8/3/2019 2254027

1/13

Mind Association

Luck and DesertAuthor(s): Norvin RichardsReviewed work(s):Source: Mind, New Series, Vol. 95, No. 378 (Apr., 1986), pp. 198-209

Published by: Oxford University Press on behalf of the Mind AssociationStable URL: http://www.jstor.org/stable/2254027 .

Accessed: 26/11/2011 09:15

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

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

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

of scholarship. For more information about JSTOR, please contact support@jstor.org.

Oxford University Press andMind Association are collaborating with JSTOR to digitize, preserve and extend

access toMind.

http://www.jstor.org

http://www.jstor.org/action/showPublisher?publisherCode=ouphttp://www.jstor.org/action/showPublisher?publisherCode=mindhttp://www.jstor.org/stable/2254027?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/2254027?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=mindhttp://www.jstor.org/action/showPublisher?publisherCode=oup

8/3/2019 2254027

2/13

Luckand DesertNORVIN RICHARDS

... the intuitivelyplausibleconditionsofmoral udgmenthreateno underminet all.'One'conditionof moral udgment' s thatmattersbeyondaperson'scontrolcannotbearon whathe deserves.For example, o be born n Pakistan annotmakeyoudeserve o fare ess wellthan f you hadbeen bornsomewhere lse.Nor can you deservecredit for winning a lottery throughsheer luck, orblamefor bleedingwhen you are cut. Those are all matters beyondyourcontrol,and thus beyond your responsibility: connectionso fundamentalthatNagelcan say'It seems irrationalo takeor dispensecreditor blameformattersover whicha person hasno control ....And yet, Nagel also argues,'If the conditionof controlis consistentlyapplied, t threatens o underminemost of the moralassessmentswefinditnatural o make.'3For one thing, says Nagel, we takethe consequencesofa person'sbehaviour o bearon justhow bad a thinghe has done,and thuson howharsharesponsehe should receive-even thoughtheconsequencesmay be determinedby somethingbeyondhis control!Consider he driverwho roarsthrougha school zone at 70 m.p.h., oblivious to the childrendartinga few feet from his path. If he is lucky enough to pass throughwithouthitting anyone,he is guiltyof recklessdriving,andwillbe criticizedaccordingly.Butwhat f a child choosesprecisely he wrongmoment odashafteraball,andendsupcrushedbeneathour driver'swheels?Thenit willbemanslaughtere hascommitted,a far worsethingto havedone,and we willtake him to deservemuchharsher reatment.What makesthe difference,though, is something beyond the driver'scontrol:his luck in what the childrendo. So here, Nagel would argue,we give luck the very power we deny it can have: the power to makea person deserve harsher or milder treatment.And, says he, we couldnot stop doing so without abandoningan equally fundamental tenetthat those who do more harm deserveharshertreatmentthan those whodo less.Similarly,The thingswe are calleduponto do, the moral ests we face,areimportantlydetermined y factorsbeyondourcontrol. t maybe trueof someone hatin

1 ThomasNagel, 'MoralLuck'in GaryWatson,ed., Free Will,New York,OxfordUniversityPress,I982, p. 176. All futurereferencesare to this edition.2 p. 177.

3 p' I76.

8/3/2019 2254027

3/13

Luck and Desert I99a dangerous ituationhe wouldbehave n a cowardlyor heroicfashion,but if thesituationnever arises,he will never have the chance to distinguish or disgracehimselfin this way, andhis moralrecordwill be different.4If his record is different, through his good or bad luck in never having beenput to the test, we will treat him differently. Indeed, such a discrimination isso basic that. . . our ordinarymoralattitudes would be unrecognizablewithout it. We judgepeopleforwhattheyactuallydoorfail todo,not just forwhattheywouldhavedoneif circumstanceshadbeendifferent.Thirdly, according to Nagel, there is 'the phenomenon of constitutive luck-[luck in] the kind of person you are ... your inclinations, capacities, andtemperament'.6 It is because of your set of inclinations and capacities andyour temperament that you act as you do. Thus, in holding you responsiblefor your actions, we allow your good or bad luck in having that set andtemperament to affect your deserts.In sum, the practice of moral judgement appears to Nagel to be a para-doxical one. On the one hand, it requires that matters beyond one's controlcan have no bearing on one's deserts. On the other, it requires that theyhave enormousbearing after all. The paradox is not dispelled, it seems tome, by observing that our practice has both Kantian and Aristotelianelements, as Judith Andre has done in her otherwise illuminating article.7At most, that only explains why our practice is paradoxical in the wayNagel claims it is.I want to argue instead that it is not paradoxical after all, and to offera different explanation for the several ways in which luck appearsto bear ondeserts. Central to my argument will be a claim that what a person deservesfor a particulardeed can differ from the criticism we are actually entitled tolevel against him for doing it. I think, that is, that our epistemic positionregarding the matters which determine an agent's deserts is so imperfectthat (for example) someone can have acted much more culpably than any-one has grounds to realize. If he has, no one is entitled to criticize himas harshly as he deserves. For criticism should reflect not a pretendedomniscience but one's actual grasp of what has been done.A culprit may thus be lucky or unlucky in how clear his deserts are. Andwe must allow his luck to make a difference in how we treat him, if we are notto change our practice into one in which we pretend omniscience. But indoing so we do not contradict the contention that his luck cannot affect whathis deserts are. So, we do not act paradoxically, as Nagel contends, butonly reflect our epistemic shortcomingsS,and the agent's good or bad fortunein those.

4 p. I82. 5 p. I82. 6 p 177.7 'Nagel,Williams,and Moral Luck',Analysis,I983.

8/3/2019 2254027

4/13

200 Norvin Richards

My case for thisconclusionrequiresme to speakof an agent'scharacter.Bythis I mean his set of preferences,n so far as these arerela-tivelytable overtime: the fact that he prefers company to solitude, or telling the truth tobeing misleading,and so on. Such a preferencecanbe fortunateor unfor-tunate nthe actions t makesone more ikelyto performand the subsequentconsequences or oneself andothers.Now it is scarcelyradical o say that when we are concernedwith whatapersondeserves,we are nterestednhis behaviour s a displayof character.It is easy to understand he standardexcuses in that light. Having harmedsomeone,I urgethat I was not maliciousbut only ignorant: .e., thatI amnotsomeone or whomthe harmwas a positiveattraction,but only someonewho did not realize t would occur. Ideally, I hope this ignorancewill notitself show meto beinsufficiently verse oharmofthe sortI have done. Or,in a differentcase, I mightplead that I acted under duress: hat I did thisharmonly to avoidanother,a more innocent weighting of outcomesthanmight have appeared.Stillanother ime, my plea might be provocation:hatI amnot a bully or a hotheadbut onlysomeonewhose aversion o violencecan be overcome.One reason to be concernedabout the characterenacted in a bit ofbehaviours thatcharacter,by definition, s relatively table.Qua displayofcharacter,the deed involves a taste or level of aversionwhich may beindulged again, n anynumberof ways,if the currentepisode is not madeunrewardingor the agent.Whatevermeasureswe do take to protect our-selves from futureindulgencesmaybe more or less unpleasant or him. Itwould be unreasonable or the measures o do him moreharm than theyprotected others fromsufferingat his hands.But the more dangeroushisfutureindulgencesare, the more costly to him those protectivemeasuresmay reasonablybe.Thus, behaviourwhichenactsaparticularraitcallsfor,or 'deserves',protectivemeasures n proportion o the riskof future harmsat whichit putsotherpeople.The fact that a persondoes deservea particular ort of responsedoesnot settle how he shouldbe treatedby thosewhowish to givehim what hedeserves,however.For they are not omniscient: hey havebnlytheirgraspof his deserts to enact,Iand they are obliged to be careful in a matterthis important.So, for example,his actingin a way which deserves theirrebukinghim does not entitle themto rebukehim, since theirgroundsforbelievinghim to have done so may be inadequate.The rebuke could beepistemicallyrresponsible, venthoughdeserved.Moreover, mattersbeyondthe agent's control bear on how clearly we seewhathe deserves-and, thus, on the legitimacyof ourtreatinghim in thatway. His luck in those mattersaffectshow we ought to treathim, not bychangingwhat he deserves,but by changingthe groundson which we are

8/3/2019 2254027

5/13

LuckandDesert 201obligedto judge.For example, t maybe a person'sgood luck thatthere areno eyewitnesses o her misdeed.Jealousof your new carpet,she emptiesherwineglassoverit-just as the ring of a telephonedistracts he only personwho would have seen her do so. A momentlater,she slips quietlyinto thenext room.Much later,you discoverthe stain-and have no grounds forblamingher for it. Were you to treather as (in fact) she deservesto betreated,youwould be actingquiteirresponsibly, ndher luckhasmadethisthe case.I thinkwe can also imagine mattersdifferently,so that addingan eye-witness wouldnot strengthenyour groundsfor inferringthat this womandeserves whateversuch people do deserve.Perhaps you enter the roombeforeshe has time to depart,whileshe is stillsmilingdownat thespreadingstain and beforeshe removes he randomdropletswhich splashedher shoe.Bad luckfor her-again, not in thatyour arrival hangeswhat shedeserves,but in thatit changesthe legitimacyof yourtreatingher as she deserves.I think thatNagel's first sort of luck, luckin the consequencesof one'sactions, playsthis samerole. Like luck in the presenceor absenceof eye-witnesses, it sometimes affects our grounds for seeing what a persondeserves,and, thereby, helegitimacyof ourtreatinghiminthatway.Othertimes, we are as clear without he tell-taleconsequences:we do not alwaysneed successo see what was intended,any more than we alwaysneed aneyewitness.This wouldnot bean inconsiderable ole for luck nconsequencesoplay.But neitherwouldit be a paradoxical ne. It wouldnot amount o allowingluck to affectdeserts,but only to recognizing hat we are not omniscientaboutsuchmatters,and thatwemust beresponsiblen our inferencesaboutthem.Consider, orexample,the recklessdriver.Sometimes, f a driverharmsno one,there s room forthe belief thathis speed anddegreeof attentionareadequate:hat he is sufficiently lert oavoidaccidents f the need arises, hathisspeed s safe forapersonof his skills,and so on.Evenif these contentionsarequite false, so that in fact his driving enactsa very dangerous evel ofunconcernandhis behaviourdeservesaveryharshresponse, he mattermaybe debatable o longas he does no actualharm.For such a driver, t is onlywhen he hits someonethat it becomes plainhe is not sufficientlyattentive,that his speed istoo fast,forhim,that he is not givingsufficientweightto theriskof harmto otherswhen he drivesas he does. It is only whenhis luck-and his victim's-runs out, that we have sufficientgrounds or treatinghimin the wayhe has deserved o be treatedall along.On the otherhand,on other occasions he accident s superfluous,n thesamewayas aneyewitnesscan besuperfluous.Thatis, the accidentdoes notstrengthenourgrounds or believinghisdriving o be reckless, s not crucialto the legitimacyof our treatinghimas he deserves.Instead,commonsenseabout human reaction time, the limitations of human eyesight, and the

8/3/2019 2254027

6/13

202 Norvin Richardsnatureof local conditions s enough:we are entitledto conclude hat anyonedrivingthatway,in thatplace,at thattime, is drivingrecklessly.If so, weneed not wait forthe tragic mpact,to be entitledto treathim as he deserves.In short,if we ought to treat a certaindriverdifferently rom one whoseidenticalrecklessness nds in manslaughter,hat will notbe because he twodeservedifferent reatment.It will be because(a)their behaviourdoesnotmaketheirequaldeserts equally clearand(b) ourtreatmentshould reflectour understanding f those deserts.The agent's uckin the consequencesofhis actionscan affectour understanding f his deserts,and,thereby, he waywe oughtto treathim. It does not affectthe characterhe enacted,however.So, it doesnot affecthis deserts, for those are determinedby the characterenacted.

IIBut isn't this wayof dealingwith luckin consequences elf-defeating, inceNagel also maintainsthat what characterone has to enact is also beyondone's control? f the 'kind of personyou are ... yourinclinations, apacitiesand temperament's determinedby mattersbeyond yourcontrol,andyourdesertsare determinedby whichsuch featuresyou enact,then (once again)yourdesertsappear o be determinedby somethingbeyond yourcontrol.This time, their ultimate determinant s whatevermade you the personyou are.This argumentsucceeds,I think,if one's character s to no extent one'sown artefact.But if the individualmakesany contributionwhatever o thesort of personhe is, that contributioncan be the basis for his deservingpraiseorblame or whathe does. Nagelleavesroom for sucha contribution.Speakingof suchqualitiesas conceit,he saysTosomeextent ucha qualitymaybe theproduct f earlierhoices;o some xtentit maybe amenableo changeby current ction.But it is largelya matter f badfortune.8

Of course,he could be wrongabout this. It couldbe that one's characteris shaped entirely by forces beyond one's control. If so, the practiceofattributing esponsibilitys undermined,with no needfor us to referto theother sorts of luckNagel bringsto our attention.But he does not argueforthis traditionaldeterminism/incompatibilism,ppearingnsteadto wanttooffera differentpuzzle.It could also be thatalthoughone's characters only 'largely'one'sgoodor badfortune,andpartlyalsoone'sowndoing,in assigningdesertswe takethewrongpartseriously.For example, t could be that twoindividualswhohad the samecharacter,one throughno fault of his own andthe otherfarmoreas the resultof his own contributions,wouldbe takento deservethe

8 p. i8i.

8/3/2019 2254027

7/13

Luckand Desert 203sameresponse or enacting his character.But,in fact, wherewe do see suchdifferenceswe do not take the deserts to be the same-because we do notallow uckto affectdeserts.Wherewe do notsee such differences,we do takethe deserts o be the same.Since our evidenceaboutsuch matters s farfromperfect,no doubt we sometimesdo not treatpeople as they deservewhen wefollow our graspof the situation.That is not the same as allowing heir luckin our grasp o affectwhat their desertsare,however,but only the result ofdesertsturningon somethingdifficult o discern.

IIIThirdly, according o Nagel, one's opportunitiesor agencyare also mattersof luck. Perhapsyou would have saveda drowningswimmer,had you beennear by when he began to struggle. But you were miles away at the time,caughtin a traffic amon the freeway.You never even knew a rescue wasneeded, until the next day's newspaper.So you missed your chance to beahero, and, thus, your chance o deserve hepraisebestowedon the womanwho was at the right place at the righttime. Her luck and yours broughtyou differentopportunities o act and, it seems, differentopportunities odeserveto be treated n a certainway by yourfellows.What if you were to complainabout this, insistingthat you should bepraised for the heroism you would have performed, and praised just ashighly as the heroine s for her actualheroism?Nagel thinks this would notbe merelyimmodestand insecureof you, as it might be if you demandedpraisewhich actuallywasyourdue. Rather,your demandwould be deeplyconfused,because hepraise s notyourdue:'We judgepeoplefor whattheyactuallydo or fail to do, not just for what they would have done if thecircumstanceshad been different.'9Of course,if Nagel meansby this that we do notjudge peoplefor whattheywould havedone,he is mistaken.Supposeyouhadbeenabandonedna time of need by someoneyou took to be a friend.Supposethatin talkingthis over with a thirdparty, t struckyouhow much moresteadfasthe wouldhave been than the swine who desertedyou. Might not this realizationpromptyou to praiseyourfriendquitewarmly?Or supposeyou were con-vincedthat a repairmanwouldhave stolenyourwallethad he not realizedhe wasunderscrutiny.Might not that be a sufficientbasis fortreatinghimratherharshly? n bothsorts of cases,it seemsthat someonecan deservetobe treated n a certainway, not becauseof whatthis personhas done butbecauseof what t is plausible o thinkhe woulddo, if giventheopportunity.Still, actualagentsdo differ rompotentialonesin theirdeserts,asNagelurges.The difference s thatonlythe actualherois praiseworthyfor avingtheswimmer,only the actual Nazi is blameworthyfor the behaviourat Belsenwhichheperformed. t wouldbeludicrous osaythepotentialherodeserves

9 p. I82.

8/3/2019 2254027

8/13

204 NorvinRichardsourwarmresponseforthatdeed,with whichhe hadnothing to do, or thatthe potentialNazi deservesa harshresponse or whathappened t Belsen.However,I willargue, hisdifference n deserts s much lesssignificant hanNagel thinks.My centralcontentionwillbethat f thepotentialagent s asmuchlike theactualone as we are imagining,then there will be something else in hisbehaviourwhichwill call forthe sameresponse.If so, his luck in not doingaparticulardeed will not affectthe treatmenthe deserves.He will deservethe same sort of encouragement rreconditioning,buthe willdeserve t fora differentenactmentof virtuousor vicious character.His luck will thusaffectnot what he deserves but the time at which he deserves t and, onceagain,the claritywithwhichhe can be seen to deserve t.

Let us focusupon the followingpair of actualandpotentialmiscreants:Someonewhowasanofficern a concentrationampmighthave eda quietandharmlessife ftheNazishadnever ome opowernGermany. ndsomeonewholed a quietandharmlessifeinArgentinamighthavebecome nofficern a con-centrationampf hehadnot leftGermanyorbusiness easonsn 1930.10We are magininghere ane6migre-hoshares he traitsofcharacterwhich theotherman enacted n beinga concentration ampofficer-so thatonlytheemigre'sgeographicalgood luck prevented his playing that same role.Perhaps,for example, both men have an extremelystrongdesireto pleasepersons nauthority,andlittle or no aversion o thesufferingofanyone heyconsider inferior.To call the e'migre"sesire to pleaseauthorities extremelystrong'is torank trelative o hisotherdesires.Forhim, thechance opleaseanauthorityis agreaterattractionn analternativehan mostof the otherthingshe con-sidersattractions.So he is quicker o notice whathe takes o besignsof theirwishes, adoptscourses he takes to pleasethem with greateralacrity,andpursuesthose courses more persistently n the face of what are for himtemptations o do otherwise.Similarly,to say he is not veryaverseto suffering n thosehe considersinferior s to rank hataversion elative ohis others.Forhim,theirsufferingis notmuch of amarkagainstacourseofaction,but is somethingmoreeasilyoverridden han otheraversive eatures.So he is not veryalertabouttheirsuffering,perhapsfails to notice it entirely, is not much deterredfroma coursebyrealizing hat t threatens hisharm,persists n suchcourses,andso on. Finally,since we arespeakinghereof the e'migre"sharacter,we aredescribing eatureswhich arerelativelystableover time.Now, I havealready uggestedthat it is not the harma persondoesthatdetermineshis deserts,butthe featuresof characterwhich he enacts.Thus,it is someone's ndifferenceo mysufferinghatcalls for attemptsto changehim, or to denyhim futureopportunities o harmme, or to showhim that

10 p. '75.

8/3/2019 2254027

9/13

Luckand Desert 205such behaviour will be costly in the future. The harm he actually does me inhis indifference is only a sign of how indifferent he is and how dangerous hisindifference is: the same reactive measures may be appropriate even thoughhe has done me very little harm. "

If all that is right, and if each of the e'migreand the concentration campofficer enacts the character described, then each will deserve the sameresponse to those enactings. For each will have enacted the same dangerous,stable features of character, calling for the same protective and re-educativeresponses. The emigre'sbehaviour will have been less striking, perhaps, andwe will then be less clear that it does call for this response. But his desertswill be the same, just as the deserts of the luckyreckless driver do not differfrom those of the unlucky one. Once again, the reason we should treat thetwo differently is not because they deserve different treatment, but because(a) their behaviour does not show their deserts equally clearly and (b) weshould treat people according to our understanding of their behaviour.But, isn't it possible that the emigre has the requisite character to bea camp officer but never enacts it, in any way?So that he never deserves theseresponses after all, by virtue of his good luck in never having this part of hischaracter called into action? Nagel appears to think of character in sucha way:A personmay be greedy,envious, cowardly,cold, ungenerous,unkind, vain, orconceited,but behaveperfectly .. an enviouspersonhatesthe greatersuccessofothers.He can be morallycondemned as enviouseven if he congratulateshemcordiallyand does nothingto denigrateor spoil their success. Conceit, likewise,neednot bedisplayed. t is fully present n someonewho cannothelp dwellingwithsecret satisfactionon the superiorityof his own achievements,talents, beauty,intelligence,or virtue. 2

'In fact, however, the paragraphdoes not describe people who never enactthe traits in question, but only people who do not enact these in particularways. To dwell with secret satisfaction on one's own superiority is a con-ceited action, albeit a different one from bragging. And envy can play a rolein one's 'cordial congratulations'-making the words ring false, the hand-shake last a bit too long, and the smile be a bit too broad. Qua enactments ofconceit and envy, on my view these call for the same protective and re-educative responses as the more blatant displays. Since their call is so muchmore muted, however, it might well be pretentious to respond to them as ifit were clear. If so, we should treat such actions differently, although notbecause their agents deserve a different response.Moreover, the idea of unenacted character is a mistaken one, it seems to

11 Similar views about desert can be found in Hume's EnquiryConcerninghePrinciples f Morals(see, in particular,Sec. V, partII); in RichardBrandt's AUtilitarianTheory of Excuses',PhilosophicalReview, 969 (see, in particular, p. 353-8); and in MichaelBayles's Character, urpose,and CriminalResponsibility',Law and Philosophy,982.12 p. i8i.

8/3/2019 2254027

10/13

206 NorvinRichardsme. One's character s the relative weightvariousmattershave for one inone's actions, in what comes to one's notice, and in the hesitationsandbearingwith which one acts,no less thanin whetherone actsin a recogniz-ably conceitedway (for example).Characters not a control-boxdetermin-ing whatmatters hallhave which weight,but the fact that they do havethatweightas one movesthroughthe world.Even if we do grantthatthe e'migre'ill enact the character n question,though, he willnotenact t by servingas an officer n a concentrationamp.Canwe concludethat, in him, the characterwill not be so unfortunateasit was in the German? f so, on the view I have offeredhis enactingsof thecharacterwill not deserve so harsh a response,for they will not indicatesomethingso threatening.His desertswould then havebeen changedby hishaving had this character n Argentina ather han in Germany: omethingwhich might have been a matterof luck. However,I do not thinkit doesfollow fromthe emigre"sot serving n a campthat the traits are not so bada thingin him.Indeed,a traitwhoseenactingsare not spectacularly ainfulfor others standsa greaterchance to persist. It is less likely to draw theoutraged eactionswhichcan makea manchange.The emigres likelyto liveanentire ife in which he takesthe pleasuresof authorities oo seriouslyandthe pain of certainotherstoo lightly.That will be a stunted life, as well asa damagingone.

IvI have agreedwithNagel that our verdicts njudgingpeopleturnonmattersbeyondthosepeople'scontrol,andthatourpracticeswouldbeverydifferentif they did not do so. Unlike him, however,I do not thinkthis amounts oourparadoxically llowing uckto affectdeserts.Instead,I haveargued, oa very largeextentpeoplearelucky only in our verdicts,not in whattheydeserve,andtheir uckmust affectour verdictsbecause t affectsourgroundsforreaching hem.This way of accommodatingNagel's paradoxhas carriedcertainphilo-sophicalcommitments: o one analysisof characterandanotherof desert,and to the belief that one's character s to some extent one's own artefact.This is not the placeto defend these commitmentsagainstevery objectionthey might provoke.However,thereis onepossiblemisgivingso deepas torequirecomment.It amounts o adoubtthatI havebeentalkingaboutdesertat all. Clearly,if it is not desert I have striven to prove independentofluck, then even if my strivingshave been successful they will not haveproveddesertunparadoxical.The reasonfordoubtingthatit is de-sert have beendiscussing,whenIhave discussedseeing actions as enactments of character, s the purposeI gaveforlookingatthemin thatway.The idea wasthat,asanenactmentofcharacter, nactionmakes uturebehaviourmorelikely, therebycallingfor

8/3/2019 2254027

11/13

Luckand Desert 207re-educativeand protective responses against that future behaviour.Butthis focus on the future is the markof utilitarian hinking;whereas, radi-tionally, a person's desertis supposed to have nothing to do with theconsequencesof treatinghim,anyparticularway. Desert i's upposed o flowfrom the natureof the deed; in my way of talking, t does so only in so farasdeeds indicate uturedeeds by being symptomsof patterns n the agent'sbehaviour.Now, personally,I am not convinced that my way of talking does notdescribedesert. It may be that when someone has the intuition that aparticular deed just calls for a harsh response, this is nothing morethan a commendablyquick realizationof how dangerous he agent is. Butlet us suppose that nothing like this is true and that it is not desert I havedefendedagainstparadoxbut somethingelse altogether.Two possibilitiesarise.First, it may be possible for the objector o recastmy chief line of argu-ment so asto defend desert.Essentially,I have argued hatdesert s a func-tion of character nacted; hat t is our understandingfthecharacter nactedthatdependson theagent's uck n such mattersas his opportunities nd theharmhe actually does; and that this is quite differentfrom allowinghisdeserts hemselves odependonhis luck.Perhaps he samecould be saidbysomeone who did not share my views as to why we should care aboutthecharacterenacted.Such a personwants to say instead that to enact somecharacter-traitss intrinsically orse hantoenactothers,andthus deserving(in itself) of a harsher esponse.Perhaps t might still be argued, hat s, thatthe agent's uckdoesnotaffect he wickednessof the character e enactsbutonly how clear we are aboutit-and, thus, does not affect his deserts butonlyourgraspof those deserts, whichis all we have to implement.Alternatively,t maybe that no such variation an be developed.Perhapsourgraspof intrinsicwickednesshasnothingto do withthe agent'soppor-tunitiesto actorthe harmhe actuallydoes,so that variationsnthesecannotexplainwhywe should treatcertainagents differently.Or, perhaps here issome otherobstacle. f so, myeffortsdo not showthatdesert sunparadoxicaland cannotbe adapted o do so. However,theywouldthen showsomethingequally interesting,in my opinion. They would show that we can avoidparadoxf weshift to thewayof thinkingwhichI have set out: anargumentfordoing so, regardlessofwhether hat amounts o a reconstrual f desertoran abandoningof it.Still, it might be objectedthat this new way of thinkingis seriouslyincomplete: hat some of ourdeepest feelingsaboutdeedsin which harm sdone have no placein it. For althoughwe do sometimescriticize othersorfeel ashamedof ourselves or actionsquadishonest,orcowardly,with littleinterestin how much harm wasactuallydone, on other occasions t is justthe reverse.If youhadcrushedsomepoorchild withyour automobile, hatis what would prey on your mind, and you would get no relief from our

8/3/2019 2254027

12/13

208 Norvin Richardssayingthat no one could have seen her dart from that shadow.If the neigh-bour girl had droppedyour baby so that its neck broke,that would changefor ever your feelings about her-no matter how clear it was that no onecould have held that tiny, slippery,wrigglingbody.

Now, it is true thatthesearereactions o theharmdone rather han to thecharacterexpressed n doing it, and, thus, that they do not connect withdesertas I haveanalysed t. That is not a flawin the analysisbut a virtue,however, for the reactions are not judgements that a harsh response isdeserved!Yourfeelingsabout the neighbourgirl obtaineven though youknow it was not her fault your baby died. Runningover the child is terribleforyou even thoughyou knowyou could not have helpedit.13 One can evenfeel a similarrevulsion oward nanimateobjects-the rope the lynch-mobused, the ripper'sknife-even thoughit is plainly absurdto think of suchobjectsasmorallyresponsible or the awfuluses to whichtheywereputandthus to deserveour reactions o them. Indeed, to construesuchreactionsasjudgementsaboutdesert,and thus to want an analysisof desert to coverthem, is to miss what is most intriguingabout them: the fact that we feelat leastsomewhat ustified n havingsuchfeelings despite recognizing hatthey arenot deserved.Whatplacecan be foundfor the idea that we sometimesoughtto havesuch feelings even though they are undeserved?Perhaps they could beunderstoodand evaluatedas displays of the characterof those who havethem. For example, t seemsplausible o saythatthe girlwho causedyourbaby'sdeathought to be distressing or you to encounterbecause(a) thesightof her should remindyouof the eventand(b) thememoriesshouldbevery painfulones for you. So, your inabilityto tolerateher in your sightwould express your quite appropriate eelings about your baby's death.Were she justanotherpersonto you, despiteher intimateconnectionwiththe tragedy,that would mean it was not the tragedy or you thatit shouldhave been.Herhavingactedfaultlesslys irrelevanto allthis,of course.That is whythe soapwhichmadethe baby slipperyoughtalso to arousestrong feelingsin you. The idea is not that we are called to protectourselvesfrom suchobjectsandpeople,orto re-educate hem,aswe are whenthey displaybadcharacter nddeserveharshtreatment. nstead,what calls forourresponseis ourownabhorrence f theharmdone.The differencebetweenthese twosometimesmakesus uneasyaboutthe reactionwhichseems somehowrightandwhich we are nclinedto saywe 'cannothelp having', orinhaving t wemayalso feel that we arebeingunfair.I have notmeant o givea full accountof thesespecialreactions,however.The pointhasbeenthattogivethem noplace nan accountof desert s notto

13 'The lorry driver who, throughno fault of his, runs over a child, will feel differently rom anyspectator . .' (BernardWilliams,'Moral Luck', in Moral Luck, Cambridge,CambridgeUniversityPress, I98I, p. 28, my emphasis).

8/3/2019 2254027

13/13

Luck and Desert 209give a flawed accountof desert, because they are not reactions o it. If so,they do not providea reason for dissatisfactionwith the suggestedway ofescapingNagel's paradox.4DepartmentfPhilosophy NORVIN RICHARDSUniversity fAlabamaP.O. Box 6289University,Alabama 5486U.S.A.

14 I am grateful or the commentsof Scott Hestevold andfor suggestions rom the editors of Mind.