Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

The Pronunciation of Iraq


Published on

Published in: News & Politics, Business
  • Be the first to comment

The Pronunciation of Iraq

  1. 1. Variation in the I r a q Vowels: Conservatives vs. Liberals Lauren Hall-Lew, Elizabeth Coppock, & Rebecca Starr Stanford University { dialect, rlstarr, coppock } @ stanford . edu NWAV36 , University of Pennsylvania, October 13, 2007
  2. 2. How do YOU say it?
  3. 3. Observation <ul><li>Iraq varies according to the second vowel: </li></ul><ul><ul><li>r/æ/k vs. r/ah/k [1], [2] </li></ul></ul><ul><li>Iraq also varies according to the first vowel: </li></ul><ul><ul><li>/ay/ vs. /Ih/ vs. /iy/ </li></ul></ul>[1] Shapiro 1997 [2] Boberg 1997; 1999
  4. 4. Hypotheses <ul><li>Second Vowel: Ir a q </li></ul><ul><ul><li>/æ/ variant indexes political conservativism. </li></ul></ul><ul><ul><li>/ah/ variant indexes political liberalism. </li></ul></ul><ul><li>First Vowel: I raq </li></ul><ul><ul><li>/ay/ variant indexes political conservatism. </li></ul></ul><ul><ul><li>/iy/ and /Ih/ index political liberalism. </li></ul></ul>
  5. 5. Motivation <ul><li>Identity & Political Psychology intersect: </li></ul><ul><ul><li>“ ... intergroup conflict, conformity to group norms, the effects of low group status ...” [3] </li></ul></ul><ul><li>Recent Work: </li></ul><ul><ul><li>American identity, Patriotism, & Ethnicity [4] </li></ul></ul><ul><li>Identity & Phonological Variables </li></ul><ul><ul><li>potential resource for the expression of a political identity </li></ul></ul>[3] Tajfel & Turner 1979; Huddy 2001 [4] Citrin et al, 1990 & 2000, cited in Huddy 2001
  6. 6. Motivation <ul><li>“ Foreign (a)” variable, as in plaza , is often realized as /ah/ in US English due to attitudinal factors rather than phonological factors. [2] </li></ul><ul><li>US English speakers evaluate /ah/ to be “more correct, educated, and sophisticated than /æ/ as a nativization of foreign (a).” [2] </li></ul><ul><li>“ Respect engenders imitation [retention of foreign sounds]; disrespect integration [full nativization]” [5] </li></ul>[5] Weinreich 1968, cited & comments added in Boberg 1999
  7. 7. Pilot Study: Methods <ul><li>Two networks with different political leanings: FoxNews and KQED (Bay Area NPR station) television and radio. </li></ul><ul><li>Collected all instances of Iraq occurring in top news stories at the time, as well as in all news stories within a 24-hour period. </li></ul><ul><li>Classified speakers as publicly conservative, publicly liberal, or unknown. </li></ul>
  8. 8. Pilot Study: Data <ul><li>686 tokens of Iraq </li></ul><ul><li>248 speakers </li></ul><ul><ul><li>44 publicly conservative </li></ul></ul><ul><ul><li>32 publicly liberal </li></ul></ul><ul><ul><li>172 unknown </li></ul></ul><ul><li>Distribution of conservatives and liberals not significantly different on KQED or FoxNews </li></ul><ul><li>No significant differences in Iraq pronunciation between networks </li></ul><ul><li>Network data was combined </li></ul>
  9. 9. Pilot Study: Results <ul><li>Significant finding: </li></ul><ul><li>83% of conservatives, but only 31% of liberals, had at least one occurrence of the /æ/ variant. ( F =17.7, df=1, p < 0.0001) </li></ul>
  10. 10. Pilot Study: Other Results <ul><li>Military personnel: </li></ul><ul><ul><li>All 15 military personnel used /æ/ </li></ul></ul><ul><ul><li>14 of 15 used /ay/ for the first vowel, while all non-military speakers used /Ih/ </li></ul></ul><ul><ul><li>The 1 military member who used /Ih/ was explicitly an anti-war Democrat, General Wesley Clark. </li></ul></ul>
  11. 11. Pilot Study: Other Results <ul><li>Virtually all news readers and news anchors use /ah/, confirming the prediction by Boberg (1999) of /ah/ in more formal styles </li></ul>
  12. 12. Pilot Study: Interpretation <ul><li>Indications that both vowels in Iraq may index political orientation. </li></ul>
  13. 13. Limitations of Pilot Study <ul><li>Political leanings of speakers not always clear. </li></ul><ul><li>Sampling was not highly systematic. </li></ul><ul><li>Not controlled for: </li></ul><ul><ul><li>Register </li></ul></ul><ul><ul><li>Audience </li></ul></ul><ul><ul><li>Regional dialect variation </li></ul></ul>
  14. 14. Main Study: Speakers <ul><li>Members of the US House of Representatives: </li></ul><ul><ul><li>Representative of all dialect regions of the US </li></ul></ul><ul><ul><li>Clear political affiliations </li></ul></ul>
  15. 15. Data: Iraq Troop Surge Debate <ul><li>Three days of debate in February 2007 </li></ul><ul><li>Debate topic: Resolution 63, stating that the House disapproves of troop surge </li></ul><ul><li>All speeches aired on C-SPAN, posted online at </li></ul><ul><li>Speeches given by 304 out of 435 total members of the House </li></ul>
  16. 16. Surge Debate Study: Methods <ul><li>Total number of speakers included: 259 </li></ul><ul><ul><li>Only included speakers who said Iraq 3 times or more. </li></ul></ul><ul><li>Total number of tokens coded: N=1959 </li></ul><ul><li>Mean = 8 tokens per speaker </li></ul><ul><li>Median = 6 tokens per speaker </li></ul>
  17. 17. Speaker Factors Coded: Political Stance measures <ul><li>Party affiliation </li></ul><ul><li>Vote on anti-surge resolution (to capture war stance) </li></ul><ul><li>Economic and Social Liberalism rating: </li></ul><ul><ul><li>percentage ratings from </li></ul></ul><ul><ul><li>based on voting record </li></ul></ul><ul><ul><li>Conservatives have a higher economic % </li></ul></ul><ul><ul><li>Liberals have a higher social % </li></ul></ul>
  18. 18. Speaker Factors Coded: Speech variety measures <ul><li>[+/-] Southern Accent </li></ul><ul><ul><li>/ay/ monophthongization </li></ul></ul><ul><ul><li>perceptual measures </li></ul></ul><ul><li>Regional accent / Speech variety </li></ul><ul><ul><li>subjective, with some inter-rater judgments </li></ul></ul><ul><li>State they represent </li></ul><ul><ul><li>Region of the US, based on US Census </li></ul></ul><ul><ul><li>Red/Blue status, based on 2004 election </li></ul></ul>
  19. 19. Speaker Factors Coded: Other demographic info <ul><li>Ethnicity </li></ul><ul><ul><li>Particular attention to AAE speakers </li></ul></ul><ul><li>Age </li></ul><ul><ul><li>Here, 4 age categories, approx same N per cat. </li></ul></ul><ul><li>Sex Class </li></ul><ul><ul><li>Subjective classification </li></ul></ul>
  20. 20. Surge Debate Study: Methods <ul><li>Logistic regression analysis based on binary dependent variable: </li></ul><ul><ul><li>over 50% use of /æ/ in Iraq for any given speaker </li></ul></ul><ul><li>Only tokens of Iraq included </li></ul><ul><ul><li>Iraqi and Iran also coded for, but not grouped with Iraq for analysis. </li></ul></ul>
  21. 21. %/æ/ use by number of speakers <ul><li>85% of speakers consistently pronounce Iraq always with one vowel or the other. </li></ul><ul><li>15% of the speakers varied in their pronunciation of Iraq. </li></ul><ul><ul><li>This does not include Iraqi , which was more likely to be pronounced differently </li></ul></ul><ul><li>In contrast to news readers from pilot, Representatives favor /æ/ over /ah/ </li></ul>
  22. 22. %/æ/ use by number of speakers
  23. 23. Results by Party <ul><li>Political party is a highly significant predictor of the Iraq pronunciation </li></ul><ul><ul><li>correlation=7.29, p=0.025 </li></ul></ul><ul><li>Republicans are statistically more likely to use the /æ/ variant </li></ul><ul><li>Political party significantly predicts Iraq pronunciation even when controlling for region, regional accent, age group, sex class, and ethnicity. </li></ul>
  24. 24. Vowel Use by Political Party
  25. 25. Economic & Social Liberalism <ul><li> </li></ul><ul><li>Economic & Social Liberalism scores based on voting record. </li></ul><ul><li>High Economic Lib = more conservative </li></ul><ul><li>High Social Lib </li></ul><ul><li>= more liberal </li></ul>
  26. 26. Economic & Social Liberalism <ul><li>Economic Liberalism score: </li></ul><ul><ul><li>Speakers with high Econ % (conservatives) are significantly more likely to say /æ/ in Iraq, when controlling for all factors except party. </li></ul></ul><ul><ul><li>consistent with our political party findings </li></ul></ul><ul><li>No significant effect of Social Liberalism score </li></ul>
  27. 27. Results by Age <ul><li>Age grouped into 4 categories, each including approximately 68 speakers: </li></ul><ul><ul><li>≤ 1944, 1945-1949, 1950-1957, ≥ 1958 </li></ul></ul><ul><li>No significant effect of age, although older people appear to use less /æ/. [1] </li></ul><ul><li>Political Party within each age category: consistent pattern, not significant. </li></ul>[1] cf. Shapiro 1997, that /æ/ is the earlier form
  28. 29. Results by Region <ul><li>Test according to: </li></ul><ul><ul><li>region of representation </li></ul></ul><ul><ul><li>the presence of monophthongized /ay/ in closed syllables </li></ul></ul><ul><li>Potential interaction between region, surge stance, and political party: </li></ul><ul><ul><li>More pro-surge conservatives in Southern states </li></ul></ul>
  29. 30. Results by Region <ul><li>4 Regions: </li></ul><ul><ul><li>South </li></ul></ul><ul><ul><li>Midwest </li></ul></ul><ul><ul><li>Northeast </li></ul></ul><ul><ul><li>West </li></ul></ul><ul><li>Defined by the U.S. Census Bureau </li></ul>
  30. 31. Results by Region <ul><li>No significant effect of Region: </li></ul><ul><ul><li>No effect of southern accent, broadly coded </li></ul></ul><ul><ul><ul><li>any presence of monophthong /ay/ </li></ul></ul></ul><ul><ul><li>No effect of southern accent, narrowly coded </li></ul></ul><ul><ul><ul><li>all /ay/ are monophthongized </li></ul></ul></ul><ul><li>Standard vs. Regional accents: not significant </li></ul>
  31. 33. Variation by Speaker Sex <ul><li>40 of 271 (15%) Congresspeople in this sample are Congresswomen. </li></ul><ul><li>Women tend to use /ah/ more than /æ/, but not significantly more than men. (p < .08 as a main effect) </li></ul><ul><li>Party remains significant when controlling for speaker sex class. </li></ul>
  32. 35. What About The First Vowel? <ul><li>Variants we heard for the first vowel in Iraq : </li></ul><ul><ul><li>/Ih/ </li></ul></ul><ul><ul><li>/iy/ </li></ul></ul><ul><ul><li>/ay/ </li></ul></ul><ul><li>Grouped /Ih/ and /iy/ together as (i) in final analysis because they were difficult to distinguish perceptually, and /iy/ was rare. </li></ul>
  33. 36. The First Vowel in ‘Iraq’
  34. 37. The First Vowel in Iraq <ul><li>Same bimodal distribution as in the histogram for % /æ/: </li></ul><ul><ul><ul><li>46 Congresspeople have >50% /ay/ </li></ul></ul></ul><ul><ul><ul><li>141 Congresspeople have 50% or less /ay/ </li></ul></ul></ul><ul><li>18% of the speakers vary between /ay/ and (i) </li></ul><ul><li>Within all second-vowel /æ/ tokens, 24% use first-vowel /ay/ </li></ul><ul><li>2 speakers say Iraq as /ay/r/ah/q categorically. </li></ul>
  35. 38. The First Vowel in Iraq <ul><li>There is no statistically significant predictive variable for the first vowel. </li></ul><ul><li>Non-significant trends: </li></ul><ul><ul><li>Within people who have Southern accents, /ay/ correlates with anti-surge </li></ul></ul><ul><ul><li>Omitting AAE speakers overall, within non - Southern accents, /ay/ correlates with pro-surge </li></ul></ul>
  36. 39. SUMMARY <ul><li>In the US House of Representatives, variation of the second vowel in Iraq correlates with the Political Party of the speaker. </li></ul><ul><ul><li>/ah/ correlates with Democrats </li></ul></ul><ul><ul><li>/æ/ correlates with Republicans </li></ul></ul><ul><li>No factors significantly predict the pronunciation of the first vowel. </li></ul>
  37. 40. Indexing Political Identity <ul><li>Implications for variation </li></ul><ul><li>Potential complications </li></ul><ul><ul><li>Complex: How to operationalize? </li></ul></ul><ul><li>Political psychology [6] </li></ul><ul><ul><li>Political views may be a critical aspect of an individual’s identity </li></ul></ul><ul><ul><li>Correlated with Political Identity: </li></ul></ul><ul><ul><ul><li>Age (old vs. young) </li></ul></ul></ul><ul><ul><ul><li>Class Mobility & Social Networks </li></ul></ul></ul><ul><ul><ul><li>Personality Type & Value System </li></ul></ul></ul><ul><ul><ul><li>Religion </li></ul></ul></ul>[6] See, e.g.: Brewer 2001; Conover and Feldman 1981; De Graaf et al. 1995; Huddy 2001; Huddy and Khatib 2007; Mackenzie 1978; Kymlicka 2001, cited in Charney 2003
  38. 41. Operationalizing Political Identity <ul><li>3 measures here: </li></ul><ul><ul><li>Political Party </li></ul></ul><ul><ul><li>Troop Surge Stance </li></ul></ul><ul><ul><li>Econ/Social Liberalism </li></ul></ul><ul><li>All 3 correlated highly </li></ul><ul><li>All 3 also differed in terms of predictive power. </li></ul>
  39. 42. Operationalizing Political Identity <ul><li>Anti-Surge Republicans (N=10) </li></ul><ul><li>Pro-Surge Democrats (N=1) </li></ul><ul><li>8 of the Republicans use /æ/ & the Democrat uses /ah/. </li></ul><ul><li>2 Republicans have less than 50% /æ/, so we can take a closer look at them. </li></ul>
  40. 43. Future Directions <ul><li>People whose voting record diverges from their political party’s </li></ul><ul><li>Synchronic patterns of variation </li></ul><ul><ul><li>within individuals </li></ul></ul><ul><ul><li>speech communities </li></ul></ul><ul><li>What's going on with /ay/r/ah/k? </li></ul><ul><li>Will Iraq data from Congress change as the war continues? </li></ul>
  41. 44. Future Directions <ul><li>Other Terms: </li></ul><ul><ul><li>Iraqi: more variable within speaker, what’s the pattern? </li></ul></ul><ul><ul><li>Iran & Iranian: do their vowels pattern like Iraq ’s? </li></ul></ul><ul><ul><li>Vietnam: consider data from the 60s/70s vs. now </li></ul></ul><ul><li>Work with non-politicians: </li></ul><ul><ul><li>Attitudinal Surveys ( cf. Boberg 1999) </li></ul></ul><ul><ul><li>Production Experiment </li></ul></ul><ul><ul><li>Perception Experiment </li></ul></ul>
  42. 45. Thank You! <ul><li>We would like to thank: </li></ul><ul><ul><li>Ben Munson </li></ul></ul><ul><ul><li>Laura Staum Casasanto </li></ul></ul><ul><ul><li> </li></ul></ul><ul><ul><li>“ Eric,” Canadian software architect, age 28, a.k.a. the guy who assembled the videos on: </li></ul></ul>
  43. 46. References <ul><li>Boberg, C. (1997). Variation and change in the Nativization of Foreign (a) in English. PhD thesis, University of Pennsylvania. </li></ul><ul><li>Boberg, C. (1999). The Attitudinal Component of variation in American English foreign (a) Nativization. Journal of Language and Social Psychology , 18:49–61. </li></ul><ul><li>Brewer, M. B. (2001). The many faces of social identity: Implications for political psychology. Political Psychology , 22:115–25. </li></ul><ul><li>Charney, E. (2003). Identity and Liberal Nationalism. American Political Science Review. 97(2): 295-310. </li></ul><ul><li>Conover, P. J. and Feldman, S. (1981). The origins and meaning of liberal/ conservative self-identification. American Journal of Political Science , 25:617–45. 2 </li></ul><ul><li>De Graaf, N. D., Nieuwbeerta, P., and Heath, A. (1995). Class mobility and political preferences: Individual and contextual effects. The American Journal of Sociology , 100:997–1027. </li></ul><ul><li>Huddy, L. (2001). From social to political identity: A critical examination of social identity theory. Political Psychology , 22:127–56. </li></ul><ul><li>Huddy, L. and Khatib, N. (2007). American patriotism, national identity, and political involvement. American Journal of Political Science , 51:63–77. </li></ul><ul><li>Kymlicka, W. (2001). Politics in the Vernacular: Nationalism, Multiculturalism, and Citizenship. Oxford: Oxford University Press. </li></ul><ul><li>Mackenzie, W. J. M. (1978). Political Identity . New York: St. Martin’s Press. The Pew Research Center (2004). Overview: News audiences increasingly politicized. </li></ul><ul><li>Rhodebeck, L. A. (1993). The Politics of Greed? Political Preferences among the Elderly. The Journal of Politics, 55(2):342-364. </li></ul><ul><li>Shapiro, M. (1997). Broad and Flat A in Marked Words. American Speech. 72(4): 437-439. </li></ul>
  44. 47. Variation by Ethnicity <ul><li>23 of 271 (8.5%) Congresspeople in this sample were identified as speakers of AAE. </li></ul><ul><li>Given the low N, no significant effect was found for the /ah/ vs. /æ/ variable between AAE and non-AAE speakers. </li></ul><ul><li>Impressionistically, AAE speakers tended to favor /ay/ for the first vowel more than non-AAE speakers, who appeared to favor /iy/ or /ih/. </li></ul><ul><li>However, no significant effect was found. </li></ul>
  45. 48. What about Shrubby? <ul><li>George W. Bush is a little bit variable </li></ul><ul><ul><li>Based on data from Pilot Study ( NPR and FoxNewsRadio ): speeches, interviews, etc. </li></ul></ul><ul><ul><li>From N=20 tokens, 2 = /ah/ & 18 = /æ/ </li></ul></ul><ul><ul><li>Possible gradual switch from /æ/ to /ah/? </li></ul></ul><ul><ul><li>Call for longitudinal analysis!! </li></ul></ul>
  46. 49. No Interactions <ul><li>Are there interactions between any of the variables? </li></ul><ul><li>Checking for interactions: </li></ul><ul><ul><li>Geography is correlated with Party, </li></ul></ul><ul><ul><li>Party is correlated with Vowel use, </li></ul></ul><ul><ul><li>But Geography is not statistically correlated with Vowel use. </li></ul></ul>
  47. 50. No Interactions <ul><li>When controlling for all other factors, we still get a highly significant effect for political party. </li></ul><ul><li>No interaction between factors is significant for predicting vowel use. </li></ul><ul><li>In an additive statistical model there is only an effect from political party. </li></ul>