutory protection of 12 individual liberties that exceed the Supreme Court’s 13 interpretation of constitutional protection.” In re Young, 141 14 F.3d at 860 (collecting examples); Guerrero, 290 F.3d at 1221 15 (“Certainly Congress can provide more individual liberties in the 16 federal realm than the Constitution requires without violating 17 vital separation of powers principles.”). That the RFRA provides 18 more protection from federal actors and statutes than may be 19 required by the First Amendment hardly undermines separation of 20 powers principles. 21 With respect to appellant's Establishment Clause argument, 22 the Clause provides that "Congress shall make no law respecting 23 an establishment of religion." U.S. Const. amend. I. The 24 Supreme Court has established a three-prong test to determine 20
    206. 1 whether a statute violates the Clause. 2 First, the statute must have a secular legislative 3 purpose; second, its principal or primary effect must 4 be one that neither advances nor inhibits religion; 5 finally, the statute must not foster an excessive 6 government entanglement with religion. 7 8 Lemon v. Kurtzman, 403 U.S. 602, 612-13 (1971) (quotations and 9 citations omitted). Applying this test, the Court has held that 10 exempting religious organizations from compliance with neutral 11 laws does not violate the Constitution. E.g., Corp. of the 12 Presiding Bishop of the Church of Jesus Christ of Later-Day 13 Saints v. Amos, 483 U.S. 327, 338-40 (1987) (exemption from 14 federal antidiscrimination laws for religious organizations does 15 not violate Establishment Clause); see also Gillette v. United 16 States, 401 U.S. 437, 460 (1971) (exemption from military draft 17 for religious conscientious objectors does not violate 18 Establishment Clause); Walz v. Tax Comm'n, 397 U.S. 664, 680 19 (1970) (state property tax exemption for religious organizations 20 does not violate Establishment Clause). 21 Given these holdings, appellant faces an unwinnable battle 22 in claiming that the RFRA -- a limited exemption for religious 23 organizations from compliance with neutral laws -- violates the 24 Establishment Clause. The RFRA had a secular legislative purpose 25 within the meaning of Lemon -- namely, to protect individual 26 First Amendment rights as interpreted by the Congress. As noted, 27 this purpose was not only permissible but was also required by 21
    207. 1 Congress’s duty to interpret the Constitution. Boerne, 521 U.S. 2 at 535. A “secular legislative purpose” need not be “unrelated 3 to religion”; rather, Lemon’s first prong aims to prevent 4 Congress “from abandoning neutrality and acting with the intent 5 of promoting a particular point of view in religious matters.” 6 Amos, 483 U.S. at 335; Gillette, 401 U.S. at 454 ("'Neutrality' 7 in matters of religion is not inconsistent with 'benevolence' by 8 way of exemptions from onerous duties, so long as an exemption is 9 tailored broadly enough that it reflects valid secular 10 purposes.") (citation omitted). The RFRA reflected no purpose to 11 promote a particular religious point of view. 12 The RFRA also satisfies the other two prongs of the Lemon 13 test. Its principal effect neither advances nor inhibits 14 religion within the meaning of Lemon. “For a law to have 15 forbidden ‘effects’ under Lemon, it must be fair to say that the 16 government itself has advanced religion through its own 17 activities and influence,” rather than simply by granting an 18 exemption to religious organizations. Amos, 483 U.S. at 337-38 19 (“Where . . . government acts with the proper purpose of lifting 20 a regulation that burdens the exercise of religion, we see no 21 reason to require that the exemption come packaged with benefits 22 to secular entities.”). Although the RFRA certainly provides 23 some benefit to religious organizations, “a law is not 24 unconstitutional simply because it allows churches to advance 22
    208. 1 religion, which is their very purpose.” Id. at 337. Finally, 2 there is no question that the RFRA decreases rather than fosters 3 government entanglement with religion, as required by the third 4 prong of Lemon. Amos, 483 U.S. at 339 (An exemption “effectuates 5 a more complete separation of [church and state] and avoids . . . 6 intrusive inquiry into religious belief.”). 7 We note in general that the Supreme Court approved of and 8 invited legislative enactments of religious exceptions to neutral 9 laws in Smith itself. 494 U.S. at 890. The court pointed to 10 state exceptions to drug laws for sacramental peyote use and 11 noted with approval that “a society that believes in the negative 12 protection accorded to religious belief can be expected to be 13 solicitous of that value in its legislation as well.” Id. 14 (“[T]o say that a nondiscriminatory religious-practice exemption 15 is permitted, or even that it is desirable, is not to say that it 16 is constitutionally required.”). We therefore hold that the 17 RFRA, as applicable to federal law, does not violate the 18 Establishment Clause of the Constitution. 19 Having found the portions of the RFRA applicable to the 20 federal government and federal law constitutional, we have little 21 difficulty finding those portions severable from the RFRA’s 22 unconstitutional sections. A court must sever the invalid parts 23 of a statute from the valid parts “unless it is evident that the 24 Legislature would not have enacted those provisions which are 23
    209. 1 within its power, independently of that which is not.” Chadha, 2 462 U.S. at 931-32 (internal quotation marks, citations, and 3 alterations omitted); Alaska Airlines, Inc. v. Brock, 480 U.S. 4 678, 684 (1987) (“A court should refrain from invalidating more 5 of the statute than is necessary.”) (alteration omitted). We 6 know of no evidence that Congress would not have applied the RFRA 7 to the federal government unless it could also be applied to 8 state and local governments. We therefore hold the portion of 9 the RFRA applicable to the federal government severable from its 10 unconstitutional portions. See Kikumura, 242 F.3d at 959-60 11 (finding federal portions of the RFRA severable); In re Young, 12 141 F.3d at 859 (same). 13 CONCLUSION 14 The RFRA is an amendment to the ADEA and, as such, is 15 constitutional. The parties have not briefed the issue of how it 16 impacts the merits of this case. The district court did not 17 apply the RFRA, relying instead on the “ministerial exception” to 18 the ADEA. We believe that, while the RFRA's application is a 19 matter of law, it would be appropriate to hear from the district 20 court first, rather than seek yet further briefing in this court. 21 We therefore vacate and remand for reconsideration under the 22 RFRA standards. 23 24 24
    210. FOOTNOTES 1. Appellant initially moved for a preliminary injunction requiring appellees to restore his active status, but he withdrew the motion after the NYAC and Lyght appointed another clergy member to fill his vacant position. 2. The district court did not address this issue, but because it raises purely legal questions, we do so here. See McGinty v. New York, 251 F.3d 84, 90 (2d Cir. 2001) (addressing question not decided by district court where facts were undisputed and legal question was briefed). 3. Section 2000e-5(f)(1) provides in pertinent part: If a charge filed with the Commission . . . is dismissed by the Commission, or if within one hundred and eighty days from the filing of such charge . . . the Commission has not filed a civil action . . . or the Commission has not entered into a conciliation agreement to which the person aggrieved is a party, the Commission . . . shall so notify the person aggrieved and within ninety days after the giving of such notice a civil action may be brought against the respondent named in the charge. . . . 4. No court appears to have addressed the issue squarely, but some suggestive caselaw exists. Some courts seem to have assumed 25
    211. without discussion that the RFRA may be asserted as a defense by a private party against another private party. See, e.g., Guinan v. Roman Catholic Archdiocese of Indianapolis, 42 F. Supp. 2d 849, 853 (S.D. Ind. 1998) (permitting the private party defendant to assert a RFRA defense but rejecting it after first finding that the ministerial exception negated the need to discuss the RFRA defense); Urantia Found. v. Maaherra, 895 F. Supp. 1335, 1336-37 (D. Ariz. 1995) (permitting the defendant to raise a RFRA defense but rejecting it because the defendant did not contest the constitutionality of the trademark and copyright laws in general or as applied to her). Bankruptcy courts have also generally permitted a private-party defendant to assert a RFRA defense against a Chapter 7 trustee. See Christians v. Crystal Evangelical Free Church (In re Young), 82 F.3d 1407, 1418-19 (8th Cir. 1996) (permitting a defendant to assert a RFRA defense and recover debtors’ tithes to the church because “the government action in question meaningfully curtails, albeit retroactively, a religious practice”), vacated, 521 U.S. 1114 (1997), reaff’d, 141 F.3d 854 (8th Cir. 1998); see also In re Tessier, 190 B.R. 396 (Bankr. D. Mont. 1995); Newman v. Midway Southern Baptist Church (In re Newman), 183 B.R. 239 (Bankr. D. Kan. 1995), aff’d, 203 B.R. 468 (D. Kan. 1996). A bankruptcy trustee is arguably "acting under color of law" and therefore falls within the RFRA's definition of "government." 42 U.S.C. § 2000bb-2(1). United 26
    212. States trustees are part of the executive branch and protect the interests of the United States in the liquidation. See 28 U.S.C. § 586(a); 11 U.S.C. §§ 701(a)(1), 703(b)-(c) and 704(9); In re Shoenewerk, 304 B.R. 59, 62-63 (Bankr. E.D.N.Y. 2003). 5. The RFRA’s stated purposes include "restor[ing] the compelling interest test as set forth in Sherbert v. Verner, 374 U.S. 398 (1963) and Wisconsin v. Yoder, 406 U.S. 205 (1972)." 42 U.S.C. § 2000bb(b)(1). The Supreme Court noted that "Congress enacted RFRA in direct response to the Court’s decision in" Smith. City of Boerne v. Flores, 521 U.S. 507, 512 (1997). 6. We find no principled constitutional distinction between Congress’s ability to amend statutes on an individual basis and its power to do so in a wholesale manner through an enactment such as the RFRA. See Guerrero, 290 F.3d at 1221 n.18. 7. Appellant also argues that Boerne explicitly invalidated all of the RFRA due to separation of powers concerns. Specifically, appellant relies upon the statement that the “RFRA contradicts vital principles necessary to maintain separation of powers and the federal balance.” Boerne, 521 U.S. at 536. The argument is entirely unconvincing. The quoted language simply explained why Congress could not enact the RFRA pursuant to its Section 5 27
    213. power. The quoted phrase reads in full as follows: “Broad as the power of Congress is under the Enforcement Clause of the Fourteenth Amendment, RFRA contradicts vital principles necessary to maintain separation of powers and the federal balance.” Id. This analysis has no application to any separation of powers concerns raised by the RFRA’s enactment and application to the federal government under the Necessary and Proper Clause. See Guerrero, 290 F.3d at 1220 (Boerne’s “discussion of the separation of powers doctrine was entirely within the framework of its section 5 analysis -- not an independent rationale.”). 28
    214. 04-5711(L)-cv, 04-5943-cv Walczyk v. Rio UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT August Term, 2005 (Argued: April 3, 2006 Decided: August 1, 2007) Docket Nos. 04-5711-cv(L), 04-5943-cv(XAP) T HOMAS W ALCZYK, E LIZABETH W ALCZYK, M AXIMINA W ALCZYK, EACH INDIVIDUALLY AND AS P.P.A. FOR M ICHELLE W ALCZYK, A MINOR CHILD,1 Plaintiffs-Appellees-Cross-Appellants, —v.— J AMES R IO, B RIAN K ILLIANY, J AMES J EPSEN, W ILLIAM T YLER, A NGELA D ESCHENES, AND S HAWN B ROWN, Defendants-Appellants-Cross-Appellees. Before: C ABRANES, S OTOMAYOR, and RAGGI, Circuit Judges. Interlocutory appeal from so much of an order of the United States District Court for the District of Connecticut as (1) denied defendants qualified immunity on plaintiffs’ federal 1 We direct the Clerk of Court to change the official caption to comport with this decision. 1
    215. and state constitutional challenges to the execution of arrest and search warrants. Cross- appeal from so much of the order as (2) denied plaintiff Elizabeth Walczyk summary judgment on the liability element of her illegal search claim and (3) granted defendants summary judgment on Thomas Walczyk’s claim of excessive bail. R EVERSED AND R EMANDED on part of defendants’ qualified immunity appeal. A FFIRMED in all other respects. Judge Sotomayor concurs in a separate opinion. T HOMAS R. G ERARDE (John J. Radshaw, III, on the brief), Howd & Ludorf, LLC, Hartford, Connecticut, for Defendants-Appellants-Cross- Appellees. J ON L. S CHOENHORN (Jennifer L. Bourn, on the brief), Jon L. Schoenhorn & Associates, Hartford, Connecticut, for Plaintiffs-Appellees-Cross- Appellants. R EENA R AGGI, Circuit Judge: In 2001, plaintiff Thomas Walczyk (“Walczyk”) was convicted after a jury trial in Connecticut on state law charges of disorderly conduct, see Conn. Gen. Stat. § 53a-182(a)(2); reckless endangerment, see id. § 53a-64(a); and improper firearm storage, see id. § 29-37i. On appeal, the Connecticut Appellate Court reversed, holding that Walczyk’s conviction violated federal and state law because it was based on incriminating evidence obtained through search warrants that were not supported by probable cause. See State v. Walczyk, 76 Conn. App. 169, 182, 818 A.2d 868, 876 (Conn. App. Ct. 2003). Thereafter, Walczyk, 2
    216. his wife Maximina, his minor child Michelle, and his mother Elizabeth initiated this civil action, suing defendants, all members of the Farmington, Connecticut Police Department, in the United States District Court for the District of Connecticut (Robert N. Chatigny, Chief Judge), pursuant to 42 U.S.C. §§ 1983 and 1988 and Connecticut law for money damages arising from events relating to Walczyk’s reversed conviction. Although the district court granted defendants’ motion for summary judgment with respect to some of plaintiffs’ claims, defendants now pursue an interlocutory appeal from so much of the district court’s order, entered on September 30, 2004, as denied them qualified immunity from plaintiffs’ unlawful arrest and search claims. See Walczyk v. Rio, 339 F. Supp. 2d 385, 389-91 (D. Conn. 2004). Not surprisingly, plaintiffs defend that denial. At the same time, Elizabeth Walczyk cross-appeals the district court’s denial of her motion for summary judgment on the liability element of her challenge to the search of her home. See id. at 391. Meanwhile, Thomas Walczyk cross-appeals the award of summary judgment to defendants on his Eighth Amendment claim that he was detained on excessive bail. See id. at 390. For the reasons discussed herein, we conclude that the arrest of Thomas Walczyk and the search of the home he shared with his wife and daughter were supported by probable cause. We reverse so much of the district court’s order as concluded otherwise, and we remand with directions to enter summary judgment in favor of defendants on that part of plaintiffs’ complaint. With respect to defendants’ search of Elizabeth Walczyk’s home, we 3
    217. affirm the district court’s denial of qualified immunity to defendants because the warrant authorizing that search was procured on the basis of plainly stale information and questions of fact remain as to whether any or all defendants acted knowingly or recklessly in misleading the issuing magistrate as to the currency of that information. Those same questions of fact prompt us to affirm the district court’s denial of summary judgment to Elizabeth Walczyk on the liability element of her unlawful search claim. Finally, with respect to Walczyk’s excessive bail claim, we affirm the award of summary judgment in favor of defendants on the ground of absolute immunity. I. Factual Background A. The Underlying Land Dispute 1. Barberino Realty Acquires the Land This case has its origins in a longstanding property dispute between the Walczyk family and Barberino Realty and Development Corporation (“Barberino”). The property — undeveloped land adjacent to the Farmington homes of Thomas Walczyk at 28 Tunxis Street and of his parents, Elizabeth and Lucien Walczyk,2 at 27 Tunxis Street — was acquired by Barberino in 1973. Over the next two decades, Barberino encountered various difficulties developing the land, only some of which are relevant to this appeal. 2. 1981: Walczyk Brandishes a Rifle at Barberino Workers Sometime in 1981, a Barberino work crew entered onto the undeveloped land to drill 2 Lucien Walczyk is not a party to this lawsuit, having died prior to its commencement. 4
    218. for soil samples preliminary to development. Thomas Walczyk, who was licensed to possess numerous firearms, brandished a rifle at the workmen and challenged their actions. The workmen sought police assistance, after which their work proceeded apparently without interruption and without any official action being taken against Walczyk. The incident nevertheless prompted Barberino’s counsel to seek assurances from Elizabeth and Lucien Walczyk that there would be no further attempts to hinder development of the land. In response, an attorney for the elder Walczyks advised that his clients were claiming title to the undeveloped land by adverse possession. The claim was based on the Walczyks’ long use of a portion of the undeveloped land for vegetable gardening and cattle grazing. 3. 1988: Walczyk Again Brandishes a Gun at a Barberino Worker Despite these 1981 events, the relationship between the Walczyks and Barberino appears to have remained uneventful until January 1988 when, in response to a Barberino demand that the Walczyks remove certain items from the property, the elder Walczyks reiterated their adverse possession claim. A few months later, in March 1988, a Barberino worker equipped with a bulldozer attempted forcibly to remove various items from the disputed property. Once again, Thomas Walczyk confronted the worker with a licensed firearm, specifically, an AR-15 automatic assault rifle loaded with thirty rounds of ammunition, and ordered him off the property. Police responded to the scene and directed Walczyk to drop his weapon. Walczyk initially 5
    219. ignored several such directives, “yell[ing] about trespassing and some statute.” Police Rpt., Mar. 24, 1988, at 2. When Walczyk finally put down the rifle, a “wrestling match[]” ensued as he tried to prevent the police from taking him into custody. Id. at 3. Charged with threatening, reckless endangerment, and interfering with police, Walczyk eventually pleaded guilty to the lesser infraction of creating a public disturbance. 4. The Walczyks’ Lawsuits Claiming Adverse Possession The following month, in April 1988, Elizabeth and Lucien Walczyk sued for adverse possession of the undeveloped property. The action settled in 1991 with Barberino paying the elder Walczyks $20,000 and granting them a perpetual agricultural easement over a portion of the disputed land. In return, Elizabeth and Lucien Walczyk signed a quitclaim of any right, title, or interest in the property and promised not to oppose Barberino’s development plans before the town planning and zoning commission. Four years later, in January 1995, Thomas Walczyk sued Barberino, as well as his parents, claiming that he held title to the disputed property through adverse possession. On March 14, 1997, the Connecticut Superior Court rejected Walczyk’s claim as a matter of law.3 In granting judgment to Barberino and quieting title in its favor, the Connecticut court 3 In addition to holding that Walczyk failed to adduce evidence sufficient to establish any of the elements of adverse possession, the court ruled that his claim was barred by the equitable doctrines of unclean hands and laches because he had “initiated and directed” his parents’ 1988 suit against Barberino and made the strategic choice to pursue that claim and its settlement only in their name. See Walczyk v. Barberino Realty, Inc., No. cv- 950465712S, 1997 Conn. Super. LEXIS 715, at *35 (Conn. Super. Ct. Mar. 14, 1997) (“The plaintiff, by his failure to disclose his own claim in a timely fashion, effectively misled the 6
    220. stated: “Thomas Joseph Walczyk has no estate, interest in or encumbrance of said real property or any part thereof.” Walczyk v. Barberino Realty & Dev. Corp., No. cv- 950465712S, 1997 Conn. Super. LEXIS 718, at *2 (Conn. Super. Ct. Mar. 14, 1997). This judgment was affirmed on direct appeal, see Walczyk v. Barberino Realty & Dev. Corp., 48 Conn. App. 911, 719 A.2d 1233 (Conn. App. Ct. 1998), and the Connecticut Supreme Court declined further review, see Walczyk v. Barberino Realty & Dev. Corp., 245 Conn. 904, 719 A.2d 1165 (1998). B. Events Relating to the Challenged Arrest and Searches 1. Walczyk’s April 1999 Threat to “Take Matters Into My Own Hands” Despite the state courts’ unequivocal rejection of his adverse possession claim, Walczyk persisted in asserting a superior interest in the disputed land. On April 9, 1999, he visited the Farmington Police Department to complain about Barberino’s development efforts, insisting to defendant Captain James Rio that he (Walczyk) “had a common law right to the land because he had been farming and maintaining it for some time.” Arrest Warrant Aff. at 2. When Rio explained that the police had received notice of the court order to the contrary, Walczyk stated that he expected to secure reversal of that judgment based on witness perjury and judicial misconduct. More significantly for purposes of the issues raised on this appeal, Walczyk told Rio that, “[i]f you guys don’t comply with what I’m telling you defendants [in the settlement of that claim] . . . and now seeks to take unfair advantage of a strategic decision he made, with the advice of counsel, to prosecute the first adverse possession claim in his parents’ names only.”). 7
    221. I’ll take matters into my own hands.” Id. Rio warned Walczyk not to do anything illegal, but Walczyk stated that he would “do what [he] had to do to protect his property.” Id. (alteration in original). 2. Walczyk’s August 1999 Reference to a Potential “Bloodbath” Some five months later, on August 30, 1999, Walczyk called the Farmington police to complain again that Barberino personnel were trespassing on the disputed property. Responding to the scene, Officer David Hebert explained to Walczyk that the police could not act on his trespass complaint without some documentary support for his property claim. In his report of the encounter, Hebert noted that W alczyk made “some off color com[m]ents that the police were not taking the action needed to avoid a ‘blood bath.’” Police Rpt., Aug. 30, 1999, at 1.4 At Walczyk’s subsequent criminal trial, Hebert explained that, although he considered this remark offensive, he did not immediately place Walczyk under arrest because the officer did not feel any direct threat to himself. Nevertheless, he did understand Walczyk’s comment as a threat toward “the Barberino Corporation and who[m]ever they were going to have down there working.” Trial Tr. vol. 1, 56, Mar. 23, 2001. Indeed, Hebert informed Barberino of the bloodbath statement, prompting its counsel to contact Captain Rio to request 4 Walczyk acknowledged the comment at his deposition in this case: “I said well here’s the law. It’s got a thousand years of history behind it and it’s clearly designed to prevent blood shed and disputes over property. Now if a blood bath ensues, are you going to be responsible, being the police, because they refused to uphold the law.” Walczyk Dep. at 77. 8
    222. police protection at the disputed property site during any work periods. 3. The Challenged Arrest and Search W arrants Soon thereafter, Rio reviewed Hebert’s report of his August 30, 1999 encounter with Walczyk. Rio was, of course, aware of Walczyk’s earlier statement that, if police did not assist him in his land dispute, he would take matters into his “own hands” and do what he “had to do” to protect his rights. Moreover, Rio knew that Walczyk was the licensed owner of a variety of firearms and that, over the years, he had responded to various situations by displaying, discharging, or threatening to discharge a firearm. In addition to the 1981 and 1988 incidents, detailed supra at [5-6], during which Walczyk had brandished firearms specifically at Barberino workers, these situations included a 1990 road-rage incident during which Walczyk, armed with a loaded AK-47 rifle, confronted an angry motorist who had followed him home;5 a 1992 complaint by neighbors that Walczyk shot a cat in his backyard;6 and a 1996 argument during which Walczyk threatened to shoot his brother John for using the undeveloped land for driving practice.7 5 Although Walczyk was initially charged with threatening, the state’s attorney ultimately declined prosecution. 6 Originally charged with unlawful discharge of a firearm, cruelty to animals, and conspiracy to commit cruelty to animals, Walczyk ultimately pleaded guilty to breach of the peace. 7 John Walczyk told police that he did not fear imminent physical injury during this argument in which no firearms were actually displayed; nevertheless, he felt nervous because he knew his brother owned guns and could act on his threat. No charges were filed in connection with this incident. 9
    223. Viewing the bloodbath comment in this larger context, Rio concluded that the events of August 30, 1999, demonstrated probable cause to arrest Walczyk for the Connecticut Class A misdemeanor of threatening, see Conn. Gen. Stat. § 53a-62(a),8 and to search Walczyk’s home and that of his parents for firearms that could be used as instrumentalities of the bloodbath threat. Rio did not speak with Officer Hebert before making this determination, but he did consult with a state’s attorney, who concurred in the captain’s probable cause assessment. Rio then communicated the relevant circumstances to defendant Sergeant William Tyler and directed him to prepare the paperwork necessary to procure an arrest warrant for Walczyk. After Tyler completed and signed the arrest warrant, he and Rio used the same information to prepare search warrant applications for the homes of Thomas Walczyk and his parents. These papers were then given to defendants Corporal Angela Deschenes and Officer Shawn Brown, who acted as affiants for the search warrant applications. On 8 At the time of the events in question, section 53a-62(a) stated: A person is guilty of threatening when: (1) By physical threat, he intentionally places or attempts to place another person in fear of imminent serious physical injury, or (2) he threatens to commit any crime of violence with the intent to terrorize another, to cause evacuation of a building, place of assembly, or facility of public transportation, or otherwise to cause serious public inconvenience, or (3) he threatens to commit such crime in reckless disregard of the risk of causing such terror or inconvenience. Conn. Gen. Stat. § 53a-62(a) (1999) (amended 2001 & 2002). All citations herein are to this version of the provision. 10
    224. September 4, 1999, a magistrate authorized the arrest and search warrants.9 4. Facts Supporting the Warrants Because plaintiffs claim, inter alia, that the warrant affidavits, on their face, fail to establish probable cause to support the challenged arrest and searches, we here reproduce the facts as detailed in the affidavits: On 08-30-99, Officer Hebert of the Farmington Police Department responded to [a] trespassing complaint made by Mr. Thomas Walczyk . . ., 27 Tunxis Street, Farmington, CT. Walczyk complained of trespassing by employees of the Barberino Realty & Development Corporation on property located on Tunxis Street. Walczyk has had a long standing dispute over that property and has made claims in the past that he has common law rights to the land. Walczyk told Officer Hebert that the Farmington Police were not taking the action needed to avoid a “bloodbath.” Officer Hebert reported Walczyk’s complaint to Mr. [Stephen] Barberino Jr., the owner of the land in question. As a result of Walczyk’s threat of a “bloodbath,” Atty Robert Reeve, representing Barberino, contacted Capt. James Rio of the Farmington Police Department. Reeve expressed concerns for the safety of employees during imminent construction work planned for the Tunxis Street property. He requested extra police presence during work periods. In the early spring of 1999, the Farmington Police Department received a letter from Barberino Jr.’s attorney along with a copy of a March 14, 1997 decision made by the State of Connecticut Superior Court regarding the issue of the land in question. The decision by Judge Christine E. Keller was in favor of Barberino Realty & Development Corp., and stated that “Thomas Joseph Walczyk has no estate, interest in or encumbrance of said real property or any part thereof.” During late winter of 1998 and early spring of 1999 Walczyk came to the 9 We use the term “magistrate” as a general term for a judicial officer in the State of Connecticut. 11
    225. Farmington Police Department to speak with Capt. Rio about the land dispute and the impending land development. He told Capt. Rio at that time that he had a common law right to the land because he had been farming and maintaining it for some time. Walczyk said that he was in the process of getting a Superior Court ruling to reverse the one made in Barberino’s favor. He claimed that Stephen Barberino Jr. had perjured himself and the presiding Judge had acted inappropriately. Capt. Rio explained that the Farmington Police had been advised of the ruling in favor of Barberino and that until we were officially notified otherwise, all parties and the police department would have to abide by the last court ruling. W alczyk responded that, “If you guys don’t comply with what I’m telling you I’ll take matters into my own hands.” Capt. Rio advised Walczyk against any illegal actions to which Walczyk responded that he would, “ . . . do what [he] had to do to protect his property.” The Farmington Police Department has investigated Walczyk on previous occasions for incidents involving threatening during which times he has either threatened the use of or displayed a gun. On 09-15-96 Officer Charette of the Farmington Police Department investigated a disturbance on Tunxis Street involving Walczyk and his brother. The argument was over property on Tunxis Street for which Thomas Walczyk was suing his parents. His brother claimed that Walczyk pushed him and threatened to shoot him. His brother felt no imminent threat but was concerned because he knew Walczyk owned numerous guns. Walczyk denied making the threat and no arrest was made. On 02-15-92 Walczyk was arrested in Farmington for cruelty to animals and unlawful discharge of a firearm. Neighbors reported seeing him shoot a cat on his property with a handgun. On 12-08-90 neighbors complained that Walczyk was shooting guns on property at the end of Tunxis Street. He was shooting but was not in violation. On 07-30-90 Walczyk was arrested by the Farmington Police for threatening. A motorist followed Walczyk home to complain about the way he was driving. Walczyk went inside and came back with a A-K assault rifle and an argument ensued. The assault rifle was taken as evidence. It was loaded with twenty rounds of ammunition. On 03-24-88 the Farmington Police Department responded to a disturbance at 12
    226. Tunxis Street. The disturbance was over the same land dispute between Walczyk and Barberino employees. Barberino employees were working on the land. Walczyk approached them carrying an AR 15 assault rifle ordering them to get off his property. One of the employee[]s complained that Walczyk had pointed the gun directly at him. Walczyk was arrested for threatening, reckless endangerment, and interfering with police. He denied actually pointing the gun at anyone. The gun was seized as evidence. It contained one .223 round in the chamber and twenty-nine rounds in the magazine. Walczyk fought with officers prior to being arrested. As of 09-03-99, Walczyk, according to Connecticut State Police records, has the following handguns registered in his name. 1. Colt model 1903, 32 caliber, ser. #: 354507. 2. Colt Govt. Model, 45 caliber, ser. #: 40562G70. 3. Smith & Wesson model 629, 44 caliber, ser. #: N872450. 4. Walthers Woodsman model, .22 caliber, ser. #: 142639. He also has the following assault weapons registered in his name. 1. Non-classified, 86S, ser. #: A000316. 2. Ruger (SR), Mini-14, ser. #: 18465824. 3. Colt, AR15-A2 H-BAR, ser. #: 325465. 4. Non-classified, MAK-90, ser. #: 9362979. That a review of Farmington Police records indicate[s] that Walczyk has maintained residences at both 27 and 28 Tunxis St., Farmington, CT. That town of Farmington property records show that 27 Tunxis Street is owned by Lucian Walzak [sic], and 28 Tunxis Street is owned by Thomas Walzak [sic]. Search Warrant Aff. and App. at 2-3.10 5. Execution of the Warrants The challenged warrants were executed on September 7, 1999. On that date, defendant Sergeant James Jepsen contacted Thomas Walczyk and, on the pretense of wishing to discuss the land dispute, proposed a meeting at the police station. Upon Walczyk’s arrival, 10 The arrest warrant affidavit contained identical facts, but did not include the last paragraph describing the Tunxis Street residences. 13
    227. Sergeant Tyler placed him under arrest and detained him in a cell block on $10,000 bail.11 Walczyk remained in custody until later that day, when his mother posted bail. While Walczyk was in custody, a team of officers, including Sergeant Jepsen and defendant Detective Brian Killiany executed the challenged search warrants, seizing nearly 60 licensed firearms from Thomas Walczyk’s home and 18 licensed firearms from his parents’ home, as well as approximately 2,600 rounds of ammunition, gun clips, ammunition belts, and other items related to firearms’ use. C. Connecticut’s Invalidation of the Challenged W arrants After a Connecticut jury found Walczyk guilt of disorderly conduct, see Conn. Gen. Stat. § 53a-182(a)(2); reckless endangerment, see id. § 53a-64(a); and two counts of improper firearm storage, see id. § 29-37i, 12 he was sentenced to pay a fine of $100 for each count of conviction. The Appellate Court of Connecticut reversed Walczyk’s conviction, holding that the search warrant that resulted in seizure of the charged guns was not supported by probable cause. See State v. W alczyk, 76 Conn. App. at 180-82, 818 A.2d at 875-76.13 In reaching 11 Under Connecticut law, police officers are empowered to set temporary bail. See Conn. Gen. Stat. § 54-63c(a) (discussed infra at [43-45]). 12 Walczyk was acquitted on four other counts of improper firearm storage; risking injury to a child, see Conn. Gen. Stat. § 53-21; and threatening, the charge that had initially prompted his arrest and the search of his and his parents’ homes. 13 Although plaintiffs note that this decision was authored by former Connecticut Supreme Court Chief Justice Ellen A. Peters, they do not argue that this ruling is dispositive in this case. See Norton v. Sam’s Club, 145 F.3d 114, 117 (2d Cir. 1998) (noting that 14
    228. this conclusion, the court faulted the warrant’s supporting affidavit for failing “to reconcile a construction of the ‘bloodbath’ statement as a threat . . . with the statement . . . of what the defendant actually had said.” Id. at 180, 818 A.2d at 875. Specifically, the court held that “[a] statement to a police officer that the police needed to act to avoid a ‘bloodbath’ cannot be the basis of probable cause to believe that the defendant, at that time or in the immediate future, would engage in threatening behavior.” Id. at 181-82, 818 A.2d at 876 (emphasis in argument not raised on appeal is deemed waived). Nor is it likely that such an argument would be convincing in light of numerous decisions declining to hold individual state officials bound, in their individual capacities, by determinations adverse to the state in prior criminal cases. See Jenkins v. City of New York, 478 F.3d 76, 85-86 (2d Cir. 2007) (holding that ruling in New York state criminal proceeding that defendant’s arrest was not supported by probable cause did not collaterally estop police from relitigating question when defendant sued them under § 1983); Bilida v. McCleod, 211 F.3d 166, 170-71 (1st Cir. 2000) (holding that collateral estoppel did not bar Rhode Island officers sued under § 1983 from asserting legality of searches found unconstitutional in earlier criminal proceedings: “[T]he interests and incentives of the individual police [officers] . . . are not identical to those of the state, and the officers normally have little control over the conduct of a criminal proceeding”); McCoy v. Hernandez, 203 F.3d 371, 374-75 (5th Cir. 2000) (same re: suit against Texas officers); Kinslow v. Ratzlaff, 158 F.3d 1104, 1105-06 (10th Cir. 1998) (same re: Oklahoma officers); Kraushaar v. Flanigan, 45 F.3d 1040, 1050-51 (7th Cir. 1995) (reaching same result regarding an Illinois officer); Duncan v. Clements, 744 F.2d 48, 51-52 (8th Cir. 1984) (same under Missouri law); Davis v. Eide, 439 F.2d 1077, 1078 (9th Cir. 1971) (same with respect to California officers). While we need not conclusively decide the issue, there is no reason to think Connecticut law would support a different conclusion regarding estoppel in this case. See, e.g., State v. Fritz, 204 Conn. 156, 173, 527 A.2d 1157, 1166 (1987) (observing that privity necessary to trigger collateral estoppel is not established “from the mere fact that persons may happen to be interested in the same question or in proving or disproving the same facts. While the concept of privity is difficult to define precisely, it has been held that a key consideration for its existence is the sharing of the same legal right by the parties allegedly in privity.” (internal quotation marks and citation omitted)), overruled on other grounds by State v. Crawford, 257 Conn. 769, 779-80, 778 A.2d 947, 954 (2001); accord Tevolini v. Tevolini, 66 Conn. App. 16, 22 n.6, 783 A.2d 1157, 1163 n.6 (Conn. App. Ct. 2001). 15
    229. original). The Connecticut court also noted that the affidavit failed to establish probable cause because it did not state that any of Walczyk’s earlier misconduct had resulted in his “conviction of threatening or of any other crime,” distinguish “between recent incidents and those that ha[d] become stale,” or state that his firearm possession was in any way unlawful. Id. at 180, 818 A.2d at 875 (emphasis in original). Nor did the court think that Walczyk’s prior statement to Captain Rio — that “he would take matters into his own hands and do what he had to do to protect his property” — established probable cause because Rio advised Walczyk “not to do so,” and, on August 30, 1999, Walczyk “followed instructions to report any possible trespass to the police.” Id. at 181, 818 A.2d at 876 (emphasis in original). D. The District Court Action On August 30, 2002, plaintiffs commenced this action, charging defendants with (1) violating their federal and state constitutional rights to have arrests and searches supported by probable cause, (2) depriving them of their federal rights to equal protection of the laws and to free expression, (3) violating their federal and state constitutional rights to bear arms, and (4) holding Walczyk on excessive bail. Defendants moved for summary judgment, which the district court granted with respect to all claims except those challenging Thomas Walczyk’s arrest, the searches of his and his parents’ homes, and Walczyk’s First Amendment claim, which apparently had been previously abandoned. We need not here discuss the district court rulings with respect to the plaintiffs’ equal protection or right to bear arms claims because neither is challenged on this appeal. As for 16
    230. Walczyk’s excessive bail claim, the district court ruled “as a matter of law” that “when a police officer sets temporary bail under Conn. Gen. Stat. § 54-63c, he performs a judicial function and hence has absolute immunity from suit.” Walczyk v. Rio, 339 F. Supp. 2d at 390. With respect to plaintiffs’ unlawful arrest and search challenges, the district court concluded that, although the supporting warrants were “facially valid,” defendants were not entitled to summary judgment on the ground of qualified immunity because questions of fact existed as to whether they had “knowingly and deliberately, or with reckless disregard of the truth, made material misstatements or omissions in the warrant affidavit[s] that were necessary to the finding of probable cause.” Id. at 389; see Franks v. Delaware, 438 U.S. 154, 155-56 (1978). The court identified three such material omissions: the affidavits’ failure to disclose that (1) the defendants “had not spoken with Officer Hebert about his conversation” with Walczyk, (2) none of Walczyk’s previous arrests “had resulted in a conviction for threatening,” and (3) Walczyk “had not lived at 27 Tunxis Street for seven years.” Walczyk v. Rio, 339 F. Supp. 2d at 389. The court concluded that “[a] reasonable juror could find that the omission of the first two items of information was critical to the finding of probable cause for the arrest, and that the omission of all three items was critical to the finding of probable cause for the searches of the houses and the seizures of the firearms.” Id. Thus, it ruled that defendants were not entitled to summary judgment on the basis of qualified immunity under either federal or state law (assuming arguendo that 17
    231. Connecticut would afford immunity to state constitutional claims14) because, when the record was viewed in the light most favorable to the plaintiffs, “a jury could find that the defendants lacked even arguable probable cause to believe that [Walczyk] had committed the crime of threatening or that the firearms to be seized were connected with criminal activity.” Id. at 390 (noting that ruling was consistent with Connecticut Appellate Court’s determination that affidavits “fell well short of establishing probable cause”).15 Nevertheless, the district court 14 We need not decide on this appeal whether Connecticut affords qualified immunity in cases of unlawful searches or seizures. See Rustici v. Malloy, No. cv-970164460S, 2004 Conn. Super. LEXIS 1734, at *46 n.26 (Conn. Super. Ct. July 1, 2004) (assuming that “qualified immunity appl[ies] to state constitutional claims”). Our holding that probable cause supports the arrest of Walczyk and the search of his home, see infra at [31-38], obviates the need for an immunity shield — state or federal — on those claims. See, e.g., Weyant v. Okst, 101 F.3d 845, 852 (2d Cir. 1996) (“The existence of probable cause to arrest constitutes justification and is a complete defense to an action for false arrest, whether that action is brought under state law or under § 1983.” (internal quotation marks and citation omitted)); Beinhorn v. Saraceno, 23 Conn. App. 487, 492, 582 A.2d 208, 210-11 (Conn. App. Ct. 1990) (stating that claim for false arrest under Connecticut law requires proof that arresting officer lacked probable cause). As to Elizabeth Walczyk’s unlawful search claim, because we identify factual issues, see infra at [41-43], that could moot defendants’ qualified immunity claim, we conclude that the availability of state law immunity need not be addressed unless and until these issues are resolved favorably to the defendants. 15 To the extent Captain Rio claimed that the evidence was insufficient to ascribe supervisory responsibility to him for any warrant deficiencies, the district court concluded that sufficient evidence had been adduced to support a jury finding that Rio was actually “involved in preparing the warrant affidavit.” Walczyk v. Rio, 339 F. Supp. 2d at 390. Because we agree with this conclusion, see Hayut v. State Univ. of New York, 352 F.3d 733, 753 (2d Cir. 2003) (noting that supervisory liability is supported by “evidence of a supervisory official’s personal involvement in the challenged conduct” (internal quotation marks omitted)), it is unnecessary for us to distinguish between Rio and the other defendants in discussing the probable cause question that is the crux of plaintiffs’ unlawful search and arrest challenges. 18
    232. denied Elizabeth Walczyk’s cross-motion for summary judgment on the liability phase of her search warrant challenge, concluding that she had not shown that no reasonable juror could find defendants’ acts lawful. II. Discussion A. Jurisdiction and the Standard of Review Because the denial of a motion for summary judgment is not a final judgment, it is generally not immediately appealable. See, e.g., Jones v. Parmley, 465 F.3d 46, 54 (2d Cir. 2006). An exception obtains, however, when the denied motion was based on a claim of immunity, at least to the extent the immunity claim presents a “purely legal question.” Mitchell v. Forsyth, 472 U.S. 511, 530 (1985) (recognizing jurisdiction to review “purely legal question on which . . . claim of immunity turns”); see O’Bert ex rel. Estate of O’Bert v. Vargo, 331 F.3d 29, 38 (2d Cir. 2003) (observing that “[u]nder the collateral order doctrine . . . the denial of a qualified-immunity-based motion for summary judgment is immediately appealable to the extent that the district court has denied the motion as a matter of law, although not to the extent that the defense turns solely on the resolution of questions of fact”); accord Jones v. Parmley, 465 F.3d at 54. The rationale for this exception is the law’s recognition that immunity shields a defendant from suit itself, not merely from liability. See Saucier v. Katz, 533 U.S. 194, 199 (2001) (“The privilege is ‘an immunity from suit rather than a mere defense to liability; . . . it is effectively lost if a case is erroneously permitted to go to trial.’” (quoting Mitchell v. Forsyth, 472 U.S. at 526 (emphasis in original))). In this 19
    233. case, defendants’ appeal from the district court’s denial of qualified immunity on plaintiffs’ search and arrest claims can be decided as a matter of law; accordingly, our jurisdiction is established. Although an interlocutory appeal would not be available from either the denial of Elizabeth Walczyk’s motion for summary judgment on her unlawful search claim or the district court’s dismissal of Walczyk’s excessive bail claim on the ground of absolute immunity, we elect to exercise pendent jurisdiction over both. “[W]e may exercise pendent jurisdiction over . . . issues that are not ordinarily subject to interlocutory review only when: (1) they are ‘inextricably intertwined’ with the determination of qualified immunity; or (2) their resolution is ‘necessary to ensure meaningful review’ of the district court’s ruling on qualified immunity.” Jones v. Parmley, 465 F.3d at 64 (quoting Swint v. Chambers County Comm’n, 514 U.S. 35, 51 (1995)). As we explain further below, see infra at [42-43], Elizabeth Walczyk’s claim is “inextricably intertwined” with the determination of defendants’ entitlement to qualified immunity in that the same disputed factual issues that preclude a finding of qualified immunity on this claim at this stage also make summary judgment inappropriate. As to Walczyk’s excessive bail claim, if we were to determine that the district court improperly granted defendants absolute immunity, they might nonetheless be entitled to qualified immunity, and thus review of this determination is also inextricably intertwined with our resolution of defendants’ entitlement to qualified immunity. Cf. Clynch 20
    234. v. Chapman, 285 F. Supp. 2d 213, 219 n.6 (D. Conn. 2003) (characterizing absolute immunity in excessive bail claim as a “cousin issue” to qualified immunity). We review de novo defendants’ legal challenge to the district court’s qualified immunity ruling, see Jones v. Parmley, 465 F.3d at 55, as well as Elizabeth and Thomas Walczyks’ cross-appeals from other summary judgment rulings, see, e.g., Root v. Liston, 444 F.3d 127, 130 (2d Cir. 2006). B. Plaintiffs’ Search and Arrest Claims 1. The Qualified Immunity Standard When a defendant officer charged with violations of federal constitutional rights invokes qualified immunity to support a motion for summary judgment, a court must first consider a threshold question: Do the facts, viewed in the light most favorable to the plaintiff, show that the officer’s conduct violated a constitutional right? If the answer to this question is no, “there is no necessity for further inquiries concerning qualified immunity.” Saucier v. Katz, 533 U.S. at 201; see X-Men Sec., Inc. v. Pataki, 196 F.3d 56, 66 (2d Cir. 1999) (observing that resolution of this first question favorable to defendant “moots” further inquiry into qualified immunity). The reason for this rule is that, where there is no viable constitutional claim, defendants have no need of an immunity shield. See generally Farrell v. Burke, 449 F.3d 470, 499 n.14 (2d Cir. 2006) (“Because we have found no cognizable violation of [p]laintiff’s rights in this case, we need not reach the question of qualified immunity.”); Holcomb v. Lykens, 337 F.3d 217, 223-25 (2d Cir. 2003) (declining to decide 21
    235. qualified immunity question and affirming summary judgment on ground that, as a matter of law, defendants did not violate plaintiff’s due process rights). Only if the answer to the first question is yes must a court proceed to the inquiry for qualified immunity: Was the right at issue clearly established at the time of the defendant’s actions? As the Supreme Court has explained, this question is not answered by reference to how courts or lawyers might have understood the state of the law: “The relevant, dispositive inquiry in determining whether a right is clearly established is whether it would be clear to a reasonable officer that his conduct was unlawful in the situation he confronted.” Saucier v. Katz, 533 U.S. at 202 (emphasis added). If the right at issue was not clearly established by then existing precedent, then qualified immunity shields the defendant. Even if the right at issue was clearly established in certain respects, however, an officer is still entitled to qualified immunity if “officers of reasonable competence could disagree” on the legality of the action at issue in its particular factual context. Malley v. Briggs, 475 U.S. 335, 341 (1986); accord Iqbal v. Hasty, – F.3d –, 2007 WL 1717803, at * 20 (2d Cir. June 14, 2007). Cerrone v. Brown, 246 F.3d 194, 202 (2d Cir. 2001); Lennon v. Miller, 66 F.3d 416, 420 (2d Cir. 1995); see Saucier v. Katz, 533 U.S. at 208 (holding officer entitled to qualified immunity if “[a] reasonable officer in [his] position could have believed that [the challenged conduct] was within the bounds of appropriate police responses”). In this respect, the Supreme Court has observed that qualified immunity protects “all but the plainly incompetent or those who knowingly violate the law.” Malley v. Briggs, 475 U.S. at 341 (quoted 22
    236. approvingly in Saucier v. Katz, 533 U.S. at 202). 16 16 Our concurring colleague takes exception to these standards at the same time that she acknowledges their reiteration of well established precedent. Her concern that a bifurcation of the “clearly established” inquiry might allow a defendant to secure qualified immunity for conduct already held unconstitutional “in the particularized sense,” post at [SS Concurrence at 6-7] is, in fact, unwarranted. If controlling authority has already established the unlawfulness of the challenged conduct in the particularized circumstances presented in the pending case, then no reasonable officer could think otherwise and, thus, qualified immunity would not shield the defendant. See, e.g., Groh v. Ramirez, 540 U.S. 551, 564 (2004). Judge Sotomayor’s further criticism — that determining clearly established law by reference to disagreements among reasonably competent officers, as indicated in Malley v. Briggs, 475 U.S. at 341, is “more permissive of defendants” than a single reasonable officer standard, post at [SS Concurrence at 8]— might merit attention if Malley contemplated “officers of reasonable competence” disagreeing based on unreasonable views of existing law. In fact, neither Malley nor today’s decision supports that conclusion. Instead, what Malley does is provide courts with a useful tool for assessing when pre-existing law that did not recognize the invoked right in the particularized context at issue, nevertheless, “must” have alerted the defendant to the unlawfulness of his action. See Anderson v. Creighton, 483 U.S. 635, 640 (1987) (holding that, where particular action at issue has not previously been held unlawful, defendant is not entitled to qualified immunity if, “in the light of pre-existing law the unlawfulness must [have] be[en] apparent”). By instructing courts to focus on whether “officers of reasonable competence could disagree” about the illegality of the challenged conduct, Malley sounds a useful reminder: because law enforcement work relies on probabilities and reasonable suspicions in an almost infinite variety of circumstances, many requiring prompt action, there can frequently be a range of responses to given situations that competent officers may reasonably think are lawful. W ithin this range, an officer enjoys qualified immunity for “reasonable mistakes.” Saucier v. Katz, 533 U.S. at 205, 206. To the extent Judge Sotomayor’s objection to Malley’s formulation relies on recent Supreme Court habeas jurisprudence, we note that the considerations informing limitations on habeas review are sufficiently distinct from those prompting recognition of qualified immunity to preclude easy analogy. See, e.g., Harlow v. Fitzgerald, 457 U.S. 800, 807 (1982) (recognizing qualified immunity because of “need to protect officials who are required to exercise their discretion and the related public interest in encouraging the vigorous exercise of official authority”). Nothing in the Supreme Court’s qualified immunity jurisprudence signals a narrowing of these considerations or an abandonment of Malley’s analysis. To the contrary, as we note in text, Saucier v. Katz cites approvingly to Malley’s 23
    237. 2. The Legality of the Challenged Arrest and Searches Applying these principles to this case, we consider first whether defendants’ actions violated plaintiffs’ rights under both the United States and Connecticut Constitutions to be free from unreasonable searches and arrests. See U.S. Const. amend. IV;17 Conn. Const. art. First, §§ 7, 9.18 Ordinarily, an arrest or search pursuant to a warrant issued by a neutral magistrate is presumed reasonable because such warrants may issue only upon a showing of probable cause. See Franks v. Delaware, 438 U.S. at 171; United States v. Awadallah, 349 F.3d 42, 64 (2d Cir. 2003); Golino v. City of New Haven, 950 F.2d 864, 870 (2d Cir. 1991); observation that qualified immunity is intended to shield “‘all but the plainly incompetent or those who knowingly violate the law.’” 533 U.S. at 202 (quoting Malley v. Briggs, 475 U.S. at 341). Finally, insofar as Judge Sotomayor expresses some concern about courts contemplating persons reaching different reasonable conclusions about the same facts, we note simply that courts do so routinely in upholding jury verdicts as long as “any rational trier of fact” could have reached the challenged result. United States v. MacPherson, 424 F.3d 183, 187 (2d Cir. 2005) (and cases cited therein). Accordingly, we hold that courts may continue to rely on Malley in resolving qualified immunity disputes. 17 “The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.” U.S. Const. amend. IV. 18 “The people shall be secure in their persons, houses, papers and possessions from unreasonable searches or seizures; and no warrant to search any place, or to seize any person or things, shall issue without describing them as nearly as may be, nor without probable cause supported by oath or affirmation.” Conn. Const. art. First, § 7. “No person shall be arrested, detained or punished, except in cases clearly warranted by law.” Id. art. First, § 9. 24
    238. see also United States v. Leon, 468 U.S. 897, 913-14 (1984). Plaintiffs nevertheless insist that the presumption is defeated in this case because (1) the warrant affidavits, on their face, fail to demonstrate probable cause, see generally United States v. Leon, 468 U.S. at 923; and (2) the issuing magistrate was, in any event, misled into finding probable cause by material omissions for which defendants were knowingly or recklessly responsible, see Franks v. Delaware, 438 U.S. at 155-56; Golino v. City of New Haven, 950 F.2d at 870-71. a. The Probable Cause Standard Before discussing these two contentions, we observe that federal and Connecticut law are identical in holding that probable cause to arrest exists when police officers have “knowledge or reasonably trustworthy information of facts and circumstances that are sufficient to warrant a person of reasonable caution in the belief that the person to be arrested has committed or is committing a crime.” Weyant v. Okst, 101 F.3d 845, 852 (2d Cir. 1996); see State v. James, 261 Conn. 395, 415, 802 A.2d 820, 833 (2002) (“Probable cause exists when the facts and circumstances within the knowledge of the officer and of which he has reasonable trustworthy information are sufficient in themselves to warrant a man of reasonable caution to believe that a [crime] has been committed.” (internal quotation marks omitted)). Similarly, under both federal and state law, probable cause to search is demonstrated where the totality of circumstances indicates a “fair probability that contraband or evidence of a crime will be found in a particular place.” Illinois v. Gates, 462 U.S. 213, 238 (1983); accord United States v. Gaskin, 364 F.3d 438, 456 (2d Cir. 2004); State v. 25
    239. Vincent, 229 Conn. 164, 171, 640 A.2d 94, 98 (1994) (noting that valid search requires “probable cause to believe that the particular items to be seized are connected with criminal activity or will assist in a particular apprehension or conviction” and “that the items sought to be seized will be found in the place to be searched”); State v. Orellana, 89 Conn. App. 71, 80, 872 A.2d 506, 516 (Conn. App. Ct. 2005) (citing Illinois v. Gates, 462 U.S. at 231-32)). Accordingly, we need not separately discuss federal and state law in assessing probable cause for the challenged arrest and searches. As the Supreme Court has famously observed, probable cause is “a fluid concept . . . not readily, or even usefully, reduced to a neat set of legal rules.” Illinois v. Gates, 462 U.S. at 232; see United States v. Gaskin, 364 F.3d at 456. While probable cause requires more than a “mere suspicion,” of wrongdoing, Mallory v. United States, 354 U.S. 449, 455 (1957), its focus is on “probabilities,” not “hard certainties,” Illinois v. Gates, 462 U.S. at 231. In assessing probabilities, a judicial officer must look to “‘the factual and practical considerations of everyday life on which reasonable and prudent men, not legal technicians, act.’” Id. (quoting Brinegar v. United States, 338 U.S. 160, 175 (1949)); accord United States v. Gaskin, 364 F.3d at 456. “Finely tuned standards such as proof beyond a reasonable doubt or by a preponderance of the evidence, useful in formal trials, have no place” in a probable cause determination. Illinois v. Gates, 462 U.S. at 235. Nor can probable cause be analogized to a prima facie case. See id. (observing that “‘only the probability, and not a prima facie showing, of criminal activity’” is necessary to establish probable cause to search 26
    240. or arrest (quoting Spinelli v. United States, 393 U.S. 410, 419 (1969))). In sum, probable cause does not demand any showing that a good-faith belief be “correct or more likely true than false.” Texas v. Brown, 460 U.S. 730, 742 (1983). It requires only such facts as make wrongdoing or the discovery of evidence thereof probable. It has long been recognized that, where there is no dispute as to what facts were relied on to demonstrate probable cause, the existence of probable cause is a question of law for the court. See Stewart v. Sonneborn, 98 U.S. 187, 194 (1878) (observing that whether facts alleged to show probable cause are true is a matter of fact, “but whether, supposing them to be true, they amount to a probable cause, is a question of law” (internal quotation marks omitted)); accord Director Gen. of R.R.’s v. Kastenbaum, 263 U.S. 25, 28 (1923) (observing that where facts are in dispute, court submits the question of probable cause to the jury, but “with instructions as to what facts will amount to probable cause if proved”); Sanders v. Palmer, 55 F. 217, 220 (2d Cir. 1893) (holding that question whether facts, “supposing them to be true, . . . amount to a probable cause, is a question of law”; when the facts are disputed, “it is the duty of the court to instruct the jury what facts, if established, will constitute a probable cause . . . and to submit to them only the question as to the existence of those facts”); see also United States v. Awadallah, 349 F.3d at 65 (distinguishing between de novo review of legal question whether undisputed facts support probable cause and deferential review of factual question whether, in case of misstated or omitted material facts, affiant’s actions were deliberate or reckless). In this case, there can be no dispute as to what facts the 27
    241. defendants relied on to establish probable cause for the challenged arrest and searches; they are memorialized in warrant affidavits. Thus, whether the affidavits, on their face, demonstrate probable cause, is a question of law. In answering that question, however, a reviewing court must accord considerable deference to the probable cause determination of the issuing magistrate, see Illinois v. Gates, 462 U.S. at 238-39 (holding that duty of reviewing court “is simply to ensure that the magistrate had a substantial basis” for probable cause determination (internal quotation marks omitted)); accord Velardi v. Walsh, 40 F.3d 569, 574 n.1 (2d Cir. 1994), mindful of the well established principle that a showing of probable cause cannot be negated simply by demonstrating that an inference of innocence might also have been drawn from the facts alleged, see United States v. Webb, 623 F.2d 758, 761 (2d Cir. 1980); see also United States v. Forero-Rincon, 626 F.2d 218, 222 (2d Cir. 1980). To the extent plaintiffs argue, in addition to their facial challenge, that material omissions infected the issuing magistrate’s probable cause determination, there is no dispute between the parties as to what purported omissions are appropriately considered. They are the three non-disclosures identified by the district court, i.e., that (1) defendants had not spoken with Officer Hebert about his understanding of Walczyk’s bloodbath statement; (2) none of Walczyk’s prior arrests had resulted in a conviction for threatening, and (3) Walczyk had not lived at his parents’ home for seven years. See Walczyk v. Rio, 339 F. Supp. 2d at 389. The materiality of these omissions presents a mixed question of law and fact. See 28
    242. Velardi v. Walsh, 40 F.3d at 574. Whether omitted information is “relevant to the probable cause determination” is a question of law that we review de novo. Id. If we identify relevancy, then questions of fact may arise as to what “weight . . . a neutral magistrate would likely have given such information,” id., and whether defendants acted “deliberately or recklessly” in omitting the information from the warrant affidavits, United States v. Awadallah, 349 F.3d at 65 (internal quotation marks omitted). Even in such circumstances, however, a court may grant summary judgment based on qualified immunity where “the evidence, viewed in the light most favorable to the plaintiffs, discloses no genuine dispute that a magistrate would have issued the warrant on the basis of the ‘corrected affidavits.’” Velardi v. Walsh, 40 F.3d at 574 (emphasis in original). Mindful of these principles, we proceed to consider plaintiffs’ warrant challenges, focusing first on the arrest of Thomas Walczyk and the search of his 28 Tunxis Street residence and then on the search of Elizabeth Walczyk’s 27 Tunxis Street home. b. Thomas Walczyk (1) The Facial Challenge We reject as without merit Thomas Walczyk’s contention that the challenged warrant affidavits, on their face, fail to state probable cause for his arrest or the search of his 28 Tunxis Street home. The facts alleged establish probable cause to believe (1) that W alczyk had violated Connecticut law by “threaten[ing] to commit . . . [a] crime [of violence] in reckless disregard of the risk of causing . . . terror” to another person, Conn. Gen. Stat. § 53a- 29
    243. 62(a)(3), and (2) that Walczyk maintained in his residence firearms that, in light of past use, were relevant evidence that he intended to threaten violence and recklessly disregarded the threat’s terrorizing effect. The facts reveal that, on August 30, 1999, Walczyk complained to Officer Hebert that the police were not taking the action necessary to avoid a bloodbath. A reasonable person would understand the bloodbath reference as a prediction of probable violence between Walczyk and Barberino. More to the point, a reasonable person would understand from other facts alleged in the affidavits that Walczyk would likely be the person initiating any such violence. A few months earlier, Walczyk had stated to Captain Rio that, if the police did not assist him in his property dispute with Barberino, he would take matters into his “own hands,” doing whatever he “had to do” to protect his property rights. Rio knew that what Walczyk frequently put into his hands to resolve disputes were loaded firearms that he stored in his home. In the past, Walczyk had brandished firearms retrieved from his home at various individuals, including Barberino workers on two occasions. On one of those occasions, the brandished weapon was a loaded automatic rifle and,19 when police intervened, Walczyk initially defied their orders to put down the weapon and resisted arrest. Moreover, the affidavits demonstrated that Walczyk plainly knew how to fire his weapons; 19 Although the warrant affidavits note that Walczyk denied pointing the rifle at Barberino workers, the magistrate judge had probable cause to conclude otherwise given that one of the workers specifically “complained that Walczyk had pointed the gun directly at him.” Arrest Aff. at 3; Search Warrant Aff. and App. at 3. 30
    244. he had used them to kill a cat on his property. Rio further knew that Walczyk’s efforts to vindicate his property rights peaceably through the courts had failed. Under the totality of these circumstances, the issuing magistrate certainly had a substantial basis to conclude that, when Walczyk told police that their continued failure to assist him in his property dispute with Barberino would result in a bloodbath, he was effectively threatening to employ violence against Barberino employees with reckless disregard for the terror such a threat would cause when communicated to the intended victim. We are, of course, mindful that a Connecticut appellate court has ruled otherwise. Observing that Walczyk’s bloodbath statement was made to secure police assistance, that court concluded: “A statement to a police officer that the police needed to act to avoid a ‘bloodbath’ cannot be the basis of probable cause to believe that the defendant, at the time or in the immediate future, would engage in threatening behavior.” State v. W alczyk, 76 Conn. App. at 181-82, 818 A.2d at 876 (emphasis in original). We respectfully disagree. Walczyk may have desired police assistance in his land dispute, but how he sought to compel that assistance was by threatening violence. Walczyk was, after all, the only person to have used an instrument of violence in connection with the land dispute. Given his prior brandishing of loaded firearms, it was certainly probable that Walczyk’s bloodbath statement was a threat to use violence against Barberino workers if the police did not intervene in his favor (something they could not do in light of state court rulings). Whether Walczyk would, in fact, have acted on his threat is not determinative of whether it was probable that he had 31
    245. made the threat with reckless disregard of the terror it would cause Barberino.20 We conclude that the affidavits, on their face, state facts reasonably supporting such a finding by the issuing magistrate. Walczyk submits that the search warrant affidavit nevertheless failed to demonstrate that there was any connection between his present lawful possession of firearms and the alleged crime of threatening. We are persuaded that the warrant affidavit states probable cause to believe that a search of Walczyk’s home for firearms would produce evidence relevant to demonstrating that Walczyk had committed the offense of threatening. At the time the search warrant affidavit was prepared, Walczyk’s apparent possession of firearms constituted relevant evidence which could suggest that his intent in making the bloodbath remark was, in fact, to threaten violence. See State v. Crudup, 81 Conn. App. 248, 260 n.14, 838 A.2d 1053, 1062 n.14 (Conn. App. Ct. 2004) (discussing intent element of threatening). Specifically, a seizure of firearms from Walczyk’s home could have shown that, at the time Walczyk made the bloodbath remark, he had the actual capacity to cause bloodshed. Moreover, such a seizure following the authorized search could have served to corroborate witness accounts that Walczyk had used weapons against Barberino workers and others in the past, which in turn could have helped establish his reckless disregard of the bloodbath 20 Walczyk does not fault the police for communicating his bloodbath statement to Barberino, much less suggest that he did not intend or foresee such communication. Indeed, given the totality of circumstances, it may well have been irresponsible of the police not to have communicated the statement. 32
    246. remark’s terrorizing effect. In sum, Walczyk’s possession of firearms was evidence relevant to the mens rea element of the crime because a factfinder could reasonably infer from such possession and from Walczyk’s past use of firearms that his bloodbath statement was not idle hyperbole, but an intentional threat of violence made with reckless disregard of its potential to cause terror. As the search warrant affidavit makes clear, the police were aware that Walczyk had previously used his home to store the firearms he brandished in confrontations with others, including Barberino workers, and thus they had probable cause to believe that evidence relevant to his alleged threatening would turn up in a search of his home. Accordingly, we hold that plaintiffs’ facial challenge to the warrant affidavits in this case necessarily fails as a matter of law.21 (2) Purported Omissions Walczyk asserts that two of the three identified material omissions misled the issuing magistrate into erroneously finding probable cause to support his arrest and the search of his home. 21 We also reject as without merit plaintiffs’ argument that the search warrant, which simply sought “Firearms,” was insufficiently particular and that the seizure of ammunition, gun clips, ammunition belts, and other items not named in the warrant violated the Fourth Amendment. Although requests to search for “evidence of a crime” violate the proscription against general warrants, see Groh v. Ramirez, 540 U.S. 551, 557-58 (2004), defendants’ application to search for “Firearms” was sufficiently particular because any firearms in Walczyk’s possession were relevant evidence that his bloodbath remark was an intentional threat of violence. Because we reach the same relevancy conclusion with respect to the ammunition, gun clips, and related firearms paraphernalia found in Walczyk’s home, we conclude that these items were properly seized under the “plain view” doctrine. United States v. $557,933.89, 287 F.3d 66, 81 (2d Cir. 2002). 33
    247. (a) Failure to Speak With Officer Hebert First, the district court pointed to defendants’ failure to disclose that no officer had spoken directly with Officer Hebert, who would have revealed that he did not himself feel threatened by Walczyk’s bloodbath statement. The conclusion is unconvincing both as a matter of law and fact. Preliminarily, we observe that the law permitting one law enforcement officer to rely on the report of another in applying for a warrant nowhere requires direct consultation to ensure that the officer reviewing the report ascribes no more weight to the described facts than the reporter intended. See generally Panetta v. Crowley, 460 F.3d 388, 395 (2d Cir. 2006); Martinez v. Simonetti, 202 F.3d 625, 634 (2d Cir. 2000); Velardi v. Walsh, 40 F.3d at 574. Indeed, we have specifically ruled that “a police officer is not required to explore and eliminate every theoretically plausible claim of innocence before making an arrest.” Martinez v. Simonetti, 202 F.3d at 635 (internal quotation marks omitted). Thus, we reject the suggestion that a law enforcement officer is guilty of a material omission when, in applying for a warrant, he fails to disclose that he has not spoken directly with a fellow officer on whose report he relies to establish probable cause. Even if we were to assume, however, that the applicant officer’s failure to ascertain that the reporting officer did not consider words he heard to be threatening was relevant to a determination of probable cause, that is not this case. Officer Hebert did understand Walczyk’s bloodbath statement as a threat. At Walczyk’s criminal trial, Hebert testified that 34
    248. he did not understand Walczyk to be threatening him, but he most certainly did understand Walczyk to be threatening any Barberino employees who came onto the disputed property. As we have already noted, this understanding finds ample support in the totality of facts recounted in the warrant affidavits. In sum, because Hebert’s subjective view of Walczyk’s statement, if it had been solicited by defendants and reported to the issuing magistrate, would actually have reinforced rather than undermined probable cause, we conclude that a “corrected” warrant affidavit would raise no genuine dispute as to the magistrate’s issuance of warrants for the arrest of Walczyk or the search of his home. (b) The Lack of a Prior Conviction for Threatening A second purported omission is the defendants’ failure to disclose that none of Walczyk’s prior conduct had resulted in a conviction for threatening. This omission is hardly relevant. Certainly, nothing in the challenged affidavits wrongly insinuates that Walczyk had such a conviction. Absent such conduct, we expect that when a magistrate, mindful of the government’s burden to demonstrate probable cause, reviews a warrant application that does not report a prior conviction for a particular crime, the magistrate assumes for purposes of determining whether the government has carried its burden that no such conviction exists. See generally Minnesota v. Dickerson, 508 U.S. 366, 376 (1993) (noting that probable cause requirement ensures against government action based on speculation). Moreover, as this court observed in Brown v. D’Amico, the law does not demand that an officer applying for a warrant “volunteer every fact that arguably cuts against the existence of probable cause,” 35
    249. as long as he does “not omit circumstances that are critical” to its evaluation, 35 F.3d 97, 99 (2d Cir. 1994). It was the particulars of Walczyk’s past conduct in using or threatening to use firearms to resolve disputes that was critical to the determination of the probability that his bloodbath statement constituted a threat of violence. This probability is in no way undermined by the lack of a prior conviction for threatening. Because we identify no merit in Thomas Walczyk’s facial challenge to the warrant affidavits authorizing his arrest and the search of his home, and because we determine as a matter of law that no alleged omissions were material to the issuance of these warrants, we conclude that he (as well as his wife and daughter) fails to demonstrate a viable unlawful search or arrest claim under federal or state law. Accordingly, we reverse the district court order denying defendants’ summary judgment with respect to these plaintiffs’ unlawful search and arrest claims, and we remand with directions to enter such a judgment. c. Elizabeth Walczyk (1) The Lack of Probable Cause According to defendants, the theory for searching Elizabeth Walczyk’s residence was that it probably contained firearms accessible to her son, constituting some further evidence that his bloodbath statement was a threat of violence. To the extent Elizabeth Walczyk joins in her son’s facial challenge to the warrant affidavits and to his charged material omissions regarding Officer Hebert and Walczyk’s criminal record, we have already explained why we reject these arguments. The district court, however, identified another omission that raises 36
    250. greater concern with respect to the search of Elizabeth Walczyk’s home. The warrant affidavit reported that Thomas Walczyk was licensed to possess various firearms and that he maintained two neighboring residences where such firearms would likely be found: “[A] review of Farmington Police records indicate[s] that Walczyk has maintained residences at both 27 and 28 Tunxis St., Farmington, CT.” Search Warrant Aff. and App. at 3. The implication was that Walczyk had maintained the residences recently. What the affidavit omitted, however, was the apparently undisputed fact that Walczyk had not resided at his mother’s 27 Tunxis Street residence for more than seven years. There can be no question that the omitted information was relevant to any assessment of probable cause. In evaluating probable cause, a magistrate is always required to consider whether the facts adduced in the warrant application “appear[] to be current, i.e., true at the time of the application,” or whether they have “become stale.” Rivera v. United States, 928 F.2d 592, 602 (2d Cir. 1991). The law sensibly draws no bright-line rule for staleness. Rather, a magistrate is expected to consider the age of the facts in light of the conduct at issue with a view toward ensuring that probable cause exists at the time the warrant is to be executed, not simply at some past time. See id.; see also United States v. Martino, 664 F.2d 860, 867 (2d Cir. 1981) (observing that, in circumstances of continuing or ongoing conduct, as contrasted with isolated illegal acts, “the passage of time between the last described act and the presentation of the application becomes less significant”). Thus, where information is seven years old, a magistrate must be alerted to that fact to make a reasonable probable 37
    251. cause determination. As we have already observed, the question of what weight a magistrate would have given omitted relevant evidence is generally a question for the finder of fact. See Velardi v. Walsh, 40 F.3d at 574. In this case, however, we can conclude as a matter of law that non- disclosure of the staleness of the dual residency allegation was fatal to a demonstration of probable cause. Not only was the allegation seriously outdated, it was the sole support for a search of Elizabeth W alczyk’s home. A comparison best makes this point. With respect to 28 Tunxis Street, the issuing magistrate could have inferred that Walczyk maintained guns at that address because he was licensed to possess such weapons and, on at least one occasion, he was actually seen retrieving a loaded assault rifle from that premises to brandish at a person with whom he had a dispute. Further, neighbors had reported seeing him fire a handgun on that property. By contrast, no facts were alleged indicating that Walczyk had ever stored or retrieved firearms from his parents’ home, much less that he had done so in the seven years since last residing there. Defendants urged that inference simply from Walczyk’s license to possess firearms and his purported residence at 27 as well as 28 Tunxis Street. Whatever questions might be raised about the strength of such an inference in any circumstance, it could not be drawn from a dual residency allegation that was seven years old. Defendants submit that the search of Walczyk’s 28 Tunxis Street residence “would have been meaningless” if he were “free to cross the street to his parents[’] home, where he 38
    252. formerly lived for many years, and have free access to other weaponry.” Appellant’s Br. at 30. This argument overlooks the fact that the warrant affidavit fails to allege any facts — apart from Walczyk’s dual residency — suggesting that anyone ever stored guns at 27 Tunxis Street. The affidavit makes no mention of the fact that Lucien or Elizabeth Walczyk was licensed to possess guns. As we have recently reiterated, probable cause to search must be based on particularized information about the place to be searched, not simply on a target’s “‘mere propinquity to others independently suspected of criminal activity.’” United States v. Martin, 426 F.3d 86, 88 (2d Cir. 2005) (quoting Ybarra v. Illinois, 444 U.S. 85, 91 (1979)). Once the dual residency allegation is “corrected,” we can conclude as a matter of law that the affidavit is devoid of any particularized information establishing probable cause to search Elizabeth Walczyk’s home. (2) Defendants’ Claim of Qualified Immunity; Elizabeth Walczyk’s Motion for Summary Judgment on the Issue of Liability Despite our ruling that the search of Elizabeth Walczyk’s home was not supported by probable cause, defendants might still be entitled to claim qualified immunity from liability for damages if the search was supported by “arguable probable cause.” Escalera v. Lunn, 361 F.3d 737, 743 (2d Cir. 2004) (holding that, even in the absence of probable cause, officer “will still be entitled to qualified immunity from a suit for damages if he can establish that there was ‘arguable probable cause’”). “Arguable probable cause exists ‘if either (a) it was objectively reasonable for the officer to believe that probable cause existed, or (b) officers 39
    253. of reasonable competence could disagree on whether the probable cause test was met.’” Id. (quoting Golino v. City of New Haven, 950 F.2d at 870); see also Caldarola v. Calabrese, 298 F.3d 156, 162 (2d Cir. 2002). Like the district court, we conclude that questions of disputed fact preclude a judicial resolution of this issue in favor of either side. While no competent officer who knew that Thomas Walczyk had not resided in his parents’ home for more than seven years could reasonably think that the stale allegation of dual residency established probable cause to search that premises for firearms, it cannot be determined from the present record which officers, if any, possessed — or even should have possessed — such knowledge. Further record development and factfinding are necessary to determine, among other things, (1) which of the defendants, if any, actually reviewed the police records that purportedly established W alczyk’s residence at both 27 and 28 Tunxis Street; (2) whether these records would have alerted a reasonable officer to the staleness of the dual residency allegation; (3) which defendants, if any, possessed independent knowledge that the dual residency statement was not accurate; (4) the circumstances under which the dual residency allegation was communicated along the chain of defendants; and (5) whether any defendant’s reliance on such communication without further inquiry was reasonable in light of the totality of the circumstances.22 Because a resolution of some of these matters in 22 As the Supreme Court has explained: [P]olice officers called upon to aid other officers in executing . . . warrants are entitled to assume that the officers requesting aid offered the magistrate the information requisite to support an independent judicial assessment of 40
    254. favor of Elizabeth Walczyk could preclude one or more defendants from claiming that they acted with arguable probable cause in searching her home, the district court correctly concluded that defendants did not yet establish their entitlement to qualified immunity. On the other hand, because a resolution favorable to one or more defendants could afford them the benefit of a qualified immunity shield, the district court also correctly denied summary judgment on liability to Elizabeth W alczyk. Accordingly, we affirm both these rulings. C. Walczyk’s Excessive Bail Claim Connecticut law allows state police to set temporary bail in certain cases. See Conn. Gen. Stat. § 54-63c(a). 23 The district court dismissed Walczyk’s excessive bail claim as a matter of law, holding that “when a police officer sets temporary bail” pursuant to this statute, “he performs a judicial function and hence has absolute immunity from suit.” probable cause. Where, however, the contrary turns out to be true, an otherwise illegal arrest [or search] cannot be insulated from challenge by the decision of the instigating officer to rely on fellow officers to make the arrest [or search]. Whiteley v. Warden, 401 U.S. 560, 568 (1971); see also Varrone v. Bilotti, 123 F.3d 75, 81 (2d Cir. 1997) (concluding that subordinate officers carrying out search directive of superior were entitled to qualified immunity although none had determined basis for order or reliability of information on which it was based). 23 The statute, in pertinent part, requires a police officer “promptly [to] order release of the arrested person upon the execution of a written promise to appear or the posting of such bond as may be set by the police officer, except that no condition of release set by the court or a judge thereof may be modified by such officer.” Conn. Gen. Stat. § 54-63c(a) (emphasis added). An officer may set bail only after conducting an interview with the individual concerning the terms and conditions of release, at which interview counsel may be present. See id. 41
    255. Walczyk v. Rio, 339 F. Supp. 2d at 390. Other district courts in Connecticut have similarly ruled. See Sanchez v. Doyle, 254 F. Supp. 2d 266, 269-73 (D. Conn. 2003); accord Machuca v. Canning, No. 3:00-cv-1722, 2006 WL 2828160, at *6 (D. Conn. Sept. 29, 2006); Minney v. Kradas, No. 3:01-cv-1543, 2004 U.S. Dist. LEXIS 5520, at *9-11 (D. Conn. Mar. 31, 2004); Bacciocchi v. Chapman, No. 3:02-cv-1403, 2004 U.S. Dist. LEXIS 1077, at *18-20 (D. Conn. Jan. 26, 2004); Clynch v. Chapman, 285 F. Supp. 2d at 219-23. Citing Sanchez and Clynch by analogy, this court recently concluded that absolute judicial immunity shielded a prosecutor who ordered a defendant’s bond increased pursuant to Conn. Gen. Stat. § 54- 63d because the prosecutor was performing a judicial function. See Root v. Liston, 444 F.3d at 132 (observing that courts apply “functional approach to immunity questions”). Following Root, we now hold what the citations to Sanchez and Clynch implied: police officers setting bail under Conn. Gen. Stat. § 54-63c(a) are engaged in a judicial function that affords them absolute immunity. “It is . . . well established that officials acting in a judicial capacity are entitled to absolute immunity against § 1983 actions, and this immunity acts as a complete shield to claims for money damages.” Montero v. Travis, 171 F.3d 757, 760 (2d Cir. 1999) (extending absolute immunity to parole board officials performing a quasi-judicial function in making parole decisions); see also Butz v. Economou, 438 U.S. 478, 511 (1978) (granting absolute immunity to administrative hearing examiners performing adjudicatory functions within federal agencies). As the Supreme Court has explained, it is “the nature of the function 42
    256. performed, not the identity of the actor who performed it, that inform[s] our immunity analysis.” Forrester v. White, 484 U.S. 219, 229 (1988) (holding that judges do not enjoy absolute immunity when performing administrative, legislative, or executive functions). Following this “functional approach to immunity questions,” this court in Root v. Liston observed that “[o]rdinarily, it is judges who set bail, and judges enjoy absolute immunity when they do so.” 444 F.3d at 132 (internal citations omitted). In short, because the setting of bail is a judicial function, see Cleavinger v. Saxner, 474 U.S. 193, 205 (1985), absolute immunity extends to police officers when they perform that function pursuant to statute. Accordingly, we affirm the district court’s dismissal of Walczyk’s excessive bail claim. III. Conclusion To summarize: 1. The unlawful search and arrest challenges of Thomas, Maximina, and Michelle Walczyk are without merit as a matter of law because the warrants for Walczyk’s arrest and for the search of these plaintiffs’ home were supported by probable cause. Defendants are entitled to have summary judgment entered in their favor on these claims. 2. The warrant authorizing the search of Elizabeth Walczyk’s home was plainly not supported by probable cause; nevertheless, the district court correctly denied summary judgment to both Elizabeth Walczyk and defendants because disputed questions of fact must be resolved before it can be determined whether defendants’ actions are shielded by qualified 43
    257. immunity or whether plaintiff is entitled to have a liability judgment entered in her favor. 3. Summary judgment was correctly entered in favor of defendants on Thomas Walczyk’s claim of excessive bail because police officers, when setting bail pursuant to Conn. Gen. Stat. § 54-63c(a), perform a judicial function, which affords them absolute immunity from suit for money damages. That part of the district court order denying defendants’ summary judgment motion with respect to the unlawful search and arrest claims of plaintiffs Thomas, Maximina, and Michelle Walczyk is hereby R EVERSED and the case is R EMANDED for entry of such a judgment. In all other respects the appealed summary judgment rulings of the district court are A FFIRMED. 44
    258. Sotomayor, J., concurring: I agree fully with the outcome of this case, and I concur with most of the majority’s reasoning; however, I disagree with its description of the qualified immunity standard we should apply and its related discussion of “arguable probable cause.” A long line of decisions of this Court features the same doctrinal misstatements, and it is time we stopped repeating uncritically this particular language and gave it the attention it deserves.1 I join all of the majority opinion except Part II(B)(1),2 and I write separately to call the Court’s attention to our collective failure to harmonize our qualified immunity analysis with the Supreme Court’s directives. The portion of the majority’s qualified immunity discussion that I find objectionable reads as follows: “If the right at issue was not clearly established by then existing precedent, then qualified immunity shields the defendant. Even if the right at issue was clearly established in certain respects, however, an officer is still entitled to qualified immunity if ‘officers of reasonable competence could disagree’ on the legality of the action at issue in its particular factual context.” Maj. Op. at [23] (quoting Malley v. Briggs, 475 U.S. 335, 341 (1986)). These two sentences and the citation to Malley reveal the two flaws I see in this circuit’s approach to qualified immunity. First, our approach splits the single question of whether a right is “clearly established” into two distinct steps, 1 1 To the extent the majority holds that courts may continue to rely on this language 2 from Malley, Maj. Op. at [25 n.16], I note that such a holding is unnecessary and serves no 3 purpose in this case except to complicate the law further. 2 1 Although I also disagree with the majority’s use of the term “arguable probable 2 cause” and its reliance on whether “officers of reasonable competence could disagree,” M aj. 3 Op. at [41], I join Part II(B)(2)(c)(2) of the majority opinion because I agree with its 4 conclusion that questions of disputed fact preclude judicial resolution of whether the officers 5 are entitled to qualified immunity for their search of Elizabeth Walczyk’s house. 45
    259. contrary to Supreme Court precedent. Second, we demand a consensus among all hypothetical reasonable officers that the challenged conduct was unconstitutional, rather than positing an objective standard of reasonableness to which defendant officers should be held, as the Supreme Court has repeatedly instructed us to do. I address both of these points in turn. The Supreme Court has made clear that “[t]he relevant, dispositive inquiry in determining whether a right is clearly established is whether it would be clear to a reasonable officer that his [or her] conduct was unlawful in the situation he [or she] confronted.” Saucier v. Katz, 533 U.S. 194, 202 (2001). That is, whether a right is clearly established is the same question as whether a reasonable officer would have known that the conduct in question was unlawful. This Court’s case law, in contrast, bifurcates the “clearly established” inquiry into two steps. See, e.g., Cerrone v. Brown, 246 F.3d 194, 199 (2d Cir. 2001) (“A police officer is entitled to qualified immunity from liability for his [or her] discretionary actions if either (1) his [or her] conduct does not violate clearly established statutory or constitutional rights of which a reasonable person would have known, or (2) it was objectively reasonable for him [or her] to believe that his [or her] actions were lawful at the time of the challenged act.” (emphasis added; internal quotation marks and citation omitted)). By splitting the “relevant, dispositive inquiry” in two, we erect an additional hurdle to civil rights claims against public officials that has no basis in Supreme Court precedent. Whether a reasonable officer would know his or her conduct to be unlawful requires an inquiry into the state of the law at the time of the conduct and “in light of the specific context of the case.” Saucier, 533 U.S. at 201. If the right at issue has not previously been articulated, or had been addressed only in a factual context that is “distinguishable in a fair way,” id. at 202, a reasonable 46
    260. officer might not have known that the challenged conduct was unlawful.3 See also id. (“‘[T]he right the official is alleged to have violated must have been “clearly established” in a more particularized, and hence more relevant, sense: The contours of the right must be sufficiently clear that a reasonable official would understand that what he [or she] is doing violates that right.’” (quoting Anderson v. Creighton, 483 U.S. 635, 640 (1987)). Contrary to what our case law might suggest, the Supreme Court does not follow this “clearly established” inquiry with a second, ad hoc inquiry into the reasonableness of the officer’s conduct. Once we determine whether the right at issue was clearly established for the particular context that the officer faced, the qualified immunity inquiry is complete. Wilson v. Layne, 526 U.S. 603 (1999), illustrates the inquiry that the Supreme Court contemplates for qualified immunity. The Wilsons brought suit against law enforcement officers who permitted members of the media to accompany them in an early morning raid of the Wilsons’ home. Id. at 607. Having concluded that the officers’ actions violated the Fourth Amendment, the Court considered whether the officers were entitled to qualified immunity. Id. at 614. The Court explained that “what ‘clearly established’ means in this context depends largely ‘upon the level of generality at which the relevant “legal rule” is to be identified.’” Id. at 614 (quoting Anderson, 483 U.S. at 639). That is, to be clearly established, “[t]he contours of the right must be sufficiently clear 3 1 The Supreme Court has stated, however, that the “clearly established” standard does 2 not mean that “an official action is protected by qualified immunity unless the very action in 3 question has previously been held unlawful,” Anderson v. Creighton, 483 U.S. 635, 640 4 (1987), nor does the standard necessarily require that the facts of earlier cases be “materially 5 similar” to the case under consideration, Hope v. Pelzer, 536 U.S. 730, 739-41 (2002). The 6 standard is one of “fair warning,” id. at 741, such that “unlawfulness must be apparent” in 7 light of pre-existing law, Anderson, 483 U.S. at 640. 47
    261. that a reasonable official would understand that what he [or she] is doing violates that right.” Id. at 615 (internal quotation marks omitted). The Court concluded that the officers were entitled to qualified immunity because it was “not obvious from the general principles of the Fourth Amendment,” id. at 615-16, or judicial decisions that the presence of the media was unlawful, and “[g]iven [the] undeveloped state of the law, the officers . . . cannot have been expected to predict the future course of constitutional law,” id. at 617 (internal quotation marks omitted). Wilson confirms that whether an officer’s conduct was objectively reasonable is part and parcel of the inquiry into whether the law was clearly established at the time of the challenged conduct and for the particular context in which it occurred. To ask whether an officer’s violation of an individual’s right was objectively reasonable after we have found that the right was clearly established in the particularized sense finds no warrant in Wilson, Saucier, or any other recent Supreme Court discussion of qualified immunity. See also Brosseau v. Haugen, 543 U.S. 194, 199- 200 (2004) (per curiam); Groh v. Ramirez, 540 U.S. 551, 563 (2004); Hope v. Pelzer, 536 U.S. 730, 739-46 (2002). I suspect that our bifurcation of the “clearly established” analysis derives from the eminently reasonable principle that whether a right is clearly established “is not answered by reference to how courts or lawyers might have understood the state of the law.” Maj. Op. at [23]. We do not expect law enforcement officers to keep abreast of every development in the case law or to recognize every implication of legal precedent for police conduct that courts have not previously considered. See Saucier, 533 U.S. at 205 (“It is sometimes difficult for an officer to determine how the relevant legal doctrine . . . will apply to the factual situation the officer confronts.”). But our bifurcated approach 48
    262. makes too much of this principle by divorcing the reasonableness inquiry from the state of the law at the time of the conduct in question. The inquiry described by the Supreme Court already incorporates a recognition that police officers should not be expected to anticipate every application of legal principles because it requires that the right be clearly established with particularity for the conduct at issue. In this case, the particularity requirement means that our “clearly established” inquiry is not complete upon reaching the indisputable conclusion that an individual has the right to be free from arrest, search, or seizure absent probable cause. See Anderson, 483 U.S. at 640-41. Rather, we must determine whether it was clearly established that the situation the officer confronted did not give rise to probable cause. See id. at 641 (holding that officials who “reasonably but mistakenly conclude that probable cause is present . . . should not be held personally liable”); see also Saucier, 533 U.S. at 206 (“[E]ven if a court were to hold that the officer violated the Fourth Amendment by conducting an unreasonable, warrantless search, Anderson still operates to grant officers immunity for reasonable mistakes as to the legality of their actions.”). In other words, based on the law at the time the conduct occurred, would a reasonable officer have known that his or her actions were not supported by probable cause, and therefore were in violation of the Fourth Amendment? The majority opinion takes this question wholly outside of the “clearly established” inquiry and asks whether the officers had “arguable probable cause.” See Maj. Op. at [41]. This Court has used the term “arguable probable cause” to describe the standard for finding that a defendant officer is entitled to qualified immunity for his or her reasonable but mistaken determination that probable cause existed in a particular context. See Caldarola v. Calabrese, 298 F.3d 156, 162 (2d Cir. 2002) 49
    263. (“[I]n the context of a qualified immunity defense to an allegation of false arrest, the defending officer need only show arguable probable cause. This is because at its heart, [t]he concern of the immunity inquiry is to acknowledge that reasonable mistakes can be made as to the legal constraints on particular police conduct.” (internal quotation marks and citations omitted; alteration in original)); Cerrone, 246 F.3d at 203.4 We have also stated that “arguable probable cause” falls under the objective reasonableness determination of our qualified immunity test. See Jenkins v. City of New York, 478 F.3d 76, 87 (2d Cir. 2007). Yet reasonableness—and therefore the existence of “arguable probable cause”—are considerations that properly fall within the clearly established inquiry as the Supreme Court has described it. See Anderson, 483 U.S. at 640-41; Brosseau, 543 U.S. at 199-201. It is not surprising, then, that “arguable probable cause” finds no mention in any Supreme Court opinion; the need for a separate term to describe this concept arises only once we have improperly splintered the “clearly established” inquiry. Because I believe “arguable probable cause” is both imprecise and an outgrowth of the first flaw in our qualified immunity analysis, I do not agree with the majority’s use of the term. I recognize that the distinction I am drawing is a fine one, but I believe it has real consequences. Our approach does not simply divide into two steps what the Supreme Court treats singly, asking first, whether the right is clearly established as a general proposition, and second, whether the application of the general right to the facts of this case is something a reasonable officer could be expected to anticipate. Instead, we permit courts to decide that official conduct was 4 1 Other courts of appeals have also used the term “arguable probable cause” in a 2 similar way as this Court. See, e.g., Williams v. Jaglowski, 269 F.3d 778, 781-82 (7th Cir. 3 2001); Jones v. Cannon, 174 F.3d 1271, 1283 & n.3 (11th Cir. 1999). 50
    264. “reasonable” even after finding that it violated clearly established law in the particularized sense. By introducing reasonableness as a separate step, we give defendants a second bite at the immunity apple, thereby thwarting a careful balance that the Supreme Court has struck “between the interests in vindication of citizens’ constitutional rights and in public officials’ effective performance of their duties.” Anderson, 483 U.S. at 639 (quoting Davis v. Scherer, 468 U.S. 183, 195 (1984)). My second objection to the majority’s formulation of the qualified immunity standard is that it treats objective reasonableness as turning on whether “officers of reasonable competence could disagree.”5 Maj. Op. at [23]. This language, which our cases frequently recite, see, e.g., Iqbal v. Hasty, – F.3d –, 2007 WL 1717803, at *20 (2d Cir. June 14, 2007); Cerrone, 246 F.3d at 202; Lennon v. Miller, 66 F.3d 416, 420 (2d Cir. 1995), derives from the Supreme Court’s 1986 decision in Malley, 475 U.S. at 341. Whether reasonably competent officers could disagree about the lawfulness of the conduct at issue, however, is not the same question the Supreme Court has repeatedly instructed us to consider: whether “it would be clear to a reasonable officer that his [or her] conduct was unlawful in the situation he [or she] confronted.”6 Saucier, 533 U.S. at 202 5 1 This language also appears in the majority opinion as part of the explanation of 2 “arguable probable cause.” Maj. Op. at [41]. Notably, however, by stating that “[a]rguable 3 probable cause exists if either (a) it was objectively reasonable for the officer to believe that 4 probable cause existed, or (b) officers of reasonable competence could disagree on whether 5 the probable cause test was met,” id. (emphasis added; internal quotation marks and citation 6 omitted), this passage seems to support the argument that whether “officers of reasonable 7 competence could disagree” is not an objective reasonableness test. In addition, by adding 8 two steps to the qualified immunity analysis beyond whether the particular right was clearly 9 established, the majority’s discussion of “arguable probable cause” further splinters our 10 qualified immunity test. 6 1 Although Saucier does not specifically refer to the reasonable officer’s competence, 2 I have no quarrel with the assumption that a “reasonable officer” is also a competent officer. 51
    265. (emphasis added); see also Brosseau, 543 U.S. at 199 (quoting Saucier); Groh, 540 U.S. at 563 (same); Hope, 536 U.S. at 746 (same). As with our bifurcation of the “clearly established” inquiry, our requirement of consensus among all reasonable officers departs from Supreme Court dictates and unjustifiably raises the bar to liability for violations of constitutional rights. Reasonable person standards are familiar constructs in the law. They define the level of prudence, care, or knowledge that the law will require of a defendant called to task for his or her actions. In the qualified immunity context, the reasonable officer embodies the minimum degree of judgment and awareness of the law that courts expect law enforcement officials to exercise in the conduct of their duties. That is, the reasonable officer standard sets the threshold beyond which a defendant officer will not be entitled to immunity. As I have discussed, to determine what conduct a reasonable officer should have known to be unlawful in the situation presented, a court must decide whether the law was sufficiently clear regarding the conduct at issue, such that the reasonable officer, and thus the defendant, would have had “fair notice that [his or] her conduct was unlawful.” Brosseau, 543 U.S. at 198. Asking whether “officers of reasonable competence could disagree” shifts this inquiry subtly but significantly. Instead of asking whether the defendant’s conduct was beyond the threshold of permissible error, as the reasonable officer standard does, this inquiry affords a defendant immunity unless a court is confident that a range of hypothetical reasonably competent officers could not disagree as to whether the defendant’s conduct was lawful. This standard is not only more permissive of defendants seeking to justify their conduct; it also takes courts outside their traditional domain, asking them to speculate as to the range of views that reasonable law enforcement officers 52
    266. might hold, rather than engaging in the objective reasonableness determination that courts are well- equipped to make. The Supreme Court has specifically criticized the conflation of an objective reasonableness standard with a requirement of unanimous consensus in the context of a petition for a writ of habeas corpus. In Williams v. Taylor, 529 U.S. 362 (2000), the Court interpreted the statutory provision allowing a federal court to grant review of a petition for a writ of habeas corpus when a state court judgment “involved an unreasonable application of[] clearly established Federal law,” 28 U.S.C. § 2254(d)(1). The Fourth Circuit had previously held that a state court’s adjudication involved an “unreasonable application” of federal law only if “the state court has applied federal law ‘in a manner that reasonable jurists would all agree is unreasonable.’” Williams, 529 U.S. at 409 (quoting Green v. French, 143 F.3d 865, 870 (4th Cir. 1998) (emphasis added)). Both the majority and plurality Supreme Court opinions rejected the Fourth Circuit’s interpretation of the “unreasonable application” standard, explaining that whether an application of the law is objectively unreasonable is a different, less stringent standard than one that asks whether reasonable jurists would unanimously find an application of law unreasonable. See id. at 409-10 (majority opinion) (“[A] federal habeas court making the ‘unreasonable application’ inquiry should ask whether the state court’s application of clearly established federal law was objectively unreasonable. The federal habeas court should not transform the inquiry into a subjective one by resting its determination instead on the simple fact that at least one of the Nation’s jurists has applied the relevant federal law in the same manner the state court did in the habeas petitioner’s case.”); id. at 378 (plurality opinion) (“As Congress is acutely aware, reasonable lawyers and lawgivers regularly disagree with one 53
    267. another. Congress surely did not intend that the views of one such judge who might think that relief is not warranted in a particular case should always have greater weight than the contrary, considered judgment of several other reasonable judges.”).7 Our Court similarly has adopted an unjustifiably stringent standard in the qualified immunity context by prohibiting liability for constitutional violations where a court believes that one reasonably competent officer would find the conduct at issue lawful, even if the overwhelming majority would not. Finally, I note that although we repeat Malley’s “officers of reasonable competence” test with regularity, and it appears frequently in the decisions of other federal courts of appeals, see, e.g., Brittain v. Hansen, 451 F.3d 982, 988 (9th Cir. 2006); Armstrong v. City of Melvindale, 432 F.3d 695, 701 (6th Cir. 2006); Wollin v. Gondert, 192 F.3d 616, 625 (7th Cir. 1999), it has not appeared a second time in any majority opinion of the Supreme Court. It seems curious that we would continue to rest our qualified immunity standard on language the Supreme Court has carefully eschewed for over twenty years since Malley was decided. In sum, the Supreme Court has struck a careful balance between the vindication of constitutional rights and government officials’ ability to exercise discretion in the performance of their duties. Our case law, in subtle but important ways, has altered this balance in favor of defendants by adding another analytic step to the qualified immunity analysis and equating objective reasonableness with unanimity among “officers of reasonable competence.” In the vast majority of 7 1 This caution against transforming a reasonableness inquiry into a consensus 2 requirement is instructive notwithstanding the plurality’s statement that the particular statute 3 at issue was not meant to codify the standard for qualified immunity into the law of habeas 4 review. See Williams, 529 U.S. at 380 n.12 (plurality opinion). 54
    268. cases, including this one, the particular phrasing of the standard will not alter the outcome of the qualified immunity analysis. There is no doubt in this case that a reasonable officer would believe that the arrest of Thomas Walcyzk, as well as the search of his home and the seizure of firearms found there, were lawful. Yet the effect in future cases may not always be so benign. What is more, the majority’s framework introduces unnecessary complications into an already complicated qualified immunity analysis. It is time to eliminate these complications and reconcile our qualified immunity analysis with the Supreme Court’s most recent, authoritative jurisprudence. 55
    269. 02-4611-ag, 02-4629-ag, 03-40837-ag Lin et. al. v. U.S. Dept. of Justice 1 UNITED STATES COURT OF APPEALS 2 FOR THE SECOND CIRCUIT 3 August Term, 2006 4 (Argued: March 7, 2007 Decided: July 16, 2007) 5 6 Docket Nos. 02-4611-ag, 02-4629-ag, 03-40837-ag 7 _____________________ 8 SHI LIANG LIN , 9 10 Petitioner, 11 12 v. 13 14 UNITED STATES DEPARTMENT OF JUSTICE ; ATTORNEY GENERAL GONZALES, 15 16 Respondents; 17 _____________________ 18 19 ZHEN HUA DONG , 20 21 Petitioner, 22 23 v. 24 25 UNITED STATES DEPARTMENT OF JUSTICE ; ATTORNEY GENERAL GONZALES 26 27 Respondents; 28 _____________________ 29 30 XIAN ZOU , 31 32 Petitioner, 33 34 v. 35 36 ATTORNEY GENERAL GONZALES, 1
    270. 1 Respondent. 2 _____________________ 3 Before: JACOBS , Chief Judge, CALABRESI, CABRANES, STRAUB , POOLER , SACK , SOTOMAYOR , 4 KATZMANN , PARKER , RAGGI, WESLEY , and HALL , Circuit Judges. 5 Petitions for review of orders of the Board of Immigration Appeals denying applications 6 for asylum, withholding of removal, and relief under the Convention Against Torture. The 7 petition for review of Zhen Hua Dong is DENIED. Xian Zou’s and Shi Liang Lin’s petitions are 8 DISMISSED for lack of jurisdiction. 9 10 Judge PARKER delivered the opinion of the Court, in which JACOBS , C.J., and CABRANES, 11 SACK , RAGGI, WESLEY , and HALL , JJ., joined. 12 13 Judge KATZMANN filed a concurring opinion, in which STRAUB , POOLER , and 14 SOTOMAYOR , JJ., joined. 15 16 Judge SOTOMAYOR filed a concurring opinion, in which POOLER , J., joined. 17 18 Judge CALABRESI filed an opinion concurring in part and dissenting in part. 19 20 21 BRUNO JOSEPH BEMBI, Hempstead, NY, for Petitioners Shi 22 Liang Lin and Zhen Hua Dong. 23 24 ALEKSANDER MILCH , Christophe & Associates, P.C., New 25 York, NY, for Petitioner Xian Zou. 26 27 KATHY S. MARKS, Assistant United States Attorney, (Sara 28 L. Shudofsky, Assistant United States Attorney, of counsel) 29 for Michael J. Garcia, United States Attorney for the 30 Southern District of New York, New York, NY, for 31 Respondents the United States Department of Justice and 32 Attorney General Gonzales. 33 34 BARRINGTON D. PARKER , Circuit Judge: 35 36 In 1997 the Board of Immigration Appeals (“BIA”) held in Matter of C-Y-Z-, 21 I. & N. 37 Dec. 915 (B.I.A. 1997) (en banc) that an individual whose spouse has been forced to abort a 38 pregnancy, undergone involuntary sterilization, or been persecuted under a coercive population 2
    271. 1 control program could automatically qualify for asylum as a “refugee” under § 601(a) of the 2 Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (“IIRIRA”) (amending 8 3 U.S.C. § 1101(a)(42), Immigration and Nationality Act (“INA”) § 101(a)(42)). See In re S-L-L-, 4 24 I. & N. Dec. 1, 3 (B.I.A. 2006) (en banc) (“In Matter of C-Y-Z-, . . . we held that a husband 5 whose wife was forcibly sterilized could establish past persecution under this amendment to 6 section 101(a)(42) of the [INA].”). This appeal considers whether the BIA’s interpretation of the 7 statute was correct. We conclude it was not. 8 Petitioners Shi Liang Lin, Zhen Hua Dong, and Xian Zou are citizens of the People’s 9 Republic of China and unmarried partners of individuals allegedly victimized by China’s 10 coercive family planning policies. Each seeks review of an order of the BIA summarily 11 affirming the denial of an application for asylum based, in part, on the BIA’s holding in C-Y-Z.1 12 We remanded these petitions to the BIA to afford it the opportunity to explain its rationale in C- 13 Y-Z- for reading § 601(a) to say that the spouses of those directly victimized by coercive family 14 planning policies are per se eligible for asylum as if they were directly victimized themselves 15 and also to clarify the status of boyfriends and fiancés under that statute. See Lin v. U.S. Dep’t of 16 Justice, 416 F.3d 184, 187 (2d Cir. 2005). We retained jurisdiction. Id. 17 On remand, the BIA reaffirmed its holding in C-Y-Z- that spouses are entitled to 18 automatic eligibility under § 601(a) but limited this per se eligibility to legally married 19 applicants. S-L-L-, 24 I. & N. Dec. 1. Eschewing a text-based analysis, the BIA elected to 1 See In re Shi Liang Lin, No. A70 895 638 (B.I.A. Sept. 29, 2002), aff’g No. A70 895 638 (Immig. Ct. N.Y. City May 9, 2000); In re Zhen Hua Dong, No. A 7 293 661 (B.I.A. Sept. 25, 2002), aff’g No. A77 293 661 (Immig. Ct. N.Y. City Oct. 12, 2000); In re Zou, No. A77 322 595 (B.I.A. Aug. 27, 2002), aff’g No. A77 322 595 (Immig. Ct. N.Y. City Apr. 4, 2002). 3
    272. 1 interpret the forced abortion and sterilization clause of the section “in light of the overall purpose 2 of the amendment” to include both parties to a marriage. Id. at 8. The Board reaffirmed the 3 dismissal of the appeals of petitioners Lin, id., and Dong, In re Zhen Hua Dong, No. A77 293 4 661 (B.I.A. Nov. 27, 2006), and remanded Zou’s petition for a determination of whether he 5 qualified for asylum based on the “other resistance to a coercive population control program” 6 clause in § 601(a), In re Xian Zou, No. A73 178 541 (B.I.A. Nov. 21, 2006). 7 Following the BIA’s decision, we ordered rehearing en banc to consider two issues: 8 First, whether § 601(a)’s provisions are ambiguous, so that the BIA’s construction of them 9 warrants Chevron deference; and second, whether the BIA reasonably construed § 601(a) to 10 extend automatic asylum eligibility to a petitioner whose legally married spouse was subjected to 11 an involuntary abortion or sterilization but not to a domestic partner or fiancé whose claim is 12 derivative unless the petitioner engaged in “other resistance” to a coercive population control 13 policy. Lin v. U.S. Dep’t of Justice, Nos. 02-4611, 02-4629, 03-40837 (2d Cir. Nov. 13, 2006) 14 (order) (“En banc order”). See S-L-L-, 24 I. & N. Dec. 1; Chevron U.S.A. Inc. v. Natural Res. 15 Def. Council, Inc., 467 U.S. 837 (1984). 16 We now conclude that the BIA erred in its interpretation of 8 U.S.C. § 1101(a)(42) by 17 failing to acknowledge language in § 601(a), viewed in the context of the statutory scheme 18 governing entitlement to asylum, that is unambiguous and that does not extend automatic refugee 19 status to spouses or unmarried partners of individuals § 601(a) expressly protects. Accordingly, 20 the petition of Zhen Hua Dong is denied. The petition of Shi Liang Lin is dismissed as moot.2 2 Subsequent to oral arguments, we learned that Lin has had no contact with his attorneys since “early 2004,” and that his attorney believes that he has either returned to China or is deceased. (Aff. of Yee Ling Poon, ¶¶ 3-5). Accordingly, Lin’s case is moot. See 8 C.F.R. § 4
    273. 1 The petition of Xian Zou is dismissed for lack of jurisdiction.3 We recognize that this decision 2 creates a split among the circuits.4 3 I. BACKGROUND 4 Congress has given the Attorney General the discretionary authority to grant asylum to 5 an alien who qualifies as a “refugee” because he or she “is unable or unwilling to avail himself 1208.8. We no longer have jurisdiction over Zou’s petition because the BIA has remanded the 3 case to the immigration court for further findings. See In re Xian Zou, No. A77 322 295 (B.I.A. Nov. 21, 2006). 4 A number of our sister circuits have deferred to the BIA’s interpretation of § 601(a). See, e.g., Zhang v. Gonzales, 434 F.3d 993, 1001 (7th Cir. 2006); Huang v. Ashcroft, 113 Fed. App’x 695, 700 (6th Cir. 2004) (unpublished opinion); He v. Ashcroft, 328 F.3d 593, 604 (9th Cir. 2003); Li v. Ashcroft, 82 Fed. App’x 357, 358 (5th Cir. 2003) (unpublished per curiam opinion). While the Third Circuit had questioned the BIA’s reading of the plain language of the amendment, stating that “[i]t takes some effort to reconcile [the BIA’s] interpretation with the language of the 1996 amendment, since the phrase ‘a person who has been forced to abort a pregnancy or to undergo involuntary sterilization’ is most naturally read as referring only to a person who has personally undergone one of those procedures,” Chen v. Ashcroft, 381 F.3d 221, 226 (3d Cir. 2004) (Alito, J.), a divided panel of the Third Circuit recently validated the BIA’s interpretation of § 601(a) over a vigorous dissent. See Sun Wen Chen v. U.S. Att’y Gen., – F.3d –, 2007 WL 1760658, at *4-*6 (3d Cir. June 20, 2007). The circuits are already split over whether § 601(a) provides protection for individuals who marry in traditional ceremonies not recognized by their government and later seek asylum based on the forced abortion or sterilization of their “common law spouses.” The Seventh and Ninth Circuits have held that the statute covers spouses from traditional marriage ceremonies, see Zhang, 434 F.3d at 999; Zhu v. Gonzales, 465 F.3d 316, 321 (7th Cir. 2006); Ma v. Ashcroft, 361 F.3d 553, 559-61 (9th Cir. 2004). In contrast, the Third Circuit, in Chen, supra, held that the amendment does not cover unmarried partners, even when they have been prevented from marrying by their government’s family planning policy. 381 F.3d at 232-34; see also Chen v. Gonzales, 418 F.3d 110, 111 (1st Cir. 2005) (acknowledging circuit split on the issue). While they have not reached the issue of traditional marriage ceremonies, the Fifth and Eleventh Circuits have declined to extend IIRIRA § 601 to cover boyfriends of individuals who have been subjected to a forced abortion or sterilization. See Zhang v. Ashcroft, 395 F.3d 531, 532 (5th Cir. 2004); Wang v. U.S. Att’y Gen., 152 Fed. App’x 761, 767 (11th Cir. 2005) (unpublished opinion). 5
    274. 1 or herself of the protection of [his or her native country] because of persecution or a well- 2 founded fear of persecution on account of race, religion, nationality, membership in a particular 3 social group, or political opinion.” 8 U.S.C. § 1101(a)(42). A showing of past persecution gives 4 rise to a rebuttable presumption of a well-founded fear of future persecution. 8 C.F.R. § 5 1208.13(b)(1). 6 In 1996, Congress passed IIRIRA § 601(a), which amended 8 U.S.C. § 1101(a)(42) by 7 broadening its definition of “refugee,” adding the following language: 8 [A] person who has been forced to abort a pregnancy or to undergo involuntary 9 sterilization, or who has been persecuted for failure or refusal to undergo such a 10 procedure or for other resistance to a coercive population control program, shall 11 be deemed to have been persecuted on account of political opinion, and a person 12 who has a well founded fear that he or she will be forced to undergo such a 13 procedure or subject to persecution for such failure, refusal, or resistance shall be 14 deemed to have a well founded fear of persecution on account of political 15 opinion. 16 17 8 U.S.C. § 1101(a)(42). 18 The next year, the BIA held that “past persecution of one spouse can be established by 19 coerced abortion or sterilization of the other spouse,” so that spouses of individuals directly 20 victimized by coercive family planning policies are per se eligible for asylum pursuant to § 21 1101(a)(42). See Matter of C-Y-Z-, 21 I. & N. Dec 915, 917-18 (B.I.A. 1997) (en banc). The 22 BIA gave no reasons for reading the statute to compel this result. 23 Petitioner Lin entered the United States in January 1991 and filed an application for 24 asylum and withholding of removal in June 1993. According to Lin’s application, he had sought 25 the required governmental permission to marry his girlfriend and have children with her, but she 26 was too young under Chinese law. After his girlfriend became pregnant and was forced to have 6
    275. 1 an abortion, Lin left China. His girlfriend remained in China because she was too weak to travel. 2 Following a hearing, the IJ found Lin credible, but concluded that he did not qualify for asylum 3 based on his girlfriend’s forced abortion and denied the petition. The BIA affirmed without 4 opinion. See In re Shi Liang Lin, No. A70 895 638 (B.I.A. Sept. 29, 2002), aff’g No. A70 895 5 638 (Immig. Ct. N.Y. City May 9, 2000). 6 Petitioner Dong attempted to enter the United States in October 1999, and was detained 7 by INS officials. When the INS commenced removal proceedings, Dong requested asylum, 8 withholding of removal, and relief under the Convention Against Torture (“CAT”). His asylum 9 petition was based on a claim that his fiancée (who continued to reside in China) had been forced 10 to undergo two abortions and that he would be jailed and fined for having left China illegally 11 were he to be deported. The IJ denied Dong’s petition, finding that, although he was credible, he 12 did not qualify for refugee status as a fiancé of a woman who had undergone forced abortions, 13 and had not established other grounds for relief. The BIA affirmed the IJ’s decision. See In re 14 Zhen Hua Dong, No. A77 293 661 (B.I.A. Sept. 25, 2002), aff’g No. A77 293 661 (Immig. Ct. 15 N.Y. City Oct. 12, 2000). 16 Petitioner Zou was taken into custody by the INS when he attempted to enter the United 17 States in September 2000. After removal proceedings commenced, Zou filed for asylum. Under 18 Chinese law, Zou and his girlfriend had been too young to marry. He claimed that she had been 19 forced to have an abortion and he had been threatened with arrest after protesting to family 20 planning officials. An IJ denied the application. The IJ found Zou’s testimony concerning his 21 threatened arrest incredible, doubting that Zou would be able to evade the Chinese authorities 22 and travel to the United States with an outstanding warrant of arrest from the Chinese 7
    276. 1 government. Left only with Zou’s claim that his girlfriend had undergone a forced abortion, the 2 IJ held that Zou did not qualify for relief under § 1101(a)(42) on that basis. Zou appealed to the 3 BIA, which affirmed the decision of the IJ. See In re Zou, No. A77 322 595 (B.I.A. Aug. 27, 4 2002), aff’g No. A77 322 595 (Immig. Ct. N.Y. City Apr. 4, 2002). The three petitioners sought 5 review here and the cases were heard in tandem. 6 Three different interpretations of the application of “refugee” status to the facts in these 7 cases have been proposed by the parties throughout the litigation. When we heard the petitions 8 in tandem, the Department of Justice argued that § 601(a) of IIRIRA should be understood to 9 confer per se refugee status on spouses of individuals who had undergone forced abortions or 10 involuntary sterilizations, but not to boyfriends or fiancés, who were covered, if at all, under the 11 “other resistance” language of the amendment. Second, the petitioners argued, both before this 12 Court and before the BIA, that the distinction between spouses and unmarried partners was 13 arbitrary and that the statute should be interpreted to extend per se refugee status to the 14 committed partners of individuals who have been forced to undergo an abortion or involuntarily 15 sterilization. 16 The third, a text-based interpretation, was put forth before the BIA by the Department of 17 Homeland Security (“DHS”), which is “charged with the administration and enforcement of . . . 18 laws relating to the immigration and naturalization of aliens,” 8 U.S.C § 1103(a)(1). In its brief 19 to the BIA, the DHS insisted that no support existed in the statute for the BIA’s automatic 20 extension of § 601(a) to spouses. Specifically, the DHS asserted that the BIA’s interpretation 21 was “at odds with the most natural reading of the statute and with established principles of 22 asylum law.” Brief of DHS on Remand at 5, In re S-L-L-, 24 I. & N. Dec. 1 (B.I.A. 2006) (en 8
    277. 1 banc) (No. A 70 895 638) (“DHS brief”). Carefully considering the text of § 601(a), the DHS 2 concluded that a rule conferring per se refugee status on spouses of individuals persecuted under 3 coercive family planning policies contradicted the unambiguous language of the amendment. 4 Further, the DHS stated that such a rule was at odds with the legislative history of the 5 amendment, as well as “a fundamental precept of U.S. refugee law . . . that, to qualify for 6 protection, an applicant must demonstrate that he will be targeted for his own protected belief or 7 characteristic.” DHS brief at 8-9, 11. The DHS understood that the statute required a “case-by- 8 case” application to individuals who have not themselves been victims of a forced abortion or 9 involuntary sterilization. Under this approach, spouses, boyfriends, and fiancés would have to 10 demonstrate their qualification for refugee status under the “for other resistance to a coercive 11 population control program” provision of § 601(a). 12 On remand, in S-L-L-, 24 I. & N. Dec 1, the BIA affirmed its earlier decision in C-Y-Z 13 that under § 601(a) an applicant whose spouse was forced to undergo an abortion or sterilization 14 procedure is automatically entitled to asylum, while limiting its interpretation to individuals who 15 were opposed to their legal spouse’s abortion or sterilization. Id. at 4. The BIA reasoned that § 16 601(a) provided “no clear or obvious answer to the scope of the protections . . . to partners of 17 persons forced to submit to an abortion or sterilization.” Id. As a result of this perceived 18 ambiguity, the majority chose to ground its interpretation “in the context of the history and 19 purposes of the Act as a whole,” finding that 20 absent evidence that the spouse did not oppose an abortion or sterilization 21 procedure, we interpret the forced abortion and sterilization clause of section 22 101(a)(42) of the Act, in light of the overall purpose of the amendment, to include 23 both parties to a marriage. 24 9
    278. 1 Id. at 8. While the decision rested on “the forced abortion and sterilization clause of section 2 101(a)(42),” the Board noted that it was applying “general principles regarding nexus and level 3 of harm.” Id. at 5. 4 The majority based its conclusion on a number of policy-based factors, including the 5 positions taken by the litigants and the fact that C-Y-Z-, as a ten-year-old decision, constituted 6 long-standing precedent. The majority was also influenced by the shared responsibility of 7 married couples regarding family planning decisions under Chinese law and the “profound 8 impact” that a forced abortion or sterilization has on both parties to a marriage. Id. at 6-8. 9 The majority stopped short of extending a per se presumption to boyfriends, fiancés, and 10 other unmarried partners. It concluded that they were not comparable to husbands because “the 11 sanctity of marriage and the long term commitment reflected by marriage place the husband in a 12 distinctly different position from that of an unmarried father.” Id. at 9. The BIA dismissed the 13 appeals of Lin, id., and Dong, In re Zhen Hua Dong, No. A77 293 661 (B.I.A. Nov. 27, 2006), 14 and remanded Zou’s case to the immigration court “for further evidence on the issues of 15 resistance and harm” based on the “other resistance” clause of § 1101(a)(42), In re Xian Zou, 16 No. A73 178 541 (B.I.A. Nov. 21, 2006). 17 Board Member Pauley concurred. While deferring to the BIA’s prior decision in C-Y-Z 18 as long-standing and widely accepted precedent, he conceded that: 19 [w]ere we writing on a clean slate, I would adopt the lately arrived at position of 20 the Department of Homeland Security . . . that whether or not the spouse of a 21 forcibly sterilized or aborted individual can be found to have been persecuted 22 depends on a case-by-case assessment of whether that spouse was persecuted on 23 account of “other resistance” to a coercive population control system, because the 24 language of the Act does not support extending refugee status to any person other 25 than the one sterilized or aborted, aside from the “other resistance” ground. 10
    279. 1 Id. at 13 (Pauley, B.M., concurring). 2 Board Member Filppu, joined by Board Member Cole, concurred in the result but 3 dissented from the majority’s reaffirmation of C-Y-Z. They reasoned that the language of the 4 statute was unambiguous in “focus[ing] on ‘a person’ who has been forced to abort a pregnancy, 5 not on a ‘couple,’ let alone a married couple . . . .” Id. at 16 (Filppu, B.M., concurring and 6 dissenting). Understanding that “statutory interpretation must begin with reference to the 7 language and structure of the statute,” id. at 15, and “Congress expresses its intent through the 8 language it chooses,” id. at 19 (quoting INS v. Cardoza-Fonseca, 480 U.S. 421, 432 n.12 9 (1987)), the dissenters concluded that an individual who has not been subjected to a forcible 10 abortion or sterilization procedure can qualify for refugee status in relation to a coercive family 11 planning policy only if he proves that he was persecuted or has a well-founded fear of future 12 persecution “for other resistance to a coercive population control program.” Id.5 13 Following the BIA’s decision, we ordered an en banc rehearing to determine whether § 14 601(a) is ambiguous, whether the BIA’s interpretation of “refugee” should be accorded Chevron 15 deference, and “[w]hether the BIA reasonably construed IIRIRA § 601’s definition of “refugee” 16 to: (a) include a petitioner whose legally married spouse was subjected to an involuntary 17 abortion or sterilization . . . ; and (b) not include a petitioner whose claim is derivatively based 18 on any other relationship with a person who was subjected to such a procedure, unless the 19 petitioner has engaged in ‘other resistance’ to a coercive population control program . . . .” See 5 While here, as throughout the opinion, we refer to a male petitioner with a wife or girlfriend who has been forced to undergo an abortion or sterilization, our reasoning applies with equal force to the perhaps more uncommon situation in which a female petitioner’s male spouse or boyfriend has been forced to undergo sterilization. 11
    280. 1 En banc order. We now conclude that the BIA’s interpretation of the statute is not correct. 2 II. DISCUSSION 3 When reviewing the BIA’s interpretation of statutes that it administers, we apply the 4 Chevron principles. We first ask if Congress has spoken. “If the intent of Congress is clear, that 5 is the end of the matter; for the court, as well as the agency, must give effect to the 6 unambiguously expressed intent of Congress.” Chevron, 467 U.S. at 842-43; see also INS v. 7 Cardoza-Fonseca, 480 U.S. 421, 447-48 (1987). Only if the statute is silent or ambiguous do we 8 turn to the second inquiry – whether the BIA’s interpretation constitutes “a permissible 9 construction of the statute.” Chevron, 467 U.S. at 843. 10 While the petitioners in this case are unmarried partners, and not spouses, of individuals 11 who have been subjected to forced abortions, we review the BIA’s interpretation of § 601, as 12 articulated in its decision below, to extend a per se presumption of persecution to spouses, but 13 not to non-married partners, of individuals who have been involuntarily subjected to an abortion 14 or sterilization.6 It is the existence of this spousal policy that the petitioners argue is an arbitrary 15 and capricious interpretation of the statute. If the BIA’s policy is at odds with the plain language 16 of the statute, it makes little sense to consider only whether it can reasonably be limited to 17 couples who are formally married. 6 We announced our intention to reach this question in our order that this case be reheard en banc. In that order we instructed the parties to address the BIA’s interpretation of § 601 as it related to both spouses and non-married partners of individuals subjected to an involuntary abortion or sterilization. All members of this en banc panel joined the order, including those who now express confusion as to why we reach the question. We are particularly perplexed by the position taken by our colleagues Judge Katzmann and Judge Sotomayor, who contend we are overreaching by considering whether the BIA’s per se rule survives review under Chevron step one, but who then proceed to assess, and declare valid, the same rule under Chevron step two. Op. of Judge Katzmann at 16, Op. of Judge Sotomayor at 1 & n.1. 12
    281. 1 Accordingly, we start our Chevron analysis of § 601(a) by asking whether Congress has 2 spoken directly to the question of whether an individual can establish past persecution based 3 solely on his spouse or partner’s forced abortion or sterilization. 467 U.S. at 842-43. In S-L-L-, 4 the BIA, without elaboration or explanation, observed that “[t]here is no clear or obvious answer 5 to the scope of the protections afforded by the amendment to partners of persons forced to 6 submit to an abortion or sterilization.” 24 I. & N. Dec. at 4. We disagree. We conclude that 7 Congress has spoken to this issue and that it has done so unambiguously. 8 In the past, this Court, when following the BIA’s holding in C-Y-Z-, has deferred to the 9 BIA’s interpretation without performing a threshold Chevron analysis of the ambiguity of the 10 statute. In Yuan v. U.S. Dep’t of Justice, for example, we stated of IIRIRA § 601 that, “[b]y its 11 plain language, the law would seem to extend refugee status only to actual victims of persecution 12 – for example, a woman who was ‘forced to abort a pregnancy,’ but not her husband.” 416 F.3d 13 192, 196 (2d Cir. 2005). Despite this recognition of the amendment’s plain language, we went 14 on to defer to the interpretation of the BIA, stating that, “we have held that spouses of people 15 actually subject to persecution under coercive family planning policies are per se eligible for 16 asylum. . . . In doing so, however, we did not explain why. Rather, we followed the lead of the 17 BIA.” Id. at 196-97 (internal citation omitted); see also Zhang v. I.N.S., 386 F.3d 66, 73 (2d Cir. 18 2004); Qiu v. Ashcroft, 329 F.3d 140, 149 (2d Cir. 2003). To the extent that deference implicit 19 in these cases can be read to say that deference is due, they are overruled. 20 The amendment provides: 21 [(1)] a person who has been forced to abort a pregnancy or to undergo involuntary 22 sterilization, or [(2)] who has been persecuted for failure or refusal to undergo 23 such a procedure or [(3)] for other resistance to a coercive population control 13
    282. 1 program, shall be deemed to have been persecuted on account of political opinion, 2 and [(4)] a person who has a well founded fear that he or she will be forced to 3 undergo such a procedure or [(5)] subject to persecution for such failure, refusal, 4 or [(6)] resistance shall be deemed to have a well founded fear of persecution on 5 account of political opinion. 6 7 8 U.S.C. § 1101(a)(42). The language of the first clause, which refers to “a person who has been 8 forced to abort a pregnancy or to undergo involuntary sterilization,” could not be more clear in 9 its reference to “a person,” rather than “a couple,” who has been subjected to a forced abortion or 10 involuntary sterilization. This interpretation follows two cardinal rules: first that “[s]tatutory 11 construction must begin with the language employed by Congress and the assumption that the 12 ordinary meaning of that language accurately expresses the legislative purpose,” Park ‘N Fly, 13 Inc. v. Dollar Park & Fly, Inc., 469 U.S. 189, 194 (1985); and, second, that “we begin with the 14 understanding that Congress says in a statute what it means and means in a statute what it says 15 there,” Hartford Underwriters Ins. Co. v. Union Planters Bank, N.A., 530 U.S. 1, 6 (2000) 16 (internal quotation marks omitted). 17 As the statute is written, “a person who has been forced to abort a pregnancy” 18 unambiguously refers to a woman who has been physically subjected to a forced abortion, and “a 19 person who has been forced . . . to undergo involuntary sterilization” means an individual who 20 has physically undergone an involuntary medical procedure intended to result in infertility. Had 21 Congress intended this clause to refer to a spouse or partner of someone who has been physically 22 subjected to a forced procedure, “it could simply have said so.” Id. at 7. 23 Similarly, the second clause of the amendment refers to “a person” who “has been 24 persecuted for failure or refusal to undergo [an abortion or involuntary sterilization.]” 8 U.S.C. § 14
    283. 1 1101(a)(42). Like the preceding clause, this language refers to individuals who have failed or 2 refused to undergo (i.e., “submit to”) a procedure affecting their own bodies. Under the 3 language used by Congress, having someone else, such as one’s spouse, undergo a forced 4 procedure does not suffice to qualify an individual for refugee status. 5 A parallel analysis governs the categories of § 601(a) relating to the establishment of a 6 well-founded fear of future persecution. The fourth category covers “a person who has a well 7 founded fear that he or she will be forced to undergo [an abortion or involuntary sterilization].” 8 This category plainly excludes an individual who does not fear that she herself will be subjected 9 to a forced abortion or sterilization. Not only does it refer to “a person,” and indicate that that 10 person will “undergo” the procedure his- or herself, the use of the pronouns “he” and “she” 11 reinforces the intention of Congress to limit the application of the clause to individuals who are 12 themselves physically forced to undergo an abortion or sterilization. 13 Finally, the fifth category of individuals refers to “a person who has a well founded fear 14 that he or she will be . . . subject to persecution for such failure [or] refusal [to undergo an 15 abortion or involuntary sterilization].” Like the language of the previous clauses, this phrase is 16 unambiguous in its reference to “a person” who fears that “he or she” will be persecuted for his 17 or her reaction to a threat of “undergo[ing]” an abortion or sterilization to his or her own body. 18 This section, like the previous three categories discussed, cannot be read reasonably to cover an 19 individual’s fears arising from a coercive procedure performed on someone else.7 7 No member of the en banc Court disputes this reading of IIRIRA § 601(a). Instead, our colleagues assert that other language in 8 U.S.C. § 1101(a)(42) – language predating adoption of the amendment – might (in the case of our colleague, Judge Calabresi) or does (in the case of our colleagues Judge Katzmann and Judge Sotomayor) somehow support the BIA’s per se rule. 15
    284. 1 In S-L-L-, the BIA noted that it was applying “general principles regarding nexus and 2 level of harm,” 24 I. & N. Dec. at 5 – principles derived from the general refugee definition 3 subsection of § 1101(a)(42) – to reach its holding. We believe that this approach was not 4 correct. The general definition of a “refugee” under § 1101(a)(42) permits “any person” who 5 experiences “persecution or a well-founded fear of persecution on account of race, religion, 6 nationality, membership in a particular social group, or political opinion” to obtain asylum. 8 7 U.S.C. § 1101(a)(42). Nothing in the general definition of refugee would permit “any person” 8 who has not personally experienced persecution or a well-founded fear of future persecution on a 9 protected ground to obtain asylum, as the BIA’s per se rule would permit. Indeed, an 10 examination of the overall statutory scheme reassures us that, pursuant to Chevron, we must 11 conclude that Congress has clearly and unambiguously spoken to the issue at hand. See FDA v. 12 Brown & Williamson Tobacco Corp., 529 U.S. 120, 132-33 (2000) (“In determining whether 13 Congress has specifically addressed the question at issue [under Chevron step one], a reviewing 14 court should not confine itself to examining a particular statutory provision in isolation. . . . A These assertions are incorrect, we believe, for reasons discussed infra. By contrast, the Third Circuit’s recent decision in Sun Wen Chen, which our concurring colleagues Judge Katzmann and Judge Sotomayor cite extensively even though they apparently disagree with its analysis, asserts that § 601(a) contains an ambiguity that the BIA is empowered to fill. See Sun Wen Chen, 2007 WL 1760658, at *4 (“The C-Y-Z- rule . . . fleshes out an issue germane to the application of [§ 601(a)] that was not addressed by Congress, and so poses no Chevron step one problem.”). But see Sun Wen Chen, 2007 WL 1760658, at *10 (McKee, J., dissenting) (“Rather than accept the language [of § 601(a)] as drafted, the majority concludes that the absence of ‘spouse’ in the statute creates a vacuum that the Attorney General may rush in and fill, even though this results in amending the statute.”); id. at *11 (McKee, J., dissenting) (“Our analysis should therefore begin and end with the language of [§ 601(a)]. There is no room here for a step two inquiry under Chevron. . . . I believe Congress meant what it said, and I do not assume that the omission of any reference to a ‘spouse’ is accidental or insignificant.”). 16
    285. 1 court must . . . interpret the statute as a symmetrical and coherent regulatory scheme.” (citation 2 and internal quotation marks omitted)). 3 It is apparent to us that when Congress rejected the BIA’s view in Matter of Chang, 20 I. 4 & N. Dec. 38 (B.I.A. 1989), of birth control policies in other nations as an avenue for asylum, it 5 did so in clear and unmistakable language. It identified those to whom asylum could be granted 6 and reaffirmed the need for direct personal persecution. Congress’s specific designation of some 7 persons (i.e., those who fear, resist, or undergo particular medical procedures) is incompatible 8 with the view that others (e.g., their spouses) should also be granted asylum per se because of 9 birth control policies. The inclusion of some obviously results in the exclusion of others. See 10 TRW Inc. v. Andrews, 534 U.S. 19, 28 (2001); Leatherman v. Tarrant County Narcotics 11 Intelligence and Coordination Unit, 507 U.S. 163, 168 (1993).8 12 The language of § 601(a) does nothing to alter the pre-IIRIRA definition of “political 13 opinion” in § 1101(a)(42), and this further demonstrates the exclusivity of the group of persons 14 entitled to asylum per se under § 601(a). Congress could have announced that the term “political 15 opinion” included any reproductive act in violation of a coercive population control program, but 16 instead it chose to create a specific exception to the general statutory requirement that a person 17 claiming refugee status based on past persecution has the burden of demonstrating that the 18 particular conduct experienced by him rose to the level of persecution and the persecution had a 8 Judge Katzmann contends that we have employed this well-known canon of statutory construction to conclude that IIRIRA § 601(a) restricts, rather than expands, the availability of asylum relief. Op. of Judge Katzmann at 5. This contention is misplaced. We acknowledge that IIRIRA § 601(a) expands the availability of asylum, but find it significant that it does so only to specifically-enumerated categories of asylum applications – categories that do not include the beneficiaries of the BIA’s per se rule. 17
    286. 1 specified impermissible nexus. See 8 U.S.C. § 1158(b)(1)(B)(i) (stating that “[t]he burden of 2 proof is on the applicant to establish that the applicant is a refugee”). IIRIRA § 601(a) states that 3 certain individuals affected by coercive population control programs “shall be deemed” 4 persecuted by reason of political opinion. In using the word “deem” in this context,9 § 601(a) 5 makes clear that those who benefit from the amendment would not be entitled to per se political 6 opinion asylum relief absent the amendment. In other words, their political opinion exists de 7 jure rather than as a matter of fact on which the applicant bears the burden of proof. For an 8 asylum applicant who does not fall within this limited exception, the burden remains on the 9 applicant – and the opportunity remains open – to demonstrate, in light of the particular facts of 10 the case, that he has (i) a well-founded fear of personal persecution (ii) based on political opinion 11 or some other impermissible ground. 12 This is consistent with what we know: While it is plain that suffering a forced medical 13 procedure can be a persecution if it is on account of a protected ground, the conception of a child 14 is no more an expression of political opinion than birth, death, sleep, or the taking of 15 nourishment. If the language of § 601(a) indicates that the woman who is subjugated to the 16 outrage of a forced abortion has not herself been persecuted for the “political opinion” of 17 conceiving a child under pre-IIRIRA § 1101(a)(42), then so much less the man who has 9 To “deem” is “[t]o treat (something) as if (1) it were really something else, or (2) it has qualities that it doesn’t have,” or, in the alternative, “[t]o consider, think, or judge.” Black’s Law Dictionary 446 (8th ed. 2004). Black’s Law Dictionary further notes that the word deem “has been traditionally considered to be a useful word when it is necessary to establish a legal fiction either positively by 'deeming' something to be what it is not or negatively by 'deeming' something not to be what it is. . . . All other uses of the word should be avoided.” Id. (quoting G.C. Thornton, Legislative Drafting 99 (4th ed. 1996)). 18
    287. 1 impregnated her; but unlike his wife or partner, he is not “deemed” under § 601(a) to hold a 2 political opinion and he must prove the existence of a political opinion or other protected ground 3 under § 1158(b)(1)(B)(i). Accordingly, we conclude that the statutory scheme unambiguously 4 dictates that applicants can become candidates for asylum relief only based on persecution that 5 they themselves have suffered or must suffer. See Sun Wen Chen, 2007 WL 1760658, at *11 6 (McKee, J., dissenting) (“Congress could have easily drafted [§ 601(a)] to extend to ‘married 7 couples who have been subjected to a forced abortion or involuntary sterilization.’ So drafted, 8 an actual victim of persecution under a coercive population control program, as well as his/her 9 spouse, would qualify for relief under the statute. However, Congress did not draft the statute in 10 this way, and we can not rewrite the statute's explicit text to achieve that result.”). 11 Indeed, the critical defect in the BIA’s policy of according per se refugee status to 12 spouses of individuals explicitly protected by § 601(a) is its creation of an irrebuttable 13 presumption of refugee status for a new class of persons. This policy effectively absolves large 14 numbers of asylum applicants of the statutory burden to prove that they have (i) a well-founded 15 fear of persecution (ii) based on an impermissible nexus. Such a presumption is contrary to the 16 text of 8 U.S.C. § 1158(b)(1)(B), which specifies that “[t]he burden of proof is on the applicant,” 17 that “the applicant must establish that . . . political opinion was or will be at least one central 18 reason for persecuting the applicant,” and that an applicant’s testimony may be sufficient to meet 19 this burden only if it “refers to specific facts sufficient to demonstrate that the applicant is a 20 refugee.” The law is clear that “an agency is not free to ignore statutory language by creating a 21 presumption on grounds of policy to avoid the necessity for finding that which the legislature 19
    288. 1 requires to be found.” United Scenic Artists v. NLRB, 762 F.2d 1027, 1034 (D.C. Cir. 1985). 2 The creation of such a presumption “is beyond the [agency’s] statutory authority.” Id. at 1035; 3 see also Cerrillo-Perez v. INS, 809 F.2d 1419, 1426-27 (9th Cir. 1987) (holding that the “BIA 4 cannot adopt a general presumption” unfavorable to applicant but instead “must consider the 5 specific facts and circumstances of each case”). Thus, the BIA lacks authority to adopt a policy 6 that presumes that every person whose spouse was subjected to a forced abortion or sterilization 7 has himself experienced persecution based on political opinion. 8 Our concurring colleagues Judge Katzmann and Judge Sotomayor suggest that the 9 political nexus prong of this presumption is, in fact, inherent in Congress’s 1996 amendments. 10 Op. of Judge Katzmann at 15; op. of Judge Sotomayor at 7. However, this contention is 11 irreconcilable with the language of § 601(a), in which Congress created this presumption for 12 specifically identified persons – those who were themselves subjected to or threatened with a 13 forcible abortion or sterilization. To the extent that the amendments overruled Matter of 14 Chang’s categorical rejection of such a nexus, 20 I. & N. Dec. 38, this means that an asylum 15 applicant is no longer foreclosed from relying on evidence of a forced abortion or sterilization – 16 whether personal or spousal – in attempting to demonstrate persecution based on political 17 opinion. However, the fact remains that Congress has relieved only persons who actually 18 experienced, or are threatened with, a forcible abortion or sterilization from the burden of 19 proving a political nexus in their particular cases. 20 We do not deny that an individual whose spouse undergoes, or is threatened with, a 21 forced abortion or involuntary sterilization may suffer a profound emotional loss as a partner and 20
    289. 1 a potential parent. But such a loss does not change the requirement that we must follow the 2 “ordinary meaning” of the language chosen by Congress, according to which an individual does 3 not automatically qualify for “refugee” status on account of a coercive procedure performed on 4 someone else. See Moskal v. United States, 498 U.S. 103, 108 (1990) (stressing the importance 5 of “giving the ‘words used’ their ‘ordinary meaning.’”); Cardoza-Fonseca, 480 U.S. at 453 6 (“Where the language of [a] law[] is clear, we are not free to replace it with an unenacted 7 legislative intent.”).10 8 Under Chevron, once it is apparent that the statute is unambiguous, our inquiry stops. “If 9 the intent of Congress is clear, that is the end of the matter,” and we are required to refrain from 10 deferring to an agency’s contradictory interpretation. 467 U.S. at 842-43; see, e.g., Barnhart v. 11 Sigmon Coal Co., Inc., 534 U.S. 438, 462 (2002); Brown & Williamson Tobacco Corp., 529 U.S. 12 at 160-61; Nat’l Credit Union Admin. v. First Nat. Bank & Trust Co., 522 U.S. 479, 500 (1998); 13 MCI Telecommunications Corp. v. American Tel. & Tel. Co., 512 U.S. 218, 226-28 (1994); 14 Cardoza-Fonseca, 480 U.S. at 447-48. Congress’s choices of language in the phrases, “a person 15 who has been forced to abort a pregnancy or to undergo involuntary sterilization,” “[a person] 16 who has been persecuted for failure or refusal to undergo such a procedure,” and “a person who 17 has a well founded fear that he or she will be forced to undergo such a procedure or subject to 18 persecution for such failure [or] refusal,” is uniformly unambiguous in its reference to an 19 individual who is subjected to, or threatened with, an involuntary abortion or sterilization If this conclusion is inconsistent with Congress’s intentions, it can, if it so chooses, of 10 course, amend the statute, as it did when it adopted IIRIRA § 601(a) in response to the BIA’s decision in Matter of Chang, 20 I. & N. Dec. 38 (B.I.A. 1989). 21
    290. 1 affecting his or her own body, and the statutory scheme of § 1101(a)(42) only reinforces that 2 conclusion. 3 As a result, we conclude that the statute does not provide that a spouse – and a fortiori, a 4 boyfriend or fiancé – of someone who has been forced to undergo, or is threatened with, an 5 abortion or sterilization is automatically eligible for “refugee” status. Instead, to qualify for 6 refugee status under the amendment, such an individual must turn to the two remaining 7 categories of § 601(a), which provide protection to petitioners who demonstrate “other resistance 8 to a coercive population control program” or “a well founded fear that he or she will be . . . 9 subject to persecution for such . . . resistance . . . .” 8 U.S.C. § 1101(a)(42).11 10 In S-L-L, the BIA was influenced by the fact that C-Y-Z- “is a precedent of long standing 11 at this point . . . and numerous court decisions have deferred to the holding.” S-L-L-, 24 I. & N. 12 Dec. at 4; see also id. at 14 (Pauley, B.M., concurring) (“[N]otwithstanding my belief that Matter 13 of C-Y-Z-, . . . was wrongly decided, I would not overrule it now, clearly a decade later and in the 14 aftermath of thousands of decisions applying it . . . .”). However, the Supreme Court has made 15 clear that “[s]tare decisis is not an inexorable command; rather, it is a principle of policy and not 16 a mechanical formula of adherence to the latest decision.” Payne v. Tennessee, 501 U.S. 808, 17 828 (1991) (internal quotation marks omitted). While no party before us argues that the rule in S-L-L- fails under Chevron step one, 11 DHS did argue this point before the BIA, and the BIA considered and rejected the argument over a persuasive dissent by two members of the Board. In any event, we cannot defer to the Department of Justice’s argument (opposed below by DHS, the agency charged with enforcing immigration laws) that the rule in S-L-L- survives review under Chevron step one if the rule finds no support in the statutory text. Accordingly, we assume that the Solicitor General will take appropriate action to recommend or assure that the views of DHS and this Court will be represented in any future proceedings. 22
    291. 1 While stare decisis is undoubtedly of considerable importance to questions of statutory 2 interpretation, the Supreme Court “ha[s] never applied stare decisis mechanically to prohibit 3 overruling . . . earlier decisions determining the meaning of statutes.” Monell v. Dep’t of Social 4 Servs., 436 U.S. 658, 695 (1978). We should not do so either. The fact that we have failed to 5 follow the plain language of a law of Congress for ten years does not require that we do so 6 indefinitely. That would “place on the shoulders of Congress the burden of the Court's own 7 error.” Girouard v. United States, 328 U.S. 61, 70 (1946). 8 Given the clarity of the statute, there is no need to resort to legislative history, which is a 9 tool of construction that we employ only if the statutory text at issue in the context of the statute 10 as a whole is ambiguous. However, were we to examine the statute’s legislative history, we 11 would find that our interpretation of § 601(a) comports with Congress’s stated purpose in 12 passing the amendment. The House Report accompanying the passage of the amendment states 13 that its 14 primary intent . . . is to overturn several decisions of the Board of Immigration 15 Appeals, principally Matter of Chang and Matter of G- . . . which . . . hold that a 16 person who has been compelled to undergo an abortion or sterilization, or has 17 been severely punished for refusal to submit to such a procedure, cannot be 18 eligible on that basis for refugee or asylee status unless the alien was singled out 19 for such treatment on account of factors such as religious belief or political 20 opinion. 21 22 H.R. Rep. 104-469(I) at 173-74 (1996) (emphasis added). 23 The report mentions as examples of victims of coercive family planning policies women 24 who have been subjected to involuntary abortions, men and women who are forcibly sterilized, 25 and “couples with unauthorized children [who are] subjected to excessive fines,” but not spouses 23
    292. 1 of individuals who have been subjected to forced abortions or sterilizations. Id. While Congress 2 disapproved of coercive family planning policies as a whole, the amendment was meant to 3 provide protection for individuals who were subjected to persecution themselves. As the report 4 goes on to state: 5 The Committee emphasizes that the burden of proof remains on the 6 applicant, as in every other case, to establish by credible evidence that he 7 or she has been subject to persecution - in this case, to coercive abortion or 8 sterilization - or has a well-founded fear of such treatment. The Committee 9 is aware that asylum claims based on coercive family planning are often 10 made by entire groups of smuggled aliens, thus suggesting that at least 11 some of the claims, if not the majority, have been "coached." Section 12 [601(a)] is not intended to protect persons who have not actually been 13 subjected to coercive measures or specifically threatened with such 14 measures . . . . 15 16 Id. at 174 (emphasis added). There is nothing in the legislative history that leads us to question 17 “the strong presumption that Congress expresses its intent through the language it chooses.” 18 Cardoza-Fonesca, 480 U.S. at 432 n.12. Here, the language Congress employed in § 601(a) 19 demonstrates that it wanted to cover “a person,” not “a couple,” not a “significant other” and not 20 an “intimate friend.” Moreover, Congress’s “emphasi[s]” on its intention that “the burden of 21 proof remains on the applicant” demonstrates that no other subsection of § 1101(a)(42) could 22 support the BIA’s interpretation of the statute.12 See Sun Wen Chen, 2007 WL 1760658, at *16 12 Judge Katzmann asserts that Congress, when it adopted IIRIRA § 601(a) in 1996, intended to accomplish the same goal as a 1993 order of former Attorney General Barr that was never enacted and never reviewed by any court. That order would have explicitly granted asylum to spouses of coercive family planning policy victims. Op. of Judge Katzmann at 8-9. Judge Katzmann’s assertion is squarely contradicted by the plain language of IIRIRA § 601(a), which – unlike Attorney General Barr’s order – does not grant asylum to spouses of persecution victims, as well as by the legislative history recounted above, which emphasizes that IIRIRA § 601(a) was not intended to make asylum available to those not explicitly protected by the amendment. Even more precarious is Judge Katzmann’s reliance on the various messages he 24
    293. 1 (McKee, J., dissenting) (“The House Report . . . expresses a congressional intent to restrict 2 asylum to the ‘person’ who undergoes the coercive procedure just as clearly as the text of the 3 statute.”). 4 This reading of the statute is further supported by the Supreme Court’s pronouncement 5 about what “refugee” means. In INS v. Elias-Zacharias, the Supreme Court held that under the 6 plain language of the general refugee definition subsection of § 1101(a)(42), “‘persecution on 7 account of . . . political opinion’ in [§ 1101(a)(42)] is persecution on account of the victim’s 8 political opinion,” 502 U.S. 478, 482 (1992) (emphasis in original), not persecution on account 9 of a political opinion held by someone else. Similarly, a spouse who has not demonstrated that 10 he himself is a victim of persecution cannot be entitled to asylum under this section of the 11 statute. It would be unreasonable to conclude otherwise, whether under § 601(a), or under § 12 1101(a)(42) as a whole.13 13 Our conclusion that Congress never intended § 601 – or § 1101(a)(42) – to apply 14 automatically to spouses is reinforced by the fact that Congress already provides for family 15 members elsewhere in the statute by authorizing derivative asylum status for spouses and 16 children of individuals who qualify as “refugees.” 8 U.S.C. § 1158(b)(3)(A). Specifically, under hears in the sounds of Congressional silence. Op. of Judge Katzmann at 13. 13 In her concurring opinion, Judge Sotomayor suggests that today’s holding casts doubt on Jorge-Tzoc v. Gonzales, 435 F.3d 146 (2d Cir. 2006) (remanding asylum claim to BIA on determination that a “combination of circumstances” experienced by applicant in Guatemala as a dependent child – including the massacre of close family members – “could well constitute [past] persecution” of the child). We do not address this concern because today’s decision does not preclude the BIA from considering the totality of circumstances in any particular case to determine if an asylum applicant has carried his statutory burden. 25
    294. 1 § 1158(b)(3)(A), an individual whose spouse or parent has been granted asylum on the basis of 2 having undergone or been threatened with the prospect of a forced abortion or sterilization is 3 automatically eligible for derivative asylum: “[a] spouse or child . . . of an alien who is granted 4 asylum under this subsection may, if not otherwise eligible for asylum under this section, be 5 granted the same status as the alien if accompanying, or following to join, such alien.” 6 What Congress did by providing in § 1158(b)(3)(A) for automatic asylum for spouses of 7 individuals who have been granted “refugee” status as a result of having been forced to undergo 8 an abortion or involuntary sterilization reflects a policy decision to encourage the preservation of 9 families. Under § 1158(b)(3)(A) the benefits of asylum are extended first to the person Congress 10 understood to be most deserving of protection – the direct victim. Once the victim gains asylum, 11 so does the spouse, and so do their children. This structure encourages couples to remain 12 together, or, in circumstances where this is not possible, facilitates reunion. 13 The BIA’s interpretation of the statute in S-L-L cuts in a different direction since it has 14 the perverse effect of creating incentives for husbands to leave their wives. As hundreds of 15 cases in the courts illustrate, the policy allows a married man to “capitalize on the persecution of 16 his wife to obtain asylum even though he has left his wife behind and she might never join him 17 and he might intend that she not do so,” Chen v. Ashcroft, 376 F.3d 215, 223 n.2 (3d Cir. 2004). 18 It is highly unlikely – indeed, inconceivable – that Congress would approve of, much less 19 intended, any of this. These counterintuitive results reinforce our conclusion that in § 601(a) 20 Congress intended to grant automatic asylum to an individual directly victimized by a coercive 21 birth control policy, and that no part of § 1101(a)(42) extends this automatic asylum to spouses. 26
    295. 1 Congress did not, at the same time it adopted § 601(a), intend to gut or to undermine § 2 1158(b)(3)(A), which already contained a synchronous, sensible way of addressing the 3 compelling problems faced by spouses and children of direct victims. 4 Although we conclude that Congress has spoken unambiguously to whether an asylum 5 applicant is per se eligible for asylum if his spouse or partner has suffered as a result of a 6 coercive population control program, the phrase “other resistance” is ambiguous and leaves 7 room for the BIA’s reasonable interpretation where the applicant relies on something beyond his 8 spouse’s or partner’s persecution.14 See Yang v. U.S. Att’y Gen., 418 F.3d 1198, 1203 (11th Cir. 9 2005) (“There is very little case law analyzing the ‘other resistance’ clause in the asylum statute. 10 Additionally, a review of the legislative history behind the 1996 Amendment does not reveal any 11 clear intent from Congress on the scope of the ‘other resistance clause.’”). 12 In its decision, the BIA held that an applicant claiming persecution for “other resistance” 13 must demonstrate (1) “resistance” to a coercive family planning policy, which can “cover[] a 14 wide range of circumstances, including expressions of general opposition, attempts to interfere 15 with enforcement of government policy in particular cases, and other overt forms of resistance to 16 the requirements of the family planning law”; and (2) that the applicant has “suffered harm 17 amounting to persecution on account of that resistance.” S-L-L-, 24 I. & N. Dec. at 10. An 18 individual whose spouse or partner has been subjected to a forced abortion or involuntary 14 For an analysis of what “resistance” might mean when someone has not been forcibly sterilized himself, see Li v. Ashcroft, 356 F.3d 1153, 1159-61 (9th Cir. 2004) (en banc) (analyzing the meaning of the “other resistance” clause and holding that it applies to a woman who announced her opposition to government population control policies and was thereafter subjected to a forced gynecological exam and threatened with future abortion, sterilization of her boyfriend, and arrest). 27
    296. 1 sterilization can therefore qualify for “refugee” status under this interpretation if that individual 2 can prove past persecution or a fear of future persecution for “resistance” that is directly related 3 to his or her own opposition to a coercive family planning policy. 4 Whatever interpretation the BIA chooses to give to the meaning of “resistance,” it is clear 5 that the fact that an individual’s spouse has been forced to have an abortion or undergo 6 involuntary sterilization does not, on its own, constitute resistance to coercive family planning 7 policies. See Zhang, 395 F.3d at 532 (“[M]erely impregnating one's girlfriend is not alone an act 8 of ‘resistance.’”). Nor could the resistance of an individual’s spouse or partner to a family 9 planning policy – whether by failure or refusal to undergo a procedure, or for “other resistance” 10 – constitute, on its own, “resistance” under regulations implementing § 1101(a)(42). Instead, as 11 the DHS has argued, “where the applicant himself has not resisted [coercive family control 12 policies], he would need to demonstrate, though persuasive direct or circumstantial evidence, 13 that his wife’s, fiancee’s, or girlfriend’s resistance has been or will be imputed to him.” DHS 14 brief at 17 (citing Singh v. INS, 134 F.3d 962, 970 (9th Cir. 1998)). The fact that someone’s 15 spouse has been subjected to a forced abortion or sterilization would not be irrelevant to such an 16 analysis, it simply could not provide for asylum status per se.15 15 Judge Calabresi asserts first that our holding conflicts—if not in actuality, then in “spirit,” Op. of Judge Calabresi at 6 —with the Supreme Court’s recent decisions in INS v. Orlando Ventura, 537 U.S. 12 (2002) (per curiam), and Gonzales v. Thomas, 547 U.S. 183, 126 S. Ct. 1613 (2006) (per curiam), and second that we rush to reach a particular result. We disagree with both of these assertions. First, Orlando Ventura and Thomas held that a reviewing court should ordinarily remand rather than pass upon a matter that is (1) primarily committed to the BIA’s discretion, and (2) has not yet been considered by the BIA. See Orlando Ventura, 537 U.S. at 16-17; Thomas, 126 S. Ct. at 1615. Neither of these conditions is present in this case. We, rather than the BIA, have primary authority under Chevron to determine whether a particular agency interpretation is consistent with the unambiguously expressed intent of 28
    297. 1 Before turning to the dispositions of the petitioners’ claims, we address some practical 2 implications of our decision. We affirm the result of the BIA’s decision in S-L-L- denying per 3 se refugee status to boyfriends or fiancés of individuals who have been persecuted directly under 4 coercive family planning policies. A necessary predicate for this result is our conclusion that § 5 601 does not confer automatic asylum eligibility on spouses, whether legal spouses or spouses 6 from a traditional marriage, but only on individuals who themselves have undergone or been 7 threatened with coercive birth control procedures. Thus, although none of the petitioners before 8 us is legally married, we understand that our reading of the statute would necessarily exclude 9 spouses of those directly victimized from per se asylum eligibility as well. 10 We emphasize that our holding today should not be read to presage the reopening of 11 cases of aliens who have already been granted asylum based on the BIA’s interpretation of § 601 12 in C-Y-Z-. Under 8 U.S.C. § 1158(c)(2)(A), the DHS is permitted to seek the termination of 13 asylum when an alien no longer qualifies for refugee status “because, owing to a fundamental 14 change in circumstances relating to the original claim, the alien's life or freedom no longer would 15 be threatened on account of race, religion, nationality, membership in a particular social group, Congress. See, e.g., Cardoza-Fonseca, 480 U.S. at 447-48. Second, the BIA has had ample opportunity to consider the statutory interpretation question in the first instance. The per se rule that we now invalidate was first announced by the BIA in 1997, in its opinion in C-Y-Z-, 21 I. & N. Dec. at 915. In 2005, we remanded this case to the BIA to give it the opportunity to reconsider whether the rule in C-Y-Z- could find support in the language of § 601(a). See Lin, 416 F.3d at 187. On remand, DHS explicitly argued to the BIA that its per se rule was foreclosed by the plain language of § 601(a) and the statutory scheme. A majority of the BIA considered and rejected this argument over the forceful, and persuasive, objections of a minority of the Board. See S-L-L-, 24 I. & N. Dec. at 15-21 (Filppu, B.M., concurring and dissenting). We see no reason to remand yet again – ping pong style – when the BIA has had ten years and several opportunities to reconsider a rule that has no basis in statutory text. 29
    298. 1 or political opinion in the country from which deportation or removal was withheld.” 8 C.F.R. § 2 208.24(b)(1) (emphasis added). Just as a change in United States asylum law does not qualify as 3 a “change in circumstances” sufficient to reopen an asylum case under 8 C.F.R. § 4 1003.2(c)(3)(ii) (permitting motion to reopen “based on changed circumstances arising in the 5 country of nationality or in the country to which deportation is ordered”), see Azanor v. Ashcroft, 6 364 F.3d 1013, 1022 (9th Cir. 2004), a change in the BIA’s interpretation of section 601(a) as a 7 result of our decision should not be seen as a “fundamental change in circumstances relating to 8 the original claim” under 8 C.F.R. § 208.24(b)(1) so as to allow the termination of an asylum 9 claim that has already been granted. See S-L-L-, 24 I. & N. Dec. at 21 n.2 (Filppu, B.M., 10 concurring and dissenting) (“We are not now concerned with reopening past cases.”). 11 III. PETITIONERS’ CLAIMS 12 We agree with the BIA that none of the petitioners can qualify for automatic refugee 13 status as a result of the treatment of their girlfriends or fiancées. Instead, each petitioner must 14 demonstrate “other resistance to a coercive population control program” or “a well founded fear 15 that he . . . will be . . . subject to persecution for such . . . resistance . . . .” 8 U.S.C. § 16 1101(a)(42). 17 Petitioner Dong’s application for asylum was based upon his fiancée’s two forced 18 abortions and threats from family planning officials that they would fine and sterilize Dong if his 19 fiancée became pregnant again. But Dong failed to demonstrate that he acted in a manner that 20 could constitute “resistance” or opposition to a coercive family control program. Nor can we 21 find that Dong has a fear of future persecution as a result of the threat that the Chinese 30
    299. 1 government would sterilize him if his fiancée became pregnant again. Dong submitted evidence 2 to the immigration court that his fiancée had moved to Taiwan to be with her father. The IJ thus 3 found that Dong was unlikely to return to China, and his fear of sterilization was conjectural. 4 Moreover, as the BIA notes, were Dong’s fiancée to return to China, Dong and his fiancée would 5 now meet the age requirements for marriage. See In re Dong, A. 77 293 661 (B.I.A. Nov. 27, 6 2006). In addition, the immigration judge correctly determined that Dong did not demonstrate 7 that he would “more likely than not” be persecuted as grounds for a withholding of removal, or 8 that he would be subjected to torture within the meaning of the Convention Against Torture. See 9 id. Accordingly, we deny Dong’s petition for review. 10 Petitioner Lin’s claim for asylum is that he and his girlfriend were refused permission to 11 marry or have a child out of wedlock, and his girlfriend was forced to undergo an abortion. Lin 12 did not claim before the immigration court, the BIA, or this Court that he had “otherwise 13 resisted” China’s coercive family planning policies. Even if he had, we agree with the BIA that 14 a request, through the appropriate legal channels, for permission to have a child, combined with 15 the subsequent abortion performed on his girlfriend, does not constitute “resistance to a coercive 16 population control program.” S-L-L-, 24 I. & N. at 10-12. However, we do not have jurisdiction 17 over Lin’s petition for review, as we find that his petition is moot. Lin’s attorney has informed 18 this Court that he has never had any contact with the petitioner, and that Lin’s prior attorney has 19 not spoken to him in approximately three years. In addition, an individual from Lin’s village in 20 China has told the attorney that “he heard from other villagers that Lin was terminally ill and had 21 returned to China not too long ago and died.” Aff. of Yee Ling Poon ¶ 5. When the possibility 31
    300. 1 of relief is “so remote and speculative that any decision on the merits . . . would amount to a 2 ‘declar[ation of] principles or rules of law which cannot affect the matter in issue in the case 3 before [us],’” United States v. Blackburn, 461 F.3d 259, 262 (2d Cir. 2006) (quoting Mills v. 4 Green, 159 U.S. 651, 653 (1895) (alterations in original)), we would run afoul of Article III were 5 we to decide the claim. Given that we do not retain jurisdiction over Lin’s claim if he has 6 returned to China and has provided no explanation to overcome the presumption that his asylum 7 application has been abandoned, see 8 C.F.R. § 1208.8, or if he is deceased, Lin’s petition is 8 dismissed as moot. 9 Petitioner Zou’s petition has been remanded by the BIA to the immigration court to 10 review its findings of adverse credibility and determine whether Zou qualifies as a refugee for 11 “resistance” to a coercive family planning policy. See In re Xian Zou, No. A77 322 295 (B.I.A. 12 Nov. 21, 2006). Accordingly, we no longer have jurisdiction over Zou’s petition. See 8 U.S.C. 13 § 1252(a)(1) (giving courts of appeals jurisdiction to review only final orders of removal). 14 CONCLUSION 15 For these reasons, the petition for review of Zhen Hua Dong is DENIED. The petitions 16 for review of Shi Liang Lin and Xian Zou are DISMISSED for lack of jurisdiction. 17 18 19 20 21 32
    301. 1 2 3 4 5 6 7 8 9 10 11 12 33
    302. 1 KATZMANN , Circuit Judge, with whom Judges Straub, Pooler, and Sotomayor join, concurring in 2 the judgment: 3 With the majority’s emphasis on denying asylum relief to legal spouses under 4 § 101(a)(42) of the Immigration and Nationality Act (the “INA”), it is easy to lose sight of one 5 fact central to the disposition of these cases: Not one of the petitioners in these consolidated 6 cases was married. In each case, it was the plaintiff’s girlfriend, not his wife, who was forced to 7 abort her pregnancy. It is thus unnecessary for us to resolve whether the BIA can legally extend 8 asylum relief to legal spouses; indeed, in doing so, we are addressing not only an issue not 9 presently before us, but also one that the parties in these cases do not even dispute. In their 10 briefs before us, both the petitioners and the Government agree that the statute is ambiguous.1 11 The question the parties dispute, and the only one that these cases require us to answer, is 12 whether the BIA’s interpretation of the statute as applied to boyfriends and fiancés is reasonable. 13 Every judge on this Court who reaches this issue agrees that it is. 14 Thus, this case could have been resolved simply and nearly unanimously by assuming the 15 reasonableness of the BIA’s construction of the statute as applied to legal spouses and then 16 holding that it was also reasonable as applied to boyfriends and fiancés. See Cai Luan Chen v. 1 The Department of Homeland Security (“DHS”) advanced a different view before the BIA. The majority “assume[s] that the Solicitor General will take appropriate action to recommend or assure that the views of DHS and this Court will be represented in any future proceedings.” Maj. Op. at 23 n.11. In so doing, the majority ignores the historic independence of the Office of the Solicitor General in determining the executive branch’s position before the Supreme Court. When agencies of the executive branch have taken inconsistent positions, as they have here, the Solicitor General may choose among those positions, or it may adopt any other available litigation position. The Solicitor General has no obligation to endorse the preferred legal theory of the court below. To the extent the majority attempts to influence the position the Solicitor General will take in future proceedings, that effort is inappropriate. 34
    303. 1 Ashcroft, 381 F.3d 221, 227 (3d Cir. 2004) (Alito, J.) (assuming “for the sake of argument” that 2 “C–Y–Z–’s interpretation is permissible” as applied to legal spouses and then determining 3 whether the BIA was reasonable in distinguishing “between married and unmarried couples”). 4 Instead, the majority has gone out of its way to create a circuit split where none need exist, see 5 Maj. Op. at 5 n.4, thereby frustrating the BIA’s uniform enforcement of a national immigration 6 policy.2 Finding in textual silence an expression of unambiguous congressional intent, the 7 majority has rejected the BIA’s determination that § 101(a)(42) is ambiguous. 8 When a governmental body with substantial experience in interpreting a complex 9 statutory scheme concludes that a statute is ambiguous, that determination should give us pause. 10 Here, the fact that the BIA concluded that the INA is ambiguous with respect to the question we 11 are called upon to answer suggests that we would do well to probe further, to consider whether 12 the seemingly plain language belies a more complicated meaning. It suggests that we should 13 consider carefully not only the text of the statute, but also the context—both the events that gave 14 rise to that text and the various agency and judicial responses to it. Text without context can 15 lead to confusion and misunderstanding. The majority’s analysis is testament to that 16 proposition.3 2 In a related context, we remanded to the BIA to address the proper scope of the term “refugee” in the first instance, noting that “it would be unsound for each of the several Courts of Appeals to elaborate a potentially nonuniform body of law” and describing uniformity as “especially desirable in cases such as these.” Jian Hui Shao v. BIA, 465 F.3d 497, 502 (2d Cir. 2006). 3 The majority finds my decision to engage in this analysis “perplex[ing]” given my belief that we need not—and should not—answer today a question that this case does not require us to answer and the proper resolution of which the parties do not dispute. Although I believe we should have limited our decision to the BIA’s treatment of boyfriends, the majority has 35
    304. 1 In answering the first of the questions set out in Chevron, U.S.A., Inc. v. Natural 2 Resources Defense Council, 467 U.S. 837 (1984), that is, whether “the statute is silent or 3 ambiguous with respect to the specific issue,” id. at 843, we must look to the text of § 101(a)(42) 4 of the INA. That text provides, in pertinent part, that a refugee is 5 any person who is outside any country of such person’s nationality 6 . . . and who is unable or unwilling to return to . . . that country 7 because of persecution . . . on account of . . . political opinion . . . . 8 For purposes of determinations under this chapter, a person who has 9 been forced to abort a pregnancy or to undergo involuntary 10 sterilization . . . shall be deemed to have been persecuted on account 11 of political opinion . . . . 12 13 8 U.S.C. § 1101(a)(42). The majority focuses its textual analysis on the final sentence of that 14 provision, that is, § 601(a) of the Illegal Immigration Reform and Immigration Responsibility 15 Act of 1996 (“IIRIRA”), concluding that the BIA’s decision “rested on” that sentence, and not 16 the general definition of “refugee” in 8 U.S.C. § 1101(a)(42). Maj. Op. at 10. Yet, as the 17 majority notes, the BIA found that “‘[t]here is no clear or obvious answer to the scope of the 18 protections afforded by the amendment to partners of persons forced to submit to an abortion or 19 sterilization.’” Maj. Op. at 13.4 Reasoning that the “lack of . . . a reference” to spouses in the nonetheless chosen to address its treatment of husbands. I would be remiss if I did not discuss both why I believe the majority’s discussion of this issue is unnecessary and also why I believe it is wrong. 4 Although we remanded to the BIA to “more precisely explain its rationale for construing IIRIRA § 601(a)” to protect spouses, Shi Liang Lin v. U.S. Dep’t of Justice, 416 F.3d 184, 187 (2d Cir. 2005), nothing in that decision precluded the BIA from looking to the more general language in the statute in determining whether to extend relief to spouses. Indeed, the BIA had the authority to interpret the meaning of 8 U.S.C. § 1101(a)(42) independent of our remand in Lin. See INS v. Aguirre-Aguirre, 526 U.S. 415, 425 (1999) (“The Attorney General . . . has vested the BIA with power to exercise the discretion and authority conferred upon the Attorney General by law in the course of considering and determining cases before it. Based on 36
    305. 1 1996 amendment “does not necessarily preclude an applicant from demonstrating past 2 persecution based on harm inflicted on a spouse when both spouses are harmed by government 3 acts motivated by a couple’s shared protected characteristic,” the Board looked to the “general 4 principles regarding nexus and level of harm” for guidance. In re S–L–L–, 24 I. & N. Dec. 1, 5 5 (B.I.A. 2006); see also id. at 5 n.5 (citing to the general regulatory framework that defines when 6 an applicant may qualify as a refugee); id. at 6 (considering the “well-established principles 7 regarding nexus and level of harm for past persecution”). 8 Thus, although the majority places great emphasis on its view that the “language in § 9 601(a) . . . is unambiguous and . . . does not extend automatic refugee status to spouses or 10 unmarried partners of individuals § 601(a) expressly protects,” Maj. Op. at 4, I find the focus on 11 the amendment misplaced. I believe we must look instead to the entirety of 8 U.S.C. 12 § 1101(a)(42) to determine whether the statute is ambiguous. See SEC v. Chenery Corp., 332 13 U.S. 194, 196 (1947) (“[A] reviewing court, in dealing with a determination or judgment which 14 an administrative agency alone is authorized to make, must judge the propriety of such action 15 solely by the grounds invoked by the agency.”).5 That section provides that a “refugee” is any this allocation of authority, we [have] recognized . . . that the BIA should be accorded Chevron deference as it gives ambiguous statutory terms concrete meaning through a process of case-by- case adjudication . . . .” (quotation marks and citation omitted)); see also Kuhali v. Reno, 266 F.3d 93, 102 (2d Cir. 2001) (granting “Chevron deference to the Board’s construction of the INA, which it . . . administer[s]”). 5 Even if the BIA construed only the 1996 amendment, it would still be necessary to consider the text of the entire statute and the context against which that amendment was enacted to determine whether the 1996 amendment is itself ambiguous within the meaning of Chevron. See Natural Res. Def. Council v. Abraham, 355 F.3d 179, 198 (2d Cir. 2004) (“In interpreting the plain language of the statute, we must look to the particular statutory language at issue, as well as the language and design of the statute as a whole, and, where appropriate, its legislative 37
    306. 1 individual who cannot return to his or her home country because of “persecution . . . on account 2 of . . . political opinion.” None of these terms is defined in any way, and none explicitly 3 addresses whether the spouses of those who have been forced to undergo an abortion or 4 sterilization are entitled to asylum relief.6 Hence, the statute, on its face, does not “directly 5 address[] the precise question at issue.” Chevron, 467 U.S. at 843; cf. Sun Wen Chen v. U.S. 6 Att’y Gen., No. 05-4011, 2007 WL 1760658, at *4 (3d Cir. June 20, 2007) (“[T]he C–Y–Z– rule 7 thus fleshes out an issue germane to the application of § 1101(a)(42)(B) that was not addressed 8 by Congress, and so poses no Chevron step one problem.”). 9 Indeed, the majority points to no language in the statute that explicitly denies asylum 10 relief to these spouses, or that precludes the BIA from extending it to them. Cf. 8 U.S.C. § 11 1101(a)(42) (explicitly precluding from the definition of “refugee” individuals “who ordered, 12 incited, assisted, or otherwise participated in the persecution of any person on account of race, 13 religion, nationality, membership in a particular social group, or political opinion”). Rather, the 14 majority takes the position that “Congress’s specific designation of some persons (i.e., those who 15 fear, resist, or undergo particular medical procedures) is incompatible with the view that others 16 (e.g., their spouses) should also be granted asylum per se because of birth control policies. The history.” (quotation marks omitted)). Because I believe the BIA held that the entire provision was ambiguous, it is not necessary to determine here whether the 1996 amendment alone is ambiguous. Thus, contrary to Judge Calabresi’s suggestion, Judge Calabresi Op. at 16 n.6, I do not necessarily agree that the BIA could not have relied on § 601(a) to provide asylum relief to spouses. 6 I agree with Judge Calabresi that this “general language” in 8 U.S.C. § 1101(a)(42) means that the BIA, with its expertise in this area, is particularly well-suited to exercise its discretion and decide how that language should be interpreted. Judge Calabresi Op. at 8. Judge Calabresi and I differ as to whether the BIA has already exercised that discretion. 38
    307. 1 inclusion of some obviously results in the exclusion of others.” Maj. Op. at 17-18. But where 2 the majority reads the language of the 1996 amendment and sees it as a limitation on the 3 availability of asylum relief, I see it, in the context in which it was enacted, as an expansion of 4 that relief. 5 I believe one question is fundamental: What was Congress’s purpose in enacting the 1996 6 amendment? Cf. Ai Feng Yuan v. U.S. Dep’t of Justice, 416 F.3d 192, 197 (2d Cir. 2005) (noting 7 the “canon of statutory construction that requires us to reconcile a statute’s plain language with 8 its purpose”). In answering that question, we should not limit our inquiry to the language of the 9 amendment; instead, we must look at the statutory scheme of which that amendment is a part and 10 the legislative activity that led to its enactment. Congress’s intent in enacting IIRIRA § 601(a) 11 was to clarify that, contrary to the BIA’s prior rulings, the imposition of some aspects of China’s 12 family planning policy can constitute persecution on the basis of political opinion, and that 13 certain victims of that persecution are entitled to protection under our asylum laws. Nothing in 14 the amendment suggests that Congress intended to prevent the BIA from extending relief to 15 victims other than those explicitly identified in the amendment. See Sun Wen Chen, 2007 WL 16 1760658, at *6 (“We are not convinced that Congress, in expanding asylum to include more 17 reproductive rights-based claims, intended to define the outer limits of relief in such cases.”). 18 There is thus some irony in the majority’s approach: By giving short shrift to context, it infers an 19 intent to limit the availability of asylum relief; had it looked more closely at context, it would 20 have recognized Congress’s intent was only to expand that relief.7 Context makes clear what The majority’s recognition that § 601(a) expanded the availability of asylum relief does 7 nothing to explain why this language, which by its terms only expands relief, should also be read 39
    308. 1 text alone fails to convey. 2 The INA provides asylum relief to individuals who have been “persecut[ed] . . . on 3 account of . . . political opinion,” but does not define those terms. In Matter of Chang, 20 I. & 4 N. Dec. 38 (B.I.A. 1989), the BIA held that whatever else “persecution” might mean, it did not 5 encompass any retribution visited upon individuals who violated China’s “one couple, one child” 6 policy. The BIA ruled that the policy “is [not] on its face persecutive” and does not “persecut[e] 7 any portion of the Chinese citizenry on account of one of the reasons enumerated in section 8 101(a)(42)(A) of the Act.” Id. at 43, 44. 9 Others in the executive branch took a different view. The next year, the Department of 10 Justice issued “interpretative rules and general statements of policy for establishing statutory 11 eligibility for asylum or withholding of deportation on the basis of political opinion for aliens 12 who express a fear of coercive population control policies in their homeland.” Refugee Status, 13 Withholding of Deportation and Asylum; Burden of Proof, 55 Fed. Reg. 2803, 2804 (Jan. 29, 14 1990).8 President George H.W. Bush reaffirmed his Administration’s support of the interim rule to have limited the BIA’s preexisting authority to further expand it. Although the majority suggests that I rely on “messages [I] hear[] in the sounds of Congressional silence,” Maj. Op. at 26 n.12, all that I actually find in silence, especially given the context, is ambiguity. See Chevron, 467 U.S. at 843 (noting that we turn to Chevron step two “if the statute is silent or ambiguous with respect to the specific issue”); see also Sun Wen Chen, 2007 WL 1760658, at *3 (“Chevron deference embodies the judgment that agencies, rather than courts, ought to serve as gap-fillers in situations of statutory silence.”). It is the majority that appears to find in silence clear evidence of Congress’s intent. 8 These amendments to the asylum regulations appear to contemplate the possibility that asylum relief would be available to the spouses of those who were subject to forced abortion or sterilization. See Refugee Status, 55 Fed. Reg. at 2805 (to be codified at C.F.R. § 208.5(b)(2)) (“An applicant who establishes that the applicant (or applicant’s spouse) has refused to abort a pregnancy or to be sterilized in violation of a country’s family planning policy, and who has a 40
    309. 1 with the promulgation of Executive Order 12,711, which provided for “enhanced consideration 2 under the immigration laws for individuals from any country who express a fear of persecution 3 upon return to their country related to that country’s policy of forced abortion or coerced 4 sterilization.” Exec. Order No. 12,711, 55 Fed. Reg. 13,897, 13,897 (Apr. 11, 1990). 5 Although the INS, in July 1990, set forth a final rule that did not address this issue, see 6 Aliens and Nationality; Asylum and Withholding of Deportation Procedures, 55 Fed. Reg. 7 30,674 (July 27, 1990), the Office of the General Counsel of the INS subsequently sent a letter to 8 Regional and District Counsel stating that “Department of Justice and INS ‘policy with respect 9 to aliens claiming asylum or withholding of deportation based upon coercive family planning 10 policies is that the application of such coercive policies does constitute persecution on account of 11 political opinion.’” Xin-Chang Zhang v. Slattery, 55 F.3d 732, 740 (2d Cir. 1995), superseded 12 by 8 U.S.C. § 1101(a)(42) (quoting INS letter to Regional and District Counsel). 13 In January 1993, Attorney General William P. Barr signed a final rule that would have 14 made this view law. It provided, in pertinent part, that “[a]n applicant (and the applicant’s 15 spouse, if also an applicant) shall be found to be a refugee on the basis of past persecution on 16 account of political opinion if the applicant establishes that, pursuant to the implementation . . . 17 of a family planning policy . . . the applicant has been forced to abort a pregnancy or to undergo 18 sterilization or has been persecuted for failure or refusal to do so.” AG Order No. 1659-93, at 14 well-founded fear that he or she will be required to abort the pregnancy or to be sterilized or otherwise persecuted if the applicant were returned to such country may be granted asylum.”); see also id. (to be codified at C.F.R. § 242.17(c)) (“Eligibility for withholding of deportation on account of political opinion is established by the respondent who establishes that he or she (or respondent’s spouse) will be required to abort a pregnancy or to be sterilized . . . .”). 41
    310. 1 (Jan. 15, 1993) (to be codified at 8 C.F.R. 208) (“January 1993 Rule”). Although this rule was 2 sent to the Federal Register, where it was made available for public inspection and scheduled for 3 publication, it was never published due to the change in presidential administrations. Xin-Chang 4 Zhang, 55 F.3d at 741. In February, additional regulations pertaining to asylum were published, 5 but these made no mention of the January rule. Id. 6 Against the background of these conflicting BIA decisions and administrative 7 regulations, 8 we were asked to determine whether asylum relief was available to victims of China’s family 9 planning policy. We held that such relief was not available, explaining that “[e]ven were we to 10 accept [the] view that the ‘administrative cacophony’ surrounding the issue justified diminished 11 deference to Chang, our result would not change. It is difficult to frame a result different from 12 the holding of Chang that would be ‘reasonable’ under both Elias-Zacarias and the existing 13 immigration laws.” Id. at 752; see also INS v. Elias-Zacarias, 502 U.S. 478 (1992). We did not 14 stop there, however. Instead, we called upon Congress and the President to determine whether 15 relief should be available to the victims of China’s family planning policy: “No doubt, the 16 President and the Congress acting together have power to create an exception to the existing 17 immigration laws for PRC citizens . . . .” Id. 18 The next year, Congress enacted IIRIRA § 601(a). The House Committee Report 19 explained that Congress’s “primary intent” in amending the definition of refugee was “to 20 overturn several decisions of the [BIA], principally Matter of Chang and Matter of G–,” H.R. 21 Rep. No. 104-469(I), 1996 WL 168955, at *173 (1996), in which the BIA had held that “the 42
    311. 1 Chinese Government’s implementation of its family planning policies is not on its face 2 persecutive and does not by itself create a well-founded fear of persecution on account of one of 3 the five grounds delineated in the Act, even to the extent that involuntary sterilization may 4 occur,” Matter of G-, 20 I. & N. Dec. 764, 778 (B.I.A. 1993); Matter of Chang, 20 I. & N. Dec. 5 at 43-44; see also Junshao Zhang v. Gonzales, 434 F.3d 993, 995 (7th Cir. 2006) (“The impact 6 of that amendment was to overrule Chang, and allow for the granting of asylum applications in 7 cases in which the claim of persecution stemmed from the enforcement of China’s coercive 8 population control policies.”); H.R. Rep. No. 104-469(I), 1996 WL 168955, at *174 (criticizing 9 the BIA decisions for “effectively preclud[ing] from protection persons who have been 10 submitted to undeniable and grotesque violations of fundamental human rights”); id. (noting that 11 “the BIA’s rationale for these opinions—that policies of coercive family planning are ‘laws of 12 general application’ motivated by concerns over population growth, and thus are not 13 ‘persecutory’—is unduly restrictive”).9 14 When Congress stated that “[f]or purposes of determinations under this [Act], a person 15 who has been forced to abort a pregnancy or to undergo involuntary sterilization . . . shall be 16 deemed to have been persecuted on account of political opinion,” 8 U.S.C. § 1101(a)(42)(a), it 17 was not providing an exhaustive list of those who could claim asylum relief because they were 9 As previously noted, Attorney General Barr had attempted to “supersede the [BIA] decision in Matter of Chang,” January 1993 Rule, at 4-5, but his efforts fell victim to a change in presidential administrations. See Xin-Chang Zhang, 55 F.3d at 741. Noting that the next administration had thus far failed to take action, Congress stepped in to accomplish the same goal through legislation. See H.R. Rep. No. 104-469(I), 1996 WL 168955, at *174 (“[T]he Administration, which has the authority to overrule the BIA decisions through regulation or through decision of the Attorney General, has not done so. Nor has it offered adequate relief to persons who have undergone such coercion.”). 43
    312. 1 victimized by China’s family planning policy. Rather, it was expressing a congressional 2 determination that, contrary to the BIA’s prior rulings, China’s “one couple, one child” policy is 3 on its face persecutory, and victims of that policy who experienced persecution should be able to 4 qualify for asylum relief without making an additional showing of their own political opinion. 5 The majority takes the position that § 601(a) does not “alter the pre-IIRIRA definition of 6 ‘political opinion’ in § 1101(a)(42)” and that, for the individuals identified in § 601(a), “their 7 political opinion exists de jure rather than as a matter of fact on which the applicant bears the 8 burden of proof.” Maj. Op. at 18, 19. It notes, in this regard, that the amendment provides that 9 those who have been subjected to forced procedures “‘shall be deemed’ persecuted by reason of 10 political opinion,” Maj. Op. at 18-19, and that “[t]o ‘deem’ is ‘[t]o treat (something) as if (1) it 11 were really something else, or (2) it has qualities that it doesn’t have,’” Maj. Op. at 18-19 n.9 12 (quoting Black’s Law Dictionary 446 (8th ed. 2004)). But, as the majority acknowledges, there 13 is more than one definition of the word “deem”; it may also mean “[t]o consider, think, or 14 judge.” Black’s Law Dictionary 446 (8th ed. 2004). Thus, in this context, Congress’s use of the 15 word “deemed” may mean that these individuals should be “judged” as having been persecuted 16 on account of political opinion, just as the proposed 1993 rule provided that these applicants 17 “shall be found to be . . . refugee[s] on the basis of past persecution on account of political 18 opinion.” January 1993 Rule, at 14.10 Indeed, the legislative history suggests that Congress was 10 That two definitions of the term exist suggests, at the very least, that there is ambiguity in Congress’s use of the term, and the fact that Black’s Law Dictionary may favor one definition is hardly sufficient to dispel that ambiguity. Indeed, although Black’s Law Dictionary may describe the majority’s preferred usage as the more “traditional[]” one and may discourage other uses, the word “deem” has long been used to mean “consider, think, or judge.” See, e.g., U.S. Const. art. V (“The Congress, whenever two thirds of both Houses shall deem it necessary, shall 44
    313. 1 not attempting to create an exception to Matter of Chang, but to overrule it; and if Congress were 2 attempting to carve out an exception to the normal requirement that applicants must establish 3 that they have faced persecution on account of one of the protected grounds, it could have done 4 so in much plainer language. In any event, the statute is, at a minimum, ambiguous, and it is the 5 BIA, not the courts, that is charged with construing the statute in the face of that ambiguity. 6 Other circuits to have considered this issue have held that when Congress enacted the 7 1996 amendment it intended to protect both members of couples that are targeted under China’s 8 family planning policy. See, e.g., Junshao Zhang, 434 F.3d at 999 (“Congress passed § 9 601(a)(1) of the IIRIRA to ensure that families who are victims of forced abortion and 10 sterilization under China’s population control policy would receive asylum . . . .” (emphasis 11 added)); Kui Rong Ma v. Ashcroft, 361 F.3d 553, 559 (9th Cir. 2004) (identifying “Congress’s 12 goal in passing the amendments—to provide relief for ‘couples’ persecuted on account of an propose Amendments to this Constitution . . . .” (emphasis added)); 15 U.S.C. § 80b-3(j) (“The Commission is authorized to adopt rules, regulations, and orders . . . as it deems appropriate to implement this subsection.” (emphasis added)). Indeed, Congress has repeatedly used the term “deem” in the INA itself to mean “consider, think, or judge.” See, e.g., 8 U.S.C. § 1101(a)(27)(J)(i) (defining the term “special immigrant” to include “an immigrant who is present in the United States . . . who has been declared dependent on a juvenile court located in the United States or whom such a court has legally committed to, or placed under the custody of, an agency or department of a State and who has been deemed eligible by that court for long-term foster care due to abuse, neglect, or abandonment” (emphasis added)); id. § 1103(a)(3) (providing that the Secretary of Homeland Security may “perform such other acts as he deems necessary for carrying out his authority under the provisions of this chapter” (emphasis added)). A “normal rule of statutory construction” provides that “identical words used in different parts of the same act are intended to have the same meaning.” Beharry v. Ashcroft, 329 F.3d 51, 61 (2d Cir. 2003) (quoting Gustafson v. Alloyd Co., 513 U.S. 561, 570 (1995) (quotation marks omitted)). Even if this rule does not conclusively resolve any ambiguity created by the existence of the two definitions, Congress’s use of the allegedly disfavored definition elsewhere in the INA renders the meaning of “deemed” in § 1101(a)(42), at the very least, ambiguous. 45
    314. 1 ‘unauthorized’ pregnancy and to keep families together” (emphasis added)). We need not go as 2 far as these courts to affirm the BIA’s decision here. Even if Congress did not specifically 3 intend to protect “couples,” there is nothing in the text of the amendment, or the context that 4 gave rise to it, that indicates that it intended to preclude the BIA from extending asylum relief to 5 both members of a couple. See Sun Wen Chen, 2007 WL 1760658, at *5 (granting Chevron 6 deference to the BIA’s interpretation in In re S–L–L–).11 7 Just as nothing in the language or history of the amendment indicates a congressional 8 intent to foreclose the extension of relief to spouses, Congress has done nothing to indicate such 9 an intent in the years since the amendment’s enactment, notwithstanding that the BIA interpreted 10 § 1101(a)(42) to cover spouses a decade ago and numerous courts of appeals have upheld this 11 interpretation as reasonable. See, e.g., Yuan, 416 F.3d at 197; Junshao Zhang, 434 F.3d at 999; 12 Kui Rong Ma, 361 F.3d at 559 (“The BIA and the courts have uniformly applied the statute’s 11 Although “[g]eneral language of a statutory provision . . . will not be held to apply to a matter specifically dealt with in another part of the same enactment,” D. Ginsberg & Sons, Inc. v. Popkin, 285 U.S. 204, 208 (1932), that rule has no applicability here where the specific provision does not address whether spouses should be entitled to relief. In the 1996 amendment, Congress identified a particular category of individuals entitled to relief, but left to the BIA the task of determining whether to expand upon that relief, just as the BIA routinely defines what individuals are entitled to asylum relief in a myriad of other contexts. Indeed, there is some language in the legislative history which suggests that Congress may have considered the possibility that individuals other than those who had been forced to undergo an abortion or sterilization might qualify for asylum. The House Committee Report on the amendment noted that “[d]etermining . . . whether the actual or threatened harm rises to the level of persecution is a difficult and complex task, but no more so in the case of claims based on coercive family planning than in cases based on other factual situations. Asylum officers and immigration judges are capable of making such judgments.” H.R. Rep. No. 104-469(I), 1996 WL 168955, at *174. If only those subjected to the procedures could claim asylum relief, immigration judges and the BIA would never have needed to consider whether an applicant’s harm rises to the level of persecution, given that the amendment explicitly provides that a forced abortion or sterilization does rise to this level. 46
    315. 1 protections to husbands whose wives have undergone abortions or sterilization procedures, as 2 well as to the wives themselves.”); In re C–Y–Z–, 21 I. & N. Dec. 915, 918-19 (B.I.A. 1997) (en 3 banc). There are obscure areas of public policy, largely hidden from public attention and 4 concern, in which it makes little sense to ascribe meaning to the absence of congressional 5 response to administrative and judicial interpretations of a statute. Immigration is hardly one of 6 those areas. To the contrary, immigration—and the issue of the appropriate scope of asylum 7 relief—have consistently been on Congress’s radar. Immigration is frequently in the news, and 8 Congress has repeatedly legislated in this area. Indeed, as recently as 2005, Congress revisited 9 this very provision and removed the annual cap on the number of asylees who could be admitted 10 under it. See REAL ID Act of 2005, Pub. L. No. 109-13, Div. B, §§ 101(g)(2), (h)(5), 119 Stat. 11 231, 305-06 (May 11, 2005). While the fact that Congress, in the course of its active attention to 12 immigration issues and legislation, has not amended 8 U.S.C. § 1101(a)(42) in light of the 13 interpretation it has been given by the BIA and the courts does not definitively mean that 14 Congress intended to protect spouses, it does suggest, at the very least, that it was not Congress’s 15 intent to foreclose that relief. 16 The majority nonetheless holds that it was “not correct” for the BIA to construe the 17 general definition of “refugee” to allow for the provision of this relief because “the statutory 18 scheme unambiguously dictates that applicants can become candidates for asylum relief only 19 based on persecution that they themselves have suffered or must suffer.” Maj. Op. at 19. I agree 20 that an individual must have personally experienced persecution to be entitled to asylum relief, 21 but that statement begs the question of what constitutes persecution. 47
    316. 1 The statute does not, in either the 1996 amendment or in its general definition of the term 2 “refugee,” prescribe exactly how much harm or what kind of harm an individual must experience 3 to have been “persecuted” within the meaning of the statute. See Ivanishvili v. U.S. Dep’t of 4 Justice, 433 F.3d 332, 340 (2d Cir. 2006) (noting that the term “persecution” is “not defined by 5 the Immigration and Nationality Act”); see also Matter of Acosta, 19 I. & N. Dec. 211, 222 6 (B.I.A. 1985), overruled on other grounds by Matter of Mogharrabi, 19 I. & N. Dec. 439 (B.I.A. 7 1987) (defining persecution as “either a threat to the life or freedom of, or the infliction of 8 suffering or harm upon, those who differ in a way regarded as offensive” (emphasis added)); see 9 also Ivanishvili, 433 F.3d at 341 (acknowledging the BIA’s definition of “persecution” as “the 10 infliction of suffering or harm upon those who differ on the basis of a protected statutory 11 ground.”). The 1996 amendment states that when one is forced to undergo an abortion or 12 sterilization, the harm thereby experienced is sufficient to qualify for asylum, but it hardly makes 13 that level of harm necessary. 14 The majority may view the 1996 amendment as providing that only individuals who have 15 undergone a forced abortion or sterilization have experienced “persecution.” If so, we again 16 differ on our interpretation of the significance of the 1996 amendment. I believe Congress 17 enacted the 1996 amendment not primarily to define the term “persecution,” but to clarify what it 18 means to be persecuted “on account of political opinion.” As noted above, Congress sought to 19 make clear, contrary to the BIA’s earlier decision in Matter of Chang, that the imposition of 20 penalties through the implementation of China’s family planning policy can constitute 21 persecution “on account of political opinion” by effectively adopting the position taken in the 48
    317. 1 commentary to the 1993 rule that, “[a]lthough . . . prosecution and punishment under a law of 2 general applicability will not ordinarily constitute persecution ‘on account of’ one of the 3 statutory grounds, . . . [p]ersecution on account of political opinion encompasses persecution of 4 people whose violation of laws may not be motivated by their political opinions but is regarded 5 by the state as political disloyalty.” January 1993 Rule, at 8. 6 Thus, I do not think that § 601(a) unambiguously defines the term persecution, and the 7 majority has pointed to nothing in the statute that suggests that the emotional and psychological 8 harm one suffers when one’s spouse is forced to undergo an abortion or sterilization is not severe 9 enough to constitute persecution. Nor does anything in the statute preclude the BIA from 10 considering the effect that China’s family planning policies may have on a couple’s shared right 11 to reproduce and raise children. Because Congress did not specifically address these issues, the 12 statute is ambiguous. It therefore falls to the BIA to determine whether the harm an individual 13 experiences when his or her spouse is subjected to a forced abortion or sterilization is sufficient 14 to constitute persecution. See, e.g., Kuhali, 266 F.3d at 102 (granting “Chevron deference to the 15 Board’s construction of the INA, which it . . . administer[s]”); cf. Sun Wen Chen, 2007 WL 16 1760658, at *5 (noting that the BIA “exercised its delegated gap-filling authority reasonably” 17 when it recognized the harms an individual experiences as a result of the forced abortion or 18 sterilization of his spouse). By holding that persecution cannot encompass such individuals, the 19 majority, as Judge Sotomayor cogently explains, usurps the BIA’s task of giving meaning to 20 ambiguous statutory terms. Further, by suggesting that the BIA is creating a presumption which 21 allows individuals to be granted asylum without proving that they were “persecuted . . . on 49
    318. 1 account of political opinion,” the majority errs. The BIA here created no presumption; rather, 2 fulfilling a basic responsibility, it simply discerned the meaning of those ambiguous terms.12 3 Finally, the majority finds support for its reading of the statute in its perception that there 4 is some tension between this rule and the provision of derivative asylum status for spouses of 5 individuals who qualify as “refugees.” Maj. Op. at 26-28. But in adopting this rule, the BIA 6 provided a basis by which individuals could claim asylum relief in their own right for harm they 7 suffered as a result of their spouses’ forced abortion or sterilization. There is no apparent tension 8 in providing derivative asylum status to spouses who have not themselves suffered any harm and 9 providing an additional basis of relief to those spouses who have, that is, those who have 10 themselves suffered harm when their partners were subjected to a forced abortion or sterilization. 11 See 8 U.S.C. § 1158(b)(3)(A) (providing derivative asylum to spouses who are “not otherwise 12 eligible for asylum under this section”); cf. Junshao Zhang, 434 F.3d at 1001 (noting that “it 13 would be particularly perverse for courts to treat a subsequent break-up of the marriage as 14 somehow lessening the impact of [the prior] persecution [of the husband]”). Because there is no 15 tension in providing these two separate, distinct forms of relief, it does not seem to me that the 16 availability of derivative asylum relief unambiguously precludes the BIA from providing an 12 The majority’s assertion that § 601 “relieve[s] . . . persons who actually experienced a forcible abortion or sterilization from the burden of proving a political nexus” is similarly problematic. Maj. Op. at 21. In my view, Congress did not relieve anyone of the burden of proving political nexus. Rather, it determined that an applicant can meet this burden by establishing proof of persecution pursuant to a coercive family planning policy. Given the majority’s agreement that any asylum applicant may “rely[] on evidence of a forced abortion or sterilization—whether personal or spousal—in attempting to demonstrate persecution based on political opinion,” Maj. Op. at 21 (emphasis added), I fail to see what additional evidence an applicant whose spouse was subjected to a forced abortion or sterilization would have to show to carry his burden of proving political nexus. 50
    319. 1 additional basis of relief to those whose spouses have been subjected to a forced abortion or 2 sterilization. See Sun Wen Chen, 2007 WL 1760658, at *4 (“We . . . do not believe that the 3 existence of derivative asylum status under a statute implies that Congress intended to foreclose 4 additional pathways to asylum specific to spouses.”). 5 At bottom, the majority seems to be motivated by a concern that the BIA’s rule will “gut 6 or . . . undermine” the availability of derivative asylum relief or have “the perverse effect of 7 creating incentives for husbands to leave their wives.” Maj. Op. at 28, 27. But nothing in the 8 BIA’s rule denies relief of any kind to husbands who come to this country with their wives; 9 moreover I think it very likely that husbands will find it advantageous to come with their wives 10 when possible because doing so will buttress the credibility of their claims. In the end, however, 11 my views of the policy consequences of the BIA rule are immaterial; so, too, are those of the 12 majority. Once we determine that the statute is ambiguous, we must defer to the BIA’s 13 interpretation of the statute if it is reasonable, whatever our own personal policy preferences. 14 See, e.g., Udall v. Tallman, 380 U.S. 1, 16 (1965) (“To sustain the Commission’s application of 15 this statutory term, we need not find that its construction is the only reasonable one, or even that 16 it is the result we would have reached had the question arisen in the first instance in judicial 17 proceedings.” (quotation marks omitted)). 18 In considering the reasonableness of the BIA’s interpretation under step two of Chevron, 19 our charge is not to determine whether the BIA’s interpretation of the statute is the one we would 20 have adopted in the first instance. Instead, “the question for the court is whether the agency’s 21 answer is based on a permissible construction of the statute”; we must defer to “legislative 51
    320. 1 regulations . . . unless they are arbitrary, capricious, or manifestly contrary to the statute.” 2 Chevron, 467 U.S. at 843, 844. Here, in determining whether to extend asylum relief to spouses, 3 the BIA reasonably considered the general principles underlying the definition of persecution 4 and concluded that a husband is persecuted “when the government forces an abortion on a 5 married couple.” In re S–L–L–, 24 I. & N. Dec. at 6; see also id. (“When the government 6 intervenes in the private affairs of a married couple to force an abortion or sterilization, it 7 persecutes the married couple as an entity.”). I see no reason why the BIA could not reasonably 8 conclude that one has suffered harm or injury sufficiently severe to constitute persecution when 9 one’s spouse is forced to undergo an abortion or sterilization. Indeed, this determination finds 10 support in the decisions of a number of courts that have explicitly recognized that non-physical 11 harm may support a finding of past persecution in at least some circumstances. See Junshao 12 Zhang, 434 F.3d at 1001 (rejecting explicitly the “notion that [a husband] suffers no persecution 13 independent of his wife, as the result of the forcible abortion of his child” and holding that 14 “[a]lthough his wife was certainly a very direct victim of China’s population control measures, 15 Zhang was a victim as well. The forcible abortion has deprived him of his unborn child, of the 16 ability to realize the family that his wife and he had desired, and forever deprived him of the 17 ability to become a parent to that unborn son or daughter with his wife”); see also Ouk v. 18 Gonzales, 464 F.3d 108, 111 (1st Cir. 2006) (noting that “[u]nder the right set of circumstances, 19 a finding of past persecution might rest on a showing of psychological harm” (quotation marks 20 omitted)); Mashiri v. Ashcroft, 383 F.3d 1112, 1120 (9th Cir. 2004) (“Persecution may be 21 emotional or psychological, as well as physical.”); Abay v. Ashcroft, 368 F.3d 634, 642 (6th Cir. 52
    321. 1 2004) (holding that the applicant was entitled to asylum “based on her fear that her daughter will 2 be forced to undergo female genital mutilation” because her “fear of . . . being forced to witness 3 the pain and suffering of her daughter is well-founded”). 4 The BIA also determined that there were not “convincing reasons to extend the nexus and 5 level of harm attributed to a husband who was opposed to his wife’s forced abortion to a 6 boyfriend or fiancé.” In re S–L–L–, 24 I. & N. Dec. at 9.13 Recognizing that “marriage place[s] 7 the husband in a distinctly different position from that of an unmarried father,” id., the BIA noted 8 that unmarried fathers do not bear the same legal and societal responsibility for violations of 9 family planning policies. Indeed, because their relationships with their partners are not 10 registered with the government and may not even be known within the community, the 11 government may often be unaware of their identities. See id. at 9-10. The BIA thus presumed 12 that the family planning officials target legal spouses for persecution to a greater extent than 13 boyfriends and fiancés. See id. Furthermore, “[p]roof or presumption of paternity . . . may be 14 considerably more difficult when a boyfriend claims to have fathered a child who was forcibly 15 aborted by government officials.” Id. at 10. Although, as the BIA itself acknowledges, 16 “drawing the line at marriage is not” perfect, id. at 9, and reasonable policymakers could differ 17 as to how to draw the line, I cannot say that, under the deferential standard which guides us, the 13 Because petitioner Dong, the only petitioner whose claim we address on this appeal, had not participated in a traditional marriage ceremony, I need not determine now whether the BIA’s rule would also be reasonable as applied to individuals who were not old enough to marry under Chinese law and who participated in such a ceremony. Compare Junshao Zhang, 434 F.3d at 999 (holding that the BIA rule is unreasonable in this context), and Kui Rong Ma, 361 F.3d at 560 (same), with Cai Luan Chen, 381 F.3d at 231 (holding that the BIA rule is reasonable). 53
    322. 1 BIA’s reading is not based on a permissible construction of the statute. See Chevron, 467 U.S. at 2 843.14 If Congress disagrees with the BIA’s interpretation, it can overturn the decision.15 3 This case presents difficult and challenging questions at the heart of our immigration 4 laws. How we respond will affect the hopes and dreams of human beings seeking to live in 5 freedom. In enacting the INA, Congress established a framework for determining when asylum 6 relief should be provided to such individuals, and in doing so, it delegated considerable authority 7 to the BIA to fill in statutory gaps and define the broad language used in the INA. It is in 8 situations such as these that we should be particularly mindful of the views of the agency 9 charged by Congress with administering the statute, views that will reflect the agency’s 10 considerable experience and expertise. We should recognize that in such circumstances what is 11 advanced as the obvious answer may not be the right one. Here, the meaning of the text 12 becomes much less clear when one examines context, and the BIA, recognizing that ambiguity, 13 has offered a reasonable interpretation of the statute. I would defer to that interpretation. 14 Because petitioner Dong was not married and has not otherwise established his eligibility for asylum relief, I agree with the majority that his petition for review should be denied. I also agree that the petitions of Lin and Zou should be dismissed. 15 The BIA, too, remains free to revisit its decision. Although Judge Calabresi suggests that my approach “preclude[s] the agency from thinking deeply and fully about the matter,” Judge Calabresi Op. at 16, I do not understand why this would be so. Unlike the majority, I do not purport to offer an authoritative constructive of 8 U.S.C. § 1101(a)(42); I would hold only that the statute is ambiguous and that the BIA’s interpretation is reasonable. As Judge Calabresi elsewhere acknowledges, id. at 2, the BIA may always change its own interpretation of statutory law, so long as the change is not inconsistent with that law. See, e.g., Nat’l Cable & Telecomms. Ass’n v. Brand X Internet Servs., 545 U.S. 967, 981 (2005) (“For if the agency adequately explains the reasons for a reversal of policy, change is not invalidating, since the whole point of Chevron is to leave the discretion provided by the ambiguities of a statute with the implementing agency.” (quotation marks omitted)). 54
    323. 1 SOTOMAYOR, Circuit Judge, with whom Judge POOLER joins, concurring in the judgment: 2 Today’s decision marks an extraordinary and unwarranted departure from our 3 longstanding principles of deference and judicial restraint. Instead of answering the limited 4 question before us – whether the BIA’s denial of asylum to the unmarried partners of women 5 forced to undergo abortions or sterilization was unreasonable – the majority has chosen to go far 6 beyond it to address an issue that is unbriefed, unargued, and unnecessary to resolve this appeal. 7 Indeed, the cases before us, which involve only unmarried petitioners, are inappropriate vehicles 8 through which to opine on the merits of the BIA’s position with respect to spouses under 9 8 U.S.C. § 1101(a)(42).1 See Carducci v. Regan, 714 F.2d 171, 177 (D.C. Cir. 1983) (Scalia, J.) 10 (“The premise of our adversarial system is that appellate courts do not sit as self-directed boards 11 of legal inquiry and research, but essentially as arbiters of legal questions presented and argued 12 by the parties before them.”); see also Turner Broad. Sys., Inc. v. FCC, 520 U.S. 180, 224 (1997) 13 (declining to reach question which received little or no attention from the parties and noting 14 “prudence dictates that we not decide this question based on such scant argumentation”). 15 Moreover, as Judge Katzmann’s concurrence, in which I fully join, cogently notes, today’s 16 holding simply ignores the context animating § 601’s enactment and further upends 1 The majority claims to be perplexed by my concern that today’s decision reaches a question it need not, particularly because, the majority reasons, all judges – including myself – who agreed to hear these cases en banc joined an order instructing “the parties to address the BIA’s interpretation of § 601 as it related to both spouses and non-married partners.” Maj. Op. at 13 n.6. My concern, however, is not a quibble over the semantics of the en banc order, but rather the majority’s unnecessary but apparently pressing need to decide a question which the facts of petitioners’ appeals simply do not present. Like Judge Katzmann, I engage the question the majority answers because I would be remiss in not voicing my profound disagreement with the majority’s conclusions. 55
    324. 1 congressional intent by reading the statute too narrowly. Section 601 was, after all, expressly 2 enacted to expand, not contract, the availability of asylum under § 1101(a)(42) in the context of 3 coercive population control programs. 4 I will not reiterate what Judge Katzmann has already ably stated. I write separately to 5 highlight the potentially ill-considered breadth of the majority opinion, which appears to cast 6 doubt on our own circuit’s caselaw, as well as to create further circuit conflicts when such 7 outcomes are easily avoided. Most importantly, however, I write because the majority’s zeal in 8 reaching a question not before us requires the unprecedented step of constricting the BIA’s 9 congressionally delegated powers – a decision whose ramifications we are ill-prepared, given the 10 procedural posture of this case, to understand or appreciate fully. 11 The majority analyzes § 601 within the broader framework of the INA and concludes that 12 “[n]othing in the general definition of refugee would permit ‘any person’ who has not personally 13 experienced persecution or a well-founded fear of future persecution . . . to obtain asylum.” Maj. 14 Op. at 17. Again, I agree fully with Judge Katzmann’s discussion of the fatal flaws in this 15 analysis, and I seek only to emphasize the majority’s apparent failure to appreciate that this 16 deceptively simple proposition may unduly and inappropriately limit the BIA not merely in cases 17 under § 601 but in others as well. 18 In coming to its conclusion, the majority endorses the view that “persecution” can only 19 be direct and personal, by which it appears to mean that the granting of asylum can never be 20 based on, in whole or in part, harm to others, no matter how closely related the harm or the 21 person harmed is to the applicant or whether harm to another is directed in whole or in part 56
    325. 1 toward the applicant.2 The majority tries to anchor this limiting principle to the text of the 2 statute, but such a reading is unwarranted and unsupportable. It is pellucidly clear from the text 3 of § 1101(a)(42) that Congress did not define nor intend to define “persecution” to exclude 4 harms “not personally” suffered by an applicant. The statute instead reads that “any person” 5 who “because of persecution or a well-founded fear of persecution” is “unable or unwilling” to 6 return to his or her country is entitled to asylum. There is no indication whatsoever of how 7 personal or direct the harm or injury must be, only that persecution to an individual can merit 8 asylum protection.3 We should, moreover, eschew the limiting construction of § 601 and 9 § 1101(a)(42) urged in today’s opinion because such a construction could lead to absurd results. 10 United States v. Dauray, 215 F.3d 257, 264 (2d Cir. 2000) (“A statute should be interpreted in a 11 way that avoids absurd results.”). If government officials shot and killed an asylum applicant’s 12 child to force him or her to convert to another religion, would that harm, which the majority 13 would ostensibly label “not personal,” be insufficient in itself to demonstrate persecution of that 14 applicant? Or what if the parent of an adult applicant was kidnapped and tortured to force the 15 applicant to renounce an opposition political party or endorse a government candidate? In the 16 end, I see no unambiguous language in the text of § 1101(a)(42) that compels the limiting 17 construction of the INA that the majority now divines. 2 In another portion of the majority opinion, the Court states “we conclude that the statutory scheme unambiguously dictates that applicants can become candidates for asylum relief only based on persecution that they themselves have suffered or must suffer.” Maj. Op. at 19. By this pronouncement, the opinion suggests that harm to others cannot form a part of the rationale for granting asylum. 3 I further agree with Judge Katzmann’s analysis of the majority’s misplaced reliance on 8 U.S.C. § 1158 to support its conclusion here. 57
    326. 1 Requiring an applicant’s eligibility for asylum to rest only on instances where he or she 2 suffers persecution “personally” merely begs the question of what personal harm is and how to 3 define it. As with any ambiguous statutory term, it is for the BIA to determine within its 4 expertise what exactly constitutes “persecution” so long as its interpretation is reasonable.4 5 Chevron, U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 843-44 (1984). From its 6 decision in In re S-L-L-, 24 I. & N. Dec. 1 (B.I.A. 2006), the BIA clearly construed 7 “persecution” as not only entailing the spouse forced to undergo the procedure but also including 8 the other spouse who, while physically unharmed, was nevertheless also targeted by the 9 government for punishment and persecution. The BIA reached this conclusion by utilizing its 10 traditional tests of nexus and level of harm, id. at 5, that is, by examining how the procedure 11 affected each spouse’s respective health and emotional well-being as well as the couple’s interest 12 in procreation and child-rearing. Perhaps most importantly of all, the BIA also considered to 13 whom the government’s actions were directed. 14 It is this last factor that is crucial. The majority concedes that both spouses suffer a 15 “profound emotional loss” as a result of a forced abortion or sterilization, but it never sufficiently 16 explains why the harm of sterilization or abortion constitutes persecution only for the person 17 who is forced to undergo such a procedure and not for that person’s spouse as well. Maj. Op. at 18 21. This conclusion rests on two principal conclusions, both equally flawed. First, the majority 4 I note that this and other circuits have found “persecution” to be an ambiguous term in other asylum cases. See, e.g., Mirzoyan v. Gonzales, 457 F.3d 217, 220 (2d Cir. 2006) (per curiam) (finding that the INA does not “unambiguously explain[] what the word ‘persecution’ means” in the economic context); Corado v. Ashcroft, 384 F.3d 945, 947 (8th Cir. 2004) (per curiam) (“The BIA is entitled to deference in interpreting ambiguous statutory terms such as ‘persecution.’”); cf. Singh v. INS, 134 F.3d 962, 967 (9th Cir. 1998) (noting that the INA “does not define ‘persecution’ or specify what acts constitute persecution”). 58
    327. 1 clings to the notion that the persecution suffered is physically visited upon only one spouse, but 2 this simply ignores the question of whom exactly the government was seeking to persecute when 3 it acted. On this point, the harm is clearly directed at the couple who dared to continue an 4 unauthorized pregnancy in hopes of enlarging the family unit. Indeed, the majority’s conclusion 5 disregards the immutable fact that a desired pregnancy in a country with a coercive population 6 control program necessarily requires both spouses to occur, and that the state’s interference with 7 this fundamental right “may have subtle, farreaching and devastating effects” for both husband 8 and wife. See Skinner v. Oklahoma ex rel. Williamson, 316 U.S. 535, 541 (1942). The 9 termination of a wanted pregnancy under a coercive population control program can only be 10 devastating to any couple, akin, no doubt, to the killing of a child. Similarly, as to sterilization, 11 the Ninth Circuit has aptly observed that: 12 In addition to the physical and psychological trauma that is common to many 13 forms of persecution, sterilization involves drastic and emotionally painful 14 consequences that are unending: The couple is forever denied a pro-creative life 15 together. As the BIA explained, 16 17 The act of forced sterilization should not be viewed as a discrete onetime 18 act, comparable to a term in prison, or an incident of severe beating or 19 even torture. Coerced sterilization is better viewed as a permanent and 20 continuing act of persecution that has deprived a couple of the natural 21 fruits of conjugal life, and the society and comfort of the child or children 22 that might eventually have been born to them. 23 24 Qili Qu v. Gonzales, 399 F.3d 1195, 1202 (9th Cir. 2005) (quoting In re Y-T-L-, 23 I. & N. Dec. 25 601, 607 (B.I.A. 2003)). Viewed in this light, the harm here is directed as much at the husband 26 as at the wife. By its action, after all, the state is preventing both members of the couple from 27 procreating as a unit, and as the BIA found in In re Y-T-L-, such harm is not limited in time to 59
    328. 1 the moment of sterilization, but is an ongoing harm that affects a married couple indefinitely. 23 2 I. & N. Dec. at 607. As the Third Circuit recently observed in Sun Wen Chen v. Attorney 3 General of the United States, --- F.3d ---, 2007 WL 1760658, at *5 (3d Cir. June 20, 2007), the 4 “persecution of one spouse can be one of the most potent and cruel ways of hurting the other 5 spouse.”5 In the end, I fail to understand how the majority can claim that the harm caused by a 6 spouse’s forced abortion or sterilization is not a personal harm to both spouses – either or both of 7 whom can be sterilized for violations of the population control programs – especially given the 8 unique biological nature of pregnancy and special reverence every civilization has accorded to 9 child-rearing and parenthood in marriage. I similarly fail to understand how the majority 10 justifies limiting the BIA’s ability to take this special and egregious harm into consideration and 11 to determine within its expertise that such acts constitute persecution against both a wife and 12 husband.6 5 The majority incorrectly asserts that the Third Circuit’s analysis in Sun Wen Chen is incongruent with my own analysis because the court there held that § 601(a) contains an ambiguity. Maj. Op. at 16 n.7. The Third Circuit found that § 601(a) is ambivalent on the treatment of spouses and that the existence of derivative asylum was not “intended to foreclose additional pathways to asylum specific to spouses.” Sun Wen Chen, --- F.3d ---, 2007 WL 1760658, at *4 (emphasis added); see also id. (“[W]e see nothing in the statute evincing Congressional intent to establish a particular policy regarding spousal eligibility.”). The Third Circuit then reasoned that because § 601(a) “establishes that forced abortion and sterilization constitute persecution,” id., it is entirely within the BIA’s authority to “interpret[] the scope of that persecution,” id., including its applicability to spouses. It is my contention that we should defer to the BIA’s interpretation regarding the broader scope of persecution under § 1101(a)(42) and not impose, as the majority here does, an unfounded requirement that persecution be direct and personal and that harm to another, even if directed at the applicant, is never sufficient for the purposes of § 1101(a)(42). These analyses are not incongruent because they both center on the deference we owe to the BIA on defining persecution. 6 The majority notes that its decision corrects the “perverse effect of creating incentives for husbands to leave their wives” inherent in the BIA’s determination in In re S-L-L-, 24 I. & N. Dec. 1 (B.I.A. 2006). Maj. Op. at 27. I note first that one of the petitioners before us was not, in 60
    329. 1 Second, the majority argues that the BIA has impermissibly created an irrebuttable 2 presumption that relieves applicants from the statutory burden of proving that they have a well- 3 founded fear of persecution based on an impermissible nexus. As Judge Katzmann correctly 4 observes, however, the presumption argument is merely a red herring. In enacting § 601, 5 Congress was not creating a presumption but merely expanding the BIA’s unduly restrictive 6 definition in Matter of Chang of the terms “persecution” and “political opinion.” Thus, § 601 7 defined “persecution” and “political opinion” to include an individual’s forced abortion or 8 sterilization under a restrictive population control policy. No presumption was created, however, 9 as the applicant still bears the burden of establishing that he or she was subject to the conduct 10 that qualifies under this expanded definition of persecution. And, while the majority places great 11 weight on the fact that § 601 supposedly creates a presumption that “relieved only persons who 12 actually experienced a forcible abortion or sterilization from the burden of proving a political 13 nexus,” and not those married to such persons, Maj. Op. at 21, his conclusion ignores the clear 14 congressional intent of § 601, expressed in the legislative history, that 15 [n]othing in [§ 601] is intended to lower the evidentiary burden of proof for any 16 alien, no matter how serious the nature of the claim. The Committee emphasizes 17 that the burden of proof remains on the applicant, as in every other case, to 18 establish by credible evidence that he or she has been subject to persecution-in fact, attempting to flee alone; he and his fiancée planned to leave China together but she was unfortunately caught before she could escape. More importantly, however, the majority’s assertion here is based on nothing but speculation as to the decisionmaking in which couples, persecuted by coercive population control programs, must engage before attempting to flee. We simply have no foundation on which to conclude that all couples have the financial resources to escape at the same time, and as the government stated at oral argument, it is not uncommon for Chinese couples to separate and have one spouse go abroad in order to amass the necessary resources to bring over the other spouse. I believe the majority here is opining on a subject – imbued with potentially significant cultural differences – with which it has no expertise or empirical evidence. 61
    330. 1 this case, to coercive abortion or sterilization-or has a well-founded fear of such 2 treatment. 3 4 H.R. Rep. No. 104-469(I), 1996 WL 168955, at *174 (1996). Quite simply, there is no way to 5 read in § 601 the presumption the majority discusses, nor a limitation on the BIA in extending 6 § 601’s reach to spouses. Either the persecution occurred or it did not, and the applicant retains 7 the burden of proving such circumstances. The BIA accordingly did not err in interpreting 8 ambiguous terms in the INA to determine that either spouse may qualify as a refugee where one 9 of them has in fact undergone forced abortion or sterilization. 10 Finally, if adopted, the majority’s limiting construction may have significant, unintended 11 consequences, broader than the Court today acknowledges. By claiming categorically that an 12 applicant cannot rely on the harm inflicted on others in § 601 cases – no matter how closely 13 related the harm or the person harmed is to the applicant or whether the harm is directed in 14 whole or in part towards the applicant – to establish persecution or entitlement to asylum, this 15 Court suggests that the BIA is precluded from ever considering harm to others as evidence of 16 persecution to the applicant. While I agree that there are certainly limits as to when harm to 17 another may inform persecution or a well-founded fear of persecution of an applicant, I cannot 18 endorse the majority’s apparent per se conclusion. As noted above, this is a question of statutory 19 construction properly answered by the BIA, which, in its administrative expertise, may interpret 20 the ambiguity inherent in “persecution” to determine when a harm is direct or personal enough to 21 be considered persecution of an applicant. And, in certain limited contexts pertinent to this 22 appeal, the BIA has done precisely this, examining the harm to family members in determining 23 whether an asylum applicant has in fact suffered past persecution, particularly where an 62
    331. 1 immediate family member has been subjected to significant and enduring mistreatment. In 2 Matter of Chen, 20 I. & N. Dec. 16, 19-21 (B.I.A. 1989), the seminal BIA decision recognizing 3 the availability of humanitarian asylum for victims of severe past persecution, the BIA recited a 4 litany of horrific acts visited on and suffered by Chen’s parents during the Cultural Revolution, 5 and such evidence – when coupled with the Chinese government’s treatment of the petitioner 6 himself – supported the BIA’s conclusion that “the respondent has clearly established that he and 7 his family were severely persecuted in the past in China.” Id. at 21. Similarly, in In re H-, 21 I. 8 & N. Dec. 337, 345 (B.I.A. 1996), where an applicant had testified about the severe physical 9 beatings he had suffered as the member of a subclan in Somalia, the BIA based its finding of past 10 persecution in part on the applicant’s testimony that his father and brother, also members of the 11 same subclan, were beaten and killed. In examining the allegations concerning the deaths of his 12 father and brother, the BIA specifically noted that “evidence of treatment of persons similarly 13 situated is persuasive of an applicant’s claim of political persecution.” Id. at 345 (quoting Matter 14 of Mogharrabi, 19 I. & N. Dec. 439 (B.I.A. 1987)); see also In re N-M-A-, 22 I. & N. 312, 326 15 (B.I.A. 1998) (finding in the context of humanitarian asylum that the applicant had suffered past 16 persecution in part because of “the disappearance and likely death of his father”). The BIA has 17 thus identified specific situations in which the harm to close family members could be central to 18 the finding of persecution and the granting of refugee status. The majority’s misguided exercise 19 in statutory interpretation, however, undermines this agency determination and suggests that 20 because the years-long harassment and egregious mistreatment of Chen’s parents or the deaths of 63
    332. 1 H-’s brother and father were “not personally experienced” by the applicants, neither Chen nor H- 2 could base their asylum applications on such harm after today’s decision.7 3 The holding today also calls into question our own caselaw – as well that of other circuits 4 – in which appellate panels have recognized that harm inflicted upon one individual may give 5 rise to, or at least help establish, persecution of another in certain circumstances. In Jorge-Tzoc 6 v. Gonzales, 435 F.3d 146 (2d Cir. 2006) (per curiam), this Court, acknowledging that petitioner 7 had not been “victimized directly” when as a young boy, his sister and her family were killed for 8 their political activities, nevertheless remanded the case to the BIA for further proceedings to 9 determine whether his age, coupled with the harm to his family members, helped to establish 10 past persecution.8 Id. at 150 (internal quotation marks omitted). While the decision rested in 11 great part on the petitioner’s age, this decision illustrates another category of asylum cases where 12 it might be appropriate to consider harm to others in determining past persecution.9 Similarly, 7 The regulations governing the claims under the Convention Against Torture explicitly recognize that torture encompasses not only physical harm to the individual but also “mental pain or suffering” that results from the threat of infliction of physical pain or suffering on another person. See 8 C.F.R. § 1208.18(a)(4)(iv). By analogy, this regulation further supports the position that persecution is not limited to direct and physical harm upon an individual but can encompass harm inflicted on others as well. See, e.g., Yan Chen v. Gonzales, 417 F.3d 268, 275 (2d Cir. 2005) (“Certainly . . . torture can constitute persecution . . . .”). 8 The passing statement in a footnote in Melgar de Torres v. Reno, 191 F.3d 307, 313 n.2 (2d Cir. 1999), that the death of the applicant’s uncle did not constitute political persecution of her is not to the contrary. As noted in Jorge-Tzoc, the petitioner in Melgar de Torres “was an adult who offered no objective evidence that her uncle’s killing was politically motivated.” 435 F.3d at 150. To the extent that Melgar de Torres suggested, furthermore, that even if the petitioner had established the link between her uncle’s killing and his political activities, such killing could then not be considered part of her past persecution, this suggestion was clearly dicta. 9 The majority professes no opinion on the continued vitality of our holding in Jorge-Tzoc and claims that nothing in today’s decision “preclude[s] the BIA from considering the totality of circumstances in any particular case to determine if an asylum applicant has carried his statutory 64
    333. 1 other circuits have confronted situations where they found persecution relying in whole or in part 2 on harm to others in certain circumstances. In Sun Wen Chen, the Third Circuit upheld In re S- 3 L-L-, the very BIA determination the majority strikes down today, in part by acknowledging that 4 the physical harm to one’s spouse is nevertheless harm to both spouses in the family planning 5 context. --- F.3d ---, 2007 WL 1760658, at *5 (“In a great many cases, forced abortion or 6 involuntary sterilization of one spouse will directly affect the reproductive opportunities of the 7 other spouse . . . . And persecution of one spouse can be one of the most potent and cruel ways of 8 hurting the other spouse . . . .”). The Sixth Circuit in Abay v. Ashcroft, 368 F.3d 634, 642 (6th 9 Cir. 2004), determined that an applicant was entitled to asylum because she had fled Ethiopia 10 with her teenage daughter to protect the teenager from undergoing forced genital mutilation. 11 The Abay court specifically noted that derivative asylum under § 1158 was not available to Abay 12 – as she was neither a spouse nor a child of a persecuted individual – but granted her asylum, 13 observing that several oral IJ and BIA decisions “suggest a governing principle in favor of 14 refugee status in cases where a parent and protector is faced with exposing her child to the clear 15 risk of being subjected against her will to a practice that is a form of physical torture causing 16 grave and permanent harm.” Id. at 642. 17 Having carefully weighed the law and arguments presented in this appeal, I must concur 18 in the judgment for the reasons already stated ably by Judge Katzmann. I agree in particular burden.” Maj. Op. at 26 n.13. Yet, the majority fails to explain why the totality of circumstances may not be applied in the context of married couples who suffer under coercive population control programs. If the BIA could consider the direct harm to Jorge-Tzoc’s family members in determining whether Jorge-Tzoc himself had been persecuted, it should be able to consider the targeting of and effect on an individual when his or her spouse is forced to undergo an abortion or sterilization. 65
    334. 1 with Judge Katzmann that the Third Circuit in Cai Luan Chen v. Ashcroft, 381 F.3d 221 (3d Cir. 2 2004) (Alito, J.), did what we should have done here. In that case, then-Judge Alito found no 3 need to reach the question of whether § 601 or § 1101(a)(42) were ambiguous because Chen, 4 who was not married to his fiancée on whom the forced abortion was performed, could prevail 5 only if the BIA’s distinction between married and unmarried couples was unreasonable.10 Id. at 6 227. Judge Alito ultimately ruled that the distinction was reasonable and denied the petition. Id. 7 at 235. This analysis should control our own very similar cases here.11 Given the above, the majority should never have reached the question it has taken upon itself to resolve, particularly in the immigration context where the Supreme Court has long recognized “that judicial deference to the Executive Branch is especially appropriate . . . where officials ‘exercise especially sensitive political functions that implicate questions of foreign relations.’” INS v. Aguirre-Aguirre, 526 U.S. 415, 425 (1999) (quoting INS v. Abudu, 485 U.S. 94, 110 (1988)). In reaching this question, the majority has, I fear, started a domino effect that may have significant and unforeseen repercussions. Finally, the majority, in noting that “[i]f this conclusion is inconsistent with Congress’s intentions, [Congress] can, if it so chooses, of course, amend the statute,” Maj. Op. at 22 n.10, seems to take comfort that its conclusion, if wrong, may 10 While I ultimately agree with Third Circuit’s recent pronouncement in Sun Wen Chen that the BIA properly interpreted an ambiguity in § 601 and § 1101(a)(42) to extend asylum protection to spouses of individuals forced to abort pregnancies or undergo sterilization, Sun Wen Chen involves a question we need not reach here because the petitioner here is not married. Had the majority dealt only with the question presented, I would adopt the approach as outlined by then-Judge Alito in Cai Luan Chen. 11 Unlike my esteemed colleague Judge Calabresi, I do not find that the BIA limited its analysis to § 601 in In re S-L-L-, but rather was grounding its interpretation in both the specific language of § 601 and the more general definition of “refugee” in § 1101(a)(42). Because the statute is ambiguous, I would defer to the BIA’s interpretation here. 66
    335. be simply overturned. But for those petitioners who fled a draconian population control program because their spouses had been forced to undergo an abortion or sterilization, the majority’s caveat must be cold comfort indeed. 67
    336. CALABRESI, Circuit Judge, concurring in part and dissenting in part: What is remarkable about this case is that essentially everyone on this court agrees that IIRIRA § 601(a), codified at 8 U.S.C. § 1101(a)(42)(B), cannot be read to grant per se asylum to spouses. This is remarkable because it is in direct conflict with every other circuit, the BIA, and ten years of rulings. See Maj. Op. at 5 & n.4. Yet we all agree. And we are correct. Moreover, it follows that, because § 601(a) does not grant that kind of asylum to spouses, it also cannot be read as granting asylum to non-spouses — like the petitioners in this case. This part of the majority’s analysis is admirable, and I join it. Because § 601(a), in unambiguous terms, grants per se “refugee” status only to the “person who has been forced to abort a pregnancy or to undergo involuntary sterilization,” the panel in this case — consisting of the author of the majority opinion, the principal concurrence, and me — sent the case back to the BIA, to allow the agency to reexamine its decision in In re C-Y-Z- (C-Y-Z-), 21 I. & N. Dec. 915 (B.I.A. 1997) (en banc) (construing § 601 as granting per se “refugee” status to spouses). The BIA stuck to its decision, see In re S-L-L- (S-L-L-), 24 I. & N. Dec. 1 (B.I.A. 2006) (en banc), but has not convinced us that C-Y-Z-’s rule can be squared with the plain text of § 601(a), and so we now appropriately say that the BIA was wrong. Unfortunately, both the majority and concurrences are not willing to stop with that, which was the issue clearly before us and fully considered by the BIA. For reasons that are quite understandable, but nonetheless wrong — both in terms of results and in terms of what the Supreme Court has said about our relationship to the BIA — the majority and the concurrences 68
    337. go further. They do so in different directions, and that fact is, to me, simply additional evidence that going further was inappropriate. I The majority says that if the BIA were to construe the general definition of “refugee” found in 8 U.S.C. § 1101(a)(42)(A) as granting per se refugee status to certain categories of people — e.g., spouses or non-spouses — that would be an impermissible reading of § 1101(a)(42)(A). This seems to me to be mistaken on several counts. A First, the majority relies on the fact that § 601(a) was enacted to overturn Matter of Chang, 20 I. & N. Dec. 38, 44 (B.I.A. 1989) (adopting the rule that victims of a coercive family planning regime could claim refugee status only if the victims demonstrated that the family- planning policy had been “selectively applied” to them on the basis of a protected ground). See Maj. Op. at 17. The majority posits that § 601(a) was not intended to do more than overturn Chang, and that Chang is therefore left in place as to spouses and partners who are not themselves forcibly aborted or involuntarily sterilized. See Maj. Op. at 18 (“The inclusion of some obviously results in the exclusion of others.”). And, under Chang, spouses and partners are not entitled to per se refugee status. Fair enough, but Chang is not an opinion of the Supreme Court, or even of a Court of Appeals; it is an interpretation of underlying statutory law by the BIA. As such, the agency is perfectly free to change it — so long as the change is not inconsistent with the underlying law. 69
    338. Thus, any suggestion that the BIA could not, because of Chang, now grant per se status to spouses pursuant to § 1101(a)(42)(A) is a non sequitur, plain and simple. B Second, the logical consequences of what the majority seems to be saying appear to me to be untenable. Suppose the BIA were to issue an interpretation of § 1101(a)(42)(A) that said, categorically, that any child who sees his parents tortured and murdered before him by a totalitarian government — say, the Nazis — is persecuted, and therefore eligible for asylum. Would such a ruling be invalid under § 1101(a)(42)(A)’s broad definition of refugee? If that is what the majority is saying, it is, in my judgment, manifestly absurd. There is nothing in the language or history of § 1101(a)(42)(A) that suggests the BIA could not adopt such a per se rule. But, if the BIA could adopt the kind of per se rule I described above — and I believe a majority of our court would agree with me that such a rule would indeed be proper — then it is improperly premature to say — as today’s governing opinion does — that the agency could not adopt an analogous per se rule with respect to individuals in the situation of the petitioners in this case. It may be that if the BIA did adopt such a per se rule, I would ultimately agree with the majority that, in the context of coercive family planning laws, such an interpretation of § 1101(a)(42)(A) is “unreasonable” at Chevron’s Step Two.1 But once it is admitted that some categorical per se asylum rules — like the one involving my hypothetical children — might be valid under § 1101(a)(42)(A) (i.e., would get by Chevron Step One) — it is, I believe, impermissible to say that an equivalent per se interpretation dealing with spouses would necessarily be invalid if it were adopted — which is in effect what the majority’s holding 1 See Chevron U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837 (1984). 70
    339. amounts to. It is impermissible given the Supreme Court’s unanimous decisions in INS v. Orlando Ventura, 537 U.S. 12 (2002) (per curiam), and Gonzales v. Thomas, 547 U.S. 183, 126 S. Ct. 1613 (2006) (per curiam). 1 In Ventura, the Supreme Court held that, “[g]enerally speaking, a court of appeals should remand a case to an agency for decision of a matter that statutes place primarily in agency hands.” 537 U.S. at 16. “This principle,” the Court explained, “has obvious importance in the immigration context,” id. at 16-17, because “[w]ithin broad limits the law entrusts the agency to make the basic asylum eligibility decision here in question.” Id. at 16. Subsequent case law has only strengthened Ventura’s reasoning. The “ordinary remand rule” was recently reaffirmed by a unanimous Supreme Court in Thomas, 126 S. Ct. 1613, and has been followed by our court in a series of cases, most notably in Ucelo-Gomez v. Gonzales, 464 F.3d 163, 168-70 (2d Cir. 2006). Yet, despite the “obvious importance” of the ordinary remand rule in the immigration context, the majority insists on precluding the BIA from interpreting § 1101(a)(42)(A)’s general provisions in the first instance. In my view, this aspect of the majority’s holding is dangerously in tension with Ventura’s command. In Ventura — much as in the case before us — the Ninth Circuit reversed a holding of the BIA, and then “went on to consider an alternative argument that the Government had made before the Immigration Judge,” but which “the BIA itself had not considered . . . .” 537 U.S. at 13. Specifically, the Ninth Circuit reversed the BIA’s holding that the petitioner was not persecuted “on account of” a “political opinion,” but then, rather than remanding to the BIA for 71
    340. further proceedings, the court evaluated for itself, and rejected, the government’s alternative argument that the petitioner failed to qualify for asylum because of changed country conditions in Guatemala. Id. In reversing the Ninth Circuit’s judgment, the Supreme Court found that the court of appeals seriously disregarded the agency’s legally-mandated role. Instead, it independently created potentially far-reaching legal precedent about . . . a highly complex and sensitive matter. And it did so without giving the BIA the opportunity to address the matter in the first instance in light of its own expertise. Id. at 17. More recently, the Court in Thomas reversed a Ninth Circuit decision which had decided, without first remanding the issue to the BIA, “that in principle ‘a family may constitute a social group for the purposes of the refugee statutes,’ . . . [and] that the particular family at issue . . . fell within the scope of the statutory term ‘particular social group.’” 126 S. Ct. at 1614 (quoting Thomas v. Gonzales, 409 F.3d 1177, 1187, 1189 (9th Cir. 2005) (en banc)) (emphasis added). Quoting Ventura — and echoing the basic principle of SEC v. Chenery Corp. (Chenery I), 318 U.S. 80 (1943), that “an appellate court cannot intrude upon the domain which Congress has exclusively entrusted to an administrative agency,” id. at 88 — the Thomas Court reiterated that “[a] court of appeals is not generally empowered to conduct a de novo inquiry into the matter being reviewed and to reach its own conclusions based on such an inquiry.” Thomas, 126 S. Ct. at 1615 (quoting Ventura, 537 U.S. at 16 (internal quotation marks omitted)). 72
    341. In Ucelo-Gomez, a panel of this court concluded that Thomas and Ventura establish the rule that “where (as here) the agency has yet to decide whether a group, a thing, or a situation falls within the ambit of a statutory term, the proper course is for the reviewing court to remand the matter to the agency in accordance with the well-worn ordinary remand rule.” Ucelo-Gomez, 464 F.3d at 169 (internal quotation marks omitted). Moreover, the panel in Ucelo-Gomez asserted that “the agency interpretation required by Thomas and Ventura is ‘in the first instance’ a particularized interpretation by the agency.” Id. (emphasis omitted). As a purely formal matter, the approach taken by the majority today is perhaps reconcilable with Ventura and Thomas. But it is fundamentally incompatible with the spirit of those cases. Even if the majority is convinced that C-Y-Z-’s rule would be an unreasonable construction of § 1101(a)(42)(A), the correct approach would be to allow the agency to make a determination on that matter first. Instead, the majority opinion — perhaps realizing that it could not, at this time, authoritatively speak on the question of C-Y-Z-’s reasonableness as a construction of § 1101(a)(42)(A) — by a preemptive strike strips the BIA of its capacity to consider the issue under § 1101(a)(42)(A). In so doing, the majority precludes the BIA from examining thoroughly this “highly complex and sensitive matter,” Ventura, 537 U.S. at 17, and “independently create[s] . . . far-reaching legal precedent . . . . without giving the BIA the opportunity to address the matter in the first instance in light of its own expertise.” Id. Significantly, Ventura and Thomas are designed to prevent just such judicial preemption of BIA positions, even when that preemption reaches what is arguably the correct result. 2 73
    342. Moreover, even if the majority were not required — as I believe it was — to remand Zhen Hua Dong’s case to the BIA,2 it should have remanded his case as a matter of wise discretion. Cf. Jian Hui Shao v. Bd. of Immigration Appeals, 465 F.3d 497, 501, 503 (2d Cir. 2006) (concluding that “the BIA is better situated than we are to decide the statutory interpretation question in the first instance,” and noting that “[o]ur decision to remand this question of law to the BIA for resolution in the first instance is supported by recent decisions of the Supreme Court of the United States and our Court” (emphases added)); Yuanliang Liu v. U.S. Dep’t of Justice, 455 F.3d 106, 116 (2d Cir. 2006) (“Because we conclude, as a matter of discretion, that it is prudent and useful for us to remand the issue of frviolousness, we need not address the more complicated question of when remands to the BIA are required by elementary principles of administrative law.”). 2 The question of whether, as a matter of Chevron Step Two “reasonableness” review, the BIA could base its C-Y-Z- decision on § 1101(a)(42)(A), is arguably neither a pure question of fact, nor of statutory interpretation. And the extent to which such mixed questions may be resolved by a Court of Appeals, without first remanding to the agency for its consideration, has not been clearly settled by the Supreme Court. Compare Thomas, 126 S. Ct. at 1615 (“[T]he proper course, except in rare circumstances, is to remand to the agency for additional investigation or explanation.” (quoting Ventura, 537 U.S. at 16) (internal quotation marks omitted)) with id. (requiring remand, and observing that “[t]he matter requires determining the facts and deciding whether the facts as found fall within a statutory term”); Hussain v. Gonzales, 477 F.3d 153, 157-58 (4th Cir. 2007) (distinguishing between factual issues not considered by the BIA, and statutory issues, and reasoning that Ventura and Thomas were directed only toward factual issues); Fernandez-Ruiz v. Gonzales, 466 F.3d 1121, 1132-35 (9th Cir. 2006) (considering a variety of factors — including the fact that the issue being decided would, following a remand to the BIA, have been reviewed by the court de novo anyway — in concluding that the Thomas-Ventura remand rule did not apply to the particular issue in question); Ucelo-Gomez, 464 F.3d at 170 (“[I]f a reviewing court can state with assured confidence (absent agency guidance as to its protectability under the INA) that a group would or would not under any reasonable scenario qualify as a ‘particular social group,’ it need not remand, and may rule on the issue in the first instance.”). 74
    343. I believe that the majority’s haste in narrowly construing § 1101(a)(42)(A) — and, therefore, in cabining the BIA’s discretion — is particularly troubling given the circumstances of this case. At stake is a rule of the BIA that has been in place, and relied upon, for over ten years. See S-L-L-, 24 I. & N. Dec. at 14 (Board Member Pauley, concurring) (“[N]otwithstanding my belief that Matter of C-Y-Z- . . . was wrongly decided, I would not overrule it now, nearly a decade later and in the aftermath of thousands of decisions applying it to grant asylum on a derivative basis.”). In addition, the invalidation of C-Y-Z-’s rule will have sweeping ramifications for this court’s immigration law docket; by one estimate, “70-80 percent of the [petitioners in our court] are Chinese seeking asylum to escape their homeland’s family planning policies.” BIA Appeals Remain High in 2nd and 9th Circuits, The Third Branch: Newsletter of the Fed. Cts. (Admin. Office of the U.S. Cts. Office of Pub. Affairs, D.C.), Feb. 2005, available at http://www.uscourts.gov/ttb/feb05ttb/bia/index.html (citing statement of Elizabeth Cronin). Given all this, our court should have approached the question of C-Y-Z-’s permissibility — either as it was or with nuanced modifications — not with haste, but with trepidation. For the truth is that we cannot foretell how the BIA would have interpreted the general definition of § 1101(a)(42)(A), had it been asked to focus on that language. By trying to decide something that is not yet before us, the majority bars the BIA from bringing its expertise to bear on this sensitive issue. In the process, the majority does not only preclude the BIA from reenacting the per se rule of C-Y-Z- — a rule which, at the proper time, I might well have concluded was “unreasonable” at Chevron Step Two, for all of the reasons the majority recites. The majority also prevents the agency from interpreting the general language 75
    344. of § 1101(a)(42)(A) in ways which might have suffered from none of the problems the majority properly associates with the current per se rule — ways, incidentally, which might truly have promoted congressional policy goals.3 In this respect, the majority opinion keeps the agency from doing what administrative agencies do best, namely, using their expertise to convert general statutes into specific rules that best reflect an underlying legislative intent.4 **** 3 To cite just one of the many possibilities which the majority prematurely forecloses: had the BIA not relied on § 601(a)’s automatic persecution rule, but instead focused on the general notion of “persecution,” the agency might have interpreted § 1101(a)(42)(A) as providing (1) that partners who had tried to marry, and were prevented from doing so, but who stayed together, are jointly eligible for asylum (which conclusion would both (a) promote the congressional policy of keeping families together, and (b) extend asylum eligibility to individuals not already covered by § 601(a)); but (2) that husbands who are legally married at the time of a wife’s forced abortion, but who choose to leave their wives behind for good, are not. 4 I am mindful that the Supreme Court has cautioned that respect for the role and expertise of agencies does not “require that we convert judicial review of agency action into a ping-pong game,” and that, therefore, remand is not required when it “would be an idle and useless formality.” NLRB v. Wyman-Gordon Co., 394 U.S. 759, 766 n.6 (1969); see also Li Zu Guan v. INS, 453 F.3d 129, 135-38 (2d Cir. 2006) (discussing futility standards); Alam v. Gonzales, 438 F.3d 184, 187-88 (2d Cir. 2006) (per curiam) (same). Moreover, and relatedly, the Supreme Court has clarified that a reviewing court must “uphold a decision of less than ideal clarity if the agency’s path may reasonably be discerned.” Bowman Transp., Inc. v. Arkansas- Best Freight Sys., Inc., 419 U.S. 281, 286 (1974); see also Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (same). But, regardless of whether these cases, which limit the necessity to remand, are understood to be “exceptions” to the Chenery and Ventura-Thomas requirements, or merely a reflection of the deeper truth that formulaic statements cannot substitute for sound judgment in particular cases, see Li Hua Lin v. U.S. Dep’t of Justice, 453 F.3d 99, 112 (2d Cir. 2006), it remains clear on which side of the line the case before us falls. We simply do not know — and, because the majority and concurring opinions make it almost impossible for the BIA to consider the general language of § 1101(a)(42)(A) in the first instance, in relation to spouses and partners of directly victimized persons, we are not likely to learn — how the BIA would have interpreted § 1101(a)(42)(A) had it been asked to do so. This is not a case in which the agency’s path, while not perfectly clear, can “reasonably be discerned”; nor is it a case in which the agency’s likely response to a remand can be predicted with confidence. Rather, it is a case in which (1) the BIA has not yet spoken — at all, and certainly not clearly — on § 1101(a)(42)(A)’s breadth in this area, and (2) our court has, unfortunately, chosen to make further inquiry impossible. 76
    345. Ironically, it was precisely because of the above reasons that the panel in Shi Liang Lin — comprised, as I mentioned earlier, of the author of the majority opinion, the principal concurrence, and me — sent it back originally. Yet if the majority’s reasoning were valid, then there would have been no reason for the panel to do so. Nonetheless, we sent it back then, and the BIA ruled only as to whether per se refugee status could be granted directly under § 601. See infra Part II.B. If the case were sent back again, to allow the agency to consider whether to extend per se protection under § 1101(a)(42)(A), it is possible that the BIA would have agreed with the majority that no such protection should be adopted. Or the BIA might have adopted a more sensible rule. Under the majority’s approach, we will never know. Accordingly, I respectfully, partially, dissent from the majority opinion. II But I cannot join the concurrences either. They act as if the BIA, because it mentioned “nexus” in passing, made a ruling under § 1101(a)(42)(A). It didn’t. Since the agency has yet to interpret the broad language of that section, it is wrong for us to say — as the concurrers do — that the agency expressed views to which we owe deference. And this is so, regardless of whether such a ruling, had it been made, would have passed the requirements of Chevron Step Two. A In SEC v. Chenery Corp. (Chenery I), 318 U.S. 80 (1943), and SEC v. Chenery Corp. (Chenery II), 332 U.S. 194 (1947), the Supreme Court articulated, and then reaffirmed, “a simple but fundamental rule of administrative law”: “[A] reviewing court, in dealing with a 77
    346. determination or judgment which an administrative agency alone is authorized to make, must judge the propriety of such action solely by the grounds invoked by the agency.” Chenery II, 332 U.S. at 196. And “[i]f those grounds are inadequate or improper, the court is powerless to affirm the administrative action by substituting what it considers to be a more adequate or proper basis.” Id. The reason for this rule is obvious: “If an order is valid only as a determination of policy or judgment which the agency alone is authorized to make and which it has not made, a judicial judgment cannot be made to do service for an administrative judgment,” because “an appellate court cannot intrude upon the domain which Congress has exclusively entrusted to an administrative agency.” Chenery I, 318 U.S. at 88. The Chenery decisions also recognized “an important corollary of the foregoing rule”: “If the administrative action is to be tested by the basis upon which it purports to rest, that basis must be set forth with such clarity as to be understandable.” Chenery II, 332 U.S. at 196. As the Court explained, “[i]t will not do for a court to be compelled to guess at the theory underlying the agency’s action; nor can a court be expected to chisel that which must be precise from what the agency has left vague and indecisive.” Id. at 196-97. If it were otherwise, an appellate court could impose its own policy judgments under the guise of “review.” Our court has repeatedly recognized and applied these fundamental rules of administrative law: (1) we may only review that which an agency itself has stated; and (2) the agency must make those statements in clear terms. See, e.g., Riverkeeper, Inc. v. EPA, 475 F.3d 83, 105 (2d Cir. 2007) (“We cannot opine on this subject, because we must consider only those justifications that the [agency] offered at the time of the rulemaking.”); Singh v. U.S. Dep’t of 78
    347. Justice, 461 F.3d 290, 294 n.3 (2d Cir. 2006) (“[W]e cannot, on appeal, substitute an argument — even one the BIA made in another context — for those that the BIA actually gave to support the conclusion . . . dispute[d] on appeal.”); Cao He Lin v. U.S. Dep’t of Justice, 428 F.3d 391, 400 (2d Cir. 2005) (“[W]e will limit our review of the [agency’s] decision to the reasons [it] actually articulates . . . . To assume a hypothetical basis for the [agency’s] determination, even one based in the record, would usurp [the agency’s] role.”); Shi Liang Lin v. U.S. Dep’t of Justice, 416 F.3d 184, 192 (2d Cir. 2005) (“The government suggests that we may simply supply our own rationale for the BIA’s decision in C-Y-Z- and then act accordingly. But the Supreme Court has made clear that ‘[i]t will not do for a court to be compelled to guess at the theory underlying [a particular] agency’s action; nor can a court be expected to chisel that which must be precise from what the agency has left vague and indecisive.’ It is not difficult to understand why. Were courts obliged to create and assess ex-post justifications for inadequately reasoned agency decisions, courts would, in effect, be conscripted into making policy.” (quoting Chenery II, 332 U.S. at 196-97)). B I recite these well-known tenets of administrative law because I believe that they preclude us from taking the route advocated by the concurring opinions. Because the BIA’s opinion in S-L-L- is lacking in clarity, it is certainly possible, with some creativity, to construe the decision as having been based on rationales which the BIA itself did not invoke. But we are not empowered to invoke those reasons. The BIA is required to speak for itself. 1 79
    348. The precise basis of the BIA’s decision in C-Y-Z- was anything but clear. But the BIA and this court have in the past stated that it was based on a construction of § 601(a). See Shi Liang Lin, 416 F.3d at 188 (noting that, in C-Y-Z-, “the BIA held that, under IIRIRA § 601(a), the forced sterilization or abortion of one spouse is an act of persecution against the other spouse . . . .” (emphasis added)); see also id. at 191 (“[A] fresh look at C-Y-Z- reveals that the BIA never adequately explained how or why, in the first instance, it construed IIRIRA § 601(a) to permit spouses of those directly victimized by coercive family planning policies to become eligible for asylum themselves.” (emphasis added)); see S-L-L-, 24 I. & N. Dec. at 3 (“In Matter of C-Y-Z-, supra, we held that a husband whose wife was forcibly sterilized could establish past persecution under this amendment [i.e., IIRIRA § 601(a)]” (emphasis added)). In Shi Liang Lin, the panel remand[ed] the instant petitions to the BIA so that the BIA [could]: (a) more precisely explain its rationale for construing IIRIRA § 601(a) to provide that the ‘forced sterilization of one spouse on account of a ground protected under the Act is an act of persecution against the other spouse’ and that, as a result, the spouses of those directly victimized by coercive family planning policies are per se as eligible for asylum as those directly victimized themselves; and (b) clarify whether, when, and why boyfriends and fiancés may or may not similarly qualify as refugees pursuant to IIRIRA § 601(a). Shi Liang Lin, 416 F.3d at 192 (emphases added). Thus, in remanding, the Shi Liang Lin panel plainly assumed that the BIA’s ruling in C-Y-Z- was based on a construction of § 601(a), and accordingly, requested that the BIA explain how § 601(a) might plausibly be read in such a manner. Consistent with these instructions, the BIA’s response in S-L-L- focused on the scope of § 601(a). See S-L-L-, 24 I. & N. Dec. at 1 (“The United States Court of Appeals for the Second Circuit has remanded this case 80
    349. with a request that we further explain our rationale in Matter of C-Y-Z-, ‘for construing IIRIRA § 601(a) to provide that the “forced sterilization of one spouse on account of a ground protected under the Act is an act of persecution against the other spouse” . . . .’” (internal citation omitted)); id. at 4 (“[W]e reaffirm our holding in Matter of C-Y-Z- . . . .”). Perhaps sensing that § 601(a)’s text had little to offer, however, the BIA’s decision in S-L-L- provided little analysis of that text. Instead, the BIA asserted, conclusorily, that its decision in C-Y-Z- “reflects the significant tensions inherent in the IIRIRA amendment”5 and that “[t]here is no clear or obvious answer to the scope of the protections afforded by the amendment to partners of persons forced to submit to an abortion or sterilization.” S-L-L-, 24 I. & N. Dec. at 4. The BIA did not hint at what these “tensions” might plausibly be, or why the plain language of § 601(a) should not be taken as establishing the proper scope of “the IIRIRA amendment.” Instead, the BIA “reaffirm[ed] [its] holding in Matter of C-Y-Z-,” id., largely on the basis of stare decisis and Congress’s supposed acquiescence. It is the BIA’s reticence to engage with § 601(a)’s text — and the resulting ambiguity in the BIA’s opinion in S-L-L- — that the concurring opinions now seek to convert into an argument that S-L-L- was based, not on § 601(a), but on the general definition of “refugee” found in § 1101(a)(42)(A). It is true that, at one point in S-L-L-, the BIA obscurely remarked that “[a]lthough there is no specific reference in the statutory definition of a refugee to a husband’s claim based on harm inflicted upon his wife, the general principles regarding nexus and level of harm apply in determining such a claim.” S-L-L-, 24 I. & N. Dec. at 5. 5 The “IIRIRA amendment” refers, of course, to § 601(a). 81
    350. But this phrase cannot, I believe, establish — as the concurring opinions would have it — that the BIA’s decision in S-L-L- was based on the general terms “persecution” and “political opinion” found in § 1101(a)(42)(A). And, even if it did, it would not do so clearly (as required by Chenery II). Indeed, one can say, as to that: manifestly not. Notably, in its very next breath, after using the nexus phrase relied on by the concurrers, the BIA in S-L-L- stated that it was applying “general principles requiring nexus and level of harm for past persecution in assessing a claim under the IIRIRA amendment.” Id. (emphasis added). It is permissible to read § 601(a) in this way, the BIA argued, because “[a]lthough the wife is obviously the individual subjected to the abortion procedure, Congress was concerned not only with the offensive assault upon the woman, but also with the obtrusive government interference into a married couple’s decisions regarding children and family.” Id. at 6. Therefore, the BIA concludes, “[w]hen the government intervenes in the private affairs of a married couple to force an abortion or sterilization, it persecutes the married couple as an entity.” Id. (emphasis added). It seems to me patent that the BIA reached this conclusion under § 601(a), and not § 1101(a)(42)(A). 2 In my view, then, the BIA’s decisions in C-Y-Z- and S-L-L- were grounded in a (mistaken) belief that, based on an “entity theory” of persecution, spouses of those directly victimized by coercive family planning policies could themselves become directly eligible for asylum under § 601(a). And it is not enough for the concurring opinions to cast doubt on my conclusion; Chenery II’s “clarity corollary” requires that the agency make clear its decision to 82
    351. rest upon a purported ground. Thus, to restate my problem with the concurring opinions: They would use the fact that the BIA refused to engage clearly with the text of § 601(a) as a basis for concluding that the BIA was relying on something else. But the incompatibility of this approach with Chenery II is apparent. Moreover, the (at best) ambiguousness of the BIA’s decision in S-L-L- results in precisely the problems adverted to in Chenery II. For it is far from clear that, had the BIA focused on the general definition of § 1101(a)(42)(A), the agency would have preserved C-Y-Z- ’s rule in its current form. That is, had the BIA been asked to examine, not § 601(a)’s automatic- eligibility rule, but instead the more general definition of “refugee,” it is quite possible that the BIA would have come up with a different per se rule, and perhaps even one that would have avoided the many problems inherent in its C-Y-Z- approach. See supra at 8-9. Under the concurring opinions’s approach, we are unlikely to know. For, by reading the agency’s opinion as deciding that which it did not decide — and certainly did not decide clearly — the concurring opinions, in effect, preclude the agency from thinking deeply and fully about the matter. And that is the very thing which the clarity requirement of Chenery II is meant to make the agency do. III In the end, as at the beginning, the BIA read us to ask — what we in fact asked: whether C-Y-Z-’s rule could be based upon § 601(a)’s text, and if so, what its reasons were. The agency could, under our remand, have turned more broadly to § 1101(a)(42)(A). It didn’t, and it certainly didn’t do so clearly. Today, we properly reject the BIA’s ruling interpreting the 83
    352. coverage of § 601(a). But in the spirit of Ventura, Thomas, and our own tradition of sending things back to the BIA for a first reading, we should now ask the BIA something that it has never been asked by any court: What would you do under § 1101(a)(42)(A), given that § 601(a) does not give you the authority to do what you did in C-Y-Z- and S-L-L-?6 We do not know what answer the BIA would give to that question for the simplest of reasons. The agency has never been specifically asked. And we should not, indeed cannot properly, assume that what it would say in response — one way or another — would be either a reasonable or an unreasonable interpretation of the statute. Moreover, since it is possible that such interpretation might have covered Zhen Hua Dong, I cannot concur with the majority and concurrences that his case is now hopeless. The sad thing is that, in their rush to reach a result in terms of who gets asylum and who does not, both the majority and the concurrers sanction bad law and bad practices with 6 The majority, attempting to answer my opinion, says, at footnote 15, that remanding Zhen Hua Dong’s case to the BIA would be engaging in useless “ping pong.” With great respect, the majority in that footnote simply repeats its conflation of two quite separate things. It is certainly true that the BIA has had multiple occasions to consider the “spousal” question under § 601(a), and has answered (incorrectly, we all agree) that per se persecutee status is available to spouses under that section. But it has never been asked what the status of spouses or of people situated like Zhen Hua Dong would be under the general terms of § 1101(a)(42)(A), if § 601(a) did not cover spouses. And, in view of its consistent – but incorrect, we today hold – rulings that § 601(a) did apply to spouses, the BIA never had any reason to address that question on its own. The concurrers, nevertheless, act as if the BIA had addressed the question and had validly given spouses per se persecutee status under § 1101(a)(42)(A). The majority holds that even if the BIA were to consider the question, it could not validly say that spouses et al. were covered per se. Both the majority and the concurrers seem to me to overstep, and for precisely the reasons indicated in Ventura and Thomas. It is not proper for appellate courts to speak for the BIA and to decide the validity of that “speech,” before the agency has had a full and focused opportunity to make its position clear. On § 1101(a)(42)(A), the agency has not yet had that opportunity. It is not ping pong when only one player has been invited to the relevant table. 84
    353. respect to our relationship with the BIA. The reason they do this is certainly understandable. But it is all unnecessary. It’s just being in a hurry. **** For all these reasons, while I concur with the majority opinion insofar as it (1) dismisses the petition of Xian Zou for lack of jurisdiction; (2) denies the petition of Shi Liang Lin as moot; and (3) persuasively interprets 8 U.S.C. § 1158(c)(2)(A) as being limited to a “fundamental change” in country conditions, I must respectfully dissent from the premature denial of Zhen Hua Dong’s petition. 85
    354. E.E.O.C. v. J.B. HUNT TRANSPORT, INC. 69 Cite as 321 F.3d 69 (2nd Cir. 2003) cal in directing that any lack of clarity must be resolved in favor of the insured. EQUAL EMPLOYMENT OPPOR- See Ferraiolo Const. Co., 584 A.2d at 609 TUNITY COMMISSION, (‘‘Any ambiguity must be resolved in favor Plaintiff–Appellant, of a duty to defend.’’) (Maine law); Wil- kin Insulation Co., 161 Ill.Dec. 280, 578 v. N.E.2d at 930 (‘‘All doubts and ambiguities J.B. HUNT TRANSPORT, INC., must be resolved in favor of the insured.’’) Defendant–Appellee. (Illinois law); Lime Tree Vill. Cmty. Club Ass’n, 980 F.2d at 1405 (‘‘If the allegations Docket No. 01–6084. of the complaint leave any doubt as to the duty to defend, the question must be re- United States Court of Appeals, solved in favor of the insured.’’) (Florida Second Circuit. law). Regardless of which of the three Argued: Jan. 9, 2002. state’s laws applied, the potential for cov- erage existed at the time CI refused to Decided: Feb. 5, 2003. defend Auto Europe. The duty to defend was therefore ‘‘clear’’ and, accordingly, the district court properly awarded attorney’s Equal Employment Opportunity Com- fees. mission (EEOC) commenced action pursu- ant to the Americans with Disabilities Act VI. Conclusion (ADA), alleging that truckload motor carri- er violated the ADA by discriminating The district court properly concluded against over-the-road truck drivers who that this insurance coverage dispute used certain prescription medications. On should be heard in Maine and resolved parties’ cross-motions for summary judg- pursuant to Maine law. Because CI’s duty ment, the United States District Court for to defend was clear, the district court the Northern District of New York, Nor- properly awarded attorney’s fees to Auto man A. Mordue, J., 128 F.Supp.2d 117, Europe. granted summary judgment in favor of The judgment of the district court is carrier, and EEOC appealed. The Court of therefore affirmed. Appeals, F.I. Parker, Circuit Judge, held that applicants perceived as unsuitable for position of over-the-road truck drivers were not perceived as substantially limited in major life activity of working, as would , establish ‘‘disability’’ under the ADA. Affirmed. Sotomayor, Circuit Judge, dissented and filed opinion. leaves some ambiguity on whether allegations based on non-intentional conduct. See, e.g., of intentional conduct eliminate the duty to Applestein, 377 So.2d at 231 (holding that defend pursuant to an intentional acts policy allegations of malice and deliberate ‘‘ ‘attempt exclusion even when facts could be developed to discredit’ ’’ negated coverage). at trial to support judgment for the plaintiff
    355. 70 321 FEDERAL REPORTER, 3d SERIES 1. Federal Courts O776, 802 5. Civil Rights O173.1 The Court of Appeals reviews a dis- Comments made by people other than trict court’s grant of summary judgment ultimate hiring authorities suggesting that de novo, construing the evidence presented certain applicants for position of over-the- below in the light most favorable to the road truck drivers were not suited to any non-moving party. form of professional driving, based on their use of prescription medications with side 2. Federal Courts O759.1, 766 effects that could impair driving ability, While the Court of Appeals may af- were not sufficient to indicate that truck- firm a district court’s grant of summary load motor carrier thought applicants were judgment on any ground with adequate more broadly limited in major life activity support in the record, it may not affirm of working, so as to regard them as dis- summary judgment where any evidence in abled within meaning of ADA, where carri- the record would support a reasonable in- er had its own safety requirements above ference in favor of the opposing party. and beyond those of federal standards, and 3. Civil Rights O173.1 it did hire some applicants on medications at issue. Americans with Disabilities Act Applicants’ perceived unsuitability for of 1990, § 3(2), 42 U.S.C.A. § 12102(2); 29 position of over-the-road truck drivers, C.F.R. § 1630.2(j)(3)(i). based on their use of prescription medi- cations with side effects that could impair 6. Civil Rights O173.1 driving ability, was not a perceived inabili- A finding of perceived disability, for ty to perform broad range or class of jobs, purposes of a discrimination action under but rather was limitation on particular job the ADA, may not rest merely on a single within larger group of jobs, and thus appli- employer’s failure to hire a candidate. cants failed to establish that they were Americans with Disabilities Act of 1990, perceived as substantially limited in major § 2 et seq., 42 U.S.C.A. § 12101 et seq. life activity of working, as would establish ‘‘disability’’ under the ADA. Americans 7. Civil Rights O240(2) with Disabilities Act of 1990, § 3(2), 42 For purposes of a discrimination ac- U.S.C.A. § 12102(2); 29 C.F.R. tion under the ADA, courts will not pre- § 1630.2(j)(3)(i). sume a mistaken assumption of disability based only on an employer’s decision not 4. Civil Rights O173.1 to hire certain candidates. Americans Truckload motor carrier did not view with Disabilities Act of 1990, § 2 et seq., applicants who were perceived unsuitable 42 U.S.C.A. § 12101 et seq. for position of over-the-road truck drivers based on their use of prescription medi- 8. Civil Rights O173.1 cations with side effects that could impair Applicants for over-the-road truck driving ability as unable to drive any driver positions who suffered from condi- trucks, so as to regard them as disabled tions treated with prescription medications within meaning of ADA, but rather per- with side effects that could impair driving ceived applicants as unfit to perform spe- ability were not regarded as substantially cific job of long-distance, freight-carrying, limited in major life activity of working tractor-trailer driving. Americans with based on underlying condition itself, as Disabilities Act of 1990, § 3(2), 42 U.S.C.A. would establish ‘‘disability’’ under the § 12102(2); 29 C.F.R. § 1630.2(j)(3)(i). ADA; rather, employer perceived appli-
    356. E.E.O.C. v. J.B. HUNT TRANSPORT, INC. 71 Cite as 321 F.3d 69 (2nd Cir. 2003) cants as unfit for positions based on use of that the applicants in question had been medications with dangerous side effects. denied over-the-road driving positions with Americans with Disabilities Act of 1990, Hunt because of their use of medications § 3(2), 42 U.S.C.A. § 12102(2); 29 C.F.R. with potentially harmful side effects, and § 1630.2(j)(3)(i); 49 C.F.R. § 391.41. not as a result of an actual or perceived disability or a record of disability as con- templated by the Americans with Disabili- ties Act of 1990, 42 U.S.C. § 12101, et seq. (‘‘ADA’’). On appeal, the EEOC argues Julie L. Gantz, Equal Opportunity Em- exclusively that Hunt regarded the reject- ployment Commission (Nicholas M. Inzeo, ed applicants as disabled, i.e., substantially Acting Deputy General Counsel, Philip B. limited from a major life activity, as de- Sklover, Associate General Counsel, Vin- fined by 42 U.S.C. § 12102(2)(C) because cent J. Blackwood, Assistant General of their use of certain medications. Be- Counsel, on brief), for Plaintiff–Appellant. cause we find that Hunt did not regard the James H. Hanson, Scopelitis, Garvin, applicants as disabled as defined by the Light & Hanson, Indianapolis, IN (Laurie ADA, we affirm the decision of the district T. Baulig, Scopelitis, Garvin, Light & Han- court. son, Washington, DC, Thomas J. Grooms, Bond Schoeneck & King, Syracuse, NY, on II. brief), for Defendant–Appellee. J.B. Hunt Transportation, Inc. (‘‘Hunt’’) Before: JACOBS, F.I. PARKER, is the nation’s largest publicly held motor SOTOMAYOR, Circuit Judges. carrier company. Hunt operates for-hire property transport services in the forty- F.I. PARKER, Circuit Judge. eight contiguous states, the District of Co- J.B. Hunt Transport, Inc. chose not to lumbia, Canada, and Mexico. Its fleet in- employ over-the-road truck drivers who cludes 8,000 tractors, and it employs ap- used prescription medications with side ef- proximately 12,000 individuals to drive the fects that might impair driving ability. trucks. Of these employees, approximate- The Equal Employment Opportunity Com- ly 10,000 are the over-the-road (‘‘OTR’’) mission argued that under the Americans drivers whose positions are at issue in this with Disabilities Act, Hunt’s decision vio- case. These OTR drivers operate vehicles lated the rights of job applicants using weighing approximately 80,000 pounds those medications. We disagree. over irregular routes under particularly difficult work conditions, including sleep I. deprivation, irregular work and rest cycles, Plaintiff–Appellant Equal Employment inclement weather, long driving periods, Opportunity Commission (‘‘EEOC’’) ap- long layovers, irregular meal schedules, peals from the February 8, 2001 decision tight delivery schedules, en route delays, of the United States District Court for the night driving, accumulated fatigue, stress, Northern District of New York (Norman and extended periods of loud noise and A. Mordue, Judge ) granting defendant vibrations. According to Hunt, the large J.B. Hunt Transport Inc.’s (‘‘Hunt’’ ’s) mo- vehicle size and extreme driving conditions tion for summary judgment and denying faced by its OTR drivers warrant height- plaintiff EEOC’s cross-motion for sum- ened safety evaluations of those OTR driv- mary judgment. The district court found ers.
    357. 72 321 FEDERAL REPORTER, 3d SERIES Like other motor carriers, Hunt is sub- (‘‘Whiteside’’), based entirely on notations ject to federal regulation under the De- in the 1993 edition of the Physician’s Desk partment of Transportation’s Federal Mo- Reference (‘‘PDR’’). Whiteside divided tor Carrier Safety Act Regulations the DRL into six columns labeled ‘‘name,’’ (‘‘FMCSAR’’). 49 C.F.R. § 301, et seq. ‘‘class,’’ ‘‘comment,’’ ‘‘restriction,’’ ‘‘treats,’’ (2001). These regulations establish mini- and ‘‘1993 PDR page number.’’ In the mum qualifications for any person driving ‘‘restriction’’ column, Whiteside indicated a commercial motor vehicle, as well as the impact a particular drug might have minimum duties for motor carriers using on an applicant’s eligibility. Whiteside OTR drivers. The regulations specifically designated five categories of restrictions: allow an operator to require and enforce ‘‘Rule Out Side [E]ffects,’’ ‘‘Not Permit- ‘‘more stringent requirements relating to ted,’’ ‘‘Unsafe [E]ffects,’’ ‘‘Heart Condi- safety of operation and employee safety tion,’’ and ‘‘Disqualifying Condition.’’ 1 An and health’’, 49 C.F.R. § 390.3(d), and re- applicant whose medication had a ‘‘Rule quire operators to restrict drivers from Out Side Effects’’ notation was required to operating vehicles ‘‘while the driver’s abili- obtain a release from the prescribing doc- ty or alertness is so impaired, or so likely tor certifying that the applicant could to become impaired, through fatigue, ill- safely drive a tractor trailer truck while ness, or any other cause, as to make it using the medication. An applicant taking unsafe for him/her to begin or continue to a ‘‘Not Permitted,’’ ‘‘Unsafe Effects,’’ operate the commercial motor vehicle.’’ Id. ‘‘Disqualifying Condition,’’ or ‘‘Heart Con- at § 392.3 (2001). A motor carrier is re- dition’’ medication could not drive for quired to ensure that drivers do not oper- Hunt while using the indicated medi- ate unless they are in compliance with the cation.2 The notation ‘‘Unsafe Effects’’ in- DOT regulations. 49 C.F.R. §§ 391.11, dicated either that the PDR cautioned 392.3, 392.4(b)(2001). users against operating heavy equipment or driving automobiles while taking the A. The Drug Review List drug (noted as ‘‘warning on driving’’ in the Between September 1993 and May comment column) or that the drug caused 1994, in an effort to comply with the drowsiness, sedation, or a high incidence FMCSAR in its hiring processes, Hunt of dizziness. A ‘‘Rule Out Side Effects’’ created a Drug Review List (‘‘DRL’’) of notation indicated that a medication could medications known to have side effects cause side effects similar to, but less per- that might impair driving ability. The vasive than, those warranting an ‘‘Unsafe list, thirty-seven pages in length and in- Effects’’ label. Finally, ‘‘Heart Condition’’ cluding over 836 medications, was com- indicated that the medication was general- piled by Hunt’s Safety Department Di- ly used for heart problems that could dis- rector of Compliance, David Whiteside qualify drivers under DOT regulations. 1. The Court will use the corrected labels these medications in the ‘‘Not Permitted’’ cat- ‘‘Rule Out Side Effects’’ for ‘‘Rule Out Side egory. Equal Employment Opportunity Affects’’ and ‘‘Unsafe Effects’’ for ‘‘Unsafe Af- Comm’n v. J.B. Hunt Transp., Inc., 128 fects’’ throughout the opinion. F.Supp.2d 117, 120 n. 2 (N.D.N.Y.2001); see 49 C.F.R. § 391.42(b)(12)(i) (prohibiting use 2. The district court found that Whiteside mis- of Schedule I drugs, amphetamines, narcot- takenly believed that the DOT prohibited ics, and other habit-forming drugs); 49 drivers from using any Schedule II–V medi- cations, rather than only Schedule I medi- C.F.R. § 392.2 (same). cations, and that he therefore included all of
    358. E.E.O.C. v. J.B. HUNT TRANSPORT, INC. 73 Cite as 321 F.3d 69 (2nd Cir. 2003) B. The Hunt Hiring Process III. Upon receiving an application for a com- EEOC filed its complaint in the United mercial driving position, Hunt forwarded States District Court for the Northern the application to its Corporate Driver District of New York on October 24, 1997. Personnel Department in Lowell, Arkan- Both sides moved for summary judgment. sas for screening of motor vehicle, crimi- EEOC alleged that Hunt violated the ADA nal, and prior employment records and for by discriminating against individuals with a review of listed references. If an appli- disabilities or ‘‘perceived’’ disabilities. cant passed this first level of screening and Hunt alleged that the DRL was a safety- received a conditional employment offer, related qualification standard addressing the applicant underwent medical screen- serious business concerns. The district ing, including questioning regarding the court granted summary judgment for applicant’s use of prescription medication Hunt and denied EEOC’s summary judg- for the last five years. Hunt used non- ment motion based on (1) its conclusion medical personnel to conduct these screen- that ADA protections did not extend to the ings. If the applicant indicated use of a excluded driver-applicants because the ap- prescription drug, the reviewing employee plicants were not, by virtue of their use of consulted Hunt’s medical guidelines 3 and certain medications, disabled within the the DRL to determine the applicant’s med- meaning of the ADA, and (2) its finding ical eligibility. that the EEOC had failed to contradict Hunt’s assertion that its use of the DRL as a safety measure was reasonable within C. The EEOC Claim DOT guidelines. Equal Employment Op- EEOC claims that Hunt improperly re- portunity Comm’n. v. J.B. Hunt Transp., jected 546 applicants in violation of the Inc., 128 F.Supp.2d 117, 135–36 (N.D.N.Y. ADA on the basis of a ‘‘blanket’’ exclusion- 2001). On appeal, EEOC abandoned its ary policy. EEOC admits, however, that argument that the excluded applicants Hunt hired several applicants who were were ‘‘disabled’’ under the ADA, claiming using drugs prohibited under the DRL—in only that the district court erred by grant- 1995, two applicants using drugs labeled ing summary judgment to Hunt when the ‘‘Disqualifying Condition’’ and eleven using evidence supported the conclusion that drugs labeled ‘‘Unsafe Effects,’’ and in Hunt regarded the applicants as disabled 1996 and 1997, one applicant taking a ‘‘Dis- because of their use of medications on the qualifying Condition’’ drug and thirteen DRL. using drugs with ‘‘Unsafe Effects.’’ Prior to commencing work, each of these new IV. employees provided Hunt with medical [1, 2] We review a district court’s documentation from a treating physician grant of summary judgment de novo, con- or health care provider certifying that he struing the evidence presented below in or she did not suffer from the potentially the light most favorable to the non-moving problematic side effects and could operate party. Manning v. Utils. Mut. Ins. Co., a truck safely while taking the drug. 254 F.3d 387, 391 (2d Cir.2001). While 3. As the district court found, Hunt maintained for Hunt unless he or she had been off such a restrictive policy on the use of drugs for drugs for at least thirty days before commenc- psychological conditions separate from the ing work. Hunt Medical Guidelines, April 11, DRL. An applicant was not eligible to drive 1996.
    359. 74 321 FEDERAL REPORTER, 3d SERIES this Court may affirm on any ground with ‘‘disability,’’ on appeal, EEOC alleges only adequate support in the record, we may that the rejected OTR driver applicants not affirm summary judgment where any were ‘‘regarded as’’ disabled by Hunt evidence in the record would support a based on their use of certain medications, reasonable inference in favor of the oppos- invoking the statutory definition of disabil- ing party. See McCarthy v. Am. Int’l ity under § 12102(2)(C). As the Supreme Group, Inc., 283 F.3d 121, 124 (2d Cir. Court explained in Sutton v. United Air 2002); VKK Corp. v. Nat’l Football Lines, Inc., ‘‘[t]here are two apparent League, 244 F.3d 114, 119 (2d Cir.2001). ways in which individuals may fall within this [§ 12102(2)(C) ] statutory definition: A. The Statutory Framework and the (1) a covered entity mistakenly believes Definition of ‘‘Disability.’’ that a person has a physical impairment The ADA provides a deceptively simple that substantially limits one or more major definition of disability: life activities, or (2) a covered entity mis- The term ‘‘disability’’ means, with re- takenly believes that an actual, nonlimiting spect to an individual— impairment substantially limits one or more major life activities.’’ 527 U.S. 471, (A) a physical or mental impairment 489, 119 S.Ct. 2139, 144 L.Ed.2d 450 that substantially limits one or more (1999). of the major life activities of such individual; Evaluating the evidence before the dis- (B) a record of such an impairment; trict court, this Court agrees with the or court below that EEOC failed to put forth evidence sufficient to demonstrate that the (C) being regarded as having such an rejected applicants were ‘‘disabled’’ within impairment. the meaning of the ADA. Specifically, 42 U.S.C. § 12102(2)(1995). EEOC regu- EEOC failed to set forth evidence suffi- lations further develop this definition, ex- cient to establish that Hunt perceived re- plaining ‘‘physical or mental impairment’’ jected applicants as substantially limited in as: their ability to perform a major life activi- (1) Any physiological disorder, or condi- ty. tion, cosmetic disfigurement, or anatomi- cal loss affecting one or more of the B. The Evidence Is Insufficient To following body systems: neurological, Support the Inference that Hunt musculoskeletal, special sense organs, Regarded Applicants as Having a respiratory (including speech organs), ‘‘Substantial Limitation’’ on a ‘‘Ma- cardiovascular, reproductive, digestive, jor Life Activity.’’ genito-urinary, hemic and lymphatic, To qualify for ADA protections, a per- skin, and endocrine; or son’s ‘‘impairment’’ must ‘‘substantially (2) Any mental or psychological disor- limit’’ a ‘‘major life activit[y].’’ 42 U.S.C. der, such as mental retardation, organic § 12102(2). Major life activities may in- brain syndrome, emotional or mental ill- clude ‘‘caring for oneself, performing man- ness, and specific learning disabilities. ual tasks, walking, seeing, hearing, speak- 29 C.F.R. § 1630.2(h) (2001). ing, breathing, learning,’’ and, pertinent to Although EEOC initially challenged this appeal, ‘‘working.’’ 29 C.F.R. Hunt’s reliance on the DRL under all § 1630.2(i). An activity is ‘‘substantially three prongs of the statutory definition of limited’’ when an individual cannot per-
    360. E.E.O.C. v. J.B. HUNT TRANSPORT, INC. 75 Cite as 321 F.3d 69 (2nd Cir. 2003) form the activity that an average person in of a Hunt OTR driver does not mean that the general population could perform or one could not successfully engage in other faces significant restrictions in the ‘‘condi- types of truck driving, let alone in other tion, manner, or duration under which the kinds of safety-sensitive work. individual can TTT perform [the] activity.’’ In Sutton, the Supreme Court consid- 29 C.F.R. § 1630.2(j)(i)-(ii). The activity ered the claims of pilots who had been of ‘‘working’’ is further defined by the denied positions as ‘‘global pilots’’ with regulations as follows: United Airlines. The Supreme Court held With respect to the major life activity of that the position of ‘‘global pilot’’ was ‘‘a working— single job’’ and, therefore, was not suffi- (i) The term substantially limits means ciently broad to satisfy the ‘‘major life significantly restricted in the ability to activity requirement’’. Sutton, 527 U.S. at perform either a class of jobs or a broad 493, 119 S.Ct. 2139. The Court reasoned range of jobs in various classes as com- that ‘‘there are a number of other positions pared to the average person having com- utilizing petitioners’ skills, such as regional parable training, skills and abilities. pilot and pilot instructor to name a few.’’ The inability to perform a single, partic- Id. ular job does not constitute a substantial limitation in the major life activity of Like the limitation that United Airlines working. placed on global airline pilots in Sutton, 29 C.F.R. § 1630.2(j)(3)(i). Thus, unless the limitation that Hunt placed on appli- Hunt perceived the applicants in question cants for the position of OTR driver was a as limited from a class of jobs or a broad limitation on a particular job within a larg- range of jobs, the EEOC’s claim must fail. er group of jobs, and not a substantial 1. Driving 40–Ton, 18–Wheel Trucks limitation on working. See Baulos v. Over Long Distances for Extended Roadway Express, Inc., 139 F.3d 1147,- Periods is Neither a ‘‘Class of Job’’ 1154 (7th Cir.1998) (driving sleeper trucks nor a ‘‘Broad Range of Jobs’’ Within is a specific job within the broader class of the Meaning of the ADA. truck driving jobs). Therefore, the appli- cants’ perceived unsuitability for the posi- [3] Driving freight-carrying tractor- tion of OTR driver cannot be characterized trailer trucks over long distances for ex- as a perceived inability to perform a broad tended periods of time is neither a ‘‘class range or a class of jobs. This is true even of jobs’’ nor a ‘‘broad range of jobs,’’ as the assuming that truck-driving in general is a EEOC alleges, but rather a specific job sufficiently broad range or class of jobs to with specific requirements. Such a posi- constitute a ‘‘major life activity’’, an issue tion requires specific abilities, especially we do not need to reach. As the dissent the ability to stay alert over long hours readily acknowledges, persons licensed to under difficult conditions. A Hunt OTR drive the types of vehicles driven by Hunt driver’s alertness cannot flag. He or she OTR drivers are also qualified to drive must be able to stay alert and withstand ‘‘various types of small and large trucks, the mesmerizing affect of driving an eigh- including tractor-trailers, moving trucks, teen-wheel vehicle for hours at a stretch, and cargo vans.’’ Dissent page 80. sometimes at night, with continuous vibra- tion over long distances. Given these de- Accordingly, to show that Hunt per- manding requirements, the fact that one ceived applicants rejected under the DRL may not be able to perform the specific job as substantially limited in a major life ac-
    361. 76 321 FEDERAL REPORTER, 3d SERIES tivity, the EEOC must show that Hunt did not have another, less demanding driv- viewed such applicants as limited from a ing position to offer the candidates does broader range or class of jobs than merely not indicate that Hunt perceived the candi- OTR positions at Hunt. dates as being unqualified for any driving position at all. Giordano v. City of New 2. The Evidence Is Not Sufficient To York, 274 F.3d 740, 748–50 (2d Cir.2001) Support a Reasonable Inference that (finding inability of the New York Police Hunt Regarded Applicants Rejected Department to offer light duty, non-patrol Under the ‘‘Not Permitted’’ and position to officer taking anti-coagulation ‘‘Unsafe Effects’’ Categories as Sub- medication did not demonstrate that offi- stantially Limited in a Broad Range cer was substantially limited in working or Class of Jobs. where other security and law enforcement [4] The EEOC argues that Hunt re- jobs in the area had such positions); see garded applicants who took particular also Baulos v. Roadway Express Inc., 139 medications as incapable of driving trucks, F.3d 1147, 1154 (2d Cir.1998) (concluding which according to the EEOC constitutes that truck drivers unable to operate sleep- either a ‘‘class of jobs’’ or a ‘‘broad range er trucks did not show that they were of jobs.’’ The record, however, only shows regarded as disabled where employer did that Hunt saw the applicants as unfit to not offer them less demanding, non-over- perform a job for which they were seeking night positions that were taken by drivers applicants: long-distance, freight-carrying, with more seniority). tractor-trailer driving. The Supreme [5] EEOC references a few comments Court has clearly stated that ‘‘[t]he inabili- from Hunt’s evaluators to candidates sug- ty to perform a single, particular job,’’ gesting that certain candidates were not however, ‘‘does not constitute a substantial suited to any form of professional driving. limitation in the major life activity of work- These comments, made by people other ing.’’ Sutton v. United Air Lines, Inc., than the ultimate hiring authorities, simply 527 U.S. 471, 493, 119 S.Ct. 2139, 144 are not sufficient to indicate that Hunt L.Ed.2d 450 (1999). Here, Hunt dismissed thought the applicants were more broadly the applicants as unable to meet Hunt’s limited given the heightened nature of own safety requirements—requirements Hunt’s standards and the fact that Hunt above and beyond the DOT’s industry-wide did hire some applicants on DRL medi- standards and unique from the require- cations. Although a few evaluators’ com- ments of other trucking companies. See ments could be more broadly interpreted, Compl. ¶ 8c; Def.’s Statement of Material there is no evidence that Hunt’s reviewers, Facts at 7; see also Adair Dep. at 85–86; relying on Hunt’s own DRL and drug lists J.B. Hunt Transp., Inc., 128 F.Supp.2d at to make a judgment on qualification for a 129 n. 17 (noting drivers were employed by position at Hunt, intended to make an other trucking companies while taking evaluation beyond Hunt’s specific guide- same medications). lines. Nor is there sufficient evidence to The evidence suggests that Hunt found support a finding that Hunt viewed the the applicants unsuited for long-distance driving limitation as extending beyond driving of Hunt’s 40–ton trucks on irregu- Hunt. Furthermore, as the Supreme Court lar, stressful schedules, but does not indi- has clearly stated, ‘‘[i]t is not enough to cate that Hunt perceived the applicants as say that if the physical criteria of a single more broadly limited. The fact that Hunt employer were imputed to all similar em-
    362. E.E.O.C. v. J.B. HUNT TRANSPORT, INC. 77 Cite as 321 F.3d 69 (2nd Cir. 2003) ployers one would be regarded as substan- an employer’s decision not to hire certain tially limited in the major life activity of candidates. working only as a result of this imputa- tion.’’ Sutton, 527 U.S. at 493, 119 S.Ct. 3. The Evidence Is Also Not Sufficient 2139. To Support a Reasonable Inference [6] In short, EEOC demonstrated only that Hunt Regarded Applicants Re- that Hunt refused to hire certain appli- jected Under the ‘‘Disqualifying cants according to its own hiring criteria; Condition’’ and ‘‘Heart Condition’’ however, a finding of perceived disability Categories as Substantially Limited may not rest merely on a single employer’s in a Broad Range or Class of Jobs. failure to hire a candidate. Baulos, 139 As noted above, two of the categories in F.3d at 1154 (‘‘Courts have uniformly held the DRL appear, at least superficially, to that an employer does not necessarily re- refer to the condition causing the reliance gard an employee as handicapped simply on a DRL drug, not merely the applicants’ by finding the employee to be incapable of use of a DRL medication.4 Although the satisfying the singular demands of a par- evidence suggests that these applicants ticular job.’’ (internal citation omitted)). were, like those in the other categories, [7] Thus, we affirm the district court’s often told that they were disqualified on grant of summary judgment in favor of the basis of the drug they were using Hunt as to the applicants rejected under at rather than on the basis of the condition least the ‘‘Not Permitted’’ and ‘‘Unsafe supporting their use of the drug, we brief- Effects’’ categories because EEOC has ly consider whether applicants using drugs failed to demonstrate that Hunt mistaken- from these two categories warrant a differ- ly perceived that the rejected applicants’ ent legal conclusion. We conclude that had impairments that substantially limited they do not. a ‘‘major life activity.’’ Accordingly, EEOC has failed to show that the appli- [8] Individuals suffering from the con- cants were ‘‘disabled’’ within the meaning ditions treated with the ‘‘Heart Condition’’ of the ADA. In so holding, we emphasize or ‘‘Disqualifying Condition’’ drugs are po- that this Court will not presume a mistak- tentially explicitly barred from truck driv- en assumption of disability based only on ing by 49 C.F.R. § 391.41.5 Hunt therefore 4. As noted in the discussion of the pertinent tery occlusion, and severe headache, while facts, the DRL contained five categories of the label ‘‘Heart Condition’’ attached to medi- drugs: ‘‘Rule out Side Effects,’’ ‘‘Not Permit- cines treating heart failure, thrombosis, ede- ted,’’ ‘‘Unsafe Effects,’’ ‘‘Heart Condition,’’ ma, congestive heart failure, ischemia, and and ‘‘Disqualifying Condition.’’ The EEOC ventric arrythmia. 49 C.F.R. § 391.41 ap- does not represent in this appeal any appli- pears to exclude persons with all of these cants rejected under the ‘‘Rule Out Side Ef- conditions from driving a commercial vehicle fects’’ category, thus removing that category where those conditions are likely to interfere from our consideration. J.B. Hunt Transp., with their ability to safely drive a commercial Inc., 128 F.Supp.2d at 122, n. 7. vehicle. 49 C.F.R. §§ 391.41(b)(3) (diabetes), 391.41(b)(4) (‘‘myocardial infarction, angina 5. Review of the DRL reveals that the label pectoris, coronary insufficiency, thrombosis, ‘‘Disqualifying Condition’’ attached to drugs or any other cardiovascular disease of a vari- treating Parkinson’s Disease, serious arryth- ety known to be accompanied by syncope, mia, alcoholism, epilepsy, seizure, migraines, dyspnea, collapse or congestive heart fail- dementia, depression, schizophrenia, diabe- ure’’), 391.41(b)(6) (high blood pressure), tes, severe arthritis, severe hypertension, opi- 391.41(b)(7) (‘‘rheumatic, arthritic, orthope- ate addiction, subarachnoid hemorrhage, ar- dic, muscular, neuromuscular, or vascular
    363. 78 321 FEDERAL REPORTER, 3d SERIES potentially regarded applicants using these V. drugs as substantially limited not just Although Hunt admittedly rejected the from driving Hunt vehicles according to applicants for its OTR driving positions the rules of the DRL and other company because of their use of certain prescription regulations, but as prevented from driving medications, the EEOC cannot succeed in legally for any commercial trucking com- its ADA claim on behalf the rejected appli- pany. As with the other categories in the cants. The record only shows that Hunt DRL, however, the restrictions on the regarded the applicants in question as ine- medications labeled ‘‘Disqualifying Condi- ligible for a specific position within Hunt, tion’’ or ‘‘Heart Condition’’ were placed on not that Hunt regarded them as ‘‘disabled’’ applicants taking the drug, not on appli- within the meaning of the ADA. The appli- cants with the underlying condition itself. cants, through the EEOC, therefore do not Even though in some cases, the company, have a valid ADA claim. under 49 C.F.R. § 391.41, could have cre- ated a policy excluding the applicant on For the reasons set forth above, this the basis of the underlying condition, the Court affirms the district court’s grant of basis for the exclusion from employment summary judgment to defendant Hunt and was the use of a listed drug, not any its denial of the cross-motion by plaintiff potential ‘‘disability’’ created by the treat- EEOC. ed disease.6 The judgment of the district court is AFFIRMED. We conclude, therefore, that any claims arising under the ‘‘Heart Condition’’ and ‘‘Disqualifying Condition’’ categories are SOTOMAYOR, Circuit Judge, dissenting. not distinguishable from the claims under the ‘‘Not Permitted’’ and ‘‘Unsafe Effects’’ This case is quite straightforward. categories and so they must also fail. Based upon a list of drugs and their poten- disease’’), 391.41(b)(8) (epilepsy or ‘‘any other current regulations. See 49 C.F.R. condition which is likely to cause loss of con- § 391.41(b)(4). sciousness’’), 391.41(b)(9) (‘‘mental, nervous, organic, or functional disease or psychiatric 6. For example, Amandtadine Hydrochloride, disorder’’), 391.41(b)(12)(i) (controlled sub- a drug to which the ‘‘Disqualifying Condi- stances), 391.41(b)(13) (alcoholism). ‘‘Mi- tion’’ label attaches treats both Parkinson’s granes’’ or ‘‘severe headaches,’’ as ‘‘vascular Disease and the flu. While an applicant tak- headache[s],’’ DORLAND’S ILLUSTRATED ing the drug for Parkinson’s might be dis- MEDICAL DICTIONARY 1042 (28th abled on the basis of the disease within the ed.1994), potentially fall within meaning of the ADA, an applicant using the § 391.41(b)(7)’s restriction on vascular dis- drug for the flu would not likely so qualify. ease. Hunt, however, would have excluded either EEOC alleged that Hunt misinterpreted a applicant because of the drug usage. Fur- DOT report cautioning about the effects of thermore, some ‘‘Disqualifying Condition’’ drugs used to treat heart conditions, claiming drugs treat the same underlying diseases as that the report merely required individual as- drugs given other labels. For example Zoloft, sessment of each patient. The DOT subse- a drug used for treatment of, inter alia, de- quently issued a report clarifying that the use pression, is listed as ‘‘Unsafe Effects,’’ while of Coumadin, a anticoagulator previously questioned, was not automatically disqualify- Prozac, also for depression, is listed as ‘‘Dis- ing. J.B. Hunt Transp., Inc., 128 F.Supp.2d qualifying Condition.’’ This further supports at 120 n. 3 (describing reports). According to the idea that the drug, not the condition itself, the DRL, however, Coumadin treats thrombo- was the true basis of Hunt’s hiring ban. sis, a condition specifically prohibited by the
    364. E.E.O.C. v. J.B. HUNT TRANSPORT, INC. 79 Cite as 321 F.3d 69 (2nd Cir. 2003) tial side effects compiled by David White- truck driving in general is such a specific side, a Hunt employee with no medical class of jobs that a substantial limitation training, and a Medical Guidelines policy on truck driving would fail to imply a developed by Michael Gray, a former Red disability; its holding relies solely upon an Lobster cashier with no medical training inappropriately narrow view that Hunt who was, nevertheless, Hunt’s Medical Ad- perceived the applicants as limited only in visor, Hunt determined that certain appli- their ability to work as long haul truckers cants were unfit to be truck drivers. The for Hunt. EEOC has provided substantial evidence that Hunt believed that these individuals Contrary to the majority’s assertion, the were unfit to drive a truck, or, for that EEOC has produced significant evidence matter, to drive at all and were incapable that Hunt regarded the applicants as sub- of performing the broad class of jobs that stantially limited in the major life activity fall under the classification ‘‘truck driving.’’ of working as truck drivers in general. An Based upon this showing, I would vacate employer perceives an employee to be sub- the district court’s grant of summary judg- stantially limited in his or her ability to ment and hold that there is a genuine work if it believes the employee is: dispute of material fact with respect to significantly restricted in the ability to whether the EEOC has established a pri- perform either a class of jobs or a broad ma facie case of disability discrimination. range of jobs in various classes as com- I therefore respectfully dissent. pared to the average person having com- I agree with the majority that the issue parable training, skills and abilities. in this appeal is whether the applicants The inability to perform a single, partic- were denied truck driving positions at ular job does not constitute a substantial Hunt because of their perceived disability limitation in the major life activity of within the meaning of the ADA. Ignoring working. significant evidence that Hunt perceived 29 C.F.R. § 1630.2(j)(3)(i); see also Bart- the applicants as more broadly limited, lett v. N.Y. State Bd. of Law Exam’rs, 226 however, the majority holds that the F.3d 69, 82–83 (2d Cir.2000). Factors that EEOC has only provided evidence that may be considered under this standard Hunt perceived the rejected applicants as include the geographical area to which an ‘‘ineligible for a specific position within individual has reasonable access; the num- Hunt.’’ Ante at 78. In doing so, the major- ber and types of jobs utilizing similar ity reasons that long haul trucking is not a training, knowledge, skills or abilities as sufficiently broad class of jobs such that a the job from which the applicant has been substantial limitation on an individual’s disqualified; and the number and types of ability to be a long haul trucker would jobs not utilizing similar training, knowl- imply that the individual was disabled edge, skills or abilities from which the within the meaning of the ADA. See ante applicant will also be disqualified. 29 at 75–76. The majority asserts that a C.F.R. § 1630.2(j)(3)(ii). limitation on an individual’s ability to be a long haul truck driver does not substan- If other jobs utilizing an individual’s tially limit his or her ability to engage in skills are available, that person is not sub- the major life activity of working, as many stantially limited in a class of jobs, even if other truck driving jobs are available for this alternate employment would not allow these individuals. See ante at 75–76. The the individual to showcase his or her spe- majority does not, however, hold that cial talents. Sutton v. United Air Lines,
    365. 80 321 FEDERAL REPORTER, 3d SERIES Inc., 527 U.S. 471, 492, 119 S.Ct. 2139, 144 Labor Statistics, Occupational Outlook L.Ed.2d 450 (1999). In Sutton, the Su- Handbook 577 (2002–03), available at preme Court determined that plaintiffs ap- http://www.bls.gov/oco/pdf/ocos246.pdf. plying for positions as global airline pilots This evidence demonstrates that truck could use their particular skills to obtain driving is a general field of employment other piloting positions from which they rather than a specific position. Accord were not disqualified, so these plaintiffs Baulos v. Roadway Express, Inc., 139 were not regarded as being shut out from F.3d 1147, 1154 (7th Cir.1998) (holding an entire occupational class. Id. at 492–93, that driving a sleeper car is a specific job 119 S.Ct. 2139. In applying this rubric, within the class of truck drivers); Best v. the Second Circuit has found that practic- Shell Oil Co., 107 F.3d 544, 548 (7th Cir. ing law is a broad occupational class, see 1997) (holding that truck driving is a class Bartlett, 226 F.3d at 84, but that working of jobs). as a policeman is a specific position within the class of investigative or security jobs, The majority does not reach the ques- see Giordano v. City of New York, 274 tion whether truck driving is a class of F.3d 740, 749 (2d Cir.2001). jobs. Instead, the majority argues that The EEOC has proffered evidence that Hunt only dismissed the applicants be- the members of the plaintiff class have cause ‘‘Hunt found the applicants unsuited undergone specialized driver training, for long-distance driving of Hunt’s 40–ton earned commercial drivers’ licenses, trucks on irregular, stressful schedules.’’ passed road tests and received medical Ante at 76. Such hyperbole is inapposite. certifications pursuant to DOT regulations. Whether long haul trucking is, in fact, The set of jobs that call for these qualifica- different from other types of truck driving tions includes driving various types of is not the central issue in this appeal; small and large trucks, including tractor- Hunt’s perception of the applicants as sub- trailers, moving trucks, and cargo vans. stantially limited in their ability to drive See Office of Management & Budget, Stan- trucks, without further limitation to long dard Occupational Classification Manual haul truck driving, is the central issue. 220 (2000), available at http:// www.bls.gov/soc/soc v3d0.htm. The De- Beyond this basic misconception, the partment of Labor classifies truck driving majority also misrepresents the record by as a separate occupation within the overall asserting that the evidence ‘‘does not indi- category of ‘‘Transportation and Material cate that Hunt perceived the applicants as Moving Occupations,’’ as does the Office of more broadly limited.’’ Ante at 77. To the Management and Budget. Id. The De- contrary, the EEOC provided significant partment of Labor estimated that in 2000 evidence that Hunt believed that the appli- there were more than 3.3 million jobs that cants were unfit to drive trucks. Numer- came under the heading of ‘‘Truckdriver ous drugs were listed on the DRL as ‘‘Not and Driver/Sales Workers.’’ 1 Bureau of Permitted,’’ 2 reflecting a belief that the 1. Driver/Sales Workers drive trucks and work from the two condition-based categories, as sales agents for the goods they haul; both ‘‘Disqualifying Condition’’ and ‘‘Heart Condi- of these aspects are integral to their jobs. See tion.’’ Ultimately, the majority finds no legal Occupational Outlook Handbook 576–77 distinction between the ‘‘condition’’ catego- (2002–03). ries and the others. See ante at 78. 2. The majority discusses the categories ‘‘Not Permitted’’ and ‘‘Unsafe Effects’’ separately
    366. E.E.O.C. v. J.B. HUNT TRANSPORT, INC. 81 Cite as 321 F.3d 69 (2nd Cir. 2003) applicant was prohibited by DOT regula- employee that he would ‘‘never drive for tions from driving a commercial vehicle anybody,’’ and numerous other applicants while taking that particular medication. were told that the medications they were Dr. Cooper, Hunt’s physician consultant, taking made it unsafe for them to drive a testified with respect to one applicant that truck, or drive in general. See, e.g., Curtin he did not feel it was ‘‘in this patient’s best Decl., Exh. 13 (reviewer told applicant interest to pursue this profession.’’ Inter- ‘‘that she could not be on [the medication] view records show that the company be- and drive [because] it can cause unsafe lieved another applicant ‘‘would most likely affect [sic]’’); id. (reviewer told applicant have difficulty functioning in the lifestyle ‘‘that he could not drive[ ] while on this of a trucker.’’ Similarly, Dr. Cooper indi- medication’’); id. (reviewer told applicant cated with regard to another applicant that ‘‘that she cannot be on [the medication] her ‘‘problems with sleep and concentra- and drive’’); Curtin Decl., Exh. 20 (drug is tion under stress are not very compatible ‘‘not permitted for driving’’); id. (‘‘[b]oth with the lifestyle expected of a driver.’’ drugs are not approved for driving’’); id. Applicant Joseph Lisa was told by a Hunt (reviewer ‘‘informed applicant that he can- I agree that all four categories of medi- Condition’’ category). In addition, Hunt’s cations on the DRL at issue here should be Medical Guidelines relating to mental and treated identically. I disagree, however, psychological conditions required that in or- with the majority’s statement that ‘‘the basis der to qualify for a job, an applicant taking for the exclusion from employment was the medication for depression must remain off use of a listed drug, not any potential ‘dis- the medication for thirty days and submit a ability’ created by the treated disease.’’ Ante letter from a doctor stating that he or she no at 77–78. The EEOC has produced signifi- longer suffers from the underlying condition. cant evidence that demonstrates the link be- Contrary to the majority’s assertion, this evi- tween the drug categories and potential un- dence provides a direct causal link between derlying conditions. For example, one of the applicants’ underlying conditions and Hunt’s interviewers noted that the ‘‘applicant Hunt’s perception of the applicants as sub- did not indicate the reason he is taking [the stantially limited in their ability to work as medication]. [N]eed to verify why he is on truck drivers. this medication.’’ Similar comments were To make a further distinction that it ulti- made by reviewers with respect to applicants mately finds insignificant, the majority asserts taking medications in each of Hunt’s catego- that ‘‘[i]ndividuals suffering from the condi- ries. See, e.g., Curtin Decl., Exh. 18 (appli- tions treated with the ‘Heart Condition’ or cant needs to provide ‘‘headach [sic] release ‘Disqualifying Condition’ drugs are potentially TTT [and a] statement that she is not taking explicitly barred from truck driving by 49 [the medication] for depression’’) (‘‘Not Per- C.F.R. § 391.41.’’ Ante at 76–77 & n. 5 Many mitted’’ category); id. (‘‘Sent to Brenda for of the conditions listed in this regulation only review on cardiovasular [sic] condition’’) disqualify an individual if the condition is (‘‘Not Permitted’’ category); Curtin Decl., ‘‘likely to interfere with his/her ability to con- Exh. 20 (‘‘the diagnosis and severity of her trol and drive a commercial motor vehicle condition for which she takes the medication safely,’’ 49 C.F.R. § 391.41(b)(5), or otherwise is considered disqualifying’’) (‘‘Unsafe Ef- suggest that an individualized determination fects’’ category); id. (applicant ‘‘will need to of potential safety concerns is required. See complete his treatments TTT and send in all id. § 391.41(b)(6)-(12). In relying on these records when his condition is resolved’’) regulations to support Hunt’s policy, the ma- (‘‘Unsafe Effects’’ category); Curtin Decl., jority ignores the crucial difference between Exhs. 23, 25 (applicant needs to ‘‘provide a individualized determinations of driver safety statment [sic] that TTT his condition is fine and Hunt’s explicit policy to create a per se w/out the meds’’) (‘‘Disqualifying Condition’’ bar from truck driving with respect to these or ‘‘Heart Condition’’ category); id. (Hunt individuals. Hunt’s policy simply assumes, ‘‘need[ed] all records on [applicant’s] condi- without justification, that these individuals tion’’) (‘‘Disqualifying Condition’’ or ‘‘Heart are unfit to drive trucks.
    367. 82 321 FEDERAL REPORTER, 3d SERIES not take [the medication] and drive’’); drive; a factfinder reasonably could im- Curtin Decl., Exhs. 23, 25 (reviewer ‘‘in- pute these statements to Hunt, even if formed appl[icant] that he could not take these employees were not the ultimate de- the med[ication] on [the] truck’’); Curtin cision makers. Hunt proffers no evidence Decl., Adair Depo. (representative told ap- that these unidentified ‘‘ultimate hiring au- plicant ‘‘it’s illegal to drive a truck with thorities’’ did not share the reviewers’ per- that [medication]’’); Curtin Decl., Manning ceptions or rely upon their statements Depo. (representative told applicant ‘‘it about the applicants’ limitations. Indeed, was illegal for a driver to drive while on Hunt does not argue otherwise; it simply this medication’’). The EEOC has provid- argues that its employees’ statements im- ed sufficient evidence to create a factual plicitly refer only to jobs at Hunt. A fact- issue whether Hunt perceived the appli- finder is certainly allowed to determine cants as broadly limited in their ability to whether the statement that an applicant work as a truck driver. would ‘‘never drive for anybody’’ implicitly The majority explains this evidence by refers only to jobs at Hunt; it is not, stating: however, this Court’s job to do so. In Although a few evaluators’ comments reviewing whether summary judgment is could be more broadly interpreted, there appropriate, this Court does not make fac- is no evidence that Hunt’s reviewers, tual determinations or refuse to credit le- relying on Hunt’s own DRL and drug gitimate inferences based upon the evi- lists to make a judgment on qualification dence presented, but views the evidence in for a position at Hunt, intended to make the light most favorable to the nonmoving an evaluation beyond Hunt’s specific party. See Giordano, 274 F.3d at 746. guidelines. Hunt also argues that the statements of Ante at 76–77. In reviewing a grant of Dr. Cooper should not be imputed to it. summary judgment, however, we do not The EEOC provides significant evidence refuse to credit a broad, but reasonable, that Hunt relied on Dr. Cooper’s advice, interpretation of the evidence. Giordano, including, for example, a reviewer’s state- 274 F.3d at 749–50. Even if this were the ment that the applicant was ‘‘disqualified standard, Hunt’s reviewers stated that one per Dr. Cooper.’’ This suffices to provide applicant would ‘‘never drive for anybody,’’ a direct link between Dr. Cooper’s opin- and made similar statements about many ions regarding applicants and Hunt’s view other applicants; it is difficult to imagine a of the applicants as disabled. clearer statement that the reviewers in- Finally, the majority’s argument that tended to say that the applicants were, in Hunt’s policy should not be imputed to fact, substantially limited in their ability to other companies in determining whether work as a truck driver for any company. the applicants were perceived as disabled The majority asserts that because is immaterial. Contrary to the majority’s Hunt’s reviewers were not the ultimate assertion, this is not a case in which the decision makers, the comments ‘‘simply potential imputation of Hunt’s policy to are not sufficient to indicate that Hunt other companies would result in the appli- thought the applicants were more broadly cants being regarded as ‘‘substantially lim- limited.’’ Ante at 76–77. Again, this is a ited in the major life activity of working matter for the factfinder to decide. only as a result of this imputation.’’ Sut- Hunt’s own employees stated on several ton, 527 U.S. at 493, 119 S.Ct. 2139. It is occasions that applicants were unfit to Hunt’s explicit statement that it believed
    368. CICIO v. DOES 83 Cite as 321 F.3d 83 (2nd Cir. 2003) applicants to be unfit to drive a truck that suit against plan administrator for employ- supports Hunt’s perception of these indi- ee benefits plan, its medical director, and viduals as substantially limited in their others, alleging state law claims arising ability to drive a truck; no potential impu- out of decision to deny preauthorization for tation is required. Thus, the EEOC has medical procedure recommended by treat- provided sufficient evidence that a factfin- ing physician. Defendants removed action der could reasonably conclude that Hunt and moved to dismiss for failure to state regarded the rejected applicants as sub- claim. Widow moved to remand. The Unit- stantially limited in the major life activity ed States District Court for the Eastern of working, because Hunt regarded them District of New York, Joanna Seybert, J., as unfit to be truck drivers. 208 F.Supp.2d 288, adopting the report and recommendation of United States CONCLUSION Magistrate Judge E. Thomas Boyle, grant- ed motion to dismiss on ground of preemp- Because I find ample support in the tion under the Employee Retirement In- record for the assertion that Hunt regard- come Security Act (ERISA), and widow ed the applicants as d substantially limited appealed. The Court of Appeals, Sack, Cir- in the major life activity of working, and cuit Judge, held that: (1) negligent delay thus, the applicants were disabled within and misrepresentation claims were remov- the meaning of the ADA, I respectfully able under complete preemption doctrine; dissent. (2) court had supplemental jurisdiction over medical malpractice claim; (3) negli- , gent delay and misrepresentation claims were subject to dismissal as conflict preempted; and (4) on issue of first im- pression, state law medical malpractice Bonnie CICIO, individually and as Ad- claim brought with respect to a medical ministratrix of the Estate of Car- decision made in the course of prospective mine Cicio, Plaintiff–Appellant, utilization review by a managed care orga- nization or health insurer is not preempted v. under ERISA. John DOES 1–8, Defendants, Affirmed in part, vacated in part, and remanded. Vytra Healthcare, and Brent Spears, M.D., Defendants–Appellees. Calabresi, Circuit Judge, filed an opin- ion dissenting in part. Docket No. 01–9248. United States Court of Appeals, Second Circuit. 1. Removal of Cases O107(9) District court’s denial of a motion to Argued: June 20, 2002. remand is reviewed de novo. Decided: Feb. 11, 2003. 2. Federal Courts O776 As Amended: March 12, 2003. District court’s decision to grant a motion to dismiss for failure to state a Widow, on behalf of herself and her claim is reviewed de novo. Fed.Rules Civ. late husband’s estate, brought state court Proc.Rule 12(b)(6), 28 U.S.C.A.
    369. UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT August Term, 2000 (Argued: June 20, 2001 Decided: September 18, 2002) Docket No. 00-9488 _________________________________________ VALERIE KRIMSTOCK, CHARLES FLATOW, ISMAEL DELAPAZ, CLARENCE WALTERS, JAMES WEBB, MICHAEL ZURLO, and SANDRA JONES, individually and on behalf of all other persons similarly situated, Plaintiffs-Appellants, v. RAYMOND W. KELLY, in his official capacity as Commissioner of the New York City Police Department, PROPERTY CLERK, NEW YORK CITY POLICE DEPARTMENT, and THE CITY OF NEW YORK, Defendants-Appellees. __________________________________________ Before: JACOBS, F.I. PARKER, and SOTOMAYOR, Circuit Judges. Appeal from a judgment of the United States District Court for the Southern District of New York (Michael B. Mukasey, Chief Judge) granting defendants’ motion to dismiss the complaint, which alleges constitutional violations arising from the seizure of motor vehicles by the police. We vacate, holding that plaintiffs’ due process rights were violated by the post- seizure, pre-judgment retention of vehicles absent a prompt opportunity to challenge the probable validity of and justification for that deprivation pendente lite. Vacated and remanded. THOMAS M. O’BRIEN, of counsel, The Legal Aid Society, Criminal Defense Division, New York, New York, for plaintiffs-appellants.
    370. JULIE STEINER, Assistant Corporation Counsel, New York, New York, for Michael D. Hess, Corporation Counsel of the City of New York (Stephen J. McGrath, on the brief), for defendants-appellees. SOTOMAYOR, Circuit Judge: Defendant City of New York seizes the motor vehicles of some of those accused of driving while intoxicated and of committing other crimes for which a motor vehicle could be considered an instrumentality.1 The City maintains possession of these vehicles in the hope of one day gaining title to them by prevailing in civil forfeiture proceedings. These proceedings generally await the resolution of criminal charges and can take months or even years to be finalized. The City claims and plaintiffs-appellants (“plaintiffs”) do not contest that the proceedings are authorized, as is the seizure, by provisions of the City’s Civil Administrative Code that cede to the City title to property found to be an instrumentality of crime. Plaintiffs contend that their inability to challenge, promptly after the vehicles are seized, the legitimacy of and justification for the City’s retention of the vehicles prior to judgment in any civil forfeiture proceeding violates their constitutional rights. We agree. A car or truck is often central to a person’s livelihood or daily activities. An individual must be permitted to challenge the City’s continued possession of his or her vehicle during the pendency of legal proceedings where such possession may ultimately prove improper and where less 1 The New York City Corporation Counsel has delegated to the Property Clerk of the New York City Police Department the authority, under New York City’s Administrative Code, to bring the civil forfeiture actions at issue in this appeal. See Property Clerk v. Covell, 139 Misc. 2d 707, 708-09, 528 N.Y.S.2d 299, 300 (Sup. Ct. N.Y. Co. 1988). We refer to defendants Commissioner of the New York City Police Department, the Property Clerk of the New York City Police Department, and the City of New York, collectively, as the “City.” 2
    371. drastic measures than deprivation pendente lite are available and appropriate. We vacate the judgment of the district court and remand for the court to order, after consultation with the parties, the appropriate injunctive relief. BACKGROUND Plaintiffs challenge the seizure and retention of motor vehicles under a section of the City’s Civil Administrative Code, N.Y.C. Code § 14-140. The City claims and plaintiffs do not contest that the statute authorizes the City’s Property Clerk to take custody, following seizure, of, among other things, “all property . . . suspected of having been used as a means of committing crime or employed in aid or furtherance of crime . . . .” N.Y.C. Code § 14-140(b). Seized property is retained by the Property Clerk of the New York City Police Department until the City either loses a future forfeiture suit or decides not to pursue one and someone claims the seized property. Id. § 14-140(e); 38-A New York City Rules & Regulations (“R.C.N.Y.”) § 12-36. The relevant provision of the Administrative Code states: Where . . . property . . . ha[s] been used as a means of committing crime or employed in aid or in furtherance of crime . . . , a person who so . . . used [or] employed . . . any such . . . property or permitted or suffered the same to be used [or] employed . . . or who was a participant or accomplice in any such act, or a person who derives his or her claim in any manner from or through any such person, shall not be deemed to be the lawful claimant entitled to . . . such . . . property . . . . N.Y.C. Code § 14-140(e)(1). The statute applies to all levels of crime, not just felonies, and to all types of crimes. Moreover, it applies to all property, both real and personal. Under the statute, the City can seize a motor vehicle following an arrest for the state-law charge of driving while intoxicated (“DWI”) or any other crime for which the vehicle 3
    372. could serve as an instrumentality. The arraignment of the defendant in the criminal action concerns only the prosecution of the criminal charge. A defendant charged with DWI does not have a right to a post-arrest hearing to determine whether probable cause existed either for his or her arrest or for the seizure of the vehicle. Unlike a felony charge, for which a “prompt” probable cause hearing must be held or evidence of probable cause must be presented to a grand jury, N.Y. Crim. Proc. § 180.10, a misdemeanor charge of DWI requires no post-arrest determination of probable cause, id. §§ 170.10 et seq.; People v. Green, 96 N.Y.2d 195, 199-200, 726 N.Y.S.2d 357, 361 (2001) (noting, in a DWI case, that whereas the “New York Constitution requires a Grand Jury indictment for felony offenses . . . , misdemeanor charges may be brought on a prosecutor’s information”); In re Robert L.,129 Misc. 2d 742, 744, 493 N.Y.S.2d 970, 972 (Fam. Ct. Bronx Co. 1985) (“There is no provision for a preliminary or probable cause hearing under the [New York] Criminal Procedure Law in misdemeanor cases.”). Moreover, N.Y.C. Code § 14-140 affords the vehicle owner no opportunity for a prompt post-seizure hearing to test probable cause for the vehicle’s seizure. Thus, neither New York criminal procedure nor the City’s civil forfeiture law allows a DWI defendant or the owner of a vehicle driven by a DWI defendant to challenge promptly the legitimacy of the City’s continued custody of the vehicle.2 That challenge may not be made until the City seeks the vehicle’s forfeiture in a separate civil proceeding that could take place months or even years after the seizure. Upon seizing the vehicle, the police issue the arrestee a voucher for the vehicle 2 As noted below, the remedies available to those who have had their property seized under New York state civil forfeiture law (N.Y. C.P.L.R. art. 13-A) do not apply to seizures under N.Y.C. Code § 14-140. Other remedies suggested by the City, specifically a Request for Judicial Intervention and an Article 78 proceeding brought under New York state law, do not provide a prompt retention hearing, as discussed below. 4
    373. and any other seized property. 38-A R.C.N.Y. § 12-32(a). If a claimant makes a formal demand for the return of the vehicle, the City has twenty-five days in which either to initiate a civil forfeiture proceeding under the City’s Administrative Code or to release the vehicle. Id. § 12- 36(a).3 Even if the City chooses to commence a civil forfeiture proceeding within the twenty- five day period, however, the proceeding is commonly stayed until the criminal proceeding concludes. In a forfeiture proceeding, the City “bear[s] the burden of proving by a preponderance of the evidence that [it] is legally justified to continue to retain the property.” Id. § 12-36(b).4 Vehicles belonging to the named members of the putative class in this action were seized by the City between March and May of 1999. The vehicles of six of the seven named plaintiffs—Valerie Krimstock, Charles Flatow, Ismael Delapaz, Clarence Walters, James 3 The applicable rules state that “[i]f a timely demand is made for the return of the property before the forfeiture proceeding is instituted, such proceeding shall be brought no later than . . . within 25 days after the date of demand. If such proceeding is not commenced within this time period, the property clerk shall give written notice to the claimant or the claimant’s representative, at his or her last known address . . . that the property will be returned forthwith to that person.” 38-A R.C.N.Y. § 12-36(a). Absent a demand for the return of the property, the City may commence forfeiture proceedings on its own initiative, often at the conclusion of the criminal proceedings. See id. (“If such proceeding is instituted before the termination of criminal proceedings against the claimant, this subchapter shall not be construed to effect [sic] any right of a party to the forfeiture proceeding to have the forfeiture proceeding stayed for such period as the court may determine.”). 4 This provision and several others found in the rules governing forfeitures under the present statute were adopted as a result of this Court’s determination that certain aspects of the predecessor statute were unconstitutional. See McClendon v. Rosetti, 460 F.2d 111, 114-16 (2d Cir. 1972) (striking down procedures for reclaiming property after termination of criminal proceedings charging crimes unrelated to seized property). As the district court below indicated, the challenge in the instant case presents claims not considered in our previous review of the statute. Krimstock v. Safir, No. 99 Civ. 12041, 2000 WL 1702035, at *4 (S.D.N.Y. Nov. 13, 2000). 5
    374. Webb, and Michael Zurlo—were seized after each was arrested for DWI.5 In some cases, the Property Clerk did not institute civil forfeiture actions against plaintiffs’ vehicles until well over two months after the seizures. For example, Clarence Walters’ car was seized on March 15, 1999. He had never been arrested for any offense before, and his DWI case in criminal court ended on June 1, 1999, when he pleaded guilty to the lesser charge of driving while impaired, a non-criminal violation. N.Y. Veh. & Traf. § 1192(1). Mr. Walters paid a fine, performed community service, and completed a Drinking Driver program that rendered him eligible for restoration of his driver’s license. On June 4, more than two-and-a-half months after the seizure of his vehicle, he was served with a forfeiture complaint. At no time between June 1999 and May 2001, when his vehicle was finally released, was Mr. Walters given an opportunity to challenge the City’s retention of the vehicle. Each of the five other DWI arrestees also pleaded guilty to the lesser charge of driving while impaired. Valerie Krimstock, who at the age of forty-eight had never been arrested before, entered her plea to the lesser charge in September 1999—some four months after she had been served with a forfeiture complaint. It was not until eleven months later, in August 2000, that a judge dismissed the forfeiture action and ordered that her 1995 Toyota, on which she had continued to make monthly payments of $273.00, be returned to her. In the case of Charles Flatow—a retired sales manager whose car was seized on April 3, 1999 in connection with a first-time DWI arrest—the Property Clerk commenced a 5 The state statute prohibits the operation of a motor vehicle if the driver “has .10 of one per centum or more by weight of alcohol in the person’s blood.” N.Y. Veh. & Traf. § 1192(2). A violation is a misdemeanor punishable by a fine or by imprisonment for not more than one year, or both. Id. § 1193(1)(b). 6
    375. forfeiture action on June 15, 1999. Mr. Flatow also pleaded guilty to the lesser charge, paid a fine, and completed the required community service and Drinking Driver program. Yet by December 1999, he still had received no hearing in the forfeiture action and his car remained in police custody. As a result, Mr. Flatow had not been given an opportunity to present evidence that a prescription anti-depressant medication he was taking at the time of the arrest caused the Breathalyzer test to exaggerate the percentage of alcohol in his bloodstream. To take yet another example, the 1995 Plymouth van owned by the seventh named plaintiff, Sandra Jones, was seized in March 1999 when her estranged husband, to whom she had lent the vehicle, was arrested for drug and weapon possession. Even though these charges were later dismissed, the Property Clerk initiated a forfeiture action in May 1999, alleging that Ms. Jones had “consented, suffered or permitted” her vehicle to be used by her husband in the commission of crimes. During the pendency of the action, Ms. Jones continued to make monthly payments of $428.06 on her impounded vehicle. In January 2000, some ten months after the police had seized it, the Property Clerk returned her van. In sum, at the time this suit was filed in December 1999, no court had yet ruled on the legitimacy of the retention of any of plaintiffs’ vehicles, all of which remained in the possession of the police even though the criminal cases underlying the forfeiture actions had concluded and none had resulted in a conviction of a crime that would serve as a predicate for forfeiture.6 6 Judge Jacobs does not subscribe to the four preceding paragraphs which recount and rely upon the circumstances particular to the named plaintiffs in this putative class action. His reasons are as follows: (1) The opinion implicitly generalizes from circumstances of the individual plaintiffs—for example, that this one had never been arrested before, that this one is a retired person, and so forth. The named plaintiffs in a putative class action are frequently selected to be especially sympathetic and appealing, but the rule we make will also govern vehicles seized from persons who are habitual drunks. (2) The fact that the six “DWI arrestees 7
    376. Plaintiffs brought this action under 42 U.S.C. § 1983, alleging violations of the Due Process Clause of the Fourteenth Amendment and seeking a prompt hearing following the seizure of vehicles, at which the City “must demonstrate probable cause that the car was used in furtherance of a crime and that it is necessary that the vehicle remain in the City’s custody until the conclusion of the forfeiture proceeding.”7 Named plaintiffs moved for class certification pursuant to Fed. R. Civ. P. 23(a) and (b)(2) and for a preliminary injunction under Fed. R. Civ. P. 65. The City cross-moved to dismiss the complaint pursuant to Fed. R. Civ. P. 12(b)(6). The district court granted the City’s motion to dismiss pursuant to Fed. R. Civ. P. 12(b)(6). Krimstock, 2000 WL 1702035, at *7. The court rejected the City’s request to dismiss plaintiffs’ constitutional claims on abstention grounds, finding that the civil forfeiture proceedings under the New York City Administrative Code did not provide an adequate forum for raising these claims. Id. at *3.8 Applying the balancing test set forth in Mathews v. Eldridge, 424 U.S. 319 (1976), to determine what procedural safeguards are required to satisfy the Due Process Clause of the Fourteenth Amendment in the circumstances of this case, the district court . . . pleaded guilty to the lesser charge of driving while impaired” does not remotely call into question the arresting officer’s probable cause to arrest them for drunk driving. It goes without saying that plea bargains in misdemeanor cases are offered for many reasons other than factual innocence or even insufficiency of evidence. 7 Even though plaintiffs invoked Fourteenth Amendment protections in their complaint, their claims involve both the City’s probable cause to believe their vehicles were subject to seizure and forfeiture (a Fourth Amendment concern, as applied through the Fourteenth Amendment) and the City’s justification for retaining their vehicles pendente lite (a Fourteenth Amendment concern). We address both issues. 8 The court did abstain from hearing plaintiffs’ claim that they had a due process right to a meaningful opportunity to request court-appointed counsel. Krimstock, 2000 WL 1702035, at *3. Plaintiffs have not raised this issue on appeal, and we therefore deem it abandoned. 8
    377. held that plaintiffs were not entitled to prompt post-seizure hearings on the question of probable cause or the legitimacy of retention pendente lite. Id. at *6-*7. Specifically, the court held that “plaintiffs’ due process right to a meaningful hearing at a meaningful time does not require the additional safeguard of a probable cause hearing.” Id. at *7. The court concluded that the plaintiffs’ interests were adequately protected by “a probable cause arrest” and the eventual forfeiture proceeding. Id. at *6-*7. This appeal followed. DISCUSSION A federal court may dismiss a complaint for “failure to state a claim upon which relief can be granted.” Fed. R. Civ. P. 12(b)(6). “We review de novo a district court’s dismissal of a complaint pursuant to Rule 12(b)(6), accepting all factual allegations in the complaint as true and drawing all reasonable inferences in the plaintiff’s favor.” Ganino v. Citizens Utils. Co., 228 F.3d 154, 161 (2d Cir. 2000). Under Rule 12(b)(6), we uphold a district court’s dismissal only if “it appears beyond doubt that the plaintiffs can prove no set of facts in support of his [or her] claim which would entitle him [or her] to relief.” Id. (internal citation omitted). Our primary focus today is the City’s continued retention of vehicles after their warrantless seizure by the police and prior to the ultimate resolution of the forfeiture action in court. It is this intermediate deprivation, lasting months or sometimes years without any prompt hearing before a neutral fact-finder, that we deem constitutionally infirm. In the absence of a showing that continued impoundment constitutes a valid deprivation, seized vehicles must be released during the pendency of civil proceedings. We reach this conclusion in light of the dictates of the Fourth and Fourteenth Amendments. In Part I, we establish a framework for analyzing plaintiffs’ challenge to the 9
    378. probable validity of the City’s post-seizure, pre-judgment retention of their vehicles, and we define “probable validity” as a due process concept that in the present case embraces the City’s probable cause for initially seizing the vehicles. In Parts II and III, we discuss the applicability of the Fourth and Fourteenth Amendments to seizure and retention of personal property under civil forfeiture laws. In the course of discussing the Fourteenth Amendment, we give special attention to three areas of due process concern raised by N.Y.C. Code § 14-140 as applied to the present facts: the temporal gap that typically exists between vehicle seizure and the eventual forfeiture proceeding; the plight of innocent owners;9 and the inadequacy of the remedies suggested by the City for addressing the propriety of continued retention of vehicles seized under § 14-140. In Part IV, we employ the three-factor inquiry prescribed by the Supreme Court in Mathews v. Eldridge to determine what procedural safeguards are required to satisfy the Due Process Clause in this case. Finding that plaintiffs’ right to due process has been violated, we conclude in Part V by offering general guidance as to the prompt post-seizure retention hearing that we deem to be constitutionally required. I. The Probable Validity of Continued Deprivation of Vehicles Plaintiffs in this action essentially seek an early opportunity to test the City’s likelihood of success on the merits of the forfeiture action, or what the Supreme Court has termed the “probable validity” of continued deprivation of a claimant’s property during the 9 Here and throughout this opinion, we use the phrase “innocent owner” as a term of art denoting a person who has an ownership interest in property threatened with civil forfeiture but who neither participated in nor permitted or suffered the alleged illegal use of the property, and persons who claim that status. Our use of the term is not intended to suggest that drivers or other persons initially charged with a vehicle-related crime are not presumed innocent until proven guilty in a criminal proceeding or are presumptively unlikely to prevail in a civil forfeiture proceeding. 10
    379. pendency of legal proceedings. Cf. Comm’r v. Shapiro, 424 U.S. 614, 629 (1976) (“[A]t least where irreparable injury may result from a deprivation of property pending final adjudication of the rights of the parties, the Due Process Clause requires that the party whose property is taken be given an opportunity for some kind of predeprivation or prompt post-deprivation hearing at which some showing of the probable validity of the deprivation must be made.”); Fuentes v. Shevin, 407 U.S. 67, 97 (1972) (“Since the essential reason for the requirement of a prior hearing is to prevent unfair and mistaken deprivations of property, . . . it is axiomatic that the hearing must provide a real test. ‘(D)ue process is afforded only by the kinds of “notice” and “hearing” that are aimed at establishing the validity, or at least the probable validity, of the underlying claim against the alleged debtor before he can be deprived of his property.’”) (quoting Sniadach v. Family Fin. Corp., 395 U.S. 337, 343 (1969) (Harlan, J., concurring)). For present purposes, we treat probable validity as a comprehensive due process concept that includes the City’s probable cause for initially seizing vehicles. Clearly, the legality of a warrantless seizure is a component of the larger question of the probable validity of continued retention of a seized vehicle. If a seizure lacked probable cause, and the City could offer no untainted post-seizure evidence to justify further retention, the claimant’s vehicle would ordinarily have to be released during the pendency of proceedings. Cf. Marine Midland Bank, N.A. v. United States, 11 F.3d 1119, 1125 (2d Cir. 1993) (“[C]ourts in this circuit have ordered the return of seized property before the commencement of a [federal] forfeiture trial on the ground that the government lacked probable cause to seize the property at the time of the seizure.”). Although there is an obvious overlap between probable cause for a seizure and the 11
    380. probable validity of a retention, the two are not necessarily coextensive. For example, at a retention hearing, the City might succeed in showing that police officers had probable cause for seizing the vehicle of a DWI arrestee, yet be unable to establish the probable validity of continued deprivation pendente lite in the face of proof of innocent ownership or evidence that the Breathalyzer test had registered inaccurate results. Similarly, the City might establish probable cause for a seizure but fail to persuade the court that its interest in the accused instrumentality would not be protected by measures less drastic than continued deprivation. Conversely, the City might fail to establish probable cause for an initial seizure yet be able to offer post-seizure evidence showing the probable validity of retention during the pendency of proceedings. Thus, the legality of a seizure typically will be a subset of the larger due process question of the legitimacy of continued impoundment pendente lite. II. The Role of the Fourth Amendment in Civil Forfeiture The Supreme Court has held that the Fourth Amendment protects claimants against unreasonable seizures of their property in the civil forfeiture context. See United States v. James Daniel Good Real Prop., 510 U.S. 43, 49 (1993) (“The Fourth Amendment does place restrictions on seizures conducted for purposes of civil forfeiture . . . .”); see also One 1958 Plymouth Sedan v. Pennsylvania, 380 U.S. 693, 696, 700 (1965) (holding that the exclusionary rule under the Fourth Amendment applies to civil forfeiture proceedings); cf. In re Seizure of All Funds in Accounts in Names Registry Publ’g, Inc., 68 F.3d 577, 580 (2d Cir. 1995) (“In order to seize property under [federal civil forfeiture law], the government must demonstrate that there was probable cause to believe that the property is subject to forfeiture.”); United States v. Daccarett, 6 F.3d 37, 49 (2d Cir. 1993) (“[T]his circuit requires seizures made pursuant to 12
    381. [federal civil forfeiture law] to comport with the fourth amendment.”); United States v. $37,780 in U.S. Currency, 920 F.2d 159, 162 (2d Cir. 1990) (same). The Supreme Court has not said that a probable cause hearing is required after a warrantless seizure of property and before trial of a government’s claim to title under a civil forfeiture law. Yet many of our laws are fashioned to ensure that a prompt hearing before a neutral judicial or administrative officer is held after the warrantless seizure of property and before full adjudication of the merits of a claim. One example is the federal civil forfeiture regime as administered by the courts of this Circuit in the context of seizures of illicit proceeds and instrumentalities. As this Court has recognized, “the seizure and forfeiture of property are two distinct events under the [federal] civil forfeiture laws.” Marine Midland Bank, 11 F.3d at 1124 (citing Daccarett, 6 F.3d 37 at 46).10 “While both events require the government to have probable cause, the government is not required to demonstrate probable cause until the forfeiture trial unless a claimant challenges the validity of the seizure before trial.” Id. (emphasis added). If the government, once challenged, cannot establish probable cause for the initial seizure or offer post-seizure evidence to justify continued impoundment, retention of the seized property runs afoul of the Fourth Amendment. Id. at 1125; see also United States v. U.S. Currency in Amount of One Hundred Forty-Six Thousand, Eight Hundred Dollars, No. 96-CV-4882, 1997 WL 10 Marine Midland Bank involved the seizure of interbank funds under the federal civil forfeiture laws and upon a warrant issued by a federal magistrate judge, who had determined that the government had probable cause to believe that interbank accounts were being used to launder narcotics proceeds. Marine Midland Bank, 11 F.3d at 1121. The banks moved for return of the seized funds by order to show cause pursuant to Fed. R. Crim. P. 41(e). Holding that “[t]he magistrate judge’s probable cause determination should be subject to judicial review,” this Court found that the warrant had not been supported by probable cause and therefore ordered the release of certain of the interbank funds. Id. at 1125. 13
    382. 269583, at *3 (E.D.N.Y. Apr. 28, 1997) (“[I]f the claimant [in a federal civil forfeiture action] properly raises the issue of the government’s probable cause for seizure before the forfeiture trial, and if the claimant demonstrates that the government lacked probable cause at the time of seizure, the property may be returned to the claimant until the forfeiture trial is held.”). We recognize that the likelihood of illegal seizure is reduced in the context of DWI arrests and that the City’s burden of proving probable cause in such cases is not onerous.11 We cannot agree with the district court, however, that a warrantless arrest is sufficient by itself to ensure the legality of the initial seizure. Some risk of erroneous seizure exists in all cases, and in the absence of prompt review by a neutral fact-finder, we are left with grave Fourth Amendment concerns as to the adequacy of an inquiry into probable cause that must wait months or sometimes years before a civil forfeiture proceeding takes place. Our concerns are heightened by the fact that the seizing authority in this case “has a direct pecuniary interest in the outcome of the proceeding.” James Daniel Good Real Prop., 510 U.S. at 55-56; see also Property Clerk v. Hyne, 147 Misc. 2d 774, 780, 557 N.Y.S.2d 244, 248 (Sup. Ct. N.Y. Co. 1990) (noting that § 14- 140’s “remedial purposes” include the fact that “revenue is generated and applied toward the cost of law enforcement”), aff’d, 171 A.D.2d 506, 567 N.Y.S.2d 603 (1st Dep’t 1991). 11 See People v. Earley, 244 A.D.2d 769, 770-71, 666 N.Y.S.2d 223, 224 (3d Dep’t 1997) (holding that arresting officer’s testimony at suppression hearing supported finding of probable cause to arrest defendant for DWI, despite contradictory testimony of defendant and his passenger and despite minor inconsistencies in officer’s testimony); Grinberg v. Safir, 181 Misc. 2d 444, 455, 694 N.Y.S.2d 316, 325 (Sup. Ct. N.Y. Co.) (“The seizure is simultaneous with a DWI arrest for which the police must have probable cause. The arresting officer evaluates an offense committed in his or her presence. Indicia of alcohol consumption and objective tests confirming the presence of alcohol minimize the risk of erroneous deprivation.”), aff’d, 266 A.D.2d 43, 698 N.Y.S.2d 218 (1st Dep’t 1999); see also People v. Mathison, 287 A.D.2d 384, 384, 732 N.Y.S.2d 2, 2-3 (1st Dep’t 2001) (noting that probable cause may be demonstrated by reliable hearsay, including information provided by fellow officer present at the arrest scene). 14
    383. III. The Role of the Fourteenth Amendment in Civil Forfeiture The government’s seizure and retention of property under civil forfeiture statutes, in the absence of a meaningful hearing at a meaningful time, raise serious due process concerns under the Fourteenth Amendment. See James Daniel Good Real Prop., 510 U.S. at 62 (holding that, absent exigent circumstances, “the Due Process Clause requires the Government to afford notice and a meaningful opportunity to be heard before seizing real property subject to civil forfeiture”); Fuentes, 407 U.S. at 80 (holding, in a case involving state prejudgment replevin statutes that permitted seizure of chattels without a prior opportunity to be heard, that it is “fundamental that the right to notice and an opportunity to be heard must be granted at a meaningful time and in a meaningful manner”) (quotation marks omitted). The fundamental right to notice and a meaningful hearing at a meaningful time has been recognized in many different contexts. See, e.g., James Daniel Good Real Prop., 510 U.S. at 43 (seizure of real property under federal forfeiture law); Connecticut v. Doehr, 501 U.S. 1 (1991) (state ex parte attachment procedures); Memphis Light, Gas & Water Div. v. Craft, 436 U.S. 1 (1978) (termination of municipal utility service); N. Ga. Finishing, Inc. v. Di-Chem, Inc., 419 U.S. 601 (1975) (prejudgment garnishment of bank account); Fuentes, 407 U.S. at 67 (state prejudgment replevin statutes); Sniadach, 395 U.S. at 337 (state wage-garnishment procedure). Due process is inevitably a fact-intensive inquiry. Doehr, 501 U.S. at 10 (“[D]ue process, unlike some legal rules, is not a technical conception with a fixed content unrelated to time, place and circumstances.”) (quotation marks omitted). The “timing and nature of the required hearing will depend on appropriate accommodation of the competing interests involved.” Logan v. Zimmerman Brush Co., 455 U.S. 422, 434 (1982) (quotation marks omitted). 15
    384. A. Temporary Deprivations of Property Pendente Lite Temporary deprivation of real or personal property pendente lite in a forfeiture action must satisfy the demands of the Fourteenth Amendment. The Supreme Court has held that even a brief and provisional deprivation of property pending judgment is of constitutional importance. See Fuentes, 407 U.S. at 84-85 (“[I]t is now well settled that a temporary, nonfinal deprivation of property is nonetheless a ‘deprivation’ in the terms of the Fourteenth Amendment.”); see also United States v. Monsanto, 924 F.2d 1186, 1192 (2d Cir. 1991) (noting that a “temporary and nonfinal” removal of a defendant’s assets, pursuant to a federal criminal forfeiture statute and pending resolution of the criminal case, “is, nonetheless, a deprivation of property subject to the constraints of due process”) (quotation marks omitted). Plaintiffs here have not challenged the procedural safeguards under New York law that guarantee the accuracy of any final judgment of forfeiture. Instead, they question the legitimacy of and justification for the intermediate deprivation of their property occasioned after seizure of the vehicle and before judgment in civil forfeiture proceedings under N.Y.C. Code § 14-140, and, indeed, before those proceedings are even commenced. See James Daniel Good Real Prop., 510 U.S. at 56 (“The question in the civil forfeiture context is whether ex parte seizure is justified by a pressing need for prompt action.”); Fuentes, 407 U.S. at 80-81 (stating that due process is intended “to minimize substantively unfair or mistaken deprivations of property”). The district court in this case collapsed the separate issues of probable cause and due process into a single analysis and, applying the test for due process set forth in Mathews v. Eldridge, 424 U.S. 319 (1976), concluded that plaintiffs had alleged no facts to suggest that a “probable cause arrest” is a “procedure” that is “unusually unreliable,” Krimstock, 2000 WL 16
    385. 1702035, at *6, and further concluded that “plaintiffs’ due process right to a meaningful hearing at a meaningful time does not require the additional safeguard of a probable cause hearing,” id. at *7. In reaching this determination, the court applied the “speedy trial” test as deployed in the federal customs case of United States v. $8,850, 461 U.S. 555 (1983), and held that plaintiffs’ due process interests are fully protected by the eventual forfeiture proceeding. Id.12 The district court’s analysis resembles the approach taken by the New York Supreme Court in Grinberg v. Safir, in which a DWI arrestee brought an Article 78 proceeding to contest the City’s seizure of his 1988 Acura for forfeiture. In response to Grinberg’s Fourth Amendment challenge to the seizure and retention of his vehicle, the court, citing various warrantless arrest and seizure exceptions, held that “[o]nce an object is permissibly seized as an instrumentality during an arrest, no warrant, pretrial hearing or judicial approval is needed for retention during the criminal action.” Grinberg, 181 Misc. 2d at 452, 694 N.Y.S.2d at 323. The court also found that Grinberg’s Fourteenth Amendment right to a meaningful hearing at a meaningful time had not been violated. Citing the speedy trial test as applied in $8,850, the court reasoned that “[i]f pendency of a criminal action is a legitimate reason for the delayed filing of a forfeiture proceeding, then retention of the subject vehicle without a hearing, while the criminal 12 The district court also cited United States v. Von Neumann, 474 U.S. 242, 249 (1986), in declaring that “‘the forfeiture proceeding, without more, provides the postseizure hearing required by due process.’” Krimstock, 2000 WL 1702035, at *5. The Supreme Court in Von Neumann, however, was addressing the different issue of what process was due in proceedings for remission or mitigation under U.S. customs laws when a claimant could challenge the seizure of his or her property in judicial forfeiture proceedings. Von Neumann, 474 U.S. at 249-50. We note that under the customs laws applicable in Von Neumann, the claimant could file a motion under Fed. R. Crim. P. 41(e) for return of the seized vehicle if he or she “believe[d] the initial seizure was improper,” id. at 244 n.3, and that, in the case of Von Neumann, Customs released the claimant’s vehicle after he had posted a bond pursuant to 19 U.S.C. § 1614, id. at 246. 17
    386. action is pending, is also permissible.” Grinberg, 181 Misc. 2d at 456, 694 N.Y.S.2d at 326. For reasons discussed more fully below, we disagree with these courts’ conclusions. Contrary to the district court’s determination in the present case, a warrantless arrest by itself does not constitute an adequate, neutral “procedure” for testing the City’s justification for continued and often lengthy detention of a vehicle which may be owned by the arrestee or by someone entirely unconnected with the conduct that gave rise to the arrest. Further, to say that the forfeiture proceeding, which often occurs more than a year after a vehicle’s seizure, represents a meaningful opportunity to be heard at a meaningful time on the issue of continued impoundment is to stretch the sense of that venerable phrase to the breaking point. We also consider it a non sequitur to hold, as the Grinberg court did, that because postponing the commencement of a forfeiture action pending the underlying criminal proceeding may not offend due process, retention of the seized vehicle without a hearing throughout that same period, or longer, is constitutionally permissible. The issues of a speedy trial and a prompt retention hearing are not parallel in this context, particularly when less restrictive methods for protecting the City’s interest in the allegedly offending res are available. Cf. Lee v. Thornton, 538 F.2d 27, 32 (2d Cir. 1976) (“Deprivation of means of transportation for [substantial] periods requires an opportunity to be heard.”); DeBellis v. Property Clerk, 79 N.Y.2d 49, 57, 580 N.Y.S.2d 157, 161 (1992) (“The core principle of the Second Circuit’s McClendon decision is that, although the government may seize and hold a citizen’s property for a variety of reasons in connection with a criminal or related proceeding, once those proceedings have terminated or it is determined that the property is not related to or is otherwise not needed for those proceedings, due process requires that the property be returned upon demand unless the government can 18
    387. establish a new basis for its detention.”). In sum, just as in the attachment and seizure cases cited above, the purpose of requiring due process in the present circumstances “is not only to ensure abstract fair play to the individual,” but “more particularly, . . . to protect his [or her] use and possession of property from arbitrary encroachment—to minimize substantively unfair or mistaken deprivations of property.” James Daniel Good Real Prop., 510 U.S. at 53 (quotation marks and citation omitted). We conclude that plaintiffs have a right under the Fourteenth Amendment to ask what “justification” the City has for retention of their vehicles during the pendency of proceedings, cf. id. at 56, 61, and to put that question to the City at an early point after seizure in order to minimize any arbitrary or mistaken encroachment upon plaintiffs’ use and possession of their property. B. Special Due Process Concerns in the Present Case Our concern that plaintiffs be provided an early opportunity to test the propriety of the City’s retention of their vehicles, after seizure pursuant to N.Y.C. Code § 14-140 and prior to any eventual civil forfeiture judgment, is heightened by several factors. These factors include the temporal gap that typically exists between seizure of the vehicle and the forfeiture proceeding, the inability of innocent owners to challenge promptly the City’s retention of their vehicles pendente lite, and the inadequacy of other suggested remedies for providing prompt post-seizure review under New York’s administrative and civil codes. 1. The Temporal Gap Between Seizure and Forfeiture Proceedings N.Y.C. Code § 14-140 and the applicable rules leave a significant temporal gap between the moment a vehicle is seized and the time the City commences forfeiture proceedings. 19
    388. Although it is possible for the City to initiate a forfeiture action earlier, it need not bring such an action until twenty-five days after a claim is made for the vehicle. 38-A R.C.N.Y. § 12-36(a) (“If a timely demand is made for the return of the property before the forfeiture proceeding is instituted, such proceeding shall be brought no later than . . . within 25 days after the date of demand.”).13 If no demand is made, the Property Clerk may initiate the action at its discretion. In the present case, forfeiture proceedings were commenced, at the earliest, three weeks after seizure of a vehicle, and, at the latest, well over two months after seizure. Thus, there typically exists a significant period after seizure and before the filing of the forfeiture action when the City is not held responsible for the legality of the warrantless seizure or the continued retention of the vehicle. The period between the seizure and the holding of a hearing in the forfeiture action is, of course, considerably longer.14 It can take months or even years. 13 The 25-day rule under 38-A R.C.N.Y. § 12-36(a) represents a legislative expansion of the time period originally ordered in McClendon v. Rosetti, 369 F. Supp. 1391, 1394 (S.D.N.Y. 1974), which required the Property Clerk to initiate a forfeiture proceeding within 10 days after a claimant’s timely demand. 14 In contrast, while the statute here requires only that a forfeiture proceeding be initiated within 25 days after a claim is made, the California controlled substances act requires that, once a person claiming an interest in seized property files a verified claim, “the [civil] forfeiture proceeding shall be set for hearing on a day not less than 30 days therefrom, and the proceeding shall have priority over other civil cases.” Cal. Health & Safety Code § 11488.5(c)(1) (emphasis added). Similarly, the civil forfeiture provision of the Wisconsin controlled substances act states that “the action shall be set for hearing within 60 days of the service of the answer but may be continued for cause or upon stipulation of the parties.” Wis. Stat. § 961.555(2)(b); see also State v. Rosen, 240 N.W.2d 168, 171-72 (Wis. 1976) (holding that the 60-day rule under the Wisconsin controlled substances act is mandatory and that failure to set the hearing within that period deprived the trial court of jurisdiction). In South Carolina, when a motorist is convicted of a third or subsequent DUI violation within 10 years, the vehicle may be seized and forfeited, but the vehicle’s registered owner, who must be given notice within 72 hours of the confiscation, has 10 days to request a judicial hearing to determine issues concerning the owner’s knowledge and authorization. “The hearing must be held within ten days from the date of receipt of the request.” S.C. Code Ann. § 56-5-6240(A). 20
    389. Many state forfeiture statutes, unlike N.Y.C. Code § 14-140, provide an early opportunity to challenge the governmental authority’s probable cause for seizing property or the legitimacy of its retaining seized property during the pendency of proceedings. Florida’s contraband forfeiture statute is one example. In upholding the Florida statute in a case involving police seizure of a vehicle from a public place, the U.S. Supreme Court observed that, although the police had not needed to obtain a warrant to seize the vehicle, the statute required that “‘the person entitled to notice is notified at the time of the seizure . . . that there is a right to an adversarial preliminary hearing after the seizure to determine whether probable cause exists to believe that such property has been or is being used in violation of the Florida Contraband Forfeiture Act.’” Florida v. White, 526 U.S. 559, 562 n.2 (1999) (quoting Fla. Stat. § 932.703(2)(a)). The Florida statute further provides: Seizing agencies shall make a diligent effort to notify the person entitled to notice of the seizure. Notice provided by certified mail must be mailed within 5 working days after the seizure and must state that a person entitled to notice may request an adversarial preliminary hearing within 15 days after receiving such notice. . . . The seizing agency shall set and notice the hearing, which must be held within 10 days after the request is received or as soon as practicable thereafter. Fla. Stat. § 932.703(2)(a); see also Cochran v. Harris, 654 So. 2d 969, 972 (Fla. Dist. Ct. App. 1995) (holding that a delay of twenty-three days beyond the ten-day limit for a hearing under Some legislatures have sought to mitigate the depreciation of property values and other costs incurred through delays of civil forfeiture proceedings. See, e.g., Ariz. Rev. Stat. § 13- 4310(I) (“Before staying civil discovery [in a forfeiture proceeding until a related criminal trial is concluded], the court shall make adequate provision to prevent any loss or expense to any victim or party resulting from the delay, including loss or expense due to maintenance, management, insurance, storage or preservation of the availability of the property or due to depreciation in the value of the property.”). 21
    390. § 932.703 violated the claimants’ right to due process); cf. Ariz. Rev. Stat. § 13-4310(B) (providing that, upon timely application by an owner of or interest holder in property threatened with forfeiture, the court “may issue an order to show cause to the seizing agency for a hearing on the sole issue of whether probable cause for forfeiture of the property then exists”); Cal. Health & Safety Code § 11488.4(h) (providing that “[i]f there is an underlying or related criminal action, a defendant may move for the return of the property [threatened with civil forfeiture] on the grounds that there is not probable cause to believe that the property is forfeitable . . . .”). Nothing like the procedural safeguards contained in the Florida contraband forfeiture act and similar state statutes is built into the New York forfeiture law. In addition, many state statutes afford avenues of interim relief for claimants who are adversely affected by seizure and retention of property. For example, the Florida contraband forfeiture statute provides that if the court determines that probable cause exists to seize property, “the court shall order the property restrained by the least restrictive means to protect against disposal, waste, or continued illegal use of such property pending disposition of the forfeiture proceeding.” Fla. Stat. § 932.703(2)(d). These means include “a bond or other adequate security equivalent to the value of the property.” Id.; cf. Ariz. Rev. Stat. § 13-4306(G) (“An owner of property seized for forfeiture may obtain the release of the seized property by posting . . . a surety bond or cash . . . .”); Cal. Health & Safety Code § 11492(c) (providing for various remedies to preserve the status quo pendente lite, including “a surety bond or undertaking to preserve the property interests of the interested parties”); N.J. Stat. Ann. § 2C:64-3(g) (providing that persons with an interest in property seized for forfeiture, except defendants prosecuted in connection with the seized property, may, after posting a bond, secure release of the property pending the 22
    391. forfeiture action). Again, no protections for a claimant’s practical interests in seized property are provided for under the New York forfeiture law. 2. Seizure of Property of Innocent Owners With respect to innocent owners, the City’s authority to seize property may be broader than its authority to cause the forfeiture of the property. In the due process context, the Supreme Court has shown special concern for the risk of erroneous deprivation posed to innocent owners. See James Daniel Good Real Prop., 510 U.S. at 55.15 The impact of N.Y.C. Code § 14- 140 on innocent owners is vividly illustrated by the predicament of plaintiff Sandra Jones, whose Plymouth van, which she had lent to her estranged husband, was seized in connection with her husband’s arrest on drug and weapon charges. Although these charges were later dismissed, Ms. Jones was deprived of her vehicle for some ten months while continuing to make monthly auto payments on the vehicle. Ms. Jones was given no early opportunity to test the probable validity of the City’s continued impoundment of her vehicle. The forfeiture provision operates against those persons who “shall not be deemed 15 In Bennis v. Michigan, 516 U.S. 442 (1996), the Supreme Court held that a state nuisance-abatement law under which authorities sought the forfeiture of a vehicle used for illegal sexual activity did not violate the Due Process Clause of the Fourteenth Amendment by not permitting an innocent co-owner of the vehicle to contest the abatement. Id. at 453. Unlike the federal forfeiture law at issue in James Daniel Good Real Property, however, the statute in Bennis did not provide for an affirmative defense of innocent ownership. Moreover, the Court in Bennis suggested that the apparent unfairness of the Michigan law was mitigated by the trial court’s “remedial discretion” to consider whether forfeiture of a vehicle would leave its owner or owners without transportation and to order payment of one-half of any sale proceeds to an innocent co-owner. Id. at 444-45, 453. We also note that the nuisance-abatement statute in question, Mich. Comp. Laws § 600.3825, authorized “removal” and “sale” of property upon entry of an order of abatement only after the property had been deemed a nuisance in a civil action. Id. at 444 n.3. Nothing on the face of the provision, or in the Bennis case, suggests that the statute permitted seizure and retention of property prior to adjudication of its status as a nuisance. 23
    392. to be the lawful claimant” to the property that has been seized by the police department. N.Y.C. Code § 14-140(e)(1).16 The statute identifies two principal groups of those who are not “lawful claimant[s]” and whose property therefore is forfeitable. One is the person who has “used” the property “as a means of committing crime or employed [it] in aid or in furtherance of crime”; the other is the person who “permitted or suffered the same to be used or employed.” Id.17 Thus, the seizure provision authorizes the Property Clerk to take custody, following seizure, of “all property or money suspected of having been used as a means of committing crime,” N.Y.C. Code § 14-140(b), without regard to whether or not an owner who took no part in the crime “permitted or suffered” the vehicle to be used as an instrumentality of 16 This cumbersome construction, defining those subject to forfeiture proceedings in terms of who is not a “lawful claimant” to the property, appears to reflect the former practice under which the burden rested on a claimant to bring a civil action in replevin to obtain the return of seized property. See Hofferman v. Simmons, 290 N.Y. 449, 454 (1943) (“Since these are replevin actions, we concern ourselves not so much with the [City’s] right to hold as with the [claimant’s] right to recapture.”). In response to a constitutional challenge to the statute, the burden of proof was shifted from the claimant to establish a lawful entitlement, to the City to establish the forfeitability of the property. McClendon, 369 F. Supp. at 1394. 17 The relevant portion of the section provides: Where moneys or property . . . have been used as a means of committing crime or employed in aid or in furtherance of crime or held, used or sold in violation of law, . . . a person who . . . so used, employed, sold or held any such moneys or property or permitted or suffered the same to be used, employed, sold or held . . . or a person who derives his or her claim in any manner from or through any such person, shall not be deemed to be the lawful claimant entitled to any such moneys or property . . . . N.Y.C. Code § 14-140(e)(1). The statute also deems not a lawful claimant any “person who derives his or her claim in any manner from or through any such person.” Id. This formulation evidently could include other potentially innocent owners and interest holders, such as secured lenders, mortgagees, and lienholders, as well as purchasers, inheritors, and other transferees of property interests. 24
    393. the crime.18 A statute that authorizes the police to seize property to which the government has not established a legal right or claim, and that on its face contains no limitation of forfeiture liability for innocent owners, raises substantial constitutional concerns.19 Because plaintiffs in 18 It is not entirely clear whether the City has the burden of proving that the owner “permitted or suffered” a vehicle to be used as an instrumentality of crime, or whether innocent ownership is an affirmative defense to be raised by the claimant. Without deciding that question, we note that we have found one New York state court opinion that has held that the City has the burden. See Property Clerk v. Pagano, 170 A.D.2d 30, 35, 573 N.Y.S.2d 658, 661 (1st Dep’t 1991) (“[A] proceeding brought under § 14-140 differs substantially from one brought pursuant to a forfeiture statute which expressly places the burden on a claimant to establish his innocent state of mind. Clearly, § 14-140 contains no such language and, on the contrary, must be construed by its own terms to place an affirmative burden on [the City] . . . of proving that respondent ‘permitted or suffered’ the illegal use of the property.”). 19 Many state statutes, while permitting seizure of forfeitable property, are much more explicit than N.Y.C. Code § 14-140 in limiting or exempting the forfeiture liability of innocent owners. These statutes generally provide for affirmative defenses by innocent owners or make proof of the owners’ culpable knowledge part of the seizing authority’s burden. For example, a Maine statute that permits forfeiture of a motor vehicle upon a defendant’s simultaneous conviction for operating under the influence (“OUI”) and operating after suspension for a prior OUI conviction, requires that the defendant be the “sole owner-operator of that vehicle.” Me. Rev. Stat. Ann. tit. 29-A, § 2421(1); see also State v. One Blue Corvette, 732 A.2d 856 (Me. 1999) (holding that the statutory requirement that defendant be the “sole owner-operator of that vehicle” precluded forfeiture of a vehicle jointly owned by defendant’s wife). Other state statutes contain express exceptions for a variety of innocent parties: Ariz. Rev. Stat. § 13-4304(1)-(2), (4)-(5) (common carriers; owners whose vehicles were “unlawfully in the possession of a person other than the owner” during commission or omission of an act; owners who “did not know and could not reasonably have known of the act or omission”; and bona fide purchasers for value “not knowingly taking part in an illegal transaction” and “reasonably without notice of the act or omission giving rise to forfeiture”); Cal. Health & Safety Code §§ 11470(e)-(h) (persons other than defendant with a community property interest in the vehicle; bona fide purchasers and encumbrancers), 11488.5(e) (owners who did not have “actual knowledge that the seized property would be or was used for a purpose for which forfeiture is permitted and consented to that use”); Fla. Stat. § 932.703(6)-(7) (spouses, lienholders, lessors, joint owners, and other parties lacking knowledge of the criminal activity alleged); N.J. Stat. Ann. § 2C:64-5(a)-(b) (unwitting lessors and security-interest holders; owners who show that they were “not involved in or aware of the unlawful activity and . . . had done all that could reasonably be expected to prevent the proscribed use of the property by an agent”); La. Rev. Stat. Ann. § 14:98(D)(2)(b) (unwitting owners; owners of stolen vehicles); Minn. Stat. § 169A.63(7) (unwitting owners, bona fide security-interest holders, and lessors). Notably, New York state civil forfeiture statutes 25
    394. this action seek only a prompt and effective means to test the legitimacy of and justification for the City’s retention of their vehicles following the seizure but prior to the forfeiture proceeding, we have no occasion to rule on the constitutionality of the seizure provision itself. Nevertheless, the scope of the police seizure authority granted under § 14-140(b), together with the City’s “direct pecuniary interest in the outcome of the proceeding,” James Daniel Good Real Prop., 510 U.S. at 55-56, further convinces us of the need to provide a prompt opportunity for independent and neutral judicial review of the probable validity of the City’s retention of vehicles pendente lite.20 likewise provide a defense for innocent owners. See N.Y. C.P.L.R. 1311(4)(d)(ii)-(iii) (innocent persons subject to “adverse impact of a forfeiture”; defendants acquitted of the underlying crime), 1311(3)(b)(iv) (non-criminal defendants who either did not know “that the instrumentality was or would be used in the commission of a crime” or did not knowingly obtain an “interest in the instrumentality to avoid forfeiture”). In Property Clerk v. Ferris, 77 N.Y.2d 428, 431, 568 N.Y.S.2d 577, 579 (1991), the New York Court of Appeals stated that the “interests of justice” exception in N.Y. C.P.L.R. 1311 “is ‘unique’ . . . and nothing in the article suggests that it applies in the limited forfeiture proceedings available pursuant to Administrative Code § 14-140”; see also Property Clerk v. Deans Overseas Shippers, Inc., 275 A.D.2d 204, 205-06, 712 N.Y.S.2d 492, 493 (1st Dep’t 2000) (same). Moreover, in contrast to § 14-140, the primary federal civil drug forfeiture statute, 21 U.S.C. § 881, contained an express “innocent owner” defense, which stated that “no conveyance shall be forfeited . . . to the extent of an interest of an owner, by reason of any act or omission established by that owner to have been committed or omitted without the knowledge, consent, or willful blindness of the owner.” 21 U.S.C. § 881(a)(4)(C). Replacing this provision, the Civil Asset Forfeiture Reform Act of 2000, Pub. L. No. 106-185, 114 Stat. 202 (codified as amended at 18 U.S.C. § 983), creates a uniform innocent owner defense whereby a claimant may affirmatively prove innocent ownership by a preponderance of the evidence. 18 U.S.C. § 983(d). The statute specifically mentions owners who “did not know of the conduct giving rise to forfeiture”; bona fide purchasers or sellers for value; claimants who acquired a property interest through marriage, divorce, or legal separation; spouses or legal dependents who acquired property by inheritance or probate; and joint tenants and others with a partial interest in property. 20 In contrast, for orders of attachment granted without notice under New York state civil forfeiture law, contained at N.Y. C.P.L.R. art. 13-A, the claiming authority must move for a hearing within five days of the seizure. N.Y. C.P.L.R. 1317, 1329 (providing for ex parte attachment of assets for five days, after which the attaching authority must move for an order 26
    395. In sum, there is a heightened potential for erroneous retention where an arrestee, whether for DWI or some other suspected criminal conduct, is not the owner of the seized vehicle. The plight of innocent owners, as exemplified by the experience of plaintiff Sandra Jones, persuades us that an early retention hearing following seizure under N.Y.C. Code § 14- 140 is constitutionally required. 3. Other Suggested Remedies Do Not Provide Prompt Post-Seizure Review. In prosecuting vehicle forfeiture actions under N.Y.C. Code § 14-140, the City has consistently opposed motions for interim relief in the form of a retention hearing. For example, in its Memorandum of Law opposing a motion for an “immediate retention hearing” in the case of Property Clerk v. Ali, the City stated that under the CPLR there is no basis upon which defendant can even make this motion. Likewise, the CPLR does not provide for such a hearing. Defendant has circumvented the traditional rules of civil procedure by asking this court to entertain, and plaintiff to defend, against a motion that has no legal basis and a hearing that would clearly be improper under the rules. Memorandum of Law in Support of Plaintiff’s Opposition to an Immediate Retention Hearing, at 12, Property Clerk v. Ali, No. 413408/99 (Sup. Ct. N.Y. Co.). In Ali, as here, the City maintained that due process was satisfied by a resolution of the merits at the eventual civil confirming the attachment and permitting the defendant to challenge the attachment, with the burden of proof on the attaching authority to establish the grounds for the attachment, its need, and the probability of success on the merits). In upholding these provisions against doubts as to the constitutionality under the Fourteenth Amendment of ex parte pre-judgment attachment, the New York Court of Appeals relied, inter alia, on available means to challenge promptly the basis for the attachment. Morgenthau v. Citisource, Inc., 68 N.Y.2d 211, 222, 508 N.Y.S.2d 152, 157 (1986). N.Y. C.P.L.R. art. 13-A “does not limit or supersede” N.Y.C. Code § 14-140, however, and applies only to felonies and therefore has been held not to apply to forfeitures under § 14- 140. Grinberg, 181 Misc. 2d at 449, 694 N.Y.S.2d at 320-21 (quotation marks omitted). 27
    396. forfeiture hearing. Nevertheless, defendants here suggest that plaintiffs may assert their constitutional rights and challenge the City’s continued retention of their vehicles through the procedural means of a Request for Judicial Intervention (“RJI”) or an Article 78 proceeding brought under New York state law. We disagree. Under current law, any review of the legitimacy of the City’s continued retention of a vehicle would likely come, at the earliest, months after its seizure. This delay is a result, first, of the City’s need to initiate forfeiture proceedings. Assuming that a claimant requests the return of the property immediately upon seizure by the police, the City has twenty-five days in which to initiate proceedings. Only after a civil forfeiture proceeding is commenced and process is served can an owner submit a RJI. See 22 N.Y. Comp. Codes R. & Regs. § 202.6(a). Along with the RJI, a claimant files a request for a preliminary conference. See id. § 202.12(a) (“If the action has not been assigned to a judge, the party shall file a request for judicial intervention together with the request for a preliminary conference.”). Under the New York rules, a “preliminary conference” is held no later than forty- five days from the request “unless the court orders otherwise.” Id. § 202.6(b). The rules do not explicitly permit a determination of probable cause or the legitimacy of continued retention at the preliminary conference, or even provide for the taking of evidence, indicating that, at most, the preliminary conference may serve (as plaintiffs suggest) to set a future date for a probable cause hearing.21 Under the current rules, therefore, any determination of probable cause for the initial 21 The New York court rules include among “matters to be considered” at the conference the simplification of issues, setting a schedule for discovery, addition of necessary parties, and settlement, 22 N.Y. Comp. Codes R. & Regs. § 202.12(c)(1)-(4), and “any other matters that the court may deem relevant,” id. § 202.12(c)(5). 28
    397. seizure or the legitimacy of continued deprivation might come sometime within three months after the seizure, or perhaps much later.22 The Fourth and Fourteenth Amendments demand a more expeditious determination of a vehicle owner’s rights. The City also suggests that an Article 78 proceeding under New York state law is available to claimants to redress any constitutional grievances. Requiring plaintiffs to resort to an Article 78 proceeding, however, would place the onus on each plaintiff to bring a separate civil action in order to force the City to justify its seizure and retention of a vehicle. This civil action provides the “[r]elief previously obtained by writs of certiorari to review, mandamus or prohibition.” N.Y. C.P.L.R. 7801. To petition for mandamus, a claimant would have the burden of showing a clear legal right to the release of his or her vehicle. See Ass’n of Surrogate & Sup. Ct. Reporters v. Bartlett, 40 N.Y.2d 571, 574, 388 N.Y.S.2d 882, 884 (1976) (“[P]etitioners’ success in this proceeding in the nature of mandamus requires a showing of a clear legal right to the relief sought.”) (quotation marks omitted). The City cites Grinberg v. Safir as proof that relief is currently available in an Article 78 proceeding. To challenge the City’s forfeiture proceeding and retention of his vehicle, the petitioner in Grinberg filed an Article 78 action and, with it, a request for a temporary restraining order. Although the court held argument only two days after the action was filed, it denied the temporary restraining order, observing that “[l]ikely inconvenience is not proof of immediate and irreparable injury,” Grinberg, 181 Misc. 2d at 447 n.1, 694 N.Y.S.2d at 320 n.1, and eventually decided the case in favor of the City more than two months later. 22 Plaintiffs have represented that, in the New York Supreme Court, First Judicial Department, Civil Branch, all forfeiture actions under N.Y.C. Code § 14-140 are assigned to the same judge, making further delays likely. 29
    398. In sum, we conclude that the suggested remedy of an Article 78 proceeding does not provide a prompt and effective means for claimants to challenge the legitimacy of the City’s retention of their vehicles pendente lite. Cf. Fuentes, 407 U.S. at 80 (finding unconstitutional a Pennsylvania statute that “allows a post-seizure hearing if the aggrieved party shoulders the burden of initiating one”). Furthermore, inasmuch as plaintiffs claim that the federal Constitution requires the state court to offer a remedy that is currently not available under state or local law, this constitutional challenge need not proceed through the state court before it reaches the federal courts. See Kraebel v. N.Y. City Dep’t of Hous. Pres. & Dev., 959 F.2d 395, 404-06 (2d Cir. 1992) (addressing the claim that an Article 78 proceeding provided all the process plaintiff was due, and finding that “[i]t is well-established that [42 U.S.C.] § 1983 generally allows plaintiffs with federal or constitutional claims the right to sue in federal court without first resorting to state judicial remedies”); cf. Logan, 455 U.S. at 432 (“Each of our due process cases has recognized, either explicitly or implicitly, that because minimum procedural requirements are a matter of federal law, they are not diminished by the fact that the State may have specified its own procedures that it may deem adequate for determining the preconditions to adverse official action.”) (quotation marks and alterations omitted). IV. The Mathews v. Eldridge Inquiry The Supreme Court has set forth three factors to weigh in deciding whether the demands of the Due Process Clause are satisfied where the government seeks to maintain possession of property before a final judgment is rendered. See Mathews, 424 U.S. at 335. These factors should be used to evaluate the adequacy of process offered in post-seizure, pre- judgment deprivations of property in civil forfeiture proceedings. Cf. James Daniel Good Real 30
    399. Prop., 510 U.S. at 53 (finding that the Mathews inquiry “provides guidance” in determining whether to “tolerate” an exception to the rule requiring pre-deprivation notice and hearing). The factors include (1) the private interest affected; (2) the risk of erroneous deprivation through the procedures used and the value of other safeguards; and (3) the government’s interest. A. The Private Interest Affected The first factor to be considered in the Mathews inquiry is “the private interest affected by the official action.” Mathews, 424 U.S. at 335. The deprivation of real or personal property involves substantial due process interests. See James Daniel Good Real Prop., 510 U.S. at 53-54 (recognizing that “Good’s right to maintain control over his home, and to be free from governmental interference, is a private interest of historic and continuing importance”); id. at 81 (Thomas, J., concurring in part and dissenting in part) (agreeing with the majority that “[i]ndividual freedom finds tangible expression in property rights”); Doehr, 501 U.S. at 11 (“[T]he property interests that attachment affects are significant.”); Fuentes, 407 U.S. at 70-71 (holding that loss of household furniture and appliances warrants a pre-deprivation hearing). The particular importance of motor vehicles derives from their use as a mode of transportation and, for some, the means to earn a livelihood. An “individual has an important interest in the possession of his [or her] motor vehicle,” which is “often his [or her] most valuable possession.” Lee, 538 F.2d at 31; see also Perry v. McDonald, 280 F.3d 159, 174 (2d Cir. 2001) (noting that an individual’s interest in driving a vehicle represents a due process concern); Coleman v. Watt, 40 F.3d 255, 260-61 (8th Cir. 1994) (stating, in the course of applying the Mathews factors to impoundment of a car under state law, that “[a]utomobiles occupy a central place in the lives of most Americans, providing access to jobs, schools, and 31
    400. recreation as well as to the daily necessities of life”); Stypmann v. City & County of San Francisco, 557 F.2d 1338, 1342-43 (9th Cir. 1977) (finding a “substantial” interest in the “uninterrupted use of an automobile,” upon which the owner’s “ability to make a living” may depend); cf. Justice Robert H. Jackson, The Task of Maintaining Our Liberties: The Role of the Judiciary, 39 A.B.A.J. 961, 963 (1953) (“My equal right to drive an automobile may be only a claim to use of property, but it concerns my personal freedom as well.”). Other considerations as well bear on the importance of the private interest at stake. One is the availability of hardship relief under the applicable law. Cf. Dixon v. Love, 431 U.S. 105, 113 (1977) (noting the availability, under an Illinois statute, of provisions for “hardship and for holders of commercial licenses, who are those most likely to be affected by the deprival of driving privileges”). Under the New York City Civil Administrative Code, no provision is made for situations in which the seizure and retention of a vehicle would cause particular hardship. See N.Y.C. Code § 14-140 (authorizing seizure of all property used as an instrumentality of crime).23 Another consideration is the length of deprivation, which increases the weight of an owner’s interest in possessing the vehicle. See Logan, 455 U.S. at 434 (noting 23 In contrast, a provision of the federal civil forfeiture statute allows release of property pendente lite under certain circumstances, including when “continued possession by the Government . . . will cause substantial hardship to the claimant, such as preventing the functioning of a business, preventing an individual from working, or leaving an individual homeless; [and] the claimant’s likely hardship . . . outweighs the risk that the property will be destroyed, damaged, lost, concealed, or transferred if it is returned to the claimant during the pendency of the proceeding.” 18 U.S.C. § 983(f)(1)(C)-(D). Similarly, N.Y. C.P.L.R. 1311(4)(d) permits the dismissal or mitigation of a forfeiture action “in the interests of justice”—for example, in light of “the seriousness and circumstances of the crime to which the property is connected relative to the impact of forfeiture of property upon the person who committed the crime.” As noted earlier, however, N.Y. C.P.L.R. 1311 does not apply to “the limited forfeiture proceedings available pursuant to Administrative Code § 14-140.” Ferris, 77 N.Y.2d at 431, 568 N.Y.S.2d at 579. 32
    401. the Court’s concern under Mathews for “the importance of the private interest and the length or finality of the deprivation”). As noted above, the City retains seized vehicles for months or sometimes years before the merits of a forfeiture action are addressed. Finally, the importance of the claimant’s possessory interest post-seizure and pre-judgment is not diminished by the likelihood that the government will eventually prevail in forfeiture proceedings. See James Daniel Good Real Prop., 510 U.S. at 62 (“Fair procedures are not confined to the innocent. The question before us is the legality of the seizure, not the strength of the Government’s case.”); Fuentes, 407 U.S. at 87 (“The right to be heard does not depend upon an advance showing that one will surely prevail at the hearing.”). For these reasons, we cannot agree with the district court’s cursory assessment of the interest at stake based solely on its observation that the seizure of the vehicles occurred “in a jurisdiction that abounds in mass transit facilities.” Krimstock, 2000 WL 1702035, at *6. The seizure authority under the statute extends not only to cars registered in New York City, but to any found there; it also encompasses commercial as well as noncommercial vehicles. If the named class members serve as any indication, motor vehicle use is often found among those for whom mass transportation is inadequate. Valerie Krimstock, for example, states that the seizure of her vehicle hindered her from traveling from her residence in the Bronx to her job in North Tarrytown and from visiting her daughter who suffers from mental illness and lives in Pennsylvania. The seizure and retention of Clarence Walters’ vehicle made it difficult, he reports, to reach his construction job sites—some located in areas of Long Island or New Jersey inaccessible by mass transit—and as a consequence he lost a certain amount of work. James Webb, a 77-year-old retiree, states that the seizure and retention of his vehicle made it difficult 33
    402. for him and his wife to see their doctors and to visit friends, and prevented him from driving his granddaughter to school. B. The Risk of Erroneous Deprivation Through the Procedures Used and the Probable Value of Other Safeguards The second factor to be considered under the Mathews test is “the risk of an erroneous deprivation of such interest through the procedures used, and the probable value, if any, of additional or substitute procedural safeguards.” Mathews, 424 U.S. at 335. The particular deprivation with which we are concerned here is the City’s post-seizure, pre-judgment retention of plaintiffs’ vehicles. The district court concluded that the procedures used by the City—a warrantless arrest and the ultimate forfeiture proceeding—adequately protect plaintiffs against erroneous deprivation of their vehicles. Krimstock, 2000 WL 1702035, at *6-*7. We are troubled by this conclusion. Neither the arresting officer’s unreviewed probable cause determination nor a court’s ruling in the distant future on the merits of the City’s forfeiture claim can fully protect against an erroneous deprivation of a claimant’s possessory interest as his or her vehicle stands idle in a police lot for months or years. Nevertheless, we conclude that, on balance, the second Mathews factor weighs in favor of the City. We acknowledge that the risk of erroneous seizure and retention of a vehicle is reduced in the case of a DWI owner-arrestee, because a trained police officer’s assessment of the owner-driver’s state of intoxication can typically be expected to be accurate. See People v. Bennett, 238 A.D.2d 898, 899, 660 N.Y.S.2d 772, 774 (4th Dep’t 1997) (holding that the court properly “instruct[ed] jurors that the police officers were experts in determining a person’s state of intoxication”). Yet the City’s victory on the second Mathews factor is a narrow one. As noted 34
    403. earlier, the risk of erroneous deprivation that is posed to innocent owners is a substantial one. Moreover, our inquiry into the risk of error is partly informed by the City’s pecuniary interest in the outcome of § 14-140 proceedings. As the Supreme Court has observed, greater procedural safeguards are “of particular importance . . . where the Government has a direct pecuniary interest in the outcome of the proceeding.” James Daniel Good Real Prop., 510 U.S. at 55-56; cf. United States v. Funds Held in the Name or for the Benefit of Wetterer, 210 F.3d 96, 110 (2d Cir. 2000) (“We have previously observed the government’s ‘virtually unchecked use of the civil forfeiture statutes and the disregard for due process that is buried in those statutes.’”) (quoting United States v. Statewide Auto Parts, Inc., 971 F.2d 896, 905 (2d Cir. 1992)).24 Under the City’s Administrative Code, property found to have “been used as a means of committing crime or employed in aid or in furtherance of crime” may, at the discretion of the police commissioner, “be used or converted to use for the purpose of the [police] department or any city, state or federal agency.” N.Y.C. Code § 14-140(e)(2); see also Hyne, 147 Misc. 2d at 780, 557 N.Y.S.2d at 248 (noting that the forfeiture law’s “remedial purposes” include the fact that “revenue is generated and applied toward the cost of law enforcement”).25 24 On several occasions, the Second Circuit has stressed that “[f]orfeiture is a harsh and oppressive procedure which is not favored by the courts.” United States v. $31,990 in U.S. Currency, 982 F.2d 851, 856 (2d Cir. 1993) (quotation marks omitted); cf. United States v. Lasanta, 978 F.2d 1300, 1305 (2d Cir. 1992) (“While congress may have intended civil forfeiture to be a powerful weapon in the war on drugs, it would, indeed, be a Pyrrhic victory for the country, if the government’s relentless and imaginative use of that weapon were to leave the constitution itself a casualty.”) (internal quotation marks and citations omitted), abrogated on other grounds, Florida v. White, 526 U.S. 559 (1999). 25 Some state statutes expressly admonish against governmental opportunism in seizing and forfeiting property. See Cal. Health & Safety Code § 11469(a) (“Law enforcement is the principal objective of forfeiture. Potential revenue must not be allowed to jeopardize the effective investigation and prosecution of criminal offenses, officer safety, the integrity of 35
    404. The Supreme Court has expressed additional concern when, as here, the “erroneous deprivation” cannot be recompensed by the claimant’s prevailing in later proceedings: [T]he availability of a postseizure hearing may be no recompense for losses caused by erroneous seizure. Given the congested civil dockets in federal courts, a claimant may not receive an adversary hearing until many months after the seizure. And even if the ultimate judicial decision is that the claimant was an innocent owner, or that the Government lacked probable cause, this determination, coming months after the seizure, “would not cure the temporary deprivation that an earlier hearing might have prevented.” James Daniel Good Real Prop., 510 U.S. at 56 (quoting Doehr, 501 U.S. at 15); cf. Shapiro, 424 U.S. at 629 (noting that where irreparable injury may result from a deprivation of property pendente lite, “the Due Process Clause requires . . . an opportunity for some kind of predeprivation or prompt post-deprivation hearing at which some showing of the probable validity of the deprivation must be made”). In contrast, for example, to benefits for which full retroactive relief is awarded if a plaintiff ultimately prevails, see, e.g., Mathews, 424 U.S. at 340, an owner cannot recover the lost use of a vehicle by prevailing in a forfeiture proceeding. The loss is felt in the owner’s inability to use a vehicle that continues to depreciate in value as it stands idle in the police lot. Cf. Property Clerk v. Duck Jae Lee, 183 Misc. 2d 360, 363, 702 ongoing investigations, or the due process rights of citizens.”); id. § 11469(f) (“Seizing agencies shall avoid any appearance of impropriety in the sale or acquisition of forfeited property.”); Fla. Stat. § 932.704(1) (“It is the policy of this state that law enforcement agencies shall utilize the provisions of the Florida Contraband Forfeiture Act to deter and prevent the continued use of contraband articles for criminal purposes while protecting the proprietary interests of innocent owners and lienholders and to authorize such law enforcement agencies to use the proceeds collected . . . as supplemental funding for authorized purposes. The potential for obtaining revenues from forfeitures must not override fundamental considerations . . . . It is also the policy of this state that law enforcement agencies ensure that, in all seizures made under [the Act], their officers adhere to federal and state constitutional limitations regarding an individual’s right to be free from unreasonable searches and seizures . . . .”). 36
    405. N.Y.S.2d 792, 795 (Sup. Ct. N.Y. Co. 2000) (permitting a secured lender to intervene in a vehicle forfeiture proceeding brought against a DWI arrestee who had defaulted on monthly car payments, so that “the value of the subject car will not continue to depreciate if plaintiff lets the action languish”). In sum, because we recognize that the risk of erroneous deprivation in the context of DWI owner-arrestees is in many cases a reduced one, we conclude that the second Mathews factor weighs in favor of the City. The scales are very nearly in equipoise, however, in light of the comparably greater risk of error that is posed to innocent owners, the City’s direct pecuniary interest in the outcome of forfeiture proceedings, and the lack of adequate recompense for losses occasioned by erroneous seizures of vehicles. C. The Government’s Interest The third Mathews factor examines “the Government’s interest, including the function involved and the fiscal and administrative burdens that the additional or substitute procedural requirement would entail.” Mathews, 424 U.S. at 335. The City argues that it has three principal interests in continuing to retain the vehicles post-seizure and pre-judgment. The first, and the most compelling among those the City has adduced, is to prevent a vehicle from being sold or destroyed before a court can render judgment in future forfeiture proceedings. The City cites Calero-Toledo v. Pearson Yacht Leasing Co., 416 U.S. 663 (1974), for the proposition that when property is easily transportable to another jurisdiction, the City must retain custody pending resolution of the forfeiture proceedings. The Supreme Court, indeed, recognized that immediate seizure was necessary in Calero-Toledo because otherwise the yacht at issue “might have disappeared had the Government given advance warning 37
    406. of the forfeiture action.” James Daniel Good Real Prop., 510 U.S. at 57. We note initially that the Puerto Rican statute under which the yacht in Calero-Toledo was seized and forfeited provided that notice of the seizure must be served upon interested parties within ten days following the seizure and that those parties have fifteen days following service within which to challenge the seizure by serving a complaint on the confiscating officer. The complaint, filed in the Superior Court, “shall be heard without subjection to docket.” 34 P.R. Laws Ann. § 1722(a) (repealed 1988). In Calero-Toledo, the yacht was automatically forfeited when no challenge was made to the seizure within fifteen days after service of the notice of seizure. Calero-Toledo, 416 U.S. at 668. The critical difference between Calero-Toledo and the present case is that plaintiffs’ vehicles have already been seized and are in the hands of the police. Just as with real property seized by the government in forfeiture proceedings, there is no danger that these vehicles will abscond. See James Daniel Good Real Prop., 510 U.S. at 56-57 (discussing the need for seizure of movable property). Plaintiffs seek a determination only of whether continued retention of their vehicles by the City is valid and justified. Continued retention may be unjustified when other means of restraint would accomplish the City’s goals. See id. at 59 (“In the usual case, the Government . . . has various means, short of seizure, to protect its legitimate interests” in forfeitable property). To ensure that the City’s interest in forfeitable vehicles is protected, claimants could post bonds, or a court could issue a restraining order to prohibit the sale or destruction of the vehicle. See id. at 58-59 (suggesting judicial means to ensure that real 38
    407. property is not sold or destroyed pendente lite).26 The need to prevent forfeitable property from being sold or destroyed during the pendency of proceedings does not necessarily justify continued retention of all vehicles when other means of accomplishing those goals are available. A bond is in some respects a superior form of security because it entails no storage costs or costs of sale. A second reason offered by the City for maintaining custody of vehicles prior to judgment in forfeiture proceedings is that the City’s in rem jurisdiction over the vehicles depends upon its unbroken possession from seizure to judgment. The Supreme Court has held, however, that possession of a res during the entire course of the proceedings is unnecessary to preserve jurisdiction. See Republic Nat’l Bank of Miami v. United States, 506 U.S. 80, 88-89 (1992) (“We hold that, in an in rem forfeiture action, the Court of Appeals is not divested of jurisdiction by the prevailing party’s transfer of the res from the district.”). Noting that the in rem rules had their origin in admiralty law, where a court established its jurisdiction by seizure, the Supreme Court found that “the court must have actual or constructive control of the res when an in rem forfeiture is initiated.” Id. at 87 (emphasis added); cf. The Palmyra, 25 U.S. (12 Wheat.) 1, 10 (1827) (Story, J.) (“Whenever a stipulation [bond] is taken in an admiralty suit, for the property subjected to legal process and condemnation, the stipulation is deemed a mere substitute for the thing itself, and the stipulators liable to the exercise of all those authorities on the part of the 26 We note that in contrast to the forfeiture statute at issue in the present case, the Puerto Rican statute in Calero-Toledo required that seized motor vehicles be appraised for their monetary value by the Office of Transportation and that plaintiffs “have the right to give bond . . . to answer for the seizure if the lawfulness of the latter is upheld.” 34 P.R. Laws Ann. § 1722(b) (repealed 1988). 39
    408. Court, which it could properly exercise if the thing itself were still in its custody.”).27 The final interest adduced by the City is the need to prevent the offending res—here, the seized vehicle—from being used as an instrumentality in future acts of driving while intoxicated. Of course, at the time of initial seizure and retention the “offending res” is only an allegedly offending res, inasmuch as the owner’s or owner-arrestee’s misconduct in connection with the instrumentality has yet to be established in either a criminal or a civil proceeding. Moreover, although the Supreme Court has found that certain situations of “executive urgency,” James Daniel Good Real Prop., 510 U.S. at 60, call for action that dispenses with normal due process guarantees, this case does not fall within that narrow category. In defining situations of “executive urgency,” the Court has provided the examples of summary seizures during wartime, seizures of contaminated food, and, formerly, the collection of taxes when the very existence of the government depended upon them. See id. at 59-60. To take one example, the Court allowed the seizure, without prior judicial process, of forty-seven barrels of poultry from a Chicago food storage warehouse after city inspectors determined they were “putrid, decayed, poisonous, or infected in such a manner as to render it unsafe or unwholesome for human food.” N. Am. Cold Storage Co v. City of Chicago, 211 U.S. 306, 308 (1908). The threat to the public was immediate, and the spoiled poultry, like contraband, was unlikely to be used for some other legitimate purpose. Motor vehicles, in contrast, present no such threat and maintain their usefulness. Cf. Austin v. United States, 509 U.S. 602, 621 (1993) (“‘There is nothing even remotely criminal in possessing an automobile.’”) (quoting One 1958 Plymouth 27 In addition, we note that the City does not caption these civil forfeiture actions as proceedings in rem. Instead, they are styled as in personam actions—for example, Property Clerk v. Ismael Delapaz. 40
    409. Sedan v. Pennsylvania, 380 U.S. 693 (1965)). In James Daniel Good Real Property, for example, the Supreme Court found that enforcement of the drug forfeiture laws did not present “a plausible claim of urgency” strong enough to dispense with normal due process guarantees. James Daniel Good Real Prop., 510 U.S. at 61. Even if driving while intoxicated were considered a matter of “executive urgency,” the response the City has chosen, requiring the impoundment of vehicles until forfeiture proceedings are terminated, is ill-suited to address the urgency. While initial seizure of a vehicle serves the constructive purpose of keeping an individual from driving in an inebriated condition, that purpose often loses its basis in urgency once the individual has regained sobriety on the morrow.28 Furthermore, the remedy of continued impoundment leaves the alleged offender free to drive while intoxicated in any other vehicle when the opportunity presents itself, while depriving some potentially innocent owners of the often indispensable benefits of daily access to their vehicles. Finally, the City’s asserted interest in removing dangerous drivers from the road is weakened if it extends not to all vehicles seized in connection with DWI arrests, but only to those that might yield an attractive price at auction. The November 1988 Forfeiture Guide produced by the Legal Bureau of the City of New York Police Department instructs that “[c]ertain categories of property do not warrant forfeiture litigation due to their small value or the near impossibility of a successful outcome,” including “[n]on-owner operated vehicles ten years old or older,” unless, inter alia, “the vehicle has a special value, e.g., an expensive import.” 1988 Forfeiture 28 A claimant’s proven history of persistent drunkenness or repeated DWI violations, however, might justify a fact-finder in denying release of the vehicle pendente lite. 41
    410. Guide, at 24-25. We do not know whether this passage reflects current policy, but we note that the City’s interest in safety cannot be paramount if it seeks to remove from the road only a lucrative subset of the vehicles seized from intoxicated drivers. D. Balancing the Mathews v. Eldridge Factors Balancing the Mathews factors, we find that the Fourteenth Amendment guarantee that deprivations of property be accomplished only with due process of law requires that plaintiffs be afforded a prompt post-seizure, pre-judgment hearing before a neutral judicial or administrative officer to determine whether the City is likely to succeed on the merits of the forfeiture action and whether means short of retention of the vehicle can satisfy the City’s need to preserve it from destruction or sale during the pendency of proceedings. In James Daniel Good Real Property, the Supreme Court concluded that to seize real property without notice and hearing, the “Government must show that less restrictive measures—i.e., a lis pendens, restraining order, or bond—would not suffice to protect the Government’s interests in preventing the sale, destruction, or continued unlawful use of the real property.” James Daniel Good Real Prop., 510 U.S. at 62; cf. Statewide Auto Parts, 971 F.2d at 905 (urging district courts “whenever possible . . . [to] favor less drastic measures, such as occupancy agreements, bonds, receiverships, lis pendens, or other means for preserving the status quo ante seizure until the criminality underlying the claimed forfeiture can be established in the context of a proper criminal proceeding with its attendant constitutional protections to the accused”).29 Here, once the vehicles have been seized, and concerns for establishing jurisdiction 29 In Statewide Auto Parts, this Court considered a due process challenge to the federal government’s seizure of real and personal commercial properties pursuant to an ex parte warrant and the federal civil forfeiture statute. Statewide Auto Parts, 971 F.2d at 898-99. In other cases 42
    411. and immediate prophylactic custody are satisfied, we find that the Due Process Clause requires that claimants be given an early opportunity to test the probable validity of further deprivation, including probable cause for the initial seizure, and to ask whether other measures, short of continued impoundment, would satisfy the legitimate interests of the City in protecting the vehicles from sale or destruction pendente lite. Whether the vehicle is in the hands of the police the morning after it has been seized, as in this case, or whether James Daniel Good’s property is still in his hands the morning before the marshals arrive with a warrant, the question is what reason the government has for refusing to exercise some means short of continued retention after seizure to guarantee that property will be available to satisfy a civil forfeiture judgment. involving seizure of real property, this Court has been careful to emphasize the value of less drastic means than seizure for protecting the government’s interest in forfeitable property. See, e.g., United States v. Premises & Real Property at 4492 South Livonia Road, Livonia, N.Y., 889 F.2d 1258, 1265 (2d Cir. 1989) (“Any exigency that might be posed by the threat of an encumbrance on, or transfer of, the property may be met by less restrictive means than seizure, for example, by the filing of a lis pendens, as was done in this case, along with a restraining order or bond requirement.”); see also United States v. Puello, 814 F. Supp. 1155, 1164 (E.D.N.Y. 1993) (stating that, “[m]indful of [the Second Circuit’s] warning [to use less restrictive means than outright seizure], the Court has explored the possibility of these less drastic means with the parties.”); United States v. Joseph Health & Beauty Supply, 807 F. Supp. 323, 325 (S.D.N.Y. 1992) (permitting a claimant to occupy seized business premises for the conduct of lawful business pending trial, as a less restrictive measure than outright seizure of the leasehold under federal forfeiture law); cf. 18 U.S.C. § 983(f)(7), (j) (providing for restraining orders, injunctions, bonds and other means to preserve property during litigation); Supplemental Rules for Certain Admiralty and Maritime Claims, Rule E(4)(f) (providing, except where the United States is seeking forfeiture, for a “prompt hearing at which the plaintiff shall be required to show why the arrest or attachment [of property] should not be vacated or other relief granted consistent with these rules”); Fla. Stat. § 932.703(d) (providing that, if the court determines that probable cause exists to believe that property was used in violation of the state contraband forfeiture statute, “the court shall order the property restrained by the least restrictive means to protect against disposal, waste, or continued illegal use,” including such means as a “bond or other adequate security equivalent to the value of the property”). 43
    412. E. Inapplicability of United States v. $8,850 and the Speedy Trial Test The City argues that the Mathews v. Eldridge balancing test is displaced by the Supreme Court’s decision to apply the speedy trial test, and not the Mathews inquiry, in examining the constitutionality of any delay in the return of property subject to future civil forfeiture proceedings. See United States v. $8,850, 461 U.S. 555 (1983) (applying the speedy trial test set forth in Barker v. Wingo, 407 U.S. 514 (1972), in finding that an eighteen-month delay in filing a customs forfeiture action did not violate constitutional due process guarantees). We disagree. As discussed in Section III.A above, plaintiffs’ claim does not concern the speed with which civil forfeiture proceedings themselves are instituted or conducted. Instead, plaintiffs seek a prompt post-seizure opportunity to challenge the legitimacy of the City’s retention of the vehicles while those proceedings are conducted. The application of the speedy trial test presumes prior resolution of any issues involving probable cause to commence proceedings and the government’s custody of the property or persons pendente lite, leaving only the issue of delay in the proceedings. The impoundment of property—or the incarceration of a criminal defendant—certainly increases the hardship worked by any delay. The Constitution, however, distinguishes between the need for prompt review of the propriety of continued government custody, on the one hand, and delays in rendering final judgment, on the other.30 30 We also note that this Court has observed, in the context of the federal forfeiture laws, that “under the Barker [speedy trial] test, which we think applies to the holding of the forfeiture trial as well as to the filing of the action, there is a due process violation at some point.” United States v. Banco Cafetero Panama, 797 F.2d 1154, 1163 (2d Cir. 1986). “The Government argues that, once an action is filed, delays of any length may be granted to allow completion of related criminal proceedings. To require prompt filing of a forfeiture action but allow indefinite postponement of the trial would reduce the filing requirement to a nullity.” Id. 44
    413. V. The Prompt Vehicle Retention Hearing As a remedy, we order that claimants be given a prompt post-seizure retention hearing, with adequate notice,31 for motor vehicles seized as instrumentalities of crime pursuant to N.Y.C. Code § 14-140(b).32 There is no universal approach to satisfying the requirements of meaningful notice and opportunity to be heard in a situation such as this. See Fuentes, 407 U.S. at 96 (acknowledging that the “nature and form of [fair prior hearings] are legitimately open to many potential variations”); cf. Green v. County Sch. Bd., 391 U.S. 430, 439 (1968) (“There is no universal answer to the complex problems of desegregation; there is obviously no one plan that will do the job in every case.”); Int’l Salt Co. v. United States, 332 U.S. 392, 400-01 (1947) (“The framing of decrees should take place in the District rather than in Appellate Courts. They are invested with large discretion to model their judgments to fit the exigencies of the particular case.”). Moreover, the cost of additional procedures and the details of their implementation are matters peculiarly suited to the experience of the district court and the knowledge of the parties. Therefore, as we did in McClendon, 460 F.2d at 116, we leave it to the district court, in 31 Pursuant to 38-A R.C.N.Y. § 12-32, the voucher issued to an arrestee by the police upon arrest and seizure must notify the arrestee of procedures for demanding the return of property and the possible actions that the Property Clerk may take after the demand has been made. See also Butler v. Castro, 896 F.2d 698, 702 (2d Cir. 1990) (“[A] voucher must be given to an arrestee for non-contraband property seized [and] must also give notice of the procedures to be followed to recover such property.”). Adequate notice of the right to a post-seizure retention hearing could readily be added to this information. This procedure may not be adequate, however, where the arrestee and the vehicle owner are not the same person. 32 We note that both parties appear to agree that plaintiffs’ vehicles were not seized as “arrest evidence” pursuant to N.Y.C. Code § 14-140(b) but rather as instrumentalities of crime. According to the City: “The seven named plaintiffs [have had their vehicles] impounded and held by the Property Clerk for forfeiture, because they were used in furtherance of a crime.” In any event, it is hard to imagine how an arrestee’s vehicle could serve as evidence in the ordinary DWI case. 45
    414. consultation with the parties, to fashion appropriate procedural relief consistent with this opinion. See Fuentes, 407 U.S. at 97 n.33 (“Leeway remains to develop a form of hearing that will minimize unnecessary cost and delay while preserving the fairness and effectiveness of the hearing . . . .”); cf. United States v. City of Yonkers, 197 F.3d 41, 57 (2d Cir. 1999) (noting that a district court has “broad equitable discretion to apportion remedial costs” in desegregation cases). Although we decline to dictate a specific form for the prompt retention hearing, we hold that, at a minimum, the hearing must enable claimants to test the probable validity of continued deprivation of their vehicles, including the City’s probable cause for the initial warrantless seizure. In the absence of either probable cause for the seizure or post-seizure evidence supporting the probable validity of continued deprivation, an owner’s vehicle would have to be released during the pendency of the criminal and civil proceedings. We hasten to point out that we do not envision the retention hearing as a forum for exhaustive evidentiary battles that might threaten to duplicate the eventual forfeiture hearing.33 Inasmuch as the purpose of the hearing is the limited one of determining whether the 33 For example, we do not expect that these limited hearings will resemble the sometimes prolonged pretrial hearings that district courts in this Circuit have conducted to address probable cause for the government’s seizure of property under federal forfeiture laws. See, e.g., Names Registry Publ’g, 68 F.3d at 579 (noting that the district court below had held a two-day pretrial hearing to examine probable cause for an ex parte seizure of bank funds under federal forfeiture law); Daccarett, 6 F.3d at 45 (three-day pretrial evidentiary hearing to examine probable cause for seizure of electronic transfer funds under federal forfeiture law); Statewide Auto Parts, 971 F.2d at 900 (oral argument followed by further written evidentiary submissions on the issue of probable cause to seize real and personal commercial properties under federal forfeiture law); see also In re Seizure of All Funds in Accounts in Names Registry Publ’g, Inc., 887 F. Supp. 435, 449 (E.D.N.Y.) (“[I]f a claimant challenges the validity of a seizure, . . . then the merits of the forfeiture trial are expedited and the government must establish probable cause for the forfeiture prior to the forfeiture trial.”), vacated on other grounds, 68 F.3d 577 (2d Cir. 1995). Unlike federal forfeiture actions, which often involve complex evidentiary issues, multiple parties and witnesses, and elaborate underlying criminal allegations, actions instituted against vehicle owners 46
    415. vehicle should be returned to its owner during the pendency of proceedings, due process should be satisfied by an initial testing of the merits of the City’s case. In addition, the retention hearing will allow the court to consider whether less drastic measures than continued impoundment, such as a bond or a restraining order, would protect the City’s interest in the allegedly forfeitable vehicle during the pendency of proceedings. On remand, in addition to fashioning appropriate relief, the district court should ascertain how many of the named plaintiffs have recovered their vehicles from the Property Clerk and determine whether their claims have thereby been rendered moot.34 To the extent that mootness may affect the claims of any of the named plaintiffs, the court, in addressing the issue of class certification, should consider whether exceptions to the mootness doctrine preserve the under the New York City forfeiture statute typically present relatively discrete, straightforward issues. 34 At the time of oral argument before this Court, three of the seven named plaintiffs—Jones, Krimstock, and Walters—had recovered their vehicles. Inasmuch as plaintiffs primarily seek relief in the form of a preliminary injunction and class certification, the question arises whether the claims of some plaintiffs have been rendered moot, and this is a question that a court must address sua sponte. The Supreme Court has held in some cases, however, that “the termination of a class representative’s claim does not moot the claims of the unnamed members of the class. . . . That the class was not certified until after the named plaintiffs’ claims had become moot does not deprive us of jurisdiction. . . . In such cases, the ‘relation back’ doctrine is properly invoked to preserve the merits of the case for judicial resolution.” County of Riverside v. McLaughlin, 500 U.S. 44, 51-52 (1991) (internal quotation marks and citations omitted); see also Sosna v. Iowa, 419 U.S. 393, 402 n.11 (1975) (same); cf. United States Parole Comm’n v. Geraghty, 445 U.S. 388, 399 (1980) (“Some claims are so inherently transitory that the trial court will not have even enough time to rule on a motion for class certification before the proposed representative’s individual interest expires.”) (citing Gerstein v. Pugh, 420 U.S. 103, 110 n.11 (1975)); Comer v. Cisneros, 37 F.3d 775, 798-99 (2d Cir. 1994) (noting that in some cases in which the claims of named plaintiffs have become moot prior to class certification, “the courts permit the class certification to relate back to the filing of the complaint and hold that the plaintiffs have properly preserved the merits of the case for judicial resolution”); Robidoux v. Celani, 987 F.2d 931, 938-39 (2d Cir. 1993) (same). 47
    416. merits of the case for judicial resolution of the unnamed class members’ claims. CONCLUSION In conclusion, we hold that promptly after their vehicles are seized under N.Y.C. Code § 14-140 as alleged instrumentalities of crime, plaintiffs must be given an opportunity to test the probable validity of the City’s deprivation of their vehicles pendente lite, including probable cause for the initial warrantless seizure. We remand to the district court to rule on plaintiffs’ request to certify their class pursuant to Fed. R. Civ. P. 23, and to formulate, in consultation with the parties, the appropriate injunctive relief needed to redress the constitutional violations examined in this opinion. Vacated and remanded. 48
    417. 1 UNITED STATES COURT OF APPEALS 2 FOR THE SECOND CIRCUIT 3 4 August Term, 2001 5 6 (Argued: March 13, 2002 Decided: September 13, 2002) 7 8 Docket No. 01-6168 9 10 _____________________________________________ 11 12 THE CENTER FOR REPRODUCTIVE LAW AND POLICY, JANET BENSHOOF, ANIKA 13 RAHMAN, KATHERINE HALL MARTINEZ, JULIA ERNST, LAURA KATZIVE, MELISSA 14 UPRETI, CHRISTINA ZAMPAS, 15 16 Plaintiffs-Appellants, 17 18 v. 19 20 GEORGE W. BUSH, in his official capacity as President of the United States, COLIN 21 POWELL, in his official capacity as Secretary of State, ANDREW NATSIOS, in his official 22 capacity as Administrator of the United States Agency for International Development, 23 24 Defendants-Appellees. 25 _____________________________________________ 26 27 Before: McLAUGHLIN, LEVAL, and SOTOMAYOR, Circuit Judges. 28 29 The United States District Court for the Southern District of New York (Preska, 30 J.) dismissed for lack of Article III standing plaintiffs’ constitutional challenge to the federal 31 government’s “Mexico City Policy” restricting the disbursement of funds to foreign non- 32 governmental organizations. Finding that this case falls under an exception to the Supreme 33 Court’s rule against assuming the existence of jurisdiction, we dismiss plaintiffs’ First 34 Amendment claim on the merits without reaching the question of constitutional standing. We 35 dismiss plaintiffs’ due process claim for lack of prudential standing, and we dismiss plaintiffs’ 36 equal protection claim on the merits. 37 Dismissal affirmed on different grounds.
    418. 1 2 _______________________________________________ 3 4 SIMON HELLER, The Center for Reproductive Law & 5 Policy, New York, NY (Janet Benshoof, on the brief), for 6 Plaintiffs-Appellants. 7 8 GREGORY G. KATSAS, Deputy Assistant Attorney 9 General, Washington, D.C. (Robert D. McCallum, Jr., 10 Assistant Attorney General; Robert M. Loeb and Sharon 11 Swingle, Attorneys, Department of Justice Civil Division; 12 James B. Comey, United States Attorney; Gideon A. Schor, 13 Chief Appellate Attorney, on the brief), for Defendants- 14 Appellees. 15 ______________________________________________ 16 17 SOTOMAYOR, Circuit Judge: 18 This suit was brought by a domestic organization that advocates reproductive 19 rights and by attorneys employed by the organization. Plaintiffs challenge the so-called “Mexico 20 City Policy,” pursuant to which the United States government requires foreign organizations, as a 21 condition of receiving government funds, to agree neither to perform abortions nor to promote 22 abortion generally. Plaintiffs maintain that these restrictions violate their First Amendment rights 23 to freedom of speech and association. The district court dismissed the case for lack of subject 24 matter jurisdiction, finding that plaintiffs lack standing under Article III of the Constitution. The 25 district court was following the general rule, set forth by the Supreme Court in Steel Co. v. 26 Citizens for a Better Environment, 523 U.S. 83, 94-95 (1998), that a federal court may not 27 assume it has jurisdiction over a matter and proceed directly to the merits. The instant case is 28 exceptional, however. Some twelve years ago we entertained and rejected, on the merits, the 29 same constitutional challenge to the provision at issue here. We therefore find that this case falls 2
    419. 1 within an exception recognized by the Supreme Court in Steel Co., and we dismiss the First 2 Amendment claim on the merits without deciding the standing question. Plaintiffs also bring 3 claims under the Due Process Clause and the equal protection component of the Fifth 4 Amendment. We dismiss the due process claim under the doctrine of prudential standing, as 5 plaintiffs’ alleged harm does not fall within the zone of interests protected by the Due Process 6 Clause. We dismiss the equal protection claim as without merit; while plaintiffs do have 7 standing for this claim under the concept we have dubbed “competitive advocate standing,” the 8 classification they challenge does not constitute an equal protection violation. 9 BACKGROUND 10 We accept the allegations in the complaint as true on this motion to dismiss. The 11 facts of this case, which are set forth in greater detail by the district court, see Center for 12 Reproductive Law & Policy v. Bush, No. 01 CIV. 4986, 2001 WL 868007 (S.D.N.Y. July 31, 13 2001) (“CRLP”), are as follows. Plaintiff The Center for Reproductive Law & Policy (“CRLP”) 14 is a nonprofit advocacy organization devoted to the promotion of reproductive rights. Individual 15 plaintiffs Janet Benshoof, Anika Rahman, Katherine Hall Martinez, Julia Ernst, Laura Katzive, 16 Melissa Upreti and Christina Zampas are CRLP staff attorneys engaged in the organization’s 17 global mission of reproductive law reform. Defendant George W. Bush is the President of the 18 United States. Defendant Colin Powell is the U.S. Secretary of State and is thus responsible for 19 “ensuring program and policy coordination among agencies of the United States Government in 20 carrying out the policies set forth in the Foreign Assistance Act. . . .” 22 U.S.C. § 6593(b)(2). 21 Defendant Andrew Natsios is the Administrator of the United States Agency for International 3
    420. 1 Development (“USAID”). At issue in this case is the so-called “Mexico City Policy”1 of the 2 United States government, whereby foreign non-governmental organizations (“NGOs”) receiving 3 U.S. government funds must agree to a provision called the “Standard Clause,” which prohibits 4 the organizations from engaging in activities that promote abortion (also referred to as the 5 “challenged restrictions”). 6 The Foreign Assistance Act of 1961 (“FAA”) authorizes the President “to furnish 7 assistance, on such terms and conditions as he may determine, for voluntary population 8 planning.” 22 U.S.C. § 2151b(b). The President’s authority to allocate FAA funding has been 9 delegated to the Secretary of State and, in turn, to the Administrator of USAID. See Exec. Order 10 No. 13,118, 64 Fed. Reg. 16,595 (Mar. 31, 1999); State Department Delegation of Authority No. 11 145-1, 45 Fed. Reg. 51,974 (Aug. 5 1980); International Development Cooperation Agency 12 Delegation of Authority No. 7, 45 Fed. Reg. 52,470 (Aug. 7, 1980). In 1973, Congress enacted 13 the Helms Amendment, which prohibits the use of foreign assistance funds to pay for, among 14 other things, “the performance of abortions as a method of family planning or to motivate or 15 coerce any person to practice abortions.” 22 U.S.C. § 2151b(f)(1). This restriction applies only 16 to the use of U.S. government funds; foreign NGOs receiving assistance may still promote 17 abortion with non-U.S. government funds without violating the terms of the statute. The 18 executive branch, however, has attached additional conditions to the granting of foreign 19 assistance funds, as it is authorized to do by the FAA. See 22 U.S.C. § 2151b(b). These 20 additional conditions are the subject of this suit. 1 The term derives from a United Nations conference held in Mexico City in 1984, at which the United States delegation presented a policy statement outlining the type of abortion- related restrictions at issue in this case. CRLP, 2001 WL 868007, at *2 n.1. 4
    421. 1 The challenged restrictions originated in August 1984, when President Ronald 2 Reagan announced the Mexico City Policy (“the Policy”). The Policy expressed the 3 government’s disapproval of abortion as an element of family planning programs and set forth 4 various ways in which the government would prohibit its funds from being used to support 5 abortion overseas. Among these, it was announced that “the United States will no longer 6 contribute to separate nongovernmental organizations which perform or actively promote 7 abortion as a method of family planning in other nations.” CRLP, 2001 WL 868007, at *4 8 (citations omitted). 9 Pursuant to the Mexico City Policy, USAID incorporated the “Standard Clause” 10 into its family planning assistance agreements and contracts. The Standard Clause provides that 11 in order to be eligible for USAID funding, a foreign NGO must certify in writing that it “will not, 12 while receiving assistance under the grant, perform or actively promote abortion as a method of 13 family planning in AID-recipient countries or provide financial support to other foreign 14 nongovernmental organizations that conduct such activities.” Id. at *5 (quotation marks 15 omitted). The restrictions established in the Standard Clause extend to all activities of recipient 16 NGOs, not merely to projects funded by USAID. Thus, in order to receive U.S. government 17 funds, a foreign NGO may not engage in any activities that promote abortion. These restrictions 18 do not apply to domestic NGOs such as plaintiff CRLP. 19 The Mexico City Policy was rescinded by President Bill Clinton in January 1993, 20 but was reinstated by President George W. Bush in March 2001. President Bush issued an 21 official memorandum that restored the abortion-related restrictions discussed above, including 22 the Standard Clause. See Memorandum, Restoration of the Mexico City Policy, 66 Fed. Reg. 5
    422. 1 17,303, 17,309 (Mar. 28, 2001) (“Restoration Memorandum”). Accordingly, as a condition of 2 receiving U.S. government funds, foreign NGOs again are required to agree not to perform or 3 actively promote abortion as a method of family planning.2 4 Plaintiffs bring this suit for injunctive and declaratory relief. Plaintiffs’ primary 5 claim, and the one with which the district court appears exclusively to have concerned itself, is 6 based on the First Amendment. The thrust of this claim is that, as a result of the challenged 7 restrictions, foreign NGOs are chilled from interacting and communicating with domestic 8 abortion rights groups such as plaintiff CRLP, thus depriving plaintiffs of their rights to freedom 9 of speech and association in carrying out the mission of the organization. Plaintiffs also allege 10 that the restrictions violate the Equal Protection Clause of the Fifth Amendment by preventing 11 plaintiffs from competing on “equal footing” with domestic anti-abortion groups, and that they 12 violate the Due Process Clause by failing to give clear notice of what speech and activities they 13 prohibit and by encouraging arbitrary and discriminatory enforcement. Finally, plaintiffs attempt 14 to bring a claim under customary international law, the substance of which appears to be identical 15 to their First Amendment claim. 16 The district court dismissed the action in its entirety on the ground that plaintiffs 17 lack standing under Article III of the Constitution. The court first noted that because the 18 challenged restrictions apply only to foreign NGOs, not to domestic organizations such as CRLP, 19 the Mexico City Policy does not affect plaintiffs directly. CRLP, 2001 WL 868007, at *7. The 2 “Abortion as a method of family planning” does not include “abortions performed if the life of the mother would be endangered if the fetus were carried to term or abortions performed following rape or incest (since abortion under these circumstances is not a family planning act).” Restoration Memorandum, 66 Fed. Reg. at 17,306. 6
    423. 1 court then applied the three-pronged standing test set out by the Supreme Court in Lujan v. 2 Defenders of Wildlife, 504 U.S. 555 (1992), and concluded that plaintiffs had failed to 3 demonstrate that (1) concrete injury in fact, (2) a causal connection between the alleged injury 4 and the government’s conduct, and (3) that the alleged injury is sufficiently redressable by a 5 favorable decision. CRLP, 2001 WL 868007, at *8-*12. 6 Our review is de novo. See Connecticut v. Physicians Health Servs. of Conn., 7 Inc., 287 F.3d 110, 114 (2d Cir. 2002). “The reviewing court may, of course, affirm on any 8 ground appearing in the record below.” MFS Sec. Corp. v. New York Stock Exch., Inc., 277 F.3d 9 613, 617 (2d Cir. 2002). 10 DISCUSSION 11 I. First Amendment Claim 12 A. Plaintiffs’ Allegations 13 The crux of plaintiffs’ First Amendment claim is their contention that the 14 restrictions chill foreign NGOs from collaborating with domestic NGOs like CRLP because such 15 collaboration may be viewed as promoting abortion and thus would jeopardize the foreign 16 NGOs’ receipt of U.S. government funds. Plaintiffs argue that such collaboration is essential to 17 their ability to carry out their mission as advocates of reproductive rights and that depriving them 18 of this ability violates their freedom of speech and association. 19 Specifically, plaintiffs allege that they depend on collaboration with foreign 20 NGOs in order to advocate abortion law reform in foreign countries; to gather reliable 21 information regarding abortion laws; to disseminate publications and reports; to reach audiences 22 worldwide in order to promote abortion law reform; to access victims and witnesses of human 7
    424. 1 rights abuses; to lobby the United States government to rescind the Restoration Memorandum; to 2 influence international conferences, international legal tribunals, and world public opinion; to 3 increase protection for the right to abortion in the United States; and to engage in open and free 4 discussion about abortion. See Am. Compl. ¶¶ 7, 85, 88, 90, 91, 105-107. 5 Plaintiffs list several countries in which they currently have projects involving 6 these activities and where foreign NGOs have agreed to the Standard Clause, id. ¶ 71, and they 7 allege that all of these activities are significantly hindered in those countries. The use of the 8 Standard Clause, according to plaintiffs, “prevents Plaintiffs from forming alliances with 9 potential partner organizations in order to increase their abortion-related advocacy efforts’ 10 effectiveness.” Id. ¶ 100. One of the ways in which this problem manifests itself is by depriving 11 plaintiffs of their audience for reproductive rights advocacy. Plaintiffs allege that the use of the 12 Standard Clause “interferes with Plaintiffs’ conveyance of their ideas and political speech about 13 abortion by chilling or prohibiting [foreign NGOs] from attending presentations given by 14 Plaintiffs and from listening to Plaintiffs’ political advocacy.” Id. ¶ 106. These hindrances, 15 according to plaintiffs, violate their right to freedom of speech and association. Similarly, 16 plaintiffs allege that the challenged restrictions impede their ability to disseminate publications 17 and reports “because [foreign NGOs] that would otherwise distribute the publications in foreign 18 countries are prohibited or chilled from doing so.” Id. ¶ 103. Plaintiffs argue that this harm is 19 actionable under Supreme Court precedent holding that “[t]he First Amendment protects 20 [individuals’] right not only to advocate their cause but also to select what they believe to be the 21 most effective means for so doing.” Meyer v. Grant, 486 U.S. 414, 424 (1988). Plaintiffs also 22 invoke their right to receive information, claiming that the Standard Clause “interferes with 8
    425. 1 Plaintiffs’ ability to obtain information necessary to accomplish their abortion law reform efforts 2 from USAID recipient [foreign NGOs],” and impedes plaintiffs’ access to victims and witnesses 3 of human rights abuses related to reproductive issues. Am. Compl. ¶¶ 101-102. Plaintiffs 4 explain that foreign NGOs are often the only vehicle to provide access to both general 5 information and first-hand accounts regarding conditions in foreign countries, id., and that 6 obtaining such information is necessary for domestic NGOs to fulfill their mission of advocating 7 reproductive rights—including their ability to lobby the United States government, id. ¶ 108. 8 B. The Planned Parenthood Case 9 We have been over this ground before. In Planned Parenthood Federation of 10 America, Inc. v. Agency for International Development, 915 F.2d 59 (2d Cir. 1990), this Court 11 entertained a constitutional challenge to the same Standard Clause incorporated by the USAID 12 into financial assistance agreements with foreign NGOs. Like the instant case, Planned 13 Parenthood involved a First Amendment challenge, based on freedom of speech and association, 14 brought by domestic NGOs. As in the instant case, the plaintiffs argued before this Court that the 15 Mexico City Policy effectively prevented them from associating and collaborating with foreign 16 NGOs, which in turn prevented them from fulfilling their mission regarding reproductive rights 17 advocacy. Planned Parenthood, 915 F.2d at 62-63. 18 This Court rejected the challenge on the merits, finding “no constitutional rights 19 implicated” by the Policy and the Standard Clause. Planned Parenthood, 915 F.2d at 66. The 20 Court reasoned that the domestic NGOs remained free to use their own funds to pursue abortion- 21 related activities in foreign countries and that “[t]he harm alleged in the complaint is the result of 22 choices made by foreign NGOs to take AID’s money rather than engage in non-AID funded 9
    426. 1 cooperative efforts with plaintiffs-appellants.” Id. at 64. “Such an incidental effect” on the 2 activities of the domestic NGOs, the Court held, did not rise to the level of a constitutional 3 violation. Id. The Court concluded that “the Standard Clause does not prohibit plaintiffs- 4 appellants from exercising their first amendment rights.” Id. Moreover, the Court explained that 5 whatever one might think of the Mexico City Policy, “the wisdom of, and motivation behind, this 6 policy are not justiciable issues,” and the Court found the restrictions to be rationally related to 7 the “otherwise nonjusticiable decision limiting the class of beneficiaries of foreign aid.” Id. at 8 64-65. Having rejected plaintiffs’ claims on the merits, this Court declined to address the 9 question of whether plaintiffs had standing under Article III. Id. at 66. 10 Planned Parenthood not only controls this case conceptually; it presented the 11 same issue. Planned Parenthood rejected the same First Amendment challenge to the same 12 provision—the Standard Clause that was first instituted by President Reagan in the 1980s and 13 was reinstated by President George W. Bush in 20013—and no intervening Supreme Court case 14 law alters its precedential value. 15 Plaintiffs’ attempts to distinguish Planned Parenthood are unavailing. First, 16 plaintiffs argue that Planned Parenthood did not involve an equal protection challenge. This is 17 true, but does not affect the First Amendment question. Second, plaintiffs argue that Planned 18 Parenthood “mischaracterizes the [restrictions’] effect as ‘incidental.’” This argument does not 3 The Standard Clause as restored under President George W. Bush contains minor alterations from the original version challenged in the Planned Parenthood case. They are not of significance here. The only substantive difference in the restored Standard Clause is that “treatment of injuries or illnesses caused by legal or illegal abortions” is now excluded from the definition of prohibited abortion-related activities. Restoration Memorandum, 66 Fed. Reg. at 17,311. Other minor alternations include the change from “AID” to “USAID,” “grant” to “award,” and “birth spacing” to “child spacing.” CRLP, 2001 WL 868007, at *6 n.6. 10
    427. 1 distinguish Planned Parenthood at all, but simply disagrees with its holding. Third, plaintiffs 2 argue that the effect on their speech is more substantial than in Planned Parenthood because the 3 provision “impedes Plaintiffs’ entire mission, not just one component of that mission.” The 4 significance of this point is not clear to us as a legal matter and, in any event, the allegations 5 made in the two cases are far too similar to support this distinction as a factual matter. 6 Finally, plaintiffs argue that Planned Parenthood “did not assess the right to 7 obtain and impart information,” and that the litigants in Planned Parenthood “did not claim that 8 their law reform advocacy in the United States and the United Nations was impeded.” By 9 rejecting plaintiffs’ claim that the Mexico City Policy prevented them from associating and 10 collaborating with foreign NGOs, however, this Court’s opinion in Planned Parenthood did, in 11 fact, assess and reject the claim that plaintiffs’ right to obtain and impart information was 12 impeded. See Planned Parenthood, 915 F.2d at 63-64 (noting and rejecting plaintiffs’ argument 13 that “it is impractical for United States citizens or organizations to engage in abortion-related 14 activities abroad without the cooperation of foreign organizations and that the Standard Clause 15 deters ‘many of the most logical and effective foreign partners’”); Planned Parenthood Fed’n of 16 Am., Inc. v. Agency for Int’l Dev., No. 87 CIV. 0248, 1990 WL 26306, at *5 (S.D.N.Y. Mar. 7, 17 1990) (“[P]laintiffs argue that [the Standard Clause] has the effect of preventing foreign NGOs 18 that receive AID funds and domestic NGOs from associating with each other for purposes of 19 receiving or disseminating abortion information using non-U.S. government money . . . .” 20 (emphasis added)). Likewise, although this Court’s opinion in Planned Parenthood did not 21 explicitly describe the scope of plaintiffs’ claim regarding the restrictions on their law reform 22 advocacy to include advocacy in the U.S. and in international tribunals, our holding clearly 11
    428. 1 contemplated and rejected that claim. See Planned Parenthood, 915 F.2d at 62 (noting that the 2 Standard Clause does not hinder plaintiffs’ use of non-AID funds “in the United States or 3 abroad”); Planned Parenthood, 1990 WL 26306, at *7 (“[Plaintiffs] also allege that the ‘reason 4 for the promulgation of the policy and the Standard Clause was to advance the Reagan 5 Administration’s effort to suppress pro-choice views and activities in the United States . . . and 6 not for any purported concern with foreign policy . . . .’” (quoting complaint)). 7 C. The Standing Issue 8 The district court dismissed the instant case, not on the merits as we did in 9 Planned Parenthood, but for lack of constitutional standing. A federal court has jurisdiction only 10 if a claim presents a “case” or “controversy” under Article III of the U.S. Constitution. This 11 “irreducible constitutional minimum” of standing requires (1) that the plaintiff has suffered an 12 “injury in fact,” i.e., an invasion of a judicially cognizable interest which is concrete and 13 particularized as well as actual or imminent, rather than conjectural or hypothetical; (2) that there 14 is a causal connection such that the injury is fairly traceable to the challenged conduct; and (3) 15 that it is likely, as opposed to merely speculative, that the injury will be redressed by a favorable 16 decision. Lujan, 504 U.S. at 560-61. “Since this case remains at the pleading stage, all facts 17 averred by the plaintiffs must be taken as true for purposes of the standing inquiry—as they must 18 be for any other issue presented.” Lerman v. Bd. of Elections, 232 F.3d 135, 142 (2d Cir. 2000). 19 The district court held that plaintiffs failed to show these elements of standing. 20 See CRLP, 2001 WL 868007, at *8-*12. In reaching its conclusion, the court relied heavily on 21 our analysis in Planned Parenthood. See, e.g., id. at *10 (“The Court of Appeals has already 22 held that the government is within its constitutional authority in imposing restrictions or 12
    429. 1 conditions on the receipt of USAID funding by [foreign NGOs].”). In particular, the district 2 court placed great weight on our statement in Planned Parenthood that the harm alleged by 3 domestic NGOs is the result not of the Mexico City Policy itself, but of “choices made by foreign 4 NGOs to take AID’s money rather than engage in non-AID funded cooperative efforts with 5 plaintiffs-appellants.” Id. at *10, *11 (quoting Planned Parenthood, 915 F.2d at 64). Based on 6 this language from Planned Parenthood, the district court found that “plaintiffs have failed to 7 show that their alleged harms are caused by the challenged government policies.” Id. at *12. 8 It is not clear, however, that the district court’s reliance on Planned Parenthood is 9 entirely justified in this context. We found in Planned Parenthood that the alleged harm suffered 10 by domestic NGOs is attributable to independent decisions of foreign NGOs, but only for 11 purposes of the merits of plaintiffs’ First Amendment claims. It does not necessarily follow that 12 Planned Parenthood answers the question of causation with respect to constitutional standing. 13 One reason why Planned Parenthood might be deemed to resolve the standing 14 question is that Planned Parenthood, though adjudicated on the merits, was decided on the 15 pleadings. Thus, one could argue that this Court decided as a matter of law that the Mexico City 16 Policy could not be deemed the legal “cause” of the alleged harm to domestic NGOs. Although 17 this finding was used to form a different conclusion in Planned Parenthood—that plaintiffs’ 18 claims failed on the merits—it arguably could be employed in our standing analysis here. On the 19 other hand, it could be argued that Planned Parenthood is not dispositive, particularly in light of 20 an intervening Supreme Court case that clarified the causation aspect of the standing inquiry. In 21 Bennett v. Spear, 520 U.S. 154, 167 (1997), plaintiffs argued that a Biological Opinion by the 22 Fish and Wildlife Service influenced the Bureau of Reclamation to reduce the quantity of 13
    430. 1 irrigation water available to plaintiffs. Rejecting the government’s contention that plaintiffs 2 lacked standing because the Bureau’s conduct constituted an “independent” act breaking the 3 chain of causation under Lujan, the Supreme Court explained that “[t]his wrongly equates injury 4 ‘fairly traceable’ to the defendant with injury as to which the defendant’s actions are the very last 5 step in the chain of causation.” Id. at 168-69. The Court stated that while “it does not suffice if 6 the injury complained of is the result of the independent action of some third party not before the 7 court . . . that does not exclude injury produced by determinative or coercive effect upon the 8 action of someone else.” Id. at 169 (quotation marks and alterations omitted). Bennett can be 9 read to support plaintiffs’ standing argument in the instant case. 10 We are thus faced with a situation of a sui generis nature, inasmuch as our 11 conclusion depends in large part on how much weight one places on our language in Planned 12 Parenthood—a case that analyzed essentially the same factual allegations as the instant case but 13 in a somewhat different context. As Planned Parenthood does not, as the district court implied, 14 resolve the standing issue conclusively, we are confronted with a novel question of Article III 15 standing. 16 D. The Steel Co. Case and Our Authority to Proceed to the Merits 17 Because we believe that our decision in Planned Parenthood dooms plaintiffs’ 18 First Amendment claims on the merits, we must decide whether we should first address 19 plaintiffs’ novel theory of constitutional standing with respect to these claims. 20 Between the time that we decided Planned Parenthood and the filing of the 21 instant action, the Supreme Court issued a decision in which it criticized the practice whereby a 22 court proceeds directly to the merits of a case while assuming arguendo that the plaintiffs have 14
    431. 1 constitutional standing to bring the suit. See Steel Co. v. Citizens for a Better Env’t, 523 U.S. 83, 2 94-101 (1998). This practice, referred to by some courts as “hypothetical jurisdiction,” United 3 States v. Troescher, 99 F.3d 933, 934 n.1 (9th Cir. 1996), was often used by federal courts 4 seeking to avoid a difficult or novel issue of standing in favor of a relatively easy merits question. 5 In Steel Co., however, Justice Scalia explained that the determination of standing is a question of 6 subject matter jurisdiction, and that a court lacks the authority to rule on a case unless it 7 determines that jurisdiction exists. Steel Co., 523 U.S. at 94. “For a court to pronounce upon the 8 meaning or the constitutionality of a state or federal law when it has no jurisdiction to do so is, 9 by very definition, for a court to act ultra vires.” Id. at 101-02. Justice Scalia’s decision in Steel 10 Co. commanded a five-Justice majority, although two of the five issued a concurring opinion, 11 which expressed a more permissive view toward the practice of assumed jurisdiction. See id. at 12 110-11 (O’Connor, J., concurring, joined by Kennedy, J.) (stating that “the Court’s opinion 13 should not be read as cataloging an exhaustive list of circumstances under which federal courts 14 may exercise judgment in reserving difficult questions of jurisdiction when the case alternatively 15 could be resolved on the merits in favor of the same party”) (quotation marks and alterations 16 omitted). This Court has heeded the admonitions of Steel Co., acknowledging that ordinarily we 17 are not to assume the existence of jurisdiction in favor of reaching an “easier” merits issue. 18 Fidelity Partners, Inc. v. First Trust Co. of N.Y., 142 F.3d 560, 565 (2d Cir. 1998); see also In re 19 Rationis Enters., Inc. of Panama, 261 F.3d 264, 267-68 (2d Cir. 2001) (citing Steel Co. rule). 20 The Steel Co. majority opinion, however, discussed several previous Supreme 21 Court decisions which, according to the Court, “must be acknowledged to have diluted the 22 absolute purity of the rule that Article III jurisdiction is always an antecedent question.” Steel 15
    432. 1 Co., 523 U.S. at 101. Moreover, the Court chose not to state simply that, to the extent that 2 previous cases might be read to permit assumed jurisdiction, those cases are overruled. Instead, 3 the Court distinguished the cases on various grounds, thus leaving their precedential value 4 intact.4 Two such cases are of particular relevance here. 5 The first is Norton v. Matthews, 427 U.S. 524 (1976), in which the Court declined 6 to address a jurisdictional issue and answered the merits question regarding whether certain 7 aspects of the Social Security Act were unconstitutional. The Steel Co. Court distinguished 8 Norton on the ground that, in Norton, “[w]e declined to decide th[e] jurisdictional question, 9 because the merits question was decided in a companion case, Mathews v. Lucas, with the 10 consequence that the jurisdictional question could have no effect on the outcome.” Steel Co., 11 523 U.S. at 98 (internal citation omitted). The Steel Co. Court explained that the outcome in 12 Norton was “foreordained by Lucas” and thus “Norton did not use the pretermission of the 13 jurisdictional question as a device for reaching a question of law that otherwise would have gone 14 unaddressed.” Id. 15 The Steel Co. Court also distinguished and did not overrule Secretary of Navy v. 16 Avrech, 418 U.S. 676 (1974). The Court explained, “Avrech also involved an instance in which 17 an intervening Supreme Court decision definitively answered the merits question.” Steel Co., 18 523 U.S. at 98-99. Avrech involved a constitutional challenge to a provision of the Code of 19 Military Justice. When another case, Parker v. Levy, 417 U.S. 733 (1974), rejected a similar 4 See Joan Steinman, After Steel Co.: “Hypothetical Jurisdiction” in the Federal Appellate Courts, 58 Wash. & Lee L. Rev. 855, 862 (2001) (noting the Steel Co. Court’s “embrace, rather than disavowal,” of cases such as Norton and Avrech, both discussed infra). 16
    433. 1 constitutional challenge to the same provision, the Court decided to dispose of Avrech on the 2 merits, stating that it was “unwilling to decide the difficult jurisdictional issue which the parties 3 have briefed.” Avrech, 418 U.S. at 677. The Avrech Court explained its rationale: “We believe 4 that even the most diligent and zealous advocate could find his ardor somewhat dampened in 5 arguing a jurisdictional issue where the decision on the merits is thus foreordained.” Id. at 678. 6 The Steel Co. Court thus distinguished Avrech, finding that the “peculiar circumstances” of 7 Avrech did not permit the case to be cited for the more general proposition that any “easy” merits 8 question may be decided on the assumption of jurisdiction. Steel Co., 523 U.S. at 99. 9 Thus, the majority opinion in Steel Co. appears to allow an exception to the rule 10 against assuming the existence of standing in those “peculiar circumstances” where the outcome 11 on the merits has been “foreordained” by another case such that “the jurisdictional question could 12 have no effect on the outcome,” provided the court “d[oes] not use the pretermission of the 13 jurisdictional question as a device for reaching a question of law that otherwise would have gone 14 unaddressed.” Id. at 98;5 cf. Nippon Steel Corp. v. United States, 219 F.3d 1348, 1352-53 (Fed. 15 Cir. 2000) (using the Steel Co. Court’s approval of Norton as authority to bypass a jurisdictional 16 question and decide the merits in an “unusual situation” where the two issues are intertwined). 17 We find ourselves in largely the same situation as the Supreme Court found itself in Norton and 5 The Steel Co. Court seems to acknowledge this when, after recognizing that cases such as Norton and Avrech “have diluted the absolute purity of the rule that Article III jurisdiction is always an antecedent question,” the Court urges that these cases do not support a rule that “enables a court to resolve contested questions of law when its jurisdiction is in doubt.” Steel Co., 523 U.S. at 101 (emphasis added). Moreover, a majority of the Justices in Steel Co. cited Norton approvingly for the proposition that a court may assume the existence of jurisdiction in certain circumstances. See id. at 110-11 (O’Connor, J., concurring, joined by Kennedy, J.); id. at 111 (Breyer, J., concurring in part and concurring in the judgment); id. at 122 n.15 (Stevens, J., concurring in the judgment, joined in relevant part by Souter, J.). 17
    434. 1 Avrech: plaintiffs in this case challenge a governmental provision (the use of the Standard 2 Clause) as unconstitutional, and there is a controlling case in which this Court entertained and 3 rejected the same constitutional challenge to the same provision. Our outcome on the merits is 4 thus “foreordained” by Planned Parenthood. Under the Norton/Avrech exception acknowledged 5 in Steel Co., we need not reach the academic question of Article III standing in this case. 6 Our approach not only comports with the language of the Steel Co. majority 7 opinion, but also advances the underlying rationale of Steel Co. and makes good sense as a 8 constitutional matter. The concern of the Steel Co. majority was that deciding a case on the mere 9 assumption of jurisdiction can lead to the rendering of advisory opinions in violation of Article 10 III: “Hypothetical jurisdiction produces nothing more than a hypothetical judgment—which 11 comes to the same thing as an advisory opinion, disapproved by this Court from the beginning.” 12 Steel Co., 523 U.S. at 101 (citations omitted). Turning to the instant case, we note that where the 13 precise merits question has already been decided in another case by the same court, it is the 14 adjudication of the standing issue that resembles an advisory opinion—the very concern that 15 animates the Steel Co. rule. It would be ironic if, in our desire to avoid rendering an advisory 16 opinion, we were to address a novel standing question in a case where the result is foreordained 17 by another decision of this Court. See id. at 123-24 (Stevens, J., concurring in the judgment) 18 (noting that by addressing a standing issue unnecessarily “the Court is engaged in a version of the 19 ‘hypothetical jurisdiction’ that it has taken pains to condemn”). We further note that the question 20 of Article III standing is itself of constitutional dimensions, see id. at 124, and “the Supreme 21 Court has for generations warned against reaching out to adjudicate constitutional matters 22 unnecessarily,” Horne v. Coughlin, 191 F.3d 244, 246 (2d Cir. 1999). 18
    435. 1 We hold that where, as here, a governmental provision is challenged as 2 unconstitutional, and a controlling decision of this Court has already entertained and rejected the 3 same constitutional challenge to the same provision, the Court may dispose of the case on the 4 merits without addressing a novel question of jurisdiction. The Supreme Court followed this 5 approach in Norton and Avrech, and approved of those cases in Steel Co. Plaintiffs’ First 6 Amendment claims are therefore dismissed for failure to state a claim.6 7 II. Due Process Claim: Lack of Prudential Standing 8 Because Planned Parenthood did not address due process claims brought by 9 domestic NGOs in this context, we address the due process claim separately and dismiss it on the 10 alternative ground of prudential standing. 11 “The doctrine of standing, which addresses the question of whether the plaintiff is 12 entitled to have the court decide the merits of the dispute or of particular issues, embraces both 13 ‘constitutional’ and ‘prudential’ requirements.” Sullivan v. Syracuse Hous. Auth., 962 F.2d 14 1101, 1106 (2d Cir. 1992) (quotation marks and brackets omitted). The constitutional 15 requirements, derived from Article III, are the injury in fact, causation, and redressability 16 elements set out by the Supreme Court in Lujan. On the other hand, “[t]he prudential 17 requirements of standing have been developed by the Supreme Court on its own accord and 18 applied in a more discretionary fashion as rules of judicial ‘self-restraint’ further to protect, to the 19 extent necessary under the circumstances, the purpose of Article III.” Id. (internal citations 20 omitted). Pursuant to the doctrine of prudential standing, a court must ask whether a plaintiff’s 6 As plaintiffs’ claims based on customary international law are substantively indistinguishable from their First Amendment claims, they are dismissed on the same ground. We express no view as to whether those claims are otherwise viable. 19
    436. 1 claim rests on the legal rights of a third party, asserts only a generalized grievance, or asserts a 2 claim that falls outside the zone of interests protected by the legal provision invoked. See Valley 3 Forge Christian Coll. v. Ams. United, 454 U.S. 464, 474-75 (1982); In re Appointment of Indep. 4 Counsel, 766 F.2d 70, 74 (2d Cir. 1985). Of particular concern in the instant case is “the 5 requirement that a plaintiff’s complaint fall within the zone of interests protected by the law 6 invoked,” Crist v. Comm’n on Presidential Debates, 262 F.3d 193, 195 (2d Cir. 2001) (quoting 7 Allen v. Wright, 468 U.S. 737, 750-51 (1984)), coupled with the rule against asserting the rights 8 of a third party. Plaintiffs’ claims do not fall within the “zone of interests” protected by the Due 9 Process Clause. 10 Plaintiffs’ due process claim is based on their allegation that the challenged 11 restrictions fail to give clear notice of what political speech, public education, and law reform 12 activities they prohibit and that they encourage arbitrary and discriminatory enforcement. Am. 13 Compl. ¶ 140. It is not the plaintiffs, however, who are allegedly left uncertain of their rights by 14 unconstitutionally vague language in a government provision; it is the foreign NGOs who are 15 allegedly left in this position. Plaintiffs’ harm is derivative of this due process-type harm, and 16 their alleged injury (albeit an unactionable one) concerns First Amendment interests. Plaintiffs’ 17 allegation, simply put, is that the vague language of the Standard Clause causes the foreign 18 NGOs to be overly cautious in avoiding interaction with plaintiffs, which in turn harms 19 plaintiffs’ speech and association interests. On appeal, plaintiffs expressly acknowledge that 20 “[t]his vagueness claim is premised on the [restrictions’] chilling effect on protected speech and 21 association.” As plaintiffs do not assert a harm to their own interest in receiving due process of 22 law, this is precisely the sort of claim that the prudential standing doctrine is designed to 20
    437. 1 foreclose. Plaintiffs cannot make their First Amendment claims actionable merely by attaching 2 them to a third party’s due process interests. See Haitian Refugee Ctr. v. Gracey, 809 F.2d 794, 3 809 (D.C. Cir. 1987) (explaining that because due process rights “do not protect a relationship” 4 between a third party and a litigant, a plaintiff “could never have standing to challenge a statute 5 solely on the ground that it failed to provide due process to third parties not before the court”). 6 Plaintiffs’ due process claim is therefore dismissed for lack of prudential standing. 7 III. Equal Protection Claim 8 A. Plaintiffs Have “Competitive Advocate Standing” 9 Plaintiffs argue that the district court failed to undertake a separate analysis of 10 their Article III standing to bring an equal protection claim. Because we agree with plaintiffs that 11 the case law regarding constitutional standing for equal protection claims is distinct, and because 12 Planned Parenthood does not foreclose this claim on the merits, we address the question of 13 Article III standing with respect to this claim. As the case law and the legal theories involved are 14 quite different, this constitutional standing analysis does not inform the question on which we 15 reserved judgment above regarding constitutional standing to bring the First Amendment claims. 16 We find that plaintiffs do have constitutional standing to bring an equal protection claim. 17 With respect to the equal protection claim, the relevant portion of the complaint 18 reads: 19 The [use of the Standard Clause] violates the Equal Protection component of the Fifth 20 Amendment to the United States Constitution because it prohibits plaintiffs from 21 associating with USAID-recipient [foreign NGOs] for the purpose of promoting abortion 22 law reform, but permits other United States citizens and residents to associate with 23 USAID-recipient [foreign NGOs] for the purpose of opposing abortion law reform, and, 24 more generally, permits association with USAID-recipient [foreign NGOs] for the 25 purpose of rendering speech opposed to abortion more effective. 21
    438. 1 Am. Compl. ¶ 138. On appeal, plaintiffs flesh out the equal protection claim by explaining that 2 the use of the Standard Clause, “by prohibiting [foreign NGOs] from collaborating with 3 Plaintiffs, denies Plaintiffs the opportunity to compete on an equal footing with opponents of 4 abortion law reform.” 5 Though plaintiffs do not employ the term, this argument is essentially a theory 6 that this Court has dubbed “competitive advocate standing.” We have acknowledged the 7 possibility that a plaintiff may have standing to bring an equal protection claim where the 8 government’s allocation of a particular benefit “creates an uneven playing field” for 9 organizations advocating their views in the public arena. In re United States Catholic 10 Conference, 885 F.2d 1020, 1029 (2d Cir. 1989). In order to “satisfy the rule that he was 11 personally disadvantaged,” a plaintiff must “show that he personally competes in the same arena 12 with the party to whom the government has bestowed the assertedly illegal benefit.” Id. 13 Plaintiffs have standing under this theory. CRLP is an advocacy organization that 14 communicates its viewpoint regarding issues of abortion and reproductive rights, and it competes 15 with anti-abortion groups engaged in advocacy around the very same issues. The Standard 16 Clause has bestowed a benefit on plaintiffs’ competitive adversaries by rewarding their suppliers 17 of information, the foreign NGOs, with government grants, while withholding those grants from 18 suppliers of information who would deal with CRLP. This is precisely the type of situation that 19 the doctrine of competitive advocate standing contemplates. See id.; cf. Adarand Constructors, 20 Inc. v. Pena, 515 U.S. 200, 211 (1995) (finding, under the test for standing articulated in Lujan, 21 504 U.S. at 560, that a non-minority subcontractor had standing to contest a government policy 22 that gave a financial incentive to general contractors to give preference to minority 22
    439. 1 subcontractors in awarding subcontracts). 2 B. The Equal Protection Claim is Without Merit 3 Because this classification “neither proceeds along suspect lines nor infringes 4 fundamental constitutional rights,” it must “be upheld against equal protection challenge if there 5 is any reasonable state of facts that could provide a rational basis for the classification.” F.C.C. 6 v. Beach Communications, Inc., 508 U.S. 307, 313 (1993); see also Weinstein v. Albright, 261 7 F.3d 127, 140 (2d Cir. 2001). Here there can be no question that the classification survives 8 rational basis review. The Supreme Court has made clear that the government is free to favor the 9 anti-abortion position over the pro-choice position, and can do so with public funds. See Rust v. 10 Sullivan, 500 U.S. 173, 192-94 (1991). Plaintiffs’ equal protection challenge is thus without 11 merit. 12 CONCLUSION 13 For the reasons stated, we affirm the district court’s dismissal of this action, 14 though on different grounds. 23
    440. SOTOMAYOR, Circuit Judge, dissenting: The central issue in this case is whether the ne exeat provision in the Hong Kong custody order confers on either Mr. Croll or the Hong Kong court “rights of custody” within the meaning of the Hague Convention on the Civil Aspects of International Child Abduction (“Hague Convention” or “Convention”). The majority concludes that it does not and, therefore, that the district court lacked jurisdiction to order Christina’s removal to Hong Kong. Interpreting the text of the Convention in light of its object and purpose, and taking into account the relevant case law in this area, I reach the opposite conclusion. In my view, the majority seriously misconceives the legal import of the ne exeat clause and, in so doing, undermines the Convention’s goal of “ensur[ing] that rights of custody . . . under the law of one Contracting State are effectively respected in the other Contracting States.” Hague Convention, art. 1, done Oct. 25, 1980, T.I.A.S. No. 11670 at 4, 1343 U.N.T.S. 89, 98, reprinted in 51 Fed. Reg. 10,494, 10,498 (1986), implemented by the International Child Abduction Remedies Act (“ICARA”), 42 U.S.C. § 11601 et seq. I therefore respectfully dissent. The Hague Convention seeks “to protect children internationally from the harmful effects of their wrongful removal or retention and to establish procedures to ensure their prompt return to the State of their habitual residence, as well as to secure protection for rights of access.” Hague Convention, Preamble, 51 Fed. Reg. at 10,498. Significantly, the Convention draws a clear line between “rights of custody” and “rights of access,” reserving the remedy of return solely for breaches of the former. Compare Hague Convention, arts. 1, 3, id. (providing for the return of children removed or retained in violation of custody rights), with Hague Convention, art. 21, id. at 10,500 (providing that a party may petition for arrangements, short of the child’s return, to secure the effective exercise of
    441. access rights). In this regard, the majority correctly observes that “an order of return is available only for wrongful removals or retentions, and removals or retentions are wrongful only if they are ‘in breach of rights of custody.’” Ante at [11] (quoting Hague Convention, art. 3, 51 Fed. Reg. at 10,498) (emphasis in original). Article 3 of the Convention provides that the removal or retention of a child is “wrongful” where: (a) it is in breach of rights of custody attributed to a person, an institution or any other body, either jointly or alone, under the law of the State in which the child was habitually resident immediately before the removal or retention; and (b) at the time of removal or retention those rights were actually exercised, either jointly or alone, or would have been so exercised but for the removal or retention. Hague Convention, art. 3, 51 Fed. Reg. at 10,498 (emphasis added). Thus, Mr. Croll cannot succeed in securing Christina’s return to Hong Kong unless he can demonstrate that her removal was “in breach of rights of custody” and, furthermore, that at the time of Christina’s removal from Hong Kong, those rights of custody “were actually exercised, either jointly or alone, or would have been so exercised but for the removal.” Hague Convention, art 3, id. For the reasons discussed below, I conclude that Christina’s removal to the United States was “wrongful” under the Convention because (1) it constituted a “breach of rights of custody” jointly held by Mr. Croll and the Hong Kong court, and (2) Mr. Croll would have exercised his custody rights under the ne exeat clause in the custody order but for Christina’s removal from Hong Kong. 2
    442. I. Was Christina Removed from Hong Kong “in Breach of Rights of Custody”? Under the terms of the Hong Kong custody order, Ms. Croll is vested with “[t]he custody, care and control” of Christina,1 and Mr. Croll is vested with rights of “reasonable access.” Particularly relevant to this case, however, is the order’s further grant of rights to Mr. Croll under the ne exeat clause. The parties agree that under this provision, Ms. Croll may not remove Christina from Hong Kong without the consent of either Mr. Croll or the Hong Kong court.2 In other words, the ne exeat clause confers a veto power on Mr. Croll to block Christina’s international relocation, unless the Hong Kong court explicitly approves such removal. In essence, the ne exeat clause endows Mr. Croll with significant decisionmaking power: absent an order of the Hong Kong court to the contrary, he can require that Christina remain in Hong Kong or, alternatively, he can use his veto power as leverage to influence Ms. Croll’s selection of the destination country. Because Mr. Croll may not invoke the Convention’s return remedy based on his “reasonable access” rights, the issue in this case is whether he may secure Christina’s return under the Convention by virtue of his rights under the ne exeat clause. 1 The majority states that the custody order “confers the sole ‘custody, care and control’” upon Ms. Croll. Ante at [2] (emphasis added); see also id. at [18] (stating that “the custody order awards custody solely to the mother”) (emphasis added). However, nowhere does the Hong Kong court use the word “sole” or “solely” in connection with Ms. Croll’s custody rights. 2 The Hong Kong custody order also provides that “[e]ither parent may request the Immigration Department not to issue passports allowing the said child to go abroad without his/her knowledge.” 3
    443. The majority mischaracterizes the issue as being a question of whether the ne exeat clause “transmute[s] access rights into custody rights under the Convention.” Ante at [28]. Clearly, the ne exeat clause works no such magic. In my view, the question presented is whether the ne exeat clause—wholly independent of Mr. Croll’s access rights—confers “rights of custody” under the Convention. The Convention’s text, object and purpose, as well as the relevant case law in this area, convincingly direct an answer in the affirmative. A. The Text, Object, and Purpose of the Convention The critical interpretive challenge in this case involves the definition of “rights of custody” as used in the Convention. The majority begins this undertaking by surveying a host of American dictionaries to support its “intuition that custody is something other and more than a negative right or veto.” Ante at [14]. Relying on these sources, the majority finds that the “custody of a child entails the primary duty and ability to choose and give sustenance, shelter, clothing, moral and spiritual guidance, medical attention, education, etc., or the (revocable) selection of other people or institutions to give these things.” Ante at [15]. While traditional American notions of custody rights are certainly relevant to our interpretation of the Convention, the construction of an international treaty also requires that we look beyond parochial definitions to the broader meaning of the Convention, and assess the “ordinary meaning to be given to the terms of the treaty in their context and in the light of [the Convention’s] object and purpose.” Vienna Convention on the Law of Treaties, May 23, 1969, art. 31.1, 1155 U.N.T.S. 331, 340 (stating general rule on the interpretation of treaties); see 4
    444. also Restatement (Third) of Foreign Relations Law § 325 (1987) (same). Contrary to the majority’s position that “[n]othing in the Hague Convention suggests that the drafters intended anything other than this ordinary understanding of custody,” ante at [15], the Convention and its official history reflect a notably more expansive conception of custody rights. The report containing the official history and commentary on the Convention clarifies that “the intention [of the Convention] is to protect all the ways in which custody of children can be exercised.” Elisa Pérez- Vera, Explanatory Report to the Hague Conference on Private International Law, in 3 Acts and Documents of the Fourteenth Session (Child Abduction) 426, para. 71 (1980) (emphasis in original) (“Pérez-Vera Report”). This broad notion of custody rights is also consistent with Article 3, which provides that “rights of custody” may arise from a variety of sources, including by “operation of law or by reason of a judicial or administrative decision, or by reason of an agreement having legal effect under the law of [the child’s country of habitual residence].” Hague Convention, art. 3, 51 Fed. Reg. at 10,498. In this way, the Convention plainly favors “a flexible interpretation of the terms used, which allows the greatest possible number of cases to be brought into consideration.” Pérez-Vera Report, para. 67. Consequently, in determining whether the rights arising under a ne exeat clause constitute “rights of custody” under the Convention, I discern an intent of inclusion rather than exclusion, so as to effectuate the drafters’ goal of making the treaty applicable to all possible cases of wrongful removal. 5
    445. Although the treaty does not generally define its legal terms, see Pérez-Vera Report, para. 83, the risk that “an incorrect interpretation of [custody and access rights] would . . . compromis[e] the Convention’s objects” led the drafters to include Article 5, which offers further guidance on the meaning of the term “rights of custody.” See Pérez-Vera Report, para. 83. I note, however, that the provision was left deliberately vague due to the drafters’ failure to agree on a more precise definition. See Pérez-Vera Report, para. 84 (“[S]ince all efforts to define custody rights in regard to [particular situations] failed, one has to rest content with the general description given [in the text].”). Article 5 provides that: For the purposes of this Convention – (a) “rights of custody” shall include rights relating to the care of the person of the child and, in particular, the right to determine the child’s place of residence; . . . Hague Convention, art. 5, 51 Fed. Reg. 10,498 (emphasis added); see also Pérez-Vera Report, para. 84 (noting that under Article 5, “rights of custody” include those rights relating to the care of the child, and that the Convention seeks to clarify this otherwise general definition “by emphasizing, as an example of the ‘care’ referred to, the right to determine the child’s place of residence.”). As I interpret the Convention, rights arising under a ne exeat clause include the “right to determine the child’s place of residence” because the clause provides a parent with decisionmaking authority regarding a child’s international relocation. Thus the ne exeat clause vests both Mr. Croll and the Hong Kong court with “rights of custody” for the purposes of the Convention. See Hague Convention, art. 5, 51 Fed. Reg. at 10,498. 6
    446. A parent’s ne exeat rights fit comfortably within the category of custody rights the Convention seeks to protect. The Convention states at its outset that its object is, along with returning children wrongfully removed from their habitual residence, “to ensure that rights of custody . . . under the law of one Contracting State are effectively respected in the other Contracting States.” Hague Convention, art. 1, 51 Fed. Reg. at 10,498. The Pérez-Vera report explains that the problem with which the Convention deals . . . derives all of its legal importance from the possibility of individuals establishing legal and jurisdictional links [in the new country] which are more or less artificial. In fact, resorting to this expedient, an individual can change the applicable law and obtain a judicial decision favourable to him. [Such a decision] bears a legal title sufficient to ‘legalize’ a factual situation which none of the legal systems involved wished to see brought about. Pérez-Vera Report, para. 15. At its core, therefore, the Convention’s return remedy targets those individuals who cross international borders, presumably in search of a friendlier forum, flouting the custody law of the child’s home country in the process. See Blondin v. Dubois, 189 F.3d 240, 245- 46 (2d Cir. 1999) (describing the Convention’s purpose as “‘preserv[ing] the status quo and . . . deter[ring] parents from crossing international boundaries in search of a more sympathetic court.’”) (quoting Friedrich v. Friedrich, 983 F.2d 1396, 1400 (6th Cir. 1993)). In light of the Convention’s broad purpose, the concept of “wrongful removal” clearly must encompass violations of ne exeat rights. When a parent takes a child abroad in violation of ne exeat rights granted to the other parent by an order from the country of habitual residence, she nullifies that country’s custody law as effectively as does the parent who kidnaps a child in violation of the rights of the parent with physical custody of that child. 7
    447. Moreover, where, as here, the parent seeks a custody order in the new country, she seeks to legitimize the very action—removal of the child—that the home country, through its custody order, sought to prevent. To read the Convention so narrowly as to exclude the return remedy in such a situation would allow such parents to undermine the very purpose of the Convention. B. The Majority’s Approach In reaching the opposite conclusion, the majority contends that “rights of custody,” as used in the Convention, refers to a “bundle of rights” of which a parent must possess a certain portion in order to be protected by the Convention, and that possession of only one of those rights — in this case, the “right to determine the child’s place of residence” by exercising and leveraging a veto power over the child’s international relocation —is insufficient to confer custody on the party possessing that power. See ante at [16]. In my view, however, the Convention’s definition of “rights of custody” contemplates a bundle of rights that are protected regardless of whether a parent holds one, several or all such custody rights, and whether the right or rights are held singly or jointly with the other parent. In fact, the Convention expressly protects joint custody rights, see Hague Convention, art. 3, 51 Fed. Reg. at 10,498, which may assume a number of forms, including situations in which one parent possesses sole physical custody of the child but shares certain decisionmaking authority with the other parent. The Convention contains no indication that in such an arrangement, a parent must possess some minimum number of rights of custody in order to qualify for protection. The majority also maintains that a parent’s ne exeat right does not equate with Article 8
    448. 5’s “right to determine the child’s place of residence” because the latter right necessarily entails “specific choices” regarding the child’s living situation rather than simply decisions regarding the country in which she lives. See ante at [16-18]. Like the majority’s definition of “custody,” however, this conclusion ignores the basic international character of the Hague Convention. While such “specific choices” certainly constitute facets of custody, the broader decision as to whether a child will live in England or Cuba, Hong Kong or the United States, is precisely the kind of choice the Convention is designed to protect. See Pérez-Vera Report, para. 56 (“Although the Convention does not contain any provision which expressly states the international nature of the situations envisaged, such a conclusion derives as much from its title as from its various articles. . . . [T]he international nature of the Convention arises out of a factual situation, that is to say the dispersal of members of a family among different countries.”). The Hague Convention provides a remedy not when a parent moves the child from city to suburb or from home to boarding school, but when he or she transports the child across national borders. In light of this international context, the term “place of residence,” as used in the Convention, logically contemplates decisions regarding international relocation. Accordingly, the right to choose the country in which a child lives, like the authority over the child’s more specific living arrangements, constitutes a “right to determine the child’s place of residence” under Article 5, and thus a “right of custody” under the Convention.3 3 To be sure, the right to prevent a child’s removal from her home country does not constitute an absolute right “to determine the child’s place of residence.” That a right is limited, however, does not render it meaningless for purposes of the Hague Convention. See Pérez-Vera Report, para. 71 (characterizing “joint custody” as “dividing the responsibilities inherent in custody 9
    449. The majority avoids this conclusion by asserting that the power to determine a child’s country of residence “protects rights of custody and access alike, and [gives] no clue as to who has custody.” Ante at [17]. But while such a power may have the effect of ensuring a parent’s reasonable access, and in fact may be included in a custody order for precisely that purpose, ne exeat rights circumscribe the choices of the parent with physical custody of the child in a way that “reasonable access” rights do not. Absent a ne exeat clause, the international relocation of a child does not necessarily violate the other parent’s access rights; the parents still may work out an arrangement that satisfies the rights of “reasonable access” even across international borders. On the other hand, when a parent expatriates her child without securing the necessary consent, she has, by definition, violated the other parent’s ne exeat rights. The majority also posits that the Convention would be “unworkable” if it provided the return remedy for violations of a parent’s ne exeat rights. See ante at [20-21]. Because an order of return can require only Christina, and not Ms. Croll, to return to Hong Kong, the majority claims that “we cannot plausibly read the Convention to compel the removal of a child from a parent who exercises all rights of care to a country in which no one has that affirmative power or duty.” Ante at [21]. The majority mistakenly assumes that the custody order in a given case is the sole source of a parent’s rights and duties vis-a-vis his or her child. To the contrary, a parent’s duty to care for a child, like his or her rights between both parents”). Furthermore, that a right is a veto or “negative right” does not diminish its status as a right. See, e.g., Cruzan v. Director, Missouri Dep’t of Health, 497 U.S. 261, 281 (1990) (recognizing the due process right to refuse life-sustaining medical treatment). 10
    450. rights of custody, may arise from many sources, including the law of the country of habitual residence. That the custody order in this case granted “custody, care and control” of Christina to Ms. Croll, therefore, does not direct the conclusion that Mr. Croll will have no responsibility to care for Christina upon her return to Hong Kong. 4 I therefore reject the majority’s dire forecast that ordering Christina’s return, without Ms. Croll at her side, risks leaving Christina helpless in Hong Kong without parental care. 4 The majority faults the dissent for its “assumption” that a court will “alter custody rights” upon the child’s return, and raises fears that absent such alteration, the child will be uncared for upon her return to Hong Kong. See ante at [23]. Christina’s care upon her return is neither premised on assumptions nor relevant to the issue before us. First, it strains credulity to suggest that a father who, as here, searches the world for his child to get her back and files a petition in a foreign forum in order to do so, would, upon succeeding in his efforts, simply permit his child to stand abandoned in the airport upon her return. The majority confuses physical care of a child with legally-ordered custody. Furthermore, if Christina’s care upon her return to Hong Kong were really a concern in the instant case, the appropriate remedy would not be reversal but a remand to the district court to assess the parties’ intentions. See Feder v. Evans-Feder, 63 F.3d 217, 226 (3d Cir. 1995) (“[I]n order to ameliorate any short-term harm to the child, courts in the appropriate circumstances have made return contingent upon ‘undertakings’ from the petitioning parent.”); In re Walsh, 31 F. Supp. 2d 200, 207 (D. Mass. 1998) (“Numerous courts granting petitions under the Convention have recognized the legitimacy of exacting appropriate undertakings from the parents designed to ensure that the children will be cared for properly during transit and that no harm will come to the children pending disposition in the country of habitual residence.”) (citations omitted), aff’d in part, rev’d in part on other grounds sub nom. Walsh v. Walsh, __ F.3d __, 2000 WL 1015863 (1st Cir. July 25, 2000). Second, the issues concerning Christina’s custody upon her return are beyond the scope of the Convention, which deals solely with returning a child to the country of habitual residence so that its courts, and not a foreign court in the country to which the child has been wrongfully removed, can adjudicate custody rights with respect to that child. This concept underlying the Convention — that the child is best served by entrusting decisions regarding his or her custody to the courts of the child’s country of habitual residence — stands in direct contradiction to the majority’s parochial view that foreign courts cannot be trusted in the same manner as American courts to competently make necessary decisions regarding the child. See ante at [23] (“on this point the dissent is generalizing from local American law”). 11
    451. Moreover, the majority’s characterization of a return remedy for violations of ne exeat rights as unworkable fails to account for the Convention’s protection of any number of joint custody arrangements in which the parents trade physical custody or in which one parent possesses physical custody and the other parent contributes to decisions about the child’s upbringing. By the majority’s reasoning, were the parent with physical custody to remove the child from the country of habitual residence, the court would have no power to return the child, because no adult would be required to care for him or her upon return. Such a conclusion, however, would largely eviscerate the Convention’s protection of joint custody rights. Far from being unworkable, the application of the return remedy in the context of ne exeat violations directly and fully advances the Convention’s goal of preventing parents from unilaterally circumventing the home country’s custody law. In contrast to access right violation cases where returning the child to her country of habitual residence would not itself guarantee the effective exercise of such rights, ordering the return of a child based on a ne exeat violation will, in and of itself, give full effect to a parent’s ne exeat rights. Finally, the majority cites to a string of authorities under the caption “Intent of the Drafters” to support its narrow reading of the Convention. See ante at [21-26]. With one exception, these authorities stand only for the unremarkable proposition that under the Convention, the return remedy is unavailable for breaches of parents’ access rights.5 In sum, those authorities shed no light on 5 The majority quotes A.E. Anton, the former chairman of the Hague Conference Commission, who opines that “breach of a right simply to give or to withhold consent to changes in a 12
    452. the issue relevant here, i.e., whether ne exeat rights constitute “rights of custody” for the purposes of the Convention. C. International Case Law While not essential to my conclusion that ne exeat rights constitute “rights of custody” under the Convention, I note that my analysis is consistent with the decisions of most foreign courts to consider the issue. See generally Air France v. Saks, 470 U.S. 392, 404 (1985) (in construing the terms of a treaty, “the opinions of our sister signatories [are] entitled to considerable weight.”) (quoting Benjamins v. British European Airways, 572 F.2d 913, 919 (2d Cir. 1978)). Given the desirability child’s place of residence is not to be construed as a breach of rights of custody in the sense of Article 3.” Ante at [22] (quoting A.E. Anton, The Hague Convention on International Child Abduction, 30 Int’l & Comp. L.Q. 537, 546 (1981)). Although Mr. Anton’s views support the majority’s interpretation of the Convention, the majority neglects to emphasize that his article represents only his personal views and not the official legislative history of the Convention. See Anton, supra at 537 (“This paper, however, must not be taken to reflect any views other than those of the author.”). His article is therefore appropriately viewed as simply the opinion of one scholar. Other scholars, in contrast, have concluded that ne exeat rights do constitute “rights of custody” under the Convention. See, e.g., Paul R. Beaumont & Peter E. McEleavy, The Hague Convention on International Child Abduction 72- 73 (1999) (“If an individual is a child’s sole custodian there are, prima facie, no restrictions on him relocating with that child. If there are such restrictions, that implies that the custody right must in some way be limited. Where this is so it must be that another body or individual holds a corresponding right in relation to the child. Therefore, should the primary carer remove the child abroad, . . . he would have breached the custody rights of the other party, if their consent had not been sought.”) (emphasis in original); Linda Silberman, “Custody Orders Under the Hague Abduction Convention,” in A New Vision for a Non-Violent World: Justice for Each Child (Proceedings of the 4th Biennial International Conference of the International Association of Women Judges, 1999), at 236 (arguing that in cases of non-removal order violations, “a failure to recognize the Convention remedy of return would be inconsistent with the careful compromise [between custody and access rights] that the Convention definition [of custody rights] has put in place.”). 13
    453. of uniformity in treaty interpretation, see Denby v. Seaboard World Airlines, Inc., 737 F.2d 172, 176 n.5 (2d Cir. 1984), these cases lend support to my understanding of the Convention. Most foreign courts addressing this question have interpreted the notion of “rights of custody” broadly in light of the Convention’s purpose and structure. The Family Court of Australia, for example, has characterized the “spirit of the Convention” as ensuring “that children who are taken from one country to another wrongfully, in the sense of in breach of court orders or understood legal rights, are promptly returned to their country so that their future can properly be determined within that society.” In the Marriage of: Jose Garcia Resina Appellant/Husband and Muriel Ghislaine Henriette Resina Respondent/Wife, Appeal No. 52, 1991 (Fam.) (Austl.), para. 26. Accordingly, the court held that the custody order at issue—which provided reciprocal ne exeat rights for each parent—created “rights of custody” in the otherwise non-custodial father. The English Court of Appeal has employed a similarly broad reading of the Convention, holding that Article 5 “may in certain circumstances extend the concept of custody beyond the ordinarily understood domestic approach” so as to ensure “that within its scope [the Convention] is to be effective.” C. v. C., [1989] 1 W.L.R. 654, 658 (C.A.) (Eng.). In C. v. C., the court ordered the return of a child where an Australian order granted custody to the child’s mother, but also provided that the father and mother would remain “joint guardians” and that neither parent could remove the child from Australia without the consent of the other. See id. at 656. Interpreting the language of Article 5, the court found that because the Australian custody order allowed the father to exercise a measure of control over the child’s place of residence, the father possessed “custody rights” within the meaning of 14
    454. the Convention:6 [T]he father had, in my judgment, the right to determine that the child should reside in Australia or outside the jurisdiction at the request of the mother. . . . [He has] some control over not only the child leaving the jurisdiction, but also as a place to which the child was going, and not only the country; for instance, to live in London under suitable circumstances. . . . The father does not have the right to determine the child’s place of residence within Australia but has the right to ensure that the child remains in Australia or lives anywhere outside Australia only with his approval. Id. at 658. The Israeli High Court of Justice, when presented with facts nearly identical to those in the instant case, similarly interpreted “rights of custody” to encompass a parent’s rights under a non- removal order. See C.A. 5271/92, Foxman v. Foxman (H.C. 1992) (Isr.) (finding that the Convention’s definition of “custodial rights” should be “broadly construed,” so as to cover cases in which parental consent is required before a child is taken out of the country); cf. C.A. 1648/92, Tourna v. Meshulem (H.C. 1992) (Isr.) (finding “rights of custody” in a father who, by virtue of a joint custody order, had authority to refuse consent to the child’s change in residence). In addition to these cases, which address custody rights held by a parent with ne exeat rights, the English Court of Appeal has also held that a court entering the custody order in the child’s 6 The majority attempts to distinguish C. v. C. as involving an order of joint guardianship, see ante at [27], but the C. v. C. court explicitly relied on the language of the ne exeat provision and not the joint guardianship clause in determining that the father possessed “rights of custody” under the Convention. See [1989] 1 W.L.R. at 657-58 (“[The lower court judge] heard argument as to the effect of . . . joint guardianship. . . . [Accordingly, t]he judge’s attention does not appear to have been sufficiently drawn to the effect on the definition in article 5 of the Convention of clause 2 of the November 1986 order, that neither parent should remove the child from Australia without the consent of the other.”). 15
    455. place of habitual residence may itself possess “rights of custody” under the Convention in certain circumstances. See B. v. B., [1993] 2 All E.R. 144 (C.A.) (Eng.). The court in B. v. B. noted that under Article 3 an “institution or other body” as well as an individual may hold custody rights and thus concluded that the removal of the child by the parent with physical custody in breach of an interim custody order conditioned on the child remaining within the jurisdiction violated the rights of both the other parent and the court issuing the interim custody decree. See id. at 148-49. Echoing the reasoning of C v. C, the court found that because the restriction required the parent with physical custody to remain in the court’s jurisdiction, and thus impliedly gave the court and the parent without physical custody the right to veto an international move, it vested both with the power to determine the child’s residence. See id. at 148-49. The court therefore affirmed the order of return on the ground that the child’s removal had been wrongful within the meaning of the Convention. See id. at 153.7 These cases reflect strong support among our sister signatories for the proposition that “rights of custody” are implicated where a custody order vests either a parent or the court with the power to block a parent with physical custody from deciding to expatriate her child.8 While there are 7 It is not apparent to me how the majority turns this case — involving a veto power over international relocation possessed by a court and a parent — into a decision that requires return of the child “whenever a court enters a custody order” that is violated, regardless of the terms of that order. See ante at [19]. This dissent has never suggested that Mr. Croll or the court would be entitled to an order of return if they did not possess ne exeat rights, nor does B v. B or this dissent suggest that return would be required in the scenario posed by the majority of “expatriation in derogation solely of parental rights of access.” See ante at [19]. 8 Although the majority states that “we and the district court are the only courts in the United States,” ante at [9], to consider the issue before us, I note that two American courts have also recognized “rights of custody” in connection with custody orders containing a ne exeat clause. See 16
    456. several cases in other jurisdictions that are certainly in tension with this view, I find the reasoning in those cases unpersuasive. At least one French court has determined that a custody order requiring the mother to raise her children in England and Wales did not create custodial rights in the father because such a reading would infringe on the mother’s right to expatriate. See T.G.I. Periguex, Mar. 17, 1992, Ministere Public v. Mme Y., D.S. Jur. 1992 (Fr.). However, the court in Mme. Y. did not address the meaning of Article 5’s “right to determine the child’s place of residence” provision and instead focused on the mother’s expatriation rights under the European Convention for the Protection of Human Rights and Fundamental Freedoms. See id. at 315-16. But deciding a Hague Convention case on the ground that the custodial parent must remain free to expatriate her child begs the crucial interpretive question of who, for purposes of the Convention, are “custodial parents” in the first place. Nothing in the Convention suggests that one parent’s right to expatriate overrides another parent’s rights of custody. On the contrary, the paramount importance the Convention places on custodial rights suggests that where custodial rights and expatriation rights conflict, the latter must yield to the former. To adopt another reading would, among other difficulties, make the Convention’s protection of joint custody impossible, as joint custody by its very nature limits each parent’s unilateral decisionmaking power, David S. v. Zamira S., 151 Misc.2d 630, 574 N.Y.S.2d 429 (N.Y. Fam. Ct. 1991); Janakakis- Kostun v. Janakakis, 6 S.W.3d 843 (Ky. Ct. App. 1999), review denied (Dec. 9, 1999), petition for cert. filed, 68 U.S.L.W. 3595 (U.S. Mar. 8, 2000) (No. 99-1496). In my view, however, these cases are of limited utility because they fail precisely to define “custodial rights” or to differentiate them from access rights. 17
    457. including his or her power to relocate to another country with the child. In my view, therefore, the legal presumption against restrictions on expatriation answers little.9 Apart from the Mme. Y. decision, two other cases are in tension with the proposition that ne exeat rights constitute “rights of custody” under the Convention. In two separate decisions, the Canadian Supreme Court has suggested — in dicta — that the Convention’s “wrongful removal” provision does not cover cases in which a parent acts in violation of an express provision in a custody order granting ne exeat rights. In the first, Thomson v. Thomson, [1994] 119 D.L.R. 4th 253 (Can.), the court ordered a child’s return based on an interim non-removal order in order to “preserve jurisdiction in the Scottish court to decide the issue of custody on its merits in a full hearing at a later date,” but noted in dicta that such a remedy would be unavailable for violation of a final non-removal order because the purpose of such an order was simply to “ensure permanent access to the non- custodial parent.” Id. at 281. In the second case, D.S. v. V.W. [1996] 134 D.L.R. 4th 481 (Can.), the court held that a return remedy was not available under the Convention for violation of an implicit removal restriction in a custody order, and — relying in part on the dicta in Thomson relating to express provisions in permanent custody orders — stated that a violation of such an implicit restriction would concern only access rights, not custodial rights. Id. at 501-06. However, the court nevertheless ultimately upheld the lower court’s order of return on the alternate ground that such return was in the 9 I also note that courts in France appear divided on this issue. See Martha Bailey, “Rights of Custody” Under the Hague Convention, 11 B.Y.U. J. Pub. L. 33, 40 (1997) (discussing French cases). 18
    458. best interests of the child under Quebec domestic legislation. Id. at 516-17. For the reasons explained above, supra I.A., I am unpersuaded by the argument that ne exeat clauses in permanent non-removal orders relate solely to access rights, the view endorsed by the Canadian Supreme Court. Nor do I consider significant the Canadian Supreme Court’s emphasis on the distinction between interim and permanent custody orders. To be sure, a court issuing an interim custody order has a strong interest in preventing a child’s removal before it has the opportunity to make its final custody determination. But nothing in the Convention’s language or official history supports the notion that this interest is any more important than the court’s interest in enforcing the final custody order once issued. The dichotomy between an interim and permanent custody order is, therefore, for the purposes of the Convention, a distinction without a difference. I note also that while the D.S. decision to uphold the order of return was unanimous, six (out of nine) justices expressed reservations regarding the opinion’s analysis of custodial rights and obligations, see 134 D.L.R. 4th at 484, 518; see also Bailey, supra, at 49, thereby raising serious doubts as to whether the opinion’s conception of ne exeat clauses in relation to the Convention truly represents the rule in Canada. Scholars have also strongly criticized the Canadian interpretation of custody rights under the Convention. See, e.g., Bailey, supra, at 42-50; Linda Silberman, “Custody Orders Under the Hague Abduction Convention,” in A New Vision for a Non-Violent World: Justice for Each Child (Proceedings of the 4th Biennial International Conference of the International Association of Women Judges, 1999), at 235-240. Therefore, following what I consider to be the more compelling reasoning of the English, Australian, and Israeli cases, I would join the courts of those 19
    459. countries in finding that rights arising under a ne exeat clause constitute “rights of custody” for the purposes of the Hague Convention. II. Did Mr. Croll Or The Hong Kong Court “Actually Exercise” Ne Exeat Rights? Apart from the central issue of whether ne exeat rights constitute “rights of custody” under the Convention, the majority also holds that Mr. Croll’s petition fails to satisfy the requirement of Article 3(b) of the Convention, which provides: The removal or retention of a child is to be considered wrongful where . . . at the time of removal or retention those rights [of custody] were actually exercised, either jointly or alone, or would have been so exercised but for the removal or retention. Hague Convention, art. 3(b), 51 Fed. Reg. at 10,498 (emphasis added). According to the majority, “[t]he right conferred by the ne exeat clause is not one that Mr. Croll ‘actually exercised,’ and it is circular to say that he would have exercised it but for Christina’s removal, because the right itself concerns nothing but removal itself, and would never have been exercised had Mrs. Croll been content to stay in Hong Kong during Christina’s minority.” Ante at [19] (emphasis in original).10 This description mischaracterizes the right that a ne exeat clause creates. The right given to Mr. Croll and the Hong Kong court by the ne exeat clause was the authority to withhold or grant consent to removing Christina from Hong Kong. Had they refused to grant Ms. Croll permission to take Christina to the United States, or even had they agreed to grant 10 Ms. Croll does not argue on appeal that Mr. Croll’s petition was defective under Article 3(b) of the Convention. The majority reaches this issue sua sponte. 20
    460. permission, they would have “actually exercised” the custody rights granted by the custody order. While I agree that neither Mr. Croll nor the court did, in fact, “actually exercise” this right, it seems clear to me that Ms. Croll’s conduct in removing Christina without the necessary consent was precisely what prevented them from doing so. Because Ms. Croll deprived Mr. Croll and the court of the opportunity to exercise their veto power by surreptitiously removing Christina from Hong Kong without first seeking consent, the ne exeat right is one that “would have been so exercised” but for Christina’s unlawful removal. Article 3(b) therefore poses no barrier to finding that Christina’s removal was wrongful under the Convention. For the foregoing reasons, I conclude that Christina’s removal from Hong Kong to the United States was “wrongful” under the Convention because her removal (1) constituted a “breach of rights of custody” jointly held by Mr. Croll and the Hong Kong court, and (2) Mr. Croll or the court — or both — would have exercised their veto rights under the ne exeat clause but for Christina’s removal from Hong Kong. Accordingly, I would affirm the district court’s decision to grant Mr. Croll’s petition for an order of return. 21
    461. United States Court of Appeals FOR THE SECOND CIRCUIT __________________ At a stated Term of the United States Court of Appeals for the Second Circuit, held at the United States Courthouse, Foley Square, in the City of New York, on the day of two thousand, ______________________________________________ Lee N. Koehler, Plaintiff-Appellant, v. 98-9624 Bank of Bermuda (New York) Ltd. Defendant-Appellee, ______________________________________________ A request for a vote as to whether the panel decision should be reconsidered sua sponte by the Court in banc having been made by a judge of the Court, and a poll of the judges in regular active service having been taken, a majority of the Court has voted not to reconsider the decision in banc. The mandate shall therefore issue. Judges Leval, Calabresi and Sotomayor dissent. FOR THE COURT: Roseann B. MacKechnie, Clerk By:_____________________ _______ Beth J. Meador, Administrative Attorney
    462. 1 UNITED STATES COURT OF APPEALS 2 FOR THE SECOND CIRCUIT 3 4 August Term, 1999 5 6 (Argued: August 30, 1999 Decided: April 10, 2000) 7 (On Reconsideration by the Court In Banc Dissent: September 28, 2000) 8 9 Docket No. 98-9624 10 11 ____________________________________________________________________________ 12 13 LEE N. KOEHLER, 14 15 Petitioner-Appellant, 16 17 v. 18 19 THE BANK OF BERMUDA (NEW YORK) LIMITED, a New York Corporation, THE BANK 20 OF BERMUDA LIMITED, a Bermuda Corporation, REEFS BEACH CLUB LIMITED, a 21 Bermuda Corporation, and A. DAVID DODWELL, a Bermuda citizen, 22 23 Defendants-Appellees. 24 ____________________________________________________________________________ 25 26 27 28 SOTOMAYOR, Circuit Judge, with whom Judge LEVAL concurs, dissenting from the denial of 29 rehearing in banc: 30 31 Judge Calabresi dissents in a separate opinion. 32 33 Federal courts may, under their alienage jurisdiction, hear controversies between 34 “citizens of a State and citizens or subjects of a foreign state.” 28 U.S.C. § 1332(a)(2) (1994). Based 35 upon a prior holding of this Court in Matimak Trading Co. v. Khalily, 118 F.3d 76 (2d Cir. 1997), 36 cert. denied, 522 U.S. 1091 (1998), the panel in this case concluded that Bermuda corporations and a 37 Bermuda citizen were not “citizens or subjects of a foreign state,” and, therefore, that a controversy 38 involving such parties was not within the alienage jurisdiction of the federal courts. Koehler v. Bank of 39 Bermuda (New York) Ltd., 209 F.3d 130, 139 (2d Cir. 2000). Because a rehearing in banc would USCA Order 4
    463. 1 provide a much-needed opportunity for the full Court to reexamine the flawed and internationally 2 troublesome position that corporations and individuals from territories of the United Kingdom do not fall 3 within the alienage jurisdiction of the federal courts, I dissent from the denial of the petition for rehearing 4 in banc. 5 6 I. 7 This is a question of “exceptional importance.” Fed. R. App. P. 35(a)(2). Its import 8 reaches well beyond our government, to our relations with foreign nations, and the access of foreign 9 entities and individuals to the federal courts. Both the Executive Branch and the government of the 10 United Kingdom of Great Britain and Northern Ireland have asked that we reconsider the reasoning we 11 employed in Matimak. This Circuit’s understanding of the scope of alienage jurisdiction is squarely in 12 conflict with that of the other circuit courts that have addressed this question. When issues of such 13 enduring significance are presented, I believe that the Court in banc should reexamine the merits of its 14 conclusion to ensure that substantial numbers of individuals and corporations are not erroneously 15 deprived of access to our federal courts. 16 The defendants in this case include Bermuda corporations and a Bermuda citizen. 17 Bermuda is not recognized by our State Department as an independent state. It is, rather, a “British 18 Overseas Territory.”1 Essential to this case is the fact that despite the myriad ways in 1 The British Overseas Territories (also referred to as “Dependent Territories”) include Anguilla, Bermuda, British Indian Ocean Territory, the British Virgin Islands, the Cayman Islands, the Falkland Islands, Gibralter, Montserrat, the Pitcairn Islands, Saint Helena and dependencies, South Georgia and the South Sandwich Islands, and the Turks and Caicos Islands. See Brief Amicus Curiae of the Government of the United Kingdom of Great Britain and Northern Ireland in Support of Matimak Trading Co. as Petitioner for Writ of Certiorari at 6 n.5, Matimak Trading Co. v. Khalily (97-893) (hereinafter U.K. Matimak Brief). Some of the British Overseas Territories have become important commercial centers. As of USCA Order 4 3
    464. 1 which the United Kingdom exercises dominion over Bermuda, British law terms Bermudan citizens and 2 corporations “nationals,” but not “subjects,” of the United Kingdom. See United Kingdom 3 Government’s Diplomatic Service Procedure Manual, Vol. 7, Annex 1, Rules 1(b), 2(a) (1996). 4 Previously, this Court in Matimak held that a corporation organized under the laws of Hong Kong 5 could not sue New York defendants in federal court because Hong Kong was, at the time, a 6 Dependent Territory of the United Kingdom, and therefore the plaintiff corporation was not a “citizen 7 or subject” of a foreign “state.”2 Relying on Matimak, the panel here concluded that Bermuda 8 corporations and a Bermuda citizen were not “citizens or subjects of a foreign state,” 28 U.S.C. § 9 1332(a)(2) (1994), and therefore not within our alienage jurisdiction. 10 The people of Bermuda would undoubtedly be surprised to learn that they are 11 “stateless.” But this is precisely the conclusion upon which these decisions rest. See Matimak, 118 12 F.3d at 86 (“[Plaintiff-Hong Kong corporation] is thus stateless. And a stateless person–the proverbial 13 man without a country–cannot sue a United States citizen under alienage jurisdiction.”). Having found 14 such entities or individuals “stateless,” the panels in this case and in Matimak had no difficulty denying 15 these litigants access to the federal courts because “[t]he raison d’etre of alienage jurisdiction is to avoid 16 entanglements with other sovereigns that might ensue from failure to treat the legal controversies of 1997, 563 banks and 34,169 other companies were incorporated in the Cayman Islands, at least 8,224 businesses were incorporated in Bermuda, at least 100,000 companies were incorporated in the British Virgin Islands, and 12,911 companies were incorporated in the Turks and Caicos. See id. at 10-11. Several of these territories, including the Cayman Islands and Bermuda, are considered significant tax havens. See Mark Baker, Lost in the Judicial Wilderness: The Stateless Corporation After Matimak Trading, 19 Nw. J. Int'l L. & Bus. 130, 132 n.8 (1998) (noting that the holding in Matimak adds an “element of unpredictability” to the world of tax structuring). 2 The Matimak decision has been extensively criticized by commentators. See III Finance Ltd. v. Aegis Consumer Funding Group, Inc., No. 99 Civ. 2579, 1999 WL 1080371, at *2 (S.D.N.Y. Nov. 30, 1999) (collecting sources). USCA Order 4 4
    465. 1 aliens on a national level.” Matimak, 118 F.3d at 82 (internal quotation omitted). These panels 2 implicitly reason that absent a “state,” there is no sovereign to offend and therefore no cause to provide 3 federal alienage jurisdiction. 4 This assurance is undermined by the strong reaction to our decisions by the United 5 Kingdom.3 Whatever other intention the panels here and in Matimak may have had, there can be no 6 doubt that the fundamental purpose of alienage jurisdiction--to void offense to foreign nations--is 7 frustrated by the Matimak decision and its further application by this panel. Paradoxically, the country 8 we offend by these holdings is not only a strong ally, but the very country the drafters of the alienage 9 jurisdiction provision had in mind more than two hundred years ago when they sought to open the 10 federal courts to foreign litigants. See Kevin R. Johnson, Why Alienage Jurisdiction? Historical 11 Foundations and Modern Justifications over Disputes Involving Noncitizens, 21 Yale J. Int’l L. 1, 12 7-8 (1996) (noting the failure of state courts to enforce debts owed to British creditors following the 13 Revolutionary War). 14 This Court, in Matimak, attempted to shift responsibility for the disturbing 15 consequences of its reasoning to the Executive Branch. Because the Department of State maintains that 16 British Overseas Territories are not independent “states,” the Matimak court reasoned that it was 3 See U.K. Matimak Brief at 9 (“The United Kingdom is keenly concerned that the citizens and corporations of its Dependent Territories be able to bring and defend suits in neutral foreign fora concerning their global commerce.”); Brief Amicus Curiae of the Government of the United Kingdom of Great Britain and Northern Ireland in Support of Appellant at 2, III Finance Ltd. v. Aegis Consumer Funding Group (No. 00-7016) (hereinafter U.K. Aegis Brief) (“The United Kingdom Government submits that it would not be in the interest of its trading relationship with the United States for the corporations of the United Kingdom Overseas Territories to be excluded from United States federal courts.”); Diplomatic Note No. 13/2000 from the British Embassy in Washington, D.C. to the United States Department of State, Feb. 2, 2000 at 1 (“The United Kingdom [] views with great concern the potential application of the Matimak rationale to individual Overseas Territories residents, as well as to commercial enterprises.”) . USCA Order 4 5
    466. 1 forced to conclude that Bermuda corporations were stateless. See Matimak, 118 F.3d at 83 2 (commenting that “it is for the Executive Branch, not the courts, to anticipate where potential 3 ‘entanglements’ with such entities are appreciable enough to recognize sovereign status”). The 4 Executive Branch, however, has urged us not to use the definition of “statehood” taken from the 5 context of diplomatic recognition as a basis for denying British Overseas Territories the benefit of 6 federal alienage jurisdiction. The Executive Branch has emphasized that to do so may cause the United 7 States to “face an international controversy with British authorities for failure to provide a neutral forum” 8 for individuals or corporations of a British Overseas Territory in federal court. Brief Amicus Curiae for 9 the United States at 8, Matimak Trading Co. v. Khalily (96-9117). 10 Our Circuit is alone in concluding that federal alienage jurisdiction does not extend to 11 citizens and corporations of British Overseas Territories. The Third Circuit, largely out of deference to 12 the Executive Branch’s position that Hong Kong corporations were considered, at the time, “subject to 13 British sovereignty,” found that they fell within the federal courts’ alienage jurisdiction. Southern Cross 14 Overseas Agencies, Inc. v. Wah Kwong Shipping Group Ltd., 181 F.3d 410, 413 (3rd Cir.1999). 15 The Seventh Circuit has held that a Cayman Islands corporation could be sued in federal court under 16 alienage jurisdiction, explaining that, “[c]ertainly, the exercise of American judicial authority over the 17 citizens of a British Dependent Territory implicates this country’s relationship with the United 18 Kingdom–precisely the raison d’etre for applying alienage jurisdiction.” Wilson v. Humphrys 19 (Cayman) Ltd., 916 F.2d 1239, 1243 (7th Cir.1990), cert. denied, 499 U.S. 947 (1991). The 20 Fourth Circuit, without discussion of the issue, has found that a Bermuda resident–apparently the same 21 individual defendant sued in this case--was a “citizen” or “subject” of a foreign state for alienage 22 jurisdiction purposes. Koehler v. Dodwell, 152 F.3d 304, 308 (4th Cir. 1998). 23 Owing to the fact that our characterization of corporations and citizens of British USCA Order 4 6
    467. 1 Overseas Territories as “stateless” has given rise to precisely the sort of damage to foreign relations the 2 statute was meant to avoid, it is questionable that this distinction has its origins in the statute. Nor is this 3 dubious characterization imposed upon us by the Executive Branch, which has advocated a contrary 4 rule. Neither has this distinction been accepted by our sister circuits. Moreover, this Circuit has 5 previously concluded, albeit without discussion, that “[t]here is no question” that alienage jurisdiction 6 existed between citizens of the United States and a Bermuda corporation. Netherlands Shipmortgage 7 Corp. v. Madias, 717 F.2d 731, 735 (2d Cir. 1983). Finally, two respected senior circuit judges from 8 the panel in this case, Judge Jon O. Newman and Judge Richard J. Cardamone, have expressed 9 disagreement with the merits of our precedent in Matimak. See Koehler v. Bank of Bermuda (New 10 York) Ltd., __ F.3d __, __ n. _ (2d Cir. 2000). All this being the case, it seems incumbent upon us, as 11 a full Court, to reexamine the basis upon which our panels both here and in Matimak reached their 12 conclusions. 13 II. 14 An examination of the merits leads to the conclusion that Matimak misapplied the terms 15 “citizens or subjects of a foreign state” in a fashion inconsistent with both the historical understanding of 16 these terms and a contemporary understanding of the relationship between the United Kingdom and its 17 Overseas Territories. 18 The panel in Matimak began its analysis with the unremarkable proposition that “a 19 foreign state is entitled to define who are its citizens or subjects.” Matimak, 118 F.3d at 85 (citing, 20 inter alia, United States v. Wong Kin Ark, 169 U.S. 649 (1898)). The court then concluded that a 21 British Overseas Territory corporation did not fall within the scope of alienage jurisdiction because 22 British law did not designate the corporation a “citizen” or “subject” of the United Kingdom or indicate USCA Order 4 7
    468. 1 that the corporation was under the control of the United Kingdom. Matimak, 118 F.3d at 85-6.4 2 None would argue with the notion that a foreign state is entitled to define what persons 3 or entities fall into its categories of “citizen “or “subject,” or any other of a variety of legal forms that 4 exist under its own domestic immigration, nationality, and commercial law. The domestic meaning that 5 any particular country may give to the terms “citizen” or “subject” does not, however, bind our courts in 6 determining whether an individual or entity falls within the statutory meaning of such terms as provided 7 by our law of alienage jurisdiction. The wide disparity in meaning that exists among countries 8 concerning such terms requires that our alienage jurisdiction be determined not according to the 9 appearance of the words “citizen” or “subject” (or translation thereof) in the pages of a country’s 10 domestic code, but according to whether United States law deems such persons or entities to be 11 “citizens or subjects” under our Constitution and statutes for the purpose of alienage jurisdiction. To 12 proceed otherwise would be to “allow foreign law to deny privileges afforded under the Constitution . . 4 Aside from the substantial authority cited for the proposition that a foreign state determines its own citizenship and nationality law, and for the relationship between the terms “citizen” and “subject,” the Matimak opinion contains little authority to support its analysis of the “citizenship” or “subjecthood” of corporations in British Overseas Territories. See Matimak, 118 F.3d at 85-86. One unpublished district court opinion is cited to support the suggestion that the corporate law of the Cayman Islands, another British Overseas Territory, is “clearly independent from the United Kingdom’s [law].” See id. at 86 (citing St. Germain v. West Bay Leasing, Ltd., No. 81-CV-3945 (E.D.N.Y. Sept. 30, 1982)). Another district court opinion from 1979 is cited to argue that corporations formed in Hong Kong were not given the benefit of British nationality. See id. (citing Windert Watch Co. v. Remex Elecs. Ltd., 468 F. Supp. 1242 (S.D.N.Y. 1979)). Although the Matimak opinion cites also to a leading treatise, the same treatise currently reaches the opposite conclusion from the panel. See 15 James Wm. Moore, et al., Moore’s Federal Practice § 102.76 (3d ed., 1999) (“A citizen of a British dependent territory is deemed to be a citizen of the United Kingdom and its Overseas Territory. Consequently, federal courts may properly invoke diversity jurisdiction over suits in which a citizen of the Cayman Islands or Bermuda is a party.”). The cases cited by the Matimak court in support of the proposition that a stateless person cannot sue a United States citizen in federal court regard an individual whose citizenship has been revoked by a sovereign and nowhere suggest that a British Overseas Territory’s people or corporations could exist in a condition of perpetual statelessness. See Matimak, 118 F.3d at 86. USCA Order 4 8
    469. 1 . [and perhaps] unintentionally promote discrimination against certain classes of people or entities.” 2 Matimak, 118 F.3d at 89-90 (Altimari, J., dissenting).5 3 As an historical matter, the drafters of the Constitution chose the words “citizens” or 4 “subjects” to refer to the broad category of those under the authority of a foreign power. See Bank of 5 the United States v. Deveaux, 9 U.S. (5 Cranch) 61, 87 (1809) (Marshall, C.J.) (recognizing that the 6 Constitution “established national tribunals for the decision of controversies between aliens and a citizen 7 [of the United States]”), overruled in part on other grounds by Louisville, Cincinnati & Charleston 8 R.R. Co. v. Letson, 43 U.S. (2 How.) 497 (1844). The Judiciary Act of 1789 used the word “alien” 9 apparently as an equivalent term to “citizens” or “subjects” in the first rendering of the statutory grant of 10 authority to exercise federal alienage jurisdiction. Compare U.S. Const. art. III, sec. 2, cl. 1 11 (extending jurisdiction to controversies “between a State, or the Citizens thereof, and foreign States, 12 Citizens or Subjects”) with Judiciary Act of 1789, ch. 20, § 11, 1 Stat. 73, 78 (extending jurisdiction 13 to suits in which “an alien is a party”).6 Oliver Ellsworth, the principal architect of the Judiciary Act of 14 1789 that contained the alienage jurisdiction provision, referred to the need to provide a federal forum 15 for controversies between United States citizens and “foreigners.” See Charles Warren, New Light on 16 the History of the Federal Judiciary Act of 1789, 37 Harv. L. Rev. 49, 60 (1932) (quoting Letter of 17 Oliver Ellsworth to Judge Richard Law, Apr. 30, 1789). “[T]he Framers often referred to [non-U.S.] 5 This task is analogous to that of deciding the state of domicile of a party in a diversity action in federal court. See 28 U.S.C. § 1332(a)(1) (1994) (providing federal jurisdiction for suits between “citizens of different States”). While a court may look to state law definitions of domicile and state citizenship for guidance, “[d]etermination of a litigant’s state of domicile for purposes of diversity is controlled by federal common law, not by the law of any state.” 15 James Wm. Moore, et. al., Moore’s Federal Practice § 102.34[3][a] (3d ed., 1997). 6 The legislative debates concerning the Judiciary Act of 1789 referred to the alienage jurisdiction provision as providing access to the federal courts for “foreigners” or “aliens.” See 1 Annals of Congress (1st Cong.) 810, 814, 825 (Joseph Gales ed., 1834) (House debates). USCA Order 4 9
    470. 1 citizens, subjects and foreigners interchangeably,” and “while foreign modes of government are hardly 2 ‘technicalities’ in any other sense, the Framers apparently did not consider them relevant to the exercise 3 of federal jurisdiction.” Southern Cross Overseas, 181 F.3d at 416 (internal quotation marks and 4 citations omitted).7 5 In 1875, the alienage jurisdiction provision was amended, replacing the term “alien” 6 with the current reference to “citizens” or “subjects.” Act of Mar. 3, 1875, 18 Stat. 470, 470. This 7 change, causing the statute to mirror the language of the Constitution, was motivated by the need to 8 clarify that an alien could not sue another alien in federal court, and not from dissatisfaction with the 9 original statutory term “alien” as impermissibly broader than the terms “citizens” or “subjects” found in 10 the Constitution. See Johnson, 21 Yale J. Int’l L. at 21. 11 Although early cases did not explore the precise boundaries of the terms “citizen” and 12 “subject” as used in alienage jurisdiction, the Supreme Court did have the opportunity to interpret these 13 same terms in other contexts. Their general use confirmed that these terms referred to a range of 14 relationships characterized by the acceptance of the authority and protection of a sovereign and an offer 15 of allegiance. In 1830, Justice Story, addressing the issue of United States citizenship for expatriates 7 At the time the Constitution was written and the first alienage jurisdiction statute was enacted, the term “subject” referred to a person who lived under the control of another. See Samuel Johnson, A Dictionary of the English Language (1755) (defining a “subject” as “[o]ne who lives under the dominion of another”). See also 2 Noah Webster, American Dictionary of the English Language at 84 (1st ed., 1828; facsimile ed. Foundation for American Christian Education 1985) (defining a “subject” as “[o]ne that owes allegiance to a sovereign and is governed by his laws. The natives of Great Britain are subjects of the British government. The natives of the United States, and naturalized foreigners, are subjects of the federal government. Men in free governments are subjects as well as citizens; as citizens, they enjoy rights and franchises; as subjects, they are bound to obey the laws.”) (emphasis in original); 2 James Kent, Commentaries on American Law 258 n.b (6th ed., 1848) (“Subject and citizen are, in a degree, convertible terms as applied to natives; and though the term citizen seems to be appropriate to republican freemen, yet we are equally with the inhabitants of all other countries, subjects, for we are equally bound by allegiance and subjection to the government and law of the land.”) (emphasis in original). USCA Order 4 10
    471. 1 noted that “[t]he rule commonly laid down in the books is, that every person who is born within the 2 ligeance of a sovereign is a subject; and, e converso, that every person born without such allegiance is 3 an alien.” Inglis v. Trustees of the Sailors' Snug Harbour, 28 U.S. (3 Pet.) 99, 155 (1830) (Story, 4 J).8 In construing the terms of the Spanish Treaty of 1795, the Supreme Court in The Pizzaro, 15 5 U.S. (2 Wheat) 227 (1817), rejected the claim that the term “subject” in the treaty applied “only to 6 persons who, by birth or naturalization owe a permanent allegiance to the Spanish government,” holding 7 more simply that, “in the language of the law of nations . . . a person domiciled in a country, and 8 enjoying the protection of its sovereign, is deemed a subject of that country.” Id. 245-46. 9 10 It has long been established that “a corporation created by the laws of a foreign state 11 may, for the purposes of suing and being sued in the courts of the Union, be treated as a ‘citizen’ or 12 ‘subject’ of such a foreign state.” National Steam-Ship Co. v. Tugman, 106 U.S. 118, 121 (1882). 13 The defendant-Bermuda corporations in this suit were created under the laws of two different 14 countries–Bermuda and the United Kingdom–but under the laws of only one recognized “state,” the 15 United Kingdom.9 While the Bermuda’s Companies Act of 1981 provides procedures for 16 incorporating companies in Bermuda, 6 Revised Laws of Bermuda, Title 17, Item 5, Part II (1989 & 8 Justice Story continued, “ [t]wo things usually concur to create citizenship; first, birth locally within the dominions of the sovereign; and secondly, birth within the protection and obedience, or in other words, within the ligenance of the sovereign. That is, the party must be born within a place where the sovereign is at the time in full possession and exercise of his power, and the party must also at his birth derive protection from, and consequently owe obedience or allegiance to the sovereign, as such, de facto.” Inglis, 28 (3 Pet.) U.S. at 155. 9 The characterization of such corporations as “stateless” by the Matimak court is particularly jarring considering that corporations are creations purely of law, and, unlike individuals, exist only through an exercise of sovereignty. See Matimak, 118 F.3d at 89 (Altimari, J., dissenting) (“A stateless corporation is an oxymoron.”). USCA Order 4 11
    472. 1 Update 1996), Bermuda and its government exist “under the sovereignty of the Crown.” 6 Halsbury’s 2 Laws of England, para. 803 (4th ed. reissue,