Sergio Tenenbaum the Judgment of a Weak Will

8/12/2019 Sergio Tenenbaum the Judgment of a Weak Will
1/38
International Phenomenological Society
The Judgment of a Weak WillAuthor(s): Sergio TenenbaumSource: Philosophy and Phenomenological Research, Vol. 59, No. 4 (Dec., 1999), pp. 875-911Published by: International Phenomenological SocietyStable URL: http://www.jstor.org/stable/2653561.
Accessed: 25/01/2011 04:41
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at.http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .http://www.jstor.org/action/showPublisher?publisherCode=ips..
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
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 support@jstor.org.
International Phenomenological Societyis collaborating with JSTOR to digitize, preserve and extend access to
Philosophy and Phenomenological Research.
http://www.jstor.org
http://www.jstor.org/action/showPublisher?publisherCode=ipshttp://www.jstor.org/stable/2653561?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=ipshttp://www.jstor.org/action/showPublisher?publisherCode=ipshttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/2653561?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=ips

2/38
Philosophy and Phenomenological ResearchVol. LIX, No. 4, December 1999
TheJudgmentof a Weak WillSERGIOTENENBAUMUniversity of New Mexico
In trying to explain the possibility of akrasia (weakness of will), it seems plausible todeny that there is a conceptual connection between motivation (what one wants) andevaluation(whatone judges to be good); akrasia occurs when the agent is (most) moti-vated to do something that she does not judge to be good (all things considered).However, it is hard to see how such accounts could respect our intuition that the akraticagent acts freely, or that there is a difference between akrasia andcompulsion. It is alsohard to see how such accounts could be extendedto the realm of theoretical reason,butthis is generallynot taken to be a problem, because it is generally assumed that there isno similar phenomenonin the realm of theoretical reason. This paper argues that thereis such a thing as theoreticalakrasia, and thatwe can find a characterization f this phe-nomenon in Descartes's Meditations. Drawing on certainpassages in the Meditations,we can construct an account of theoreticalakrasia; this account can then be adaptedtoresolve the original problem of akrasia in the realm of practicalreason. The accountasserts that there is a conceptualconnection between motivation and evaluation in freeaction; it also enables us to show how the akraticagent is still acting freely when hedoes somethingthat he does notjudge to be the best all thingsconsidered.
In the Protagoras, Socratessays thatmostpeoplemaintainthatthere are many who recognize the best but areunwilling to act on it. (...) When-ever I ask what can be the reason for this, they answer that those who act in this way areovercome by pleasureor pain.1Both Protagorasand Socratesfind this view untenable.Socrates shows thatthis view is untenableby first identifying pleasure with the good. He thencan go on to presentthe following reductio of thispopularview:Suppose we now say that a man does evil thoughhe recognizes it as evil. Why? Because he isovercome. By what? We can no longer say by pleasure because it has changed its name togood. Overcome we say. By what we are asked. By the good, I suppose we shall say. I fearthat if our questioneris ill-mannered,he will laugh and retort:Whatridiculousnonsense for aman to do evil, knowing it is evil and that he ought not to do it, because he is overcome bygood.2
I Protagoras; 352e (trans.by W. K. C. Guthrie n The CollectedDialogues of Plato, editedby E. Hamilton and H. Cairns;Princeton:PrincetonUniversityPress, 1961).2 Protagoras; 355c.
THE JUDGMENTOFA WEAK WILL 875

3/38
Even if we find the identificationof good andpleasure ess palatablethanSocrates' audiencedid, we mightstill acceptwhat Kantcalls the 'old formulaof the schools': IWe desire only what we conceive to be good; we avoid only what we conceive to be bad.But in this case, we seem to be bound to join Socrates' imaginary ill-mannered nterlocutorn findingridiculousthe vulgarbelief thatwe can pur-sue evil willingly. For, if pleasurecan overcome the agent, this is becausethe agentdesiresit, and if the agentdesiresit then theagenthas to conceive itas good. And once we make this inference it seems hard to avoid Socrates'conclusion: thatpeople can be overcome by pleasureto pursuean evil is aridiculousopinion. Indeed it is hardto avoid the conclusion thatthe opinionis as ridiculousas can be: it is a contradiction.Yet acceptingSocrates'conclusionseems to force us to deny anextremelycommon phenomenon.For we all have found ourselves yielding to tempta-tion, procrastinating, ating beyond healthy limits, avoiding pain, and evenabandoning he greatergood for the sakeof pleasure.This phenomenon s thephenomenonof weakness of the will or akrasia.Since it is not easy to live with contradictionsor to turn a blind eye toordinary phenomena, it is often tempting to do away with Socrates'identification of pleasureand good or the old formulaof the schools. Whatthese two have in common is the commitmentto a conceptual tie betweenmotivation and evaluation(or a certainkind of judgment),andI will call anyview committedto this kind of conceptualconnection a 'scholasticview.' Wecan read the old formula of schools in two ways. In one way, the formulamerely states a weak necessary condition for desiring something; anythingthatI desire I must somehow conceive to be good. On this reading, desiring(or motivation)andconceiving to be good (or evaluation)could be in impor-tantrespects out of tune;for example, I could stronglydesire somethingthatI conceive to be slightly good. However,we may readthe old formulaof theschools as makinga strongerclaim; thatis, that the desire for a certainthingshould be identified with a positive evaluation of this thing. This readingrules out the above disparity between motivation and evaluation. In thispaper, I will be interestedprimarilyin the strongerreading of the formula.Indeed one of the aims of this paperwill be to defend the plausibility of ascholastic view committed o thestrongerreadingof the formula.4
I have adaptedthe formulationand the denominationof the "old formulaof the schools"which is found in Kant's Critique of Practical Reason (trans.by L. W. Beck; New York:Macmillan, 1956; p. 61). Kant himself expresses some reservationsaboutthis saying, butI will ignore them.Of course the weaker readingwill discussed and arguedfor by implication. As we willsee in a moment,I thinkthe correct version of the scholastic view identifies desire with a
876 SERGIOTENENBAUM

4/38
Socrates identifies pleasure and good because these are the things wepursue,and the formulaof the school insists that we could only desire whatwe judge to be good. Either way, desiring,wanting or pursuingon one hand,and judging somethingto be good on the otherhand,cannotpart company.And why not find the culprit right here? If we say that the akraticagent ismotivatedby whathe does notjudge to be the best, we can accept the usualphenomenonwithout incurringany painful contradiction.Yet this move creates some difficulties which we will be in a better posi-tion to assess later. But we can have a preview of the problemsif we thinkabout a different, somewhat less common, phenomenon: compulsion. Imight be compelled to do something by other people, or I might be com-pelled to do something by my own desires, my urge gettingthe betterof me.So a drug addict mightbe incapable of resistingthe strengthof his urge, andin a moment of madness I might indeed be incapable of overcoming thestrengthof my desire for ice cream. As Aristotle says, "when nature is thecause, no one would call the people akratic."5 here seems to be a differencebetweenakrasia andcompulsion.Aristotlesays thatakrasia is "blameworthyand base", and though we might find this language too strong6we do con-ceive of an akraticpersonas a free and responsible agent (or, at least this iswhat makes theissue of akrasiapuzzling,for there s no difficultyunderstand-ing thatwe might lose control of our bodily movements). Indeed, we rarelyfind that the claim 'I was too weak to resist temptation' s enough to excusethe agent from all, if any, blame. On the other hand, an agent acting undercompulsion is not to blame for her actions; if my desires, working 'behindmy back', drive me to an action despite myself I am no more free than incases of externalcompulsion.Onemightwant to deny thatakraticactions areindeed free, but this mightbe the consequenceof despairingof having failedto find room for free akraticaction within one's philosophicalviews. At anyrate, this paper will try to explain the possibility of akrasia on the under-standing hat the akraticagentis a free agent.Donald Davidson's seminal articleon weakness of the will tries to recon-cile weakness of the will with a view of motivation in which it is identifiedwith evaluation. In the first section of this paper, I briefly examine David-son's effort to resolve the problem.The second section discusses attemptstoopen a gap between motivation and evaluation.I will endeavorto show thatthe strategyof discardingthe old formulaof the schools is not compatiblewith a properunderstanding f the relationbetween motivationandevaluation
positive evaluationfrom a certain perspective. These evaluations might be overridenupon reflection(i.e., from a reflectiveperspective).5 Nicomachean Ethics (trans. by T. Irwin; Indianapolis:Hackett Publishing Co., 1985; p.186 (1148b30)).6 Aristotle himself qualifies this claim later, and saying that the akraticagent is not base but"halfbase" (See Nicomachean Ethics, 1152al5-20).
THEJUDGMENTOF A WEAK WILL 877

5/38
in free action.Explainingthe possibility of akrasiain terms of a gap betweenevaluation and motivationseems plausible only as long as we thinkthatthispossibility is peculiar to practicalreason-if we think thatthere is no suchthing as theoreticalakrasia. Since thereis no equivalentsplit between evalua-tion andmotivationin theoreticalreason,7 f we find thatakrasia is also pos-sible in the theoreticalrealm,we have a generalreason to reject this way ofaccountingfor weakness of the will. I arguethat we can find in Descartes'sMeditations a characterization of theoretical akrasia. Although this issignificantin itself, I arguethatalso we can find in the Meditations the basicresources to account for this phenomenon:a distinction between directandoblique cognitions and a distinction between primaryand reflective cogni-tions. Finally, I arguethat a similaraccount of akrasia canbe extendedto therealm of practical reason. This account will turn out to be a scholasticaccount of akrasia that neither denies thephenomenonnor falls prey to theills that befell Davidson's account.8Althoughthe primaryaim of this paperis to explain the possibility of akrasia, I believe that examination of thisissue will throwlight on the structureof practicalreason in general.
IDavidsonpresentsthe problemof akrasia as the conjunctionof the followingprinciples,which are at the same time individually plausibleand apparentlyinconsistent:
(PI) If an agent wants to do x more than he wants to do y and he believeshimself to be free to do either x ory, then he will intentionally do xif he does either x or y intentionally.(P2) If an agent judges that it would be better forhim to do x thanto doy, then he wants to do x more than he wants to do y.(P3) There are incontinentactions.9
This is not to say that there is no such thing as a motivatedbelief, a belief that is motivatedby a certaindesire. I will be ignoring these cases of motivated belief. Apartfrom thesecases, there is little sense to be made of a parallel distinction between motivation andevaluation (moreon this later).However, I will arguethat, even if we ignore these cases,theoreticalakrasiais possible.8 I shall not try to explain the possibility of what David Pears calls "underivativebrazen
akrasia." (See "How Easy is Akrasia" Philosophia 11, 1982. On this issue, see alsoAlfred Mele's "Pears on Akrasia and Defeated Intentions", Philosophia 14, 1984) Ibelieve that there is no difficulty in assimilating these cases to momentary changes ofmind or compulsion, depending on the case, but this is a topic for a differentpaper.I willhere assume that "underivativebrazen akrasia" is impossible. This assumption is alsoimplicit in Davidson's early account of akrasia (See "How is Weakness of the WillPossible?" n Essays on Actions andEvents, New York:Oxford UniversityPress, 1980).9 "How is Weaknessof the Will Possible?", p. 23.
878 SERGIOTENENBAUM

6/38

7/38
agent finding y better than x all things considered, and thus being weak-willed.However, this solution does not seem very satisfactory. Davidson is
committed to sayingthatone acts fromthe all-out udgment, but one choosesas one's all-out judgmentthat which was a prima-facie udgment overriddenby betterclaims to the good. One is struckby the mysteryof how one wouldsuccumb to such an obvious mistake.If we follow Davidson's own analogywith judgments of probability, the akratic person is like the person whoknows that the New YorkTimes published an announcementthat the presi-dentialelections would takeplace today,and believes thatthis is a typograph-ical error, but still goes to the local polling place. This form of stupiditydeserves explanation. Davidson is aware of this problem; he presents hissolution to it in the following sentences:If r is someone's reason for holding thatp, then his holding that r mustbe, I think,a cause ofhis holdingthatp. But, andthis is what is crucialhere, his holdingthat r may cause his holdingthat p withoutr being his reason;indeed the agent may even think thatr is a reason for reject-ingp.10
So, in the case of the orthodox Jew, though he did not hold that thereasons from the point-of-view of sensuous pleasure were good ones, thesereasons nonethelesscaused him to hold that it was betterto smoke. But hereDavidson seems to be paying mere lip service to the scholastic view. For inthis case, the unconditional acceptance of a value judgment seems to beweirdly independentof whether the agent is readyto endorse thisjudgment;thejudgment operates,so to speak,behind the back of the agent, despite hiscontraryassessment of the reasons. The fact that the desires that lead theagent to act can be said to be reasonsactuallyhas no role to play in explain-ing why the agent chose this particularaction-as opposed to any other thatcan be rationalizedby her beliefs and desire-since the action is not deter-mined by the agent's overall assessmentof thereasons.Davidson'scharacter-ization of akraticactioncan equally applyto actingon compulsion;the agentjudges a certaincourseof actionbest,but some contrarydesirestakechargeofthe agent."ISince Davidsonhimself leans towardleaving the scholastic view behind,we might as well look at those philosopherswho show no scruples in doingso. In examining these views, I hope to clarify the natureof the conceptualtie between evaluation and motivationin free action, so I can lay the ground10 "How is Weakness of the Will Possible?", p. 41.For furthercriticisms of Davidson's views on akrasia, see, inter alia, William Charlton,Weakness of Will (New York: Basil Blackwell, 1988), ch. 7, and Irving Thalberg,"QuestionsAbout MotivationalStrength"n Ernest LePore & BrianMcLaughlin,Actionsand Events: Perspectives on the Philosophy of Donald Davidson (New York: BasilBlackwell, 1985).
880 SERGIOTENENBAUM

8/38
for a plausible version of the scholastic view, and to show that preservingthis tie is a desideratumof any accountof akrasia.II
Denying that there is a conceptual tie between motivation and evaluationmight seem eminently plausible. We often say that something is good butnot what we want, or even that we do not know what is good but we knowwhat we want. But in another sense, it seems that the link of motivationwith evaluation, and thus the scholastic formula, should be unproblematic.For, indeed,if I wantsomethingthere is a sense in which I make a judgment;I take it that that which I want has some value, at least from a certain per-spective. For I might regretthat I wanted something, andI might judge thatmy want involved an illusory presentationof something as good. This mightbe the 'morningafter' of the akraticagent,but notonly of the akraticagent.Imight regret having wanted something, if it turned out to be incapable ofbringing me any satisfaction (the cake was stale), or if for any reason itturnedout thattherewas no value in the objectof my desire (eatinga cake isafter all not that satisfying, and could not justify my having wanted it sobadly). In the normalcases there is no problemin identifying a want with ajudgment,which might turn out to be correct or incorrect.No wantor desirethat I take seriously can come to me simply as a pang in my stomach, anincomprehensibleyen thatpushesme somewhere.Excepting some momentof insanity, my desires and wants must be directed towards certainobjects,projectsand aims whose pointI can see. And if we see the good merelyas themost abstract haracterization f the aim of any practical udgment, ust as wecan see the true as the most abstractcharacterizationof the aim of everytheoretical udgment, the identificationof motivationwith evaluation seemsto be unproblematic.To see some pointin what I do (even if the point is justsomething expressed by the sentence 'I felt like doing it') and so to judge itto be good, is not a sufficient conditionfor free action, but it seems to be anecessary one. If I chose to do something, I must have taken it, for somereason or another, o be somethingworthdoing. And this amountstojudgingit to be good, at least in the abstract sense of 'good' described above. Ofcourse this is not to deny thatpangs or yens or even wants and desirescouldtake control of my body and issue in a piece of behavior,but this would notbe a piece of free action.If I move my hand as soon as a lit cigarettetouchesit, I do so in a way that does not involve any judgment;I am moved by theburn without necessarily seeing any value in this action. In this case, how-ever, there is also no motivationinvolved. It would be imprecise, to say theleast, to claim that the person who moves her hand in this way was movedby her desire not to be burned.These are exactly instances of our behaviorthatdo not involve the mediatingcontributionof our desires.
THE JUDGMENTOFA WEAK WILL 881

9/38
We can think of wanting or desiring something as having a perspectivefrom which one takes a certain course of action to be good, or worthy ofbeing pursued.In less cumbersome anguage,we can say thathavinga desireinvolves seeing thepointof an action.Of course,the teleological characterofa desireguarantees hata desirehas always a certainaim. To say that a desireis a perspective on the good, however, involves more than saying that adesire aims at something; it implies not only that a desire has an aim, butalso thatit presentsto the agentthepoint of this aim (froma certainperspec-tive).
We can understandbetterwhy havinga desire involves seeing somethingas good from a certainperspectiveif we look at pathologicalcases of behav-ior, cases that do not seem to involve any desires. Suppose someone obses-sively washes his hands every five minutes. In this case, we are ordinarilyreluctant o ascribeto the agenta desire" to wash his hands(let alone a desireto have his hands cleaned). And the reason is that we see no point in thisbehavior, and assume that the agent also sees no point in it; the behaviordoes not issue fromtaking anything o be good fromany perspective.Suppose, however, one were to insist on ascribing desires to such acompulsive agent,on extendingthe word 'desires' also to referto these states,states of the mind that are not connected in any way to putativeevaluativejudgment. Still, in the context relevant to the understandingof akrasia, thecontext of free action,we can safely assume that all desires will involve see-ing something as good from a certain perspective. An agent, when actingfreely, can act on a desireonly insofar as thisdesirepresentsthe agentwith aperspectiveon the good; in otherwords,desires can contribute o free actiondirectly, by giving us reasons to pursuetheirobjects, only insofar as they areperspectives on the good. This point follows from a ratherweak assumptionabout the nature of free action:thatfor somethingto count as free action itmust be possible for the agent to provide some account of this action, toprovide her reasons for acting in this way. An agent who had no account ofher actions, whose actions were incomprehensibleto herself, would not beacting freely.'3 Now we can think of this account as a kind of deliberationthat the agentcould have gone throughpriorto the action. Whetheror not weassume that the agent actuallywent throughthis deliberation,we can repre-12 At least, as Joseph Raz points out, not in the philosophical sense of 'desire', in which adesire is supposed to figure prominentlyin explanation of intentional action. Raz also
defends the claim that "one can want (...) only what appears to one to be good or ofvalue" in "The Moral Point of View" in J. B. Schneewind (ed.), Reason, Ethics andSociety (Chicago: Open Court,1996), p. 70ff. His argument s similar to the one I presentbelow.13 One might want to include as a free agent, an agent who acted from unconsciousreasons. In this case, the agent could be incapable of accounting for her free actions.This possibility wouldnot do any harm o my argumentas long as there is some account tobe given of heraction in terms of her reasons for performing his action.
882 SERGIOTENENBAUM

10/38
sent his reasons for acting as a hypothetical deliberation. We can now askhow a desirecould enterthe agent's deliberation.Could a desire that did notpresentits object as having any point,as being good from a certainperspec-tive, be a putative item in deliberation?That I find myself craving some-thing, or moved in a certain direction, is not yet a reason to do anything.Suppose I wake up in the morningand I find that I am somehow inclined toreorderthe dishes, althoughI can't see any point in doing it: they are per-fectly well-arrangedand thereis nothingpleasantor interestingaboutmovingcups and plates around.Whendeliberatingabout what to do, thisinexplicablecravingcould notgive me, directly,a reason to give anything.Of course, I might find that the cravingis taking my attentionaway frommore important asksand decide that I better ndulgethis craving.Here,how-ever, it is not the cravingthat is the source of these reasonfor the pursuitofits object, but the desire to engage in these tasks that requiremy craving togo away-what I want is not to reorder he dishes but to get rid of the crav-ing. Indeed if the craving could go away by my taking a deep breath andcounting to five, I'd probably preferthis alternative,since the craving givesme no reason to reorder he dishes. Quitethe contrary, f I would continuetofind myself thus inclined, ratherthan thinking that reordering my dishesshould be given a certainweight in my deliberations, would look forprofes-sional help. To the extent that this urge takes hold of me, I am no longer act-ing freely; in fact, I am exhibitinga paradigmatic ase of compulsive behav-ior. As Joseph Raz puts it:So if I want to count the blades of grass in my gardenI do so because I thinkthatthis will takemy mind off some upsettingevent, or because the action has some other good-making prop-erty. I findmyself drawnto count bladesof grass,but cannot think of any reason for doing so, Iwould certainly deny that this is a desire of mine. It is a force which seizes me in spite ofmyself. If I am overcome by it and performthe action, I would be right to say that I could nothelp it.14
However, the case for a scholastic view is not so simple. There are goodreasons not merely to identify the things we are motivated to do with thethings we value orjudge to be good. I proposeto proceed as follows: I willexamineGaryWatson's contention hat motivation and evaluationmay be, incertaincases, completely divorced."5This will serve a twofold purpose.Onthe one hand,it will makeit clear why the conceptualtie between motivationand evaluation cannot be simply one of mutual implication. We must con-cede to Watson thatwe do notjudge to be good all the things that we desire.14 "The Moral Pointof View", p. 71.15 "Free Action"Journal of Philosophy v. 57 (1975), pp. 205-20. Note thatWatson himselfdoes not try to make use of this point in order to account for how it is possible to actfreely against one's best judgment. See his "Skepticism about Weakness of the Will"PhilosophicalReview 86 (1977).

11/38
On the other hand, this discussion will also show why a non-scholasticanalysisof these cases, such as Watson's, is inadequate.Watson argues that there is no sense in which we value the objects ofcertain desires we have, or judge the objects of these desires to be good."6Watson gives us the example of a squashplayer who has a sudden urge tosmash a racket on her opponent's head in frustration.It would indeed behighly misleading to say that this person sees some value in smashing theracket on her opponent's head, but that this value is overriddenby othervalues. The agent sees no value in this action, not even a smidgen of value.Although this point is undeniable, t should not lead us to go too far towardsevering the ties between motivation and evaluation. This example onlyshows that such an agent, rom a reflective perspective, does not attributeanyvalue to such a desire, indeed, not even a smidgen of value. This is not anobjection,however, to the claim thatdesires arealways claims to or perspec-tives on the good, and thusjudgmentsof the good from a certain perspective.In the above case, the reflective judgment takes the claim to be merelyillusory. Similarly, in the realm of theoretical reason we can distinguishbetween a consideration hatappears to make a claimplausibleanda consid-eration that in fact lends some plausibility to a claim. A considerationpertaining o the formercategorywill not necessarilypertain o the latter.Forinstance, one often finds appealingthe form of reasoning which urges thatone is less likely to win the state lottery if one simply chooses the samenumbersthat were drawnin the previousweek, given thatit is very unlikelythat the same numberwill be the winning numbertwice in a row. But thegambler's fallacy does not yield any plausibility to a belief, not even asmidgen of plausibility, despite the fact that it is quite hard to free oneselffrom the illusion of validity. Suppose we want to spell out the conceptualconnection between our being inclined to believe a certainjudgment, and anotion such as validity or truth.The above example shows thatone might beinclined to believe' a form of reasoningthat one (reflectively) judges to bewholly invalid.But one should not thereforeconcludethat there is no concep-tual connectionin this area;we merely placedit at the wronglocation. For itis only insofaras I take the gambler'sfallacy to appear to be valid that I aminclinedto believe it. The same goes for the conceptualconnectionbetween
16 In a later paper, "Free Action and Free Will" (Mind 96, 1987), Watson distinguishesbetween valuinga certainthing andjudging it to be good. The distinction does not appearin "FreeAgency" and I doubtits intelligibility, so I will just use the two expressions inter-changeably.
17 Belief and desire are often takento be analogousterms.But this is not true, since 'belief'is an 'all-out' term (I cannot, insofar as I am rational, believe contradictorystatements),whereas 'desire' isn't (I can desire incompatible states-of-affairs, without failure ofrationality). The closest practical counterpart of belief is intention, and the closesttheoretical counterparto desireis an inclination to believe something.
884 SERGIOTENENBAUM

12/38
desiring andjudgingto be good. It is only insofar as something appears o begood that one desires it.This point is well illustrated by another of Watson's examples. Watsoncompares the man who gives up sex because he "decides that the most
fulfilling life for him will be one of abstinence", and the man who "thinkshis sexual inclinations are the work of the devil".18Watson thinks that tounderstand his difference we must take this second man to be motivated byhis sexual inclinations while not seeing any value in them. This might betrue, butthis does not mean that we cannotunderstand is desire as a perspec-tive on the good, even if he deems this perspective illusory. This is to saythat agent has the perspective (his sensuous nature provides him with thisperspectiveon the good), butdoes not endorseit. Compare his with a personwho sees the two lines in the setup of the Mtiller-Lyer llusion, but knowsthey are of the same size. This person is still, in a sense, under a perceptualillusion. The drawings still appear to her as of different sizes, but she knowsbetterthan to endorsetheperspectiveaffordedby this illusion.
The moderatecelibate, on the otherhand,does not consider he perspectiveof his sexual desire illusory. Rather,he takes it to provide him with insightinto a form of value which is overriddenby other values. Watson's descrip-tion, in fact, fails to capture he predicamentof the devil-fearingcelibate.ForWatson thinks thatan actiondone againstour evaluationis not free. But thetrue work of the devil is to inducea humanbeing into sin of her own accord.Because compulsion isn't sin, Watson's devil does no better than the likes ofStalin. The real devil lets the celibate be the agentof his own evil; he makesthe celibate take something evil to be good.I do not wish to deny that motivation and evaluation might just comecompletely apart.But when this happenswe seem to face a case of compul-sion. If we take whathappens o an addict,or a kleptomaniac,whose desiretohave one more shot of vodka or to steal is one that he wishes he could actagainst,we do have a case in which theagent'sjudgmentsand evaluationsarecompletely powerless. But these cases are differentfromthe case of akrasia,exactly because of the way in which themotivation of the agent bypassesherjudgment, makingher unfree. It is importantnot to misdescribethis case as acase in which the agent's desire did not involve a judgment, as if it did notinvolve perceiving the action she performs as good (or desirable) from acertainperspective.When motivationand evaluationcome apart, he action isnot consequent, causally or inferentially, upon this judgment. Rather, theaction bypasses thisjudgment,and the agentdoes not act freely or responsi-bly. But if our views on akrasia are motivatedby the desire to preserve anordinaryunderstanding f thisphenomenon,we needto explain why we holdthe akraticagentto bejust as (or almostas) responsibleas the agentwho acts18 Free Agency, p. 210.

13/38
accordingto herbetter udgment,andwhy we do not make excuses forhim inthe sameway that we make excuses for agentswho act undercompulsion.Alfred Mele, for instance,tries to explainthe phenomenonof akrasia as acase in which the strengthof motivations does not correspond o the strengthof evaluations; that which the agents judges best is not what the agent ismost motivated to do.19But he says that this explanation does not reduceakrasia to compulsion since it does not rule out the possibility that theakratic agent is still capable of exercising self-control. The agent mightengage in a certain action that will make the strengthof motivations equalthe strengthof evaluation.However, this seems to pushthe problemone stepfurther.We now need to ask why the akraticagent failed to engage in self-control, andthe same questions aboutweakness of the will should resurface.In order to see that this appeal to self-control does not work, it is worthexamining Mele's view in more detail, especially his attemptto answer asimilarcriticismraisedby Watson.2"According to Watson, if an agent chooses freely not to exercise self-control, the agent must have changed his judgment. That is, if the agentcould have exercised self-control in such a way that he would choose A overB, and yet decides not to do it, then it could no longer be true that the agentstill holds A to be betterthan B. It is exactly this point thatMele contests:Perhapsto choose not to implementa choice would be to abandonthe choice; but it does notfollow that one who chooses not to exercise self-controlin supportof one's better udgmentnolonger holds thatjudgment. We may, without obvious contradiction,describe a case in whichan agent judges that all thingsconsidered t is betterto do A thanB, but due in partto his takinghis reasonsfor doingA to be only slightly moreweighty thanhis reason for doing B, decides toindulge himself and to refrain from exercising self-control in supportof A. In such a case, theagent may think his doing B to be permissible, even though he judges A to be better;and hemay self-indulgently opt for the lesser alternative.21
According to Mele, Watson has failed to show that the person whochooses not to exercise self-control in a way that makes it possible for her tochoose A over B has thereby given up herjudgmentthatA is betterthan B.Mele describes a case in which an agent thinks that his moralqualms aboutstrip-teaseclubs providehim with a "slightlybetterreason,all things consid-ered, not to enter the clubs thanto enter them",22 ut he also experiences acertaindesire to enter one of the clubs. He has a "desire-eradicating evice"thathe could use to do away with this desire, thusmakingsurethat he would19 See his Irrationality (New York: Oxford University Press, 1987). See also John Heil's"MindsDivided"(Mind98, 1989).20 For Watson's version of the criticism, see "Skepticism about Weakness of the Will".Mele's response can be found in "Is Akratic Action Unfree?" (Philosophy and Phe-nomenologicalResearch,XLVI, 1986), andIrrationality,ch. 2.2.21 Irrationality,p. 28.22 Irrationality,p. 28.
886 SERGIOTENENBAUM

14/38
stay away fromthe clubs. According to Mele, however, there is nothing self-contradictory about imagining that the agent will choose not to use hisdesire-eradicatingdevice, while still judging that,all things considered, it isbetter not to enter the club. It is clear why Mele chooses a case in whichtheagent thinks there is only a 'slightly betterreason' to takea certain course ofaction, and that he stipulates that the agent will think that "his doing B ispermissible".It is much harder o make the case if we assumethatthe agentthoughtthatA was overwhelmingly better thanB, or thatB was not permis-sible.23 I will first examine why Mele will face problems if he tries toprovide an example of this kind. This will help us understandwhy Mele'sresponseto Watsonis unsatisfactory ven in the case he describes.Let us assume that Joe lives in a dictatorship,and the dictatorsare afterhis sister who is the head of an underground uerrillagroup,and Joe knowsthatthey will show no mercytowardshis sister.The governmentoffers Joe amillion dollars to turnher in andhe refuses, because he finds it abominableto trade his sister's life for a few bourgeois comforts. The governmentexplains to Joe thatthe offer will still be standingfor 48 hours,and thattheywill bring to his home tonight a suitcase full of cash, hoping he'll havechangedhis mindby then.They'll open it in front of him, grinandwait;andJoe knows he cannotresist such a vivid displayof hardcurrency.Fortunately,Joe can avoid facing this irresistibletemptationby just puttinga sign on thedoor thatsays: "Joedoes not live here anymore".Not being very bright,thegovernmentagentswill turnaroundand nevercome back.Could Joe just choose not to put a sign on the door, and yet retain hisjudgmentthathe should not turn n his sister?Now it seems that the answeris 'no', and thatis the reason why Mele did not choose an example like this.If we rest satisfiedwith this point,we would alreadyend up in a positionthatMele's view has to give up a lot. For typicalcases of akrasia do not involveour choosing against our judgment that something is "slightly better",or23 It is not clear what 'worse butpermissible' could mean in this context. We can say thatan option A is morally permissible,butworse thanB, from a moral point of view. But thisis tightly connected with the possibility of supererogatory actions. The concept ofsupererogations difficult enough in the realm of ethics (for doubts aboutits applicability,see Shelly Kagan, "Does ConsequentialismDemandToo Much?",Philosophyand PublicAffairs, 13, 1984, pp. 239-54). It is even harder o understandhow an action can be betterall things considered yet supererogatory all things considered?).To say thatB is morallybetter than A, but that A is permissible, implies that doing A is not immoral. Mutatis
mutandis,we would expect that the akraticaction would not be irrational.In what sense,then, are we still speakingof akrasia? In what sense, can we then say that option A wasbetter than B, all things considered,if it is not irrational o choose B? Perhaps, althoughchoosing B is not irrational,it is more rational to choose A. But Mele surely owes anexplanationof what it meansto say thatchoosing A is more rational hanchoosing B, andyet it is not irrationalto choose B. At any rate, since Mele gives no explanation aboutwhat it means to say that an option is permissible n this context, I will assume that it addsno content to the claim that anoptionis only slightlybetterthan another.

15/38
choosing to do something that we consider to be in any interesting way"permissible". n this case, Mele would have shown that only a small subsetof akraticactionsare free andthis would alreadyamountto giving up a lot ofthe phenomenon. However, the right answer is 'yes', but insteadof makingthings better for Mele's position, this fact just helps us illuminate why thisreply to Watson is unsatisfactory ven in the case Mele describes.
There is a clear reason why it is hardto imagine that Joe could fail tochoose to exercise self-control andnot change his judgment.For supposehestill makes this judgment, and he also judges that he cannot give in to hisdesire for cash without exercising this form of self-control. Of course, hemight think that the exercise of self-controlis painful enoughthat it changeshis assessmentof whatis betteror worse. If he thinksthat a sign on the dooris aestheticallyso repulsive that it's not worthposting it in orderto save hissister's life, then the exercise of self-control does not leave his judgmentintact; reconsideringwhat is neededin order to save his sister, he no longerthinks that saving her life is better thanaccepting a million dollars. But aslong as Joe is minimally logically competent,24and that he is otherwiseindifferent about posting or not posting the sign on the door, he must con-clude fromthe fact thatit is betterto save his sister's life than to accept thatmoney, that it is betterto post the sign than not to post the sign. So we cansafely conclude that,as long as Joe is not overcomeby tremendousstupidity,he cannotmaintainthe originaljudgmentthatA is betterthanB (that savinghis sister's life is better than accepting the money) and also judge that it isbetter not to exercise self-control (that not posting the sign is better thanposting it). However,he can fail to exercise self-control while still maintain-ing his original judgment. We can imagine that as Joe is about to post thesign he starts maginingall the wonderfulcars he could drive, the wonderfulplaces he could visit, the wonderful house he could buy if he had a milliondollars-none of which, of course, compensatesfor the loss of his sister.Wecan imagine that despite not changing his reflectivejudgment he falls intotemptationand does notpost the sign.In this case, however, Joe exhibits another instance of akraticbehavior.Since Joe acceptsthejudgment that,all things considered, t is betterto exer-cise self-control, if he doesn't do it, he will be acting against his bestjudgment.And it should now be clear thatthe possibility of exercising self-control could not, in this case, help us explaininghow the akraticagent wasfree. If the answer s "becausehe was free to exercise self-control", t must betruethat when the agentdoes act akratically,when he does notexercise self-control,he was still free to exercise self-control. But if his failureto exercise24 If he is not, we have a differentkind of problem,similar to the case in which Joe does notrealize that exercising self-control would let him act accordingto his best judgment. I'lldiscuss this case below.
888 SERGIOTENENBAUM

16/38
self-control itself involves akrasia, we cannot assume that he was free toexercise self-controlunless we explain how akrasia can be freeaction.The same analysis applies to the case Mele discusses, and we should notget confused by the fact that the agent judges the reasonto be only "slightlybetter". If he thinks that not going to the striptease show is better thangoing, and he is otherwise indifferent about using the desire-eradicatingdevice, as long as he maintainshis originaljudgment andis minimally logi-cally competent,he must conclude that it is better to use the desire-eradicat-ing device (given thathe judges it to be a necessary, painless means to theoption he judges to be better). Of course, he might still not use it, but thiswould be a case of going againsthis best judgment, since he must judge thatall things considered,it is betterto use the device. As we said above, how-ever, if his failure to exercise self-controlis due to akrasia, it cannot help usunderstand ow akraticactioncan be free.
Mele also argues that the agent who failed to exercise self-controlcan stillbe consideredfree if she overlookedthatit was possible to exercise self-con-trol or if she misjudgedthe amountof self-controlnecessary, and thus did notrealize that it was in herpower to exercise self-control. So, for instance, sup-pose Joe thoughtthat he did not need to post the sign on the door.Rather,hethoughtit would be enoughto put on his sunglasses to avoid the temptation,or he just overlooked the possibility of posting the sign. Nonetheless, it isstill true in this case that it was within Joe's power "toresist acting on thisdesire",and so "thedesire did not compel him, in which case [his takingthemoney] was a freeaction."25
It is importantto note, first, that on this account, again, only a smallsubset of cases of akrasia will turnout to be free actions. In most cases ofakrasia,we cannot attributeailure to act on one's bestjudgmentto overlook-ing a certaintechnique or self-controlormisjudgingthe amountof resistancenecessary. Moreover, is it true that the akraticagent was not compelled bythe desire in these cases? It might be useful to comparethe possibility thatMele envisages with the case of someone who is held at gunpointbut doesnot realize that he can overpower his aggressor. It seems that this agentshould not be considered as acting free from external compulsion, justbecause,unbeknownst o him, he could have successfully resisted his aggres-sor. In the same way, the agent who does not realize that she can exerciseself-controlshouldnot be seen as actingfree from internalcompulsion.Is an agent really free to do X if he can bring aboutX, but he does notknow that he can bringit about?Am I free to escape fromjail, if the door isopen but I don't realize that it is open? These are difficult questions toanswer, but to accept Mele's suggestion that the akraticagent was free toresist her desires we need to answerboth affirmatively.AlthoughI think the25 Irrationality,p. 22.
THEJUDGMENTOF A WEAKWILL 889

17/38
agent is not free in these cases, I do not wish to settle this issue here. I wishonly to point out that even if there is a sense of 'freedom' that let us answerthese questions affirmatively, Mele's suggestion is not satisfactory. For,first, this sense of freedomdoes not allow us to say that the agent is respon-sible or blameworthy or acting akratically.At least we generallydo not holdpeople responsible or blameworthy for not bringing about X if they areignorantthatthey can bring about X.26 f I could have saved someone's lifeby flippinga switch, butI didn't realizethatI could have saved him this wayI am not responsiblefor his death.
Secondly, akrasia in this case would not be irrational.For one cannot beaccused of irrationality or not doing what she did not realize she could do.27So an akraticagent who overlooks the possibility of exercising self-controlcannot be considered rrational n this account.Butonce she failed to exerciseself-control,it was no longer upto her to follow herbest judgment.
Finally, as Watson also points out, even if we think thatthe agent is freeto act according to his best judgment in this case, what prevents the agentfrom acting according to his best judgment while he is free to do so is hisignorance, not akrasia. The failure of the agent, insofar as he is free, isignorance (or misjudgment)not akrasia. Mele tries answering this objectionas follows:It may be true both thatan agentwould not have doneA if he had seen what he needed to do tomount a successful resistance against his desire to do A and that, if he had been strongerormore resolute, a special effort of resistance would have been unnecessary.28Mele is suggesting that akrasia is at least one partof the explanationof whythe agent did not do A when he was free to do A, since had he been strong-willed he would have doneA. But the failureto do A whenone was free to doA cannot be attributed o akrasia under this view. On this view, akrasia26 Unless, of course, the ignorance itself is culpable; if, for instance, the agent chose toengage in certain actions that would bringabout his ignorance.In this case we would runto the same kinds of problemat a different evel; we would have to explain why the agentdecided to engage in these actions that cause him to be ignorant of the possibility ofexercising self-control in orderto bring about X. If the agent decided to engage in theseactions, because he thought that the cost of knowing how to exercise self-control madebringing about X no longer an attractiveoption, the agent did not behave akratically-since, in this case, he thoughtthatall thingsconsidered,X was not the best option.On theotherhand,if the agent thoughtthat all things considered,it was betterto engage in theseactions that would result in his knowing that he could (how he could) exercise self-con-
trol, but didn't, then his failure to engage in these actionswas an instanceof akrasia. Thiscan be considered blameworthyonly if we have an account that makes akrasia blame-worthy,but this is what is in question.Fora recentdiscussionof the difficulties involved inthe notion of "culpable gnorance",see Michael Zimmerman,"MoralResponsibility andIgnorance",Ethics 107, 1997, pp.410-26.27 Again, unless not realizing that she could exercise self-control was itself irrational.Andthe same that was said in the note above applieshere,mutatismutandis.28 Itrationality,p. 27.
890 SERGIOTENENBAUM

18/38
explains why there weren't other options open to the agent, not why theagent failed to choose an option thatwas open to him (or thathe was free topursue).The problem of akrasia is not that we are driven by desireswhose objectswe do not value-this is merely compulsive action-but rather hat we desireand positively evaluate that which we, under our own implicit or explicitrecognition, ought not to so desireor evaluate. Thus,akrasia should be seennot as a collision between motivationandevaluation,but betweentwo kindsof evaluation.
We have seen how to understand he conceptual tie betweenmotivationand evaluation,and thus how to understand he scholasticformula.I arguedthat to desire something is to have a perspectivefrom which it appearsto begood (even if we reject this perspective upon reflection). We have found,however, that Davidson's attemptto provide an account of weakness of thewill while accepting the scholastic formula was unsatisfactory. It left uswithout an understandingof how an agent could be prone to the kind ofmistake typical of akratic behavior. We can clarify this point by taking alook at a suggestion by JohnMcDowell. Accordingto McDowell, the akraticagent makes the correct judgments about the good, but his grasp of thecontent of these judgments is defective. Let us look again at the akraticorthodoxJewish smoker.He judges, just as the virtuousperson does, that allthings considered, he should not smoke on Friday evenings; religiouscommands, after all, are more important han mere inclinations.What kind ofgraspof the content of thejudgmentlike 'All things considered,I should notsmoke' could explain why the akratic agent stops short of acting in theappropriatemanner?29Whatpartof the phrase 'all things considered' did theagentfail to understand?Whatwe need to explain is how the structureof ourpracticaldeliberation makes us prone to form unconditional udgments thatdisregardour most considered,reflectivejudgments-a kind of mistake thaton the face of it seems incomprehensible.We have also seen why the rejection of the scholastic formula leaves uswithout a satisfactory accountof akrasia. Now, if we think that akrasia is aphenomenon which has no parallel in the realm of theoretical reason, it isnatural o look at uniquefeatures of practicalreason to account for akrasia.Thus, it might be tempting to look for the source of akrasia in a sortofmotivationalbreakdown, because there is no similar form of breakdownin29 I am also changing McDowell's terminology to make it consistent with the rest of thepaper. See his "Comments on Irwin's 'Some Rational Aspects of Incontinence"'Southern Journal of Philosophy, 1988, suppl., pp. 89-102; esp. section V. Of course,McDowell also tries to explain the possibility of a defective graspof the content, but it isunclear that McDowell's characterizationof the failure of the akraticagent is compat-ible with a scholasticview. See, for instance,his claim that "Theincontinent'sproblem israther that practical thinking with this sort of content is not (perhaps not yet) fullyingrained nto his motivationalmakeup" p. 97).
THEJUDGMENTOF A WEAKWILL 891

19/38
theoreticalreason.In the next sectionI want to examineDescartes's views onthe relationshipbetween the will andthe intellect, and how they show us thatthere is a theoretical counterpart o the phenomenonof akrasia. I will alsoarguethat we can find in Descartes's work the resourcesto accountfor thispossibility. In the following section I will show how to extend this accountto practical eason-to themore oftendiscussedcases of akrasia.
But before we move on, I must clear up a possible source of misunder-standing.One might think thatby adoptingDescartes's view of the relationbetween the will and the intellect, I have adopteda highly controversial ormof voluntarism: he view that we can change our beliefs as we please, or thatwe can decide to believe whateverpropositionwe want to believe. Or in aless extreme version that in at least some significantcases, we can come tobelieve a propositionsolely by wantingto believe it. Although some peoplemight want to defend some version of this kind of voluntarism,3"1do notthink Descartes is committed to it.3"Descartes does think that affirming,denying and doubtingare modes of the will, and so thatbelieving involvesthe will. But this identifies believing with a form of activity, and in particu-lar free activity, but not necessarily with the ability of believing whateverone wants to believe.
Althoughthe natureof this activitywill be clearerbelow, it mightbe use-ful to put fortha word of warningagainsta confusion that could temptus toascribeto Descartesthis controversial orm of voluntarism.If we takejudg-ment to be the main activity of the will, we can distinguishbetweenjudgingto be true andjudging to be good. From what we said above, it should beclear that it is naturalto identify thatwhich we judge to be good with thatwhich we want. So it would be natural o equateour freedomto act in accor-dancewith what we judge to be good-a capacityof self-determination-as afreedom 'to do whatever we want'. But it would be misleadingto think thatwe must be ableto characterize ur freedomto act in accordancewith what wejudge to be true also as freedom'to do whateverwe want',sincejudgingto betrue is not conceptuallyconnectedto wanting.If we want to draw a parallelwith this expression in the realm of theoretical freedom we must say thatfreedom to act in accordanceto what we judge to be trueis our freedom toaccept whateverseems correct to us, not that this freedomis the freedom tobelieve whateverwe please.3230 As far as I know, no one claims that, for any proposition,no matterhow absurd t is, one
can affirm it or deny it at will. For some (less extreme) versions of voluntarism, seeBarbaraWinters,"Willingto Believe" Journal of Philosophy,LXXVI, 1979, pp. 225-43and Bas van Fraassen,"Belief and the Will" Journal of Philosophy,LXXXI, 1984, pp.235-56.31 For a different view, see Harry Frankfurt, Demons, Dreamers and Madmen(Indianapolis:Bobbs-Merrill,1970).32 One mightthinkthat this way of defendingDescartesfrom a controversial ormof volun-tarism could not account for passages in which he characterizes our freedom as
892 SERGIOTENENBAUM

20/38
IIIThe possibility of theoreticalakrasia comes up most clearly in Descartes'sMeditations when we try to carrythe skeptical essons of the FirstMeditationinto the later Meditations. At the end of the First Meditation, Descartesexpresses theproblem as follows:My habitualopinions keep coming back, anddespite my wishes, they capturemy belief, whichis as it were bound overto them as a resultof long occupation and the law of custom.33
This is not very far from how an akraticagent could describe his failedattempts to follow his reflectivejudgments. But whatmakes those "habitualopinions" come back, flouting the skeptical argumentsof the First Medita-tion? Why can't his "wish" be easily fulfilled in light of the doubts raisedagainsthis habitualopinions? In otherwords, why aren'tthe skepticalargu-mentsenough to make surethat old habits will notcapturehis belief?I think the answers to these questionscan be betterunderstoodwhen welook at a distinctionimplicit in Descartes's reexaminationof these doubtsinthe Third Meditation. When Descartes conceives the possibility of radicaldoubt, he notes thatwheneverhe contemplatesthoughtssuch as '2 + 3 = 5'he cannotfail to assent to them. As he says:
when I turnto the things themselves which I think I perceive very clearly, I am so convincedby them thatI spontaneouslydeclare:let whoever can do so deceive me; he will never bring itaboutthatI am nothingso long as I continue to think I am something;(...) or bringit about thattwo and threeaddedtogetheraremore or less thanfive, or anything of this kind in which I seea manifestcontradiction.34
The possibility of skeptical doubt depends upon the possibility of not"turning o the things themselves";thatis, not turningmy attentionto thosetruths which I cannot doubt when I contemplatethem, while still retainingthem in the mindsomehow. I takeit thatDescartesis heredrawinga distinc-tion between oblique and directthought.There is a way in which I conceive
"freedom of indifference". The exercise of this kind of freedom does not seem todepend on judging something to be true or good. But, since affirmingand denying arealways forms of judging, the difference between freedom of indifference and freedomconsidered merely as spontaneityis a difference between the grounds of ourjudgmentsof the good and true, not a difference about whether we makejudgments about the trueor good (since our freedom is always expressedin the activity of judging). I believe thatthe clearest discussion of these issues in Descartes's work is to be found in a letter toMesland in The Philosophical Writingsof Descartes, vol. III, trans.by Cottingham, J.,Stoothoff, R., Murdoch, D., and Kenny, A. (New York: Cambridge University Press,1991, henceforth "CSMK"), pp. 244-46, and in Oeuvres de Descartes (henceforth"AT"),edited by Ch. Adam andP. Tannery(Paris:Vrin/C.N.R.S., 1964-76), vol. IV, pp.173-75.33 ThePhilosophical Writingsof Descartes, trans.by Cottingham,J., Stoothoff,R., and Mur-doch, D., (New York:CambridgeUniversity Press, 1987; henceforth"CSM"),vol. II, p.15; AT VII, 22.34 CSM II 25; AT VII, 36.
THEJUDGMENTOFA WEAKWILL 893

21/38
of two plus threeequalingfive such thatthis thought presentsitself to me asimpervious to doubt,a thoughtin which I grasp not only that two plus threeequals five but how it is so. I will call the former an oblique cognition andthe lattera directcognition.35We can say, roughly,a directcognition is a representation f an object orclaim throughwhich one clearlyunderstandsorseems to understand)why theobject is as one represents t, or how it is that this claim is true.36An obliquecognition is a representationof a claim or an object which is not a directcognition, but one throughwhich one understands or seems to understand)that there arereasons to accept that the object is as one represents t, or thatthe claim is true.We can think of an oblique cognition as standing proxy toone or more direct cognitions. The direct cognitions carries with it theexplicit justification of that which the cognition representsas true. So, amathematicalproof of a theorem s a directcognitionof this theorem, since itnot only presents the theoremas true,but also presentsthe reasons for hold-ing this theorem to be true. But if I later remember having proven thistheorem to be true, without rememberingthe proof itself, my cognition isoblique. It representsnot the reason for holding the theorem to be true, butthatthere is a reasonto hold this theoremto be true,a reason that this cogni-tion does not make available to me. Of course, there is a sense in which mymemory is itself a reason for my acceptingthe truthof the theorem, but it isa reasonfor accepting the theorem,not a reasonwhythe theoremis true; mymemory could not be part of the proofof the theorem.In the same way, if Itrustthejudgmentof reputablephysicists on mattersof quantummechanics,my cognitions of the matter are oblique; they stand proxy not to a directcognition that I have ever had available to me, but they nonetheless standproxy to a directcognition.
In both direct and oblique cognitions we deal with the same proposition,but we conceive of this propositionin differentways. We can also see thatthis distinction is important in motivating Descartes's intuitionism.Descartes insists that we shouldtry, as much as we can, to entertainseveralsteps of a derivationat once.37This insistence can be seen as manifestingnota bizarremistrust n the powersof memorybutrathera realization hat this iswhat true understandingpartlyconsists in. The mathematicianwho does notsee the whole proof of a propositionwithin a single intuition knows that it
35 I do not intend to use the word 'cognition' here as an achievementnoun. So it does notfollow from the fact that I have a cognition that presents me an object as X on thegroundsthat Y that the object is in factX or thatYconstitutes good groundsfor acceptingthat the objectis X. My cognitionmightbe in some way delusory.36 This does not mean to say that one can say anythingmore than "I see it " to justify whythe object is as she represents t.37 See Rule 11 of "Rulesfor the Direction of the Mind"(CSM I, 37; AT X, 407).
894 SERGIOTENENBAUM

22/38
is so and even that it must be so but does not know how it is so andhow itmust be so.
Oblique cognitions seem to be especially importantin reflective judg-ments, since reflective udgmentsmighthave to weigh considerations hat arenot clearly commensurate.By looking at two objects at more or less thesame distance, I can determine, n most cases, which object is larger ust bylooking at them. But the situation s more complex, if I have to examine, forinstance, incompatibleclaims of my sensorycognitions of an object and mymore theoreticalunderstanding f a physical object. Suppose,for instance,Isee what seems to be an object floatingin the air in flagrantviolation of thelaws of gravitation.When one asks whetherthe claims of our theoreticalunderstandingof physical objects should be valid or whether the claims ofsensory perceptions should be valid, one is notjust struck by each of thesecognitions and put in the position of waiting to see which of them wouldincline the will one way or the other, or by just bringingthese different deasto one mind. I have to consider how I shouldtake each idea, when and how Ishould take my sensory perceptionsat face value, the reliabilityof my clearand distinct perception (or my theoretical beliefs about the nature of theobject), and so forth.In orderto settle the competingclaims of those cogni-tions, we need to rely on obliquecognitions, on cognitions such as: 'the factthatI have a sensory cognition of the state of a physical object is a reliablebut not infallible guide to the state of this object (it should count as someevidence to the claim that the object is in this state, but not as conclusiveevidence)'. In cases of conflictof cognitionsfrom differentcognitive sources,our reasoning involves not only the presentationof these cognitions to themind, but primarilythe representationof thefact that we are in the posses-sion of these cognitions, and of our reflective views on the relative weightthat we should accord to these cognitions. We can now reserve the term'reflective cognition' only to those cognitions that evaluate the relativeweight of a certaincognition, such as the following: 'the fact that I have asensory cognition of the state of a physical object is a reliablebut not infal-lible guide to the state of this object (it should countas some evidence to theclaim that the object is in this state, but not as conclusive evidence)'. I willcall a 'primarycognition' any cognition that is not reflective. All reflectivecognitions are oblique, but not all oblique cognitions are reflective. We cancall a 'reflectivejudgment' a judgment in which one attemptsto settle theincompatibleclaims of different sources of cognition. A reflectivejudgmentwill be based on reflective cognition, and thus on oblique cognitions. Notethatthis is not a claim aboutempirical psychology or phenomenology,but aclaim about what mustbe involvedin ourcapacityto adjudicate he claims ofincompatiblecognitions. Since each cognition presents the object as corre-sponding to this cognition, we cannot settle this question through, so to

23/38
speak, the eyes of the cognitions themselves,but only as we reflect uponthesignificanceof havingthese cognitions.We arenow in a betterpositionto understandDescartes' separationof thewill from the intellect in a way thatdoes not commit Descartes to viewingthe will as an arbitrarycapacity to pick and choose its favorite ideas asobjects of belief. It has often been pointed out that Descartes thinks that Icannot fail to assent to the fact that a trianglehas three sides.38Why doesDescartesthen insist on separating he will fromthe intellect, in distinguish-ing my assent to the idea from my understandingof it? One could limit theactivityof the will to the cases when we do not have a clear anddistinctidea,but this would fly in the face of Descartes's assertion thatwe are most freewhen we assertthatwhich we clearlyand distinctlyperceive.39
This separationwould be indeed unnecessaryhad we no ideas that, in animportant ense, compete-ideas that arealternativepresentations r apparentpresentationsof the same objector fact. The idea that we have of a corporealobject when we arewalking aroundseeing corporealbodies is very differentfrom the one that we get from learning physics. The first presents it ascolored and continuouslysolid in a way thatthe latterdoes not. Descartes'sown exampleis the two ideas we may form of the sun, one fromsensory per-ception and the other from physics. But it is important o conceive them asdifferentappearances f the samething,so thattheycan be seen as deliveringtwo incompatibleviews of the same object. Moreover, we could not under-standthem as presentationsof the sun if in some way they did not appear ous as (accurate)representations f the sun;or, in Descartes'swords,as if theywere images of things.40However, unless we want to thinkabouta subjectin ratherschizophrenicterms, we have to make room for the fact that the subject, at least in mostcases, adjudicates between competing representations-a person whosethoughtspresentincompatibleviews of the same objectshas to decide whichthoughtis the one that she shouldacceptandwhich she shouldsee as merelyanappearanceof a certaincontentbeingaffirmed ather hana true affirmationof the content. Withouta conceptof the will, we have to conceive of the sub-ject as thinking that the sun is smaller than the earthin his daily businessandas havinga totallydifferent dea of the sun whenhe is engagedin scienceor reflective thought. If it were not for the fact that the subject makesjudgments-i.e., the fact that a subject affirms some ideas and deniesothers- the competingclaims of these ideas would simply coexist. In orderto be disabusedof the daily illusions the subjectmustbe capableof bringing38 See, for instance, John Cottingham"Spinoza's critique of Descartes", Journal of theHistoryof Philosophy26:2, 1988. But see letter to Mesland (CSMK,244-46; AT IV, 173-75) for an importantqualificationof this claim.39 See CSM II, 40; AT VII 57-58.40 CSM II, 25; AT VII, 37.
896 SERGIOTENENBAUM

24/38
these two ideas togetherand of taking one to be that which provides her withthe correct representation of the sun. The activity of the will is then thisactivity of adjudication.Contrary o whatone might think,the will turns outto be a truly cognitive faculty. In a sense it is the cognitive facultypar excel-lence since it includes the capacity to weigh thedifferentclaims of differentideas of the same objectof cognition.
Under this picturewe get a conceptionof the will as evaluatingcompetingrepresentationsof an object; the will turnsout to be a faculty of the agentclosely connected to our ability to consider him rational.It is because thewill can adjudicatebetween the claims made by differentideas of the sameobject thatit can be considereda rationalcapacity, throughwhich inconsis-tency and incoherencecan be avoided. One canthinkaboutthe Cartesianwillas performing wo differenttasks. The first s to bring to the gaze of the mindprimary cognitions of ideas of the same object. The will has to be able torepresentthe claim made by, for instance,the idea of the wax given by sen-sory perception,and the one madeby the intellect.But the will, of course, isnot merely a reflective capacity but an adjudicatingone also. If it is perform-ing its job correctly, it will affirm only the clear and distinct perceptions ofthe intellect (at least when we are engagedin the search fortruth).In the caseof intuitive judgmentthis could not fail. The clarity and distinctness of theprimary cognition suffice to guarantee assent. But if, as a result of anabstruse derivation, the will has to rely upon oblique cognitions, the willmight end up making the wrongjudgment, given thatthe reflectivecognitionwill inheritthe obliqueness of the primaryone.
If, when going through the reasoning of the First Meditation all that Icarryfrom it is the thought that I had reasons for doubtingthe existence ofthe externalworld, I might not withholdmy judgment accordingly (thoughImight accept that I ought to withholdjudgment). If my will compares thismere recollection with the ideas of sensory perception-which provide me,atleast apparently,with primarycognitions that representthis world-I mightfind the latterconvincingwhile still agreeingthatI oughtto be convinced bythe former (and this might in practice be the fate of every skeptic). If thishappens,we have a case of theoreticalakrasia.This bringsus back to a pointI made earlier.The more clearly and distinctlyI understandhe groundsfor acertain truth, the less am I prone to this sort of theoretical akrasia.Descartes's advice in Rule Eleven thatwe shouldkeep as muchof a proof inmind as possible is all the more importantf we realize that this clear under-standingof a propositioncan also preventus from slipping into this form oftheoretical akrasia. At any rate, it is important o note that the fate of theskeptic is not unlike the fate of the akraticagent. In the same way, we cansay thatthe orthodoxJewish smoker thinks thathe ought to be convinced byhis reflectivejudgment, but he is actually convinced by the judgment fromthe pointof view of pleasure.
THEJUDGMENTOFA WEAK WILL 897

25/38
This distinction between what I believe andjudge as opposed to what Ibelieve I ought to believe and judge, should not be seen as a distinctionbetweenthe way states of belief strike a helpless agent,and theway the agentwould like to find himself believing. Both kinds of beliefs express (freelymade) judgments.We shouldthinkof this case as the case of someone whohas been convincedby a philosophical argument,butwho, when turningherattention from the argument,cannot fail to be persuadedby the evidence ofher senses. It is no surprisethen thatwe resort to a device such as the evildemon to make this clear. In this way the upshotof the FirstMeditation canbe surmised in a direct cognition. And here again we can see that the evildemon has a functionsimilar to the various devices thathelp us exercise self-control. The evil demonperforms he samefunction for theMeditator hattheconspicuously displayed autopsy photos of the lungs of a heavy smokerperform for the persontryingto quit smoking.And heretoo perhapswe cansay that his reflectivejudgmentis based on a reflective cognition. As such,no matterhow certain it is, its certaintyis groundedon oblique cognitions.And if this is the case, the akraticagent, just like the skeptic, might end upadoptingthe unreflectivepointof view of her directcognitions.It is important o distinguishwhat I have been calling theoreticalakrasiafrom a much discussed phenomenon:self-deception. Mary gives Joe all theindications, shortof just saying it, that she considershim a bore. She yawnswhile he speaks, she avoids his companyat parties,and she makes no effortto feign interestin his anecdotes. Joe thinksvery highly of Mary's judgmentand it would be devastatingto him to believe that she finds him a bore; hecertainly wants to go on believing that she finds his company invaluable.Despite all evidence,Joe believes thatMarydoes not considerhim a bore.We can make the distinctionclearby pointing outthattheoreticalakrasia,as I understand it, is a shortcoming wholly within theoretical reason. Itsdescription nvolves no referenceto thedesires,values orpractical udgmentsof the agent.Self-deception nvolves the interferenceof practicalreason;Joe'sdesires aredetermining he formationof an irrationalbelief. One is not moti-vated in any way to find the gambler's fallacy persuasive; t would be quitesurprising to find out that one has a stake at that issue. Yet, one finds itpersuasiveeven if one acknowledgesthatone oughtnot to find it persuasive.
Depending on how one understands elf-deception, it will be a particularcase of theoretical akrasia or a distinctphenomenon.If one thinksthat theself-deceived agent accepts that he ought to form the contrarybelief (if onethinks,for instance, that in orderto be self-deceived Joe must accept thatheought to believe thatMaryconsidershim a bore), then self-deceptionwill bea particular ase of theoreticalakrasia. If one thinks,perhapsmoreplausibly,thatself-deception does not involve acceptingthatone should formthe con-trarybelief, thenit will turnout to be a phenomenondistinctfrom theoreticalakrasia.In either case self-deceptionwill raise difficult andinteresting ssues,
898 SERGIO TENENBAUM

26/38
but these issues are not the topic of this paper. My claim is that there is aphenomenon analogous to practical akrasia that can be described whollywithin the realm of theoreticalreason that can help us illuminate its betterknownpracticalcounterpart.41
IVWe can now see how this pairof distinctions-oblique and directcognition;primaryandreflectivecognition-can helpus providea scholasticaccountofweakness of the will. We can think about practicalreason in a similarway.We can thinkthatwhat makes ourpracticalsensibilities cognitive capacitiesis thatwe can see them as primarycognitions whose claims are evaluatedinreflective cognitions in ourexercise of practicalreason.This is to say thatadesire, an urge or any other form of motivation provides us with a putativeperspective of the good, which we may call a 'cognition of the good' to beevaluated uponreflection.Underthis understanding,an urge which is takeninto considerationwhen decidinghow to act or how to lead one's life, shouldnot be seen as a mere impulse but as a putativejudgment, a way of seeingthe worthof an action. Comparativeudgmentsin which we try to settle theincompatibleclaims of differentdesires can be seen as reflectivejudgments,in which we weigh the force of the claims madeby competingcognitions ofthegood.
Any judgment, for instance, which has to settle between claims to thegood from differentperspectivespresupposesa comparisonof obliquecogni-tions of the competing cognitions of the good, which are not immediatelycommensurable.42Since different desires might present an object as goodfromdifferentperspectives hey cannotbe evaluated hroughdirectcognitions;41 For the same reason, what I take to be theoreticalakrasia is distinct from what Meletakes to be akraticbelief, since he claims that "'motivatedness'of incontinentaction isone of its definingfeatures.This is trueas well of incontinentbelieving" (Irrationality,p.
112).42 By saying that two kindsof goods arenot commensurable, do not meanto say thattherecan be no reasonablegroundsto choose betweenthem. All that is meantis that we cannotfind a commonmeasurethatwould allow for a weak orderingof all goods of these kinds.Suppose I am deciding between how to spend some extra $1000 I have received. I couldeitherinvest the money in my retirementplan (R) or use it to contribute o the election ofa certain political candidate (E). Suppose I am indifferent (or find that I can't decide)between the two alternatives. Suppose now that my retirementplan announcesa promo-tion that will, uponmy request,add a $10 bonusto my $1000 if I invest in the plan (R*). Itis clear that I prefer R* over R even if I might still be indifferent(or unable to decide)between R* and E. So we get that(R 2 E), (E ? R*), and-'(R ? R*). For a similarnotionof incommensurability,see Joseph Raz, The Morality of Freedom (New York: OxfordUniversityPress, 1986), p. 232. See also on this issue ElizabethAnderson,Valuein Ethicsand Economics (Cambridge, Mass: HarvardUniversity Press, 1993), ch. 3. For anaccount of akrasia that relies heavily on a notion of incommensurablemodes of valua-tion, see, David Wiggins, "Weakness of Will, Commensurabilityand the Objects ofDeliberation", n A. Rorty,Essays on Aristotle's Ethics;Berkeley: University of Califor-nia Press, 1980).
THEJUDGMENTOFA WEAKWILL 899

27/38
thus, the ensuing reflective cognitions will inherit the obliqueness of theprimarycognitions upon which they are based. When I recall the thrill ofwatching a mystery movie, and I imagine myself reliving this experience, Ihave a cognitionin which watchingmysterymovies is presentedas desirable.However, when evaluating what should I do tonight, I have to weigh theclaim of this cognition against other incompatible claims. When reflectingabout this option, I weigh thatwatchinga movie is desirable because of thisthrill, but in doing so I do not necessarily see how the thrill makes it desir-able; I am no longerjudging what is good fromthe perspectiveof desires forthrills.My reflectivejudgmentheretoo is based on reflective, oblique, cogni-tions. The reflectivejudgment will be groundedon cognitions such as: 'thefact that I findX pleasantis a reliable butnot infallible guide to the fact thatX makes some positive contribution to my good'. As I recognize myreflective judgments as authoritative, I recognize that my unconditionaljudgmentsshould alwayscome frommy reflectiveperspective.But unless thereflectivecognition can do awaywith the naturalplausibilityof the otherper-spectives, here, just as in the case of theoreticalakrasia, the less clear mygraspis on the reflectivecognitions, the moreproneI will be to form uncon-ditionaljudgments thatdo not infact arise from my reflective perspective.Let us now examine this claim in moredetail.Let us return o our orthodoxsmoker. Hejudges that:
(i) From a certainpoint-of-view, smokingis betterthanrefraining romsmoking.(ii) All things considered, refraining from smoking is better thansmoking.
As we said above, Davidson correctly points out that (ii) is compatible withthe unconditionaljudgment that it is better to smoke. But we argued thatthoughDavidson's account of akrasia could makethese statementscompati-ble, it could notexplainthefrequencyof instancesof theirconjunction n freeaction.The root of this difficulty lies in Davidson's account being oblivious tothe fact thatwhen we move fromjudgmentsof the form of (i) to judgmentsof the form of (ii), something gets lost. In orderto make judgments of theform of (ii), we have to step back from the perspectives under whichjudg-ments of the form of (i) are made. If we lose sight of this, we will be boundto think that(ii) makes (i) irrelevant-at this point (i) could have no bearingon his action if the agentcan follow his judgment.After all, (ii) seems to doaway with any plausibilitythat (i) might have. But we can understandwhatgets lost in this move in light of the pairof distinctions we broughtforwardin our discussion of Descartes.
900 SERGIOTENENBAUM

28/38
A desireprovides the agent with a direct and primarycognition. An objectof a desireappears n this way to be good andworthyof being pursued.Whenreflecting upon such a desire, one notices that its realization is incompatiblewith more important ends. So the end of eating chocolate cake might beincompatible with keeping oneself healthy. Or a cognition of the good isdeemed illusory upon reflection.For instance,Waltermight see his desire toavenge himself againsthis rivalwho got the poetry prize he so muchcovetedas arising from ungroundedemotions. After all, to the extent that desire forrevenge makessense, it ought to be directedagainst those who wronged him,those whom he might legitimately resent. But Walter cannot take seriouslythe thought that someone who got a poetry prize he coveted indeed haswronged him. So these reflective cognitions present him with the view thatthe primary cognition is illusory or overridden. But the primarycognitionremains, andthe reflective cognition might not have the same immediacy orclaritythatthe firstcognitionhas.So in the same way thatconvincing myself thatthe sun is bigger thantheearth does not therebydo away with the way the sunlooks to me when I stepout of my house, the judgment that he ought not to try to avenge himselfdoes not do away with the original cognitionof the good. A certaincognitionof the good that I believe I ought to accept might fall short of immediatelyprovidingme with an all-out judgment in at least two ways. I might under-stand that I should pursue a certain course of action, but have very littleinsight as to why. This could happen f I were told by someone I trustthatacertaincourse of life is worthpursuing she explainsto me, for instance, howone feels much better when one stops eating animalproducts)or if I wereconvinced by complex considerationsthat I ought to become a vegetarian.But it might also fall short of immediately providing me with an all-outjudgmentif I can see why I ought to accept this cognition of the good, but Icannotclearly graspthe alternative ognitionsas illusory.
This is similar to a situationin which a reductio makes it clear to us thata certainpropositioncannot be true,but we do not understandwhatis wrongwith the reasons we took to support this proposition. Again, we can thinkabout the fate of skepticism at the end of the First Meditationin the sameterms. The remembranceof having good reasons to doubtall of one's beliefstells us thatthat which our senses convey to us oughtto be doubted. But theconviction thatthis was shown is an obliquecognition whereas the presenta-tion of objects by the senses is immediatelyclear. And if my understandingof this argumentis unclear and vague I might be unpersuadedby such anargumentwhile still retaining my conviction that I ought to be persuaded.Similarly, realizing that the gambler's fallacy is a poor form of reasoningdoes not necessarily dissolve the temptationto avoid bettingon last week'swinning number,andthis temptation s greater he less clearmy graspof thestructure of the fallacy is. So I will probablybe differently persuaded by
THE JUDGMENTOF A WEAK WILL 901

29/38
instancesof the gambler'sfallacy if I was just assuredby a friend that this isa fallacy, or if I havemerelya vagueunderstandingf how theprincipleguid-ing my choice is an instance of the gambler'sfallacy. And if this is the case,I might find myself saying things such as: "I understand hat this is a falla-cious reasoning, but it just does not seem possible that these numberswillcome up again".43Turningbackto the akraticpersonwe may say, similarly,that the akraticagent's understandingthat A is better than B is a reflective cognition ofwhich she might have only a vague or oblique understanding.And in the
same way thatsomeone who has read the First Meditationmight believe thathe ought to be persuadedby the argument,while notactually being persuadedby it, the akraticagentbelieves that he oughtto be persuadedby his reflectiveunderstanding,buthe is not persuadedby it. And this amountsto saying thatthe akraticagentacts againsthis own recognitionthatall thingsconsideredheshould act otherwise.When Waltervividly imaginesthatpersonholdingwithpridethe awardheso much coveted, he can see the point of hating that.person and takingrevenge. Walter knows indeed that he ought not to take revenge, but hisdesire for revengemight presenta conceptionof the good whose inadequacyhe only vaguelyunderstands ightnow. Waltermight see whyhe shouldtakerevenge in a way that he cannot see why he shouldrefrainfrom doing so. OrWaltermight see that he should not takerevenge, but he cannotclearly seewhy he shouldn't-the plausibilityof the desire for revenge is still there. Ofcourse, Waltercould, so to speak,wise up. His understanding f the groundsfor theinadequacyof his desire forrevengecould become moreandmoreclearto him, in the same way thatit will requiresome effort for a maturehumanbeing to acknowledge that, in a sense, the stars in the sky 'look small'; amaturehumanbeing will oftenjust perceive the stars as distant. In the idealcase, Walter will become virtuousin the Aristotelian sense of 'virtuous',4and he will no longer be prone to akraticbehavior. His understanding hatthere is nothingto be said for this perspectivewill become flawless, andhisdesire for revenge will be lost. Walter would still understandwhy one mightbelieve thatthere is somethingto be said for revenge in these situations,how43 One might think that if a person does not bet consistently with the belief that these areinstances of the gambler's fallacy, then, at least when it comes to the time of betting, hedoes not have available(at least at thatmoment)the understandinghat this is a fallacious
form of reasoning. But the same goes for practical akrasia. One can always claim thatthe agent who acts against her best judgment no longer has available her reflectiveunderstanding hat,say, A is betterthan B. There is no more reason(at least no morepre-theoreticalreason)to deny the phenomenon n one case thanin the other.44 The virtue in question, for Aristotle,is temperance.Accordingto Aristotle,"thecontinentand the temperatepersonare both the sort to do nothingin conflict with reason (...); butthe continent person has base appetites, and the temperate person lacks them."(Nicomachean Ethics, 115lb-1152a)
902 SERGIOTENENBAUM

30/38
this understandingof what can count as good could make a claim on some-one, but it would no longer make a claim on him. Walter would not seerevenge as good from any perspective, and so, according to our scholasticview, desire for revenge would not in this case be one of his desires. What thevirtuous agent accomplishes is that his unconditional judgment no longerneeds to rely on oblique cognitions, since there are no longer any direct cog-nitions of the good that present to him a temptationas a good worth pursu-ing in this occasion. Since those things that could be objects of temptationdo not appear o be good to this ideally virtuousagent, the reflective perspec-tive is idle: it does not need to compare or weigh the claims from differentperspectives.
Perfect virtue is certainlynot easy, and likely impossible. It is no surprisethat we devise methods that are similar to Descartes's evil genius: a directcognition that can mimic the function of perfect understanding.One mightlook at a lovely pictureof one's family to stay awayfrom a gambling table,or keep large sized clothes that one used to wearin ordernot to indulge againin old eating habits. Of course these strategiesarejust the same strategiesthatcome under he headingof self-controland that are invokedby those whoascribe akrasia to a gap between motivation and evaluation. Self-controlwould be a way to line up one's motivation with one's evaluation. But onthe view I am defending self-control should be understoodas a way of gettingone's evaluationattunedwith one's conceptionof how one oughtto evaluate.It is not a process of garneringmotivation, but finding a clear and obviousway to present something that one abstractlyandperhaps vaguely judges tobe good-in the ideal case, findinga direct

Sergio Tenenbaum the Judgment of a Weak Will

Documents

Inez Tenenbaum, of counsel · Inez Tenenbaum, OF COUNSEL...

Brief of Respondent Joel Tenenbaum

Nancy Kanwisher, MIT...Brain Specializations of Social...

Barbara Tenenbaum

Bayesian models of human inductive learning Josh Tenenbaum.....

Daniel Tenenbaum Appointment Letter...2016/08/29 · August...

Record Labels' appellate brief in Sony v. Tenenbaum

Dinamica2-ExerciciosResolvidos-Tenenbaum (1).pdf

Bayesian models of human inference Josh Tenenbaum MIT.

massimo.sbs.arizona.edu file · Web viewThe present...

Artur Tenenbaum/menu... · Lodz Polen, November 2009. - Ewa...

Prof. A. Tenenbaum Director Cardioline Heart Institute

Sony v. Tenenbaum (Sony Reply Brief)

"18: ABBY HELLER BURNHAM/JEREMY ERIC TENENBAUM"

JOSHUA BRETT TENENBAUM -...

By Carl Tenenbaum David Haynes Philip Pham Rachel Wakim.