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.

An Introduction to Historical Explanation

9 views

Published on

An introduction to the literature on historical explanation, for a seminar that I organized over the 18-19 academic year at UCLouvain.

Published in: Education
  • Be the first to comment

  • Be the first to like this

An Introduction to Historical Explanation

  1. 1. An Introduction to Historical Explanation CEFISES Seminar EH, 3.10.2018 Charles H. Pence @pencechp
  2. 2. Outline 1. Historical vs. Experimental Sciences 2. Narrative and Historical Explanations 3. History and Traces 4. The Role of Analogy 5. Historical Contingency The take-home: Historical sciences like paleobiology, archaeology, and astronomy present unique challenges to traditional categories of explanation, argument, and evidence in philosophy of science. Charles H. Pence 2 / 36
  3. 3. Historical and Experimental Sciences Charles H. Pence Historical and Experimental Science 3 / 36
  4. 4. Stamp-Collecting Charles H. Pence Historical and Experimental Science 4 / 36
  5. 5. A Demarcation of Sorts Historical sciences: paleobiology, archaeology, geology, planetary science, astronomy, astrophysics Experimental sciences: everything else, but paradigmatically experimental physics, molecular biology Charles H. Pence Historical and Experimental Science 5 / 36
  6. 6. Cleland on Historical Science Fundamental difference between historical and experimental sciences: time-asymmetry of causation Control of causes ⇒experiment Ignorance of effects Knowledge of effects ⇒historical inference Ignorance of causes For more: Cleland (2002), Philos Sci 69:474 Charles H. Pence Historical and Experimental Science 6 / 36
  7. 7. Inferiority Complex Central to Cleland’s framing: there’s no sense in which historical sciences are worse than experimental, they simply operate under different epistemic constraints Charles H. Pence Historical and Experimental Science 7 / 36
  8. 8. Demarcation? Historical/experimental need not be a real “demarcation” – fine if it’s only varieties of epistemic tools fit to purpose Charles H. Pence Historical and Experimental Science 8 / 36
  9. 9. Narrative and Historical Explanation Charles H. Pence Narrative and Historical Explanation 9 / 36
  10. 10. Just-So Stories Charles H. Pence Narrative and Historical Explanation 10 / 36
  11. 11. Narratives How do we understand the role of narratives in historical explanations? A major worry: fitting token historical events into narratives is not predictive in the same sense as fitting them into natural laws or similar generalizations Charles H. Pence Narrative and Historical Explanation 11 / 36
  12. 12. Narrative and Mechanism Swaim: A narrative gives us a possibility space and a causal mechanism. A space of possibilities serves as our license proving that we have explored sufficient relevant alternatives; then a proposed causal mechanism allows us to discriminate between those alternatives Charles H. Pence Narrative and Historical Explanation 12 / 36
  13. 13. Philosophy of History Notably, this puts this literature into contact with work throughout the philosophy of history – how do historical narratives explain, in general? Charles H. Pence Narrative and Historical Explanation 13 / 36
  14. 14. History and Traces Charles H. Pence History and Traces 14 / 36
  15. 15. A “Smoking Gun” Back to Cleland: Historical scientists are trying to use evidential traces to discriminate between many competing, possible common causes Charles H. Pence History and Traces 15 / 36
  16. 16. A “Smoking Gun” Charles H. Pence History and Traces 16 / 36
  17. 17. Causal Overdetermination Events usually bear an asymmetric relation to their traces: one event produces many, many traces, each of which individually could be enough to infer its existence Charles H. Pence History and Traces 17 / 36
  18. 18. Causal Overdetermination Events usually bear an asymmetric relation to their traces: one event produces many, many traces, each of which individually could be enough to infer its existence Charles H. Pence History and Traces 17 / 36
  19. 19. Causal Overdetermination Historical science: attempt to deal with this fact by hunting for comparative evidence for the common causes of traces Experimental science: attempt to deal with the inverse problem, by carefully modifying experimental controls to exclude other possible confounding causes Charles H. Pence History and Traces 18 / 36
  20. 20. But Wait! Turner: These traces are very often destroyed, sometimes irrevocably – placing researchers not into a situation of overdetermination, but one of (perhaps irretrievable) local underdetermination! In fact, our background theories tell us just how these processes destroy historical information. Charles H. Pence History and Traces 19 / 36
  21. 21. But Wait! Charles H. Pence History and Traces 20 / 36
  22. 22. Optimism Currie: This is true in some, particular circumstances, but not enough to ground a global skepticism about method in historical science Watch this space: Adrian may well visit this spring! Charles H. Pence History and Traces 21 / 36
  23. 23. The Role of Analogy Charles H. Pence The Role of Analogy 22 / 36
  24. 24. A Brief Digression Recall the classic, deductive-nomological model of scientific explanation: 1. Scientific explanations are (sound) deductive arguments, which 2. Make (essential, non-eliminable) use of laws of nature, and 3. Invoke induction only to establish their boundary conditions or to demonstrate (lesser, inferior) statistical claims Charles H. Pence The Role of Analogy 23 / 36
  25. 25. Another Digression: IBE Cleland’s search for a “smoking gun” clearly can’t rely on anything like this kind of explanatory form. So what do we do instead? Cleland’s choice: opt for inference to the best explanation. But! Problems abound: cf. van Fraassen’s critique Charles H. Pence The Role of Analogy 24 / 36
  26. 26. Argument from Analogy Charles H. Pence The Role of Analogy 25 / 36
  27. 27. Argument from Analogy Charles H. Pence The Role of Analogy 26 / 36
  28. 28. What are they? What are we doing when we offer an argument by analogy? Trying to use the analogy to build models of shared causal processes, with the aid of empirical examples of the instantiation of those processes (Currie). Charles H. Pence The Role of Analogy 27 / 36
  29. 29. What are they? Here, then, we see the great importance of possessing a large stock of analogous instances or phenomena which class themselves with that under consideration, the explanation of one among which may naturally be expected to lead to that of all the rest. (Herschel 1830, §138) Charles H. Pence The Role of Analogy 28 / 36
  30. 30. But wait! What if the past is too historically contingent to support analogous reasoning of this sort? Charles H. Pence The Role of Analogy 29 / 36
  31. 31. Historical Contingency Charles H. Pence Historical Contingency 30 / 36
  32. 32. Path Dependence Desjardins (2011), Philos Sci 78:731 Charles H. Pence Historical Contingency 31 / 36
  33. 33. Other Options Complete convergence: If all paths eventually converge to the same end-state, then history is erased Complete chanciness: If all paths scatter populations to a massive variety of outcomes in evolutionary space, then history is erased Charles H. Pence Historical Contingency 32 / 36
  34. 34. Path Dependence To summarize, both convergence and chance can erase history. When different populations adapt similarly to a given environment, history is erased because past differences in the value of a state variable ceases to exist. Chance on the other hand can make the derived populations more scattered and thus create a situation where it is impossible to see a relationship between changes in the initial states and the probability of different evolutionary outcomes. When this happens, having different evolutionary histories will not affect distinctively the probability of reaching one or another (set of) outcome(s). (Desjardins 2011, 347) Charles H. Pence Historical Contingency 33 / 36
  35. 35. Path Dependence The upshot: Whether or not history is preserved by an evolutionary process is an exceptionally subtle matter! Charles H. Pence Historical Contingency 34 / 36
  36. 36. Contingency and Analogy This will make the applicability of analogical reasoning extremely context-dependent. Some worries: 1. What if we can’t know whether an analogy will hold in a given case? How can we evaluate their quality? Charles H. Pence Historical Contingency 35 / 36
  37. 37. Contingency and Analogy This will make the applicability of analogical reasoning extremely context-dependent. Some worries: 1. What if we can’t know whether an analogy will hold in a given case? How can we evaluate their quality? 2. Can we detect a signal of these varying impacts of contingency in examples drawn from scientific practice? Charles H. Pence Historical Contingency 35 / 36
  38. 38. Contingency and Analogy This will make the applicability of analogical reasoning extremely context-dependent. Some worries: 1. What if we can’t know whether an analogy will hold in a given case? How can we evaluate their quality? 2. Can we detect a signal of these varying impacts of contingency in examples drawn from scientific practice? 3. Surely the source and target of the analogy will also change its features; can we taxonomize these kinds of inferences? Charles H. Pence Historical Contingency 35 / 36
  39. 39. Questions? charles@charlespence.net https://charlespence.net @pencechp

×