'l'

he Exle.malist Challenge

6l

3
The ExternalistChallenge

Sometimes the idea behind internalism is tl'rirtthere is tlrt' followinq concepttral connection between moral jrtdgernrentirncl the rvill (Nagel, 1970;McDowell, 1978, 1979, l9fl5; Platts,1979, l98l ).
I'an aeent jr,rclgesthat it is right firr l.rer to Q in circtttnstances (1, thcr.r she is motivated to in C. In other worcls, moral brinss motivation with it

.juclgement

3.I

INTERNAI-ISM

VS. EXTL,RNALISM

Suppose we debate the pros and cons of givinu to faminc relief and yotr convince me that I shotrlcl give. Flowever rvhen the oc'But casiorr arises for me to hancl over my rnonev l szrr' wait! I know | .shoulrl sive to famine relief . But you haven't convinced n're that I lravc ny reeson to do so!'And so I don't. I strtlsested earlier that suclt an cltrtblrrst would occirsion sericlr.rs puzzlenrr:nt. Having convinced mer that I shciulcl give to fanrine r e l i e ' v < l r rs c e r r l t ( ) h a v e d o n e e v e r y t h i n e y < l t rn e e d t o d o t o c o n vinct' nrt' tlrat I have a reas()nto do so. And having convinced rne thut I lr:rvc rr rc:ls()n to give to firmine re lie'- arbsentwcakness of will or sonrc othcr srrch psvchological ailtrrc - \/()u seern to havc <lone evcrlthing Yorr ncc'cl to clo to rnotivatc rnc to do so. Ptyt.zlem e n t l v o t r l c l t h r r s n a t r r r u l h ' a r i s c r e < ' a t r s t 'I,t a v i n g c < l n v i n < : e dm e that I shorrld donate, you lvorrld quitc riuhtlv expect me tcl hand over mv monev. Beliruing I shoukl scems to brinq with it my being moti,urtt.cd to - at-least absent weakness clf will ancl the like. This iclea. that moral jtrclgement has a practical trpshot, is 'internalism' (Falk, 1948;Frankena, l95U; eenerally referred to as Davidson, 1970; Williams, 1980; Railton, 1986; Korsgaard, 1986; Brink, l986; Wallace, 1990; Darwall, Gibbarcl and Railton, 1992). 'interralism' is a vague label in tht' Unfortunatell', however, philosophical literature, used to refer to several quite differt'rrt claims about the connection between moral facts or.iudgernrt'nts o n t h e o n e h a n d , a n d h a v i n g r e a s o n s o r b c i n g r r ) ( ) t i v a t c c< l ln tlrr' othcr (:ls noted bv both Brink ancl Wallace). Lr:t nrt'bcgin lx s p c l l i n g ( ) u t s ( ) r n eo f t h e s e r a t h c r c l i r ' r t ' n t < ' l u i r r r s .

.sim,pliritn'. This is a very strong clairn. lt commits us to deny'ing t h a t , b r e x a i n p l e , w e a k n e s so { ' t h e w i l l a n d t h e l i k e m a y d e f e a t a n agcnt's rnoral rnotivations while leaving her appreciation of her nroral reasons intact. And for this verv reason it is, I think, a rnanistly irnplarrsible clairn irs well. However I will not hal'e anvthing rnore to sav abotrt it here; rather I der discttssion of this versi<ln of internalisrn trntil the next chapter, when it rccmcrgcs as a c()nsequenceo'one sort of anti-Htrmcan theory of' nlotivation. More plausibly, then, the idea behind internalism is sonretimes that though there is a conccptual cclnnection between nroral -jtrdgcrnent and the will, the connection involved is the follorvine deJeasihLe one (Illackburn, l9tl4: lU7-9, forthcorning; .fohnston, 1989: Pcttit ancl Srnith. 1993a). (i, If'an asent.jrrdees that it is right for l'rer to Q in circtrrnstanr:es thcn cither she is motivateclto Q in (l or she is practicallvirrational. In other words, agents who judge it right to act in various \\'aysare so motivated, and necessarily so, absent the distortins influences of we akness of the will and other similar brms of practical rrnreason on their motivations. I will have more to say about this iclea in what bllows. And sometimes the idea behind the internalisrn requirement is not, or at least is not primarily, that there is a conceptual connection o'some sort between moral judgement and mcltivation, btrt that therc is the following conceptual connection betwecn the ('()nt('nl o a nrorirl.iudgement - the nroral facts- and ()ur reas()]s i r l a < ' t i o n ( N a g < ' | ,1 9 7 0 ; K o r s g a a r c l , 1986).

62

'l'he

Extemalist Challrnge

I'he lixtnnoList ()halbnse

(;3

If it is rieht irr asents to Q in circumstatrces O, thcn there is a reas()nfirr those agents to Q in O. In othcr w()rds, rnoral cts are cts about ()Llr reasons for action: ttre!' are ll'remselvessirnply reqrrire rne nts ()f rati()nalit!' rlr reAson. -rhis last internalist clairn rnigl'rt bc oIred as an explanation .f the previous ()ne, br it plausiblv entirils thc previous clairn. The proof o' this rvill be spelled out in sornc dctail later (chaptcr I'r), brrt in general terms the idea can be prrt like this. It is a pratitrrcle that an agent has a reason to act in a c:crtirin wav.just ir-rcase she would be motivated to act in that way i' shc werc rational ( K o r s g a a r c l ,l 9 8 6 ) . A n d i t i s a c o n s e q r e n c e . l ' t h i s p l a t i t u c l e t h a t an agent who.judges herself to have a reason t() act in a certain way - who.judges that she would be so rnotivated if she were rational - is practicallv irrational if she is not motivated to act accordingll'. For if she is not motivated accordingly then she fails to be rationar bv her own lights (Srnith, 1992). But if this is risht then it is clear that the thircl irrrn o'internalism entails the seconcl. For. according to thc third brm, the judgement that it is risht ro acr in a c er t a i n r v a y i s s i r n p l y e q u i v a l e n t r o t h e . j u d g e m e n t t h a t t h e r e i s a reason to act in that way. The reverse does not hold, however. The sec<tnd internalist claim does not entail the third. Expressivists, frrr exanrple, asrce that someone who judges it right to act in a certain way is either motivated accorclingly or practically irrational in some way, btrt denv that moral requirements are requirements <lf raticlnaliw or reAson. They thus accept the second internalist claim because thel' think that a moral judgement is the expression o'a preference, or perhaps the expression of a disposition to have a preference; but they reject the thircl because they think that fully rational creatures may yet differ in the preferences that they have, or are disposed to have. Let me eive the second and third internalist clairns names. I will call the second, the clne that may be accepted even by those rvh<r cleny the third internalist claim, 'the practicality rcquirernent ()n moral judeement'. And, for obvious reasons, I rvill call tht' rhirrl internalist claim 'raticlnalism'. These two frrrnrs ()-ir)t('rnlrlisnr a l l < l w r r s t . d i s t i n g r r i s h c o r r e s p o n d i n g f < r r n r s. ' t ' x t t . r r r : r l i s r r r .

Onc filrnr o'erxternalisrnarttounts t() atclenial o' ratirlnalisrrr. T h i s k i n c l o - e x t e r n a l i s r n i s r ' o n s i s t e n tl v i t h t h e p r a c t i c a l i t v x ' q r r i r t ' r n e n t . E x p r e r s s i r . ' i sa ts rc tvpically both extcrnalists and intcrnlrlisls i n t l - r i ss t ' n s e ( A y ' e r ' ,l ! ) 3 6 ; F I u r e , 1 9 5 2 ; B l z r c k b u r n , l 9 U 4 ) . T l r t ' r ' u r t ' crterrralists in so r as thcv urt' anti-r'zrtionzrlists, anrl vet thev arc lulso internalists in s<l r as tht'y accept tl're practicalitv reqrrir-cn ) e n t o n r n o r a l . j r r d e e r n c n t .I l t r t t h e o t h e r k i n d o f e x t c r n a l i s r n , t h c str()ngcr irrrn, arnounts t() a clcnial o' the practicalitv rcqtrircrlrcnt. Since r:rtionalisrn cntails the practicalitl' requirernent, this forrn ol'externalism thcrt'irrc cxcltrdes rationalism as lvell. Manv o f t h o s e r v h o t h i n k , u g a i n s t t h e e x p r e s s i v i s t s ,t h a t r r r o r a l . j u d g e ments pllrp()rt to be rlt'scr-iptive are externalists in this stronger Brink, l9l]6, sense (Foot, 1972; Sturseon, l9tl5; Railton, 19136;

r9 8 9 ) .
Mv task in the present chapter is to clefbnd both theserirrms o internalism - both rationalism and the practicai tv rcquirenrent asainst t\4'o recent externalist challenges. The first comes i'on-r I)avicl Brink (1986). Brink's challense is directed prirnarilv n g a i n s t t h e rw e a k c r i n t e r n a l i s t c l a i m : t h a t i s , a g a i n s t t h c p r a c t i c a l i t v rcquirernent. The second comes iom Philippa Foot (1972). FIcr challer-rse is directecl prirnarily against the stronger internalist clainr: that is, against rationalism. In what bllows I will begin by clarifying the kind of rationalism to which we are committed by the stronger internalist clairn. I then consider Brink's and Foot's challengesin turn. As a matter o' ftrct both Brink and Foot accept the stronger form of externalisrn, the forrn that excludes both rationalism and the practicalitv req u i r e m e n t . H < l w e v e r ,a s w e w i l l s e e , b e i n g a n e x t e r n a l i s t o l - e i t h e r kind involves far more controversial and counter-intuitive corrrrnitments than eit.her Brink or Foot seem to realize.

3.2

RATIONALISM

AS A CONCEPTUAL AS A

CT,AIM VS. RATIONALISM SUBSTANTIVE CLAIM

.fohn N{uckie draws a clistinction between two quite different clrrirrrs a rationalist might make (Mackie, 1977: 27-30). As I sec it,

64

'fhe

lixtentalist Challcnge

I'he Extemalist Challenge

6ir

it is Mackic's appreciation o'this clistinction that allows him ro 'crror argue br his theorv': the view that all moral thought and talk is infcted with an error o-prcsupposition; the presupposition that the world cotrtains obje<'tivelv prescriptive featrrres (Srnith, 1993a). We can best introduce this <listinction by wa\, o'an analclgv. Supposewe are interestcd in whethcr ()r ll()t therc are any witches. How are we to go about answerinq orrr rlucstion? First we must ask a conceptual question. \Arhat is orlr c()ncclt( o'a witch? l,et's supposr: we answer this conceptual questi<ln as irllows. ()ur concept ()f a w i t c h i s t h e c o n c ep t o f a p e r s o n w h o c x p k r i t s h i s r t r h e r r e l a t i < l n ship with a sllpernatural aeency in ordcr t() c:rrrsc events to happen irr the natural world. Then, second, we lnust ask a subslttrttiue question. That is, having now fixed on what our concept of a witch i s ,w e m u s t a s k w h e t h e r t h e r e i s a n y t h i n g i n t h e w o r l d i n s t a n t . i a t i n s orlr concept of a witch. I'we clo not think that there are any supernatrlral agencies fbr anyone to have a relaticlnship with, then we will answer this srrbstantive question in the negative. We will say th:rt there are n() witches. Mackie's idea is that, when we ask whether there are any rnoral facts, we h:rr.'cto irllow cxactly the same procedurc. Wc must first o f u l l a s k u ( ' ( ) n ( ' ( ' l ) t u a lc l u t ' s t i o n . W h a t i s o t r r c ( ) n c e p t o f : r r n o r a l fact? Nlackie iurswcrs tllat ()rlr c()ncept of a morirl act is the cor-r'objcctivclv cept of an prcscriptivc' fattrrc o' thc world. And then, according to Mackie , wc rl)ust g() ()n to irsk a strbstantive qtrestion. Is there anything in the world answering to our concept of a rnoral fact? Mackie's answer to this question is, mously, that once we are clear about what it is that we are looking frrr, we see that there are no moral facts. For we see that our concept of an objectively prescriptive ature is not instantiated anl,where in the world. I said that Mackie draws our attention to two clifferent claims a rationalist might make. This is because,in light of'his distinction bctween conceptrral claims and substantive claims, raticlnalism mieht now be taken to be a conceptual claim: the claim that otrr concept o' a moral requirement is the concept of a rcason irr action; a requirernent of rationality or reason. Or altt'rnativeh, 'I'hut rationalism might be taken to be a substantive clairrr. is, rationalists rnight be telling us that there are rcqtrirt'nrt'nls o

rationality or reason corresponding to the various moral reqtrirt'me nts. Taken in the first way, rationalism is a claim abottt thc bcst analysis o'moral terrns. Taken in the second way, rationalistn is a claim about the deliverances of the theory of rational acti()n. As I see it, when Mackic tells us that our concept of a moral fact is the concept of ;rn objec'tively prescriptive feature o'the world, he is telling us that the rationalists' conceptual claim is true. And when he tells trs that therre are no objectively prescriptive features in the world, he is telling rrs that the rationalists' substantive claim is false. That is, as I see it, Mat'kie's argument ftrr the error theory may be reconstructed as ollclws. Conceptual trtrth: I'ascnts are rlorally required to Q irt circttms t a r ) cs e (l then there is a requirement o' ration:rlitl' ()r reason for all agents to Q in (l circunrstirnces

Srrbstantivc'claim: Thcrc is no requirement of rationalitv <lr reAs()nfrrr all agents who fincl thcmselvesitr (l to circumstances (lonclusion: Agents are not morally required to Q in circumstances(l

'fhat

we are able to reconstruct Mackie's argument in this way is important, for it shows that in defending the rationalists' concepttral clairn we do not thereby beg any questions. Even if we acccpt the rationalists' concepttral claim, we rnllst still go on to clencl the rationalists' substantive claim. Ancl conversely, even if lve deny the rationalists' sttbstantive claim, we must still enuage with the rationalists' conccptual claim. This clistinction between rationalism as a conceptual claim ancl rationalism as a substantive claim is to be central in what bllows. For note that the stronger internalist claim - what I have callecl 'rat.ionalism'is simply a claim about our concept of rightness: it is a claim about the content o' an agent's judgement that her action is right, not a claim to effect that judgernellts with suclt contents re tnre. Moreover, note that it is this conceptual claim that errtails the practicality requirement. The lruth o'the stttrstartt i v e < ' l : r i n ti s s i m p l v n o t r c q t t i r e d f < r r t h a t e n t i r i l n r c n t t o h o l c l .

66

'l'he

Lxlenrctlist Challenge

'I' he llxtem rr'list Chullense

67

It is thus ratiorralism as a conceptrral claim that is to be at issue in the present chapterr, not rationalism as a substantive claim. Rationalism as a substantive clairn will come up fcrr discussion in lirtcr ctrapters, brrt f<rr norv the focrrs is to be purely conceptual.

3.3

BRINK'S

.AMORA[,IS'[ '

CHALLENGE

In 'Extcrnalist Moral Realism' David Brirrk argues that we must rejcct the practicalityrequirernent. Sincc the rationalists'cor)cephis argument thus tu:rl claim entailsthe practicali$'requirement, threatcnsto refute rationalismas well. Here is Brink. is skepticism Much rnoralskepticisnr aboutthe objectivity of'moralitr', that is, skepticisnr about the existence of moral facts. Btrt another traditional kind of skepticism accepts the existence of mrlral cts and zrsks whv we shotrld care about these facts. Anroralists are the traditiorral way of representing this secor.rdkind o skepticisrn; the anroralist is someone who recognizes the existcnce o'nrr>ral considerations and remains unmoved. 'I'hc . . . Irlendcr of the practicality requirement] . . . must disr n i s s t h c a r r r o r - u l i sc t hallenge as incoherent... We may think that t h e a r r r o r u l i s t < ' h u l l t ' r r g ei s c o h e r e n t , b u t t h i s c a n o n l y b e b e c a u s e 'inverted rvc conirse rnor:rl st'nses o'terrns and commas' senses of' those sanretcrnls.. . Thrrs... al)parent amclralists... remain unmoved, not by n'hat they regard as moral considerations, but only b1' what otllers regard as rnorirl considerations. not take the The that...fthis] ...does problenr...is arnoralist's challenge seriously e nough . . . We can imagine someone who regards certain demands as moral demands - and not simply' as conventional moral demands - and yet remains trnrnoved . . . [If] . . . we are to take the amoralist challenge seriouslv, we nrust attempt to explain why the amoralist should care abour morality. (1986: 30)

Brink's argument is simple enough. According to defenders of the practicality requirement, it is supposed to be a conceptual truth that agents who make moral ()' judgementsare motivated accordingly,at Ieastabsentweakncss will anclthe like. But far iom this being a conceptualtrtrth, it isn't

ottt any sort of truth at all. FOr amoraliststlse nroral tel'nrsto pit'k theverysameproperticswcpick<lutwhenwettsenl<lrlrltct.llls' llv tht' Their tise crfmoral terms lt)ay tht:rcforc be reliably guidecl [Jttl terrns. ()lrr of thtlse tlses nroral facts in tlre sanrt' \\,il\, as rlr to all at arnoralistsclier iom trs in ttrat thcv see no reas()n lvorcls' what thel- thtrs uke to llc trrtlrallv required' In ottrer acmotivated being without amoralists rnake moral itrclucrrlertts irof practical sort anv cordinglr', ancl withotrt sttIrins ionl false. thus rationaity either. The prac'ticalit' requirement is As Brink notes, cleftl.clers''the requirentetlt ha'e gerreralll' to this challenge bv boldly denying that am<lralists not responclecl exist'Ancln()rC()ul(ltlrcr,withanycredibility,fbl.anroralists are amol'lg thc mot'c poptrlar heroes o' both philosophic:rl Plato's fantasy u, ,',.r,-philostiphical fiction' Brink mentions ancl Dickens'sUriah Heep' But nor are amoralists Thrasymachtts There are' ater all' re:llconfinecl to the worlcl of make-believe' story is life sociopathslike Robert Harris, tre thrill-killer whose b1' Gary Watson (1987)' Harris faithfully retolcl ancl ar"ralysecl and that he clairnstrat he knervthat what he rvasdoing waslvronq It therefore conflict' no he felt that sintply chose to do it anl'\'vay; of weakness from seemsquite wrong ,o ,.,ppnsc tltat he suffered irrationalitv will, or, perhaps, fiom nl' other kind of practical either. to itrsistis <t'the requiremetrt h:rve tettclecl what def'enclcrs nce of existe therefore rather that' properly clescribed' the amoralists is r-rotinc,t'siitent witS the practicality requiremellt. judgements at For, they clainr,amoralistsclo not real\ make moral same ProPerthe otlt pick all. Even if they clo use moral words to not really ties that we pick otlt when we use moral words' they do to be judge u.,, be right and wrong; rather they judge acts i.iglt, ancl .wrongi. Thu, is to say they use moral words in a mentions' diflerent sense;in the inverted commas senseBrink 'Q-itg is right' as sentence According to Hare, for example, the'$-ing is right'; but rather mean not does used by' an arnoralist juclge to be means'Q-itg is in accordancewith what other peoplc act that an the such' as And, right' (ar.] tgrz' 724-6,163-5)' 'right' to without beins either motivated Q airoralist mal'juclge it of will is no counter-example ttl *uktess oto $ or strffering

68

The Extemalist Chaltcnge

T'heExtemalist Challznge

(i1)

the requirement. For the requirement tells us that those who judge it right to Q are motivated accordingly, absenr weakness of will, not that those whojudge is 'right' to q u." motivated accordingly, absent weakness of will. Now Brink thinks that this i.vcrtecr commas responsedoesn't take the amoralist challenge 'scri'usly' enough. And I must confess that I share his misgivings, at least as regards the details of Hare's version of the response. For, as Brink points out, there seemsto be nothing incoherent abo.t the ide of an amoralist who claims to have special insight inro what is reailvright and wrong; an amoralist whosejudgements abour what it is ,igt t arra wrong to do are therefore, even by her own rights,out of lilre with thejudgements of orhers. Bur if this is right, i'ren the jucrgemenrs of amoralisrscan hardly be thought of ai judge-e.,ts-ubJut what other people judge to be right and wrong. Despite these misgivings, however, I itrlnt< that the inverted commas responseto the amoralist challenge is along exactly the right lines. In what follows I wanr rherefore io give a rio part reply to Brink. First I will saywhat the inverted .o-L., ."rpo.,r" ."ily amounts to; how it differs from what Hare says.And second I will saywhy defenders of the requirement are right to think that the requirement is a conceptual truth.

3.4

REPLY

TO

BRINK'S

CLAIM MAKE

THAT

AMORALISTS

REALI-Y

MORAL

JUDGEMENTS
As I see it, defenders of the practicaliryrequiremenr are right to say that amoralisrs do nor reaily make mral judgementsl trrey simply go wrong in trying ro saymore rhan ttris. fn1 point is not that amoralistsreally makejudgements of some other Lind: about what other peoplejudge to be right and wrong, for example. The point is rather that the uerybestwe can sayabout amoralisisis that they try to make moral judgements but fail. In orcler to see why this is not ad oc,consider an analogy. There is a familiar problem aboui the conditions .ncler whi<.h we should say of someone that sl-re rc:rllv nrakcs <.olorn. irr<lg<._

c:ttr 1985:chapter 2;Tawil, 1987).The problerrr ments (Peacocke, be brought out by reflecting on the caseof someone, blind iorn birth, who has a reliable methocl of using colour terms. We rnight imagine that she has been hooked uP to a nlachine from birth that allows her to feel, through her skin, when an object has thc appropriate surface reflectance properties. Now such a person certainly has a facility with colour terms, a facility that allows her to engage in many aspectsof the ordinary practice of colour ascription. For she uses terms with the same cxtension as our colour ternts, and the properties of objects that explain her uses of those tcrlrs are the very same properties as those that explain oltr uses of colour terms. (This is similar to what wc said earlier abottt the amoralist'suse of moral language.) And we can even imagine, i'we like, that her colour judgements are far more accurate and reliable than those made by sighted fblk. When she makes colour judgements, she is therefore not appropriately thought of as making judgements about what other peoplejudge to be red, green and the like. (This is again similar to rvhat we have said about the amoralist.) However, despite the facility such a blind person has with colour language,many theoristshave thought that we should still colour concepts or mastery of colour deny that she possesses terms. For, they say, the ability to have the appropriate visual experiences under suitable conditions is partially constitutive of possession of colour concepts and mastery of colour terms (Peacocke,1985: 29-30, 37-8). And what such theorists thereby comrnit themselves to saying is that, despite her facility with colorrr terms, such a blind person does noL reall) make colour juclgementsat all. They do not have to saythat she is really making judgements of some other kind, of course. Rather they can insist that though she is trying to make colourjudgements, becauseshe of colour concepts, she fails. when doesn't count as a possessor are red','Grassis green'and the like, she is she says'Fire-engines therefclre best interpreted as using colour terms in an inverted 'red', grass is cornmas sense:she is saying that fire-engines are 'green' and so on. It is, I hope, clear that the structure of this debate over the iricrltor masteryof colour terms is in cruciitlrcspt:t-ls <'ort<litions l }irrk w i t l t r '(w ' te ' l r g r t g t i' r r l i < l r lt o t l r < ' s t r - r r c t r t r t ' < l ' t l t t ' r i t ' b ai t t

70

'l'lp

lixtr,ntntist Chalkngc

T'he Extnnalist

Challznge

7|

( ) n c s i c l t s : i v st h a t a s u b j e c t h : r s nrasteryo'colour terms (moral t e r r n s ) , a n c l t h t r s r e a l l t ' m a k e s c o l o r r r . j t r c l g e r n e n t s( m o r a l j u d g e _ rrr.nts), .nlv i, rr'der cert:ri. r'..rliti'.s, being in the psycotogical_statethat we cxpress rvherr wt' rr'rke cor'u..1.,crger.rnts lmorar .irrcluerrre.tsc ) ntails tra'irrq ^. llppr()pr-iate visual experience (moti'irti.n). The othc'r siclc cr.'it's this holcling insiead that the a b i l i n ' t o . s e a t e r r n w h r s e r r s ei s r r . l i a b h ' c x p l a i n e d b v t h e r e l e v a n t p r . p e . t i e s o f ' . b j e c t s i s c r r . t r s h r . r r . r ' c l i th e r w i t h m a s t e n , o' c,lorrr tcrrns (r'.ral ternrs) .'cl thc: abilitv realll, to make ..,io,.,. . j r r d e c r n c n t s ( r n o r a l j u c l e e r n c n t s ). I l a v i r r q t h e a p p r o p r i a t e v i s u a l expe.ience (rn.ti'atir.) '.der appr.priatc co.clitions is an ent i r e l v c . . t i . s e n t , a n d o p t i c l n a r ,c x t r a . T h e c l e b a t ei s a r e a l o.e, so horv itre rve to decicle lvho wins? Ir'asire so'co.e objcctiru that th.se wh's.y that the capacity to have ccrt:rin visuarlexperiences is partiallv constitutive of mastcn o col.trr ternls d. not take 'seri.trsly' encltrgh the challenge p<rseclbl pc.ple wh. can rcliably say '()rass is-green', .Fire-ens i n e s a r e r e c l ' , a n d s o o . , w h i l e y ' e rb t eine compltely blind. Supp.se tht' ,tljc.t,r i.sists that since blind peopi. .u. reliably .ie c'l'rrr tc'nrs i. this wzrr',itjtrst f.llolvs that thy have full rnasrery of coklrr tcl'rns. \4'otrld the objecticln be a goocl one? I do not think so. F.r ther .bjecti,. si.rplv ass.mes the conclusion it is srrpp'scd ro be :.rr-sui.s irr. It assrrnresthat blind people have r l r a s t e r ao c o l o u r t e r n r s , s o r n e t h i n g t h a t t h o s e w h o think that rnast.r.!'requir-es the cap:rcity to have the appropriate visual experiences uncler the appropriate cor-rcliticlns dny. I t s e e r n sr o m e t h a t B r i n k ' s a m o r a r i s r c h a l l e n g e i s f l a w e d j u s t in this rval'. He puts a prejudiciar interprerario. n the amoralisr's .cliable .se of moral terms. He :rssumes that the amoralist's reliable use is evidence of her mastery of those terms; assunres that being suitably' motivated under the appropriate conditions is not a conclition of mastery of moral tertns. But those who accept the practicalitv reqtrirement do not accept the account of what it is to ha'e rnasterv of r'oral terms that makes this prejudicial interpretaticln of the amoralist's use of moral terms appropriate. vtlhat this srrgeestsis that, in orcler to adjudiit" " clebatc with _ Bri.k, what we really.eed is a' increpe nclent reason for acr.<,1rtirrg ()ne or the other account of mastery. I n w h a t f < l l l < l w sI w i r r r

-I'hc zrrgtrtllt'rtt therefore to provide such an independent reason. dt'i'ncl wh<l those by offered mastery is to be that the account of is it alont' the practicality requirement is to be preferred because able to provide a plausible explanation of the reliable connc<'tiott between moral judgement and motivation in the good and stron{willecl person.

3.5

AN ARGUMENT

FOR THE

PRACTICALITY

REQUIREMENT

All we have said so far about the strong externalists' account of rnoral motivation is that, bv their lights, it is a contingent and rationally optional matter whether an agent who believes that it is right to act in a certain wav is motivated to act accordingly. But more quite evidently needs to be saicl' By ail accounts, it is a striking fact about moral motivation that a change in motivation follows reliably in the wake of a change in ^t least in the good and strong-willed person. A moral jud.geme.nt, plausible theory of moral judgement must therefore explain this itriking fact. As I see it, those who accept the practicaligv requirement can, whereas strong externalists cannot, explain this striking fact in a plausible waY. Suppose I am engaged in an argument with you about a funda-"ntui moral question; a question about' say, whether we shorrld vote for the libertarian party at some election as opposed to the social democrats. In order to make matters vivid, we will suppose that I come to the argument already jgdging that we should vote for the libertarians, and already motivated to do so as well. Drrring the course of the argument, let's suppose you convince tne that I am fundamentally wrong. I should v6te for the social democrats, and not just because the social democrats will better promote the values that I thought would be promoted by the libertarians, bttt rather because the values I thought should and woulcl be pr<> moted by libertarians are themselves fundamentally mistaken. () Yorr get me to change my most fundamental values. In this s()rt sitrration,what happens to lny motives?

'f

he Extentalist Challznge

l-he Extunalist Chalbnge

Moreover, and importantly, note that defenders o-the reqtrirement are in a position to insist that what an agent is thus tn<ttivated to do when she changes her moral judgement is preciseh' what she judges it right to do, where this is read de re and not dc dicto. Thus, if an agent.judges it right to Q in C, and if she has not derived this judgement from some more fundamental judgement about what it is right to do in (1, then, absent weakness of will and the like, defenders of the practicality requirement can insist that she will be motivated non-derivatively to Q in C. This is because, on the rationalist alternativer, a notr-clerivative desire to in Cl is what her judgement that it is right to Q in C causesin her, or because, on the expressivist alternative, the judgement that it is right to Q in C is itself-just the expression of such a non-derivative desire. In the example under discussion, then, in deciding that it is right to vote for the social democrats, defenders of the practicaliW requirement can insist that I acquire a non-derivative concern for social democratic values. But now consider the' second Answer, the answer favourecl by the strong externalist. She will sa)'that the clender of the practicality requirement has convenientlv overlooked a crucial part of the story: narnell', the stipulation that I am a good and strongwilled person. She will therefore insist that what explains the reliable connection between judgement and motivation is a motivational disposition I have in virtue of which I count as a good person. In other lvords, what explains the reliability of the connection is the conlent oJ'my moral motiuation. But what exactly is the content of nry moral mcltivation, according to the strong externalist? Before the argument began I was motivatecl to vote for the libertarians. Could it be that it was r-ny having a non-derivative concern for libertarian values that rnade me count as a good person, when I judged it right to vote for the libertarians? Evidently not. Aer all, as a resltlt of the ensuing argument I have come to reject my earlier jttdsement that it is right to vote ftrr the libertarians in favour of the.iudgement that it is right vote for the social democrats. But since, on this way of seeing things, my initial motivation was not itsel rationally mandated by my earlier judge-

Though the precise a.swer to this quesrion will 'f c.urse crepend, intn alia, on the very point at issue,this rnuch at leastcan be acceptedby defendersof the practicalityrequirement and strong externalist.s alike. If I arn a goocl and strong-willecl person then a new morivation will lbllow in the wake of my new juclsement. s<r let's adcl in the assumption that I am a goocr o'r ,ti.r,-,g-willecl 'o. person. Then, since I n. l'nger the .irrdgeit right to vote Iibertarians, I will no longer be moti'atecrto clo so. Ancl since I havc cor'e t<lj.dge it right to vote for the sociarcrenrocrat.s, I r,r,iil n.w be mcltivatedto do that insteacl. Thc qrrestionis: hclw Are we t<r explain the reliabirif of this connecti.. berweenjucrgement and m.tivation in the good and strong-willecl perso.? o, o." *" to explain whv, under a range of counterfactual circumstances, the eood and strong-willedperson's moral mrtivations will arwavs all in line behincl her newly arrived at moraljrrdgements? As I seeit, there are only rwo possibleanswers. ()n the one hancl \\'e can say that the reliable connection betwee' jtrclgement ar-rcl rrrovitarionis ro be explained internally: it follows clirectlyfrom thc c()ntentof rnoral.judgementitself. The icleawill then be either that the belief rh,t :r. act is right producesa correspo.ding m.tivatio' (this is the rati.nalists' alrernative),or prhaps'ihat the attitrrde of accepting that an acr is right is itself irientcatwiththe ltate of being motivated (this is the expressivists'). ()r, <t' the other hand, wc can say that the reliable corrnection between judgenrent and motivation is to be explainecl externalfu: it bllows from the contenr of-the motivational disposirionspossessed by the so.d a'd strons-willedperson. Those who defe' the practicality requirement opt for the first answer, strong externalistsopt for the second. (lonsider the first answer.since thosewho defe'd the practicality requirement think that it is in the narure of moral jugenre.t that an agenr who .judges it right to Q in ci.cumsns C is motivated to $ in c, at leastabsentweakness of will or some other such psychologicalfairure, they will insist that it comes as n() surprise that in a strong-willedperson a change of morar r''tivation follows in the wake of a changeinmoral jirdgernent. F.r rhirr is.iusta direct consequence of the practicaliiyrequirc,rr.,,rr.

74

The Extentalist

Challenge

7'he Extentalist Challen ge

75

ment - since it wasjus_t a wholly contingent ancl rationally optional extra - so the mere fact that I have found reason to change my j_.dgementgivesme no reasort. change this motive. I may therebre quite rationally contintre to be have a desire to vote for the libcrtarians; though of course I wourd have to juctge that in so doing I arn modvated to do something that t nwj"udge wrong. Having a nonderivative concern .r liberatarian"vaes while .iudging it right ro vote for the libertarians is thus nor whar makes me a good person. For it cannot explai. wh' I cha.ge my motivation when I change my judgementwhat this forces the strong externalist t. admit is that, on their wav of seeing things, the motive in virtue of which I am to count as a good person must have a content capable .f explaining nor just why I am motivated to vote for the libertarians when I judge it right to vote for the ribertarians, bur arso whv I stop beig motivated ro vore for the liberraria.s when I give up judgi.rg ttat it is right ro do so. And the onry motivationaiconttit cJpaue or playing this role, ir seemsro me, is a rnotivation to do tire right thing, where this is now read de dictoandnot de re. Atbottom, the strong externalist will have to say,having this selconciously moral nrotive is rvhat makes me a good p"rro.' N<tte that if this were the conrent of the goocl person,s motivati.ns, then rhe srrong exrernalisrwoulcl indied b able to explain the reliabilitv <lf the conne.tion benveen m'ral judgeme't ancl nrotivation.A change in the g'ocl person's rnotitions would follow a change in her moral judgernents because her motivationswould be deriveclfrom herjiudglments togcther with her sefconciously moral motive. Thus, according to this story, when I no longer believethat it is right ro vore for the libertarians, I lose a deriueddesire to vote br them, ancl when I come to berieve that it is right to vote for the sociardemocrats,I acquire a d.eriued desire to vote for them. But my motivations are in each case derivativebecausethey are clerived from my current judgement about what the right thing to do is rogerher with my asi morar motive: a nonderivative concern to do what is righi. However, if this is the best explanation rhe stiong exrernalisr can give of the reliable connection berweenmoral.i.cigenrcnt a.cl nt'tivation in the good ancl strong-wiilecl perso' ir.,,i'it s(.(,rrs r()

rrre that we have a straightforward reductio. For the explanation is only as plausibleas the clairn that the goocl person is, at borrorn, rnotivatedto do what is right, whcre this is read de dictoand not de re, and that is surely a quitr- irnplausibleclaim. For commonsense tellsus that if gooclpeople.jrrdge it risht to be honest,or risht to care for their children and iiends and fellows,or right for people to get what they deserve, then they care non-derivativelyabout t.hesethings. Good peoplc care non-derivativelyabout honesty, the weal and woe of thcir chilclrcn and friends, the well-being of their f'ellows,people gettine what they deserve, justice, equality, ar-rdthe like, not -jtrsronc thing: doing what they believe to be rieht, where this is read de dicto and not de re. lndeed, commonsensetells us that beins so motivated is a fetish or moral vice, not the one and only rn<lralvirtue. It is worthwhile underscorinsthe presentobjection by comparing it to a related objection of Bernard Williams's to the kind of moral philosophy that ernphasizes impartiality (1976). Williams askstrs to consider a man who, when faced with a choice between savinehis wife or a stranger,choosesto savehis wife. Many rnoral philosophers think that, even in such a case, a morally good person would be moved bv impartial concern; that this man's motivating thought would therebre have to be, at best,'that it was his wife, and that in situationsof this kind it is permissiblero save one's wife'. But, Williams ob.jects, this is surely wrong. It provicles the husband with 'one thought too many'. And in order to see that this is so he asks us to consider matters from the wife's perspective. She would quite rightly hope that her husband's 'motivating thought, fully spelled out' is that the person he saved rvas is wife. lf any further motivation were required then that would simply indicate that he doesn'r have the feelings of direct love and concern br her that she rightly wants and expects.He would be alienated from her, treating her as in relevant respecrs just like a stranger; though, of course, a stranger that he is especiallywell placed to benefit (Williams,1976: l8). The present objection ro externalism is like Williams's objection to the kincl of rnoral philosophy that emphasizes impartiality, only nrore powerirl still; br it does not require the assumprion, <'<lntr<lversial by the lights of'sonle, that ntorality itsel'embra<:es

/l)

I'he Exte.ntalist Chalknge

The L)xtentalist Challenge

77

partial values like love ancl iienclship. For the objection in this caseis simply that, in taking it that a good person is motivated to do what she believesright, whcrc this is read de dictoand.not de re, externalists too providc th(' nrorally good person with 'one thought too man)". Thcv alit:nate her from the ends at which morality properly airns..|usl as it is consrirutiveof being a good lover that you have dirt'r't ('()nccrnbr the person you love, so it is constitutive of being a rn<trirllv g<loclperson that you have direct concern for what you rhink is right, where this is read dereand not de dicto.This is somerhinl rhar rntrstbe conceded even by those moral philosopherswho think that the only right course of action 'I'hcy is one of impartiality. r()() tnust agree that a morally good person will have a direct anrl norr-rlerivative impartial concern; her concern for impartiality lrlust rrot itself be derived from a more basic non-derivativeconccrrt de dictoto do the right thing. We have therefore found a decisivereasc)n to reject the strong externalists' explanation of thc reliable connection between moral judgement and motivarion in the good and strong-willed person.For, in short, the strong extcrnalists'explanation commits us to falseviewsabout the content of a good person's motivations; it elevates a moral fetish into the one and only moral virtue. And the remedy, of course, is to retreat to the alternative,internalist, explanation of the reliability of the connection between moral judgement and motivation. But if we do that then, of course, we have to accept that the practicalityrequirement is a constraint on the content of a moral judgement after all. The conclusion is important. For it means that we now have the independent reason we needed for giving an account of mastery of moral terms according to which the practicalityrequirement is itself a condition of having mastery. Only so can we explain the reliable connection between moral judgement and motivation in the good and strong-willedperson. Brink's 'amoralist' chalenge thus collapses. For despite the facility they have with moral language, amoralists do not have mastery of moral terms, and they therefore do not really make moraljudgements. The fact that they make 'moral' .judgementswithout being motivated or suffering from practical irrationaliry thus provides us with no challenge t<r the practicality requirernent.

3.6 Foor's'ntlqurrrE'

CHALLENGE

Shoulcl we accept not -iust that there is a conceptual connectioll betlveell moraljudgement artcl rnotivation, but that there is also a t'onceptual connection betwccn cts about the rightness clf actions ancl cts about what wc have reason to clo? In other words, shor.rlcl we accept the ratiotralists' conceptual claim zrs n'ell? I'hilippa Foot argues that we shoulcl not. 'Morality as a System of llypothetical lmperatives' Foot has Ir-r First, she wants to explain why we should reject rnain aims. trvo Itrnt's claim that moral fcts are cts about our reasons for ruction. Ancl second, she wants to set out, in schematic firrm at Icast, her own preferrcd altcrnativc: an anti-rationalist, instir t r t i o n a l a c c o u n t o f m o r a l c t s .I n w h a t f o l l o w s I w i l l i r s t c o t r s i d e r Foot's arguments against l(ant, and I will then sAy,lnore positivelv, rvhl I think we should accept, with lCant, the rationalists' c6ncepttral claim. Foot begins by reminding us of I(ant's distinction between ( aregorical ancl hypothetical imperarives: that is, thc distinction l)etween requirements that are binding on someone conditionally on her having a certain desire, and requirements that are binding ()n someone unconditionally, that is whether or not she has a t t'rtain clesire. The first are the hypothetical imperatives, the st'cond are the categorical imperatives. This distinction is import;lltt for l(ant because he claims that moral requirements are cat('qorical imperatives, not hyPothetical, and he seems to think (lurt granting this is tantamount to the admission that moral r t'quirements are, just as the rationalist supposes. reasons for ;r<'tion binding upon rational creatures as such. Hgwever, accordinq to Foot, when we try to make Kant's claim more precise, it lrr'<.qrnes either uncontroversial, and so uselessas a defence of the rrrtionalists' conceptual claim, or else it becomes unacceptable, rtttt ulia because simply equivalent to a statement of that claim. l.r't rne take the horns of this dilemma in turn' l"oot thinks it should be agreed on all sides that moral juclgenr('nls ar.('irppr()priately expressed by judgements employing a 'shottld'. She elx( ,rl(.gor.ir.irl r.i r t h t ' r t h a n a h y p g t h e t i c a l , u s e o f

'l'he

lixtemalist Challenge

7-hel')xtqnrtlisl OhalLengt

71)

plains this clisti.cri.. lr' t'xunrple (lg7z: 159). Suppose a man warlts to go honre and irr olrir.r to <lo so he has to catch the noon t r a i n . w e t h e r e f < r ^ r t c l l h i r r r ' v r r r s h o t r l c rc a t c h t h e n o o n t r a i n . ' H o w e v e r s u p p o s r ' t r r t l r . r ' t h ^ t , . i r r s tb e b r e n o o n , h e s t o p s w a n t i n g to go honre. Accordirrg lo Fort ir wotrlcl then be false for us to tell hirn that he should t'att'lr th. '.on train. The claim that was earlier true is now irlsr'. I lt'r. wc have the hypothetical use of 'should'. F o r t h c r . r r l r . l r l r < ' ' s h . r r l c l ' c r a i mj u i t b c f o r e n o o n i s clefeated bv the fa<'t thrrr tlrr, rrrirrr has no clesire then that will be served by his doing whal wt' sai<l ht, should clo. (lontrilst this use.''sh.rrl<l'with the case in which a r.an b e h a v e sc o n t r a r y t o s o n l ( ' r r r . r ' l r l . t ' r ; r r i r c n r e n t .S u p p o s e h e i s c r u e l . A s F o o t p o i n t s o u t , i t i s r r r r t ' r ( ) s : r v. h i m t h a t b e h a v e d a s h e shouldn't even if his bcirrg .^rt'l irr these circumstances is preciselv what he has to cl. i lrt' is r() ?rct in a way that serves is interests and desires. Btrt in that <'lrst. it irllorvs that moral requirelnents are, in one importart s('trs('lrt any rate, categorical. Since wc are not obliged to withdruw tlrt' rrr'ral 'shoulcl'when we find out that acting morally sen'cs n. irr(crcst <lr desire of the agent, moral requirenrents are appr.priaterv expressed by a categorical, as opposed to a hvpothetical, rrseo'.shoulcl'. 'shorrld'? what is the ground of the categ()ri('ar .se of F o o t t el l s 's that it has to clo with the kincls o c<lnsicleration required to support the 'shotrld' claint. More specifically, whereas hypotheti_ 'shoLrld' cal clainr.s can be strppr>rtecl only by showing'that the acti()n in question servesa desire or interest of the agent, categori_ 'sh..ld' cal claims have to be supported bv menlioning some releva.t ature of the agent's circumstances. It is the ii...,-stances an agent faces, circumstallces that might be faced by another agent with difrent desires and interes, that makes an action rnorally required. Horvever, as F.ot remi.ds us, though in this sense moral re_ qtrirements are indeed categorical, they are like many other more nrundane requirements: for example, requirements of etiquette. Strppose sor)eone acts contrary to a requireme.t of etiquetie. Hc replies i. the first person to a letter written to him in thc third. -lhen, as F.ot notes, it is true to sav of hirn that he has a<'tc<l as lr. s h . r r l r l r r ' tt ' r ' e n i ' h t ' t e l l s r r s ( t r r r l v ) r l r a t a r . t i n gi n a c t . o r r l l r n < . r . w i t h

tlre reqttire[]ents of etiqtrettc itr no \\'Av setf('s tltl\ itttt'lt'st ot' t t ' t t t t t s tI t ' 1 t l r o f h i s ; t h a t i n o r d e r t o : r c h i e v c u ' l t l t tl r e \ \ ' : r r 1 tIs <lesire s tnoralitr.'slt<lrrlrl' irr the first person. For, as rr'itltreqttir-t'rr)ctlto r l a i m s o f e t i q u e t t e a r e s t r p p o r t c d l t v t ' i t i n c s o n r e t ' e l e v i r t l t ( ' a t l l l ' ( ' thltt tnigltt bt''lrt't'tl ,it'trttttstllltccs o'the agent's circumstatrccst l r v s o m e o n e w i t h d i f f e r e n t < t c s i r t ' sl t t t r l i n t e r e s t s . \ r { l t i r t m a k t ' s a requirement a requirernent () ('(i(ltlctte is n<lt that actine irt tlrt' rt'levant w:rv serves an intt'r'cst ot' rlt'sirt: o thc ztge'ltt. T h e f z r c t t h a t r e q u i r e n l ( ' r t s o t ' t i c l t r c t t e a n c l r e q t t i r ' ( ' n t e r l t s( ) rnorality seern in this rcspt'<'t to bt' on all ixrrs providt's F<lot rr'ith t h e m a t e r i a l s t o c o m p l e t t ' h t ' r ' l l l ' q t l t l ) e r t (o t r t h e f i r s t h o r r - ro 1 ' t h t ' < l i l e m m a . l - o r n o o n c l r o l < l st l t a t r e q t r i r e l l ) e l l t s( ) ' ( ' t i q t l e t t c a l ' e lcquirements of reasotr. Sotttt'<lnt'rvh<l ails to livc rrll to tlrt' d e t ' t r t e c lt o l l t ' |equirements o-etiqtrcttc is ttOt. or that reas()l't, i r - r ' a t i o n a lB . r r t i n t h a t c a s ( ' r v t ' l t a t ' t ' a n e x a r n p l e r > ' as c t o l r e q t r i r . t ' 'shotrltl' l r r e n t s : r p p r o p r i i r t e l v t ' x p I ' t ' s s t ' r [b v a c n t e g < t r i < ' atl t s e o ' t h a t a r e n o t t h e t n s e l r , c sr t ' r l t t i r e l t l c n t s o ' r e a s o n . A t r d o t r c e t h i s i s are clltst'en, it ftrllows that adrnittirrg that rrtot'al reclttir-etncttts t ' g o r i c a l i n t h e u n c ( ) n t r ( ) \ ' e r s i a ls e n s e o f b e i t r g a p p t ' o p r i a t e l v c ' x 'shottld' is thtts insttfficicnt t<' sltotv pressed trsing a categorical a l ' e r c q t t i r e t n ents of rerasott. tlrat moral reqttircrrtents This gives Foot the materials t() al-gtleItct'casc otr tlre se'coltrl lrorn of the dilemma, Ibr those u'ho insist tltat trtoral reqtrir.t'r n e n t s a r e c a t e g o r i c a l i m p e r a t i v c s i n t h e t r t ' t h c t -s e t l s e o ' l r t ' i r r q rcquirenrents of reason have to sil\' $'hat r('as()tr tltcr' Itavc irr l hat this : r r p p o s i n g t h i s t o b e s o . T h c l ' c a t r , < l f ' c o t t r s c ,s i r r r p l vr r . s t a t i s s o . B u t , : r c c o r d i n g t o F o o t , . s t t c ha t r a s s c r t i o t t i s c v i d c n t l v c ( ) l ' l { l ' ( ) s' ( ) t t l l t c r l ( ) r t ' r s i a l a n d , f c l r t h a t v e r y r e a s ( ) n ,t t t r i r c c e p t a b l c .I t l ' r . r r t ( r a t i o n a l i n , u vic'rr.sltt' o ' p r a c t i c i t l v i e w rlre orthcldox philosophicul \ l r n ) s r . r pi n t h e i r l l o w i n g l c r n l s . l rtis s i n w h i c h i t r l r : r r il t t s o t t t t ' t r ' a r ' < l t ' f i ' a I r r : r t i o n a la c t i o t r s are thctsc ()r t() ()wn prlrposcs,doing what is calculatedto be clisarlvalltilH,^e()tls r'trstrate h i s e n c l s . F o o t . 1 9 7 2 : 1 6 2 ) Irr otht'r words, practical rationalin' is a svstenrof ltrpotlteticirl i r r r p c r a t i v c s , n ( ) t i t s ) ' s t e t n o ' c l t t e g < t r i c a l i l n p c r : r t i ' i ' t ' s .A n r l s h t ' 'lrttrttot'itlitv rlot's rtot t l r t ' r ' t ' i r r tc ' l r a w st l r c i n c v i t a b l c c o n t ' l r r s i o t r :

80

The lixtematist Chaltznge

T he I) xternali st Challenge

8l

necessarily involve any such thing' (lgz2:162). According to Foot, we should therefore re.jectthe rationalists' claim that noral requirements themselves provide us with reasonsfor action. But what should we sayinsteacl? Foot suggests that the analogy with etiquette goes all thc way clown. In her view moral facts are, like facts about what ctiqtrette requires of us, institutional facts (Foot, 1977). This explai.s why both sorts of requiremenr are categoricalin the unc()ntr()versial sense.For someone falls within the scopeof an instituti''al n()rn simply in virtue of the circumstances in which she fincls lrcrs.lf; her falling within the scope of the institutional norm is in .' way conditional .pon the desires that she happens, conringenrly, t. incl herself with in those circumstances. And it thereirrc als. cxplains why neither sort of requirement is a requirement .'re:rs.n or rationaliry. For someo n e w h o f a l l sw i t h i n t h c s c . p c , l r . i r r s t i t u t i o n a lorm requiring n her to act in a certairrway may or nlay ..t have a desire that woulcr bc satisfied by her acting in that way. In what follows I want to begin by nraki.g some remarks about Foot's own positiveview about the nature ''moral req.irements, her view that there is a substantialancl iilurninating unulogy to b. found between moral requirements on the one-hand, ncl .e_ quirements of etiquette on the other. In my view, once we spell out the analogy it becomesevident that it breaks down in cerrain fundamental respects.The upshot is that we must reject Foot's view that moral facrs are a kind of institurional fact. I will then consider in some detail her argument against the rationalists' conceptual claim.

3.7

REPLY

TO

FOOT'S

CLAIM

THAT ARE

MORALITY

AND

ETIQUETTE

ANALOGOUS Foot does not tell us how exactly we should work out the institttti.nal account of morality in detair, but it is reasonabie r<r assurne that it should be worked out along the lines o. II. L. A. llart's acc().nt .f legal rules (l96l). For H:rrt's is <.t,r(rrirrl' rlrt.

best and most thoroughly worked out account of institutional norms in the philosophical literature. 'What makes for the existence of a legal Hart's question is: system?', and he wants an answerto this question that will account fbr the normativity of law: that is, an account of how the coercive measures of a legal svstem difer from the mere use of coercive force. His answer is that a legal system is comprised by a set of institutions of a certain kind. More precisely,his answer is that a legal system existsjust in case there exist.sa system of rules that guide conduct, a system of rules that is in turn picked out by a supreme rule of recognition. The existenceof a suPreme rule of recognition, Hart tells us, is in turn constituted by a pattern of activity amongst a sub-group of those to whom the systemof rules as a whole applies, those responsible for formulating the rules, implementing them and enforcing them. The important feature of this subgroup that makes their activity constitutive of the existence of the rule of recognition, in Hart's view, is the fact that they accept the system of rules as a 'the internal point of view': that is, they are moved by whole from what they take its rules to be, and they want others to be so moved as well, because they think of the system of rules as a whole as iustifiable and authoritative. It is this last feature that Hart thinks accounts for the normativity of the law. [t explains how the coercive measuresof a legal systemdiffer from the mere use of coercive brce (Hart, 196l: esp. 97-107). As I see it, Hart's account of what makes for the existence of a lcgal system tells us something important about institutional norms, but what it tells us is bad news for the idea that moral requirements are such norms. For though it turns out to be a conceptual truth that institutional norms exist only if the members of a certain sort of sub-group of those to whom the norms :rpply are motivated in accordance with their beliefs about what the norms require, it is a wholly contingent fact about any particulirr member of the group to whom the norms aPply - members of thc sut>group included - whether or not she is motivated to act in :rtc'ordancewith her beliefs about what the norms require. It bllows that it is therefore possible for someone to believe tlurt tht'rc existsan insitutional norm requiring her to act in a

82

'l'he

Lxtnnalist Challenge

'l'he

I',xtentrtlist Challenge

u3

certirin wav witlrout bc'ing r)rotivatedto act acr:ordinely. NOt only that, it is als. p.ssiblc tlrar hcr lack of morivation signals not p r a c t i c a l i r r a t i o n a l i t \ ' , b r r r p r i n < - i p l e c lh o s t i l i t v t o t h e s y s r e m a s a w h o l e . F o r s h e r n a v c o r l ) ( ' t . h c . b c l i e f b v , s a y ,s e e i n g t h a t t h o s e i n t h e s p e c i a l s r r b - e r o r r pa ( ' ( ' ( ' p t , i ' o r r r t h e i n t e r n a l p o i n t o f v i e r v , a supreme rtrle <lf rcc<lgrrition tllat r.ntails th:rt there is zr certain l l ( ) r n r h a t a p p l i t ' s t o h c r ' , l r n < ll ' r ' t , u s a n o u t s i d e l t o t h i s s u b - g r o u p , s h e m i s h t t h i n k t h n t t h t ' i r i r t . r i v i r t 'i s p o s i t i v e l y r r n j u s t i f i a b l e . p e r h ' p s , b e c a r r s e s h e h a s r r ' r < ' . r r s t ' r r r t ' ( lt o t h o s e i . t h e s u b - g r o u p rnaking up rules that aPPlv ro ht'r', she thinks that thel, sh<luld r n i n d t h e i r o w n b u s i n e s s .l n t h i s w l r v s h t : r n i g h t n o t o n l y r o t a c c e p t the systemof rules as wholt' li'.rrr tlrt' 'internal point of view', but a c c e p t f r o n r t h e ' i n t e r n u l p o i n t < l r ' i t ' \ \ "a s y s t e n )o f n o r m s t h : i t s e t s i t s e l a ' s a i n s tt l r e i n s t i l r r r i o r r . l n s h o r t - a n d t h i s n r a y . ( ) r \ ' : i ( ' ( ' n l: r l l t o o o b v i o u s - o n H a r t ' s a c c o r u l t O f i n s t i t u t i o n a l n < > r - r rr rr sn r r l ( ' n t ' s j u d g e n r e n t t h a t t h e r e i s an inst.itutional norm rcqrririrrg lrt'r'to uct in a certain way is not s t r l r j e c tt o t h c p r a c t i c a l i $ ' r e q t r i r t ' r r r t ' r r t .f l e r r n o t i v a t i o n t o d o w h a t slrc is irrstitrrtionally requircd to rlo rrrust thereirre be externally relatedto he r judgements aborr what the institutional nornrs req t r i r e o f h e r . T h u s , b r e x a r n p l ( ' , s ( ) r n e ( ) n cw h o i s r e l i a b l y n l o t i v ated to act in accordance rvith thc larvrntrstdesire to do what she is legally oblised to do, rvhcrc this is r.r.acl r1r dicloand noL de re.She rnust have this motivation bercatrsean indepenclent sotrrce of nr<rtivirti<rn wotrlcl not explain whv the connection is reliable.Hart e m b r a c e s t h i s c o n c l u s i o n , o f c o u r s e . F o r h e i s a p o s i t i v i s t ,a n c l s o hc thinks that it i.sahvay'.s pos.siblet<l r-aisethe qrrestion whethe r or n()t we sh<-rrrld be law-abiding citize ns. Accorclir-rg to Hart, there nray be no reason at all for someone t<l do what she believes hersell'legallv required to do. A s I s e e i t , H a r t ' s a c c o u n t o f i n s t i t r r t i o n a l . o r r r r s i s w e l l - s t r i t e ctl o explaining both the conditions under which nornls of etiquettc exist and otrr rnotivations to do wtrat e tiquette requires o'trs. For thc existence of .orrns of ctiquette, like the existence .- legal rrrles as Hart sees things, does seern to be constitutc(l bv,rhc a c t i v i t ) ' o f a s u b - g r c l r p r v h i c h m a k e s , c h a n g e s , : r n c re r r [ . r c " , t h , ' * , ' r l o r n l s , a s t r b - g r o u p o f p e o p l c w h o a r e r n < l v c r lb y n o r r r r s o l ' t ' t i ( l l r c t l c b t ' < ' a u s tt' h c y a c c t ' p t t h t ' n r i - o r r rt h t ' ' i n t t ' r ' r r : r lp o i n t o r , i c r v ' :

l trthor.itrrtirt'. t l r a t i s , b e c a t r s et h e y t h i n k o ' t h c r n a s j r r s t i i a b l ez r n <a Ilreseare the people rvho tlrink tlrat it is a soocl thing that rvt'all tlr ottr la.pscs, rrrirtdour manncrs, and who thert'ilt'e ('c]stlrc r.rs r r r r c lp r a i s e u s ' o r k e e p i n u a i t h : M i s s M a n u e r s a n c l h e r k i n c l . \ \ ' i t h o r r t t h e a c t i v i t y o ' t h i s s r r b - g r ' o r r pt,h e s y s t c m a s a w h o l e r v o t t l d sinrph' collarpse. Nloreover orrr.jtrdeenl('nts:Ilx)ut rvhat etiqttette reqtrires of trs i n ' e a t b e s t < l n l ye x t e r n a l l v r t ' l u t t ' r l t ( ) ( ) u r r n o t i v a t i o n s t ( ) a c t a c c ( ) r d 'tht' rlotrc thing' - insidcrs to tlte strbi r r q l r ' .T h o s e w h r t r e l i a b l v < k r - r l o s e e t t t t o r l e s i t ' et o r l < l r v h a t gr'()up,pcople like Miss Mlrrtttt'r-s ('tiquette rcquires of thcrrr t'ltert' this is read rlc dido, anclnt> rle ra, t l r o s cw h o d o t r o t r e l i a b l r ' < l o ' t h t ' r l o t t e t h i r t g ' t l o s e c t - rtro h a v c , a t l r c s t , s o m c i n d c p e n d t . n I r i ( ) l l r ' ( (o ' 'ntotivatittn ilr cloing what thev -I'he lrclievectiqtrette rt'qttirt'so thern. indepeudcnce cl'their rn()tivat.ion e x p l a i n s l v l r r , i n ' h a t t l t e v a r - c l . t t o t i v a t e dt o d < l i s s t r licicntly at odds rvith what ('tiquette reqtriles o'thctn, they rnav t ' r ' c n b e c o n l e a c t i v c l v l t o s t i l e rt o c l o i r t g w h a t e t i q t r e t t e r e q t r i r c s o f tlrent. IJtrt n()w. bv contr:ISt with this institr.rtional acc<ltttrt olt'quirements of etiquettc, the instittrtional accottnt o' tnorlrl lt'qtrirenrents scems to have no plar.rsibilitywhatsoever. For a colls ( ' q u e n c e o ' a n i n s t i t u t i o n a l a c c o u n t t > t n o r i r l r c q u i r e m e n t s r v o t t l d lurve t<r bc th:rt our nror:rl judgerrtents t()() are <tnh. cxtcrnrtlly rcktled ( ( ) ( ) u r m o t i v a t i < l n s .N o r a l l l g o o d p e o p l c , t h o s c w h o r c l i a b l y d < r rr'hat they believe they shorrld, wotrld therefore have to bc thortsht o as being motivated to do what thev believe thel' arc morirllv r t'rlrrired to (l(), rvlrere this is re ad de ditto and not d? rr'.But rve hitve ;rlreacly s e e n t h a t t h i s i s . j u s tf a l s e .M < l r a l l y g o o c l p e o p l e a r e i t r c l e e d r t'liably rnotivated to do what they believe thev strolrld, but onlv i' rvc read this z/r rr and :not de dicto. ln slrort, the institutional accoturt o'moral reqttiremertts cnt.rils that mclral .judgerlents are rtot subject t() thc practicalirl' rt'rltrircurent; entails that morally eood people are tnotivatt:clt<t tlo what they believe they shoulcl, where this is reacl dc dirtoand n o t r 1 cr r . B r r t b o t l ' r o ' t h e s e c o n c l u s i o n s a r e f l s e. F o r r e a s o t t s w e lr:rvt':rlrcadvsccn, we have t<lthink o-good people as tlr()(ivatcd t o < l o w h a r tt h t ' y b e l i e v e t h e y s h o t r l d , w h e r e t h i s i s r e a d d e r z a t r c ln o t t l t r l i t t o , u n r l t l r c o n l v w : r v i n w h i c h w t : c : l n d o t h i s i s b y s e ei n g n r o r a l

84

'l-\rc lixtemati.st Chatbnge

'l'he

Externalist Challcnge

ill-r

judgements as subjecr r() rhe pracricaliry requirement. But if this is right then it foll.ws thar we must rejecr the institutional account of moral requirement.s. Foot's own favoured form of antirationalisnris thus a n()n-starter. There is no illuminating analogy to be found betwet'nrrtoralitv and etiquette.

3.8 THE

RH,PLY 1'O FOOT'S RAT'IONAI,IS-I'S'

OBJECTION

TO CI-AIM

CONCEPTUAL

so ar I have ilctrscd<ln F.<lt's p<lsitive claims about the nature of moral reqtrircrncnts. Lct n)c n()w bcus on her negativeclaims, her explicit arglrmcr)tag:rinst l{ont. As we will t"", h". argument turns on an assumption rvc l[N.'crro reason to accept. By all accounts,clainrs :rb.rrt what we are morally required. to do are caregoricalimperativcs:that is, if true, they are macletrue by facts about the circrrmstanccs in which we find ourselves;they cannot be defeated by the simplc otrservationthat acting ir-rthe way required will not seryeanv dcsirc or inrerest that we happen to have.And, according to the rationalist, it is a conceptual truth that claims about what we are morally required to do are claims about our reasons;claims about what we are required by rationality or reason to do. But - and now here is Foot's objection - it is a conceptual truth that requirements of rationality or reason are hypothetical imperatives: that is, they can be defeated by the simple observationthat acting in the way required does not serve any desire or interesr that we happen to have. The rationalists' conceptual claim is therefore false; for a categorical imperative can trardly be analysedin terms of a hypothetical impe.uiiu". My reply to this argument is simple. For note that Foot simply asserts that it is a conceptual truth that requirements of rationaiity or reason are hypothetical imperatives.She gives no clefenceof this claim. Perhaps she thinks that no defence is required given that it is philosophical orthodoxy. It is, after all, the view of practical rationality we have inherited from Hume. But it still seems quite wrong to me. As far as I can tell, it is in fact a conceptual truth that requirements of rationality or reAs()nare categoricalimperatives,not hypothetical imperatives.

My argument for this clairn - that requirements o'ra.ionulit\'<lr reason are categorical irnperatives - will be given later (chaptcr' ir). Here, however, I want to sh<lw how, on the assumpti<ln that requirements of rationality are categorical imperatives, we car-) clinch the argument ilr the rationalists' concepttral claim. Notc that an extra argument is <t'rtainly needed; that the fact that categorical moral rcqrrir<'rnt'nts are requirements of rationalitl, cloes not follow immcdiateh i'om the premise that requirements of rationaliry are categori<:al irnperatives. For, for all that that premise tells us, moral rcqrrircments might even be institutior-ral cts: that is, the analogy with ctiquette might hold good. What is needed is thus an ar{umcnl firr the claim that morirlity connects in some fundarnental wav rvith the categorical requirements of rationality. And strch an argunrernt can, I think, be provided.

3.9

AN ARGUMENT

FOR THE

RATIONALISTS'

CONCEPTUAL CLAIM

lt seems to me that there is single,powerful,line of argumentin support of the rationalists'conceptualclaim. The argument
trades on the truism that we expect agents to do what they are rnorally required to do. The argument can be stated as follows. Moral requirements apply to rational agentsas such. But it is a are morallyrequiredto actin conceptual truth ttrati'rational agents :r certainwaythen we expectthem to act in that way.Beingrati<lnal, as such, must therefore suffice to ground our expectation that rational agentswill do what they are morally requirecl to do. But how could this be so?It could be so only if we think of the moral requirements that apply to agents as themselvescategorical requirements of rationality or reason. For the only thing we can legitimately expect of rational agentsas such is that they do what they are rationally required to do. The crucial step in this argument is the premise that we expect rational agents to do what they are morally required to clo. It rnight be thought that this premise tradeson a pun on'expect'; that to saywe expect someoneto do something can mean either (lrat wt'believe that they utilL, or that we believe that they shoulrL.

U6

'fhe

lixtentatist Challenge

'l'he

Exteruralist Challenge

87

For the argument to work, the premise has to be interprt'tecl in t h e f o r m e r w a v ; b u t , i t m i g h r b e s a i c l ,f o r t h e p r e r n i s e t o b e t r - r r ei r has to be interpretecl in thc: lirtter way. However I think that this is just a mistake. Thc prcrnise is tnre when we interpret the claim that we expecr rati<lnala{entsto do what they are rnorallv r e q t r i r e d t o c l o a s t h t . c l r r i r r rt h l t t w c h e l i e v e t h e y w i l l . In clrder t() see this, notc that we certainly expect rationitl aqents t<r do what thev iurlgc themselves to be m<trallv requirecl to d < l : t h a t i s , w e c e r t a i n l v b c l i t , r ' c n o t . j t r s t t h a t t h e v s h o r r l c l ,b r r t t h a t they will, other things bcing t'qual. For this fcrllows clirectlv i.rrr t h e p r a c t i c a l i t l ' r e q u i r - e n l ( , r ) t , a n r l , a s w e h a v e s e e n , \ ^ , . .h i r v e n < ; alternative but to accept that: lrbsent pr:rctical irrationalitv, agenrs rvill do what they.juclge to b. righr, ar leasr other things bcins cqtral. However it might be thoLrght this is also the most that we c a r ) e x p e c t i n t h a t s e n s e ,a r r c l t h a t , t r r ' < ' a t r s o ef t h i s , t h e a r g u r n e n t c l o e s n ' te o t h r o u g h . F o r e v e n i , o t h c r t h i n g s b e i n e e q u a l , r a t i o n a l agents will clo what they jtrdge thernst'lves morally requirccl to clo, t h c a r g t r r n e n t p r o v i d e s u s w i t h n o r c A S O nt o t h i n k t h a t r a t i o r a l a g e n t s r v i l l a l l c o m e u p w i t h t h e s a r n e . j r r c l g c r nn et s a b < l t r tw h a t t h e v :rre rnorally required to do. Rational agenrs may therefore cliIr in their moral.judgements, differ without being in anv wav strbject L<> ralional criticism (Blackbur', 1984; Price, lg88). But i'asenrs rnal' difl-er in their moral judgernents without being srrb-jcct tcr rational criticism, then it cannot be that their .juclgcrrents are about what they are required to do by the catesorical requirenrents of rationality. This is the view taken by expressivists.It is the rearsonwhy, even though they accept the practicality requiremenl, t h e y n o n e t h e l e s sr e - j e c tt h e r a t i o n a l i s t s ' c o n c e p t u a l c l a i n r . I t i s , i yotr like, the reason why they are expressivists. I' act., however, tl-reobjection backfires. For, as we ha'e seerr, it is a platitude that our moral judgements at least purport to b(' objective (chapters I and 2). Thus if A says,It is right ro Q in c i r c t r m s t a n c e sc ' a n d B s a y s ' I t i s n o t r i g h t t o Q in circurnst;'l(.('s ( l ' t h e n w e t a k e i t t h a t A a n d B d i s a g r e et;h a t a t m o s r one r>'tlrt,ir jrrclgen-rents is true. And that means, in turn, that rve takc it thlrt rvt' can ult at leastone of A's ancl B'sjudgements ionr tht' r.lrriorr;rl p o i n t o l ' r ' i e w , i r r i t i s f a l s e .B u t i f t h i s i s r i g h r r l r c n i r i r l l o r r sr l r r r r . , c d o i n a c t e x [ ) ( ' ( ' tt r t l i t t t t r t l . , r g < r l() ' l(r)r(s i t r s tr r st l r t ' a r g r r r n e n ts a y s w

rvhat they are morally required to do, not just what thev juclge thernselves to be nrorallv reqtrire d to do. For we can and do expect in their lational agents to judge tntly;we expect them to conae'rge jrrdeements about what it is right to clo. Our concept of a moral requirement thus tu'ns ()ut to be the concept of a categorical requirement of rati<lnalityater all. I n c l e e c l ,t o t h e e x t c n t t h a t w e d o n o t e x p e c t a n a g e n t t o d o t h e r ight thing, btrt perhaps onlv what she believes to be right, it can rrow be seen that otu' lt'sscr expectation reflects our view of the lrlentas to some exter)t irrational; as someone who fails to live up t o t h e r e q t r i r e m e n t s o ' r e a s o n t < lt h e e x t e n t t h a t s h e s h o u l d . F o r i t rcflects our view of hcr as s()lneone who will not correct her false bclie'about what she is rnorally required to do before she acts. .\ncl this in turn suggcsts a range <lf other reasons rvhy we rnight n()t expect an agent who is rnorally required to Q in fact to Q. For there are all sort.s o'ways irr rvhich agents can fail to live up the rcquirements of reason; all sorts of ways in which they can be l)racticallv irrational. They may suer from weakness o'will, or r ornpulsion, or any of a range of other forms of practical unr'('Ason (Pettit and Smith, 1993a). Importantly, however, none o' thcsc reasons fcrr modi$ing our expectation shows that tl'rere is :ulvthing wrong with the crucial premise in the argument: the l.rrenrisethat we expect rationalagents to do what they are morallv r t'quired to do, in the sense of believing that they will, at least other things being equal. Indeed, all of these reasons for moclify,irrtr <rur expectations presuppose the truth of that crucial prenrise, l , r r l l r e v s r r g g c s tt h a t a m o d i f i c a t i o n o ' o u r e x p e c t a t i o n r e q u i r e s sornc forrn practical irrationality or unreason. As I sce it, the argument given is therefore sound. Ortr concept o a rnoral requirement is indeed the concept of a categorical r t'qrrirement of rationality or reason. Moreover, note that we can rt'uch this same conclusion from another direction. For, as I see it, tlrt' :rppropriatenessof a whole range of moral attitudes depends rrpon the truth of the rationalists'conceptual claim. Approval ancl <[islrpproval,br exarnple, must lie somewhere close to the heart o' .rnv u('c()unt o'morality. For it is a datum that we approve and rlis:rp1rr'ovt'<l'wha pte o p l e d o w h e n m o r a l m a t t e r s a r e a t s t a k e : w c .rl)l)r()\'('ot h o s t ' w h o d o t h e r i g h t t h i n g a n d d i s a p p r o v e< l ' t h o s c

BB

The Extentalist Challenge

'l'he

Extemalist Challenge

who clo the wrong thing. But, as I will now go on to argue, such attitudes themselves preslrppose thc legitimacy o our expectation that rational people will act rightll', and so, in turn, presuppose the truth of the rationalists' c()nceptual clairrt. In order to see why this is so, we need to remind ourselves of the difference between approval and clisapproval on the one l-rand, a n d m e r e l i k i n g a n c l d i s l i k i n g ( ) n t h e o t h e r . F o o t h e r s e l f r n a k e st h e following remarks about the dircnce. What anyone can want or likc is t)()t rcstricted. logically speaking, by factsabout his relationship to ()th('r pt'ople, as br instancethat he is a fi-iend or a parent of onc, arrtl cngaged in a.joint e nterprise with another. Such facts can, howcver, create possibilitiesof'apthat worrld otltcrwise not exist. (Foot, prt>ving and disapprovir.rg 1977: 194) In other words, whereas we can like ancl dislike more or less what we please, we cannot jtrst approvc atrd disapprove o what we please. Certain relationships tretween those who approve or disappro\/e on the one hand, and those who are approved or disapproved of on the other, are presupposed by the attitudes of approval ancl disapproval. [T] he attitudes of approval irnd disapproval would nol be what they are without the existenceof tacit agreement on the question o f w h o l i s t e n st o w h o m a n d a b o u t w h a t . ( F o o t , 1 9 7 7 : 1 9 8 ) [A]pproval and disapproval can, logically speaking, exist only against a backgrotrnd of agreement about the part that other people's viewsshall be given in decision nraking. (Foot, 1977: 199) Thus, according to Foot, it makes sense to say that I disapprove of your behaviour clnly if we presuppose that you are to take account of that fact in deciding what to do. But in order for it to make sense that I merely dislike your behaviour, we need to presuppose no such thing. Consider an example by way of illustration. Suppose you eat peppermint ice-cream, and that I just can't tnc' stand it when people clo that. What would it be appropriate f<rtice-crcam, or that I to say: that I dislike your eating peppermint disapprove of your eating pepperrnint ice-cream? Act'or<ling t<r t o o t , i t o n l y r n t r k e ss e n s e t ( ) s a v t h a t I c l i s a p p r o v c i ' v < l t t : t l ' ( ' t ( )t l l k ( '

account of that fact in decicling what to do' For to say that I disapprove o'your behaviotrr, as opposed to merely dislike it, thc ,ig.ri, the fact that, as I sce it, your behaviour transgresses stndards in rerms , wlrich y.u and I both acknowledge yout' behaviour is to be.itrdgetl. In other words, disapproval presupposes that your behaviottr is contrary to my legitimate expectaiio.rr; -y beliefs ab'ut how you will behave' Disliking your behaviour presupposcsno sttch thing' However, if this is right, then it follows inlmediately that apand disapproval are otrly ever in place when there. exist proval 'gtound; for legitimate exPectationsabout how solneone will beave.One obvious area in which approval and disapprovalare in the area of rational decision-making'For we all place is theref<rre L"p..t of each orher rhar we will decide what to do on rational grunds. As Foot notes, however,this is not the only area in which pproval and disapproval are in place. The members of a chess of a fellow member who moveshis castle .iub may well clisapprove on the diagonal, for instance, and their disapproval may suwive the discoverythat, in the context, this is not an irrational thing for him to do. For even if he has good reasonsfor doing what he does, he still actscontrary to their legitimate expectations,for he acts in violation of an agreement he either tacitly or explicitly entered into by becoming a member of the club, the agreement to move pieces in accordance with the rules. Their legitimate his ckress expectation is thus that thosewho do not want to play by the rules will not play at all. I began by saying that it is a datum that we approve and disapprove f *hut people do when moral matters are at stake. But now, in light of our account of the preconditions for approval and disapproval, an obvious question presents itself. consider, for irrsta.., disapproval of those who act contrary to moral requiremenrs.Such disapprovalis ubiquitous. Yet how can this be? For, as we have seen, disapproval of those who do not do what they are the legitimacy of our expectmorally requirecl to do presupposes that, as we see it' ation tirat they will act otherwise;it presuPposes their decision is a bad one in terms of the commonly acknowleclged standardsby which their decisions are to be judged' But wliat pr.vides grouncls for the legitimacy of this expectation? In

90

'l'lte

lixtentalist Chattenge

'l'he

lixtemrtList Challenge

9l

'i.ttre of what a.c tht'*' c'mrrr'.1\., ackn'wleclgecl standards by which their clecisi()ns ilr(, to bc.jrrclgecl? N . t e h o w i r n p l a r r s i l l r t 'i t w . r r r c r b e t ' s t r p p o s e t h a t w h a t g r o u n d s . the lesitimacy o'tlris ('xl)(.('t.ti()rr is the faci that rational ...n,.,.", ha'e each enter(:d i.t. lrrr aur-('erlent to act moraily, an agreer n c . t o n : r p a r w i t h { h t ' l r g r ' < ' < ' ' t ' . ( t h e c h e s sp l a y e r e i t h c r tacitly or e x p l i c i t l y e n t e r s i n t < lb ' l r t ' < ' r . r i ' r s : r m e r n b e r . f a c h e s s cl'b. For n<l suclt arrccrllents lr:rvt' t'r,t,r'bccn rnacle. perhaps we should sav inslead that what gr.lrrr<ls ()ur' ('xpcctation is ,-,ot , , r - ,a g r e e n r e n t rational creatrlres havc' irr rr< r nrirrlc, but rathcr an a[ree ment tht'v w.rrld n-rake i'tht'r' rvt'r't'. . . I'rhey r.verelvhat? I'iilev were l a t i o n a l , < l c o u r s e l B r r t i n t l l r l t r r s er v c h : r v ea b a n d o n e d the iclea that what grorrnds thc l<'girirrr:r.r ' ' ( ) l l r e x p e c t a t i o r r i . st h e f a c t o f agreement in fav'trr ,'llrt. lrll.r.uriv.. what grouncls the legitinracl'.f otrr expcctatio. is tlrr. rrrt,r. fct that pcoplc are rational agents. Beinq rati.nal s.Ii<'t's t<l gr,tr.d the expectation that p e . p l e w i l l c l o w h a t t h e v a r t ' r r r . r a l l v r t - q r r i r - e ctl. c l o . G i v e n that . r ' r a l a p p r . v a l a ' d d i s a p p r < x , a li r . t ' r r l l i q r r i t , r r s , t h e trtrth 'f the rationalists'conceptual cl;rinrthrrssct.rrrs t < lb c e n t a i l e c l b y t h e a c t that th. preconclitio.s .' r'.r'al aJrprcl'al and clisappr.rvar u.e satisfiecl. It is worth rer'arking trrat Fr,t a.kn<lwleclges that rrer views ab'trt the precorditi'.s irr rrr.rar apprclval ancl rnoral disappr<l'al zrrein apparent c'nflict with the instituti'nal accour-rt o moral requirernents she votrrs. She asks, Itlr example: What . . . :lre we t. sayab,trt those wh'altogether reject moralitv? Strrell we thi'k it p-ssible t. disappr<lve of their actins,arrhouch thel'd' n()t ag.ee t. take :')v acco.rt r>'whatwe say? This is true, lu'rl it is a. ir'porta.t ct ab'ut the phe.omc'c,rn that we call 'r,.rirlity' that wc are ready t. brirrs pressrrre to bear againstthose rvho rcje<:tit. But this no rnore shows that rnoral attitrrdes clo not c l e p e n ro l ' :rgrcer'ertwithi' hrrr'a. societv t h a n t h e p o s s i b i r i o, f assertirrg .the r ki'ds..'atrth.ritv agai'st th.se who d n.,t u..:.p, i t s h . r r ' st h a t a r r t h o r i t ' r e q r r i r e sn o a g r e e m e n t . . . [w]c wilr <.,rr i - ' . t r h er ' w i t h t h e c o . f i d e ' c e t h a t w e h a v et h e w . r l c l w i t h t r s- r h t , rv'rld rhat r)avs a t l e a s tI i p s e n , i c et o r r r . r a r i r v .( F , r t , I r ) 7 7 :r ) i r ) l } , r t h c r r c J r l v t ' r r t i r c l v n r i s . s t 't . ls r t , p o i r r r . F o r - < , r . < . irr, . r r l . i r . t ' s i r t s t i t r r r i ' r r r:lr r . . r r r r r, r a r r ' : r l i t rr' v . t ,i r r r l r t ,s P t . < . i s : rrlr l r - g r . r r p r'lr.

get to set the standards, changc thcm and cnf<rrce thern, can in some wav be said t() ('nt('r into an agreement to make otrr rlecisions on moral matters irr a ct'rtairrway, those who reject nroral'worlcl' i t y a r e , b l ' h e r a ( ' c ( ) u n t , n ( ) t u p a r t v t o t h i s a l - e e l n e n tT . he 'thern' we have with us whe n wc coni-ont is thus, for thenr, simplv a nrob forcir-rg its r'<lrrrrrronlvast'ccd stanclarcl cln anothcr- group r v h o s e a g r e e m e n t t h c l ' < l < lr r < l t h a v e . T h e i n s t i t u t i o n s t h a t F o o t thinks suEce to unckrrgir'<l a shared staltdard of clecision-nr:rking t h u s u n d e r g i r d n o s r r c l rt h i n q . T h e v p r o v i d e n o b a s i sa t a l l f o r t h e expectation that thosc rvho reit-ct morality will make their <le'wc' cisions in the rvay will. TItat reqrrires the tnlth o'the rati()ltc l a i n r, a c'laim that Foot is in no position to alists' conceptual cndorse.:

3.10

SUMMARY AND

PREVIEW

My aim in this chapter has been to consider irnd <lefendfivo fcrrms o-internalisn: the practicality requirement on nroraljuduement and the rat.ionalists' conceptualclainr.In the corlr.se of doins so I have considered and rejected two powerful arguments fbr David Brink's'amoralist'challenseto the practicalitv cxternalism: requirement, and Philippa Foot's 'etiquette' challenee to thc raticlnalist's conceptual claim. If my arguments in this chapter have been on the right track, then it follows that, contrary to both Brink and Foot, our-juclgelnents about what we are morarllyrequired to do are sirnplvjudgcrnents about what the categorical requirements of rationality or reason demand of us. However this is bound to raise evebrows. F'or it is commonlv held that the standard accolrnt of hrrman psvchology, the account of belief and desire we get from Hume, is inconsistent with the very coherenceof the claim that there are <utegorical require ments of rationaliq,. In the next two chapters I tlrt'ref<rre turn to considerHume's psychological theory in its own t('r'lns. How much o'Hume's theory should we keep, horv much slrorrlrlwe rcject? I begin by considcring Hunte's views about Ir r r r r r a n l ( ) t i v a t i ( ) n .

Hozu to Solae lhe MoraL Problcm

l83

6
How to Solve the Moral Problem

6.1

AN ANAI,YSIS OF' RIGHTNESS IN TERMS OF FACTS ABOUT NORMATIVE REASONS OUR

W e s a i c l t h a t a n a n a l y s i s o f ' r i g h t n e s s n r u s t c a p t r r r c v a r i < i t r sc o r t ' platittrdes aborrt rnoralitl', platitucles concerning tlre obiectivitr and practicality of rnoral -judgement, the superwenience of the n r o r i r l o n t h e n a t t r r a l , t h e d i s t i n c t i v e s t r b s t a n c eo f m o r a l r e c l r r i r e lnents, ancl the various procedures via which we ('olne bv rnor:rl k n < > w l c d q o( c h a p t e r 2 ) . M o r e o v e r , w e s a i c l t h a t e v e n i f s u c h a n a n a l v s i si s r r o l a t l r < l r o r r e h l y r r : d u c t i v e n a t u r a l i . s t i cd e f i n i t i o r r , e v e r r i ' i t i s r n e r el v a r t o n - r e d r r c t i v e a n a l y s i s ,w e r n u s t s t i l l b e a b l e t o r t s e it to sclrrirrt' rnorirlit.i with a broader naf-uralisrn. N{oral aturcs rnust trlrn out t() be rratrlral atrrres, i' therc are all\' moral a t r r r e sa t a l l ; n r o r a l i t y r n u s t n ( ) t r e q u i r e a n y t h i n g l r o n - n a t u r a l f i ) r ' (chapter ?). Ancl inalll', we said tlt:rt, acct>rdins to its realizati<>n such an anirlysis, it nrust trlrn ()ut that facts about rightness ha\.'t' the character of catesorical requirenrents of reilsor) (chapter 3) . Now we havc alreach' seen that r.l'ecan analyse otrr conce pt o :r norrnative reas()n quite generally in consenial, radicallv arrtiIiurre:rn, categorical inrperative terrns. To sav thirt we h:l,c lr normativc reason to $ in certain circumstances (l is t() sa)' that w<' w o r r l c lw a n t o L l r s e l v e s t r - r i n ( l i f w e w e r e i r l l y r a t i o n a l . I ' t h c l t ' i s n n<lrmativc reason for some 2rgentto Q in ccrtai tr circttnrsturt<t's (l then there is a like normative'reason filr all thosc u'ho irrrl () tcl thernselves in circrrmstances Q . S r r c h a t r a n a l t s i s t l t r r sr r l r . r ' l r l r (( ', captrlres platituclcs about objt'ctivitl', practicalitr', sup('r-\'('nr('n rr'r' a n d p r ' o c c d r r r e r .H o w e v e r r : t s : r b o r t t t h e r r o r r r t a l i r , t 'r ' t ' r r s o r r s

I r a v t -c l r r i t e g c n c r a l l v a r ' ( ' n ( ) t t h < ' r n s e l v e s i d e r t t i c ew l ith facts ab<lut i t i s r i s l r t r , v h i c ha c t s or rrsto lteritrrn. For even though all claims alrorrt rvhat it is riL{lrt or.rrs to do enteril corresponding claims ' ' c a s o nt o d o , n o t a l l c l a i m s a b o u t aborrt whirt rvt' hirvt' rrorrrr:rtivtr r v l r a t w e h a v c t r o r r r u r t i v cr t ' i r s o n t o d o e n t a i l c c l r r e s p o n c l i n g c l a i r n s a b o t r t r v h a t i t i s r i g l r t i r r r r s t o < t o .( l o n s i d e r s o r n e e x a m p l e s b v r v a v of illtrstration. I t s c e m s e a s vt ( ) i r t r l r g i r r t t ' h a t I h a v e a n o r m a t i v e r e i t s ( ) nt o d r i n k beer rather tllan rvirrt' rr'hilc rclaxinc aier rvork. For any<lne in r - e l e v u n t l y s i r n i l a r c i r t r r r r r s t : r n c e st ( ) m i n e - w h i c h i n t h i s c a s e incltrclcsthe t't th:rt I t'rrjov bcer rather than wine - this rnav be tl-re perfect wav ol r.t:laxins. To suppose otherwise wcluld be mistakcn. If I clrink rvint' to rt'l:rx uier work then I am not doing what I h a v e n o r r u a t i v c l c l r s ( ) r )t o c l o . B u t e v e n a c l n r i t t i n g t h a t t h i s i s s o , it is harcl to bclicvc rhut I anr noruilly requiredto drink beer rather thitrr rvirrc rvlrile rclaxing aier rvork; that it woulcl be. imrnoral f<tr nle to clrink winc. F<lr cvcn though reasons are t stake, the rezls()ns a t s t a k e a r c n o t t l r e m s e l v e sm o r z l r e a s o n s . Or irgain, consiclcr the lifc of someone rvho is moved whollv and soleh'bv considcrati<>ns of rishtness. Perhaps she devotes her li to the cornnrcln g<lod, never giving her own welre a single thottsht, except in so r as her welfare is part of, or implicated in, the common good. Such a life certainh seemsto have somethinr important rnissing flrom it. For a life that is pe rfect from the rnoral point of view rnav lack much of what makes a lif personally fulfilling. It may lack much that is significant, r'rot fror-n the moral point of view, but from the point of view of the person whose life it is: the point of view of personalperfection(\Iolf, 1982). Reasons are at stake but, again, they are not moral reasons. What needs t<l be adclressed is thus how precisely we are to demurcate the province of the moral, as opposed to the nonmoral, reasons. And the answer is plain enough. For what the analvsis of'r'rormative reasons quite generally leaves out of account - and rightly scl,o'course - is the distinctive substance or corilent<tf Icas()ns that makes them into moral reasons as opposed to nonr n o r a l r e : l s o l l s .T h e a c t t h a t t h e r e i s s u c h a d i s t i n c t i v e s r r b s t a n c c ( ) r ' ( ' ( ) n t c r ) ti s , v o r r w i l l r e c a l l , e v i d e n t f r o m t h e p l a t i t u d e s c o l t c e r n i r r s s r r b s t l r n c t 't h a t w e f " i n c la r n o n g s t t h e p l a t i t u d e s c o n s t i t t r t i v e o f

184

How to Soluethe MmaI Probbm

Hou lo Solue the Moral Problzm

I 85

the moral. These are platitudes like 'Right acts are often concerned to promote or sustainor contribute in some way to human flourishing', 'Right acts are in some way expressiveof equal concern and respect',and the like (chapter 2). These platitudes need not and should not be thotrght of as fixing a unique content or substance for moral reasonsall by themselves, rather they simply serveto tell us when we are in the ballpark of moral reasons,as opposed to the ballpark of non-moral reasons(Dreier 1990). It is these platitudes about substancethat need to be added in and captured if we are to turn our analysis of normative reasonsquite generally into an analysisof rightness in particular. My suggestionthen, in schematicform at least,is that our Q-ing in circumstances C is right if and only if we would desire that we in C, if we were fully rational, whereQ-ing in C is an act of the Q affropriate substantiae hind: that is, it is an act of the kind picked out in the platitudes about substance.There is clearly further work to be done here in illing out this idea of an 'appropriate substantive kind' in detail, but the schematicform of the analysis will sufice for our purposes.

6.2

THE

SOLUTION

TO THE

MORAL

PROBLEM

We now have a solution to the moral problem at hand. That is, we can now explain why the following three propositions: I Moral judgements of the form 'It is right that I $' expressa subject's beliefs about an objective matter of fact, a fact about what it is right for her to do. If someonejudges that it is right that she Qs then, cetnispaibut is motivated to Q. she

I' our concept of rightncss is the concept o'what we would desire ourselves to cl() i'we were fully rational, where this is a desire for something o' the appropriate substantive kind, then i t d o e s i n d c e d i r l l < l wt h a t o t r r l n o r a l j u d g e m e n t s a r e e x p r e s s i o n s o f o u r b e l i e . sa l x > u t a r r r > b j e c t i v em a t t e r o f f a c t . F o r o u r m o r a l judgernents arc ('xprcssior.ts o['our beliefs about what we h:rve n o r m a t i v e r c l r s ( ) nl ( ) r k r , r + l t t ' r e s u c h r c a s o n s a r e i n t r r r l l c a t e g o r i cal requirenlents o' rationality. (1) is thus true. Moreover, as we have seen, srrch bt'lir'.s clo indeed connect with motivation in the manner ol (2). And, again as we have seen, their doing so does not in any wav r'orrtponrise the claim that motivation is to b e e x p l a i n e d i n [ { r r r r ) e 2 r nt c r n ) s ; t h a t i s , i n t e r m s o f b e l i e ' a n c l d e s i r e , w h e r c b c l i c ' a n < l < l e s i r ea r e d i s t i n c t e x i s t e n c e s . ( 3 ) t o o i s thus true. So ar s<lgood. Btrt does the analysis allow us not onll' 'r solve the moral problenr, but to do so in a way that allows us Ito square morality with a broader naturalism? It seems to me that it does. The analysis tclls trs that the rightness of acts in certain circum'evalustances C - using our e:rrlier terminology, let's call this the ated possible world' - is the feature that we would want acts to have in C if we were fully rational, where these wants have the appropriate content - and, again, using our earlier terminology, let's call this world, the world in which we are fully rational, the 'evaluating possible world'. Now though, for reasons already this does not itself constitute a naturalistic definition of given, rightness though it is merely a non-reductive, summary style analysis (chapter 5) - it does provide us with the materials tcr constrllct a two-stage argument of the following kind. Conceptual claim: Rightnessin circumstances C is the feature we would want acts to have in C if we were fully rational, where thesewants have the appropriate cot)ter)t Substar-rtive claim: Fnessis the feature we would want actsto have in (l if we were fully rational, and Fness is a feature of the appropriate kind Rightnessin C is Fness

An agent is motivated tr-ract in a certain wayjust in case she has an appr<lpriate desire and a means-end belief, where belief and desirt' are, in Hume's terms, distinct existences.

are both consist.ent and true. Let me briefly explain why.

( lrrrclrrsion:

186

Hout to Solae the Moral Probkm

IIou lrt.\ohre thc,\ltmt,l Pntblcn

lu7

And this argument is, in turn, broadlv naturalistic in tw<t reFirst, it is naturalisticin so r as the featuresthat rvervould spects. want our acts to have under conditions of ull rationality, the features that we would want acts tcl instantiate in the evalrrateci possibleworld, are themselvesall naturirl eatures whenever the evaluatedworld is itself naturalistic. Our non-reductive,summary style definition of rightness, in conjunction lvith a substzrntive claim of the kind described, thus allows us to identifi' rightness with a naiural feature of acts in nattrralisticworlds like the actual world: for example, in this case,with Fness. And second,even though the analysis is not itself naturalistic even though it defines rightnessin terms of full ratior-rality rvhere this may not itself be definable in naturalistic terms - fully rational creatures in the evaluating possible w<-rrld arc then-rselves naturalistically realized. For a fully rational creature is sirnply ancl,asyou rvillrccall,a rlatur-al with a certainpsychol<>gv sor)leone feature is simply a feature that figures in one o'the natural or (chapter 2). Of c()urse, the social sciences,including psychology psvchologyof a fully rational creature is an iclealizedpsvcholow, but.such :rn idealization reqr"rires nothing non-natural for its realization. T'hus, if we wanted to, we could construct non-r-cdtrctive analyses o'the key norrnative conceptswe use to characterizethe normative featrrresof such :rn idealized creature's psychokrw the uniry,the cohere nce, and the like, of its desire s - and thcn use these analysesto construct fwo-stagearguments, much like that just given, in order to identi$ these normative features of a fully' rational creature's psychology with natrrral features of its psychology (for an analogy, see note 9 to chapter 2) . (,-oherence and unity, though not naturalistically definable are therefore just natural features of a psycholog,v. themselves The evalu:rtins possible world is therefore naturalistic in the reler,'antrespect as well. The analysisof rightness provided thus m:rkes the legitimacl' of moral talk depend r-rltimatelyupon the possibility of iclenti$ing moral features, like rightness, with naturzrl feattrres ol acts.Absent such identifications,we would have to conclucle thut moral features are simplynot instantiated at all; that nroral tirlk is, much as Mackie th<lught,based on an error of prcsrrpJrositiorr.

[ J u t c a n w ( ' s a v r t t o r t ' ?( ] l t r r r v t ' s u v r v l r t ' t ] r e rr n < l r a lt a l k i s o l i s r r o l legitirrrzrte?

6.3

ARI.,

'I'TII.,RF] ANY MORAL

FAC]'|S?

T h e s r r b s t u n t i v r< rn t l r t ' t \ \ ' ( s t a Ja C ' 'llrirri r g u n l e l t t d e . s c r i b e da b o v c t e l l s r r s t h a t n r < l n r lt a l k i s l t ' q i t i r t r a t c i r r s t i n c : r s ea c c r t a i l t c o n c l i t i o n i s n r e t . M v h a r t r l i r r ql x r c k ; r u a l l t ' t I i r r r n c li n t h e s t r e e t i n s t r c h a n c l s u c h c i r c r t t n s t u r ) ( ' (i's ,or cxanrplc,clnh'i, rrnder corrditions sr - i g l r t I o ' r r l l r a t i o n a l i t r . r r ' < 'u o t r l r l t r l l w a n t t h a t i f r v e f i n d a w a l l c t i n t h e s t r e e t i n s r r t l t l r r r <s l r r < h( i l ' c t u t r s t a n c c st.h e n w e h a n c l i t b a c k . i i s i n < l t ' r ' r lt r t r c , t h e n i t i s : r n a l t i o r i t n l t h . T h c l O'corrrse, this f a c t t h a t r t e n ' < l t r l <h l i r v t ' s r r < l rl r c l e s i r e t r n d e r c o r r c l i t i o r r s r>'irll ef t h e t h e < t n ' t h i r t s v s t e r n a t i c i r l l v r a t i o n a l i t v w i l l l r t ' t r ( ' ( ) n s ( ' q u ( ' n < 'o j u s t i f i e so r r r c l e s i r c s . anrl this tusk <lI thcorl c()nstnl(ti()nis itsel a r e l a t i v e l v t t l t n t t i t ' r r l t ' r ' p r i s t 'i:t i s : r t a s k t l r a l l c q t r i r e s r t ' l t ' c t i o r r and cclnver-sirti()n n,( ) t c m p i r i c a l i n v c s t i g a t i o n . H < t r v c v e rt h a t d ( ) e s not mean that it s ant obviot{.r tnlth. For it nrieht n<tt <lnl!' ttrke a sood deal of relection and convers:rtion br anv inclividtral tcr cliscover tlris tcl be tnre, it rnight also take tirnt' ancl cr>rt to convince an,vone elsc. Ancl of cotrrse, even acr all o that timtr and effort, it mieht tllrn ()ut that we are wrong. \rhirtwc thotrqht was an u pnon truth might have been no tnlth at all. In clcciclinI w h e t h e r ( ) r n o t m o r a l t a l k i s l e e i t i m a t e , t h e n , i t s e e n t st o m e t h a t w e l t a v e n < la l t e r r t a t i v eh r r t t o a c l m i t t h a t w e i t r e v e r t ( r r r i r r ra tn opiniorr on something about which we can have no cast-iron guarantee. However, ftlr all that, it scerns to n)c that we shorrld none thc less have some confidence in the legitimacv of moral talk. For, in short, the empirical fact that nloral argument tends tcl elicit the agreemcnt o'our llows give s tts reason to believc that there will be a c()nvergence in our desires rrnder conditions of firll raliona l i t r , . F o r t l - r eb e s t e x p l a n a t i o n o f t h a t t e n d e n c y i s o u r c o n v e r s e n c e upon a set of extrernelv unobvious a priofi moral truths. And the truth ()'these trnobvious a lriorimoral trtrths reqtrires, in trrrr-r, a ('()llver-gence in the desires that fully rational creiltures w<luld have. N o w t h i s a r g u m e n t i s l i k c l v t o m e e t w i t h s o r n e r e s i s t a n c e .A i e r rrll, isn't thcrt' ctrrrentlv much entrenched mot'al diszrgreerttcnl?

lB8

How to Solaethe Moral Probl"em

How to Solue the Moral Problzm

I fit)

And don't such disagreements constitute profound obstacles to a convergencein our desiresin fact emeruing? This is true, but it does not count againstthe force of the argument. Indeed, once we remember the following three points, such disaereements can be seen to add to the force of the argurnent. Fi;X, we must remember that alongside such entrenchecldisagrements as we in f,act find we also incl massiveareas of entrenched agreement.As I see it, this is the real significanceof the fact that we have and use the so-called 'thick' moral concepts, concepts that at once both describe some naturalistic state of affairs and positivelyor negativelyevaluateit: c<,rncepts like courage, brutaliry, honesry, dupliciry, loyalty, meanness, kindness, treachery,and the like (Williams, 1985: 129). For whar rhe prevalence of such conceptssuggests is that there is in fact considerable agreement about what is right and wrong: acts of brutality, duplicity, meannessand treachery are wrong, at least other things being equal, whereasacts of courage, honesfy, loyalty, and kindness are right, again, other things being equal. What the prevalence of such conceptssuggests is therefore that moral agreement is in fact so extensive that our language has developed in such a way as to build an evaluative component into certain I naturalistically descriprive concepts. Segond,when we look at crlrrent areasof entrenched disagreement, we must rcmember that in the past similarly entrenctred disagreementswere removed inlu alia via a process of moral argument. I arn thinking in parricular of the historical, and in some places still current, debates over slavery,worker's rights, women's rights, democracyand the like. We must not forget that there has been considerablemoral progress,and that what moral progressconsists in is the removal of entrenched disagreements of just the kind that we currently face. And third and finally, we must remember that where entrenched disagreements currently seernutterly intractable we can often explain why this is the casein waysthat make them seemless threatening to the idea of a convergencein the opinions of fulll' rational creatures.For example, one or t.he other parties t<l thc disagreementall too often forms their moral beliefsin responst, to the directivesof a religious authority rather than as the r.t:strlt ol

the exercise of thcir own free thought in concert with their t'llows. But belie.s irrmed exclusively in this way have clrrbiotrs rational credentials. They require that we privilege onc gr()up's opinions abor.rt whirt is to be done - those of the religious authority - ()ver an()ther's - those of the followers - f<rr no go<lcl reason. The firct that disagreement persists for this sort of reas()r) thus casts no cl<ttrbton the possibility of an agreement if we wer.c to engage in ii'r' and rational debate.2 In light of thcse points it seems to me that, notwithstanding such disagreerncr.lts as there are and will perhaps rernain, we shoulcl theref<rrc in fact be quite oprimistic about the possibiliry of an agreernent abotrt what. is right and wrong being reached under more idealizecl conditions of reflection and discussion. We might eventtrally come to be pessimistic, of course. Our epistemic situation might deteriorate, widespread disagreements might emerge, clisagreements that seem both unresolvable and inexplicable. Ancl if that were ro happen, then we nright well quite justifiably come to think that Mackie was right after all, that there are no moral facts, though there would still be roclrn for doubt. The point is simply that this is not our current epistemic situation.3

6.4 RIGHTNESS STANDARD

HOW

THE

ANALYSIS US TO TO

OF REPLY TO

ENABLES OBJECTIONS

RATIONALISM

The argument so far has been that, given our analysis of rightness, it follows that we should accept both the rationalist's conceptual claim and the rationalist's substantiveclaim. Our concept of a right act is the concept of a categorical requirement of reason and, substantively,there are indeed such categorical requirements of reason - or so we should think. In this next to final section I want briefly to consider how the analysiso'rightness enablesLrs to reply some of the standard objections to rationalisnt. I consider objections put forward by Hume, Foor, Harmirn. ()atrthier ancl Mackie.

190

Hou to .Sohre the Lloral Prohlent Hume's Objection

First pr-crnist': Se<'oncl pn'nrist':

Htntt to .Soltp tlrc Moral Prohknt

( )lrjt'<t x lxrs srrrace ref lectatr<:tf Pr()l)('rt\' (

l{)t

Perhaps the most fmorrs 'bjec'tion to rationirlism is t<l bc ouncl 'l'rertti.se. in the following passagc iorn Ilrrrne's In every systemof rnoralitv rvhich I lrirvc hitlrt'rto rnet rvith, I lrave a l w a v sr e m a r k e d t h a t t h c a u t l l ( ) r p r ' ' c e e < l s irr s'nre tirnc in the ordinarr''way o reasonirrg,irnd establishcs tlre being o :r qod. or makes obsen'ati<lns concerning hrrrtralrlrI:rir.s; u,hen o'a strclclen I a m s t r r p i s e dt o f i n d t h a t i n s t c a c l' ' t l r . r r s r r a lc . p r r l a t i . ' s ' l p r o p o s i t i o n si s a r - r d is not, I rncetwith no pr<lposition thirt is not c o n n e c t e dw i t h a n o u g h t o r a n o u g h t l ) ( ) t .T h i s c h : r r r g c is impcrc e p t i b l e ,b u t i s , h o w e v e r , o f t h e l a s tc o n s e q t r e . c eF . or asthis'trght or ought not expresses solne rrcw relation or arrnati.n, it is necessarythat it should be obsen'eclanrl cxplainecl; arrrl at thc same time that a reason shotrld be givcn fbr what sr:e nrs altogerther i n c o n c e i v e a b l eh , ow this rrew relatio' ca' bc a dedrrctir,rf i i,rrr others wlriclr are entirel-dilrent liom it. (l8tl|: ,169) Hume is right that an explanatirl' is needecl. But an expl.nation is also easv enough to provide. consider an anal()gv. Srrppose rvc clisr:overttr;rt a cerr-ainsurce reflectance propert\/, cr, is the propctrtv that carrsesobjects to look red to nor-mal perceivcr.s trndcr stanclard conclitions. Farniliarlv enough, we are tl'ren in a position to assert that an object's having surface reflectance property c, entails that that ob-ject is red. But how can this be? How can facrs about an object's colour .be a deduction' from, or entailed by, facts abotrt its surface reflectance properties? For these rwo sorrs of fact are 'er-rtirely different' from each other, at least in the sense that thev are not definitionallv equivalent. The answer is that, even though it is not true by definition that being red is a matter of having surface reflectance property cr, it is true by definition that being red is a marter of having rhe propcrtr that actually causes objects to look red to normal perceivers rrnclt'r. standard conditions.r rhe entailment therefore fcrllows fi.m tht' fact that the following inference can be known to be valid rrr<'rclr by knowing the meanings of rvor<ls.

S r r l l : r tt ' r ' r ' l e tt a r t t : e p r o p e r t ) ' c r i s t h e p r ( ) p ( ' l t v (lLrt lr(tuallv ('itttses O b j e c t st o l t l o k r e d l o I t o r rturl 1rt'r'ct'ivcrsttltder standard conditions

( irtrclrrsiotr:

()bjcrt x is red

For then, sin<'t'thcst'tonrl prentiseis a necessan'truth - that is, since cven' possiblc' tr'or'lrl in rvhich strrce reflectance property c e x i s t s i s a p o s s i b l t ' r v o r l < li r t w h i c h t h c p r o p e r r y t h a t c a u s e so b j e c t s ttnder standard conditions in e to loclk red to rr<lrttlll pt'r'r'r'ivcrs rttlual rr,orlrl t'xisls - il carr bt' r'crnovcd lvithout affecting the argun r e n t ' s v a l i d i t l ' . F o r i t i s i r t t p o s s i l l l e o r t h e f i r s t p r e l n i s e t o b e t r L t e 'I-hc i r s t p r e r n i s e t h u s e n t a i l s t . h ec o n c l u irnd the conchrsion lrlsr'. sion all bi'itsel{'. '<xrqht' 'is' t<l i s i n s o m e r e s p e c t st h e s a n l e . Thc transition i-orrr Suppt-rsc we cliscovt'r' that, itr certain circumstances C, eiving to farnine relic is the lt'atrrrt'rvc rvottltl want acts to have in (l if we lvere firlly rational. Then we arc itr a position to assert that sclme{)ne's qil,ing t() nrine rclie'in ccrtain circutnstances C entails that thev do thc right thing in C. Truer enough, the act that 'entirely different' frclm sornc()ne cloes thc right thing in (l is the firct that she gives to f:rmine relief in C, at least in the sense that the two are nclt clefinitionallv eqtrivalent. But, nclne the less, thc irppropriatene'ss of the entailment claim f<lllows from the fact that the ollowing infrence is one that we can know to be valid merely bv knorv.ing the meanings of words. First prcrrise: Ar agent givt's t() irrnine relief in circttmstances(l C is the Givins to f:rmine relief in circumstances featrrre that we would want acts to have in C if rve were fullv rational, where such a want is of kiltcl the appropriate srrbstirntive
(iiving to Ianrine i.clief in circumstances C is the light thing to do in (l

Second premise:

( l<r ;r<'ltrsion:

192

How lo Solae the Mmal Problrm

the Moral Probbm Hout lo Sofue

193

For, again, since the second premise is a necessary truth it too can bc' omitted without affecting the argument's validity. But in that caseit follows that the first premise entails the conclusion all by itself. For the first premise cannot be true and the conclusion false.We have therefore derived an 'ought' from an 'is'. There is, of course, a point of disanalogy between the rwo argumentsjust given, and it is worthwhile making this explicit. Whereas,in the moral case,the second premise of the argument is not just necessary,but also knowable a priori, the second premise of the argument in the colour case is , though necessary, itself only knowable a posterimi. However this feature of disanalogy should not hide the more striking features of analogy alreacly mentioned. For in neither caseis the second premise a matter of definitional equivalence.Knowledge of the second premise of the argument in the moral caserequires substantive work in the form of reflection on which of our desireswould sunrivea processof reflectiveequilibrium, and even if this is a relativelya priori enrerprise, unlike the attempt to discover a posteiori which surface reflectanceproperties of objects causethe appropriate visual experiences,it is arguablyjust as difficult and problematic, perhaps even rnore so. This is why the truth of the entailment clairn comes in both cases. assur:h:r strrprise Not everyorre rvill be entirely h^ppy with this explanation of the transition from 'is' to 'ought'. Barry Stroud, for example, tells us that Hume's real concern in the famous 'is-ought' passage is that though we . . . undoubtecllydo make transitionsfrom beliefs about the way things are to the judgement that things ought to be a certain way. . . if we understand the peculiarnatureof these'conclusions'
- if we recognize their'active' or motivational force - we see that the transitionsby which they are reached are not ones that reason determines us to make. Once we come to have certain beliefs about the way things are, then, because of natural human dispositions we come to feel certain sentiments which we express in moral j u d g e r n e n t s .( 1 9 7 7 :1 8 7 ) But the previous discussion suffices to show why, if this was Hume's concern, it too can be laid to rest.

'ought'-ltrtlgt'nt('tlt ls The active or rrro(ivalional force of an r e q u i r e m c n t o n m o r a l j t t c l g e r r r t ' r l t :i capturecl by tht' PI'a('(i('itlity is rigSt then either she is nrotiv^tt'rl lrt an agentjuclges iSrrt ur) 2r('t rvt' o.t u"...r.i.gly .r- s5t' is i'rirti<lnal. But now, contrary to Hrtmc, tlris accepting in involved have seen thar tht,r-t, is no tension requiremcnt wltilt' tlrirlking of moral -iudgements as expressiotts the o f e l i e s r a t h ( ' r -t h u . t , x p r e s s i o n so f o u r ' s e n t i m e l t t s ' . I n d e e d , veryanalvsis<l.r-ig}rtrtt'ssintermsoffactsaboutnorlnativc reasons, thc artalvsis tltirt all<lws tts to explain why the transition 'is' t6'()rrgftt' is ()ttg t[at rcason determines us to make' itself from allows us t() exJ)llrirt rn'lty In()ral beliefs are subject to the practiin cality requirenent. For. r't'ry roughly, if we believe that Q-ing to acts want (l we wOuld that featttre lrlrs lhe certain circtrnrstattt't's t h e t . o h a n d ' o n e t h e o n have in fl if wc rvt'rt'lrrlll'ratiotr:il then, c or in to Q extent that wc urt, r.ational cither we will already want to the other, the on and, want, this have t() lls carrsc our belief will cause not does belief our O and in to want extent that we clo tr<lt Q is, of Ancl that rational. irlly us to l1ave this want rve will n11tbe judgemoral collrse, just what thc practicalitv requirement on our me.t wulcl leacl us to expect about the connection between moral beliefs ancl m<ltivations.s

Foot's and Harman's Objection Foot's main objection to rationalism is that we can give no content to the idea that the immoral person is irrational'
of Attempts have sornetimesbeen made to show that some kind 'should' of morality: in irrationality is involved in ignoring the .Immoral _ so what?' as One says.Not comnlc ilfaut_ so what?' saying illegitimate some on rested But s far as I can see these have man who amoral that the assumption,as, for instance,of thinking notice no takes but is immoral ugr..i that some piece of conduct has he that con6uct rule of a oi thut, is inconsistentlyclisregarding others that desire to accepted;or again of thinking it inconsistent will not do to one what one proposesto do to them. The fact is that its the man who rejects morality because he seesno reason to obev (1972: l6l ) inconsistency. not of villainy but nrles can be c<tnvictecl

t9.1

Hozu to Sohte \he LIorul Problem

How to Solue the Moral Problzm

195

F()()t's objecti<trt herc is, in ect, th:rt there is n<l rviry o'l-ari()nirllv criticizinq rationalist the clesircs o'thosc who <:lairns that tht're is rnust bc making rvh<> rcjcr:t mr>ralitl', so the 'illicit sorne - ()r, as he calls it, 'at>

l l s s r u l rp t i ( ) n ' . ()ilbert Hlrrnrzrn objccts to rationalisnt soltttisrn' - in a sinrilirr vcin.

i r li r n i n ; r l , r v h o d o r r r i t I f ] h c r e a r e p c o p l e , s t r c h : r s t l r t ' s r r < < ' < ' s s c obser-r'c t h e a l l c g e r l r c q r r i l t ' r r r t ' r r tl i ( ) t t ( ) h a r n r o r i n j r r r e o t h e r s a r r d t h i s i s n o t r l r r c t o i n : t t t c r r t i o r r , i r i l r r l t , t o c o r r s i c l c r ' ( ) ra p p r e ( i : r t e ccrtain ilrgrrr)rcr)ts,ignorlrnct' o r'r'lcvirnt evi<leltcr', errors irr r t ' l t s o r t i l r g ,i r r a t i o n a l i n , , u n l ' e a s ( ) r ) i l l t l e r r t ' s s o, r w t ' l t k l r e s so f r v i l l . . . T h e a b s o l r r t i s tn r i e h t a r g u e t h a t t h c c r i r r r i n n l n r r r s t b e i r r a t i o n a l ()r'at least unreilsonable. Sceing tllrt a propose<l corrrsc of a<tion w i l l p r o b i r b h ' c a r r s c s ( - r ' i ( ) u \i n j r r n t o s o l n e o u t s i t l c r ' , t h c c r i r n i n ' . r l rlot's not trc:tt this ils 2l re2ts()ll n{)t t() rrnclertake that r'orrrse o' -l-his acti()n. rlrrst l>c irr:rtional oL rrnrcasonable. ttccausc strch a c o r t s i c l c r ' : r t i o ns i r n p l v i s s r r c h a r e i l s ( ) n a t r d i n d e e c l i s : r n o l l ' i o r r s r'('1rs()n a, b a s i t l ' e a s ( ) n , n o t ( ) n e t l t : r t h a s t o l t c c l e r i v c d i n s r l r r r c c o r t t P l c x \ \ ' a l t l ) f ( ) u g h a r ' ( i r r ) c r e a s o r r i r r t t .B t r t t h e r r i t r n u s t b e i r l l t t i o n a l ( ) l ' a t l ( ' i r s t t r n r e A s o n : r b l eI i r r -t h t : r ' r i r n i n i r l l ) ( ) t t ( ) < ' i r l es r r icit'rrtlv ultorrt otht'r-s, sint'<' tlte crirnin:rl's lack o' concern irr otltt'r-s is uhlrt is rt'sPonsilrlt' irr ttrt' crirrrinal's rt()t takintr the l i k e l i h o o c l o l t i t t n t t ( ) a n ( ) r r t s i ( l ( ' r ' t (l,) ( ' l r l ( ' : l s ( ) n l r g a i r r s t : r1 l r - o 1 t , r s t ' d
(()llls(' o lttlitltt . . .

imagine that he is ttot inconsistent, that he does not disregard a rtrle o'conducl that hc lrits himself already accePted, that he is not inattentivc. tltat ltt' clocs t)ot suffer from weakness of will and so on and so irrtlt . . . Ihrt all that that shows is that the flaw in his r e a s o n i n g l i e s i n t h t ' l t r t ' r r r i s t 'f r o m w h i c h h e b e g i n s : t h a t h e h a s a norrnative rc:ts()ll t() gait) w('alth no n)atter what the cost to others. For, as we ha\'(' s('('n, this is equivalent to the claim that fully rational creatrlr('s woul<l want that, if they find themselves in the criminal, then they gain wealth no circumstances o tltt' stt<'ct'ssful matter what tltt' (()sl l() others. And the successful criminal's opinion notwithstatrrling, it seems quite evident that we have no reason to belicvt' th:tt this is true. Fully rational creatures would want no such thirrg. Note what I havt' not said. I have not said that the fact that we all clisagree with tht' slr('cessirlcriminal entails that he is wrong. Perhaps we arc all nristakelt about what fully rational creatures would want. But the mere act that it is logically possible that we are wrong gives tts no more reason to endorse the opinions of the successful crinrinal and doubt our own convictions than the mere fact that it is logically possible that we are wrong when we think that the sun will rise tomorrow gives us reason to endorse the opinions of the prophets of doom. Harman tells us that the fact that the successful criminal does not observe the moral requirement not to harm others cannot be 'failure to consider or appreciate certain arguments'. traced to a But it seems to me that that should now seem quite wrong' After all, the successful criminal thinks that he has a normative reason to gain wealth no matter what the cost to others, and he sticks with this opinion despite the fact that virtually everyone disagrees with him. Moreover, he does so without good reason. For he can give no account of why his own opinion about what fully rational creatures wotrld want should be privileged over the opinion of others; he can give no account of why his opinion should be right, others' opinions should be wrong. He can give no such account Fecause he rejects the very idea that the folk Possess between lhern a stock of wisdom about such matters against which each pcrs()n's opinions should be tested. And yet, ultimately, this is the only <'otrrt of'appeal there is for claims about what we have norma-

T h e r - r ' l a t i v i s t l' t s' 1 r l vt ( ) s u (l l i l n : r r ' u l l r n c nits t h a t , o n u n v p l a r r s (s os r ible chrtractcrizlttior ot ' r ' e a s o n a b l c r r c a ss rrtlrrnreasorrablcner r a t i o n a l i t v i r n r l i r r ' : r t i o r r a l i t t 'a )s l t ( ) t i ( ) l r s thitt can bc pat't o the ) t h t ' w o r l d , t l r e a b s o l r r t i s t 'c s c i e r r t i i c ( ) l l (( ' l ) t i ( ) l o sl a i r n i s l r r s tf u l s e . Sonreorrc c a r r l > t 'c o r n p l e t e l rr a t i o n : r lu i t l r o r r t e l i r r gc o n c c r - n irrr(l (.1 9 8 5 :3 9 - 4 0 ) r ( ' s p e c t i r r o r r t s i c l e r s I n c c t , t h e n , b < l ( h F l a r - r n a na n d F o o t c h a l l e n s c t h e r : r t i o n a l i s t t < l c x p l a i n u n u n c ( ) n t r o v c r s i i r l s e n s e i n u h i c h t h e s t r c c e s s l i r lc r i r n i n a l i s b e i n g i r r i t t i < t n a l .( l ( ) n t r a t ] t o t r o t h , l r o w e v e r , i t s t ' c n t s t ( ) l l l c ( - a s v t() nleet this challcnge. T h e s r r c c c s s f i rc l r i n r i n a l r n r r s tb c s i n h i s c l c l i b e r a t i < l n s l ' o r n s < l r r r c c v l r l u a t i v ep r c m i s c o r o t l ) e r . L e t ' s i r n a s i t ) c t h a t h c b c e i n s i - o r n t h t ' p r e r n i s e t h a t h c h a l sa n o r n r a t i v e r c a s ( ) n t o g a i n \ 4 c a l t h r ) ( ) l n i t t t ( ' l ' u ' l t l r t t h e ( ' ( ) s tt ( ) ( ) t l r e r s . N c l w , s i v c n t h i s l t r c n r i s e \ v ( . ( ' u r r c t . r ' l u i r r l v

lgti

Hou to Sohte t.heL4oral Probkm

Hou ltt Soh'e tlp Moral Prohlent

197

tive reason to do. The successfirl crirninal thus seems to me to strffer fiom the nll too comlnon vice of intellzctual rrroganrc. He 'ailure to consider or appreiiierefore does indeed strffer fronr a ciate certain argulrents', firr he doesn't feel the force of arguments that cclnte fr<tm othersirt all. Gauthier's Objections ln Morals b,9Agreemntl David Clauthier contrasts two competing conceptions of rationality and two competing strategies for demonstrating the rationality of moralitv. Let it . . . be agreed that in so r as the interestsof others are not affected,a person actsrationally if and only if she seeksher greatest interestor benefit. This might be denied by some, but we wish here to isolate the essentialdifference between the opposed conceptions of practical rationaliry. And this appears when we consider rational action in which the interestsof others are involved. Propo nents of the maximizing conception . . . insist that essentially nothing is changed . . . On the other hand, proponents of the conceptiou . . . insist that what makes it rational to unhtersalistic an interest does not depend on whose interest it is. Thus the satisfy rational pe rson seeksto satisf, all interests. . . The main task of our moral theory - the generation of moral constraintsas rirtional - is thus easilvaccomplishedby proponents conception of practical reason. For them the of the universalistic relation between reasonand morals is clear.Their task is to defend their conception of rationality, since the maximizing and the conceptionsdo not rest on equal footings. The maxiuniversalistic the virtue, among conceptions, of mizing conception possesses weakness.Any consideration affording one a reason for acting on the maximizing conception, also affords one such a reason on the universalistic conception. But the conversedoes not hold. On the universalistic conception all persons have in effect the same basis for rational choice - the interestsof all - and this assumption,of the impersonality or impartiality of reason, demands defence. (1986:6-7) Since Clauthier embraces the maximizinp; conception of rational'mirin t:rsk ity, he therefore sees his own problem in achieving the

o f m o r a l t h c o r y ' t o b < ' t h t ' < l p p o s i t e .I l e t h i n k s h e d o c s n ' t h a v c t < r defencl his owrr vit'ws ulrorrt practical rationalitv at all. Rathcr he h a s t o s h o u l r o r r ' i t <o r r l r l l r t , r l t a t t h e m a x i m i z i n g c o n c e p t i o r r ' ' tnoralitv. u n d e r w r i t t ' s l l r < ' r - i r t i o r r a l i t lo ( w ( ) ( ( ) r n l ) ( ' t i r r q conceptions of practical rationality Gauthicr's :rre supp()st'rl to rt'prt'st'nt n'hat would be characterized in mclre t r a d i t i o n i r l t ( ' r ' n l sl r s t l l ( ' r a t i < l n a l i s t ' sa n d t h e a n t i - r a t i o n a l i s t ' s c o r r c e p t i o n s . ( ) a r r t h i t ' r ' i s i r r r r l v i n v o u r o f t h e a n t i - r a t i o n a l i s t ' sm a x i mizing, t>r [Itrrrrt'an, ('()n('cption clf practical rationality. He has t w o f u n d a r r r t ' n t : r l <r ' i t i < ' i s r n s ol' (he rationalist's conception. First, he objects thlrt tlrt' r'utionalist nrakes all the sante assumptions about prncti<'ul nrtionrrli$' :rs the anti-rationalst plus mor4 so that t h e r a t i < l n a l i s t ' sv i t ' w o p r i r c t i c a l r a t i o n a l i r y i s r e l a t e d t o t h e a n t i rationalist's as:r str()ngt'r view is to a weaker. And seconcl, he objects that what thc rationalist aclcls to the anti-rationalist's assumptions ab()lrt practicirl rationality is, in effect, the assumption that it is irrationirl not to h:l,c concern for other people. The rationalist thrrs hands hersell'the conclusion that rationaliry is on the sicle of morality on a plate. Horvever the analysis of normative reasons offered here, and the interpretation of rationalism, suggests that Gauthier's criricisms are radically misplaced. For, to begin with and most fundamentally, Gauthier is simply wrong to assume that the maximizing concept.ion of practical rationality is uncontroversial. Indeed, as we have seen, no one should accept the maximizing conception of rationality. Rationalists and anti-rationalists alike should reject the view that what we have most reason to do in certain circumstances is whatever we happen to desire most to clo in those circumstances, in favour of the view that what we have most reason to do, in certain circumstances, is whatever we would desire most that we do in those circumstances if we were fully rational. And once this crucial point is agreed, we see immediately what the flaws are in Gauthier's fivo criticisms of rationalism. Consider his first criticisrn. Once it is agreed that the fationalist and anti-rationalist both have to accept an analysis of reasons in terms of what would be desired under conditions of full rationality it becomes quite clear that the rationalist isn't just saying everything the anti-rationalist says plus more. Rather, the rationalist

lgfl

Hou to SoLue the LIorol Problem

Hou to Soh,ethe Moral prcblcm

lgg

and anti-rationalist can agree abottt sontething ftrndatlcntal, while yet still clisagreeing abortt sonethin{ fttndarncntal. Thcy' can both aJrce th:rt practical rationalitv is to be defincd in terms o f t h e u n c o n t r o v c r s i a l p r o c e c h r r c sv i a w h i c h w e - j u s t i $ ' o u r d e s i r es ; tl'ratonl,v so can we charactcrize rvhat ftrll riltionalin'ccinsists in. 'reflectivc equilibrium' Each rnay therebre be said to accept a concep[ion of practical rationalitl'. Rut thev still disagree abottt something fundamental becatrse thev rlisagree abottt the scope for reasoned change in our desires r.'ia the process of relective eqtrilibrium. The rationalist thinks that the existencc o'reasons presupposes that under conclitions of full rationalitv we wotrld all har,'e the same desires about what. we are to clo irt the l'aricltrs circumstances wc might ace; that absetrt such a c()nvergencc we should say that there are no reasons at all. Arrd the anti-rationalist c l en i e s t h i s , c l a i n r i n g i n s t e a d t h a t r e a s o n s n ) a y e x i s t e v e n i f w h a t an)' particular agent would desire ttnder conditions of trll rationalitv clepcndscrucially upon lvhat her actual desireswere to begin witlr. The rationalist thus accepts a non-relaliuereflective eqtrilil> ritrrn conception of practical ratiotralitr', whereas the anti-t'ationalist acccpts tr relatiuereflective equilibrium conception of practical rationalir'-. ()auther is there<lre wrong that raticlnalism and arttirationalisnr are related as a stronger doctrine is to a weaker. Rather thcv clisagree fundarnentally about whether normati'v'e reASOnsare t(.rbc anall'sed in relatir''c or trotl-relative terms (see again the discussion in chapter 5). Now consider ()atrthier's second objection. He tells us that the rationalist hancls herself the conclusion that it is irrational not to have concern for other people on a plate, fclr she sintplv deines practical ratiorralitf in these terms. This is the so-called 'universalistic' cclnception of practical rationalit)'. But, as we have t h e r a tionalist need not atrd should nclt deine pritctical s e e n , .just rationality in these terms. Rather, like the anti-rationalist, she can and should define practical rationalitf in terms of relective equilibrium, and she should then distinguish herself from the antirationalist by insisting that, on her cotrception, the existence ol r e a s o n sp r e s u p p o s e sa c o n v e r g e n c e i n t h e d e s i r e s o f l i r l l v r a t i ( ) t ) a l creatures. In other words, she should clistinguish hersel{ i'orn tht'

anti-rationalist bv insisting that, as she sees things, reasons are non-relative. For orrct' tht' rationalist has so definecl her co.cept i o n o f p r a c ' r i c u lr u r i o r u r l i t \ ' ,s h c c a n t h e n i n s i s t t h a t i t i s i r r a t i o n a l not to havt' r'..r'r'r. irr' .ther people if and only if, rvhen we ( h < : cngagc in r n t ' t l r , d . r ' t ' l c c t i v ce q u i l i b r i u m i n o r d e r t o j r r s t i i , o u r d e s i r c s ,a ( ( ) n ( ' ( ' l - n i r r 'o t h e r p e o p l e t u r n s o u t t o b e a m o n g t h e desiresthat arc rht'rrrst'lvcs r a t i o n a l l y - j u s t i f i a b l e ;o n e o f t h e d e s i r e s any persor) worrlrl lrlrvcrrndcr cclnditions of full rationalit)'. In this way, not onlv d.t's sht' not have to make the conclusion that it is irrational n()r (() lurvc r'<ln<'ern fcrrothers a trivial consequence of her definition . pr':r< tical rationalitl', but she can insist that it may w e l l b e h o t l y < o n t t ' s t t ' < lw h t ' t h c r o r n o t i t i s i r r a t i o n a l n o t t o h a v e c o n c e r n i r r o r h t ' r ' ; r r ' o P l t ' .F o r i t i s c o n t r o v e r s i a l w h a t t h e o u t c o m e o f t h e m c t h ( x l r r ' < ' l t ' < ' r i v< t ' q u i l i b r i u m w o u l d b e . I n c r e e d ,o n t h e r a t i o n a l i s t ' s r r o n - r ' t ' l l r r i v tr ' 'orrception of reasons, it rnav rvell be h o t l y c o n t e t s t t ' dw l r t ' t h c r t h c r . c a r e a n v r e a s o n s a t a l l .

Mackie's Objections John Mackic tells us rh:tr

. . . I(a.t hir.self thought rhat moral judgenrents are categorical imperati'es. . . So far as ethics is co'cerned, mv thesis that there are no objective values is specifically the denial thar any such categoricallyimperarive elemenr is objectivelvvalid. (1977: 29) Mackie's rwo fam'trs objections to the claim that there exist objective valtres are thus directed quite specifically as:rinst the rationalist. Consider his first argument. If therc were objet'ti'r'e 'alues, then thel would be e'tities .r qualities or relatio.s of a very strange sort, utterly different fr,m arrvthing else i. the rrniverse.Corresp<lndi'glv, if we were arvare.f them, it would have to be by some specialfaculty of moral perceptio. or inttrition, utterly different from ordinary waysof k.owi.g evervthing else. (1977: 3tt)

200

How to Sohte the Mornl Problern

Iltnu lo ,\olzu,tlte i\Ionil l)ntbknt

20 1

But can Mackie realll' lay a charge o-strangenessagainst riehtness, at least as that feature of acts has been analysed here? It seems not. To sltr that perfrrrntirtg an atl ol a ct't'taitts()rl in certain circumstances is risht is, I have argtted, to say intetaliu, that there is a normative reason to perform it. And this, in turn, is simply to say th:rt fullv rational creatures would clesire that such an act be perfcrrnrecl in such circuntst2ltlces, whet'e such a desire is of the zrppropriate substantive kind. ()n this Accoltnt, moral atures 'entities or qualities or relatious like rightness thus simplv are not 'special faculty <lf tnoral Percepti()n of o'a very strange s()rt'; no i n t r r i t i o n ' i s r e q u i r e d i n o r d e r t o g a i n k n o l v l e c l g eo f t h e m . A l l t h a t is required is the ability to think about what a more rational person would want. And this, in ttrrn, jttst requires the abilitv t<r ellgage in an attempt to find a systenlatic and uniletl.justification of the various desires that we have; to engage in a process <-rf reflectir,e equiliblium. The charge o'stranseness is thus entirely misplaced. Consider now Mackie's second argument. The icleahere is that . . . radical direncesbetween first order moral judgeme nts make :ls apprelrensionsof objectrve it dificult t() tretrt those.itrdgenlerlts that tells the rnere ()ccurrenceof disagreements But it is r-rot rrurths. ( ) n q u e s t i o n si I l a g a i n s t t h c o b j c c t i v i t yo f v a l u e s . D i s a g r e e n l e n t history or biologv or cosmology does not show that there are no itr these fields br investigatorsto disagree abotrt' issttes r>biective But such scientirc disagreement results i'om speculative inferences or explanatory hypothesesbased on inadequate evidence, and it is hardly plausible to interPret moral disagreenlentin the sarne wav. Disagreement about moral codes seems to reflect people's adherence to and participation in different waysof li. (1977:36) N{ackie thrrs agrees that the availability of a rational procedure for 'mere occtlrresolving moral disagreement would render the ratignalist's rence' of moral disagreements harmless, from the that the idea that there exists strch poitrt of view. Btrt he sr-rggests 'hardl,V plausible'. His actual arsutlrents ilr tIis a procedrrre is claim are, however, hardly convincing.

M a c k i c ( t ' l l s r r s t l l : l t n l ( ) l i r l r l i s l r g r c c r r r e nc t r r n n ( ) t p l a t r s i l r h ,l r c 'sltcculative interpretc<l as r.t'srrltirrg i'ortr inrenccs <tr explanatory hlpotlrt'st's lllrsr'<l orr irur<lt'<lrratc cl'idence'. But, in rtranv rvays, t h i s i s . j r t s t : r l s t ' .l i o r ' , l r s r r ' r ' l u r v t ' s c c n , w h e n w e c ( ) n s t r u c t l r t o r u l t h e o r i e s \ \ ' ( ' i r r ' ( 't r l i r r g t o i r r < lo r r t w h a t w e h a r v en o l ' r n a t i v e r e a s ( ) n t o d o , a n r l l v h t ' r r \ \ ' ( 'l r ' \ 't < l i n < l o r r t w h a t w e h a v e n o r m a t i v e r e a s o n t o d o w e l r r t ' t r l i r r g t o i r r < lo r r t r v h a t [ r . r l l vr a t i o n a l c r e a t u r e s w o u l c l lvan[, an<l ulrt'rr rr't'(rr to incl ()Lltrvhal fully rational creatures w o u l d w a n ( w ( ' t ' r r r l l l r r ko n a p r o c e d u r e o f - j u s t i f i c a t i o n o f o l r r v a r i o u s d c s i r t ' s t h r r t i s v t ' r ' r ' s i r r r i l a ri n c l e e d t o t h e e n t e r p r i s e o f t h e o r y constructi()rr irr scir.rtr t' in lcsponse to observational evidence. ; a r e s i r r r l r l vr r l r p l i < : t t i o n so f t h e i d e a o f r e f l e c t i v e e q u i l i b n u m . aBoth O f c o t t r s r ' ,1 t t ' r ' h i r l r M s rrckic is sayingthat this attempt to provide nd o o m e d t o f a i l u r e ; t h a t ollr desircsn'ith a svst('rr):rtic.iustificatio is n o s u c h . j t r s t i i < ' i r t i o nw ' i l l b c b r t h c o m i n g ; t h a t t h e r e i s n o s i n g l e t h i n g t h a t i r l l v r a t i o r r a l c r e a t r r r e sw o u l d a l l w a n t u s t o d o i n t h e various circurnstancr's in which we find ourselves. Perhaps that is the real point o his argurnent iom relativity. Br-rtif so then the reply is again simple enough. For though it is certainly logically possible that no systematicjustification of our desires is forthcoming, in light of the remarks we made in the previous section about the power of moral argument to elicit agreement, it seems more reasonable to think that such a justification is forthcoming. This is not to deny, of course, that the real test of Mackie's argument from relativity lies in the ultimate outcorne of debate in normative ethics. The real question is whether we will, by engaging in such debate, come up with answers to moral questions that secure the free agreement of those who participate. If we do not, then our conidence that such agreement is to be had may be undermined, and quite rightly so. But that gives trs no reason for scepticism now. We must give the arguments and see what their outcome is.

6.5

CONCLUSION

The taskin this chapter has been to provide an analysis of our

concept of rightness in terms of orrr concept o[ a normative reason. The analysis providecl has many advantages.

909

How to Solae the Moral Problzm

First, the analysisallows us to solve the moral problem. For it gives us a way of thinking about our moral beliefs as beliefs about an objective matter of fact, and it explains why these beliefs connect with motivation, but it does each of these things while all the while taking it for granted that Hume was right that our motivations are to be explained in terms of beliefs and desires, where beliefs and desires are distinct existences.Second, the analysis allows us to square morality with a broader naturalism. For it allows us to seewhy the rightnessof an act just is a natural And third, and finally, in conjuncfeature that that act possesses. tion with some plausible assumptions about the potential that moral argument has to bring about agreernent,the analysisalso allows us to think that our moral talk is in fact legitimate. For it is plausible to suppose that through moral argument we can in fact discover what the reasons that we all share really are. At the beginning of this book I said that, as I see things, debate in normative ethics is crucial for the final resolution o' metaethical questions.It should now be clear why I said this. For, given the analysisof rightness argued for here, the -justifiability of our commitment to morality is itself a hostageto the fortune of debate in normative ethics. As I see it we are justified in thinking that there are moral facts, and so in engaging in ordinary moral debate, but our justification is defeasible,and nray itself be defeated by the outcome of those very debates.If we are interested in the final resolution of meta-ethical questions - in whether or not there really are any moral facts - then it seems to me that we therefore have little alternative but to engage in normative ethical debate and to see where the arguments that we give ultimately lead us.

Notes

NO'I'},S

'T0 (]HAPTER I

WHAT IS THE

MORAL PROBLEM?
Crispin \A'right (11)92) and Paul Horwich (1992) have both recently irrgued thut orr a rninimal theory of truth, we have no choice but to reg:rrtl rnorirl t'lrrirnsas apt for assessment in terms of truth and falsehood. I they wcre right then, on the minimal theory of truth, we worrld havr: t<l rcgard rnoral claims as expressions of belief. A preferencr' irr the expressivists' solution to the moral problem would thcn have to depend on taking a non-minimal view of trutt^. However, as I have argued elsewhere, contrary to both Wright and Horwich, expressivists need not take a non-minimal view of truth. Minimalism about tnrth is neutral with regard to solutions to the mo,al problem. For relevant discussionsee Smith, 1994a, 1994b; Divers and Miller, 1994; Horwich, 1994;Jackson, Oppy and Smith, 1994.

NorES ro

"";;*#"

EXPRESSrvrsr

Here I follow Wiggins (1987: 193 note l2), who reminds us that Moore himself gave the following useful account of naturalism: 'By nature then I do mean . . . that which is the subject matter of the natural sciences,and also of psycholog/' (Moore, 1903: 183 ancl note 40). This certainly seemsto be what Ayer has in mind by the 'naturalism' term as well. Note two points, however.First, for this to 'natural' be a good way of defining we must assumethat ethical facts are not thernselves part of the subject matter of a natural or .social science. The assumption might seem harmless enough ()l('c w(' rrote that, even so, the mahingof ethicalclaimsmay still be part () tlr('

204

,\'ote^r
srrbjt'ct rnntter o a nertural or a social s<ience. And seconcl, note ttlat rvhen'n:rtrrral'is dt:fined in tlris wat', it trlrns ()ut that a ntrtrlral fezrture of the world rnav ()r rnav n()t be a ph1'sical feattrre (.fackson, 1992: lfll). A possible world in rvhich there irre creaturcs rvith ccrtain psv<'holouic:rl traits has natural featrrres, accorcling t() this dcfinition, xrt it mar'()r nlliv rrot be a world with phvsical fcatrrres. The creatures rnay all be non-phvsical l>cings: irngels, irr cxarnplc. A natrrralist in ethics thrrs nt't'<ln()t l)c a phvsicalist. Moore cxpresses his rclrrctarrce t() p()sit a pcrccpturl aculq' of 'l rnriral inttriion in these terrns: worrld rvish it obscned that. when I call ritrch propositions "lntuitions". I me:rn nterclv t() assert that tlrev are irrc'apablc of proof; I inIh noth,inguthnln,n as to thr ,nunn?r or origin of our cowtitiort tl them' (1903, x). tor rli.scrrssion o'thc gencr':rl rlucstit>n rvhcther rnoral fi'atrrr-cshar.'e lun t'xpl:rrratorl' role see f{arrnan, 1977; Strrr.seon, l98l-r, 1986; H a r t n a r t , l 9 f | 6 ; a r r d , e s p c r . ' i a l l r 'S , avrc-\{c('ord, 1988. Note that though, irccolding t() this st()n'. we know whiclt n'.tturitl {t'attrrt'.srr'r' liavt'()ur pr()- arrrl con-attilrrdcs tt>rvardsorr t}re basis o' rnelt' reflcction, :rnd thotrgh lve arc thcrcbv able to cxplairr holv wc <arr nr:rkc rnoral .jucluenl('nts in accorrlan('c \vitll p:lrti('uLir sup('rv t ' r t i c r r < e c o r t r l i t i o n i r l s o n t h e l > a s i so ' r n c r c r e f l e c t i o n , i t i s n ' t t h e (fts(' \\'(' krrorv n !rioi tltut olr-icc'tshai'ing ccrtailr nntrrral f'cuttrr-t-s havt'tt'r'tairr rnolal fcatrrt'es. I n o t h t ' t ' w o r c l s ,r v e h a v e n o t c r ' ( ) s s c d arl 'is<rtrglrt g l p r ' . F o r , r ' r ' r l ' r r r r r t r l t l r ' ,( l t o r r g l i r r , c k n o w ( ) n t h e b a s i s o ' r c l t ' r ' t i o r r l v h a t t h e c ( ) n t c n t s o l ' o r r r -p r < r a n d ( ( ) r ) - a t t i t u d e s arc, this is rrot krrorr'lt'rlgt' rt prirtri, btr( <>rtlra lto.:(tiori. Rt'lr'< tion helc is thus n ( ) t t ( ) b c t h o r r g h t o r r s t ' r r t u i l i n g a l t r i o f i a c ( ' e s s .R e l e c t i o n i n t h i s o f ( ) r r r '( ) \ \ ' n u ( ) n - ( ( ) g n i t i v r -s ( a t e s . < i i s ei s n t o r c ' l i k e i r t r t c r p c r ' < ' < ' p t i o n -lhorrsli l l a r e i s , I t h i n k , r ' i g h t l v d e s c r i l > e c la s i r n e x p r c s s i ' , ' i s t h , is vicrr's arc irr act ratllcr cotnplicatccl ( l9ll ). Irr ltis vicrr rnoral.jtrrlger n e n t s h a v e b o t h z r n t ' x p r t ' s s i r . ' t ' r r n c t i o r rr .vhich is;rrirlarr', ancl a < l c s c r i p t i v e i r r r c t i o n , w h i c h i s s c c o n d a r y . \ { l r e n s o r n t o n e . j u d { c s a n l() be right she is both describing it as having thc rrirtrrr:rlfeatllres 2r( that. or her, rnakt:s prescribing it appropriate, arrrl slre is also prescribing it or expressir-rghcr pro-irttitrrde towarcls it. In Hare's r,it'rv,thc prescriptivc or cxpressive urrctiorr of rnolzrl.iuclgcrnents is best c:rpturctl semanticallv by ascritrins to tht'rn tlre linerristic irrrn o an irnperirtive. Notc that thc clispositional ar-ralvsis has hcre becn irrrnrrluteri so :rs to rrrake it trarrsparcnt that wc ('an irlcntif'r' a clispositiorr rvitlr its t'att'soricul trasc (Evans. l9U0). Thotrglr this vicrv is r'orttrorr'rsiul ( s t ' t ' .I i r r < ' x r u t t l t l r M ' , < ( i i r r r r , l 1 ) l l 2 ) , l : r r l o p t i t i r r o r r l t ' r .l o r n : r k c t h t '

Nolc.s

205

p o i n t s t h i r t l i r l l o r v t ' : r s i t ' r ' 1 os t l i { ( ' .l l v e r f . ' t h i n g t h a t f i r l l c l w sc o t r l d s t i l l b c s a i r l . t l t o r r g l r t l r c r w o r r l r l h a v t ' t o b e s a i d s o n r t ' w h : r td i f f e r - t ' n t h ' ,i f t l u t t c l i s p o s i t i o n sa r e c o n s t i t u t e d b y , b t r t n c l t w e t o o k i r r s l c l r <tl l r c v i < . l v idt'ntit irl rritlr. tlrt'il crrtt'gor.i<'a bla s e s . -fhe A n a r u r l o g vr r r i s l r t l r t ' l r t ' 1 1 ; i r l . view that'risht'refers to the 'rieht' cartst'o ()ur u\('s o tlr<'worcl n r i g h t t r s e i r l l vb e c o n r p a r e c l t < r a r c l i r t t ' r l v i c n l r l r o r r tn : l t r l r a l k i n d t e r r n s ; i n d e e d , o n e w a v o f u n d e r i t i s ; r s l l r < ' r i c r vt h i r t ' r i e h t ' i s a n a t u r a l k i n d t e r r n . S t r p p o s e starrrlirrs '\r'lrt('r' r t ' t ' r ' st o r v h a t c v e r n a t u r a l k i n c l i s t h e c a r r s eo f o u r orrr rvolrl 'rllrtt'r'. Then, irsPrrtnam (19ll) fanroush'poitrts rrst'o tlrt'rr'olrl <xrt. thclt' rrrirt rvcll l>t' another cornmunity whiclr uses a u'orcl, 'rv:rlt'r't" s:r'r: .r wolrl rr'hich pl:r,va likc the s role in their langtragc.jrrst r o l t ' ' n l r t t ' r ' p l l r v s i r r o r r r l a n t l r r a g e- t h c y r r ) a y u s e i t t o r e f e r t o t h c s t r r l i t l r : r l ( ( ) r n ( ' s r ' r > r ll i v c l s , l a k e s u n d s t r e a l t r s ,i s g o r ) d t . <d l rirrk. and s() ()n - llrt rllrost' r't't'rt'nce difrrs fiom the rerence of orrr rvord '\4ill('r''. 'u'ater'', I;or rr'ltt'r('lrs tl)(' cillls:rlllist()11'of <tul rvortl giverr t h : r t i t i s : r r r : r t r r l l l k i n r l t t ' r ' n r ,c r ) s r l r e st h a t i t r c r s t o F { " O , t h e c z u r s i r l 'watt'r*', histon ol tlr<'iruol'<l given that it is a natrrrul kirid tcrrrr, -fhus, c n s l l r ' ( ' st l ) l r t i t r ( ' ( ' r s t o X \ 7 - . even though orrr words'water' 'rr'rr(t'r''1" :rrr<l p l a v t l i t ' s a r n c r o l c - t h e v a r c e a < ' h t r s c c lt o r c r ' t o t l r e natur-ul kirr<l thitt is inrncl in rivcrs, lakes ancl streanls, is gr>ocl t<r c l r i r r k , ; r r r <s l o o r r - t h i s r v i l l n o t b y i t s e l ' e r r a r a l ) t ( ' et l r a ( w e r v o t r l c lb e < l i s a g r c c i r r ei ' w c s a i d , o f c c r t a i n s t u f f , ' T l . r a t s t u f i s w a t e r ' i r n < l t h e l ' s a i c l ,o t h c s a u r c s t r r f , ' T h a t s t r r i s n ( ) t t v a t e r * ' . I n t h t . s e t e r m s . t h e 'right' objcction to the r,.iewthat the word rers to the carrse of orrr r r s e so t h e w o r d ' r i s h t ' s t a t c d in the text nlay n()w bc ptrt likc this. Wrercas the possibility o' explaining such disagrccrnents awav is acceptablc n the case of tw() cornntunities wiro rrse natrrral kind 'water' 'water*' ternls - like and to play the sanre rolc in their livcs, the possibiliry o'explaining such disagreernentsawuv is trnacct-ptable ir-rthe case of two conrmtrnities who use a word to pl:ry the same 'risht' roler in their lives as the word plays in orrr livcs. Yet nrctaphvsicerl-hut-rrot-clefinitional naturalism leaves open the possibiliry that rve shorrld erxplain such disagreements away. I owe this useul terrn to I'hilip Pettit. It rnigl-rt be thotreht that wc don't yet have an argunrent that worrld :rllow us t() square colour talk with a broadt'r physicirlisrn pn .se, zrsthe argunrent-iust given has no be:rring <tn wl.ret.hera strbject's experience of having something look red to her is itsclf a phvsical state. But the f<rregoing discussion suqgests an obviotrs stratew irr sclrr:rring talk of colrirrr experience with phvsical talk aswcll. J'hc irst srclr r v o t t l c lb c t o < ' o r r s t n l ( ' t : r na n a l y s i so ' o u r c ( ) l l c c J ) to f i r < o k r r r r - ( ' x l ) ( . r l -

(i

206

Notes

Notes

207

irllow encc. The second stagewould be to show how these analyses us to identifr colour experienceswith, say,stateso'the brain. I, as seemsplausible,our concept of a colour experielrce is the concept of a state o'a subject that, in conjunction with a relevant desire, causallyexplains our bodily nrovements- fbr exarnple,our picking out red objects from objects of other colours - then it should be clear enough how the attempt at vindicatiou would go, and why it (cornpare l,ewis, 1972). should be deerned likely to be successfitl NOTES TO CHAPTF]R 3 THE

Snrith, irrtht'orttirtg). .'\ thcory .r foregrorttrd rationality will spell ()ut tht: <'orrsirlt'r'rrtion But note that we actior.r. th sa t r a t i o n a l l y . j t r s t i f y r : a n , r r r r c l c lt l r t ' r r : r r r r t ' o u ' t h c o r y o ' r a t i o n a l i t y ' , d e s c r i b t ' s ( ) l n e t h i n g clse as rvt'll. \\'t' r'rrrr<lt'stribt' thc psvchologv of the perctlv rational agcnt; s()lll('()n(' u'lro t'rrrbritct:ssttch ftrregrotttrd reas()ns. Pettit ancl I call this tlrt'tlrt'orr'o''background'rationaliti'. Will a theory of arl agent vc l l u s a b o r . r t h e m o t i v a t i n g r e a s ( ) n s b a c k g r ' o r r r r rr l ':rliorr:rlitt h a s ?( ) r r l r . i t l r t ' r r g t ' r t l i t r r l t r e s t i o ni s p e r f e c t l y r a t i o n a l . F o r m o r e o n t h i s , s t ' t ' r ' h : r p t t ' r '5 . tll:rt tll('rt'rlrriretnent is'in the broird settse'a requireIn srrggt'stirrq m c r ) t ( ) l t : r t i ( ) n : r l i t vl : r r n i r l l o w i n s W i l l i a m s ( 1 9 8 0 : 1 0 2 - 3 ) a n d P a r f i t ( l 9 f l . 1 : i ) ( ) t n ( ) t ( ' 2 : rt ( ) I ) a r t I ; 1 1 7 - 2 0 ) . B o t h t h i n k t h a t a t h e o r y o pnrt ticul nrtiorlrlitv worrld tell trs what an agent has reason to d o , r r r r r ll r o t l r t l r i r r k t h u t t h i s i n t u r n d e p e n d s , i n l n u l i a , o n w h a t i . s t h c < ' u s t ' ,r ' i l t l r ( ' r ' t l l r r r r n t ' r e l v o n w h a t t h e a g e n t b e l i e r v e st o b e thc t ast'. N o t t ' t l r u t w ( ' ( l n r l i s t i r r g r r i s ht h e t w o c l a i m s b e i n g r n a d e h e r e i r . r terrns o'tlrt'tlistinctiorr rrtadein fcrotnote l. To sav that agettts are r a t i o n a l l v r c q t r i r t ' c l t ( ) I ) r ( ) r n ( ) t et h e i r f u t u r e i n t e r e s t s ,o r t h a t t h e r e i s of rirtiortalitv to do so, is to make a claim abotrt a nornrativt' r'(':rs()n the the<rrr ol .ftrcgrourtd rationality (a claim about the cotrsiclerations that rationally jtrstiy action), whereas to say that agents are rationallv required t() have a desire to promote their futrrrc interests is to make a claim about the theory oI' bachgroundrationaliw (a claim abotrt the psychology of the perfectly rational agent). Relatedly, note that the Humean who thinks that agents have and act upon :r desire to promote their future interests need Irot, and perhaps should not, think that that desire has to be appealed to in justif ing (Pettit and Smith, I990: 5tt9-91). their actior-rs Indeecl, it seems to me that there are rnore mttndatre counterexamples to the principle left to right. Consider cases itr which you go to the refrigerator convinced that there is sornething in particular that you want, though you aren't quite strre what it is. Then, while looking at the contents you suddenly, as we should put it, 'realize what it was that you wanted all along'. lf we wish to respect this commonsense description of such occurrences then we should reject the principle left to right. I say McDowell rejects a phenomenological conception. He trowhere says that he rejects such a conceptior.r. But given that he 'consequentially ascribed desires are indeed desires' thinks that (1978: 25), he must. For the point of consequentially ascribed

EXTERNAI,IST

CHAI,I-ENGE

this point is never admitted by the externalists To my kr.rowledge themselves,larp;elybecause the problem to which admitting this point is the solution is never explicitly addressed:that is, the prol> lern of explaining the reliability of the connection between moral judgernent and motivation in the good and strong-willed persotr. Indeed, both Brink and Foot seem to think that externalism ors explanation of the connection between moral judgement a better and rnotivation than that offered by defenders of the practicality requirement (Brink, 1989: 49; Foot, 1972: 165-7). Whether they would still think so if they were to think about the point currently under discussionI do Irot know. For this reas()nI am not convinced that Hart's irrsitutionalaccount of law adequately captlrres the normativity of law either. For absent a defence of the claim that all those whcl live inside a legal systemhave tacitly agreed to obey the law - on [{art's institutional account, there can be no legitimate expectation that all those who will act in accordancewith it; no legitimate are inside a legal system srounds for disapproving of their acts of disobedience;no reason 'us' 'them'. If incleed the law has versus fbr rejecting the image of normative content, then it seemsto me that tve have no alternative but to derive that content from the normativiry of rnorality, and so to reject the very fundamentals of positivism. NOTES TO CHAPTER OF 4 THE HUMEAN

THEORY

MOTIVATION

Normative reasonsare the subject matter of what Philip Pettit ancl I 'oreground' rationality (Pettit anrl elsewhere call the theory of

20u

Notes 'fhorrgh

Nole.s

201)

clesires is that there may be no phenonrenolosical ground for their :rscription. I clisctrssthis idea towards the encl of this chapter. Tlrtrs, I contend that a subject'sfake belie'that she desires to $ is not a state that is pote ntially explanatory o'her behaviour. Even-dav experient:e supp()rts this c<;ntcntion. Relect on occasiorrs rvhen \,otr stand at the edge of a <:old swimrning pool thinking rhat you desirc: to jump in. On some such occasions yorrr body is totally unresponsive to the desires that vou pro'ess to have. As vou stand there r)r()tionlesst'ou sometirnes colllc to the <oncltrsi<ln that, contran' t() what you thousht, you didn't really want to go swimming after al.. Tl.rus,.just as we would expect if this contention were true, there are c a s c si n w h i c h a s u b j e c t b e l i e v e s t h a t s h e d e s i r e s t o Q r i e h t u p u n r i l the tirne th:rt she is supposed to act only to discove r that she in fact has no such dcsire wher-r hcr body fails to respond to her desire. Of corrrse, we can construct cases in which it might o.ppearLha 1lr) asent's irlse belie.s irbout what she wttnt.s m()tivate her. The exarnple o'John the nrusician discussed earlier who pursues a career as a rnusician even though he does not want to, though he falscly believes he does, n-rayperhaps be such a case. But it seems to nre that we ale obligecl to think that irr cascs like this the al.sebelie itself does no motivati<Jnal work; that that work is done by other desires that are present: fitr example, in this case,John's desire to please his rnother. We are <lblised to construe matters in this wa1,in <rr<lertrr e xplair) the rnotivatittnitl diJJercnrr between two ways-fohn's case might have trlrncd out. It could have turned out as described. ()r it corrlcl have ttrrrrecl otrt likc the case of the asent at the swirnrning pool..fohn nrisht ha\.'e orrnd that he was unable to bring hirnsel't<r ensase in rntrsical pursuits. His inability rnight hal'e brought horne to him thc ct that he does not want. to be a musician after all. Whv have I said that tl.ris is very rorrgh and somewhat simplified? 'p' Because where is a propositiotr about thc past, or a pr<t;tosition t() the effect that sornethins happens though not rhrough the asent's own doing, it will n()t be true that the clesire that p is a disposition to bring p about. A rnore accrtrate and fully general characterization of the lunctiorral role of desire wotrld thereforc need to bc given along the lines suggestecl in decision theon,. A desire that p is a state that disposes a subject to rnake certain s()rts ()t bets when faced with lotteries lvhere the outcorne is inter alia that p. On this account, it is still true that the difference bctween rlcsirt. arrd belief is given by their difrential roles in the explanation o action, though in this case the acti()n is all l>ettirrg ttclravir>rrr'.

r - o r r s l rr u r < ls i r n p l i i e d , t h e s u g g c s t i o n i r r t h c t t ' x t r v i l l t l o l i r r present l)rlrl)()s('sl ,r o w t ' r ' e r . F O r t l u r i n t e r e s t h c r c i s i r r t ' x 1 r l : r i r r i r r r l the rtattrlc o rnotivirtirtg reasons, and the desircs th:rt t'orrstitrrlr' sr t ' t l t t ' r e i r r e d e s i r e s f b r s t a t e so a a i r - s r n o t i v a t i r r g t ' ( ' i l s ( ) r ru tlurt rrri' n o 1i r r t l r t ' J - l r s tl,r r r r l t l t . t t a r c t o b e b r o u g h t a b o u t t h r o u g h t h t ' a g t ' r r t ' s owrr <krirrg.r\s Irrrrclt'r-st:trrd it, this constitrrtes a replt'to (l. 1". S c h r r t ' l t ' r ' ' s( 1 1 ) 1 l) rrror.cpressinc criticisnls of an earlier prcs('nt a t i o r r o l t l r t ' s t ' i r l t ' : r s( S r n i t l r , 1 9 8 7 ) . F o r f t r r t h e r d i s c u s s i o r ro ' r l i r c r ' (1992) and Vellernan (1992). tions ol lit scr'I lrrnrlrt'rstorre retween two kinds o'statewhich are suc:lr P h i l i p I ' t ' t t i t < l i s t i r r g t r i s h t 'ls tlrat llrt' worlrI nrrrst Iit with their content de.sires ancl hnbits of infn' enrc (l{)l)li: l8-11)).'l'hrrs, Iirr example, he argtres that the rnc,r/r.r p o r t ? n . \l r l r l r i t o l i r r l t ' r r i r r g i s a s t i r t e t h a t p r ( ) d u c e s t l r e o u t p u t ( ) ' b e l i < ' r ' i r r t rl r r t < 1n ' l r t ' r r t l r t ' i r r p u t i s a b e l i e f t h : r t p a n d a t r e l i e ' t h a t l r t ' < l r r i r r r s t, h i s i s t h e r e o r e . l u s t i l s m u c h a s t a t e w i t h p -+ (1,urr<i, r v h i c l r t l r t ' r v o l l r l r r r r r s tl i t : r s a r l e s i r e ; f o r i t r e q r r i r c s t h a t o r l r p s \ L 's c h o l o t l l ) ( ' i r ( ( ' r ( : r i r ln l n ' . l r r r l o r r r p : : v c h < l l o g " t n ( ) t b e i n g t h a t w a v i s n ( ) t : r r ( ' a s ( ) n i r r c l r l r r r g i r r et h t ' h a b i t c l i n f e r e n c c , b u t r a t h c r a r c ? l s ( ) l l i r r c h u r r g i r r go u r ' p s v t ' h o l o g r " .[ ) e t t i t c l a i m s t h a t r v e t h c r e f r r r e n c c t l t o t l i s t i r r g r r i s l r l t ' s i l t ' s i o r r r l r u l r i t so [ i n r e n c e , a n r l t h a t ( ' ( ) r ) s i d e r i r t i ( ) n s o < l i r t ' t ' t i o r ro i t a r c i r r a < l c q u a t et ( ) t h i s t a s k . FIis positivc srr{gestion lrl)()ut h()w rve mig}rt nrake t}ris <listinctiorr 'belie-<'har.rnellcd': is thut rlesircs, lrnlike habits o irtferencc, art' or that the belie'that the rcsportse is sr-ritable that is, hc srrggc'sts realizing lhe c()ntcnt o the desirc nlust chirnncl the irnpa<t o'tht' dcsire, whereas the belief that the resp()nsc is stritable br realizins thc content o tlre habit of inrencc need not channel the in.rpact o the h:rbit o nference. Solneone who, dcsirinl t() \', Qs, Qs onlv bec:rtrst' her belic'that she can rlr b1' $-irte channels the impact oher rf csire to r.lr. But s()nrc()ne who has ht'. ttuxlus ponens hirbit Of i n l r r . i n s n e e r l n o t h a v e t l . r ei r n p a c t o f ' h e r h a l > i t c h a n n e l l e d i n t h i s w a v i n o r d e r t o e n d u p b e l i e v i n g t h a t q w h e r r s h e b el i c v e st h a t p a r r d that p, q. Therc nccd be no role plavecl bv the belicf that believing 'ront q is a way of'believing sotnethinr thnt follows l>y rnodusI,tonens her otlrer bclie.s. I anr, howevcr, unconvinccd bv Pettit's clainr that rvc Ireed tcr distinguish desires frorn habits of irrfrence in the way' he sttssttsts. What the exan'rplc of habits o'inference seen)s t() rne to show is.lttst 'direction of fit'. arrd tlurt, for horv rntrch slack there is irr the phrasc certirin purp()ses, we do better to talk directly itr terms o[ pattet ns o dispositions, where these eet spelled out in firnctionalist tcrlns. Fot'

210

Notes

l{otcs

?ll

' r. ' ! l0

it seems plain enough that, for example, the modusponenshabit of inference must itself be intemal to the dispositions of belieling that p, believing that p --r q, and the like, not, as Pettit's way of serring things up seemsto suggest,a separate disposition hat combines uith the dispositions constitutive of these beliefs to produce other beliefs. After all, what would the dispositions constitutive of believing p be if they did not already include dispositionslike the disposition to believeq when you believep -->q, ancl other similar dispsitions? Indeed, it seems plain enough that habits of inference must be thought of in this way, because they are themselvescriticizable if they do not fit the world. Someone who has the gambler's fallacy habit of inference, for example, should get rid of that habit and acquire a different habit instead; a habit that better allows her to derive truths from truths. And this seems to me best understood as a claim about the dispositionsconstitutiveof being a believer at all. Someone who has the gambler's fallacy habit of inferencc is a defective belieuer. When I talk of stateshaving directions of fit, I theref<rre have in mind whole packagesof dispositions constitutive of desiring and believing, not states that might be so described more loosely. I therefore ignore habits of inference in what follows. What about having truth as a goal? Does this amount to desirins truth? And if it does, then since believershave truth as a goal, does it follow that believersmust be desirersof the truth? I do not think so. As I see it, talk of believers having truth as a goal is best understood more loosely. To say that a believer has truth as a goal is not to credit her with dispositions in addition to the dispositions constitutive of being a believer. Rather, the dispositions constitutive of having truth as a goal are themselves intem,al to the dispositions constitutive of being a believer at all. Thus, as I see it, it is wrong to Tuppose that there may be a believer who does not have truth as a goal. Someone who does not have tmth as a goal is not a kosher belieun at all. That is the cash value of the claim that beliefs are states that must fit the world. See again footnote 8 for relevant discussion. Must the anti-Humean concede that beliefs and desires have or-rly one direction of fit? As I have argued elsewhere, in one way this issue is entirely terminological (Pettit, 1987; Smith, l9t38a). How'Are ever in order to ask the question discussed next in the text, there psychologicalstatesthat are both belielike and desirelike?', we must take a stand on the terminological issue.I interpret this irs asking whether there is some third kind of psychologicalslat(' -

ll

(Althanr. l1)tt6) or quasi-belie.fs as I have calletl tlrt'rn t'lst'rvltt'tt' besires (1986: 56) - rrot wltctlter anv beliefs are also desircs.Nott', ltowt'rt't, that i wc clirl choose to ask this question in these latter tetnts tlrcn w e r v o r r l r ll r a v t ' t o r c s t : r t ( ' t h e H t t m e a n ' s P l a s t h e c l a i r n t l r t t n t o t i v l t l irrg rt'irs<rrrs ar(' ( ()nstitutlrt:d, inter alia, by desires-that-zrre-ttot-l>t'lit's (Smith, l1)ftfiu). Davirl l,t'wis :rll{u('s tltat the concept of an instrumental bt'sirt is irrcolrt'r't'nt lx'(:lllsr: inconsistent with the tenets of' Ba1t'siulr decisiorr lht'orr' (11)tlU). A version of Lewis's argumert irr nonquantitati\'(' rlt'tisiort theory is developed byJohn Collins (lgUU). Flowcvt'r', t'r't'rr i l,t'wis and (lollir-rs are right about this (see Huw , at wouldn't be enough to show that there th Prict"s rt'ph (11)tJ1))) arc n() lrt'sirt's:rt all. For in order to argue for that conclusion we reqrrirt' :rrr a<l<litiorrirl argtrrnent, an argument to the effect that l>esireseither. l-ewis does attempt tc) thcrc art' rro rrorr-irrstr-rrrrrt'ntal give strch an:rrl{unl('nl, btrt the argument he gives is uncotrvitrcitrg 'imper( l9ft9: irolrrott' I ). Norr-instnrnrental besires are, he tells trs, viotrs to thanst' trrr<lt'rthe irnpact of experier-rce'. Btrt why he thinks this I <lo n()t krr()w. Surelv, i'there are non-instrumental besires, all that irllows is that thcy change under the irnpact o'experience in the rnanner ofan ordinan belie, not an instrumental desire. Thus experienccs that suffice for getting rid of the belielike part will also suffice ftrr p;etting rid of the desire-like part; and experiences that suffice for acquiring the belief-like part will also suffice for acqtriring the clesire-like part. As I understand it, Nagel's view is similar to McDowell's. Nagel thinks that certain propositions with impersonal and tenseless contents are such that, if they are entertained at all, then the entertainer is clisposed to act in a certain way. Thus, for example, Nagel thinks that if I now entertain the proposition that A is in pain at t entertain it while fully appreciating the realitv of other people and times - then I am now disposed to relieve A's pain at t. However Nagel also thinks that simply in virtue of being so disposed we can 'motivated desire' to pursue the act in say that an agent has a question. For this reason he thinks that we should concede P2 to the Humean: motivation does indeed entail the presence of desire. Btrt, 'if the desire is a motivated one the explanation of it as he puts it, will be the same as the explanation of the pursuit, and it is by n<r means obvions that a desire must enter into this further explanation' (1970: 29). The pursuit may rather be properly explained by the presence of a besire , not a desire. And, if it is, thetr we should

t2

2r2

Notes deny Pl. Motivations may be constituted by besires; they are not only constituted by desires. So saysNagel. And our reply to Nagel is the same as our reply to McDowell in the text. NOTES TO CHAPTER OF 5 AN ANTIREASONS

Notes

2l:t

FIUMEAN

THEORY

NORMATIVE

Michael Stocker puts the point this way. 'It is often held that something's being goocl or believed good - its being rational, given the agent's values and beliefs - makes intelligible (explains) why a person seeksor desires it. If what I have said . . . is correct, then this is mistaken. . . U]ust as the person may well not seek or desire the (believed) good, so, were that person to do what would produce (believed) good, that fact might well not make intelligible why the person so acted. If I am known to be sunk deeply into despair or some other depression or to have long ago ceased caring about someone'swelfare, then citing the (believed) goodness of my act will not make intelligible my act which benefits that other person' 'Thus (Stocker,1979:.746). And again later: rationality in the sense of value maximization against the background of an agent's beliefs is not the form of all action, nor even all intelligible action. Nor is the correspondingly rational person the form of all people, nor even all intelligible people. Tryt.g to understand people as if they were such rational beings involvesinadequate moral psychologies' (Stocker, 1979:fcrotnote l2). Is this argument consistent with the earlier argument against Johnston?It is. Indeed, it helps to explain why a theory like the sel interest theory may be self-effacing (Parfit, 1984: 23-4). Suppose it is rational for me to do just one thing: promote my long-term sel interest. And suppose further that it is not rational for me to desire to promote my long-term selirrterest; that my long-term selfinterest would be best served by *y desiring to act for the sake of family and friends, write books, advance humanity and so on, without having any direct concern whatever for my own long-term selfinterest.What it is desirable that I do is, in this case,not what it is desirable that I desire that I do. But now suppose I come to believe the selinterest theory. I come to believe that it is desirable to promote my long-term selinterest and undesirable to desire to prornote my long-term selinterest. From the argument just given, having these beliefs makes it rational for me to desire to promote my long-term self-interest and to desire not to desire t() pr()m()tcnry

. ince the reasonI have lht'<lt'sirt'trr o w n l o n g - t e r m s e l i n t e r e s tS my long-term self-interest, something we know irrrlt'pcrrpromote that I rationally shouldn't desire, is that I bclievc th<' st'ldently interest theory, it is no surprise to learn that I rationally shorrlrln't believc the sel-interest theory. The theory is thus self-effacing.Arrrl il since I clesircrrot to desire to promote my long-term selintercst, is no surprisc that I am motivated to get rid of that belie. I arrr indee<l rnovt:<lto do what the theory tells me it is rational to do. Once wc havc accepted C2 we have no reason to deny that our belic.s irborrtorrr rcasonsplay a causalrole in the production of our desirt:s.Ilowt:ver, notc that the argument I have given for C2 does not dcpt'rrrl at irll on accepting the non-relntiueconception of normative rcasons I rnysell'vour. It goes through as well on the relatiue con<cptiorr lavorrredby Williams. If this is right, then it follows that evcn thost' who at'ccpt a Humean, relative conception of reasons shotrl<laclrrritthat <>urbeliefs about our reasonscan play a causal role in the production of our desires.This suggests that one of the isstrcs that is supposed standardly to divide Humean from antiHtrrnean theorisls of normative reasons does not divide them at all. Accortling to (lhristine Korsgaard for instance ( 1986), antiHurneans assert,and Humeans deny, that reason can produce a motive. But, as we have seen,what is at issueis not this, but rather whether the reasons that produce motives are ihemselves relative, as the Humeans suppose, or non-relative, as the anti-Humeans suPPOSe. NOTES TO CHAPTER MORAL 6 HOW TO SOLVE

THE

PROBT,EM

Note that I am merely sa)4ngthat the existence of these concepts shows the extent to which we agree in our evaluations, not that these evaluations are coTect. As I see it, many of these evaluations - those implicit in the use of our concepts of chastity, promiscuity, piety, frugality, selreliance and the like - are or may well be incorrect. For I do not think it at all obvious that we would desire people to act in these ways if we were fully rational. For this reason, I (perhaps mistakenly) take myself to be endorsing what Susan 'centralism', Hurley calls not 'non-centralism' (1985). James Rachels has an excellent discussionof this point in his account of the Biblical story in which God asksAbraham to sacrifice Isaac ( l97l ) . You will recall that, in that story, Abraham is supposed

214

l,trotes
to bc conflicted. He believed that it would be wrong for him to kill his orvn son, but he also believed in God's perfect goodness;thar Gocl rvould not ask him to do something wrong. Yet how could Abraharn rationally maintain his belief in God's perfect goodnessif hc genuinely believesboth that it would be wrong for him to kill Isaacand that this is what God is asking him to do? The point that the story driveshome is that there is no escapingforming your own n.ror:rlbelief! - in Abraham's case, beliefs about the rightness or wrongnessof killing Isaac,and beliet about the perfect or imperfect goodnessof Gocl. The only question is whether you will form these beliefs via an exerciseof your own rational capacities, in the light of the evidence , or whether you will form them without exercisingyour rational capacities, simply in responseto the directivesof an authority. I thus find myself in substantialagreement with Derek Parfit when he writes at the end of Reasons and Pnsons.'There could clearly be higher achievements in the struggle for a wholly just world-wide community. And there could be higher achievementsin all of the Arts and Sciences. But the progresscould be greatestin what is now the leastaclvanced of theseArts or Sciences. This, I have claimed, is Non-ReligiousEthics.Belief in God, or in many gods, prevented the free development of moral reasoning. Disbelief in God, openly admitted by a majoriry, is a very recent event, not yet completed. Becausethis event is so recent, Non-Religious Ethics is at a very early stage.We cannot yet predict whether, as in Mathematics,we will all reach :rgreement.Since we cannot know how Ethics will develop, it is not irrational to have high hopes.' (1984: 454) 'actually' in this For an explanation of why we should use definition, and how that helps to facillitate the entailment claim, see Davies and Humberstone, 1980. The point is also discussedin Smith 1986b;Wiggins, 1987;Wright, 1988; Lewis, 1989;Johnston, 1989. Strpposethat we discover that our desiresare out of line with our evaluative beliefs when we deliberate, and that we are not rational. Our evaluative beliefs do not cause us to have the desires that they can and should cause us to have. What happens then? The answer is: many different things might happen (Pettit and Smith, 1993a, 1993b; Kennett and Smith, forthcoming). We might employ some techr.riqueof selcontrol we have at our disposal and thereby get ourselves to do what we believe desirable none the less: that is, we might manage to be continent. Alternatively, though we have somc techniqtre of self-controlat our disposal,we might not exercisc it:

Notes

215

that is, we might strffer from weakness of will. Or altcrnativcly, we might have tro techrrique of self-control at otrr disposal, and.so go on to act ()n a desire that we have, but a desire that causcs rrs t<l act in a way that we do not value: that is, we might suflr i'orrr compulsion.

RcJ'erences

217

References

'The Legacyof Emotivism' in Graham Macdonald Ntharn,.|. E.J. 1986: on A. J. and Mmalityr 1li.sa't.s arrd Crispin Wright, eds, I'act, Science and Logic.Basil Blackwell. 275-88. Ayn's Language,7-ru.th G. E. M. 1957: Intention. Basil Blackwell. Arrsccrnrbe, Edition 1946. Aver, A.J. 1936:Langtage,Truth and I'ogic.Gollancz.Secc'rnd Wiltson, ed', reelill. reprinted in Gary I 954:'Frecdom and Necessity' Oxford University Press,1982. 15-23. 'Moral Realism' inJohn Casev,ed., Moralit\ rtrul Simon l97l: Blackbtrrrr, Methuen. l0l-24. Moral Reasoreing. 1984: Spreadingthe Wmd. Oxford University Press. 'Errors and the Phenomenology of Value' in Ted l{onderich' l98l1a: e<1., i\Iorality and Objectioif. Routledge and Kep;anPaul. l-22. 'supervenience Revisited' in Ian Hacking, ed., Exncispsin l9tl5b: of Casimirl-eu4.Cambridge Unive rsity Press. Ana\.sis:)ssays b1Students 'Morals ar.rd Modals'in Graham Macdonaid and Crispin Wright, l9tJ6: on A. J. Ayn's I'angtage,'l-ruth and Morality: Essays eds, lcl, Scjence and Logic.Basil Blackwell. I l9-42. 'FIow to Be an Ethical Antirealist' in Peter A. French, Theodore l{)fl7: ti. Llehling,Jr. and Howard K Wettstein, eds, 1987: Miduest Studies VolumtXII: fualism and Anti-fualisrn.University o'Notre in Phiktsophy I)anre Press.361-75. '(lircles, Finks,Smells,and Biconditionals' inJames Tomberlin, 1994: ed., Philosophical Perspectiues:Volumz VII, I'anguage and ['ogtc. Ridgeriew Press.259-81. 'Flight from Reality' in Rosalind Hursthotrse, (]avin fi>rthcoming: a ['-e'st.srhill l,awrence and Warren Quinn, eds, Virtuesand Rcasons, lor Philippa f'oot. 'How to be a Moral Realist' in GeoIiev Sa1'r't'Bovd, Richard l98fl: Mc(lorcl, ed., tsason Mmal fualism.Cornell Universitv Press.lfll228.

Brink, David O. 1984:'Moral Realismand the SkepticalAr-grrnrt'rrts i.orrr Disagreernent arrd Queerness', Australa.sian ol l)hihxltlrt. Journnl I I l-25. 'Exterrralist 1986: Moral Realism',SouthemJountal of Philosoplrl Srrppk'rnenr. 23-42. 1989: Moral Realism arul thcIoundations of Ethics.Cambridge Univt.rsilv Press. 'A Campbell,.fohnl99ll: S i n r p l eV i e w o f ( l o l o u r ' i n J o h n H a l d a n e a n d Crispin Wright, cds., Ileality, Representation and Projection. Oxord Universitl l)ress.257-68. Carnap. Rudol l96i): 'Replies and SystematicExpositions' in P" A. Schilpp, ed.,'l'he I'hilosophtof Rudolf Carnap.Open Court. 'Belic, C o l l i n s , . f o h nl 9 t 3 t t : D e s i r ea n d R e v i s i o n ' , Mind.333-42. Dancv,Jonathan l9-9ll:aralRcasons. Basil Blackwell. Daniels, Norman 1979: 'Wicle Relective Equilibrium and Thec-rry Acceptarrcein L,thi<'s' .loumal of Philo.suphy.256-82. (lornell University Press. Darwall, Stephcn l\|81\:Inpartial llcason. Gibbarct,Allan and Railton, Peter 1992: 'Toward lin de siJr'12 Ethics: Strme Tre nds', Phihtsophical Ratieu. I l5-89. Dar,iclson, Donalcl 1963: 'i\ctions, Reasonsand Causes' reprintecl in D a v i d s o n1 9 8 0 . 3 - 2 0 . 'How 1970: is Weaknessof the \4lill Possible?' reprinted in Dar.idsor.r 1980.21-42. 'Intending' 1978: reprinted in Daviclsor-r 1980. 83-102. jltsa1s 1980: on Actionsand Events.Oxford Urriversin' Press. Davies,Martin ancl Hrrmberstone,Llovd 1980: 'Two Notions of Necessity', PhilosophicalStuclies.l-30. Divers, John and Miller, Alex 1994: 'Why Expressivists about Valut' Should Not Love N{ininralisrn about Truth' , Analysis. Dreier,.farnes1990: 'Internalism and Speaker Relativism',Ethics.tt-26. Dworkin, Ronalcl 1977 :'l-ahi ng Right s Seriously. Drrckworth. Evans, (lareth 1980: 'Things Without the Nind - A (lommer)tary or) Chapter J'wo of Strawson'sIndiuiduaLs' in Zak van Srraaten, r'r1., Philosophical Subjecls:Essay.s Presented to P. F-.Strauson.(llart'n<lorr Press.76-1 16. Fafk, W. D. l94U: "'Ought" and Motivation', Proceedings of the tlri.strtttlirtrt I I l-38. Society. F o o t , P h i l i p p a 1 9 5 8 :' M o r a l A r g u m e n t s 'r e p r i n t e r li n F o o t l 1 ) 7 f t1 , .) t ; - l o 1 ) 'Morality 1972: asa System < - r ' H y p o t h e t i cIa nl r p e r a t i v c sr' ' t ' p l i r r l c r i lrr F o o t 1 9 7 8 .1 5 7 - 7 3 . l ( ) 7 7 : ' A p p r o v a l a n t l f ) i s a p p r o v a lr ' t ' p r i n l t ' r l i r r l . ' o o l l 1 ) 7 1 {l H l ) ' l l ) 7

218

Rzferences
and Pettit, Philip 1988:

Ilcferrnces
'Functionalism and Broittl (lontt'nt',

2I9
Minrl.

1978: Virtuesand Vius. University of California Press. Frankena, William 1958: 'Obligation and Motivation in Recenr Moral Philosophy' in A. I. Melden, ed., tsa1son Moral Philosophy. University of Washington Press. Frankftrrt, Harry l97l 'Freedom of the \4lill and the Concept of a Person' reprinted in Gary Watson, ed., F-ree WiIl. Oxford University Press, 1982.81-95 'Reason Gauthier, David 1975: and Mirximization', CanadianJoum,al of Phil-osophy. 4ll-34. 1986: Morak by Agreemcnt (llarendon Press. Gibbard, Allan 1990: WiseChoices, Apt Feelings. Clarendon Press. Hale, Bob 1986: 'The Compleat Projectivist',Philosophical Quartnly.6584. 1993: 'Can There Be a l,ogic of Attitudes?' in John Haldane and Crispin Wright, eds., fuality, R4tresentationand Projection. Oxford University Press.337-63. Hare, R. M. 1952: The Languageof Morak. Oxford University Press. l98l: Moral Thinking. Oxford University Press. Hart, H. L. A. l96l: The Concepl of Law. Clarendon Press. Harman, Gilbert 1973: ThoughLPrinceton University Press. 1975: 'Moral Relativism Defended', Philosophical Rniau 3-22. 'fhe 1977: Nature of Mmakty Odord University Press. 1985: 'Is There a Single True Morality?' in David Copp and David Zimmerman, eds, 1985: Mmalitl, Reasonand Truth. Rowman and Allanheld. 1986: 'Moral Explanations of Natural Facts - Can Moral Claims Be Tested Against Moral Reality?', SouthernJournal of Philosopy Supplement. 69-78. Horwich, Paul 1992: 'Gibbard's Theory of Norms', Philosophland Public Alfairs. 67-78. 1994: 'The Essenceof Expressivism', Analysis. H u m b e r s t o n e ,I . L . 1 9 9 2 : ' D i r e c t i o no f F i t ' , M i n d . 5 9 - 8 3 . Hume, David 1888: A Treatise of Human Nature. Clarendon Press, 1968. Hurley, S. L. 1985:'Objectity and Disagreement'in Ted Honderich, ed., Morality ttnd Objectiaif. Routleclge and Kegan Paul. 'Critical Notice of Susan Hurley's Natural Rzosons: Jackson, Frank 1992: Pnsonali\ and Polity', AustrahtsianJournal of Phi losop hy. forthcoming: 'Armchair Metaphysics'in Michaelis Michael and.fohn O'Leary Hawthorne, eds, Phihsophyin Mind. Kluwer Press. and Pargetter, Robert 1987: 'An Objectivist'sGuide to Subjectivism aborrt (lolorn' in Ratiau Intc.ntationabde Philosobhie. 127-41.

381-,100. 'Mininralisrtt :ttttl'l'r'trtlrOppy, Graham and Smith, Michael 1994: Aptness', Mind. 'DispositionalTheories of Value', Itroreul.ingt of lhe Johnston, Mark 1989: Ari stote lian .Socie{y Supplemen tary Volume. | 39-7 4. of theMetaphysics of Mmak [.ilrrary o Kant, Imrnanuel 1786: F-oundations I-iberal Arts, 1959. Kennett,.feanette forthcoming:'Mixed Motives', Australasian Jounuil.ol Philosophy. 'Philosophy and Commonsenst': and Smith, Michael forthcoming: (lase The of Weakness of Will' in Michaelis Michael and John O'Leary Hawthornc, eds, Philosopfu in Mind. Kluwer Press. 'Skcpticism Korsgaard,Christinc l9tl6: about Practical F.eason', Jounml of Phitosophy. 5-25. Basil Blackwell. Ikipke, Saul l9tl0: Naming and Neccssity. and Culture.Clarendon Press. Kymlicka, Will 1989: Liboalism, Oommunity Political Philosophy. Oxford University Press. 1990: Contemporary 'How tr>Define Theoretical Terms' l.ewis, Davirl 1970: ,Journal of Philosu phy. a27-46. 1972:'Psychophysical and Theoretical ldentifications', Austrakrsian Joumal of Philosophy.249-58. ', 'Desire 1988: as Belief Mind.323-32. 'Dispositional 1989: Theories of Value', Proceed,ings of the Aristotelian Soaely SupplementaryVolume. ll3-37 . [,oar, Brian 7981:.Mind and Meaning. Cambridge University Press. 'Are McDowell,Johrr 1978: Moral Requirements Hypothetical Imperatives?', Proceedings of the AristotzlianSociety Supplementary Volume. I 3-29. 'fhe 'Virtue 1979: Monisr.331-50. and Reason'. 'Non-Cognitivism l98l: and Rule-Following' in Steven Holtzman and Christopher Leich, eds, Wittgenstein: To Follow a RuLe.Routledge. l4l-62. 'Values 1985: and SecondaryQualities' in Honderich 1985. ll0-29. McGinn, Colin 1982: The Character of Mind. Oxford University Press. 1983: I'he SubjectiveViatt. Oxford University Press. Mackie,J. L. 1976: ProbbmslromLockc.Clarendon Press. 1977: Ethics: Inaenting Right and Wrong.Penguin. McNaughton, David 1988l Moral Vision.Basil Blackwell. Monro, D. H. 1967: Empiricismail Ethics. Cambridge University Press. Morrre, G. E. 1903: Principia Ethica.Carnbridge University Press.

220

Rzferences

fufnences

:'l I

'l'he Nagel, Thomas 1970: Possihility of Altruism. princeton Ur-riversirv' Press. 1986: I'he View from Nouhen. Oxfbrd University press. Nozick, Robert l98l: Philo.sophical Explanations. HanttardUniversity press. Parfit, Derek 1984: Rtasorts and Per.sons. Oxford University press. Peacocke,christopher 1979 Ilolistic Expkrnatian. oxford University press. press. 1985: Slense and Content. Oxford University Pettit, Philip 1987: 'Hurnca.s, A.ti-tlumeans and Motivation,, Mirul. 530-3. 1993: T'heCommon Mind. Oxford Llniversity press. arrd Smith, Michael 1990: 'Backgrounding Desire', phitosophicat Rzuiatt.565-92. and Smith, Michael lg93a: 'Practical Unreason', Mind. b3-79. and Smith, Michael 1993b: 'Brandt <_rn SelControl' in Brad Hooker, ed,.,Rntionality,Rulzsand Utilie. Westview press. 33-b0. and Smith, Michael forthcoming: 'Parfit's p' inJonathan Dancy, ed., fuading Parft. Basrl Blackwell. Platts, Mark 1979: Waysof Meaning. Routledge and Kegan paul. l98l: 'Moral Reality and the End of Desire'in Mark plat:rl.,ed., Rzference,Tnrth and Rtalie. Routledge and Kegan paul. 69-82. Price, Huw 1988: -crs and theI'unction of Truth. Basil Blackwell. ', 1989: 'Defending Desire-as-Belief Mind. ll9-27. Prior, Elizabeth W., Pargetter,Robert andJackson, Frank 1982: ,Three Theses About Dispositions' , Amzrican Phitosophicat euarterty. Zbl-7. Prrtnam, Hilary l98l: fuason, 'fruth and History. Cambridge University Press. Rachels,Jamesl97l: 'God and Human Attitudes', reprinted in Tom L. Beauchamp,Joel Feinberg andJames M. Smith, eds,phitasophy and the Human Condition.Prentice Hatl, 198g. b0g-16. Railton, Peter 1986: 'Moral Realism', Phitosophicat Reviat. 163-207. 1993a: 'what the Noncognitivist Helps us to See the Naruralist Must Help Us to Explain' in John Haldane and Crispin Wright, eds., fuality, Representation and hojection. Oxford University press. 2?9300. 1993b:'Reply to David Wiggins' inJohn Haldane and Crispin Wright, eds., fuality, R$resentation and Projection. Oxford University press. 3l 5-28. Ramsey,Frank P. l93l: 'Theories' in his TheFoundationsof Mathcmatics. Routledge and Kegan Paul. Rawls,J<rhn l95l: 'Outline of a Decision Procedure for Ethics,, phik> sobhical Rniau 177-97.

'Moral Theory arrrl l'.x1>l;rrrrtlo (i ltttp,' Sayr-c-McCorc le , offrey 1988: ,ttt '\1,'r,tl l'.ttttyt c<1., in Sayre-McC<lrcl, reprinted Geoffrey tence' (lornell Press. 256-81. University Rzalism. 'Contractualism and Utilitariatristrt'itt,'\ttt;ttlr:t S c a n l o n , ' f h o m n s1 9 8 2 : (.;rtrt Utilitarianism and Bt',^tttttl. eds, Berrrard Williams, Serr arrd Press. 103-28. Ur.riversity briclge 'Pro-Attitudesand Directions of fit', l\lind.277-81 . Schueler,C).F. 1991: of Ethics.Hackett Ptrblishins (,()rr)Sidgn'ick, Herrry,7907: T'heMethod,s l 9 t 3 l . pany. 'The Triviality of the Debate over "Is-Orrght"anrl thc Singer, Peter 1973: Defi rri tion of "Moral"', AmericanPhilosophical Quartn\. 5l 6. 'On Some Criticisms of a PhysicalistTheory o' (l j. 1975: Smart, -1. (lolour' reprinted in his Essals Metaphysical and Moral. Basil B l a c k w e l l .1 9 U 7 . 'Peacocke on Red and Red-', Synthese.559-76. Smith, MichaeI l9iJ6a: 'should We Believein Emotivism?'in Graham Macdonalc{ancl 1986b: Crispirr Wrieht, eds, l-act, Scirnu and Morality; Essalson A. J. ,lvn's Basil Blackwell. 289-310. Language,Truth and Logzc. ' T h e H u m e a n T h e o r y o f M o t i v a t i o n ' ,M i n d . 3 6 - 6 1 . 1987: 'On 1988a: Humeans, Anti-Humeans and Motivatiolr: A Replv to Pettit', Mind.589-95. 'Reason and Desire', Proceedings Society.243-56. of theAristotelian 1988b: 'Dispositional of the Ari.stotelian Theories of Value', Proceedings 1989: SocieilSupplementary Volume. 89-l 11. 'Realism' in Peter Singer, ed., A Companion to Ethics. Basil l99l: Blackwell. 399-410. 'Valuing: Desiring or Believing?'in David Charlesand Kathleen 1992: Oxford Universitl' and Rcalism. Lennon, ed,s, fuduction, Expl.anation, Press.323-60. 'Objectivity and Moral Realism: On the Significance of thc 1993a: Phenomenologyof Moral Experience' inJohn Haldane and Crispin Oxford University Wright, eds, Rcality,R$resmtation and Projection. Press.235-6. i993b:'Colour, Transparency,Mind-Independence' inJohn Haldane and Proiectioz.Oxford arrd Crispin Wright, eds, Rzality, RePresentation University Press.269-78. 'Why about Value Should Love Minimalisnr about 1994a: Expressivists Truth', Ana\sis. l-12. 2l-61994b:'Minirnalism, Truth-Aptitude and Belief', Analysis. 'Desiring the Bad: An Essay in Moral Psvchologv', Stocker,Michael 1979: 738-53. loumal of Philosophy.

222

Rcfnmces

Stroud, tsarry lg77: Humc. Routledge and Kegan paul. sturgeon, Nicholas l9g5: 'Morar Exlranationrii., Durrid copp and David 'Rzason Zirnmerman, eds, Mmality, and Truth. Rowman and Allanheld. 49-79. 1986: 'what Difference Does It Make whether Moral Realism Is True?' , Southern Suppleme n t. I I 5_41. /oumal of phitosophy Tawil, Nathan lgSi: Reference ona' nuntlonality. ph.D. Dissertation, Princeton Universiw. Velleman, J. David 1992: ,The Guise of the G.ood,, Nos.3-26. Wallace, Jay 1983: Motiuation and Moral Reality. Bachelor of Philosophy thesis. Oxford University. 'How ro Argue about practical Reason,in Mind.267_97. _.,1990: Watson, Cary lgTb:.Free Agency' reprinted in Gary Watson, ed., _ree ll/ill. Oxford University press. l9ti2. 96_110. 1987: 'Responsibilityand the Limits of Evil' in Ferdinand schoeman, ed_, Rcsponsiklity, Character and thc Emotions: f,t* nrloy, i, tW*ot Psy cholag. Cambridge Un iversity press. 256_g6. wiggins, David lg8?:-'A_sensibleSubjectivism'in his Need,s, values,Tntth. Basil Blackwell. 185-214 'Cognitivism, 1993a: Naturalism, an. Normativiry: A Reply to perer Railton' inJohn Haldan^ea1d Crispin Wright, ,a"., n"oirg,-n"prrr* and Projection. Oxford University press. 301_314. - ?!i"" 'A 1993b: Neglected position?' inJohn irardane and crispin wright, Reresmtation and projection. Oxford Univrsiry press. :1"^., P"tlE, 329_36. williams, Bernard lg26: 'persons,characrer and Morarity' reprinted in W i l l i a m s 1 9 8 1 .l - 1 9 . 1980: 'Internal and External Reasons' reprinted in williams l9gl.

Index

Abraharn,2l3n2 absolutisrn, 194 action explanation, 9, l0l-4, 139-40, 208n6 causal vs. teleological cr>nception of, l0l-4, ll3-1.1, l3l. 139-40 i r r t c n ( i o r r a l r s . c l e l i b c r a r i v ep ( r \ p ( , ( t r \ ( , s on. l3l-3, 136-7, 139-40, 179-U0, 2l2nI addiction, 134. I.15, 143-4. I.16, l4u, 15.1-5 advice, l5l agcnt relative vs. agent neural reasons, I 68-70 A l t h a n r , . l .E . J . , l l 8 , 2 l l n l { ) amoralism. 66-71 Anscombe,G. E. M., lll, ll7 analysis, of colour concepts, 29, 3l-2, 49-53 o normative reasons, l5l-77 < r fr i g h t n e s s , 4 5 - 6 , 5 4 - 8 , 1 8 2 - 4 Jff clso ( onceptual analysis an ti-H urnean theory o modvation, l2-13, 61, 92-3, I I8-25. t)o o normative reasons, 130-81, 213n3 apprual, litl-91 ,{ver,4..1..4. 12, l6-119.63,133-5, 154 bawling baby, 134, 140. 148, I55 belief and desire are distinct existences, 7-8, I l 9 - 2 1 , 1 3 7 ,2 1 0 n l 0 functional role oi 7, 113-15, ll6-25, I 36-7, I 77-80, 208n6, 20tin?, 209n8. 210n9,2l0nl0 lrsire, I ll|-25, 179, 2l0n 10, 2l ln I I and internalisrn, 61, 120 Mcl)owell on, l2l-5 Nagel on, 2l lnl2 Blackburn, Simon, 3, 4, 12. 16, 21,23-4, 56,61,63,86 Brink. Dar.id O., 3, 4, 12, 28, 29, 35, 60, 63, 6(i71, !)1.2{)6nl

Boyd, Richard, ll, 2.<] ()ampbell,.lohrr,29 cannibals,3.3--1 ( i a r n a p . R u d o l f . . 1 4 ,4 9 ( a l e g o r i (a l i r r t p e r r t i r t . s \. / ' i t n p ( . r . r t i \ c \ ( tr( unrstances, rer reasons artd circtrmstanccs cognitivisrn,9 O o l l l n s . . | o h n ,2 l l n l I cokrur conrepts, 2tt-32. 36, 3lJ. .1.1, 4U-5.+. 68-70, 190-1, 201n6.205n9 conceptrral ;rnalvsis explanadon of, 2!)-ll2 netw()rk st1'le , .1.1-5(i, I6'.1-.1 paradox o. 37-!) -irl). reductive vs. non-rcd uc tivc. 3ir-(i, -1-1 l6l-.{, 185-(i role o platitrrdes irr, 21)-l]2 rolc of paradignrs in. I-rl-2, ir5, l(i3-.1 s u r n n r : u ) s t ' l e , 3 l - 2 . 3 t , l ,I 5 : 1 . l ( i l - . 1 , l t i i r convergence in desires, lti.l-77, 198-g irr moral judgerncnt. {). tt7, I87-1},2l{n:l criminal, | 1)4-6 D a r r c r ,J , onatharr, 3, l3 Daniels. Nornran, I ir9 Daruall, Stephcn, +, 2t{, 37, bo D a v i d s o n , D < ; n a l d , { i ) .9 2 . l O 2 - 3 , I 1 7 . l 37-4 I Davies,M:rrtirr, 2l{n-1 clclibrration e f f e c d v e n e s so , 1 3 2 , I 7 7 - l t 0 imaginatiorr and, 157-9, I(il nreanknds reasoning atrd, 157 s y s t e m a t i c . j u s t i f ia <t i o n o f d c s i r e s a r r r l .

t0r_13.

l98l: Mmal Luck. Cambndge University press. 1985: Ethics and the I'imits.oJrniutopny. Harvard university press. wilson, George l98b: 'Davidr., o., Intetio.rar Action, in Ernest Lepore and Brian Mclaughlin, eds., Actions and Events: perspectives on the of DonaH Dauidson.Basil Btackwell. 29_43. -._ .lh-ibtophy Wolf, Susan 1982: .Moral Saints',./ozmal of philnsopfu. al9_Zg. Woods, Michael 1972: .Reasorr, n.Uo., and Desire,, proceed,ings oJthe Society Supplementary Volume. lgg_201. ... . ldtt!*lian wrrght, ()rispin 1988: ,Moral Values, projection and Secondary eual_ i.tiei, proczedingsof thz Aristotzlia, Siaery Supplementary ottr_.. l-26. 1992 Tnilh and Objectiviry. Harvard Universiw press.

1 5 9 - 6 rr , 6 6 ,r 7 5 - 7
rrnderivecl desirt's prodrrced artrl destroyed b,v,158-til, 171), 2llhi.t rre clra a<-tiorrexplanation d e p r e s s i o n .1 2 0 - 1 . 1 2 4 . l 3 i r . l : l l { - 1 t . I 1 2 . 1 5 4 - r ,2 l 2 n I o e s t r . t D l l t t \ l.' ) l - / /