The Transparency of Mental Content

7/30/2019 The Transparency of Mental Content

http://slidepdf.com/reader/full/the-transparency-of-mental-content 1/19

Ridgeview Publishing Company

The Transparency of Mental ContentAuthor(s): Paul A. BoghossianSource: Philosophical Perspectives, Vol. 8, Logic and Language (1994), pp. 33-50Published by: Ridgeview Publishing Company

Stable URL: http://www.jstor.org/stable/2214162 .

Accessed: 01/09/2013 20:56

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

.

Ridgeview Publishing Company is collaborating with JSTOR to digitize, preserve and extend access to

Philosophical Perspectives.

http://www.jstor.org

This content downloaded from 168.176.5.118 on Sun, 1 Sep 2013 20:56:23 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=rpc

http://www.jstor.org/stable/2214162?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp





http://www.jstor.org/stable/2214162?origin=JSTOR-pdf

http://www.jstor.org/action/showPublisher?publisherCode=rpc



Philosophical erspectives,, LogicandLanguage, 994

THE TRANSPARENCYOF MENTAL CONTENT1

Paul A. BoghossianNew York University

Introduction

MichaelDummettncewrote:

It is an undeniable feature of the notion of meaning-obscure as that notion

is-that meaning is transparent in the sense that, if someone attachesameaning to each of two words, he must know whether these meanings are thesame.2

As the surroundingiscussionmakesclear,Dummett's laimisn't merely hefairlyweak thesisthat, oranytwoexpressionsn a person's epertoire,t mustbe possible or thatperson o come to knowwhether r nottheymean he same.Rather, he claimis that t mustbe possiblefor thatperson o come to knowsuch a fact purely introspectively, without the benefit of further empirical

investigation--apriori, as I shall also occasionallyput it.

Dummett doesn't say much about the basis for his conviction that'transparency'n this sense-epistemic transparencyas I shall sometimescall it,

to distinguish it from the referential variety-is "an undeniablefeature of the

notionof meaning."He gives the impression f finding he claimtoo obviousto need arguing. nthis respect, believe,he followsin thefootstepsof FregeandRussell,bothof whomgave epistemic ransparencypivotal, f unargued,role ntheir espectiveheories f linguistic ndmental ontent.

In anotherpaper,I hope to discuss this historicalpoint. In this paper,however,my concernwill be purelyphilosophical nd, then,only with mentalcontent. I believe that the notion of epistemic transparency oes play an

importantole in our ordinary onception f mental ontentandI want o saywhatthatrole is. Unfortunately,he task s a largeone;hereI amableonly tobeginon its outline.

I shall proceedsomewhat ndirectly,beginningwith a discussionof







34 /Paul A. Boghossian

externalistconceptions f mentalcontent. shallshow thatsuch conceptions

violateepistemic ransparencyo an extent hathasnot been fully appreciated.Subsequently, shall look at the implicationsof this violation and at thereconstructive roject hata rejection f transparencyntails.I am inclined othink it unlikelythat we will get a conceptionof propositional ontentthatunderwritespistemic ransparency.utI am concernedhatwe havenot fullyappreciatedhe role that ransparencyurrently laysand the workthatwouldneed o be donewereweto discardt.

Externalism, Self-Knowledge and Transparency

The minimal ntended ignificance f the TwinEarth hought xperimentsmay be put like this:Eithermentalcontentsarenot individuatedndividual-istically,or theyarenotindividuatedn termsof their ruth onditions.

Thus,considerOscarand twin Oscar; hey arealike in all intrinsic,non-relational espects.ButwhereasOscar rowsupin a physical nvironment hichcontainsH20, twin Oscargrows up in an environmentwhichcontainsXYZ.And widespreadntuition ppears o have it thatwhereasOscar'sutterance f'Water s wet' expressesa thought hat s true f and only if H20 is wet,twinOscar'sutterance f thatsentence xpresses thought hat s true f andonlyif

XYZis wet. Hence,either he contents f the twothoughts renot individuatedindividualisticallyr theyarenot individuatednterms f their ruth onditions.3

Thus,consider lsoBertand twin Bert; heyarealikein all intrinsic, on-relational espects.ButwhereasBertgrowsup in,anddefers o, a communitynwhich heword arthritis's used o applyonlyto a certain indof inflammationof the joints, twin Bertgrows up in, and defers to, a communityn which'arthritis's usedto apply o all rheumatoidilments, ven to thosethatmightoccur n the thigh.Andwidespreadntuition as t thatwhereasBert'sutteranceof 'I havearthritisnmy thigh'expresses thoughthat s true f andonlyif he

has arthritisn his thigh and,hence,canneverbetrue), winBert'sutterancefthatsentence xpressesa thoughthat s true f andonlyif he has(someother

disease) harthritisnhisthigh and,hence,mayoccasionally etrue).Asbefore,the moralwouldappearo be: Either hemental ontents f the twothoughts renot individuatedndividualisticallyr theyarenotindividuatedn terms f theirtruth conditions.4 Let us call the formerview externalism,and the latterindividualism.

According o externalism,hen,the psychologicallyelevant ontentsof asubject'sntentionaltatesare ndividuatednpartbycertain ortsof factsaboutthephysicaland/or ocialenvironmentnwhichhehappenso be situated.5 nd

although his wouldappear o underminehe claim thatwe can havepurelyintrospective,non-empirical ccess to the contentsof our thoughts,we areassured yanincreasing umberf philosophershat hisappearances illusory:







TheTransparencyf MentalContent 35

there s in fact no conflictbetweenan extemalist onception f mental ontent

and ntrospectiveelf-knowledge. want obeginwiththe question,Onwhat sthisconfidence ased?

It is basedon a thoughthat s succinctly xpressed y Davidson:

Showing that there is no conflict [between externalismand knowledge ofcontent] is basically simple. It depends on realizing that whatever isresponsible for the contents of our thoughts, whether known or not, is alsoresponsible for the content of the thought that we have the thought.6

Howexactly houldweunderstandavidson'shought ere?Let's assume,without oss of generality, hat we think n a languageof

thought(in fact, I'll pretend hat we thinkin English. This will make theexposition asierbut s notrequiredor anythinghat's o follow:similar laimscouldhavebeen madewithout he assumptionhatpsychological tatespossesssyntactic tructure.)Withthe convenient ssumptionn place,we maysaythatPeterbelieves hatp justin case Peterhas a tokenof a sentenceS whichmeans

thatp, inhisbeliefbox;andPeterdesires hatq just ncase Peterhasa tokenofa sentenceQ, which meansthat q, in his desirebox; andso on for the otherattitudes.

It is now easy to give a completelygeneralelaboration f Davidson'sthought. orsinceextemalisms guaranteedobe entirelyneutral n thematter,

we may safely assume hat hereare syntacticmechanismswhich ensure hatwhen,andonly when,a tokenof thesentenceS entersPeter'sbeliefbox, then

ceterisparibus,o doesa tokenof the sentence I believeS'. Andso, providedthat heembedded ndunembeddedokensof S mayalwaysbecountedupon oexpress he samecontent,Peter's econd-ordereliefwillalwaysreport orrectlyon the content of his first-orderbelief, the externalistnatureof contentindividuationotwithstanding.ence,and n thissense,extemalisms shown obe consistentwithauthoritativeirst-personnowledge f thought ontent.Thisis, I think, the best and simplest way to develop the idea that Davidson

expresses.7

Now, thereare,I think,a number f questions hatmightbe raisedaboutthisproposal.Forinstance: an we in fact alwayscounton the embedded ndunembeddedccurrencesf S to express hesamecontent?Asweshallpresentlysee, theremaybe specialreasons ordoubtinghereliability f this covariationunderextemalistassumptions.)Also: theenvisagedproposalwouldappearatbest to ensure heavailability f reliablebeliefsaboutcontent,notknowledge;and herearesometrickyquestions bout ustificationhatneedanswering. utIam not going to pursueeitherof these lines of questioninghere;for presentpurposes, wish simplyto grant heclaim thatextemalisms consistentwith

the sortof knowledgeof contentenvisagedby this elaboration f Davidson'sidea. My point will ratherbe that the kind of self-knowledge hat s therebysecured alls short way short-of thekind of self-knowledgewe normallythinkof ourselves spossessing.







36/ PaulA.Boghossian

Let usgrant heclaimthat, oranysentenceS inmybeliefbox, I canform

a correctecond-ordereliefabout tscontent yprefixinghewords Ibelieve...'to that very sentence. It is important o observe, however, that no otherepistemiccapacityis therebygrounded.For instance,supposeI have twosentencesP andQ in my belief box thatareidentical in content; t needn'tfollowthatI willknow,orbe able to a priori cometo know,that heyare,andthis despite my being said to know what the contents of P and Q areindividually. or,similarly, oes it followthat f I havetwosentencesQ andRin my belief box thatare distinct n content, hatI will know,or be ableto aprioricome toknow, hat heyare;and hisdespitemy beingsaidtoknowwhatthecontents f QandR are ndividually.nfact, tdoesn'tevenfollowfrom hesortof knowledge hat s underwritteny theproposal,hatI will know,or beable to aprioricome toknow,whether wodistinct okensof thesane sentencetype S expressthe same or distinctcontents.To put thepointconcisely,theproposal oesnothingoreconstructheintuitivedeathatwe know he contentsof ourthoughtsn the sense that we can introspectively iscriminateetweenthem, hatwe cantell themapartromeachother: t doesn'tbyitselfunderwritetheepistemicransparencyf mental ontent.

Externalism, Transparency and the Semantics of Travel

Thethesisof theepistemic ransparencyf contentmaybe usefullybrokenup into two parts: a) If two of a thinker's okenthoughtspossess the samecontent, henthethinkermustbe able to know apriori hat heydo;and(b)Iftwo of a thinker's okenthoughtspossess distinctcontents, henthe thinkermust be able to know apriori thatthey do. Call the first the thesis of thetransparency of sameness and the second the thesis of the transparencyof

difference.

Now, it is fairlyeasy to showthatextemalistcontentsdo not satisfythe

transparencyf sameness.Kripke's otorious renchman,ierre, lready howsthis forthespecialcase of Milliancontents themselves, f course,a speciesofextemalist ontent).8 hedetailsarewell-enough nown hat neednotrehearsethemhere.A plausiblestoryculminatesn Pierrehavingtwo expressions'Londres'nd London'which referto the samecity without,however,hisrealizinghat heydo.Thethought eexpresseswith

Londons pretty

is thesameas thethought eexpresseswith

Londress pretty;butthis fact s unavailableo introspectionlone.

Weneeda differentortof case,however,oillustratehegeneral laim hat







The Transparency f MentalContent 37

even a non-Millian, referentiallyopaque contentwould fail the transparency f

sameness, f it wereindividueatedxtemalistically.BrianLoarhas describeda casewhich may be adapted o just this purpose.9

Paul, an English speaker,has been raisedby a Frenchnannyin a sheltered

way. She speaks English with Paul, but amuses herself by referringto the cats

around them as "chats" and never as "cats." Paul, satisfying the usual

competence and deference constraintson concept possession, is to be credited

with beliefs involving the concept cat. In particular,he belief he would express

with "All chats have tails"would be the belief thatall cats have tails.

Now, Paul occasionally sees his parents, who speak of animals called

"cats."Because on those occasions, no cats or picturesof cats are ever present,

Paul does not realize that cats are his familiar"chats."However, Paul's parents

tell him quite a lot about cats and in particular hatall cats have tails. On this

basis, it is again true of Paul that the belief he would express with "All cats

have tails" would be thebelief thatall cats have tails.10

Intuitively, however, it is quite clear that Paul will not be able to tell a

priori that the belief he expresses with "All chats have tails"is the same belief

thathe expresses with "All cats have tails."

So much, then, by way of illustrating externalism's violation of the

transparency f sameness.What has not been generallyappreciated,however, is

thatextemalism also allows for a coherentdescriptionof the converse case: twotokens of a thinker's language of thoughtbelonging to the same syntactictype

have distinct meanings,but the thinker s in principlenot in a position to know

a priori thattheydo. Theseconverse cases, illustrating xtemalism's violationof

the transparencyof difference, are, I believe, importanteven for issues beyond

the ones presentlyat hand.1I

To see one way in which such converse cases might arise we need to

think -harder perhaps han s commonin the literatureabout the semanticsof

so-called 'one-worldtravelingcases'. SupposethatEarthand Twin Eartharepart

of the actual world and that Peter, a normal, competent adult Earthling, is

suddenlyand unwittinglytransportedo Twin Earth.An accidentaltourist,Peter

goes to sleep one night at home and wakes up in twin home in twin bed. He

suffers no discernible disruptionin the continuity of his mental life. Here on

Twin Earth,Peter happily lives out the rest of his days, never discovering the

relocation that he has been forced to undergo.How should we thinkabout the

semanticsof Peter's thoughts?

Well, one intuition that is sharedby practicallyeveryone who has thought

about these cases is that, after a while (how long is unclear), some tokens of,

say, 'water' in Peter's languageof thoughtwill shift from meaningwater and

will come to mean twater. Thus, to quote just one example, Tyler Burgewrites:

Tlhethoughtswould not switch as one is [quickly] switched from one actualsituation to another twin actual situation.The thoughts would switch only if







38 / PaulA. Boghossian

one remained long enough in the other situation to establish environmentalrelations necessary for new thoughts.So quick sWvitching ould not be a case

in which thoughts switched...But slow switching could be such a case.12

Viewed from an extemalist perspective, it's hard to deny the force of the

intuition that Burge is expressinghere. We imagine Peter moving to Twin Earth

and staying there for a while. Over time he thinks to himself a variety of

thoughtsthat he expresses with sentenceslike:

I won't go swimmingtoday;the water is too cold.

Or:

There's too much waterin this whiskey.

It seems very plausible that these expressions of beliefs about one's present

environment, expressions of current desires and current intentions, are

expressions of thoughtsthat involve the concepts twater, not water.

So it seems right to say that some tokens of 'water' will shift from

meaningwater to meaning twater. But it seems to me equally compelling to

say that certainother tokens of 'water'won't shift. Specifically, it seems to me

that those tokensof 'water'occurring n memories,andin beliefs about thepast

based upon them, will retain theirEarthly nterpretations, espite being tokened

on Twin Earth. Such thoughts,unlike, for instance,beliefs with undatedgeneral

contents,or thoughtsabout one's presentsurroundings, recausedand sustained

by previous perceptions long gone. In the normal case, they owe little, if

anything, to current perceptions and cognitive transactions with one's

environment. From a purely intuitive standpoint, they would be expected to

retain theirEarthlyinterpretations,despite the admitted shift in their syntactic

cousins. After all, why should mere travel from one place to anotherobliterate

one's capacity to have certainmemories. Normal tourismdoesn't have such an

effect; why should travelto anotherplanet with somewhatunusualgeographical

properties?Consider an example. While still on Earth, Peter goes hiking in the

mountainsof northernNew Zealand.Here he comes across Lake Taupoand is

startled o see the famoustenorLuciano Pavarotti loating on its pristinewaters.

They talk amiably fora while andPeter goes off flushedwith excitement.

Understandably enough, this experience of Peter's gives rise to many

subsequentmemories on his part,and to beliefs based upon them. Considerthe

ones that are tokened while he is still on Earth,many years priorto his trip to

Twin Earth. There can be no question about their content. They are about

Pavarottiand Lake Taupoand water.13 In fact, I take it, they would be about

Pavarottieven in a possible world in which Pavarottihas a twin living in Los

Angeles who Peter doesn't know aboutbut whom he subsequentlymeets. Even

after he does so, the memoriesof the encounteron Lake Taupo,and the beliefs







The Transparency f Mental Content 39

based upon them, would continue to be about Pavarotti,not Twin Pavarotti.

Well, some years go by and Peter is moved surreptitiously o Twin Earthand becomes happily ensconced there. Eventually, some of the tokens of his

mental names come to refer to the twin counterpartsof the familiar Earthly

figures, and some of the tokens of his general terms come to express twearthly

properties: some tokens of 'Pavorotti'will come to refer to twin Pavarottiand

some tokens of 'water' will come to mean twater.14

One day, perhapswhile reading a schedule of upcoming events, Peter is

moved to reminisceabout the occasion when he saw Pavorottiswimming in the

waters of Lake Taupo. He calls up vivid and accuraterepresentationsof the

scene. Of course, he takes himself to be rememberingscenes involving the

singer he is reading about now. But he isn't. His memories, intuitively, are

about theEarthlyPavarotti, heEarthlyLakeTaupo,andEarthlywater, previous

perceptionsof which are the sustainingcause of his laterability to recall what

the scene looked like andhow it felt.In the situationdescribed,Peter'sexternally ndividuated hought tokensare

not epistemically transparent o him. In particular,Peter's language of thought

contains tokenexpressions that possess different semanticvalues, despite being

of the same syntactic type. And yet, clearly, Peter does not know that they do.

Tokens of 'Pavarotti', 'water', and 'Lake Taupo', in sentences expressing

memories and beliefs about thatmemorableoccasion, will mean Pavarotti,waterand Lake Taupo, respectively; whereas other tokens of that type, in sentences

expressing beliefs about his current environment, or current desires, will

intuitively mean twater and twin Pavarotti.From the inside, however, therewill

be no indication of this:as faras Peteris concerned, they will appear o express

precisely the same contents.15

Externalism, then, would appear to be robustly inconsistent with the

epistemic transparency f thoughtcontents:thoughtsthat have the same content

may look to introspectionto have distinct contents (as in Pierre and Paul), and

thoughts that have distinct contentsmay look to introspectionto have the same

content (as in Peter). Externalistcontents fail, in other words, to possess thefeaturethatDummettproclaimed"undeniable."Whatproblem,if any, does this

pose?

I shall argue that the problem is this. We don't just ascribe thoughts to a

person in order to say something descriptively true of him. We use such

ascriptionsfor two relatedpurposes:on the one hand,to enable assessments of

his rationality and, on the other, to explain his behavior. As these mattersare

currentlyconceived, a thought must be epistemically transparentf it is to play

these roles. Withouttransparency, urconceptionsof rationalityand of rational

explanation yield absurd results. We manifest our recognition of this fact bybarring de re thoughts-thoughts which intuitively lack epistemic

transparency-from figuring in assessments of rationality and psychological

explanation.However, if we abandontransparency ven for de dicto thoughts,








and hence in effect altogether, then we must either jettison the notion of

rationality and with it the practice of psychological explanation that itunderwrites,or we must show these notions can be refashionedso as not to yield

absurdresults.The problem is that the first suggestionis wild and there appears

to be no obviously satisfactoryway of implementing he second.

The Apriority of Logical Properties and De Re Thoughts

We may usefully begin with a discussionof de re beliefs. By a 'de re belief'

I shall mean, by stipulation, a belief that is individuated by the objects it is

about. Such beliefs are typically reportedwith the use of an 'of-clause', ratherthan a 'that-clause',as in,

Jane believes of the piano that it is ugly,

and their contentis given by a Millian proposition.Clearly, and in contrastwith

fully conceptualized, eferentially paquede dicto beliefs, it is both necessaryand

sufficient for the distinctness of two de re beliefs applying the same predicate

thattheyconcerndistinct objects.

I should emphasizehere thatI am not presupposing, or the purposesof this

discussion, either that this account explicates the 'intuitive' notion of de re

belief, if thereis such a thing,or even thatthere actuallyarede re beliefs in this

stipulatedsense. I am only interested n the questionwhat would be true of such

beliefs, if there were any.

Well, one claim that is often made in connection with such thoughts is

nicely expressedby Burge:

individual entities referredto by [referentially]transparentlyoccurringexpressions, and, more generally, entities (however referredto orcharacterized)of which a person holds his beliefs do not in general play adirect role in characterizing he natureof the person's mental state or event.Thedifferencebetween uch entities]does not bearonAlfred'smind n any

sense thatwould mmediatelyffectexplanations f Alfred'sbehavior rassessmentof the rationality f his mentalactivity. .. Moreover, it seemsunexceptionable to claim that the obliquely occurring expressions inpropositional attitude attributionsare critical for characterizinga givenperson's mental state. Such occurrencesare the stuff of which explanationsofhis actions and assessments of his rationality are made.16

Burge's striking claim here is that de re beliefs, in contrast with fully

conceptualized, eferentially paque,de dictobeliefs, don'tenterintoassessments

of a subject's rationalityor psychologicalexplanationsof his behavior.Here,for

the purposes of further illustration, is Jerry Fodor making a similar point

(Fodor'semphasisis on psychologicalexplanation):Suppose I know that John wants to meet the girl who lives next door, andsuppose I know that this is true when "wantsto" is construedopaquely. Then,given even rough-and-readygeneralizations abouthow people's behaviors







The Transparency f Mental Content 41

are contingent upon their utilities, I can make some reasonable predictions(guesses) about what John is likely to do: he's likely to say (viz., utter), "I

want to meet the girl next door." He's likely to call upon his neighbor....On the other hand, suppose that all I know is that John wants to meet the

girl next door where "wants to" is construed [referentially]transparently; .e.,all I know is that it's trueof the girl next door that John wants to meet her.Then there is little or nothing that I can predict abouthow John is likely toproceed. And this is notjust because rough-and-readypsychologicalgeneralizations want ceterisparibusclauses to fill them in...17

Whatreasonsdo Burge and Fodoroffer in supportof theirrespectiveclaims?

Burge, actually,has very little to say on the matter(thoughin fairness to him, I

should point out that the essay from which the citation is drawn is largely

concemed with other questions).And the exact interpretation f what Fodorhas

to say would take us too far afield. So without concerning myself overmuch

with why these authors believe that de re beliefs are unfit for the purposesof

content-based psychology, let me offer my own explanation.Let's begin with

the question aboutrationality.

Suppose that Jane sees a wholesome-looking apple. She thinks de re of the

apple that it is wholesome. That is, she comes to believe the Millian

proposition

<apple,, is wholesome>.

She subsequently sees the same apple with its blemished side exposed. She

thinks de re of this apple thatit is not wholesome. Thatis, she comes to believe

the Millian proposition

<applel, is not wholesome>.

Two things are true of Jane in this case. First, her dere beliefs about the apple

logically contradicteach other: the (Millian)propositionsubtendedby the one is

p and the one subtendedby the otheris not-p. And, second, she cannot recover

from this condition on an a priori basis; to discover that the beliefs contradict

each other she would have to learn an empirical fact, namely, that the appleinvolved in the first thought is identical to the apple involved in the second.

That the two thoughts logically contradict each other is not introspectively

accessible to her.18

Clearly,a similarcase can be described to illustratethe fact thatthe logical

consistency of Millian propositions is also not necessarily introspectively

accessible. In general, then, the point is that the logical properties of de re

propositions are not knowable a priori. The question is: How might this fact

help explain why de re beliefs are unfit for the purpose of assessments of

rationality?







42 / Paul A. Boghossian

The Apriority of Logical Properties and Norms of Good

Reasoning

The answerderives from ourconceptionof the natureof rationalityand, in

particular,of what it is for someone to be a good reasoner. Whatdoes a person

have to do in order to count a good reasoner?Clearly, it is not at all a question

of knowing empirical facts, of having lots of justified true beliefs about the

external world. Rather, it is a matter of being able, and of being disposed, to

makeone's thoughtsconformto the principlesof logic on an a priori basis.19A

surreptitiously nvattedbrain transplanted from its normaladult body into a

vat and attached to a computer that seamlessly duplicated and continued its

previous course of experience-could be as good a reasoner as it ever was,

despite the sharp escalationin the numberof its false beliefs aboutthe external

world. Or so, at any rate,our conceptionof rationalityrequiresus to think.

So, rationality is a function of a person's ability and disposition to

conform to the norms of rationality on an a priori basis; and the norms of

rationalityare the normsof logic. We may, if we wish, put matters n a far less

committal way: let's say thatbeing minimallyrationalis a matterof being able

to avoid obvious violations of the principles of logic, given enough time to

reflect on the matterandso on. But even relative to this very minimal notionof

rationalitywe would appearto have stumbledonto a problem.For accordingtothis view, our Jane,who innocently believed of one andthe same apple thatit is

both wholesome and not wholesome, would appear to stand convicted of

irrationality:she believes a pair of Millian propositions that contradicteach

other, but she is unable to recoverfrom this predicamenton an a priori basis, no

matterhow long she may be given to reflect on the matter.But, intuitively,

there is nothing irrational about her. Therefore, either our conception of

rationality is mistaken or we have to find some other systematic and non-

arbitrary way of absolving Jane's cognitive behavior. Obvious conservative

solution: Bar an agent's de re thoughtsfrom entering into assessments of her

rationality.The ban on contradictory elief andinvalid inference s preserved,butonly in applicationto de dicto beliefs, just as Burgesays.

The Apriority of Logical Properties and Psychological

Explanation

Precisely parallelconsiderationsexplain why de re thoughts are also unfit

for the purposes of psychological explanation. Since rationality is taken to

consist in the ability and disposition to conformto the principlesof logic on an

a priori basis, any rationalsubject,regardlessof his externalconditions, maybeexpected to obey certain laws (or counterfactual-supporting eneralizations):

namely, those generalizations that mirror the introspectively obvious logical







The Transparencyf MentalContent 43

consequences f a person'spropositionalttitudes.Thus,ourordinary sycho-

logical practiceof explainingandpredicting ehaviors built upon appeal osuch aws as this:

If S occurrentlyelievesp andoccurrentlyntends oF if p, and f S hasno independenteason or not F'ing, henS will intend o F or,at thevery east,will be disposed o intend o F.If S intendso F iff p, butdoesnot believep, butmerelyq instead,(wherep andq are ogically ndependentropositions),henS will notintend o F.

The trouble is that perfectly rational subjects will not obey thesegeneralizations,hen heyareconstrueds quantifyingverde re thoughts.

Thus, suppose hatJack ntends o call theFBI wheneverhe is within50

feetof a spy.And et's suppose hathe believesof the deanof hisCollegethathe is a spy. The dean in fact is currently eatednext to him at the beach,disguisedas a lifeguard. o, de re, he believesof the 'lifeguard'eatednext tohim, that he is a spy. Yet he doesn't budge.Yet Jackwould appear o be aperfectlyational erson.

The elementary eneralizationsponwhich thepracticeof psychologicalexplanation epends ail to hold for dere thoughts.Therefore, itherthere s

somethingwrong with our conceptionof rationalityand, hence, with thegeneralizationshat t underwrites,rwe mustfind somewayof justifyingour

beliefin thesegeneralizations,onsistentwith their ailing o holdin this case.Obvious conservative solution: Except a subject's de re thoughts frompsychological xplanationsf his behavior.Theelementary eneralizationsnwhichpsychological ractice epends represerved,utonlyinapplicationo d

dictobeliefs, ustasBurgeandFodor ay.It seems to me, then, that we have before us the generalanswerto the

question: Whatconsiderationsmight underwriteBurge's claim that de re

thoughts are unfit for the purposes of assessments of rationalityand

psychologicalxplanation?heanswer s thatboth heseenterprisesequirehatthelogicalpropertiesf thepropositionalttitudesheymanipulatee knowableapriori;and he ogicalpropertiesf derethoughtsren't. Again, don'tclaimthat his s the answer hatBurgehimselfwouldgive.)

Reference and Epistemic Transparency

Why, though, are the logical propertiesof referentially ndividuatedthoughts-whether these be dere thoughtsnormally o-called,or thoughts

involvingMillianpropernames-not knowablea priori?Whatmakes hesethoughts pecial nthisregard?

The answer,of course, is that the logical propertiesof referentially







44 / PaulA.Boghossian

individuatedhoughtsail to be a priorijustbecause hese thoughtsail to be

epistemicallyransparent.t is preciselybecauseand onlybecause-a subjectis unable o tell a prioriwhether hethought e expresseswith

Tully s bald

is the sameas the thought e wouldexpresswith

Cicero s bald

thatheis unable otell whetherhethought ewouldexpresswith

Tully s baldandCicerosn't

is or is not a contradiction,n a Millianconstrual.Referential ransparency

entailsepistemic pacity.It is worthnoticingn thisconnectionhatviolations f thetransparencyf

sameness ndviolations f thetransparencyf differencenducedifferentortsoflogicaldefect.A thinkeror whomthetransparencyf samenesss false may

well fail, as we have seen, tobe able to tell apriorithata givenpropositions

in fact a logicalconsequence f otherpropositionshathe believes.That t is

sucha consequences a facthe will be able to discoveronly a posteriori,bylearningomeempiricalacts.

Thus,Pierrebothbelieves hat

HelivesinLondon

andbelieves hat

Ifhe livesinLondon'Londres'), elivesinthesamecityas Oscar

Wilde ived.

Yet he is unable o draw heconclusion hathe lives in the samecityas Oscar

Wilde ived n.

Such a thinker,however,need not be supposedever to actuallyreasoninvalidly.It is consistentwiththe falsityof the transparencyf sameness, n

otherwords, hatall the simple nferenceshat ook a priorito sucha thinkero

be valid,are valid;what s falsified s theclaimthatall thesimple nferences

thatarevalid,willnecessarilyo look.Onthe otherhand,a thinkerwho suffers rom heconversedefect failure

of thetransparencyf difference-will suffer rom he converseailing.Forsuch

a thinker,ertainnferencesmaywelllookvalid,when hey nfactarenot. And

thatthey are not is a fact he will be able to discoveronly a posteriori,by

learningomeempiricalacts.

Peter providesan appropriatexample.Since his languageof thoughtcontains okenexpressionhatdiffer n semantic aluedespitebeingof the same

syntactic ype,he will be temptedo think hatcertainnferences reinstances








of, e.g., modus ponens, when they in fact aren't.Thus,he might muse to

himselfas follows:Whoever loatson water, ets wet.

This thought,by virtueof expressing generalquantified roposition,s to beregarded s havingTwearthlyontent, .e. as beingabout win-floatingndtwinwater. Now, however,he combines t with a memorybelief that he wouldexpresswiththe sentence

Pavarottincefloated n water

to conclude

Pavarotti nce got wet.

The inferencewill seem validto him;but t arguablysn't.Thesecondpremise,by virtueof expressing memorybelief that s rooted n anEarthly xperience,will be aboutEarthly loatingandEarthlywater.Truepremises,aided by afailureof univocity hatPeter s inprinciple ot in a position o introspect,willcombine oproduce falseconclusion.

Returningo the main ine of argument, e see, then, hat he undamentalanswer o thequestion:Whyare de re thoughts nsuitableor thepurposes f

assessments f rationalityndpsychological xplanations,s this: t's becausedere contents(and Millianpropositionsquite generally)are not epistemicallytransparent.y contrast,ullyconceptualizede dictothoughts, owhere ubjectto co-referential ubstitution, re supposed o be, as Burgerightlysays,"thestuff of whichexplanationsf hisactionsandassessments f hisrationalityremade."

If the diagnosison offer s correct,however, hiscontrast an holdup onlyif fully conceptualized edicto houghts re transparent.ut as Loar'sexampleof Paul andmy exampleof Petershow, fully conceptualized edictocontentswill themselvesailtobe epistemicallyransparentand hencewillthemselvesfail tobe suitable orthepurposes f psychologicalxplanationndassessmentsof rationality-if theyare ndividuatedxtemalisticallynthemannerhatBurgeadvocates.On whatbasis, hen,does Burgedistinguish etween hem?Whyareextemalisticallyndividuated,nd,hence,non-transparent,edicto houghts eldto be suitable orthe purposes f rational sychology,when de rethoughtsreconceded ot obe?

The Apriority of Logical Properties and the Introduction ofSense

It seems to me that once epistemic transparencys identified as asemantically ignificant hesis, ts rolein a varietyof important isputesn the








philosophy f language ndmindbecomesobvious. shalldiscuss woof these:

therole it hasplayed n the canonical rgumentor the thesisthatnameshavesense andnot merelyreference,andthe role it plays in generatingKripke'spuzzleaboutbelief.

Beginningwiththe former,manyphilosopherswouldprobably esist the

claim hat ransparencylaysapart n the canonical rgumentmotivating non-Millianview of names,becausetheywould resist the claim thatthere s anysuchargument.Mostphilosophers riteas if it's merelyobvious-and, hence,inneedof noargument-that omeonemightbe ina state rulydescribed y

(1) Marybelieves hatAliwas a champ

butnot therebyna state rulydescribed y

(2) Marybelieves hatClaywas a champ.

It'sworth eeing,however, hat t isn't merelyobvious. t becomescompelling,as I shall now argue,only whenone makesthe assumption,eft implicitbyFrege, that beliefs involving proper names are fit for the purposesofassessments f rationalityndhencemustbe epistemicallyransparent.tis onlyunder he termsof thisassumptionhatonegets an argumentor the referentialopacityof proper ames nthefirstplace.20

For considerhowa beliefin referentialpacity s typicallymotivated.Wearegiven a case whichgoes like this:Mary incerely ssertsAliwasa champ'.She also sincerelyasserts Claywasnot a champ'.She asserts hese sentenceseventhough t is clear hat,as she is using henames Ali' and Clay'theyreferto one andthe samelegendaryboxer.Now, given the followingprinciple orreporting eliefs

Jones'sincere ssertion f 'p' expresses is beliefthatp

wemayconcludehat

(1) Marybelieves hatAliwas a champ

and hat

(3) Marybelieves hatClaywas nota champ.

It is importanto notice,however, hatnothing o far barsus fromsupposingthatbeliefsinvolvingnamesare referentiallyransparent,nd,hence, hat 1) is

equivalento

(2) Marybelieves hatClaywasa champ.

Forall thatthis wouldentail s thatMaryhas contradictoryeliefs,a stateofaffairs hat s, presumably,erfectly ossible.We need o begivena reasonwhyan ascription f contradictoryeliefs is unacceptablen thepresent nstance.








Otherwise,we wouldhave no case illustrating,ndconsequently o argument

for,thereferentialpacityof beliefs nvolvingnames.What s that eason?We get sucha reasononlyif we insist thatbeliefsinvolvingpropernames

must be fit for the purposesof assessmentsof rationality nd psychologicalexplanation,ndhencemusthave ogicalpropertieshatareknowable priori -

must, hat s, be transparent. rmedwith such an assumptionhe argumentorreferentialopacity is finally enabled. For the assumption nsists that theattributionf a contradictoryairof beliefs nvolving roper ames s acceptableonlyif the fact that heycontradictachother s apriori available o the subjectto whom they are attributed. et it seems perfectlyclearin this case thatnomatter ow muchMarymight earchherown mind, he won'tdiscover hat he

belief she expresseswith 'Ali was a champ' s referentiallyand,hence,on aMillianconstrual,ogically) nconsistent ith the belief thatshe expresseswith'Clay was not a champ'.Withthe insistencen place, then, t follows, that 1)and 3) couldnotbeattributingogically ontradictoryeliefsand,hence, hat 1)is not equivalent o (2). If we ignore,as Fregehimselfevidentlydid, views thatattempt o accommodatehisresultby analyzing eliefin termsof a three-placerelation,we get Frege'sconclusion:heremustbe a level of semantic escriptionof beliefs nvolvingnamesother han hereferential.

Kripke's Puzzling Pierre

Wemayobserve heverysamedialecticatwork nKripke'samousPierrecase.Inone scenarioI won'tdiscuss heother),Pierre ssentsboth o

Londress pretty

and o

London s not pretty.

Kripke rgues hat, n tfiiscase, there s no sayingwhat t is thatPierrebelieves,no satisfactory elief ascription.His overall dea is to protectMillian theoriesfrom the charge hatthey generateabsurdbelief ascriptions, y showing hatparallel bsurditiesanbegeneratedolelyfromprinciplesonstitutivef beliefas such,and without elianceon distinctivelyMillianprinciples for example,substitutivity):

When we enter into the areaexemplified by...Pierre,we enter into an areawhere our normalpractices of interpretationand attributionof belief aresubjected to the greatest possible strain, perhaps to the point of breakdown.So is the notion of the content of an assertion, the proposition it

expresses.21

But is it really true that we are unable to say what it is thatPierre believes

in the case as described? Why can't we say, applying our disquotational







48/ PaulA.Boghossian

principle,hatPierrebelievesboth thatLondons prettyandthatLondons not

pretty?Kripkewrites:...there seem to be insuperabledifficulties with this altemative as well. Wemay suppose that Pierre...is a leading philosopher and logician. He wouldnever let contradictorybeliefs pass. And surely anyone, leading logician orno, is in principle in a position to notice and correct contradictorybeliefs ifhe has them... .[Pierre]cannot be convicted of inconsistency: to do so isincorrect.22

Herewe see theassumptionf epistemic ransparencylayingan explicit

role:Pierrecannotbe ascribedcontradictoryeliefs because"anyone...is n

principlena positiono noticeandcorrect ontradictoryeliefs f he hasthem";

andyet in thiscase it is clearthathe will not be ableto do so. But theblameoughtnot to be placedon the veryideaof belief or propositionalontentas

such,at least notin the first nstance.Rather,he sourceof the problemies in

the fact thatKripke s workingboth with the requirementhat contentbe

transparent nd with a notion of propositionalcontent that falsifies thatassumption. heimpressionf a puzzle s generatedy keeping wo conflictingelementsat playat thesametime.No wonder, hen, hatno satisfactory elief

ascriptiono Pierres forthcoming.

A Simple Solution?

Our story thus far has unfoldedas follows. We have the view that

rationality, r at anyrate,goodreasoning,s thedispositiono conform o the

principles f logic on an a priori basis.This view, in conjunctionwitha non-

transparentonception f propositionalontent,yieldshighlycounterintuitive

results.Ourwillingness o excludede re thoughts n particular, ndMillian

thoughtsngeneral,rom heprovince f rationality-basedsychologymanifests

ourrecognition f thisfact. It commitsus to theassumption, nlesswe are to

forego rationality-based sychologyaltogether, hat de dicto contents are

epistemicallyransparent.Yet mostof usadhere o conceptionsf propositionalontent, hiefamong

them externalistconceptionsof propositionalontent, hathave it as a con-

sequencehatde dictocontents renottransparent. ecannothave t bothways.

We must eitherrejectsuch conceptions f propositionalontent,or we must

show how to refashionheideaof reasoningo that t no longerconsists n the

dispositiono conformologic on anaprioribasis.But isn't there n fact a simpleway of refashioninghe idea of reasoning

thatwill takecareof theproblemases?23Well,how would tgo?

We can't just say: A personis absolvedfor believinga contradictionprovidedhathe doesn't-or can't-'see' that t's a contradiction.rrationality

often consists n preciselysuchfailure.We want n somesenseto capturehe

fact thatPierreor PeterorPaulare blameless or notseeingthe contradictions








that he storiesattributeo them.The questions how that s to be done.

The only proposal can thinkof is this:A thinkers to be absolved orbelieving a contradiction, rovided hat the contradictoryharacter f theproposition e believes s inaccessibleo merea priorireflection n his part.

The troubles that,against he backgroundf a non-transparentonceptionof propositionalcontent, any contradictorypropositionwill satisfy thatdescription. inceon a non-transparentonception,t is preciselynot availableto mere a priori reflection hat a given belief is a belief in a contradiction,practically nycontradictoryelief will be absolvableunder he termsof thisproposal. The only exceptions will be those beliefs of which it issimultaneouslyrue i) that heyarebeliefs ncontradictions,ii) that he subjectbelieves them to be beliefs in contradictions,nd(iii) whichhe neverthelessrefuses-mysteriouslyenough to changehis mind about.Any such subjectwouldundoubtedlyountas exemplifying formof extremerrationality;uthecouldhardly e consideredhenormal ase.

Conclusion

There s a pervasive ensionbetweenour conception f rationalitynd thepracticeof psychologicalexplanation t underwrites, n the one hand,and

currentlydominant onceptionsof mentalcontent,on the other.The formerpresuppose what the latter deny. One or the other conceptionmust bereconsidered.24

Notes

1. This paper is extracted from a longer unpublished manuscript, much of whichdates back to 1987. Some portions of the present paper overlap with portions of"Externalismand Inference" in E. Villanueva (ed.): Rationality in Epistemology(Atascadero: Ridgeview Press, 1992). For helpful reactions either to this paper,or to the material in the larger manuscript, I am grateful to Barry Loewer,Stephen Schiffer, Jennifer Church, Crispin Wright and David Velleman;participants in seminars at Princeton and Michigan; and audiences at St.Andrews, Stony Brook, Campinasand Oxford.

2. Truth and Other Enigmas, (Cambridge,Mass: Harvard University Press, 1978),p. 131.

3. See Hilary Putnam: "The Meaning of 'Meaning,"' in his Mind, Language andReality (Cambridge: Cambridge University Press, 1978).

4. See Tyler Burge: "Individualism and the Mental," in Midwest Studies inPhilosophy, 1979. Burge, of course, regards externalism to be the correct moralof the thought experiments.

5. Throughout this paper, I shall use 'externalism about mental content' to referonly to the sorts of dependence on external factors that are encouraged by the

Putnam and Burge thought experiments. Any number of other sorts ofdependence are also imaginable, but I take it that no case has been made fortaking them seriously as models of our propositional attitudes.

6. "Reply to Burge," unpublishedms.7. Though I certainly don't wish to assume that Davidson himself would develop it







50 / Paul A. Boghossian

along these lines. For a similar proposal see Ernie Lepore and Barry Loewer:"Solipistic Semantics," in Midwest Studies in Philosophy, 1980.

8. See Kripke: "A Puzzle About Belief," reprinted in Salmon and Soames (eds.):Propositions and Attitudes (Oxford: Oxford University Press, 1988).

9. "Social Content and Psychological Content," in Grimm and Merrill (eds.):Thought and Content (Tucson: University of Arizona Press, 1987). As I explainbelow, Loarhimself uses his example to illustratea different thesis.

10. Loar himself uses the case to argue for individualism about thought content. Hisidea is to concede that the beliefs would be reported as the externalist claims,while still maintaining that their psychologically relevant contents would beindividualistically individuated.

11. The assumption that there could not be disambiguated tokens of an ambiguousword type in the language of thought is widespread in the philosophy of mind.See, for example, JerryFodor: Representations (Cambridge:MIT Press, 1981),p. 227.

12. "Individualismand Self-Knowledge," Journal of Philosophy, December 1989, p.652.

13. In this example, I am not taking a stand on whether the intuitive semantics ofpropernames is Millian or not. I intend the example to work on either reading.

14. Obviously, my Twin Earthis richer than Putnam's in containing counterparts ormany more objects, properties and people than his.

15. Ideally, this example would be described at much greater ength and would includea discussion of various objections and replies. For some furtherdiscussion seemy "Externalism and Inference," Stephen Schiffer's "Boghossian on External-ism and Inference" and my "Reply to Schiffer" all in Philosophical Issues, op.cit.

16. "Other Bodies," in Woodfield (ed.): Thought and Content (New York: Oxford

University Press, 1982), p. 99, my emphasis.17. "Methodological Solipsism Considered as a Research Strategy in CognitivePsychology," reprinted in Representations (Cambridge: MIT Press, 1983), p.234-5.

18. I am assuming here, and in the rest of this paper, that logical properties are, inthe first instance, properties of propositions, not of the sentences whichexpress them. I cannot defend this assumption here, except to say that it seemsto me incredible to suppose that we cannot, in point of principle, evaluate alanguageless creature's logical abilities. I hope to discuss this issue at greaterlength elsewhere.

19. Notice that I am talking primarily about 'good reasoning,' rather than'rationality' more widely construed. I take good reasoning to involve normsconcerning the manipulation of propositions already at hand; I leave it open

whether rationality involves something more.20. Here I follow a point made by Kripke n "A Puzzle AboutBelief."21. p. 135.22. p. 122.23. In discussions of this paper, I have often encountered the impression that there

may be.24. Since writing this paper, I have come across two recent discussions of epistemic

transparency-Ruth Millikan: "White Queen Psychology," unpublishedms; andJoseph Owens: "Cognitive Access and Semantic Puzzles," in Anderson andOwens: Propositional Attitudes (Chicago: CSLI, 1990). Both writers agree thatthe notion has played an important role in traditional discussions, yet bothadvocate that it be eliminated: they concur in seeing a belief in transparencyas amere residue of a 'Cartesian' conception of mind, Millikan calling it the "last

myth of the given." However, neither writer satisfactorily explains, in my view,how the crucial notions of reasoning and rationality are to be plausiblyrefashioned in its absence.

Documents

The Transparency of Mental Content