Scientific Revolutions (Catania, 3 June 2011)

Transcription

1 ScientificRevolutions (Catania,3June2011) TomNickles DepartmentofPhilosophy UniversityofNevada,Reno,USA Thesenotesroughlyfollowmyonlinearticle, ScientificRevolutions, inthestanfordencyclopediaofphilosophy. Consultthatarticleforreferences.ThebestdetaileddiscussionofKuhn sviewsisstillpaulhoyningen Huene s (1993). 1.Introduction:Interest&ImportanceoftheTopic 2.Whatisarevolution? 3.Whatisascientificrevolution? 4.ThomasKuhn smodelofscientificrevolutions 5.SomeGeneralCriticismsofKuhn 6.Incommensurability:Kuhn&hisCritics 7.HowdoesKuhnexlainscientificrevolutions? 8.SomePhilosophicalImplications a.scientificprogress b.realism&truth c.socialconstructionism 9.SomeNon KuhnianConceptionsofRevolutionandRupture 10.ComparisonwithEconomicInnovation 11.RevolutionorEvolution? 12.Kuhn sattempttounitehistoricismandnaturalism. 1.Introduction:Interest&ImportanceoftheTopic Revolutionsarebig,suddenchangesthataredramatic&fascinating. Theysignaldeephistoricalchange.Theymakehistoriographyinteresting,dynamic. InpoliticsweprobablythinkfirstoftheFrenchRevolution,thentheRussianBolshevikRevolution, theenglishrevolutionofthe1640s&thegloriousrevolutionof1688,theamerican Revolutionof ,thepossiblerevolutionsinprogresstodayaroundtheMediterranean. Insciencewethinkoftherelativityandquantummechanicsrevolutionsinphysics,therevolution inchemistryledbylavoisier,darwininbiology,andthescientificrevolutionfrom CopernicustoNewton,whichsupposedlyseparatedmodernityfromfeudalismandtheMiddle Ages. Mostrecently:theplatetectonicrevolutioningeology(1960s). Revolutionsarewatershedeventsthatgiveafuturereferencepoint: before&after therevolution. Somequestions: Whatisarevolution,moreexactly? Howcanweexplainwhy&howarevolutionoccurred? Arerevolutionsnecessary?Historicallycontingent?Bothnecessaryandcontingent? Dorevolutionsallhaveapatternoradynamicalmechanismincommon? Doesatime seriesofrevolutionsrevealanunderlyingpatternofhistoricaldevelopment? Whatdoestheexistenceofrevolutionstellusaboutthenatureofpoliticsorscience? Inparticular,whatimplicationsdoestheexistenceofscientificrevolutionshavefor... scientificrealism:areweapproachingthetruthaboutobjectivereality?

2 2 scientificprogress? theoverallpatternofscientificdevelopment? thenatureofscientificpractice? Havetherereallybeenanygenuinescientificrevolutions? WhatisascientificrevolutionaccordingtoThomasKuhninTheStructureofScientific Revolutions? HavetherebeenanyspecificallyKuhnianrevolutions? 2.Whatisarevolution? Historicalpath:Copernicus,DeRevolutionibus politicalrev thenbackintoscienceitself. Etymology: revolve areturntoapreviousstate:thewheeloffortune,copernicus. Modernmeaning:justtheopposite acompleterejectionofthepast;anewdirection. Revolution mererevolt(achievementvs.attempt)orrebellionorcoupd étatorreformation. Revolutionisdeeperthanmerechangeofgovernmentleadershipwithinthesamesystem. Soasuccessfulrevoltagainstcurrentleaders(Locke,socialcontract)isnotnecessarilya revolution! Reformers(e.g.,Luther)aimtoreturntooriginalroots;revolutionariesdestroy&replacethose roots. PeterSchouls(RoutledgeEncyclopediaofPhilosophy):revolutions... a.introduceradicallynovelchanges:extremelycreativeofnewformsorstructures b.areillegal c.havegreaterfreedomofsomekindasagoal(hegel Marxinfluence) (Isthistrue?Whataboutreligiousfundamentalistrevolutions?) Whatabouttimescale?Mustrevolutionsbeevent like,sudden?wecommonlydistinguish revolutionfromevolution! Whatabouttrajectories?Aren trevolutionssochaoticbynaturethattheiroutcomeis unpredictable? 3.Whatisascientificrevolution? LittletalkofscientificrevolutionuntilHerbertButterfield,TheOriginsofModernScience(1949)& especiallythomaskuhn,thestructureofscientificrevolutions(1962,1970with Postscript ). Butterfield,A.R.Halletal.spokemainlyofrevolutionsthatfoundmodernsciences,notrevolutions withinawell founded,maturescience.kuhngreatlyextendedtheapplicationof revolution. Logicalpositivism/logicalempiricism(Schlick,Neurath,Carnap,Reichenbach,Hempel)&even Popperwerenon historical&hadnoplaceforrevolution. KarlPoppertrivializeditwithhisslogan,inresponsetoKuhn:scienceas revolutioninperpetuity. I.BernardCohen s4criteriainthescientificrevolution:ahistoriographicalessay(1994): a.theoriginalscientistsmustperceivethemselvesasrevolutionaries. b.relevantcontemporariesmustagreewiththem. c.laterhistorians&philosophersmustalsoagree. d.somustlaterscientistsworkinginthatgeneralfield. Thisistoostrong?Cf.Copernicus,1543.Mendel1865.Planckin1900.Shouldthesefail? Is revolution aforward lookingvs.retrospectiveterm(like biologicalspeciation )? Scienceisnonlinear:Majorchangesofdirectionmaybeginwith normalscience.

3 3 Evenwhenscientistsareawarethattheyareengagedin revolutionary work,theycannotknowin advancewhereexactlyitwilllead.genuinerevolutionsofallkindshaveunpredictable trajectories. Hegel: TheowlofMinervatakesflightatdusk.

4 4 Does thescientificrevolution (roughly,copernicustonewton)failcohen stest? a.currentdebateaboutitsrevolutionarystatus:e.g.,shapin,thescientificrevolution: Therewasnosuchthingasthescientificrevolution,andthisbookisaboutit. b.160yearsishardlyevent like.butrelativetoalargertime scale:feudal modern? c.itwaspartofamuchwidersocialmovement,notarevoltwithinasinglediscipline. 4.ThomasKuhn smodelofscientificrevolutions Kuhn1957:TheCopernicanRevolution:triplemeaningforKuhn(akindofpunon revolution?): a.copernicus workitselfinvolvedrevolution:derevolutionibusorbiumcoelestium,1543. b.itwasarevolutioninthescientificsense,and... c.amodelforkuhn sfuturework:revolutionascognitive/conceptualreorganization (Whywasthisrevolutionsodelayed?Nointrinsicallynewtheoryordata.SeeHoward Margolis,ParadigmsandBarriers:HowHabitsofMindGovernScientificBeliefs.) InthatbookKuhnrequiredthatarevolutionhavebroadimpactonthegeneralpublicworldview, presagingchapterxofstructure: RevolutionsasChangesinWorldView. TheStructureofScientificRevolutionsrecognizes2kindsofrevolutions: a.revolutionsthatfoundascienceorevenaperiod.model:thescientificrevolution& Lavoisier. b.revolutionswithinanongoing,maturescience,e.g.,quantumtheory.thisistheshocking kindof revolutiontowhichkuhncalledattention,especiallyshockinginphysics! Kuhndividesmaturescienceintolongperiodsofnormalscienceunderaparadigm,interrupted bybriefperiodsofcrisisorextraordinarysciencethatsometimesbecomefullrevolutions, sometimesnot. (OnewayofchallengingKuhn overlysharpnormal revolutionarydistinctionistosuggest thatcrisesoccurmoreoftenthanheallowsbutareusuallyresolved,thatthereisacontinuum ofcrisisseverity.thenonlinearpotentialofcontagionisalwaysthere.) Normalscienceisconvergent.Scienceincrisisbecomesmoredivergentasconstraintsloosen. Kuhn:ascientificrevolutionisaparadigmchange,aparadigmswitch,usuallyprettyrapid. Sorevolutionsareratherevent like,episodic. Viathecrisis,mostorallofthepiecesarealreadyavailable,onlyconceptualreorganizationis needed. Normalscienceisguidedbyexemplars,notrules.Exemplars=paradigms,inthesmallsense. Paradigmsinthelargesense=exemplars+mainprinciples+standards+metaphysicalworldview. Theelementsofaparadigmareaccepteddogmatically&deeplyingrained,notsubjecttotest. exemplar=awidelyacceptedproblemformulation+solutionthatshowsthewaytofurtherwork. NotetheroleofrhetoricinKuhn:analogy,metaphor,similarityviathedirectmodelingofnew researchpuzzlesonexemplarsviaan acquiredsimilarityrelation. Notealsotheimportanceofheuristicappraisal=estimatesoffuturefertility&thusrelevanceto contextofdiscovery. Kuhnemphasizesthatthereisnosuchthingas thescientificmethod oralogicofdiscovery,nor whatphilosophershavecalledconfirmationtheoryorlogicofjustification. Kuhn sformoftheunderdeterminationproblemamplifiestheoldproblemofinduction: Majorscientificdecisionsareunderdeterminedbydata+logic.

5 5 Thisisespeciallytrueofrevolutions=paradigmswitch.Historicallycontingent,somewhat arbitrary. Hencerhetoricalpersuasionplaysamajorroleinreorganizingthesolidarityoftheexpert community. Innormalsciencethisisnotusuallyaproblem,sincetherhetoricalmodelingonpastexemplary workissodeeplyingrained. Butsinceeveryparadigmhaselementsofarbitrariness,ofhistoricalcontingency,&sincenormal scienceissofocusedondetail,itisboundtoproduceanomalies=resultsthatdon tfitthe paradigmwell. Long unresolvedanomaliescantriggeracrisisofconfidenceintheparadigm. WhatisthenatureofaKuhnianrevolution? Anunexpectednewempiricaldiscoverycantriggeracrisis(apotentialrevolutioninminiature): e.g.,x rays,specificheats(experimental).atheoreticalexamplewouldbetheclassical equipartitiontheoremleadingto theultravioletcatastrophe(ehrenfest). ButinStructureKuhnsawrevolutionsmainlynotasnewcontentbutnewformorstructure: a.aconceptualreorganizationoftheobjectivematerial(textbooks,teaching,etc.) b.acognitivepsychologicalreorganizationthatalterstheworldview c.anemotionalreorganization=majorchangeincommitmentw.r.t.promisevs.sterility, goodsciencevs.bad,newvs.old fashioned d.socialcommunityreorganization:in group,out group,controlofoldjournals,newjournals: i.e.,anewsetofnetworkconnections,anewcomplexsystemwithadifferentdynamic e.oftenareorganizationoftechnicalpractices,aswhatcountsas goodscience inthisfield QuotesfromStructure(1970,102,etc.):Einstein sspecialrelativityrevolutionmightserveas a prototype for revolutionary reorientation in the sciences. Just because it did not involve the introduction of additional objects or concepts, the transition from Newtonian to Einsteinian mechanics illustrates with particular clarity the scientific revolution as a displacement of the conceptual network through which scientists view the world. [1970, 102] b.psychologicalreorganization: gestaltswitch, differentperceptions, religiousconversion, politicalpersuasion,scientists liveindifferentworlds. HowdoKuhnianrevolutionscomparewiththeabovecriteriaofSchouls&Cohen? a.astrongbreakwithtraditionorareturntooriginalpurity?both?e.g.,copernicusvs.qt. (VassoKindisaysbothatdifferentlevels:thereisareturntopurityatthemetalevel,to save theintegrityofthediscipline.) b.kuhndoesstressrevolutionasoverturning,butreorganizationmorethancompleterazing. c.butkuhnianrevolutionsdoinvolveasharpbreak,rupture:theold&newparadigmsare incommensurable =nocommonmeasureforevaluation. d.kuhnrequiredcontemporaryawarenessofcrisis&recognition(bythewinners!)ofa successful revolution. e.duringthecrisisphase,revolutionariesareseekingfreedomfromtheoldparadigm. 5.SomeGeneralCriticismsofKuhn a.kuhn smodelofscienceduringcrisis(&histheoryofhumancognition)reducesciencetothe samelevelaspoliticaldebate,even mobpsychology (Lakatos)&thusfailstoexplainthe

6 6 historicalwatershedthattheemergenceofmodernsciencerepresents,thesuccess&progress ofthesciencesvs.otherhumanendeavors. b.inparticular,kuhnisguiltyofsubjectivism,irrationalism,irrealism,&relativism. Structureisanattackonscienceasweknowit.Kuhnisacultural leveler. c.thoaninternalisthimselfinhisownhistoricalwork,kuhn smodelviolatestheinternal external distinctionbymakingscientificdecisionsdependheavilyonexternalinfluencesratherthan oninternal,technicalevidence&argument. d.hisnotionofparadigmishopelesslyvague&ambiguous. e.therefore,soishisdistinctionofnormal&revolutionaryscience. f.kuhn shistoricalclaimisfalse:thehistoryofsciencedoesnotdivideintonormal&revolutionary periods. g.insofarasrevolutionsfollowanonlineardynamic,thelabels revolution & revolutionary are retrospective.planck1900asaproblemcase. h.kuhnemploysasloppytheoryofmeaning&referencethatleadstoanuntenableholismthat requiressharpbreakstoescapefromtheprisonofthepresent,paradigmaticconceptual scheme. i.inparticular,kuhn snotionofincommensurablebreaks... isvague&ambiguous;kuhnvacillatesbetweenstronger&weakerclaims. incoherentorself contradictoryineliminatinganybasisforoppositionofparadigms. leadstoaperniciousrelativism. is,again,historicallyfalse. 6.Incommensurability:KuhnandHisCritics (see,e.g.,shapereintheroutledgeencyclopediaofphil.,soler,sankey&hoyningen Huene, 2008) PaulFeyerabendonincommensurability:2mutuallycontradictorytheoriesareincommensurable, becausechanginganypartofatheoryholisticallychangesthemeaningoftheterms. Objection:Butifincommensurable,thenhowcantheycontradictoneanother? Mutuallycontradictory isacomparison. Incommensurability deniescomparability. Kuhn searlyholisminstructure:paradigmchangeinvolvesbreaksatseverallevels. Kuhn:Revolutionsinvolveincommensurablebreaksateachlevel,resultinginholistic,global incommensurability: a.linguistic conceptual;e.g., mass inclassicalmechanics&relativityarenotinterdefinable. Thelanguagesofclassicalmechanics&specialrelativityarenotintertanslatable. b.observational:scientistsworkingincompetingparadigmshavedifferentbasicperceptions. (Doesthisconfuseconfusehumanperceptionwithscientificobservation?) c.theoretical(substantivescientificclaims):theparadigmsarelogicallyincompatible d.normative:values&standards:local:whatconstitutesgoodsolidstatephysicsofthermal or electricalconduction? e.axiological:aimsorgoalsofscience,global&local StephenToulmin(1972).Likestructuralism,Kuhn saccountofrevolutionscannotaccountfor transitionsbetweenparadigms.theoverlyrigidconceptionofparadigmiswhatnecessitates theideaofrevolutionarybreakouts.[relatedly,kuhncouldalsobeaccusedofemployinga versionofanequilibriummodelpopularinamericansocialscience.revolutionsoccurwhen theanomaliesbecomesoseriousthatequilibriumcannotberestored.ironically,kuhn sown

7 7 theoryispartofthesocialsciencethatallowsdebateoverfundamentals.sowhywashe surprisedbystrongcriticism?] LarryLaudan(1984):Realscientificcasesshowthebreaksaremorelocal,notglobal,&thus rationaldiscourseacrossparadigmchangeispossible. ThelaterKuhnretreatedtolocalincommensurabilityinhisunfinishedbook:localtranslation failuresthatdon tpreventmutualinterpretation.sowhyisincommensurabilityanylonger interesting? ThelaterKuhnalsogivesuptheperceptualmetaphors,Gestaltswitch,etc.,&eveneliminates historyofscienceinfavoroflinguisticanalysis. Mypoint:creativeworkingscientistsareveryflexible,alwaysworkingatfrontiersofresearch. Theyfrequentlyfaceruptures&mini crisesofmanykinds.theirconcepts&techniquesare fluid,frequentlyshifting.yettheyunderstandeachother.thuskuhn sclaimsare exaggerated. Einsteinquote: Thescientist...mustappeartothesystematicepistemologistasanunscrupulous opportunist. 7.HowDoesKuhnExplainScientificRevolutions? a.anycreativedisciplineeventuallyoutgrowsitsoldconceptual&practicalframework.soany creativedisciplineunderminesitscurrentproducts,slowly&rapidly. (Toulmin,mycriticismofKuhnonnormalscienceastoostatic,Schumpeteroncapitalism.) b.revolutionsarehavehistoricallycontingentfeaturesthatnormalresearchisboundtoexpose eventually. c.paradigmsarepartofatwo tiersystem:adogmaticfoundation+asurfacelayerofnormalwork. d.paradigmsareholisticallycompact.achangerequiredanywherepropagatesthroughthesystem. e.thusrevolutionscanbehighlynonlinear.asmallcausemayhaveaverylargeeffect,&vice versa. (ThusKuhnisforcedtomakenormalsciencepractiallyimmutable,exceptforincremental, cumulativeadditions.fornormalscienceisaveryfragilestateonhisaccount.) Anormalscientificanomalycaneventuallyacquiresocial&logicalleveragetooverturnthe system,ifthereisamorepromisingproto paradigmtoreplaceit. E.g.,WilliamThomson,LordKelvin, BaltimoreLectures (1884): twosmallclouds marred thehorizonofnewtonianphysics:theanomaliesofblackbodyradiation&theresultofthe Michelson Morleyexperiment.Thelatterwascrucialtotherelativityrevolution&theformer tothequantummechanicsrevolution. f.thenewframeworkisexcitingbutperplexing&crude&leavesmuchnewworktobedone. g.thusenoughcreativesuccess+heuristicappraisalofitsfutureprospectsisalsorequired. ThusKuhnianrevolutionsarenecessarybecausecontingent!Theparadoxisonlyapparent. Andthusrevolutionsarenecessaryforcontinuedprogress,contrarytopositivisttradition, whichenvisionedsteady,cumulativeprogressconsistentwiththepresentconceptual framework. E.g.,P.W.BridgmanoperationisminTheLogicofModernPhysics(1927). Weshouldnowmakeitourbusinesstounderstandsothoroughlythecharacterofourpermanentmental relationstonaturethatanotherchangeinourattitude,suchasthatduetoeinstein,shallbeforever impossible.itwasperhapsexcusablethatarevolutioninmentalattitudeshouldoccuronce,becauseafter allphysicsisayoungscience,andphysicistshavebeenverybusy,butitwouldcertainlybeareproachif sucharevolutionshouldeverprovenecessaryagain.[1927,2]

8 8 Bridgman(aNobelLaureateinexperimentalphysics!)completelymisunderstandshownew concepts,language,&proposedmechanismsemergeasproductsoftheongoingresearch processratherthanprovidingafixed,antecedentfoundationforit.thisisthesamemistake astraditionalconceptionsof thescientificmethod vs.pragmaticconsequentialism. ContrarytoBridgman,Kuhnmultipliedrevolutions:everysmallspecialtyhasthem:hierarchy. Thevisibilityproblem.Ifcontemporaryscientistsmustbeawareofcrisis&revolution&ifthere havebeensomanyrevolutions,thenwhydidtheworldhavetowaitforkuhntoseethem? Because,hesaid,theyarelargelyinvisible.Butwhy? a.afterarevolution,thewinnersrewritethehistoryofsciencetomakeitlookasifthepresent paradigmisthebrilliantbutrationalsequeltopreviouswork.theyadheretotheolderview thatrespectablerationaldevelopmentrequirescumulativecontinuity. b.theimplication:onlysomeoneofkuhn shistoricalsensitivitycouldnoticethisfromoutsidethe field. Eventoscientistsinotherspecialtiesrevolutionsusuallylooklikecumulativeadvances. c.disciplinaryconsequence:anewgenerationofhistorians&historicalphilosophersfoundsubtle conceptualbreakseverywhere! 8a.SomePhilosophicalImplications:ScientificProgress a.kuhn:normalscienceisprogressiveinacumulativeway. b.revolutionaryscienceisnotcumulative,butitisstillprogressiveinhavingwiderscope,more precision,greatertheoreticalsophistication,etc.wecaneasilyorderparadigmshistorically. c.thisisprogresswithrespecttopreviousresults,notprogresstowardsomefinaltheory,some finaltruth,writtenin nature sownlanguage (RichardRorty),waitingtobediscovered. d.rortyetal.onthepragmaticimplication:scienceisjustonemorehumanprojectamongothers. It sgoalsvarywithhumaninterests&henceitscriteriaofsuccess. 8b.SomePhilosophicalImplications:RealismandTruth Revolutionsprovideoneargumentforlocalrealismandstructuralrealism:whateversurvivesa revolutionisrobust:invariantunderchange. But,onthewhole,theexistenceofrevolutions,especiallyinmaturescience,isashocktorealism. Whyinphysics,ourmostmaturescience?!IsKuhn sanswerbasicallycorrect? InimplyingthatscienceisnotgraduallyprogressingtowardafinalTruth,revolutionsprovidethe strongestformoflarrylaudan snegativehistoricalinduction(laudan1981). TheyarealsothestrongestformofKyleStanford s(2006)problemofunconceivedor underconceivedalternatives. Futurerevolutionsarebeyondourpresentconceptual linguistic&thusimaginativehorizon. Wemayeventhinkthemimpossible,bytoday slights.considerpastexamples: a.euclideangeometryasaprioritrue,noconsistentalternativeisconceivable(kant). b.darwinontheevolutionofspecies. c.quantumtheoryrejectsuniversalcausation&acceptsentanglement. DudleyShapere:modernsciencegetsweirder&weirder,weirderthanevenprevioussciencefiction writerscouldimagine. MyKuhnianpoint:theKuhnianrevolutionarypotentialofascienceisafunctionofitscompactness &henceitsmaturity,&henceitsnonlinearity.smallviolationscanpropagaterapidlythruthe network:tippingpoints,criticalpoints,phasechanges,contagion,epidemiologicalmodels. 8c.SomePhilosophicalImplications:SocialConstructionism

9 9 Kuhn semphasisonthecommunityofexpertstriggerednew wavesociologyofscience. Structuredramaticallysharpenedtheoldunderdeterminationproblem:thegapbetweenour theories&whatthedata+logicdemandisverylargewithmuchconstructiveadjustment& negotiationpossible. Evenifmetaphysicalrealismistrue(Thereisonebest,objectivelytruedescriptionofreality.),itis notgiventous;hence,agreatdealofhumanconstructionisalwaysinvolved. Moreradicalpositions: a.metaphysicalrealismiscompletelybeyondus,soshouldnotbetakenseriously. b.metaphysicalrealismisincoherent(rorty).sowearefreefromthatsortofobligationtonature. Scienceisahumanlyconstructedtooldesignedtosatisfyhumaninterests,nothingmore. 9.SomeNon KuhnianConceptionsofRevolutionandRupture CharlesPeirce,thefounderofAmericanpragmatism,saidinanearlyessay, eachchiefstepin sciencehasbeenalessoninlogic, implyingthatphilosophymustlearnfromscience;hence majorscientificlessonswillalsobringmethodological epistemologicallessons(peirce1877). ButPeirceapparentlydidnotseetherevolutionarypotentialofthislinkageofsubstance& method.mostfin de sièclescientistssawwhatwenowcalltheclassicalworldpictureas approachingthefinaltruthabouttheworld. HowardMargolis,ParadigmsandBarriers(1993),distinguishes2modelsofrevolution. a.thegapview:thereisalargelogicalgaptobejumpedbetweentheold&thenewscience, e.g.,possiblytheemergenceofthemodernconceptionofprobability:seecbelow. b.thebarrierview:thereisacognitivebarrier(adeeplyentrenched,oftenunconscious habit of mind )thatblockstheconceptualbreakthrough.whensomeonedoesbreakthroughor evade thebarrier,itconflictswitholdintuitions&causesincommensurability,e.g.,copernicus. c.apossible3rdmodel:openingupawholenewdomain,withoutdirectconflict. (SomethingsimilartoMargolis barrierviewwouldseemtofitevennormalscience,wherethe scientistsareconvincedthattheyhaveallthetoolsnecessarytosolvethepuzzlebutcannot. Butwhenonedoessolveit,recognitionisimmediate.Butno:thereneedbenoconflictor incommensurabilityinthiscase,justboundedrationality.andtheheuristicexpectationsin termsofavailabletechniquesarealreadyinplace,whereascopernicanfrontiers&intuitions donotknowwhichinformationisrelevant,&itmaybequiteperipheral.todaythereisthe knowledge pollutionproblem.) PaulThagard,ConceptualRevolutions(1993)articulatesabroadlyKuhnianmodelin computationaltermsviahiscomputerscienceprogram,echo,especiallytaxonomic reorganizationintermsoftreestructures.(1) Branchjumping relocatessomethingto anotherbranchofthesametree,e.g.,whaletomammal.(2) Treeswitching moreradically replacesthetreebyareorganizedone,aswhendarwin sbiologicalconceptionofbiological taxonomyreplacedlinnnaeus. PeterGodfrey Smith,inareviewofJablonska&Lamb,EvolutioninFourDimensions(2005), doubtsthatkuhnianrevolutionscanoccurinthebiologicalsciencesafterdarwin&perhaps post 1900Mendeliangenetics.(1)Biologyistoolooselystructuredlogically.(2)Thereare norapid,reallydecisiveoverturningevents,onlyadelugeofnewinformation&techniques, asinmoleculargenetics.

10 10 [Butcomparetheso calledscientificrevolution.]e.g.,crick scentraldogmawasnot decisivelyoverturned,justslowlyeroded.revolutioninbiologywillthusbearetrospective label,e.g.,ifthednasequenceturnsoutnottobesocentralgivenotherformsofinherited informationtransmission. TheFrenchtraditionbeforeKuhn: rupture, coupure,abreakwiththepast,isaprominent theme. ButKuhnwaslargelyunfamiliarwiththeirwork. LéonBrunschvicgandGastonBachelard:AsforPeirce,philosophymustlearnfromcurrent science. GastonBachelard(1934):Wecanunderstandscientificreasononlyviathehistoryofscience. Afruitfulprogram,suchasclassicalmechanicseventuallybecomesunfruitful,anobstacleto progress. Ascientificrevolutionbecomesnecessary&shouldproduceanepistemologicalrevolution. ButcoupuresforBachelardarenotasradicalasforKuhn.Heretainsprogresstowardtruth&a kindofcumulativityinthatanearliertheoryremainsinthelateroneasaspecialcase. Canguilhemwasmoreinterestedinthebiological&healthsciencesthanphysics&stressedthat thesesciencescannotreducetophysicsbecause,unlikephysics,theyinvolveadistinction betweenthenormal&thepathological. He&BachelardcriticizedKuhn snotionoftrans revolutionscientificnorms&scientificconsensus astooweaktoexplainthekindsofclosureweactuallyfindinthehistoryofscience. MichelFoucault,inTheArcheologyofKnowledge,etc.,introducestheideaofdiscursiveformations &ofbreaksbetweenthem.thesearemuchlargerculturalbreaksthankuhn sscientific revolutions,althoughsciencescanemergeatthesepoints,e.g.,thesciencesofsocialcontrol. IanHacking(Canadian)isstronglyinfluencedbyFoucault.Hackingwrites: FoucaultusedtheFrenchworldconnaissancetostandforsuchitemsofsurfaceknowledgewhilesavoirmeant morethanscience;itwasaframe,postulatedbyfoucault,withinwhichsurfacehypothesesgottheirsense. Savoirisnotknowledgeinthesenseofabunchofsolidpropositions.This depth knowledgeismorelikea postulatedsetofrulesthatdeterminewhatkindsofsentencesaregoingtocountastrueorfalseinsomedomain. Thekindsofthingstobesaidaboutthebrainin1780arenotthekindsofthingstobesaidaquarter century later.thatisnotbecausewehavedifferentbeliefsaboutbrains,butbecause brain denotesanewkindof objectinthelaterdiscourse,andoccursindifferentsortsofsentences.[2002,77] Hacking s historicalontology developsaversionofthisview:modernprobability&statistics&all the real quantitiestheyinvolvearehumanhistoricalconstructions,yetnotthereforeunreal. Sowholewaysofspeaking&understandingtheworld,wholelanguages&theirassociated techniques&realities,comeintoexistence&displaceolderways again,broaderthan Kuhnianrevolutions. LikeFoucault,Hackinglikestheideaofhistoricalaprioris:taken for grantedplatformsthatcan tbe meaningfullydisputedortested(likekuhnianparadigmsduringnormalscience). Thehistoricalaprioripointsatconditionswhosedominionisasinexorable,thereandthen,asKant s syntheticapriori.yettheyareatthesametimeconditionedandformedinhistory,andcanbeuprooted bylater,radical,historicaltransformations.t.s.kuhn sparadigmshavesomeofthecharacterofa historicalapriori.[hacking2002,5].

11 11 Hackingalsohasakindofincommensurability(morelikeFeyerabend sthankuhn s)&akindof relativism: JustasstatisticalreasonshadnoforcefortheGreeks,sooneimaginesapeoplefor whomnoneofourreasonsforbeliefhaveforce (2002,163). MichaelFriedman(2001)alsocanbereadasdefendingasortofhistoricalapriori,intheGerman traditionofneo Kantianism&Reichenbach.Withoutfullycommittinghimself(?)toKuhnian revolutions,heshowshowtointerpretthemintermsofahistoricallyrelativizedapriori. Friedmanhimselfdefendsa2 tieredviewofscience:itneedsanunderlying apriori framework. ThisisthemodernscientificversionofKant s2 tieredviewofcognition:anypositionthatlacksa bottomlevelofbasicdefiningorprocessingrules(categories&formsofintuition)isanaïve empiricism! Kuhn sviewinstructureisclearly2 tiered.heis akantianwithmoveablecategories. 10.ComparisonwithEconomicInnovation SeemyStanfordEncyclopediaofPhilosophyarticle. 11.RevolutionorEvolution? Arethere/havetherebeenanyKuhnianrevolutions? Kuhn saristotleepiphany:misleading:tooperceptual+hugetime jump. Isitamatterofscale?E.g.,theCopernicanRevolutionvs.thePlateTectonicRevolution. Thecloserwelookata revolution, thelessrevolutionaryitbecomes becausedozensofsmaller stepsarenowfilledin,eachofthema stretch butnotrevolutionary. TowardtheendofStructure,KuhnhimselfcomparedrevolutionwithDarwinianbiological speciation. Amodifiedviewisevenstrongerinhislastwork: [R]evolutions,whichproducenewdivisionsbetweenfieldsinscientificdevelopment,aremuchlike episodesofspeciationinbiologicalevolution.thebiologicalparalleltorevolutionarychangeisnot mutation,asithoughtformanyyears,butspeciation...[scientificspecialtiesareanalogousto reproductivelyisolatedbiologicalpopulationsinbeinga]communityofintercommunicatingspecialists,a unitwhosememberssharealexiconthatprovidesthebasisforboththeconductandtheevaluationof theirresearchandwhichsimultaneously,bybarringfullcommunicationwiththoseoutsidethegroup, maintainstheirisolationfrompractitionersofotherspecialties.[kuhn2000c,98] a.darwinianevolutionisslow&gradual&continuous,notrapid&discontinuous. b.speciationisaretrospectivecategory.soshouldn trevolutionalsoberetrospective? Onereply:treatKuhnianrevolutionsasanalogoustoNilesEldredge&StephenJayGould punctuated equilibrium conceptionofevolution.inalaterarticletheynoteaconnectiontokuhn: [C]ontemporarysciencehasmassivelysubstitutednotionsofindeterminacy,historicalcontingency,chaos andpunctuationforpreviousconvictionsaboutgradual,progressive,predictabledeterminism.these transitionshaveoccurredinfieldafterfield;kuhn scelebratednotionofscientificrevolutionsis,for example,apunctuationtheoryforthehistoryofscientificideas.[1993,227] StuartKauffman(1993)andBrianGoodwin(1994)defendreorganizationintheformofselforganization astheprimarymacro evolutionarymechanism,withevolutionaryadaptationaddingonlythe finishing

12 12 touches. GouldandRichardLewontinhadraisedthispossibilityintheirfamouspaperof1979, The Spandrelsof SanMarcoandthePanglossianParadigm. Appliedtothedevelopmentofscience,thisview implies thatrevolutionsdeterminetheoverallshape,whileordinaryscientificworkappliestheadaptive microevolution. ComplexitytheoristssuchasphysicistsPerBak(1996)&Albert LászlóBarabási(2007)propose more Hegelian modelsofcreativestructuralchange,inthesensethattheyrequirenoagencyof great men or greatdiscoveries, onlyanetworkeddistributionofrelativelydumbprocessors. Iftheyareright,thiscouldgiveadifferentwayofclassifyingrevolutions,dependingontheir aetiology: anonymousnonlineartransformationsvs.majorcontributions. Myview: revolution isausefuldescriptivetermtorecognizesciencesasdynamicalsystems,but Kuhn exaggeratesthenormal revolutiondistinction&theextentofthebreaks. Creativescientistsareusedtotheuncertainties,confusions,&dangersoffrontiers. 12.Kuhn sattempttounitehistoricismandnaturalism Roughly,evolutionisanaturalisticprocess,whilerevolutionsarehumanhistoricalprocesses. Kuhnattemptstoovercometheoldoppositionbetweendeephistoricism(whichgenerallystresses the humanlyconstructed,contingent,local&whatrequireshumaninterpretationorverstehen)vs. thoroughgoingnaturalism(stressingobjectivelynecessary&universallawsofnature& Erklären). Kuhn sstronganti whiggismisdeeplyhistoricist,asishisviewofsocialconstruction,rejectionof final truth,&thegeneration&corruptioninhistoricaltimeofallthingsscientific. Yetheimposesa lawful patternon(someof)thematuresciences:normalscience+revolution. OliverWendellHolmes,Jr.(1861)onceremarked, Revolutionsneverfollowprecedentsnorfurnish them. Giventheunpredictability,thenonlinearity,theseeminguniquenessofrevolutions, whether politicalorscientific,itisthereforesurprisingtofindkuhnattemptingtoprovideageneral Theoryof ScientificRevolutions. CompareMarx smaterialisttheoryofhistoryasasimilarattempttobebothnaturalistic& historicist. (ButunlikeMarx s,kuhn stheoryisnotaconflicttheory,nordoesittellastorywitheitheran ending oragoal.) Unfortunately,thelaterKuhngraduallygivesuphistory,movingfromHegelbacktoKant& emphasizing aprioriargumentsandlinguisticanalysis,whiletakinglittlenoticeofrapiddevelopmentsinthe empiricalcognitivesciences.

13 13