Royal Institute of PhilosophyModern Moral PhilosophyAuthor(s):
G. E. M. AnscombeSource: Philosophy, Vol. 33, No. 124 (Jan., 1958),
pp. 1-19Published by: Cambridge University Press on behalf of Royal
Institute of PhilosophyStable URL:
http://www.jstor.org/stable/3749051 .Accessed: 05/02/2014 22:09Your
use of the JSTOR archive indicates your acceptance of the Terms
& Conditions of Use, available at
.http://www.jstor.org/page/info/about/policies/terms.jsp .JSTOR is
a not-for-profit service that helps scholars, researchers, and
students discover, use, and build upon a wide range ofcontent in a
trusted digital archive. We use information technology and tools to
increase productivity and facilitate new formsof scholarship. For
more information about JSTOR, please contact [email protected].
.Cambridge University Press and Royal Institute of Philosophy are
collaborating with JSTOR to digitize,preserve and extend access to
Philosophy.http://www.jstor.org This content downloaded from
168.176.5.118 on Wed, 5 Feb 2014 22:09:18 PMAll use subject to
JSTOR Terms and ConditionsPHILOSOPHY THEJOURNALOFTHEROYALINSTITUTE
OFPHILOSOPHY VOL.XXXIII.No.I24JANUARY1958 MODERNMORALPHILOSOPHY" G.
E.M. ANSCOMBE I WILL beginby statingthreetheseswhichI presentin
thispaper. The firstis that it is not profitableforus at presentto
do moral philosophy;that shouldbe laid aside at any rate untilwe
have an adequate philosophyof psychology,in whichwe are
conspicuously lacking.The secondis that the conceptsof
obligation,and duty- moralobligationand moralduty, that is to
say-andof what is morallyrightand wrong,and ofthemoralsense of
"ought," ought to be jettisonedif this is
psychologicallypossible;because they are survivals,or
derivativesfromsurvivals,froman earlierconceptionof ethics which no
longergenerallysurvives,and are only harmful withoutit. My
thirdthesisis that the differencesbetweenthe well-
knownEnglishwriterson moralphilosophyfromSidgwickto the presentday
are of littleimportance. Anyonewho has read Aristotle'sEthicsand
has also read modern moral philosophymust have been struck by the
great contrasts betweenthem.The
conceptswhichareprominentamongthemoderns seem to be lacking,or at
any rate buriedor farin the background, in
Aristotle.Mostnoticeably,theterm"moral"itself,whichwe have by
directinheritancefromAristotle,just doesn'tseem to fit,in its
modernsense,intoan accountofAristotelianethics.Aristotledistin-
guishesvirtuesas moraland intellectual.Have someofwhathe calls
"intellectual"virtueswhatweshouldcall a "moral"aspect?It would
seemso; thecriterionis presumablythata failurein an "intellectual"
virtue-like that of having good judgmentin calculatinghow to
bringaboutsomethinguseful,say in municipalgovernment-maybe
blameworthy.But-it may reasonablybe asked-cannot any failure be
made a matterof blame or reproach?Any derogatorycriticism, say of
theworkmanship ofa productor the designofa machine,can be
calledblameor reproach. So we wantto put in theword"morally" 1 This
paper was originallyread to the Voltaire Society in Oxford. I This
content downloaded from 168.176.5.118 on Wed, 5 Feb 2014 22:09:18
PMAll use subject to JSTOR Terms and ConditionsPHILOSOPHY again:
sometimessucha failuremaybe morallyblameworthy,some- timesnot.Now
has Aristotlegotthisidea ofmoralblame,as opposed to any other?If he
has, whyisn't it morecentral?Thereare some mistakes,he
says,whichare causes,notofinvoluntarinessin actions, but
ofscoundrelism,and forwhicha manis blamed.Does thismean that
thereis amoralobligationnot to make certainintellectual
mistakes?Why doesn't he discuss obligationin general,and this
obligationin particular?If someoneprofessesto be expounding
Aristotleand talksin a modernfashionabout"moral"such-and-such, he
mustbe veryimperceptiveifhe doesnotconstantlyfeellikesome- one
whosejaws have somehowgot out of alignment:the teethdon't
cometogetherin a properbite. We cannot,then,lookto
Aristotleforanyelucidationofthemodern way of talkingabout
"moral"goodness,obligation,etc. And all the best-knownwriterson
ethicsin moderntimes,fromButlerto Mill, appearto me to have
faultsas thinkerson thesubjectwhichmakeit impossibleto hope forany
directlighton it fromthem.I will state these objectionswith the
brevitywhich their charactermakes possible. Butler exalts
conscience,but appears ignorantthat aman's consciencemay tell himto
do the vilestthings. Humedefines"truth"in sucha wayas to
excludeethicaljudgments fromit, and professesthathe has
provedthattheyare so excluded. He also implicitlydefines"passion"
in such a way that aimingat anythingis havinga passion.His
objectionto passingfrom"is" to "ought"wouldapplyequallyto
passingfrom"is" to "owes" or from "is" to "needs."
(However,becauseofthehistoricalsituation,he has a pointhere,whichI
shall returnto.) Kant introducesthe idea of
"legislatingforoneself,"whichis as absurd as if in these days, when
majorityvotes commandgreat respect,one wereto call each
reflectivedecisiona man made a vote resultingin a majority,which as
a matterof proportionis over- whelming,forit is always i-o.The
conceptof legislationrequires superiorpowerin thelegislator.His
ownrigoristicconvictionson the subjectof lyingwereso intensethatit
neveroccurredto himthata lie couldbe relevantlydescribedas
anythingbut just a lie (e.g. as "a lie in
such-and-suchcircumstances").His rule about universalizable
maximsis uselesswithoutstipulationsas to what shall count as a
relevantdescriptionofan actionwitha viewto constructinga maxim
about it. Bentham and Mill do not notice the difficultyof the
concept "Pleasure." Theyare oftensaid to have
gonewrongthroughcommit- tingthe "naturalisticfallacy";but
thischargedoes not impressme, because I do not findaccountsof it
coherent.But the otherpoint- about pleasure-seemsto me a
fatalobjectionfromthe veryoutset. This content downloaded from
168.176.5.118 on Wed, 5 Feb 2014 22:09:18 PMAll use subject to
JSTOR Terms and ConditionsMODERNMORALPHILOSOPHY The
ancientsfoundthisconceptprettybaffling.It reducedAristotle to
sheerbabble about "the bloomon the cheekof youth"because, forgood
reasons,he wantedto make it out both identicalwithand differentfrom
the pleasurable activity. Generationsof modern
philosophersfoundthisconceptquiteunperplexing,and it reappeared in
theliteratureas a problematiconeonlya yearortwoagowhenRyle
wroteabout it. The reasonis simple:sinceLocke,pleasurewas taken to
be somesortof internalimpression.But it was superficial,if that was
therightaccountofit,to makeit thepointofactions.One might
adaptsomethingWittgensteinsaidabout "meaning" andsay
"Pleasurecannotbe an internalimpression,forno internalimpression
couldhave the consequencesof pleasure." Millalso, like Kant,
failsto realizethenecessityforstipulationas to
relevantdescriptions,if his theoryis to have content.It did not
occur to him that acts of murderand theftcould be otherwise
described.He holdsthatwherea proposedactionis ofsucha kindas to
fall undersome one principleestablishedon groundsof utility, one
mustgo by that;whereit fallsundernoneor several,theseveral
suggestingcontraryviewsoftheaction,thethingto do is to calculate
particularconsequences.But prettywell any action can be so des-
cribedas to makeit fallundera varietyof principlesof utility(as I
shall say forshort)if it fallsunderany. I will now returnto Hume.
The featuresof Hume's philosophy whichI have
mentioned,likemanyotherfeaturesofit,wouldincline me to thinkthat
Hume was amere-brilliant-sophist;and his proceduresare
certainlysophistical.But I am forced,not to reverse, but to add to,
this judgmentby a peculiarityof Hume's philoso-
phizing:namelythatalthoughhe reacheshisconclusions-withwhich he is
in love-by sophisticalmethods,his considerationsconstantly open up
verydeep and importantproblems.It is oftenthe case that in theact
ofexhibitingthesophistryonefindsoneselfnoticingmatters which
deservealot of exploring:the obvious stands in need of
investigationas a resultof the pointsthat Hume pretendsto have
made. In this,he is unlike,say, Butler.It was alreadywell known
thatconsciencecoulddictatevile actions;forButlerto have written
disregardingthisdoes not open up any new topicsforus. But with Hume
it is otherwise:hence he is avery profoundand great philosopher,in
spiteofhis sophistry.For example: SupposethatI say to
mygrocer"Truthconsistsin eitherrelations of ideas, as that 2os. =
fi, or mattersof fact,as that Iordered potatoes,you
suppliedthem,and you sent me a bill. So it doesn't applyto sucha
propositionas thatI oweyou such-and-sucha sum." Now if one
makesthiscomparison,it comesto lightthatthe rela- tion of the
factsmentionedto the description"X owes Y so much money"is an
interestingone,whichI will call that of being"brute 3 This content
downloaded from 168.176.5.118 on Wed, 5 Feb 2014 22:09:18 PMAll use
subject to JSTOR Terms and ConditionsPHILOSOPHY relativeto" that
description.Further,the "brute" factsmentioned herethemselveshave
descriptionsrelativelyto whichotherfactsare "brute"-as, e.g., he
had potatoescartedto myhouseand theywere leftthereare
brutefactsrelativeto "he suppliedme withpotatoes." And the fact
Xowes Ymoneyis in turn "brute" relative to other descriptions-e.g.
"Xissolvent." Now the relation of "'relativebruteness"is a
complicatedone. To mentiona fewpoints: ifxyz is a set of
factsbruterelativeto a descriptionA, thenxyz is a set out of a
rangesome set amongwhichholdsif A holds; but the holdingof some set
among these does not necessarilyentail A,
becauseexceptionalcircumstancescan alwaysmakea difference;and
whatare exceptionalcircumstancesrelativelyto A can generallyonly be
explainedby givinga fewdiverseexamples,and no theoretically
adequateprovisioncan be made forexceptionalcircumstances,since a
furtherspecialcontextcan theoreticallyalwaysbe imaginedthat would
reinterpretany special context.Further,thoughin normal
circumstances,xyz wouldbe a justificationforA, that is not to say
thatA just comesto the same as "xyz"; and also thereis apt to be an
institutionalcontextwhichgivesits pointto thedescriptionA, of
whichinstitutionA is of coursenot itselfa description.(E.g. the
statementthat I give someonea shillingis not a descriptionof the
institutionofmoneyor ofthecurrencyofthiscountry.)Thus,though it
wouldbe ludicrousto pretendthat therecan be no such thingas a
transitionfrom,e.g.,"is" to "owes,"thecharacterofthetransition is
in factratherinterestingand comesto lightas a resultofreflecting on
Hume's arguments.' That I owe thegrocersuch-and-sucha sumwouldbe
one of a set of factswhichwouldbe "brute"in relationto
thedescription"I am a bilker.""Bilking"is of coursea speciesof
"dishonesty"or "in- justice." (Naturallythe
considerationwillnothave any effecton my actionsunlessI wantto
commitor avoid acts of injustice.) So far,in spiteof
theirstrongassociations,I conceive"bilking," "injustice"and
"dishonesty"in a merely"factual"way. That I can do
thisfor"bilking"is obviousenough;"justice"I have no idea how to
define,except that its sphereis that of actionswhichrelateto
someoneelse, but "injustice,"its defect,can for the momentbe
offeredas a genericname coveringvariousspecies.E.g.: "bilking,"
"theft" (whichis relativeto whateverpropertyinstitutionsexist),
"slander,""adultery,""punishmentof the innocent." In
present-dayphilosophyan explanationis requiredhow an unjustmanis a
bad man,or an unjustactiona bad one; to givesuch an
explanationbelongsto ethics;butit cannotevenbe begununtilwe are
equippedwitha soundphilosophyof psychology.For the proof I The
above two paragraphsare an abstract of a paper "On Brute Facts"
forthcomingin Analysis. 4 This content downloaded from
168.176.5.118 on Wed, 5 Feb 2014 22:09:18 PMAll use subject to
JSTOR Terms and ConditionsMODERNMORALPHILOSOPHY thatan unjustman is
a bad man wouldrequirea positiveaccountof justice as a
"virtue."This part of the subject-matterof ethicsis,
however,completelyclosed to us untilwe have an accountof what
typeof characteristica virtueis-aproblem,not of ethics,but of
conceptualanalysis-and how it relatesto the actionsin whichit is
instanced:a matterwhichI thinkAristotledid notsuccedin really
makingclear.For thiswe certainlyneed an accountat least ofwhat a
humanactionis at all, and howits descriptionas "doingsuch-and-
such" is affectedby its motiveand by the intentionor intentionsin
it; and forthisan accountof such conceptsis required. The
terms"should"or "ought"or "needs" relateto goodand bad:
e.g.machineryneedsoil,orshouldoroughtto be oiled,in thatrunning
withoutoil is bad forit, or it runsbadly withoutoil. Accordingto
this conception,of course,"should" and "ought" are not used in a
special"moral"sensewhenone says thata man shouldnotbilk. (In
Aristotle'ssenseof the term"moral" (q'COucK'S), they are being used
in connectionwith a moralsubject-matter:namelythat of human
passionsand (non-technical)actions.)But theyhave nowacquireda
specialso-called"moral"sense-i.e. a sensein whichtheyimplysome
absoluteverdict(likeone ofguilty/not guiltyon a man) on whatis
describedin the "ought"sentencesused in certaintypesofcontext: not
merelythe contextsthatAristotlewouldcall "moral"-passions and
actions-but also some of the contextsthat he would call
"intellectual." The ordinary(and quite indispensable)terms"should,"
"needs," "ought," "must"-acquired this special sense by
beingequated in therelevantcontextswith"is obliged,"or "is
bound,"or "is required to," in the sense in whichone can be
obligedor boundby law, or somethingcan be requiredby law. How did
this come about? The answeris in history:between Aristotleand us
came Christianity,withits law conceptionof ethics. For
Christianityderivedits ethicalnotionsfromthe Torah. (One mightbe
inclinedto thinkthat a law conceptionofethicscouldarise only
amongpeople who accepted an allegedlydivinepositivelaw; that thisis
not so is shownby the exampleof the Stoics,who also
thoughtthatwhateverwas involvedin conformityto humanvirtues was
requiredby divinelaw.) In consequenceof the dominanceof
Christianityformany cen- turies,the conceptsof
beingbound,permitted,or excusedbecame deeply embeddedin our
language and thought.The Greek word taicapcL'ELv,the aptest to be
turnedto thatuse, acquiredthesense "sin," fromhavingmeant
"mistake,""missingthe mark,""going wrong." TheLatinpeccatumwhich
roughly correspondedto ayzapnra was evenapterforthesense "sin,"
because it was already associatedwith "culpa"-"guilt"
-ajuridicalnotion.The blanket 5 This content downloaded from
168.176.5.118 on Wed, 5 Feb 2014 22:09:18 PMAll use subject to
JSTOR Terms and ConditionsPHILOSOPHY
term"illicit,""unlawful,"meaningmuchthe same as our blanket
term"wrong,"explainsitself.It is interestingthatAristotledid not
have such a blanketterm.He has blankettermsforwickedness-
"villain," "scoundrel";but of course a man is not a villain or a
scoundrelby theperformanceofone bad action,or a fewbad actions. And
he has termslike "disgraceful,""impious";and specificterms
signifyingdefectof the relevantvirtue,like "unjust"; but no term
correspondingto "illicit."The extensionof thisterm(i.e. the range
of its application)could be indicatedin his terminologyonlyby a
quitelengthysentence:thatis "illicit"which,whetherit is a thought
or a consented-topassionor an actionor an omissionin thoughtor
action,is somethingcontraryto one of the virtuesthe lack ofwhich
showsa man to be bad qua man. That formulationwould yield a
conceptco-extensivewiththe concept"illicit." To have a law
conceptionof ethicsis to hold thatwhat is needed
forconformitywiththevirtuesfailurein whichis the markofbeing bad
qua man (and not merely,say, qua craftsmanor logician)-that whatis
neededforthis,is requiredby divinelaw. Naturallyit is not
possibleto have sucha conceptionunlessyoubelievein God as a law-
giver;like Jews,Stoics,and Christians.But if such a conceptionis
dominantformanycenturies,and thenis givenup, it is a natural
resultthat the conceptsof "obligation,"of beingboundor required as
by a law, shouldremainthoughtheyhad losttheirroot;and ifthe
word"ought"has becomeinvestedin certaincontextswiththesense
of"obligation,"it toowillremainto be spokenwitha specialemphasis
and a specialfeelingin thesecontexts. It is as if
thenotion"criminal"wereto remainwhencriminallaw and
criminalcourtshad been abolishedand forgotten.A Hume
discoveringthis situation might conclude thatthere wasa special
sentiment,expressedby "criminal,"whichalone gave the wordits
sense.So Hume discoveredthesituationin whichthenotion
"obligation"survived,and the notion"ought" was investedwith that
peculiarforcehavingwhichit is said to be used in a "moral"
sense,but in whichthebeliefin divinelaw had longsincebeenaban-
doned: forit was substantiallygivenup amongProtestantsat the time
of the Reformation.'The situation,if Iam right,was the
interestingone ofthesurvivalof a conceptoutsidetheframeworkof
thoughtthatmade it a reallyintelligibleone. When Hume producedhis
famousremarksabout the transition I They did not deny the existence
of divine law; but their most charac- teristicdoctrinewas that it
was given,not to be obeyed, but to show man's incapacityto obey it,
even by grace; and thisapplied not merelyto the ramified
prescriptionsof the Torah, but to the requirementsof "natural
divine law." Cf. in this connectionthe decree of Trent against the
teaching that Christ was only to be trustedin as mediator,not
obeyed as legislator, 6 This content downloaded from 168.176.5.118
on Wed, 5 Feb 2014 22:09:18 PMAll use subject to JSTOR Terms and
ConditionsMODERNMORALPHILOSOPHY from"is" to "ought,"he was,
then,bringingtogetherseveralquite differentpoints.One I have
triedto bringout by myremarkson the transitionfrom"is" to "owes"
and on the relative"bruteness"of facts.It wouldbe possibleto
bringout a differentpointby enquiring about the transitionfrom"is"
to "needs"; fromthe characteristics ofan organismto
theenvironmentthatit needs,forexample.To say thatit
needsthatenvironmentis not to say,e.g.,thatyouwantit to have that
environment,but that it won't flourishunlessit has it. Certainly,it
all dependswhetheryou wantit to flourish!as Hume wouldsay. But
what"all depends"on whetheryouwantit to flourish is whetherthe
factthatit needsthat environment,or won'tflourish withoutit,has
theslightestinfluenceon youractions,Now thatsuch- and-such"ought"to
be or "is needed"is supposedto have an influence on
youractions:fromwhichit seemednaturalto inferthatto judge thatit
"oughtto be" was in factto grantwhatyoujudged"oughtto be"
influenceon youractions.And no amountof truthas to whatis thecase
couldpossiblyhave a logicalclaimto have influenceon your
actions.(It is not judgmentas suchthat sets us in motion;but our
judgmenton how to get or do somethingwe want.)Hence it mustbe
impossibleto infer"needs" or "oughtto be" from"is." But in the case
of a plant,let us say, the inferencefrom"is" to "needs" is
certainlynot in the least dubious. Itis interestingand worth
examining;but notat all fishy.Its interestis similarto
theinterestof the relationbetweenbruteand less
brutefacts:theserelationshave been very little considered.And while
you can contrast"what it needs" with "what it's
got"-likecontrastingde facto and de iure-that does not make its
needingthis environmentless of a "truth." Certainlyin thecase
ofwhattheplantneeds,thethoughtofa need will onlyaffectactionif you
want the plant to flourish.Here, then, thereis no
necessaryconnectionbetweenwhatyoucan judgetheplant "needs" and what
you want. But thereis some sort of necessary connectionbetweenwhat
you think you need,and what you want. The connectionis a
complicatedone; it is possiblenotto wantsome- thingthat you judge
you need.But, e.g.,it is not possibleneverto want anythingthatyou
judgeyou need. This,however,is not a fact
aboutthemeaningoftheword"to need,"but aboutthephenomenon of
wanting.Hume's reasoning,we mightsay, in effect,leads one to
thinkit mustbe about the word"to need," or "to be good for." Thuswe
findtwoproblemsalreadywrappedup in theremarkabout a
transitionfrom"is" to "ought";nowsupposingthatwe had clarified the
"relativebruteness"of factson the one hand, and the notions
involvedin "needing,"and "flourishing"on the other-therewould
stillremaina thirdpoint.For, followingHume,someonemightsay:
Perhapsyou have made out yourpointabout a transitionfrom"is" This
content downloaded from 168.176.5.118 on Wed, 5 Feb 2014 22:09:18
PMAll use subject to JSTOR Terms and ConditionsPHILOSOPHY to "owes"
and from"is" to "needs": but onlyat thecostofshowing "'owes" and
"needs" sentencesto expressa kindof truths,a kind offacts.And it
remainsimpossibleto infer"morallyought"from"is" sentences.
Thiscomment,it seemsto me,wouldbe correct.Thisword"ought,"
havingbecomea word of meremesmericforce,could not, in the
characterof havingthat force,be inferredfromanythingwhatever. It
may be objectedthat it could be inferredfromother"morally
ought"sentences:but thatcannotbe true.The appearancethatthis is so
is producedby the factthatwe say "All menare i"and "Soc- ratesis a
man" implies"Socratesis O." But here "i"is a dummy predicate.We
mean that if you substitutea real predicatefor "O" theimplicationis
valid. A real predicateis required;not just a word containingno
intelligiblethought:a wordretainingthesuggestionof force,and apt to
have a strongpsychologicaleffect,but whichno longersignifiesa real
conceptat all. For its suggestionis one of a verdicton my
action,accordingas it agreesor disagreeswiththedescriptionin
the"ought"sentence.And whereone does not thinkthereis a judge or a
law, the notionof a verdictmayretainits psychologicaleffect,but not
its meaning.Now imaginethat just thisword"verdict"wereso used-with
a charac- teristicallysolemnemphasis-as to retainits atmospherebut
not its meaning,and someonewereto say: "For a verdict,afterall,
youneed a law and a judge."The replymightbe made: "Not at all,
forifthere werea law and a judgewhogave a verdict,thequestionforus
would be whetheracceptingthatverdictis somethingthatthereis a
Verdict on." This is an analogue of an argumentwhichis so
frequently referredto as decisive:If someonedoes have a divinelaw
conception ofethics,all thesame,he has to agreethathe has to have a
judgment thathe ought(morallyought)to obeythe divinelaw; so his
ethicis in exactlythesamepositionas any other:he merelyhas a
"practical major premise"': "Divine law oughtto be
obeyed"wheresomeone elsehas,e.g.,"The
greatesthappinessprincipleoughtto be employed in all decisions." I
shouldjudge thatHume and our present-dayethicistshad done a
considerableserviceby showingthatno contentcouldbe foundin
thenotion"morallyought";ifit werenotthatthelatterphilosophers try
to findan alternative(veryfishy)contentand to retain the
psychologicalforceof the term.It wouldbe mostreasonableto drop it.
It has no reasonablesenseoutsidea law conceptionofethics;they are
not goingto maintainsuch a conception;and you can do ethics
withoutit, as is shownby the exampleof Aristotle.It wouldbe a I As
it is absurdlycalled. Since major premise=premise containing the
term which is predicatein the conclusion,it is a solecismto speak
of it in the con- nection with practical reasoning. 8 This content
downloaded from 168.176.5.118 on Wed, 5 Feb 2014 22:09:18 PMAll use
subject to JSTOR Terms and ConditionsMODERNMORALPHILOSOPHY
greatimprovmentif,insteadof "morallywrong,"one alwaysnamed a
genussuch as "untruthful,""unchaste,""unjust." We shouldno
longerask whetherdoingsomethingwas "wrong,"passingdirectly fromsome
descriptionof an action to this notion;we should ask
whether,e.g.,it was unjust;and theanswerwouldsometimesbe clear at
once. Inow come to the epoch in modernEnglish moral philosophy
markedby Sidgwick.Thereis a startlingchangethatseemsto have
takenplace betweeenMill and Moore.Mill assumes,as we saw, that
thereis no questionof calculatingthe particularconsequencesof an
actionsuchas murderor theft;and we saw too thathis positionwas
stupid,becauseit is not at all clearhowan actioncan fallunderjust
one principleof utility. InMoore and in subsequentacademic
moralistsofEnglandwe findit takento be prettyobviousthat "the
rightaction" is the actionwhichproducesthe best possibleconse-
quences(reckoningamongconsequencestheintrinsicvaluesascribed to
certainkindsofact by some"Objectivists"I).Now it followsfrom
thisthata mandoeswell,subjectivelyspeaking,ifhe acts forthebest in
theparticularcircumstancesaccordingto hisjudgmentofthetotal
consequencesof this particularaction. I say that this follows,not
that any philosopherhas said preciselythat.For discussionof these
questionscan of courseget extremelycomplicated:e.g. it can be
doubtedwhether"such-and-suchis therightaction"is a satisfactory
formulation,on thegroundsthatthingshave to existto have predi-
cates-so perhapsthebestformulationis "I am obliged";or again,a
philosophermaydenythat "right"is a "descriptive"term,and then take
a roundaboutroutethroughlinguisticanalysisto reacha view
whichcomesto thesamethingas "therightactionis theoneproduc- tive of
the best consequences"(e.g. the view that you frameyour
"principles"to effectthe end you chooseto pursue,the connexion
between"choice" and "best" beingsupposedlysuch that choosing
reflectivelymeansthat you choosehow to act so as to producethe best
consequences);further,the roles of what are called "moral
principles"and of the "motiveof duty" have to be described;the
differencesbetween"good" and "morallygood" and "right"need to be
explored,the specialcharacteristicsof "ought"sentencesinvesti-
gated.Suchdiscussionsgeneratean appearanceofsignificantdiversity
ofviewswherewhatis reallysignificantis an overallsimilarity.The
overallsimilarityis made clearifyou considerthateveryone of the I
Oxford Objectivists of course distinguishbetween "consequences" and
"intrinsicvalues" and so produce a misleadingappearance of not
being "con- sequentialists." But they do not hold-andRoss
explicitlydenies-thatthe gravityof,e.g., procuringthe
condemnationof the innocentis such that it can- not be
outweighedby, e.g., national interest.Hence theirdistinctionis of
no importance, 9 This content downloaded from 168.176.5.118 on Wed,
5 Feb 2014 22:09:18 PMAll use subject to JSTOR Terms and
ConditionsP H I L O S O P H Y best known English academic moral
philosophershas put out a philosophyaccordingto which,e.g.,it is
notpossibleto hold that it cannotbe rightto killtheinnocentas a
meansto anyendwhatsoever and thatsomeonewhothinksotherwiseis in
error.(I have to mention bothpoints;becauseMr. Hare,
forexample,whileteachinga philo- sophy whichwould encouragea
personto judge that killingthe innocentwouldbe whathe "ought"to
chooseforover-ridingpurposes, wouldalso teach,I think,thatifa
manchoosesto makeavoidingkill- ingtheinnocentforanypurposehis
"supremepracticalprinciple,"he cannotbe impugnedforerror:that just
is his "principle."But with thatqualification,I thinkitcanbe
seenthatthepointI havementioned holdsgood of
everysingleEnglishacademicmoralphilosophersince Sidgwick.)Now
thisis a significantthing:forit meansthatall these philosophiesare
quiteincompatiblewiththe Hebrew-Christianethic. For it has been
characteristicof that ethicto teach that thereare
certainthingsforbiddenwhateverconsequencesthreaten,such as:
choosingto killtheinnocentforanypurpose,howevergood;vicarious
punishment;treachery(by whichI mean obtaininga man's confi- dence
in a gravematterby promisesof trustworthyfriendshipand then
betrayinghim to his enemies);idolatry;sodomy; adultery; makinga
falseprofessionof faith.The prohibitionof certainthings simplyin
virtueof theirdescriptionas such-and-suchidentifiable kindsof
action,regardlessof any furtherconsequences,is certainly not the
wholeof the Hebrew-Christianethic;but it is a noteworthy featureof
it; and if everyacademicphilosophersince Sidgwickhas writtenin
sucha wayas to excludethisethic,it wouldarguea certain
provincialityof mind not to see this incompatibilityas the most
importantfactaboutthesephilosophers,and thedifferencesbetween
themas somewhattriflingby comparison. Itis noticeablethat none of
these philosophersdisplays any consciousnessthat thereis such an
ethic,whichhe is contradicting: it is
prettywelltakenforobviousamongthemall thata prohibition suchas
thaton murderdoesnotoperatein faceofsomeconsequences. But
ofcoursethestrictnessoftheprohibitionhas as itspointthatyou are
notto be temptedbyfearor hopeofconsequences. If you noticethe
transitionfromMill to Moore,you will suspect thatit was
madesomewhereby someone;Sidgwickwillcometo mind as a likelyname;and
youwillin factfindit goingon,almostcasually, in him.He is rathera
dull author;and the importantthingsin him occurin asides and
footnotesand smallbits of argumentwhichare
notconcernedwithhisgrandclassificationofthe"methodsofethics." A
divinelaw theoryofethicsis reducedto an insignificantvarietyby a
footnotetellingus that "the best theologians"(God knowswhom he
meant)tellus thatGod is to be obeyedin his capacityof a moral
being. ' 0OpTLKos o'C'Wcavoso; one seems to hear Aristotlesaying:
IO This content downloaded from 168.176.5.118 on Wed, 5 Feb 2014
22:09:18 PMAll use subject to JSTOR Terms and
ConditionsMODERNMORALPHILOSOPHY "Isn't the praisevulgar?",-But
Sidgwickis vulgarin that kindof way:he
thinks,forexample,thathumilityconsistsinunderestimating your own
merits-i.e. in a speciesof untruthfulness;and that the
groundforhavinglaws againstblasphemywas that it was offensive to
believers;and thatto go accuratelyintothevirtueof purityis to
offendagainstits canons,a thinghe reproves"medievaltheologians"
fornot realizing.
Fromthepointofviewofthepresentenquiry,themostimportant thingabout
Sidgwickwas his definitionof intention.Hedefines intentionin such a
way thatone mustbe said to intendany foreseen consequencesof one's
voluntaryaction.This definitionis obviously incorrect,and I daresay
thatno onewouldbe foundto defendit now. He uses it to put forwardan
ethical thesiswhichwould now be accepted by many people: the thesis
that it does not make any differenceto a man's
responsibilityforsomethingthat he foresaw, thathe feltno
desireforit, eitheras an end or as a meansto an end. Using the
language of intentionmore correctly,andavoiding
Sidgwick'sfaultyconception,we maystatethethesisthus:it doesnot make
any differenceto a man's responsibilityforan effectof his
actionwhichhe can foresee,thathe doesnotintendit.Nowthissounds
ratheredifying;it is I thinkquitecharacteristicofverybad degenera-
tionsofthoughton suchquestionsthattheysoundedifying.We can see what
it amountsto by consideringan example.Let us suppose that a man has
a responsibilityforthe maintenanceof some child.
Thereforedeliberatelyto withdrawsupportfromit is a bad sortof
thingforhimto do. It wouldbe bad forhimto withdrawits mainte-
nancebecausehe didn'twantto maintainit any longer;and also bad
forhimto withdrawit becausebydoingso he would,letus say,compel
someoneelse to do something.(We may suppose for the sake of
argumentthat compellingthat personto do that thingis in itself
quite admirable.)But nowhe has to choosebetweendoingsomething
disgracefuland goingto prison;if he goes to prison,it will follow
that he withdrawssupportfromthe child.By Sidgwick'sdoctrine,
thereis no differencein his responsibilityforceasingto maintainthe
child,betweenthe case wherehe does it forits own sake or as a
meansto someotherpurpose,and whenit happensas a foreseenand
unavoidableconsequenceofhis goingto prisonratherthando some-
thingdisgraceful.It followsthat he must weigh up the relative
badness of withdrawingsupportfromthe child and of doing the
disgracefulthing;and it mayeasilybe thatthedisgracefulthingis in
fact a less vicious action than intentionallywithdrawingsupport
fromthe childwould be; if thenthe factthat withdrawingsupport
fromthechildis a side effectofhisgoingto prisondoesnot make any
differenceto his responsibility,thisconsiderationwill inclinehimto
X E.,N. I I78bi6, II This content downloaded from 168.176.5.118 on
Wed, 5 Feb 2014 22:09:18 PMAll use subject to JSTOR Terms and
ConditionsP H I L O S O P H Y do thedisgracefulthing;whichcan
stillbe prettybad. Andofcourse, once he has startedto look at the
matterin this light,the only reasonablethingforhimto considerwillbe
theconsequencesand not the intrinsicbadnessof this or that
action.So that,giventhat he
judgesreasonablythatnogreatharmwillcomeofit,he can do a much
moredisgracefulthingthan deliberatelywithdrawingsupportfrom
thechild.Andifhis calculationsturnout in factwrong,it willappear
thathe was not responsibleforthe consequences,becausehe did not
foreseethem.For in fact Sidgwick'sthesisleads to its beingquite
impossibleto estimatethebadnessofan actionexceptin thelightof
expectedconsequences.But if so, thenyou mustestimatethebadness in
thelightoftheconsequencesyou expect;and so it will followthat you
can exculpateyourselffromthe actualconsequencesofthemost
disgracefulactions,so longas youcan makeout a case fornot having
foreseenthem.WhereasI shouldcontendthata manis responsiblefor the
bad consequencesof his bad actions,but gets no creditforthe good
ones; and contrariwiseis not responsibleforthe bad conse- quencesof
good actions. The denial of any distinctionbetweenforeseenand
intendedcon- sequences,as far as responsibilityis concerned,was not
made by Sidgwickin developingany one "methodof ethics"; he made
this importantmoveon behalfofeverybodyand just on its ownaccount;
and I thinkit plausibleto suggestthat thismove on the part of
Sidgwickexplainsthedifferencebetweenold-fashionedUtilitarianism and
that consequentialism,as I name it,whichmarkshimand every
Englishacademicmoralphilosophersince him. By it, the kind of
considerationwhichwouldformerlyhave beenregardedas a tempta-
tion,thekindofconsiderationurgeduponmenbywivesand flattering
friends,was givena statusby moralphilosophersin theirtheories. It
is a necessaryfeatureof consequentialismthat it is a shallow
philosophy.For thereare alwaysborderlinecases in ethics.Now if
youare eitheran Aristotelian,ora believerin divinelaw,youwilldeal
witha borderlinecase by consideringwhetherdoingsuch-and-suchin
such-and-suchcircumstancesis, say,murder,or is an act ofinjustice;
and accordingas you decideit is or it isn't,you judgeit to be a
thing to do or not.Thiswouldbe themethodofcasuistry;and whileit may
lead you to stretcha pointon the circumference,it will not permit
you to destroythe centre.But if you are a consequentialist,the
question"What is it rightto do in such-and-suchcircumstances?" is a
stupidone to raise.The casuistraisessucha questiononlyto ask "Would
it be permissibleto do so-and-so?"or "Would it be permis-
siblenotto do so-and-so?"Onlyif it wouldnotbe permissiblenotto do
so-and-socouldhe say " Thiswouldbe thethingtodo."I Otherwise, I
Necessarily arare case:for the positive precepts, e.g. "Honour your
parents," hardly ever prescribe,and seldom even necessitate,any
particular action. I2 This content downloaded from 168.176.5.118 on
Wed, 5 Feb 2014 22:09:18 PMAll use subject to JSTOR Terms and
ConditionsMODERNMORALPHILOSOPHY thoughhe
mayspeakagainstsomeaction,he cannotprescribeany- forin an
actualcase, the circumstances(beyondthe ones imagined)
mightsuggestall sortsofpossibilities,and youcan'tknowin advance
whatthepossibilitiesaregoingto be. Nowtheconsequentialisthas no
footingon whichto say "Thiswouldbe permissible,thisnot"; because by
his ownhypothesis,it is theconsequencesthatare to decide,and he has
no businessto pretendthathe can lay it downwhatpossible twistsa man
couldgivedoingthisor that; themosthe can say is: a man mustnot
bringaboutthisor that;he has no rightto say he will, in an
actualcase,bringaboutsuch-and-suchunlesshe does so-and-so.
Further,the consequentialist,in orderto be imaginingborderline
cases at all, has of courseto assume some sort of law or standard
accordingto whichthisis a borderlinecase, Wherethendoes he get the
standardfrom?In practicethe answerinvariablyis: fromthe
standardscurrentin his societyor his circle.And it has in factbeen
the markof all these philosophersthat theyhave been extremely
conventional;theyhave nothingin themby whichto revoltagainst the
conventionalstandardsof theirsort of people; it is impossible
thattheyshouldbe profound.But the chancethata wholerangeof
conventionalstandardswillbe decentis small.-Finally,thepointof
consideringhypotheticalsituations,perhapsveryimprobableones,
seemsto be to elicit fromyourselfor someoneelse a hypothetical
decisionto do somethingof a bad kind.I don't doubt thishas the
effectof predisposingpeople-who willneverget intothe situations for
which they have made hypotheticalchoices-to consentto similarbad
actions,or to praiseand flatterthosewho do them,so long as
theircrowddoes so too, whenthe desperatecircumstances
imagineddon'thold at all. Thosewhorecognizetheoriginsofthenotionsof
"obligation"and oftheemphatic,"moral,"ought,in thedivinelaw
conceptionofethics, but whorejectthenotionofa
divinelegislator,sometimeslookabout for the possibilityof
retaininga law conceptionwithouta divine legislator.This search,I
think,has someinterestin it. Perhapsthe
firstthingthatsuggestsitselfis the "norms"ofa society.But just as
one cannotbe impressedby Butlerwhenone reflectswhatconscience can
tellpeopleto do, so, I think,onecannotbe impressedby thisidea ifone
reflectswhatthe "norms"ofa societycan be like.That legisla- tion
can be "foroneself"I rejectas absurd;whateveryou do "for
yourself"maybe admirable;but is notlegislating.Onceone seesthis,
one maysay: I have to framemyownrules,and theseare thebest I can
frame,and I shallgo by themuntilI knowsomethingbetter:as aman
mightsay "Ishall go by the customsof my ancestors."
Whetherthisleads to good or evil will dependon the contentof the
rulesor of the customsof one's ancestors.If one is luckyit willlead
to good. Such an attitudewould be hopefulin this at any rate: it I3
This content downloaded from 168.176.5.118 on Wed, 5 Feb 2014
22:09:18 PMAll use subject to JSTOR Terms and ConditionsPHILOSOPHY
seemsto have in it some Socraticdoubtwhere,fromhavingto fall back
on such expedients,it shouldbe clear that Socraticdoubt is good; in
factrathergenerallyit mustbe good foranyoneto think "Perhapsin
someway I can't see, I maybe on a bad path,perhapsI am
hopelesslywronginsomeessentialway".-The searchfor"norms" mightlead
someoneto lookforlaws ofnature,as iftheuniversewere a
legislator;but in the presentday thisis not likelyto lead to good
results:it mightlead one to eat theweakeraccordingto the laws of
nature,butwouldhardlylead anyonenowadaysto notionsofjustice;
thepre-Socraticfeelingaboutjusticeas comparableto thebalanceor
harmonywhichkeptthingsgoingis veryremoteto us. Thereis
anotherpossibilityhere:"obligation"maybe contractual. Justas we
look at the law to findout what a man subjectto it is requiredby it
to do, so we lookat a contractto findoutwhattheman whohas madeit is
requiredby it to do. Thinkers,admittedlyremote fromus, mighthave
theidea ofa foeds rerum,oftheuniversenotas a legislatorbut as the
embodimentof a contract.Then if you could findout what the
contractwas, you would learn your obligations underit. Now,you
cannotbe undera law unlessit has beenpromul- gatedto you; and
thethinkerswhobelievedin "naturaldivinelaw" heldthatit was
promulgatedto everygrownmanin his knowledgeof good and evil.
Similarlyyou cannotbe in a contractwithouthaving contracted, i.e.
given signs of enteringupon the contract.Just possibly,it mightbe
arguedthattheuse oflanguagewhichonemakes in the ordinaryconductof
lifeamountsin somesenseto givingthe signsof
enteringintovariouscontracts.If anyonehad this theory, we
shouldwant to see it workedout. I suspectthat it would be
largelyformal;it mightbe possibleto constructa systemembodying the
law (whosestatusmightbe comparedto thatof "laws" oflogic): "what's
sauce forthe goose is sauce forthe gander,"but hardlyone
descendingto such particularitiesas the prohibitionon murderor
sodomy.Also,whileit is clearthatyou can be subjectto a law that you
do not acknowledgeand have not thoughtof as law, it does not
seemreasonableto say that you can enterupon a contractwithout
knowingthatyou are doingso; such ignoranceis usuallyheld to be
destructiveof the natureof a contract. It mightremainto
lookfor"norms"in humanvirtues:just as man has so manyteeth,whichis
certainlynottheaveragenumberofteeth menhave,but is thenumberof
teethforthe species,so perhapsthe speciesman, regardednot just
biologically,but fromthe pointof view of the activityof thoughtand
choicein regardto the various departmentsoflife-powers and
facultiesand use ofthingsneeded- "has" such-and-suchvirtues:and
this"man" withthe completeset ofvirtuesis the"norm,"as "man"
with,e.g.,a completeset of teeth is a norm.But inthissense"norm"has
ceasedto be roughlyequivalent '4 This content downloaded from
168.176.5.118 on Wed, 5 Feb 2014 22:09:18 PMAll use subject to
JSTOR Terms and ConditionsMODERNMORALPHILOSOPHY to "law." In
thissensethenotionofa "norm"bringsus nearerto an Aristotelianthana
law conceptionofethics.Thereis, I think,no harm in that; but if
someonelooked in this directionto give "norm" a sense,thenhe
oughtto recognizewhat has happenedto the notion "norm,"whichhe
wantedto mean"law-withoutbringingGod in" it has ceased to mean
"law" at all; and so the notionsof "moral
obligation,""themoralought,"and "duty"arebestputon theIndex, if he
can manageit. But meanwhile-isit notclearthatthereare
severalconceptsthat need investigatingsimplyas part of the
philosophyof psychology and,-asIshould
recommend-banishingethicstotallyfromour minds?Namely-to beginwith:
"action," "intention,""pleasure," "wanting." More will probablyturn
up if we start with these. Eventuallyit mightbe possibleto
advanceto consideringtheconcept "virtue";withwhich,I suppose,we
shouldbe beginningsomesortof a studyof ethics. I willendby
describingtheadvantagesofusingtheword"ought" in a
non-emphaticfashion,and not in a special "moral" sense; of
discardingthe term"wrongsin a"moral" sense, and using such
notionsas "unjust." It is possible,ifone is allowedto proceedjust
by givingexamples, to distinguishbetweenthe intrinsicallyunjust,and
what is unjust giventhecircumstances.To arrangeto geta
manjudiciallypunished forsomethingwhichit can be clearlyseen he has
not done is in- trinsicallyunjust.This mightbe done,of course,and
oftenhas been done,in all sortsof ways; by
suborningfalsewitnesses,by a ruleof law by whichsomethingis
"deemed" to be the case whichis ad- mittedlynot the case as a
matterof fact,and by open insolenceon the part of the judges and
powerfulpeoplewhentheymoreor less openlysay: "A figforthe factthat
you did not do it; we mean to sentenceyou forit all the same." What
is unjustgiven,e.g.,normal circumstancesis to deprive people of
their ostensibleproperty withoutlegalprocedure,not to pay
debts,notto keepcontracts,and a
hostofotherthingsofthekind.Now,thecircumstancescan clearly make a
greatdeal of differencein estimatingthe justiceor injustice of such
proceduresas these;and thesecircumstancesmay sometimes
includeexpectedconsequences;forexample,a man's claim to a bit of
propertycan becomea nullitywhenits seizureand use can avert
someobviousdisaster:as, e.g.,if you could use a machineof his to
producean explosionin whichit wouldbe destroyed,but by meansof
whichyou could diverta floodor make a gap whicha firecould not
jump. Now thiscertainlydoes not mean thatwhatwouldordinarily be an
act of injustice,but is not intrinsicallyunjust,can alwaysbe
renderedjust by a reasonablecalculationofbetterconsequences;far
fromit; but theproblemsthatwouldbe raisedin an attemptto draw '5
This content downloaded from 168.176.5.118 on Wed, 5 Feb 2014
22:09:18 PMAll use subject to JSTOR Terms and ConditionsP H I L O S
O P H Y a boundaryline (or boundaryarea) hereare
obviouslycomplicated. Andwhilethereare certainlysome
generalremarkswhichoughtto be made here,and someboundariesthatcan
be drawn,the decision on particularcases would forthe most part be
determinedKac-a TOv 3pOpov Voyov"accordingto what's
reasonable."-E.g. that such- and-sucha delay of paymentof a
such-and-suchdebt to a personso circumstanced,on the part of a
personso circumstanced,would or wouldnotbe unjust,is reallyonlyto
be decided"accordingto what's reasonable";and forthis therecan in
principlebe no canon other than givinga fewexamples.That is to say,
whileit is because of a big gap in philosophythat we can give no
generalaccountof the conceptof virtueand of the
conceptofjustice,but have to proceed, usingthe concepts,onlyby
givingexamples;still thereis an area whereit is not because of any
gap, but is in principlethe case, that thereis no accountexceptby
way ofexamples:and thatis wherethe canonis
"what'sreasonable":whichof courseis nota canon. That is all I
wishto say aboutwhatis just in somecircumstances, unjustin
others;and about theway in whichexpectedconsequences can playa
partin determiningwhatis just. Returningto myexample
oftheintrinsicallyunjust:ifa procedureis oneofjudiciallypunishing a
man forwhathe is clearlyunderstoodnot to have done,therecan be
absolutelyno argumentabout the descriptionof this as unjust. No
circumstances,and no expected consequences,which do not
modifythedescriptionoftheprocedureas one ofjudiciallypunishing a
manforwhathe is knownnotto have donecan modifythedescrip- tion of
it as unjust.Someonewho attemptedto disputethiswould onlybe
pretendingnot to knowwhat "unjust" means:forthisis a paradigmcase
of injustice. And here we see the superiorityor the term"unjust"
over the terms"morallyright"and "morallywrong."For in the contextof
English moral philosophysince Sidgwickit appears legitimateto
discusswhetherit mightbe "morallyright"in somecircumstancesto adopt
that procedure;but it cannotbe arguedthat the procedure wouldin any
circumstancesbe just. Now I am not able to do the
philosophyinvolved-and I think thatno one in
thepresentsituationofEnglishphilosophycan do the
philosophyinvolved-but it is clearthat a good man is a just man;
and a just manis a manwhohabituallyrefusesto commitor partici- pate
in any unjustactionsforfearof any consequences,or to obtain any
advantage,for himselfor anyone else. Perhaps no one will
disagree.But, it willbe said,whatis unjustis sometimesdetermined by
expectedconsequences;and certainlythat is true.But thereare
caseswhereit is not:nowifsomeonesays,"I agree,but all thiswants a
lot ofexplaining,"thenhe is right,and,whatis more,thesituation at
presentis thatwe can't do theexplaining;we lack thephilosophic This
content downloaded from 168.176.5.118 on Wed, 5 Feb 2014 22:09:18
PMAll use subject to JSTOR Terms and
ConditionsMODERNMORALPHILOSOPHY equipment.But
ifsomeonereallythinks,in advance,ithatit is open to
questionwhethersuchan actionas procuringthejudicialexecution of the
innocentshouldbe quite excludedfromconsideration-Ido not wantto
arguewithhim;he showsa corruptmind. In such cases our
moralphilosophersseek to imposea dilemma upon us. "If we have a
case wherethe term'unjust' appliespurely in virtueofa
factualdescription,can't oneraisethequestionwhether one
sometimesconceivablyoughtto do injustice?If 'whatis unjust' is
determinedby considerationofwhetherit is rightto do so-and-so in
such-and-suchcircumstances,then the question whetherit is 'right'to
commitinjusticecan't arise, just because 'wrong'has
beenbuiltintothedefinitionofinjustice.But ifwe have a case where
the description'unjust'appliespurelyin virtueof the facts,without
bringing'wrong'in, thenthe questioncan arisewhetherone 'ought'
perhapsto commitan injustice,whetherit mightnot be 'right'to? And
of course 'ought' and 'right'are being used in theirmoral
senseshere.Now eitheryou mustdecidewhat is 'morallyright'in the
light of certain other'principles,' or you make a'principle' about
thisand decide that an injusticeis never'right';but even if you do
the latteryou are goingbeyondthe facts;you are makinga decisionthat
you will not,or that it is wrongto, commitinjustice. But in
eithercase, if the term 'unjust' is determinedsimplyby the facts,it
is not the term'unjust' that determinesthat the term 'wrong'
applies, but adecision that injusticeis wrong,together withthe
diagnosisof the 'factual'descriptionas entailinginjustice. But the
man who makes an absolute decision that injustice is 'wrong'has no
footingon whichto criticizesomeonewho does not make that decisionas
judgingfalsely." Inthis argument"wrong" of course is explained
asmeaning "morallywrong,"and all theatmosphereofthetermis
retainedwhile its substanceis guaranteedquite null. Now let us
rememberthat ''morallywrong''is thetermwhichis
theheirofthenotion"illicit,"or "whatthereis an obligationnotto do";
whichbelongsin a divinelaw theoryor ethics.Here it reallydoes add
somethingto thedescription "unjust" to say there is an obligation
not to do it; for what obliges I If he thinksit in the
concretesituation,he is of course merelya normally temptedhuman
being. In discussionwhen this paper was read, as was perhaps to be
expected, this case was produced: a governmentis requiredto have an
innocent man tried, sentenced and executed under threat of
a"hydrogen bombwar."Itwouldseemstrange to me to havemuch hopeof
soaverting a war threatenedby such men as made this demand. But the
most important thingabout the way in whichcases like this are
inventedin discussions,is the assumption that only two courses
areopen:here, compliance andopen defiance.No one can say in advance
of such a situationwhat the possibilities are going to be-e.g.that
there is none of stallingby a feignedwillingnessto
comply,accompanied by a skilfullyarranged "escape" of the victim. B
I7 This content downloaded from 168.176.5.118 on Wed, 5 Feb 2014
22:09:18 PMAll use subject to JSTOR Terms and ConditionsPHILOSOPHY
is the divinelaw-asrules obligein a game. So if the divinelaw
obligesnot to commitinjusticeby forbiddinginjustice,it reallydoes
add somethingto the description"unjust" to say thereis an obliga-
tionnotto do it,Andit is because"morallywrong"is theheirofthis
concept,but an heirthatis cut offfromthefamilyofconceptsfrom
whichit sprang,that "morallywrong"bothgoes beyondthe mere
factualdescription"unjust"andseemsto have no discerniblecontent
excepta certaincompellingforce,whichI shouldcall purelypsycho-
logical.And such is the forceof the termthat philosophersactually
supposethat the divinelaw notioncan be dismissedas makingno
essentialdifferenceeven if it is held-because they thinkthat a
"'practicalprinciple"running"Iought(i.e. am morallyobliged)to
obeydivinelaws" is requiredforthemanwhobelievesin divinelaws. But
actuallythisnotionofobligationis a notionwhichonlyoperates in the
contextof law. And I shouldbe inclinedto congratulatethe
present-daymoralphilosopherson depriving"morallyought"of its
nowdelusiveappearanceof content,if onlytheydid not manifesta
detestabledesireto retainthe atmosphereof the term. It maybe
possible,ifwe areresolute,to discardthenotion"morally ought,"and
simplyreturnto the ordinary"ought",which,we ought to notice,is
suchan extremelyfrequenttermofhumanlanguagethat it is difficultto
imaginegettingon withoutit. Now ifwe do returnto it, can't it
reasonablybe asked whetherone mightever need to commitinjustice,or
whetherit won't be the best thingto do? Of courseit can. And the
answerswill be various.One man-aphilo- sopher-may say that
sincejusticeis a virtue,and injusticea vice, and virtuesand vices
are builtup by the performancesofthe action in whichtheyare
instanced,an act of injusticewill tendto make a man bad; and
essentiallythe flourishingof a man qua man consists in his
beinggood (e.g. in virtues);but forany X to whichsuchterms apply,X
needswhat makesit flourish,so a man needs,or oughtto
perform,onlyvirtuousactions;and even if,as it mustbe admitted may
happen, he flourishesless, or not at all, in inessentials,by
avoidinginjustice,his life is spoiled in essentialsby not avoiding
injustice-so he still needs to performonly just actions.That is
roughlyhow Plato and Aristotletalk; but it can be seen that
philosophicallythereis a hugegap, at presentunfillableas faras we
are concerned,whichneedsto be filledbyan accountofhumannature,
humanaction,the typeof characteristica virtueis, and above all of
human"flourishing."Andit is thelast conceptthatappearsthemost
doubtful.For it is a bit much to swallowthat a man in pain and
hungerand poor and friendlessis "flourishing,"as Aristotlehimself
admitted.Further,someonemightsay that one at least neededto stay
alive to "flourish."Anotherman unimpressedby all thatwill say in a
hard case "What we need is such-and-such,whichwe won't i8 This
content downloaded from 168.176.5.118 on Wed, 5 Feb 2014 22:09:18
PMAll use subject to JSTOR Terms and
ConditionsMODERNMORALPHILOSOPHY getwithoutdoingthis(whichis
unjust)-so thisis whatwe oughtto do."
Anotherman,whodoesnotfollowtheratherelaboratereasoning
ofthephilosophers,simplysays "I knowit is in anycase a disgraceful
thingto say thatonehad bettercommitthisunjustaction."The man who
believesin divinelaws will say perhaps"It is forbidden,and
howeverit looks,it cannotbe to anyone'sprofitto commitinjustice";
he liketheGreekphilosopherscan thinkin termsof "flourishing."If he
is a Stoic,he is apt to have a decidedlystrainednotionof what
"flourishingconsists"in; ifhe is a Jewor Christian,he neednothave
any verydistinctnotion:the way it will profithimto abstainfrom
injusticeis somethingthathe leaves it to God to determine,himself
onlysaying"It can't do me any goodto go againsthis law." (But he
also hopesfora greatrewardin a newlifelateron,e.g. at thecoming of
Messiah;but in thishe is relyingon specialpromises.) It is leftto
modernmoralphilosophy-themoralphilosophyofall
thewell-knownEnglishethicistssinceSidgwick-toconstructsystems
accordingto whichthe man who says "We need such-and-such,and
willonlygetit thisway" maybe a virtuouscharacter:thatis to say, it
is leftopen to debate whethersuch a procedureas the judicial
punishmentof the innocentmay not in some circumstancesbe the
"right"one to adopt; and thoughthe presentOxfordmoralphilo-
sopherswould accorda man permissionto "make it his principle" not
to do such a thing,theyteach a philosophyaccordingto which the
particularconsequencesof such an action could "morally"be takeninto
accountby a man who was debatingwhat to do; and if theyweresuchas
to conflictwithhis "ends,"it mightbe a stepin his moral education
to frame amoral principleunder which he "managed" (to use
Mr.Nowell-Smith's phrase') to bringthe action; or it mightbe a new
"decisionof principle,"makingwhichwas an advancein the formationof
his moralthinking(to adopt Mr.Hare's conception),to decide:in
such-and-suchcircumstancesone oughtto procurethe judicial
condemnationof the innocent.And that is my complaint.
SomervilleCollege,Oxford. I Ethics,p.308. I9 This content
downloaded from 168.176.5.118 on Wed, 5 Feb 2014 22:09:18 PMAll use
subject to JSTOR Terms and Conditions