11/23/08 1

BayesianEpistemology

AlanHjekandStephanHartmann

1.Introduction

Bayesianismisourleadingtheoryofuncertainty.Epistemologyisdefinedasthe

theory of knowledge. So Bayesian Epistemology may sound like an oxymoron.

Bayesianism, after all, studies the properties and dynamics of degrees of belief,

understoodtobeprobabilities.Traditionalepistemology,ontheotherhand,placesthe

singularlynonprobabilisticnotionofknowledgeatcentrestage,andtotheextentthat

ittraffics inbelief,thatnotiondoesnotcomeindegrees.Sohowcantherebea

Bayesianepistemology?

Accordingtooneview,therecannot:Bayesianismfailstodojusticetoessential

aspects of knowledge and belief, and as such it cannot provide a genuine

epistemology at all. According to another view, Bayesianism should supersede

traditional epistemology: where the latter has been mired in endless debates over

skepticism and Gettierology, Bayesianism offers the epistemologist a thriving

research program. We will advocate a more moderate view: Bayesianism can

illuminatevariouslongstandingproblemsofepistemology,whilenotaddressingall

ofthem;andwhileBayesianismopensupfascinatingnewareasofresearch,itbyno

meansclosesdownthestaplepreoccupationsoftraditionalepistemology.

Thecontrastbetweenthetwoepistemologiescanbetracedbacktothemid17th

century. Descartes regarded belief as an allornothing matter, and he sought

justificationsforhisclaimstoknowledgeinthefaceofpowerfulskepticalarguments.

No more than four years after his death, Pascal and Fermat inaugurated the
11/23/08 2

probabilistic revolution, writ large in the PortRoyale Logic, in which the many

shadesofuncertaintyarerepresentedwithprobabilities,andrationaldecisionmaking

isamatterofmaximizingexpectedutilities(aswenowcallthem).Correspondingly,

theCartesianconcernforknowledgefadesintothebackground,andamorenuanced

representationofepistemicstateshasthelimelight.Theisticbeliefprovidesavivid

exampleofthecontrastingorientations.Descartessoughtcertaintyintheexistenceof

Godgroundedinapodeicticdemonstrations.Pascal,bycontrast,explicitlyshunned

suchallegedproofs,arguinginsteadthatoursituationwithrespecttoGodislikea

gamble,andthatbeliefinGodisthebestbetthusturningthequestionoftheistic

beliefintoa decision problem(whichhe,unlikeDescartes,hadthetoolstosolve).

BayesianepistemologyowesitsnametotheReverendThomasBayes,who,acentury

later, published an important theorem that underwrites certain calculations of

conditional probability central to confirmation theorymore on this shortly. But

really Bayesian epistemology is something of a misnomer; Kolmogorovian

epistemologywouldbefarmoreappropriate,aswewillsee.

Caveats: When we speak of traditional epistemology, we lump together a

plethoraofpositionsasiftheyformamonolithicwhole.Otherarticlesinthisvolume

distinguishcarefullyamongvariouspositionsthatourbroadbannerconflates.Forour

purposes,theystartoutregardingknowledgeandbeliefasthecentralconceptsof

epistemology,andthentogoontostudytheproperties,grounds,andlimitsofthese

binarynotions.WealsospeakofBayesianismasifitisaunifiedschoolofthought,

wheninfacttherearenumerousintramuraldisputes.I.J.Good(1971)calculatesthat

thereare(atleast)46,656waystobeaBayesian,whilewewillmostlypretendthat

thereisjustone.Byandlarge,thevariousdistinctionsamongBayesianswillnot
11/23/08 3

matter for our purposes. As a good (indeed, a Good) Bayesian might say, our

conclusions will be robust under various precisifications of the position. Many

traditionalproblemscanbeframed,andprogresscanbemadeonthem,usingthe

tools of probability theory. But Bayesian epistemology does not merely recreate

traditionalepistemology;thankstoitsconsiderableexpressivepower,italsoopensup

newlinesofenquiry.

2.WhatisBayesianEpistemology?

Bayesian epistemology is the application of Bayesian methods to epistemological

problems.Bayesianismmodelsdegreesofbeliefas probabilities alongthelinesof

Kolmogorovs(1933)axiomatization.Let beanonemptyset.Afield(algebra)on

isaset F ofsubsetsof thathas asamember,andthatisclosedunder

complementation(withrespectto )andunion.Let P beafunctionfrom F tothe

realnumbersobeying:

1)P(a)0forallaF. (Nonnegativity)

2)P( )=1. (Normalization)

3)P(ab)=P(a)+P(b)foralla,bFsuchthata b=. ( Finiteadditivity)

CallPaprobabilityfunction,and( ,F,P)aprobabilityspace.

Onecouldinsteadattachprobabilitiestomembersofacollectionofsentencesofa

formallanguage,closedundertruthfunctionalcombinations;thisismorecommonin

Bayesian confirmation theory. A lively area of current debate concerns just how

finely grained such contents of probability attributions should be. For example,

variousproblemsofselflocationsuggestthatprobabilitiesshouldattachtocentred
11/23/08 4

propositions,e.g.<possibleworld,individual,time>triples.

Kolmogorov extends his axiomatization to cover infinite probability spaces,

requiring F tobeclosedunder countable union,andstrengthening3)to countable

additivity. He defines the conditional probability of a given b by the ratio of

unconditionalprobabilities:

P(a b)
P(a|b)= ,providedP(b)>0.
P(b)

If P(a| b) = P(a), then a and b are said to be independent (relative to P).

VersionsofBayestheoremcannowbeproven:

P( b | a)P(a )
P(a|b) =
P( b)

P(b | a)P(a)
=
P(b | a)P(a) + P(b | a)P( a)

More generally, suppose there is a partition of hypotheses {h1, h2, ... , hn}, and

evidencee.Thenforeachi,

P(e | hi )P(hi )

P(hi|e)=
P(e | h )P(h ) .
j j
j=

The P(e |hi)termsarecalled likelihoods,andtheP(hi)termsarecalled priors.See

Joyce(2008)foramoredetaileddiscussionofvarietiesandusesofBayestheorem.

Bayesianismoffersanaturalanalysisoftherelationof confirmation betweena

pieceofevidenceeandahypothesish:

econfirmsh(relativetoP)iffP(h|e)>P(h).

Wemayalsodefinevariousprobabilisticnotionsof comparative confirmation,and

various measures of evidential support (see Eells and Fitelson 2000 and Fitelson

1999).
11/23/08 5

Bayesianismcanbeunderstoodascombiningasynchronicthesisaboutthedegrees

ofbelieforcredencesofarationalagentatagiventime,andadiachronicthesisabout

howtheyevolveinresponsetoevidence.Synchronically,theagentscredencesare

probabilities. Diachronically, her credences update according to the rule of

conditionalization. Suppose that initially her credences are given by probability

function Pinitial, and that she becomes certain of e (where e is the strongest such

proposition).ThenhernewcredencefunctionPnewisrelatedtoPinitialasfollows:

(Conditionalization) Pnew(x)=Pinitial(x|e)(providedPinitial(e)>0).

Jeffreyconditionalization allowsforlessdecisivelearningexperiencesinwhich

herprobabilitiesacrossapartition{e1,e2,...}changeto{Pnew(e1),Pnew(e2),..., },

wherenoneofthesevaluesneedbe0or1:

Pnew(x)= iPinitial(x|ei)Pnew(ei)(providedPinitial(ei)>0).

(Jeffrey1983).Pnew(x)iscalledtheposteriorprobabilityfunction.

According to Bayesian orthodoxy, an agent begins with a prior probability

functionandrepeatedlyupdatesby(Jeffrey)conditionalizationasevidencecomesin.

This combines a striking permissiveness about the starting point of an agents

epistemological odyssey with considerable rigidity about how the agent should

respondtoevidence.ButBayesianismisathemethatadmitsofmanyvariationssee

Good(1971)andHjekandHall(2002)fordiscussionofsomeofthem.Here,letus

considerseveralfurtherconstraintsonpriorsthathavebeenproposed.

A probability function is said to be regular if it assigns probability 1 only to

tautologies,andprobability0onlytocontradictionstoallothersentencesitassigns

intermediatevalues.Itseemstobeanepistemologicaldesideratumthatapriorbe

regular,reflectinganopenmindednessappropriatetoanagentwhoisatabularasa.
11/23/08 6

A rationale is that to rule out (probabilistically speaking) a priori some genuine

logicalpossibilitywouldbetopretendthatonesevidencewasstrongerthanitreally

was. The principle ofindifference alsoenjoins youtoreflect thepovertyofyour

evidence in your credences: you are required to give equal probability to all

possibilitiesamongwhichyourevidencedoesnotdiscriminate(andinastateoftotal

evidentialinnocence,thatisallofthem).Asophisticatedversionoftheprincipleof

indifference,favouredbysocalledobjectiveBayesians,hasbeenexploredbyJaynes

(2003): maximize the probability functions entropy, which for an assignment of

positive probabilities p1, ..., pn to n possibilities equals i pi log(pi). See also

Williamson(2005).

Thentherearetwoprinciplesthataremeanttocodifyonesepistemiccommitment

toaligningonescredencestocertainprobabilistichypotheses.Inthefirst,Lewiss

PrincipalPrinciple(heresimplified),thehypothesesconcerntheobjectivechanceof

therelevantpropositions(Lewis1980):

C0(a|cht(a)=x)=x(forallaandtforwhichthisisdefined).

HereC0issomereasonableprior,aanarbitraryproposition,andcht(a)=xtheclaim

thatthechanceattimetofaisx.Theideaisthatoneshouldalignonescredences

withwhatonetakesthecorrespondingobjectivechancestobe,wherethelatterare

genuineprobabilitiesintheworld.Inthesecond,vanFraassensReflectionPrinciple

(van Fraassen 1984), the hypotheses concern ones own future credences for the

relevantpropositions

Ct(a|Ct (a)=x)=x(forallt,t ,aandxforwhichthisisdefined).

HereCtisonesprobabilityfunctionattimet,andCt onesfunctionatlatertimet .
11/23/08 7

Theideaisthatrationalityrequiresacertaincommitmenttoonesfutureopinions;

when all is going well, ones future selves are betterinformed versions of ones

currentself.

3.ContrastsbetweenTraditionalEpistemologyandBayesianEpistemology

WecannowbringoutseveralpointsofcontrastbetweentraditionalandBayesian

epistemology.Wehavenotedthatknowledgeandbeliefarebinarynotions,tobe

contrastedwiththepotentiallyinfinitelymanydegreesofcredence(corresponding

toalltherealnumbersinthe[0,1]interval).Knowledgeisnotmerelyjustifiedtrue

belief, asGettier hasfamously shown,butmanyepistemologists hopethatsome

fourthconditionwillcompletetheanalysissomekindofconditionthatrulesout

casesinwhichonehasajustifiedtruebeliefbyluck,orforsomeanomalousreason.

Someepistemologists(e.g.Sosa(1999))advocateversionsofsafetyasaconditionon

knowledgeroughly,attheclosestworldsinwhichagivenagentbelieves p,p is

true.Others(e.g.Nozick(1981))advocatesensitivityroughly,attheclosestworlds

inwhich p isfalse,theagentdoesnotbelieve p.(Andsomeadvocateboth.)Note

well:the closest worlds.Herewefindanotherdisjuncturebetweentraditionaland

Bayesian epistemology: nothing in the standard Bayesian apparatus reflects the

similarityofworldsthathastakencentrestageintheanalysisofknowledge.

Noticethattruth,justification,andtheseantiluckconditionsmaybecharacterized

as at least partially objective, with belief providing the only purely subjective

component. This is insharpcontrast toorthodox Bayesianism, which refines and

analyzes this doxastic notion, but which has no clear analogue of the objective

conditions.Mostimportantly,Bayesianismapparentlyhasnothingthatcorrespondsto
11/23/08 8

thefactivityofknowledge:thatonecanonlyknowtruths.Andevenwhenourbeliefs

fallshortofknowledge,stillitisadesideratumthattheybetrue;buttheBayesian

seemstohavenocorrespondingdesideratumforintermediatecredences,whichareits

stockintrade.Whenyouassign,forexample,probability0.3toitrainingtomorrow,

whatsensecanbemadeofthisassignmentbeing true? Itisalsodubiouswhether

Bayesianism can capture justification (cf. Shogenji (2009) for an interesting

proposal)oranyantiluckconditiononknowledgemoreonthisshortly.

Relatedly, allBayesianclaims mustberelativized toaprobability function,or

more precisely, to a probability spacean entire probability model. We saw this

above in the definitions of independence and confirmationthey came with

parentheticalreferencesto P. Manyauthorssuppressthesereferences,encouraging

onetoforgettheirinherentsubjectivity(andevenwesecretedthemawayinslightly

disingenuousparentheses!).Traditionalepistemologists,bycontrast,conductmuchof

their discussions in terms free of any such relativizationthey speak of one

proposition being evidence for another, of a process of belief acquisition being

reliable,andsoonwithoutanyqualification.Andagain,truth,justification,andanti

luckconditionsaretypicallysupposedtoholdornotindependentlyofwhethersome

agentthinksthattheydo,orwhethersomemodelsaystheydo.

Thesynchronicrequirementthatanagentscredencesobeytheprobabilitycalculus

may be regarded as generalizing the requirement familiar from traditional

epistemologythatonesbeliefsshouldbeconsistent.Thediachronicrequirementof

conditionalizationisreminiscentoftheQuineanprincipleofminimummutilation

(change beliefs as little as the evidence allows)a conservative recipe for belief

revision. But nothing in traditional epistemology corresponds to Jeffrey

11/23/08 9

conditionalizationit is essentially a probabilistic revision rule. The principle of

indifferencecorrespondsveryroughlytoaCartesianadmonitiontosuspendjudgment

whenonesevidenceislacking,butitisfarmorespecific.Andtherearenotraditional

analoguesofthevariousadditionalconstraintsonpriors.Goingintheotherdirection,

Bayesianismissilentaboutsomeofthecornerstonesandmorerecentconcernsof

traditionalepistemologywewilldiscussthisatgreaterlengthattheend.

GiventhestrikingdifferencesbetweentraditionalandBayesianepistemology,are

therereasonstopreferonetotheother?

4.Thesis:BayesianEpistemologyIsSuperiortoTraditionalEpistemology

Jeffrey, a famous Bayesian, suggests two main benefits accrued by the Bayesian

frameworkinhis(1992):

1.Subjectiveprobabilitiesfigureindecisiontheory,anaccountofhowouropinions

andourdesiresconspiretodictatewhatweshoulddo.Thedesirabilityofeachofour

possibleactionsismeasuredbyitsexpectedutility,aprobabilityweightedsumofthe

utilitiesassociatedwiththataction.TocompleteJeffreysargument,weshouldadd

that traditional epistemology offers no decision theory (recall Descartes versus

Pascal).Theanalysisofrationalactionsurelyneedstoadverttomorefinegrained

mentalstatesthanbinarybeliefandknowledge.(SeeErikssonandHjek2007for

morediscussionofwhytheintermediatecredencesthatarenecessaryforthatanalysis

cannotbereducedtothesebinarynotions.)

2.Observationsrarelydelivercertaintiesrather,theireffectistypicallytoraiseour
11/23/08 10

probabilitiesforcertainpropositions(andtodropourprobabilitiesforothers),without

anyreachingtheextremesof1or0.Traditionalepistemologyapparentlyhasnoway

of accommodating such lessthanconclusive experiential inputs, whereas Jeffrey

conditionalizationistailormadetodoso.

We may continue the list that Jeffrey has started of putative advantages of

Bayesianismovertraditionalepistemologyatsomelength:

3.Knowledgeisunforgiving.Itsstandardsaresohighthattheycanrarelybemet,at

leastincertaincontexts.(Thisisrelatedtothefactthatknowledgedoesnotcomein

degreesnearknowledgeisnotknowledgeatall.)Thisinturnplaysintothehands

of skeptics. But it is harder for skeptical arguments to get a toehold against the

Bayesian.Forexample,themere possibility oferrorregardingsomeproposition X

underminesaclaimofknowledgeregardingX,butitisinnocuousfromaprobabilistic

pointofview:anagentcansimplyassign X somesuitableprobabilitylessthan1.

Indeed,evenanassignmentofprobability1isconsistentwiththepossibilityoferror

underplausibleassumptions,itcanbeshownthatadartthrownatrandomata

representationofthe[0,1]intervalhasprobability1ofhittinganirrationalnumber,

eventhoughitmightfailtodoso.

4.Moreover,itisaplatitudethatdoxasticstatescomeindegrees,andthecategories

of beliefandknowledgearetoocoarsegrained todojustice tothisfact. You

believe,amongotherthings,that2+2=4,thatyouhaveahand,thatLondonisin

England, and (say) that Khartoum is in Sudan. But you do not have the same
11/23/08 11

confidenceinallthesepropositions,aswecaneasilyrevealinyourbettingbehavior

and otherdecisionmakingthatyoumightengagein.Theimpoverished natureof

belief attributions is only exacerbated when we consider the wide range of

propositionsforwhichyouhavelessconfidencethatthiscoinwilllandheads,that

itwillraintomorrowinNovosibirsk,andsoon.Wemayconflateyourattitudesto

them all as suspensions of belief (as Descartes would), but that belies their

underlyingstructure.Suchattitudesarebetterunderstoodassubjectiveprobabilities.

5. Relatedly, the conceptual apparatus of deductivism is impoverished, and

comparativelylittleofourreasoningcanbecapturedbyit,eitherinscienceorindaily

life(pace PopperandHempel).Afterall,whetherwelikeitornot,ourepistemic

practicesconstantlybetrayourcommitmenttorelationsofsupportthatfallshortof

entailment(OaksfordandChater2007).Wethinkthatitwouldbeirrationaltodeny

thatthesunwillrisetomorrow,toprojectgrueratherthangreeninourinductions,

andtocommitthegamblers fallacy.Probabilitytheoryhelpsustounderstandsuch

relations.

6. Bayesianism has powerful mathematical underpinnings. It can help itself to a

century ofworkinprobabilitytheoryandstatistics.Traditionalepistemologymay

appeal to the occasional system of epistemic or doxastic logic, but nothing

comparabletotheformidableformalmachinerythatwefindintheBayesianstool

kit.

7. Bayesian methods, in turn, have much wider application than any formal
11/23/08 12

systematization of knowledge or belief. Look at the sciences, social sciences,

engineering,andartificialintelligenceifyouneedanyconvincingofthis.

8.TherearemanyargumentsforBayesianism,whichcollectivelyprovideakindof

triangulation to it. For example, Dutch Book arguments provide an important

defenseofthethesisthatrationalcredencesareprobabilities.Anagentscredences

areidentifiedwithherbettingprices;itisthenshownthatsheissusceptibletosure

losses iff these prices do not conform to Kolmogorovs axioms. There are also

arguments from various decisiontheoretic representation theorems (Ramsey 1931,

Savage1954,Joyce1999),fromcalibration(vanFraassen1984),fromgradational

accuracyorminimizationofdiscrepancyfromtruth(Joyce1998),fromqualitative

constraintsonreasonableopinion(seeEarman1996foradiscussionofresultsofCox

and others), and so on. Moreover, there are various arguments in support of

conditionalization and Jeffrey conditionalizatione.g., Dutch book arguments

(Armendt 1980, Lewis 1999) and arguments from minimal revision of ones

credences (Diaconis and Zabell 1982). Again, there is nothing comparable in

traditionalepistemology.

9. Finally,apragmaticargumentforBayesianismcomesfromanevaluationofits

fruits.Asweshowatgreaterlengthinsection7,Bayesianismishighlyexplanatory

withminimalresourcesasimple,fecundtheoryifevertherewasone.Traditional

epistemologyishardpressedtoofferthesamerewards.Forexample,wewillseehow

various important intuitions about confirmation can be vindicated by a Bayesian

analysis,andsomeerroneousintuitionscanbecorrected.Itseemsthatnoanalysis
11/23/08 13

couchedpurelyintermsofknowledgeandbeliefcouldpaysuchdividends.

So weseevariousadvantages thatBayesianism apparently hasovertraditional

epistemology.Butthisdoesnottellthewholestory.Forstarters,thetriumphsof

Bayesian confirmation theory just touted are supposedly offset by the socalled

problem of old evidence (Glymour 1980). If P(e) = 1, then e apparently cannot

confirmanythingbyBayesianlights:inthatcase,P(h|e)=P(he)/P(e)=P(h).Yet

weoftenthinkthatsucholdevidencecanbeconfirmatory.Considertheevidenceof

theadvanceoftheperihelionofMercury,whichwasknowntoEinsteinatthetime

thatheformulatedgeneralrelativitytheory,andthus(wemayassume)wasassigned

probability 1 by him. Nonetheless, he rightly regarded this evidence as strongly

confirmatoryofgeneralrelativitytheory.ThechallengeforBayesiansistoaccount

forthis.(SeeZynda(1995)fordiscussion.)

Bayesianism,then,isnotwithoutproblemsofitsown.Soletusrevisitthecontest

betweentraditionalepistemologyandBayesianism,thistimelookingatarguments

thatinclineinfavoroftheformer.

5. Antithesis: Bayesian Epistemology Is Not Superior to Traditional

Epistemology

1.Bayesiansintroduceanewtechnicalterm,degreeofbelief,buttheystruggleto

explicateit.Tobesure,theliteratureisfullofnodstobettinginterpretations,but

these meet a fate similar to that of behaviorismindeed, a particularly localized

behaviorismthatfocusessolelyontheratherpeculiarkindofbehaviorthatismostly

foundatracetracksandcasinos.Othercharacterizationsofdegreeofbeliefthatfall

outofdecisiontheoreticrepresentationtheoremsarealsoproblematic.(SeeEriksson
11/23/08 14

and Hjek2007.)Belief,bycontrast,is sofamiliar tothefolkthatitneeds no

explication.

2.Recalltheabsenceofanynotionoftruthofanintermediatedegreeofbelief.Yet

truthistheveryaimofbelief.Itisusuallythoughttoconsistincorrespondencetothe

waythingsare.Moreover,wewantourmethodsforacquiringbeliefstobereliable,in

the sense of being truthconducive. What is the analogous aim, notion of

correspondence, and notion of reliability for the Bayesian? The terms of her

epistemology seem to lack the successgrammar of these italicized words. For

example,onecanassignveryhighprobabilitytotheperiodattheendofthissentence

beingthecreatoroftheuniversewithoutincurringanyBayesiansanction:onecando

sowhileassigningcorrespondinglylowprobabilitytotheperiodnotbeingthecreator,

andwhiledutifullyconditionalizingonalltheevidencethatcomesin.Traditional

epistemologyisnotsotolerant,andrightlynot.

3. Relatedly, the Bayesian does not answer the skeptic, but merely ignores him.

Bayesianismdoesntmakeskepticalpositionsgoaway;itmerelymakesthemharder

tostate.

4.TheBayesiansimilarlylacksanotionofjustificationortotheextentthatshe

hasone,itistoopermissive.AtleastonwhatwehavecalledBayesianorthodoxy,any

priorisasuitablestartingpointforaBayesianodyssey,yetmereconformitytothe

probabilitycalculusisscantjustification.

Now,theBayesianwillbequicktoanswerthisandthepreviousobjectionsina
11/23/08 15

singlestroke.Shewillappealtovariousconvergencetheorems.Forexample:

If observations are precise then the form and properties of the prior
distribution have negligible influence on the posterior distribution. From a
practicalpointofview,then,theuntrammeledsubjectivityofopinionceases
toapplyassoonasmuchdatabecomesavailable.Moregenerally,twopeople
withwidelydivergentprioropinionsbutreasonablyopenmindswillbeforced
intoarbitrarilycloseagreementaboutfutureobservationsbyasufficientamount
ofdata.(Edwardsetal.1963:201)

Call this convergence to intersubjective agreement; such agreement, moreover, is

often thought to be the mark of objectivity. The forcing here is a result of

conditionalizingthepeoplespriorsonthedata.GaifmanandSnir(1982)similarly

showthatforeachsuitablyopenmindedagent,thereisadatasetsufficientlyrichto

forceherarbitrarilyclosetoassigningprobability1tothetruememberofapartition

ofhypotheses.Callthis convergencetothetruth. TheBayesianmighteventryto

parlay these theorems into providing surrogates for that fourth condition for

knowledge, insisting that such convergences do not happen by luck, or for some

anomalousreason,butareprobabilisticallyguaranteed.

Thesearebeautifultheorems,butoneshouldnotoverstatetheirepistemological

significance. They are glass halffull theorems, but a simple reversal of the

quantifiersturnsthemintoglasshalfemptytheorems.Foreachdataset,thereisa

suitably openminded agent whose prior is sufficiently perverse to thwart such

convergence:afterconditionalizingherprioronthedataset,sheisstillnowherenear

assigningprobability1tothetruehypothesis,andstillnowherenearagreementwith

other people. And strong assumptions underlie the innocentsounding phrases

suitablyopenmindedagentandsufficientlyrichdataset.Nodataset,however

rich,willdriveanywhereatalladogmaticagentwhoconcentratesallcredenceona

singleworld(maximallyspecifichypothesis).Worse,anagentwithawackyenough
11/23/08 16

priorwillbedrivenawayfromthetruth.Considersomeonewhostartsbygivinglow

probabilitytobeingabraininavat,butwhosepriorregardsalltheevidencethatshe

actually gets as confirming that she is. And we can always come up with rival

hypotheses that no courses of evidence can discriminate betweenthink of the

irresolvableconflictbetweenanatheistandacreationistwhoseesGodshandiwork

ineverything.

5.AproponentofBayesianismmaydescribeitasfecund(andwedid);butan

opponentmaydescribeitasempty.Withsolittleconstraintonpriors,itisnot

surprisingthatBayesianismaccountsforsomuch.Indeeditisirrationaltodenythat

thesunwillrisetomorrow,toprojectgrueratherthangreeninourinductions,and

tocommitthegamblers fallacy.ThetroubleisthatBayesianismcondonesallsuch

inferentialpracticesforallarelicensedbysuitablyperversepriors.

6.ThetraditionalepistemologistmayprotestthatBayesiansdistancethemselvesfrom

the world. Recall our discussion of the relativization of Bayesian claims to the

subjectiveprobabilityfunctionsofagents.Ratherthanhookingupdirectlywiththe

world,thetermsoftheirepistemologyareallinternaltoprobabilisticmodelsofthe

world.Moreover,theBayesianapparentlydoesnothavemuchofastoryaboutwhat

makesamodelgood,oronemodelbetterthananother.Thisisrelatedtotheconcern

that the Bayesian does not do justice to truth, justification, and the antiluck

conditions.

WhatarewetomakeoftheseconflictingconsiderationsforandagainstBayesian

epistemology?Atthisstageofthedialectic,anygoodHegelianwillinsistthatits
11/23/08 17

timefora:

6.Synthesis

Shouldwereallypreferoneapproachtoepistemologyovertheother?Shouldone

ofthetwoapproachesbejettisoned?Wewillarguethatweshouldnotregardthemas

incompetition.Infact,thetwoapproachescomplementeachotherinbothsubject

matterandmethod.

Traditionalepistemologistssometimesstressthatphilosophydiffersfromscience

andinsistthatphilosophyhasitsowndistinctmethodofenquiry,namelyconceptual

analysis.Bayesians,ontheotherhand,typicallyconsidertheirworkmoreinlinewith

scientifictheorizing.ThisisreflectedinthemanyconnectionsBayesianepistemology

haswithBayesianstatistics,decisiontheoryandtheliteratureoncausaldiscoveryin

artificialintelligence.ItisalsoreflectedintheimportanceBayesiansgivetosolving

real problems. As we will see in the next section, Bayesianism is tremendously

successfulinthisrespect.Giventhesesuccesses,Bayesiansshouldhardlybeexpected

to give up their framework just because it is not a complete panacea for all

epistemological ills. By way of analogy, scientists rightly held onto Newtonian

mechanicseveninthefaceofsometheoreticalproblems(forexample,itscommitment

toactionatadistance).AnddespiteitsbeingsupersededbyRelativityTheory,we

understandpreciselyhowNewtonianmechanicsisstillapproximatelytrue.Webelieve

thatthesuccessorofBayesianismwillstandinasimilarrelationtoBayesianismas

RelativityTheorystandstoNewtonianmechanics.

SoletusrevisitthechargesleveledatBayesianismintheprevioussection,keeping

inmindthisviewofitasaworkinprogressthatnonethelessisclearlyearningits
11/23/08 18

theoretical keeplike any good scientific theory. We may happily take its

fundamentalconcept,degreeofbeliefasaprimitiveintheabsenceofasuccessful

analysisofit.Ithasearneditstheoreticalkeepbyitscontributiontoavirtuoustotal

theorylike anyprimitive scientific concept. Andit can live peacefully alongside

traditional epistemologys primitive concept of belief, without any expectation of

reductionofonetotheother.Indeed,theprospectsforsuchreductionstrikeusas

unpromising.(SeeErikssonandHjek2007.)

AgainstthechargeofBayesianismbeingempty,itcanpleadthegoodcompanyof

deductivelogic.Tobesure,crazysetsofbeliefcanbeconsistent,andinferences

fromabsurdpremisescanbevalidtheslogangarbagein,garbageoutisastruein

epistemologyasitisincomputerscience.Bayesianism,likelogic,cannevertheless

playasalutaryroleinkeepingourdegreesofbelief,likeourbeliefs,inharmony,and

inpolicingourelicitinferences.Afterall,deductivelogicisneverregardedasa

completesetofconstraintsonbelief;similarly,theBayesianconstraintsondegrees

ofbeliefshouldnotberegardedascomplete.Someadditionalconstraintsmaywell

findtheirinspirationintraditionalepistemology.

AgainstthechargeofBayesianismsverdictsbeingmodelrelative,itcanplead

thegoodcompanyofscience.Afterall,ourbestmethodsofenquiryinthephysical

andsocialsciencesworklikethis.Arguablyweshouldnotexpectepistemologytobe

different.

NorneedansweringskepticalchallengesbepartofBayesianismsjobdescription,

just as it is not part of traditional epistemologys job description to underwrite

rationaldecisionmaking,confirmation,andtheuseofprobabilisticandstatistical

methodsinthesciences.Thereisnoharmintheirlabourbeingdivided.Theyaretwo
11/23/08 19

differentwaystoapproachepistemologyandtheyoftenanswerdifferentquestions.

Andwheretheirquestionsareshared,theirapproaches canbecomplementary

ratherthanmutuallyexclusive.WeseethatBayesianepistemologyhelpstoaddress

somequestionsthatwefindintraditionalepistemological debates.TheBayesian

treatmentofissuessuchastestimonyandthecoherencetheoryofjustification(see

section7)arecasesinpoint.ItturnsoutthattheformalmachineryofBayesianismis

wellsuitedtomakecertainquestionsmorepreciseandtoprovideanswerswhenour

intuitionsdontgiveclearverdicts.

NowletusseewhatonecandowiththemachineryofBayesianism.Itshouldbe

assessed by the problems it solves and how much it unifies, for example, the

methodologyofscience.

7.AchievementsofBayesianEpistemology

Many of the Bayesian success stories are from confirmation theory. But

Bayesianismhasmuchmoretoofferasitsdomainofapplicabilityalsoincludesother

partsofepistemologyandphilosophyofscience.Herearefivehighlights.

1.ConfirmationTheory.Aswesawinsection2,Bayesiansbeginwiththeideathat

confirmation is a matter of probabilityraising. They then show how important

intuitionsaboutconfirmationcanbevindicated.Supposethathentailse,sothatP(e|

h)=1.ThentheposteriorprobabilityofhisP(h|e)=P(h)/P(e).Hence,forafixed

priorprobabilityofh,theposteriorprobabilityofhincreasesifP(e)decreases.From

thiswecanimmediatelyaccountforthemethodologicalinsightthatmoresurprising

evidence confirms better. Similarly, Bayesians have provided a rationale for the

varietyofevidence thesisthe more varied the evidence is, the betterand have
11/23/08 20

providedilluminatingdiscussionsoftheDuhemQuinethesis(Earman1996,ch.3).

Toaddresstheseissues,severalmodelassumptionshavetobemade.Inthecaseof

thevarietyofevidencethesis,forexample,morevariedhastobeexplicatedin

probabilisticterms.Thiscanbedoneindifferentways.ManyBayesiandiscussionsof

thevarietyofevidencethesisassumethattheevidenceiscertain.Butwesawalready

thatthisisnotalwaysthecase.ItspeaksinfavoroftheBayesianframeworkthatit

provides the tools to model more complicated testing scenarios. See Bovens and

Hartmann(2003)formorerealisticBayesianmodelsofthevarietyofevidencethesis

andtheDuhemQuinethesis.

Bayesianconfirmationtheoryconnectsnaturallywithempiricalpsychologyasa

wealthofworkinthepsychologyofreasoningunderuncertaintydemonstrates.See,

for example, Chater and Oaksford (2008) and Oaksford and Chater (2007) for

sophisticatedBayesianmodelsthataccountforempiricalfindings.Crupietal.(2008)

show how the presence of the conjunction fallacy (i.e. that experimental subjects

assignahigherprobabilitytoaconjunctionthantooneoftheconjuncts),asfamously

demonstratedinpsychologicalexperimentsbyTverskyandKahneman(1983),canbe

explainedinconfirmationtheoreticalterms.Crupietal.(2007)argueonnormative

andexperimentalgroundsforaspecificmeasureofevidentialsupport(thesocalled

Zmeasure). And Bayesian confirmation theory provides a flexible framework to

rationallyreconstructspecificepisodesfromthehistoryofscience.Whiletraditional

epistemology does not have the resources to study such episodes, the Bayesian

frameworkisideallysuitedforthesepurposes(Franklin1990).Itisalsobettersuited

thansomesystemofepistemicordoxasticlogicimaginetryingtoilluminatesome

scientificepisodesolelywithKandBoperators!
11/23/08 21

2.DynamicsofBelief.Traditionalepistemology,withitsfocusontheanalysisof

knowledge,isrelatively silentaboutquestions ofbeliefdynamics.Ifthereis talk

aboutbeliefchange,itisgenerallyassumedthatittakesplaceonthebasisoflearned

evidencethatiscertain.Traditionalepistemologysharesthisassumptionwithlogical

theories of belief revision such as the AGM theory (Grdenfors and Rott 1995).

However, Jeffrey taught us that learning often does not come in the form of

certainties.Toaddressthesecasesoflearningandbeliefchange,philosophersaswell

asresearchersinartificialintelligencehaveformulatednewupdatingrules(suchas

Jeffrey conditionalization) and developed powerful tools such as the theory of

Bayesiannetworks(Neapolitan2003).

3. Applications. Bayesianism has a symbiotic relationship with causation and

powerfulalgorithmshavebeendevelopedtolearncausalrelationsfromprobabilistic

data(KorbandNicholson2004,Pearl2000,Spirtesetal.2001).Thesealgorithmsuse

thetheoryofBayesiannetworks.ABayesiannetworkorganizesasetofvariablesinto

a Directed Acyclic Graph (DAG).ADAGis asetofnodesandasetofarrows

betweensomeofthenodes.Theonlyconstraintisthattherearenoclosedpaths

formedbyfollowingthearrows.A rootnode isanodewithoutgoingarrowsonly,

anda parent ofagivennodeisanodefromwhichanarrowpointsintothegiven

node.Eachnoderepresentsapropositionalvariable,whichcantakeanynumberof

mutuallyexclusiveandexhaustivevalues.TomakeaDAGintoaBayesiannetwork,

onemorestepisrequired:weneedtospecifythepriorprobabilitiesforthevariables

intherootnodesandtheconditionalprobabilitiesforthevariablesinallothernodes,
11/23/08 22

givenanycombinationofvaluesofthevariablesintheirrespectiveparentnodes.The

arrows in a Bayesian network carry information about the independence relations

betweenthevariablesinthenetwork.ThisinformationisexpressedbytheParental

MarkovCondition: A variablerepresented byanodein theBayesian network is

independentofallvariablesrepresentedbyitsnondescendentnodes,conditionalon

allvariablesrepresentedbyitsparentnodes. Inthecausalmodelingliterature,this

conditioniscalledtheCausalMarkovCondition.

4.TheCoherenceTheoryofJustification.ConfrontedwiththeCartesianskeptic,

coherentists pointoutthatwhenourbeliefsystemshangtogetherwell,withtheir

differentpartssupportingeachother,thenthisisanindicationofthetruthofthe

systems(BonJour1985).However,thecorrespondingtheorythecoherencetheory

of justificationsuffers from several problems. Here are two. First, the theory is

vague,asitisdifficulttomakeprecisewhatcoherenceis.Second,coherenceisnot

necessarilytruthconducive.Forexample,fairytalesaremadeup,althoughthestories

theytellmaybehighlycoherent.Hence,thecoherenceofasetofpropositionsisat

besttruthconduciveceterisparibus.Butwhatgoesintheceterisparibusclause?This

questionishardtoaddressifweonlyhavethetoolboxoftraditionalepistemology.

Bayesians can be of real help here. They have proposed and analyzed various

measures of coherence and analyzed in detail under which conditions, if at all,

coherenceistruthconducive(BovensandHartmann2003,DouvenandMeijs2007,

Olsson2009).

5. Sources of Knowledge/Belief. Traditional epistemology examines sources of

11/23/08 23

knowledge and belief such as our senses, memory and testimony. All three have

inspired Bayesian modelbuilding. First, the uncertainty of the evidence from our

senseshaspromptedthedevelopment ofamorerealistic updatingrulethanstrict

conditionalizationJeffreyconditionalization.Second,conditionalizationrepresents

anidealizedversionoftheepistemologicalroleofmemoryonewhoupdatesonly

byconditionalizationneverforgetswhileBayesianmodelsofboundedrationality

allowformemoryloss(Mehtaetal.2004).Third,Bayesianshavetheresourcesto

model the effect of combining the testimony of several witnesses (Bovens and

Hartmann 2003). There is also a growing literature on selfknowledge and self

location,asexemplifiedbytheSleepingBeautyproblem(Elga2000),andwhether

centred information can rationally induce changes inopinions about uncentred

propositionsconcerninghowtheworldis.

8.AvenuesforFutureResearch

Inthisclosingsection,webrieflypointtosometopicsthatwewouldliketosee

addressedinfutureresearch.WhilesomeofthemconcerntherelationofBayesian

epistemology to traditional epistemology and philosophy of science, others are

internaltotheBayesianprogram.

1. More bridges between Bayesian epistemology and traditional epistemology.

Thereisnoharminlabourbeingdividedbetweenthetwokindsofepistemology,as

wehavearguedbutitwouldbeallthebetteriftheycouldbecomemorecooperative

enterprises.Thinkofsomeofthetimehonoreddebatesintraditionalepistemology:

skepticism,theanalysisofknowledge,reliabilism,internalismvs.externalism.Think

of some of the currently hot topics: contextualism, subjectsensitive invariantism,

11/23/08 24

contrastivism,relativism,luminosity,knowledgehow(asopposedtoknowledge

that),knowledgewh_(who,where,when,which),Wherearethecounterpart

debatesinBayesianepistemology?Goingintheotherdirection,thinkofsomeofthe

timehonoreddebatesinBayesianepistemology:constraintsonpriors,updatingrules,

the extension of subjective probabilities to infinite spaces. And think of some

currentlyhottopics:credencesaboutchances(asinthePrincipalPrinciple),credences

aboutonesfuturecredences(asintheReflectionPrinciple),updatingcredenceson

centred or indexical propositions, ... Where are the counterpart debates in

traditionalepistemology?Eachofthesetopicssuggestsabridgewaitingtobebuilt.

Tosomeextent,suchprogressawaitsabetterunderstandingoftherelationship

betweentraditionalclaimsaboutbelief/knowledgeandBayesianclaimsaboutdegrees

ofbelief,whichisstillcontroversial.Aswehavesaid,wearenotsanguineaboutthe

prospectsofareductionineitherdirection,althoughreductionissurelynottheonly

waytoofferillumination.Andevenifthetwoepistemologiescontinueonseparate

tracks,stilldevelopmentsinonecanprovideinspirationorheuristicguidanceforthe

other.

2. More bridges between Bayesian epistemology and philosophy of science.

Bayesianismstartedasaconfirmationtheory.Andindeed,theformalmachineryto

addressconfirmationtheoreticalquestionsishighlydeveloped.Ourambitiousgoal,

however,shouldbetodevelopafullfledgedBayesianphilosophyofscience.Hereis

anincompletelistofquestionsthatshouldbeaddressedtothisend:

(i)WhichstancedoesBayesianismtakeintherealismdebate?Isitneutraltothe

debate,ordoesitfavoraversionofscientificrealismorantirealism?(Douven2005,
11/23/08 25

Earman1996)

(ii)CanscientifictheorychangebeunderstoodinBayesianterms?(Earman1996)

(iii) Can Bayesianism help to characterize the overall structure of science? Is it

epistemicallyadvantageoustoaimforunifiedtheories?(Myrvold2003)

(iv)CanaBayesianreadingofInferencetotheBestExplanationbegiven?Lipton

(2004)arguesthatexplanatoryconsiderationsareencodedinthelikelihoods,andnot

inthepriors.

(v)HowcanscientificidealizationsbeunderstoodinBayesianterms?Addressingthis

questionisimportantasidealizationsareubiquitousinscience.Thetroublebegins

whenweattachapriorprobabilityofzerotoanidealized(hencefalse)statement.The

posterioristhenarguablyalsozero,whicharguablyrendersBayesianismuseless.

3.BayesianSocialEpistemology.Bayesianepistemology,aswehavepresenteditso

far,sharesoneimportantfeaturewithtraditionalepistemology:itisindividualistic,

i.e.itisconcernedwithoneagent,whohasbeliefsandwhoupdatesherbeliefsinthe

light of new evidence. However, the doxastic unit could well be a community

comprising several individuals, or more. Kuhn (1962) argued that it is the entire

scientificcommunitythatacceptsorrejectsaparadigm.Orthinkofajurythathasto

comeupwithaconsensualverdictinamurdercase.Inlightofexamplessuchas

these, a new fieldsocial epistemologyhas been established (Goldman 1999,

Kitcher1993).Whilemuchworkinthisfieldisinformal,formaltoolshaverecently

beendevelopedthataddressissuesinsocialepistemology.Especiallynoteworthyis

the work on judgment aggregation (List and Puppe 2009) that comprises

investigationsinspiredbythediscursivedilemma.BovensandRabinowicz(2006)
11/23/08 26

and Hartmann and Sprenger (2009) have given epistemic analyses of various

aggregationrulesstudiedinthiscontext.Itishopedthatthisworkwilleventually

developintoafullBayesianaccountofgroupjudgmentandgroupdecisionmaking.

Other topics of current interest include the debate about rational disagreement

(Feldman and Warfield 2009) and, related to this, theories about consensus and

compromiseformation.

Tosumup:Bayesianepistemologyisanexcitingandthrivingresearchprogram.

TheresplentymoreworkforBayesianstodo!1

References

Armendt,B.(1980).IsThereaDutchBookArgumentforProbabilityKinematics?

PhilosophyofScience47:583588.

BonJour,L.(1985).TheStructureofEmpiricalKnowledge.Cambridge,MA:Harvard

UniversityPress.

Bovens, L. and S. Hartmann (2003). Bayesian Epistemology. Oxford: Oxford

UniversityPress.

Bovens,L.andW.Rabinowicz(2006).DemocraticAnswerstoComplexQuestions:

AnEpistemicPerspective.Synthese150:131153.

Chater, N.andM.Oaksford(eds.)(2008). TheProbabilistic Mind:Prospectsfor

BayesianCognitiveScience.Oxford:OxfordUniversityPress.

Crupi,V.,K.TentoriandM.Gonzalez(2007).OnBayesianMeasuresofEvidential

Support: Theoretical and Empirical Issues. Philosophy of Science 74(2):

1
WethankJohnCusbert,KevinKorb,RalphMiles,JonahSchupbach,Declan
SmithiesandJanSprengerforveryhelpfulcomments.
11/23/08 27

229252.

Crupi, V., B. Fitelson and K. Tentori (2008). Probability, Confirmation and the

ConjunctionFallacy.ThinkingandReasoning14(2):182199.

Diaconis,P.andS.L.Zabell(1982).UpdatingSubjectiveProbability.Journalofthe

AmericanStatisticalAssociation77(380):822830.

Douven, I. (2005). Wouldnt it Be Lovely: Explanation and Scientific Realism,

Metascience14:338343.

Douven,I.andW.Meijs(2007).MeasuringCoherence.Synthese156:405425.

Earman,J.(1996).BayesorBust?ACriticalExaminationofBayesianConfirmation

Theory.Cambridge,Mass.:MITPress.

Edwards,W.,Lindman,H.,andSavage,J.(1963).BayesianStatisticalInferencefor

PsychologicalResearch.PsychologicalReview70:193242.

Eells,E.andB.Fitelson(2000).MeasuringConfirmationandEvidence. Journalof

Philosophy97:663672.

Elga, A. (2000). SelfLocating Belief and the Sleeping Beauty Problem. Analysis

60(2):14347.

Eriksson,L.andA.Hjek(2007).WhatAreDegreesofBelief? StudiaLogica 86,

SpecialIssueFormalEpistemologyI,B.Fitelson(ed.),183213.

Feldman,R.andT.Warfield(eds.)(2009).Disagreement.Oxford:OxfordUniversity

Press.

Fitelson, B. (1999). The Plurality of Bayesian Measures of Confirmation and the

Problem of Measure Sensitivity. Philosophy of Science 66 (Proceedings):

S36278.

Franklin,A.(1990).Experiment,RightorWrong.Cambridge:CambridgeUniversity
11/23/08 28

Press.

Gaifman, H. and M. Snir (1982). Probabilities over Rich Languages. Journal of

SymbolicLogic47:495548.

Grdenfors,P.andH.Rott(1995).BeliefRevision,inHandbookofLogicinArtificial

IntelligenceandLogicProgramming.VolumeIV:EpistemicandTemporal

Reasoning, eds. D.M. Gabbay, C.J. Hogger und J.A. Robinson. Oxford:

OxfordUniversityPress,35132.

Glymour,C.(1980).TheoryandEvidence.Princeton:PrincetonUniversityPress.

Goldman,A.(1999).KnowledgeinaSocialWorld.Oxford:ClarendonPress.

Good,I.J.(1971).LettertoTheAmericanStatistician25:6263.

Hjek,A.andN.Hall(2002).InductionandProbability.InTheBlackwellGuideto

thePhilosophyofScience, P.MachamerandM.Silberstein(eds.).Oxford:

Blackwell,149172.

Hartmann, S. and J. Sprenger (2009). Judgment Aggregation and the Problem of

TrackingtheTruth.Synthese.

Howson,C.andP.Urbach(2007).ScientificReasoning:TheBayesianApproach
.

Jaynes,E.(2003).ProbabilityTheory:TheLogicofScience
.Cambridge:Cambridge

UniversityPress.

Jeffrey,R.(1983). TheLogicofDecision.2nded.Chicago:UniversityofChicago

Press.

Jeffrey, R.(1992). Probable Knowledge. In Probability andthe Art of Judgment,

Cambridge:CambridgeUniversityPress.

Joyce,J.(1998).ANonpragmaticVindicationofProbabilism.PhilosophyofScience

65:575603.
11/23/08 29

Joyce,J.(1999).TheFoundationsofCausalDecisionTheory.Cambridge:Cambridge

UniversityPress.

Joyce,J.(2008).Bayes'Theorem. TheStanfordEncyclopediaofPhilosophy(Fall

2008Edition),E.Zalta(ed.).

Kitcher,P.(1993).TheAdvancementofScience.SciencewithoutLegend,Objectivity

withoutIllusions.NewYork:OxfordUniversityPress.

Kolmogorov,A.N.(1933).GrundbegriffederWahrscheinlichkeitsrechnung.Berlin:

Springer.Englishtranslation:FoundationsoftheTheoryofProbability.New

York:Chelsea1950.

Korb,K.andA.Nicholson(2004).BayesianArtificialIntelligence.BocaRaton:CRC

Press.

Kuhn, T. (1962). The Structure of Scientific Revolutions. Chicago: University of

ChicagoPress.

Lewis,D.(1980).ASubjectivist'sGuidetoObjectiveChance.InR.Jeffrey(ed.),

StudiesinInductiveLogicandProbability,VolII.,Berkeley:Universityof

CaliforniaPress,263293.

Lewis,D.(1999).WhyConditionalize?InPapersinMetaphysicsandEpistemology.

Cambridge:CambridgeUniversityPress.

Lipton,P.(2004).InferencetotheBestExplanation.London:Routledge.

List,C.andC.Puppe(2009). JudgmentAggregation:ASurvey.In:PAnand,C.

Puppe and P. Pattanaik (eds.), Oxford Handbook of Rational and Social

Choice.Oxford:OxfordUniversityPress.

Mehta,N.,S.RajivandK.Srinivasan(2004).RoleofForgettinginMemoryBased

Choice Decisions: A Structural Model. Quantitative Marketing and

11/23/08 30

Economics2(2),107140.

Myrvold,W.(2003).ABayesianAccountoftheVirtueofUnification.Philosophyof

Science70:399423.

Neapolitan,R.(2003).LearningBayesianNetworks.PrenticeHall.

Nozick,R.(1981).PhilosophicalExplanations.Cambridge,MA:HarvardUniversity

Press.

Oaksford, M. and N. Chater (2007). Bayesian Rationality: The Probabilistic

ApproachtoHumanReasoning.Oxford:OxfordUniversityPress.

Olsson,E.(2009).AgainstCoherence:Truth,Probability,andJustification.Oxford:

OxfordUniversityPress.

Pearl,J.(2000).Causality.NewYork:CambridgeUniversityPress.

Ramsey,F.P.(1931).TruthandProbability.InR.B.Braithwaite(ed.),Foundations

ofMathematicsandotherEssays.London:Routledge,15698;reprintedinD.

H.Mellor(ed.),F.P.Ramsey:PhilosophicalPapers,Cambridge:Cambridge

UniversityPress1990.

Savage,L.J.(1954).TheFoundationsofStatistics.JohnWiley.

Shogenji, T. (2009). The Degree of Epistemic Justification and the Conjunction

Fallacy.Synthese.

Sosa,E.(1999).HowtoDefeatOppositiontoMoore.PhilosophicalPerspectives13:

141153.

Spirtes,P.,C.Glymour,andR.Scheines(2001).Causation,PredictionandSearch.

Cambridge,Mass.:MITPress.

Tversky,A.andD.Kahneman(1983).ExtensionalversusIntuitiveReasoning:The

Conjunction Fallacy in Probability Judgment. Psychological Review 90:

11/23/08 31

293315.

vanFraassen,B.(1984).BeliefandtheWill.JournalofPhilosophy81:235256.

Williamson, J. (2005). Bayesian Nets and Causality: Philosophical and

ComputationalFoundations.Oxford:OxfordUniversityPress.

Zynda,L.(1995).OldEvidenceandNewTheories.PhilosophicalStudies77:6795.