Historical Science M ethodology and D ifferences from E xperim - PowerPoint PPT Presentation

Historical Science M ethodology and D ifferences from E xperim ental S cience C arol E . C leland C arol E . C leland Philosophy D epartm Philosophy D epartm ent ent C enter for A strobiology C enter for A strobiology U niversity of C olorado (B oulder) U niversity of C olorado (B oulder)

O VE RVIE W O VE RVIE W  D ifferences in the m  D ifferences in the m ethodology of classical ethodology of classical experim experim ental science and prototypical historical science: ental science and prototypical historical science: two different patterns of evidential reasoning. two different patterns of evidential reasoning.  T he role of com  T he role of com m m on cause explanation in the evaluation on cause explanation in the evaluation of historical hypotheses. of historical hypotheses.  T he Principle of the C om  T he Principle of the C om m m on C ause and the on C ause and the asym asym m m etry of overdeterm etry of overdeterm ination. ination.  T he priority of com  T he priority of com m m on cause over separate causes on cause over separate causes explanation in historical science. explanation in historical science.

T he structure of T he structure of C lassical E xperim C lassical E xperim ental S cience ental S cience  Focus: Is on a single (som  Focus: Is on a single (som etim etim es com es com plex) hypothesis plex) hypothesis which typically has the form which typically has the form of a universal generalization of a universal generalization (A ll C ’s are E ’s). (A ll C ’s are E ’s).  C entral Research A ctivity: C onsists in repeatedly  C entral Research A ctivity: C onsists in repeatedly bringing about the test conditions specified by the bringing about the test conditions specified by the hypothesis and controlling for extraneous conditions hypothesis and controlling for extraneous conditions that m that m ight be responsible for false positives and false ight be responsible for false positives and false negatives. negatives.

T he E xperim T he E xperim ental Program ental Program vs. vs. S olitary E xperim S olitary E xperim ent ent  Failed predictions: do not result in the rejection of  Failed predictions: do not result in the rejection of hypotheses; they are best interpreted as attem hypotheses; they are best interpreted as attem pts to pts to protect the hypothesis from protect the hypothesis from false negatives. false negatives.  S uccessful predictions: A re not followed by risky tests  S uccessful predictions: A re not followed by risky tests (in Popper’s sense); they are best interpreted as (in Popper’s sense); they are best interpreted as attem attem pts to protect the hypothesis from pts to protect the hypothesis from false positives. false positives.  A cceptance/ rejection of a hypothesis: occurs only  A cceptance/ rejection of a hypothesis: occurs only after a hypothesis is subjected to a series of after a hypothesis is subjected to a series of experim experim ents controlling for plausible auxiliary ents controlling for plausible auxiliary assum assum ptions that could explain predictive successes and ptions that could explain predictive successes and predictive failures. predictive failures.

T he structure of T he structure of Prototypical H istorical S cience Prototypical H istorical S cience Focus:Is on proliferating m Focus:Is on proliferating m ultiple, rival hypotheses ultiple, rival hypotheses to explain a puzzling body of traces to explain a puzzling body of traces encountered in field world. encountered in field world. C entral Research A ctivity: C onsists in searching C entral Research A ctivity: C onsists in searching for a ‘sm for a ‘sm oking gun’ a trace(s) that sets apart oking gun’ a trace(s) that sets apart one or m one or m ore hypotheses as providing a better ore hypotheses as providing a better explanation for the observed traces than the explanation for the observed traces than the others. others.

A C ase S tudy A C ase S tudy T he A lvarez H ypothesis T he A lvarez H ypothesis   T wo pronged hypotheses: im T wo pronged hypotheses: im pact, extinction pact, extinction   Initially m Initially m any different explanations for the end- any different explanations for the end- C retaceous m C retaceous m ass extinction: pandem ass extinction: pandem ic, evolutionary ic, evolutionary senescence, clim senescence, clim ate change, supernova, volcanism ate change, supernova, volcanism , , and m and m eteorite Im eteorite Im pact. pact.   D iscovery of an iridium D iscovery of an iridium anom anom aly (“sm aly (“sm oking gun”) in oking gun”) in K -T boundary sedim K -T boundary sedim ents narrowed it down to two ents narrowed it down to two possibilities: volcanism possibilities: volcanism and m and m eteorite im eteorite im pact. pact. D iscovery of extensive quantities of a rare form D iscovery of extensive quantities of a rare form of of shocked m shocked m ineral subsequently cinched the case for ineral subsequently cinched the case for im im pact over volcanism pact over volcanism . .

T he E valuation of H istorical T he E valuation of H istorical H ypotheses H ypotheses  N ot grounded in prediction:  N ot grounded in prediction:  H istorical predictions are not ‘risky’ in Popper’s  H istorical predictions are not ‘risky’ in Popper’s sense; too m sense; too m any highly plausible extraneous any highly plausible extraneous conditions (e.g., iridium conditions (e.g., iridium poor m poor m eteorite, geological eteorite, geological processes, unrepresentative sam processes, unrepresentative sam ples) capable of ples) capable of defeating them defeating them . .  Predictions are typically vague, e.g., Ward’s  Predictions are typically vague, e.g., Ward’s ‘prediction’ about C retaceous am ‘prediction’ about C retaceous am m m onites; they onites; they serve m serve m ore as guides for looking for a sm ore as guides for looking for a sm oking gun oking gun than predictions. than predictions.

T he E valuation of H istorical T he E valuation of H istorical H ypotheses (cont.) H ypotheses (cont.)  A hypothesis m  A hypothesis m ay be rejected on the basis of ay be rejected on the basis of evidence that does not refute it, e.g., the evidence that does not refute it, e.g., the contagion hypothesis for the end-C retaceous contagion hypothesis for the end-C retaceous extinctions. extinctions.  T he acceptance of a hypothesis does not  T he acceptance of a hypothesis does not require a successful prediction, e.g., the iridium require a successful prediction, e.g., the iridium anom anom aly was not and could not have been aly was not and could not have been predicted or retrodicted. predicted or retrodicted.

T he E valuation of H istorical T he E valuation of H istorical H ypotheses(cont.) H ypotheses(cont.)  G rounded in explanatory power:  G rounded in explanatory power:  H ypotheses are accepted and rejected in virtue of  H ypotheses are accepted and rejected in virtue of their power to explain (vs.predict) puzzling bodies of their power to explain (vs.predict) puzzling bodies of traces discovered through field work. traces discovered through field work.  T he A lvarez hypothesis explains an otherwise  T he A lvarez hypothesis explains an otherwise puzzling association (correlation) am puzzling association (correlation) am ong traces better ong traces better than any of its rivals. It is for this reason that it is than any of its rivals. It is for this reason that it is viewed as ‘confirm viewed as ‘confirm ed’ and its rivals are no longer ed’ and its rivals are no longer seriously entertained by scientists. seriously entertained by scientists.

C om C om m m on C ause explanation on C ause explanation   C om C om m m on cause explanations and narrative explanations, e.g., on cause explanations and narrative explanations, e.g., sexing a T . rex. sexing a T . rex.   Reichenbach’s epistem Reichenbach’s epistem ic Principle of the C om ic Principle of the C om m m on C ause: on C ause: seem seem ingly im ingly im probable associations (correlations or probable associations (correlations or sim sim ilarities) am ilarities) am ong traces are best explained by reference to a ong traces are best explained by reference to a com com m m on cause. on cause.   Presupposes an ostensibly m Presupposes an ostensibly m etaphysical claim etaphysical claim about the about the tem tem poral structure of causal relations in our universe: m poral structure of causal relations in our universe: m ost ost (not all) events form (not all) events form causal forks opening from causal forks opening from past to future past to future (leave m (leave m any traces in the future). any traces in the future).