Smith on Moral Fetishism Author(s): Hallvard Lillehammer Source: Analysis, Vol. 57, No. 3 (Jul., 1997), pp. 187-195 Published by: Oxford University Press on behalf of The Analysis Committee Stable URL: http://www.jstor.org/stable/3328140 Accessed: 24/11/2010 08:17 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=oup. 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 [email protected]
Oxford University Press and The Analysis Committee are collaborating with JSTOR to digitize, preserve and extend access to Analysis.
Smith on moral fetishism HALLVARD LILLEHAMMER
1. Introduction Accordingto internalismabout moral judgementsthere is an interesting conceptualconnectionbetweenan agent'smakinga moral judgementand that agent'smotivation. The externalistdenies this and claims that any interesting connection between moral judgements and motivation is contingent.1The resolution of this dispute has importantconsequences. For whereasthe internalistcan construemoral judgementseither as noncognitivestates like desire or as cognitivestates like belief, the externalist is committedto construemoral judgementsas cognitivestates like belief.2 A vindicationof externalismwould thereforelend supportto those who believein the possibilityof some kind of moral reality. In his book The Moral Problem and in a recent issue of this journal, MichaelSmithclaimsto refuteany theorywhichconstruesthe relationship between moral judgementsand motivation as contingent and rationally optional.He claimsthat no suchtheoryis able to accountfor the platitude that a good and strong-willedperson is reliablymotivatedin accordance with hermoraljudgements.3Morespecifically,the claimis that althoughthe externalistmay providea reliablelink betweenthe moral judgementsand motivationsof some individual,the only link at his disposalis a basicmoral motiveto do what is right,wherethis is readde dicto. But,so Smithargues, we can readoff fromthe platitudesthat are definitionalof moraldiscourse that this self-consciouslymoral motive makesfor moral fetishismand not for moral goodness.4Good people care about what is right, where this is read de re, not de dicto. He calls this a reductioof externalism. 1 An uninterestingconceptual connection between judgement and motivation which the externalist accepts: if an agent judges that it is right for her to f in circumstances c, then she is motivated to f in c unless she isn't. 2 For an externalist theory, see Brink 1989. For internalist cognitivism, see Dancy 1993. For internalistnoncognitivism, see Blackburn1993. 3
(Smith 1994: 60-91) and (1996b). The latter discussion is part of a defence against Miller 1996, in which it is claimed, unsuccessfullyto my mind, that Smith'sargument is question-begging. Miller 'happily concedes' that externalism fails to account for the platitude (1996: 171). I refuse to make this concession.
4 (Smith 1994: 76). Smith (1996b) makes no mention of the de dicto/de re distinction. There he applies a distinction between instrumentaland non-instrumentaldesires in order to make what is essentially the same point. The two distinctions are related as follows. Someone whose motivation is explained by a desire to do what is right de ANALYSIS 57.3, July 1997, pp. 187-95. ? HallvardLillehammer
Smith'sargumentfails. In showinghow it fails, I shallmakethreeclaims. First, a concern for what is right, where this is read de dicto, does not amountto moral fetishism.Second,it is not always morallypreferableto care about what is right,wherethis is readde re. Third,the externalistcan accountfor why a good and strong-willedperson is reliablymotivatedin accordancewith her moraljudgementswithout appealingto a basicmoral motive to do what is right,wherethis is readde dicto. 2. Smith's argument
Smith'sargumenthas the form of a dilemmawhich ariseswhen we ask in virtue of what a good and strong-willedperson is reliablymotivated in accordancewith her judgementsabout what is right. The externalistsays the connectionbetweenmoraljudgementsandmotivationis contingent,so he cannot say that it obtains in virtueof the contents of the moral judgements themselves.What accounts for an agent'smoral motivation must then be that agent's motivational dispositions, more specifically the contentsof her desires.5 What do the contents of an agent'sdesireshave to be like in order for that agent to be reliably motivated in accordancewith her judgements about what is right?One answeris this: an agentwho thinks it is rightto f in circumstancesc, and who desiresto do what is right,will be motivated to f in circumstancesc, all other things being equal. The externalistmight thereforeclaim that what accountsfor the reliabilityof moral motivation in the good and strong-willedpersonis a desireto do what is right. Smith'sdilemmathen arisesfrom the fact that a sentenceof the form 'x has a desireto do what is right'may be read eitherde dicto or de re. The difference between the two readings is a difference in logical scope. Considerthe sentence:'I want a sloop'.6On a de dicto readingthe meaning of this sentenceis 'I want that:I have a sloop', wherethe sentencefollowing the colon givesthe contentof the want. On a de re readingthe meaning is 'Thereis a sloop of which I want that: I have it'. There are interesting differencesbetween the two readings. First, the difference in meaning mirrorsa differencein the intentionalityof the mentalstates.In the case of dicto has a non-instrumentalmoral desireto do whateveris right, to which his desires to perform particularright actions are merely instrumental.A person who desires what is right de re desires to performparticularright actions for their intrinsicrightmaking features. s Smith (1994: Chapter4) endorsesa Humean belief/desireaccount of action-explanations, according to which the term 'desire' is used to cover all motivating states. I follow this usage. 6 The example is Quine's (1971: 101).
SMITH ON MORAL FETISHISM
the de dicto want, but not the de re want, I have an attitudewhich includes the concept 'sloop' as part of its content. Arguably,I need to possess the concept of a sloop in orderto have a want with this content. Second,the de re reading, but not the de dicto reading, entails the existence of the objectreferredto by 'sloop'. Arguably,the de re readingattributesa desire to me which is in part individuatedwith referenceto an externallyexisting objectwhich must exist in orderfor my desireto have it to exist.7 Now for the claimthat good peopledesireto do what is right,wherethis is read de re and not de dicto. This claim requiresclarification.There are at least two attribution-sentences betweenwhich Smith'sargumentequivocates.The firstsays that 'x has a desireto do what is right'(1994: 74, 75, 76). Call this sentenceP. The second sentencesays that 'x has a desireto do what he believesis right'.8Call this sentenceQ. P and Qhave different implicationsfor what the good person is like. First,a de re readingof P entailsthe existenceof right actions, whereas a de re readingof Q only entails the existence of actions believedto be right.It follows that in the case of P but not Qthere is a constitutivelink betweengoodnessand rightactions.Second,in the case of Q but not P we can say that a person who changeshis fundamentalvalues desiresto do what is rightat both ends of the transition.For all we mean by this is that he is motivatedin accordancewith his judgement.In the case of P, on the other hand, at most the desireat one end of the transitionis a desire for what is, in fact, right. Smithmust think it is a de re readingof Qwhich characterisesthe good and strong-willedperson. For his dilemmais formulatedfor the case of a good and strong-willedpersonwho changesher most fundamentalvalues. This person is neverthelesssupposed to be good partly in virtue of her concernfor what is right, where this is read de re. This claim only makes sense if we take Smithto mean that she desiresto do what she believesis right. Otherwise,she would not be able to changeher fundamentalvalues consistentlywith remaininggood in virtueof her de re concernfor what is right. Smith'sdilemmais as follows. Supposeyou are a libertarianand have alwaysvoted for the libertarianparty.Duringthe course of a discussiona friendconvincesyou that you are wrong, whereuponyou judge that it is rightto vote for the social democratsand becomemotivatedto do so. How can the externalistaccountfor the fact that you changeyour motivationto It has been argued that no such mental states exist (Fodor 1980). If they do not, then the very idea of a de re desire is problematic.Fortunately,we can ignore this question for the sake of argument. 8 Or, what for Smith amounts to the same thing: 'X has a desire to do what he judges right' (1994: 73, 75).
accordwithyourjudgement? On the firsthornof the dilemma,Smithcorrectlyclaimsthattheexternalistcannotappealto an antecedent desireto do whatis right,wherethis is readde re. Readde re, Qsays:'Thereis an actionwhichx believesis right,and x has a desireto performthat action'.9Now priorto your conversionyou desiredto voteforthe libertarians andjudgedthatit was do For to so. the a mere externalist, right changein judgementhas no for motivation. So you might equally well implications subsequent continuedesiringto votefor the libertarians ratherthanthe socialdemocratswhilejudgingthis is wrong,ratherthanchangeyourmotivationto accordwith yourjudgement. An appealto yourantecedentdesireto do whatis right,wherethisis readde re,goesno waytowardexplainingyour subsequent changein motivation. On the secondhornof the dilemma,Smithclaimsthat a desireto do whatis right,wherethisis readde dicto,mayaccountforwhyyourmotivation changeswith your judgement,but only at the cost of moral fetishism.Now this is plausibleonly if the de dictodesirehas universal ratherthanexistentialscope.To seethis,considerthe de dictoreadingof P.Thetruthof a sentenceof the form:'x has a desirethat:thereis some actionwhichis rightandwhichx performs' wouldnot sufficeto account for a reliableconnectionbetweenmoraljudgements andmotivation.Nor wouldit justifyany chargeof moralfetishism.Nevertheless, it is still a desireto do whatis right,wherethisis readde dicto.Thecharacterwho Smiththinksis a moralfetishistmustbe someoneall of whosedesiresto performparticularrightactionsarederivedfroma generaldesireto do whatis right.Thisdesireis bestattributed by a sentenceof theform'x has a desirethat:if someactionis right,thenx performsthataction'.Inother words,Smith'smoralfetishistis someonewho desiresto do whatis right in thesenseof P.He is a personwhoseonlynon-instrumental moraldesire is a desireto do whateveris right.All his desiresto performparticular actionsbelievedto be rightaremerelyinstrumental to the satisfactionof thisdesireto do whateveris right.He desiresto performno actionon the basisof its right-making featuresalone(1996b:180ff.). Ifyouhavesucha standingdesireto do whatis rightyoumaystopdesirand startdesiringto vote for the social ing to vote for the libertarians 9 The same conclusion can be drawn for a de re reading of P. Notice that it does not follow that x is barredfrom desiringto performthe action believed to be right under some description. What is crucial is that his desire is not merely instrumentalin the pursuit of a desire to perform actions under the description of rightness. Smith can therefore claim that the person who desires to do what is right de re desires to perform right actions for their right-makingfeatures,and thus for a reason (1996b: 182).
SMITH ON MORAL FETISHISM
democrats, giventhatyounowthinkit is rightto voteforthesocialdemocrats.Thereis a rationalexplanationfor thischangein motivation,since you haveall alongdesiredto performrightactions.Theproblemis that do not havetheirmoralmotivationexplainedby goodpeopleapparently a de dictodesireto do whatis right.To showthis,Smithgivesexamples, suchas thatof the manwho desiresto do whatis right,who believesit is andwho onlythererightto savehiswiferatherthana completestranger, foredesiresto savehis wife (1994:75-76). Smiththinksthis personis a moralfetishist,becausehis desireto savehis wife is derivedfroma basic moraldesirewhichis directedtowardsrightnessquarightness,andin the pursuitof whichhiswifeplaysonlyan incidentalrole. Accordingto Smith,the way out of this dilemmais to endorsethe followinginternalist principle: If an (PR) c, thenshe agentjudgesthatit is rightto fin circumstances is motivatedto f in c unlesssheis practically irrational. to (PR)moraljudgements issuein desiresto act as theyspecify According on painsof practicalirrationality.10 Thekindsof irrationality in question arepsychological disturbances suchas weaknessof willandthelike.There is thena rationalexplanationfor whya goodandstrong-willed personis Fora goodandstrong-willed reliablymotivatedby hismoraljudgements. personwho judgesthatit is rightto fin c will knowthatit is rationalfor himto be motivatedto f in c. Andbecausehe does not sufferfromany irrationaldisturbance suchas weaknessof will, he will be motivatedto f in c. 3. Moralattitudes:de dicto Theclaimthatit is a fetishto careaboutwhatis right,wherethis is read de dicto,is false.It is falseevenfor Smith'sbasiccase, wherean agent values.Considerthe caseof someonewho changeshis mostfundamental hasalwaysbelievedthatmoralityis not verydemanding in termsof individualsacrifice.Supposehe comesto believethathe is morallyrequiredto sacrificeeverythinghe has, perhapsevenhis life. Supposefurtherthathe doesnot directlyacquirea de re desireto do whathe now thinksis right, butthata standingdesireto do whatis rightde dictoprovidesthe causal linkwhichmotivateshimto sacrifice he has.It is not a platitude everything thatthis personis a moralfetishist.Maybeit wouldbe admirableif he cameto careaboutwhatis rightin anunderived eventually way.Butgiven 10 Smithcalls the principlethe practicalityrequirement(1994: 61). Apparently,he takes
the principleof Weak Internalismin (1996b: 177) to capturethe same claim, in spite of its making no referenceto the concept of rationality.
what he now considersmoralityto demand, he might be forgiven if his immediateconcernfor what is rightis not direct. Smith is strangelysilent about the person who comes to reassesswhat moralityrequires.For althoughhis dilemmais formulatedwith reference to a fundamentalchange in values, his argumentthat a de dicto concern for what is rightis a fetishis not. This argumentis supportedby an appeal to cases like that of the man who faced with the choice of saving his wife or a strangeris motivatedto save his wife, but only conditionallyon a de dicto desire to do what is right (1994: 75-76). Smiththinks this person fails to be good becausehe fails to carenon-derivativelyabout his spouse. Butnothingfollows fromthis exampleaboutwhat explainsthe motivation of a good and strong-willedpersonwhen she changesher most fundamental values, since this exampledoes not mention a change in values, never mind a fundamentalchange. A concernfor what is right,wherethis is readde dicto, has a role to play in the psychology of good people beyond this special case. To take one example, many people go through phases when they temporarilylose affectionfor people to whom they are close. Considersomeone who goes to a partyduringa phasewhen she is tiredof her husband.At the partyshe meets a very charmingpersonand is temptedto have an affair.She judges that it would be wrong to have an affairon accountof her husband'sfeelings. But she is temporarilyindifferentto herhusband'sfeelings.However, she has a standingde dicto desireto do what is rightwhich, togetherwith her moral judgement,causes her to do the right thing, in spite of the absenceof a de re desireto do the rightthing and the presenceof a de re desireto do the wrongthing.If thereis anythingin this case whichprevents this personfrom beinggood it is not herstandingdesireto do what is right, wherethis is readde dicto. Forthis desireis playingthe role of an internalised norm that preventsher from beingtemptedto do wrong. Suchnorms are not in contradictionwith the platitudesthat are definitionalof moral discourse.Their benefitsare all too obvious. Consider next the case of the father who discoversthat his son is a murderer,and who knows that if he does not go to the police the boy will get away with it, whereasif he does go to the police the boy will go to the gas-chamber.The fatherjudgesthat it is rightto go to the police, and does so. In this case it is not a platitudethat a desireto do what is right,where this is read de re, is the markof moralgoodness.If what moves the father to informon his son is a standingdesireto do what is right, where this is readde dicto, then this could be as much of a savinggraceas a moral failing. Why should it be an a priori demandthat someone should have an underiveddesireto send his son to death?
4. Moralattitudes:de re Smith is wrong if he thinks the externalistis barredfrom attributingto good people desiresto do what is right, where this is read de re. Suppose it is rightto care for one'sfamily.Many peopledo, and they do so without havingderivedthis concernfrom a concernfor what is right,wherethis is readde dicto. The samegoes for a host of otherconcerns,from a desirefor self-preservationto desiresfor the well-beingof livingcreaturesof all sorts. Externalismis consistentwith the claim that these concernsare partially constitutiveof moral goodness. Externalismis also consistentwith the fact that de re concernsfor what is right can be acquiredby experience,educationand reflection.I might changemy previouslymercenaryattitudetowardshumanlife afterexperiencingthe horrorsof war and thuscome to carein an underivedway about other people's suffering. I might be brought to love my country after havingits valuesinculcatedin me at school. Or I may undergoa processof reflectionand acquirea belief that it is right of me to performa certain action,whereuponthat beliefcausesa desirein me to do what I now think is right, where this is read de re, not de dicto.11The externalistdoes not deny that moral beliefs directlycause desires to act in accordancewith those beliefs. Sometimesthey do and sometimesthey don't. The crucial point is that it is not necessarilya markof irrationalitywhen they don't. Smithcan press his point once more at this stage. For in virtueof what is it that moral beliefs cause desiresto do what is right in those in whom such desires are reliably produced?What is the externalist account of moralgoodnesswhich guaranteessuchmotivationin the good and strongwilled person? One does well to rememberthat the externalistdoes not have to answer this question.For he may acceptthe platitudeabout the reliabilityof motivation in the good and strong-willedperson without acceptingthat this platitude is in need of some furtherplatitudinousa priori principlelike (PR)to accountfor it. He may regardthe platitudeas primitive.If he does, then he will say that it is a fact that we call people good and strong-willed only if they are reliablymotivatedin accordancewith their moral judgements.If someonefails to be so motivated,then they do not qualifyas good and strong-willed.What makes some people motivatedin one way rather than anotheris a matterof theirpsychologicalmake-up,somethingabout which we are mainly ignoranta priori.Afterall, (PR)does not tell us very much about the psychology of good people either.All it tells us is that if arebeliefswhichmaycausedesiresto acton them 1 The thesisthatmoraljudgements is neutralbetweenSmith'sinternalismand the externalismhe wishesto refute.To employit hereis thereforenot question-begging.
they arenot directlymotivatedin accordancewith theirmoraljudgements, in some way or other,then they are practicallyirrational. 5. The PracticalityOption
Theexternalist cando better.Considerthefollowingplatitudeaboutmoral judgementsand motivation,neutralbetweenexternalismand internalism: (PO) If an agentjudgesthat it is rightfor her to f in circumstancesc,
thenif shehasa normative reasonto f in c shewillbemotivated to f in c unlesssheis practically irrational.12 (PO)differsfrom(PR)inthat(PR)does,whereas(PO)doesnot,entailthat it is alwaysirrationalnot to be motivatedin accordance withone'smoral is entails that it irrational not to be so motivated judgement.(PO)only whenone has a reasonto be so motivated.Theexternalistcan appealto (PO)to accountforwhya morallygoodandstrong-willed personis reliain with her moraljudgements on Smith'sown bly motivated accordance theexternalist cando this terms,yetwithoutendorsing(PR).Furthermore, whileavoidingan awkwardimplication of (PR),namelythatit is always irrationalnot to be motivatedin accordance withone'smoraljudgement, no matterhow poorthatjudgement is.13 On the externalistaccountwhichI am proposing,a good personis someoneforwhomit is rationalto actin accordance withhismoraljudgewho on and acts those because he knows whatmorality ments, judgements a For such there is the same rationalexplanation requires. person exactly forwhyhe is reliablymotivatedin accordance withhis moraljudgements as the one Smithproposesfor all agentson (PR).An agentwho has a normativereasonto be motivatedin accordancewith his moraljudgementswill be motivatedin accordance withthosejudgements on painsof irrationality. Butnot all agentsmaybe suchthattheyhavea normativereasonto be motivatedin accordancewith theirmoraljudgements.First,an agent whose moraljudgementsare radicallydefectiveor corruptedmay be withthem.Second, rationallyrequirednot to be motivatedin accordance an agentmightcometo thinkthatmoralityrequiressomethingwhichit cannotbe rationallydemanded thathe do. Youmight,forexample,come 12 I call thisthe practicality optionto distinguishit fromthe
practicality requirement. have a reason to act in normative agentsnecessarily withtheirjudgements accordance aboutwhatis rationalin (1996a:162, footnote1). sinceSmiththinks Thisretractioncan be extendedto thecaseof moraljudgements, of whatis rational(1994:62ff.).He goeson moraljudgements reduceto judgements to saythatagentsnevertheless 'rationallyshould'actin accordancewiththeirjudgementsaboutwhatis rational.Theforceof this'should'eludesme.
13 Smithretractshis claimthat
SMITH ON MORAL FETISHISM
to think that it is morally requiredto drown all handicappedpeople at birth or that mothers should be preventedfrom having an abortion even when the alternativeis that both motherand child will die. It is not a platitude that you would be irrationalnot to be motivatedin accordancewith those judgements. On the contrary, someone who was motivated in accordancewith his moraljudgementsno matterwhat they were, could be accused of a different kind of moral fetishism. It follows that Smith himself, in virtue of his commitmentto (PR), is committedto a kind of moral fetishism.14 Peterhouse Cambridge,CB2 1RD [email protected]
References Blackburn,S. 1993. Essays in Quasi Realism. Oxford: Oxford UniversityPress. Brink,D. O. 1989. Moral Realism and the Foundations of Ethics. Cambridge: CambridgeUniversity Press. Dancy, J. 1993. Moral Reasons. Oxford: Basil Blackwell. Fodor,J. 1980. Methodological solipsism considered as a researchstrategyin cognitive science. The Behavioraland Brain Sciences 3: 63-110. Miller,A. 1996 An objection to Smith'sargumentfor internalism.Analysis 56: 169-74. Quine, W. V. O. 1971. Quantifiersand propositional attitudes. In Referenceand Modality, ed. L. Linsky, 101-111. Smith, M. 1994. The Moral Problem. Oxford: Blackwell. Smith, M. 1996a. Response to Swanton. Analysis 56: 160-68. Smith, M. 1996b. Response to Miller.Analysis 56: 175-84.
14 Partsof this paper have been presentedat the universitiesof St. Andrews, Cambridge and Reading.I am gratefulto the audiencesthere, as well as to David Copp, Jonathan Dancy, Brad Hooker, Chris Penston, Michael Smith, Peter Smith, an anonymous referee, and especially Alex Oliver for their generous help and comments. Work for this paper has been supported by a PeterhouseResearchStudentship.