Department of Economics, Universite Catholique de Louvain

Adam Smith and Three Theories of Altruism

Author(s): Elias L. Khalil
Source: Recherches Économiques de Louvain / Louvain Economic Review, Vol. 67, No. 4 (2001),
pp. 421-435
Published by: Department of Economics, Universite Catholique de Louvain
Stable URL: http://www.jstor.org/stable/40724332 .
Accessed: 14/11/2013 04:32

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

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Department of Economics, Universite Catholique de Louvain is collaborating with JSTOR to digitize, preserve
and extend access to Recherches Économiques de Louvain / Louvain Economic Review.

http://www.jstor.org

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
 Adam Smith and Three Theories of Altruism

Elias L. Khalil*
AmericanInstitute
forEconomic Research,
ofEconomics,VassarCollege
and Department

Introduction

Duringthe past threedecades,economists havedevelopedtheoriesof al-

truismthatdepartin different ways from the narrowlyconceivedHomo
economicus model.Thesetheories intothreebroad
can be broadlyclassified
approaches.The first,calledherethe "egoistic"perspective, can be seen
as a variantof reciprocalcooperationmodelssuch as the one proposed
by Axelrod(1984). It maintains thatone maysharehis incomewithano-
therto inducea reciprocal transferin the future.The second,namedthe
"egocentric"view,as epitomized in Becker(1976),arguesthatthedonor's
utilityfunction
includestheutilityofpotential Thatis,thedonor
recipients.
woulddonatea resource ifthevicariousenjoyment ofwatching thepleasure
ofothersexceedsat themarginthedonor'ssatisfaction from consuming the
resourcehimself.The third,dubbedthe "altercentric" framework ("alter"
afterthe Latin "other"),can be surmisedfromtheworkof Mead (1934),
Etzioni(1986),Prank(1988),and Simon(1990). It viewsthebenefactor's
actionas stemming froma moraldictumas bindingas rulesofhonesty.
Smithleveleddirectcriticismsagainstthreetheories ofaltruism cur-
rentinhistime.Amazingly, thesetheories arereplicasoftheonesjustmen-
tioned.Smithput forward an alternativebased on the idea of sympathy.
* Behavioral Sciences Research
Council, a division of the American Institutefor Economic Research,
www.brc-aier.org,and Departmentof Economics, Vassar College, elkhalil@vassar.edu. A longerversion
benefitedfromthecommentsofGary Becker,UlrichWitt,RobertGoldfarb,MarkWilhelm,TimothyCrippen,
JohnDavis, Thomas Nitsch,Roger Masters,participantsof sessions at the Universityof Freiburg,Univer-
sityofChicago, and AmericanEconomic Associationmeeting,anonymousreferees,and especially Robert
Frank.This workwas made possible bya researchfellowshipfromtheAlexandervon HumboldtFoundation.
The usual disclaimerapplies.

421

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
422 Recherches
Économiques - Louvain
de Louvain Review
Economie 67(4),2001

Althougha fewmodernwriters havenotedSmith'stheory(e.g.,Collard,

(1978;Frank,(1988),a fewer realized
itspertinence
to themoderndebate.
Sectionone providesa succinctstatement of Smith'stheory.Sectiontwo
summarizes Smith'scritiqueofthethreetheoriesofhistime.Sectionthree
the modernapproachesand showshowtheystillfailto answer
identifies
Smith'scritique.

1 Back to Smith

1.1 The PrincipleofSympathy

Smith'stheory ofethicswas greatlyinfluenced byDavid Hume'snotionof
sympathy (Mackie,(1980;Haakonssen, (1981).However, in The Theoryof
MoralSentiments, SmithdepartedfromHume'sutilitarian position, which
portrays sympathy as themechanism bywhichtheagentcalculatessocial
welfare. ForSmith,sympathy stemsfrominstantaneous sentiments towards
immediate experiences;it does notgenerally arisefrommeditated calcula-
tionofthewelfare of all concerned. Smithviewedsympathy as the foun-
dationofvirtuespursuedfortheirownsake,notforthesakeofadvancing
socialwelfare. He regardedsympathy to be the foundation ofbeneficence
(altruism), self-command (thepropriety behavior),respectand admira-
of
tion, and socialrank (Khalil(1990),(1996)).
Concerning benevolence, Smithstressedthatsympathy expresses the
genuineconcernovertheinterests ofothers,in short"other-interest" . This
concernentailsthatthe benefactor has to suspendhis owninterest. The
negationofself-interest, however, does notmeanthataltruism stemsfrom
a principle, whichis radicallydifferent fromself-interest. For Smith,the
motiveto satisfyself-interest and other-interest stemsfromthe same ge-
neraltendency ofhumansto sympathize-in one case withtheselfand in
theotherwiththebeneficiary. That is, Smithdid notviewself-interest as
radically differentfrom other-interest: botharesimplydifferent instancesof
sympathy. We witnessthatmanactsmoreoftenin sympathy withtheself
(i.e.,out ofself-interest) becausemanis obviously morefamiliar withthe
circumstance ofhisownselfthanwiththecircumstance ofothers.That is,
forSmith,thereis no fundamental distinction, butonlya difference in de-
gree, between one's own as
feelings opposed to the feelingof others towards
one'sinterest.
To be precise,however, Smithappearsto notea difference between
self-interest and the sympathy of otherswithone's interest. Whileself-
interest seemsto be an "original" sensation, thesympathy ofotherswith
one's interest does not take place immediately. Rather, is a mediated
it
or "reflected" sympathy with the agent who is originallyexperiencing the
benefit or pain:

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
EliasL. Khalil 423

Everyman,as theStoicsusedto say,is firstand principally recommended

to his own care;and everyman is certainly, in everyrespect,fitterand
ablerto takecareofhimself thanofanyotherperson.Everymanfeelshis
ownpleasuresand hisownpainsmoresensiblythanthoseofotherpeople.
The former ; the latterthe reflected
are originalsensations or sympathetic
imagesofthosesensation.The former maybe said to be thesubstance,the
lattertheshadow(Smith(1976) p. 219).

Thatis,a spectator can assessa benefit orpainonlythrough thereac-

tionoftheagentwhois experiencing it.Nonetheless,thisdifferencedoesnot
changetheclaimpositedherethatthereis no fundamental distinction
bet-
weenself-interestandsympathy. Whenan agentsympathizes withsomeone
else'sfeelings
towards, e.g.,an apple,it is a reflectivesensation.Likewise,
whentheagentsympathizes withhisownfeelings towardstheapple,it is
also a reflective
sensation : That is, bothsensations involvesympathy and,
hence,bydefinition, are reflectiveof originalsensations.The onlydifference
betweenthetwocases arisesfromthedegreeof familiarity occasionedby
the usual factthatthe agentis morefamiliar withhis ownfeelings than
withthefeelings oftheother.(To note,however, thisis notalwaystrueas
in thecase whenan agentis moreattunedto thefeelings ofothersthanto
hisown.)
Foraltruism, thedegreeoffamiliarity is crucialandagentsact mostof
thetimein a self-interested manneronlybecausetheyare familiar mostly
withtheirownoriginal sensations thanwiththeoriginal sensationsofothers.
Obviously, thereis a stronger motiveto helpa stranded personiftheperson
happensto be a closeacquaintance ratherthan,ceterisparibus,a distant
associate.Andmanis moremotivated to help,afterhimself,theoneswho
livein thesamehousewithhimthan"thegreaterpartofotherpeople":
Afterhimself,themembers ofhisownfamily,thosewhousuallylivein the
same housewithhim,his parents,his children, his brothersand sisters,
are naturally
theobjectsofhiswarmestaffections.Theyare naturally and
usuallythepersonsuponwhosehappinessor misery hisconductmusthave
thegreatestinfluence.
He is morehabituatedto sympathizewiththem.He
knowsbetterhoweverythingis likelyto affectthem,and his sympathy
withthemis morepreciseand determinate,thanit can be withthegreater
partof otherpeople.It approachesnearer,in short,to whathe feelsfor
himself(Smith(1976) p. 219).
Social proximity,
in addition,illuminatesforSmithwhythe agentfeels
less affectiontowardsthe childrenof his cousinsthan towardsthe children
of his sistersand brothers:
The childrenof cousins,beingstillless connected[than"thechildrenof
brothersand sisters"],are ofstillless importance
to one another;and the
affection
graduallydiminishes as therelationgrowsmoreand moreremote
(Smith(1976) p. 220).
Social remotenessalso explains forSmithwhythe parent-childaffec-
tion is weakerifthe child was separated fromthe fatherfromits infancy:

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
424 Recherches
Économiques - Louvain
de Louvain Economie
Review
67(4),2001

apt to be lessattachedto a child,who,bysomeaccident,has been

A father
separatedfromhimin itsinfancy, and whodoes notreturnto himtillit is
grownup to manhood.The fatheris apt to feellesspaternaltenderness for
thechild;thechild,lessfilialreverence
forthefather.Brothers and sisters,
whentheyhavebeeneducatedin distantcountries, are apt to feela similar
diminutionofaffection (Smith(1976) pp. 220-221).
Smith'snotionofsocial proximity is conceptuallyanalogousto genetic
proximity at the biologicallevel. Sociobiologistshave employedsuch genetic
proximity as the explaining factor ofaltruismin humanand non-humanani-
mals, what is known as the "inclusive fitness"hypothesis(Hamilton(1964);
Wilson (1975)). The contrastbetweenSmith'stheoryand the inclusivefit-
ness hypothesiscannotbe elaboratedhere.However,it is sufficient to state
that the ideas of social proximityand geneticproximitytryto account for
the strengthor weaknessof sympatheticsentiments.Whethersympathetic
sentimentsare learnedreactionsthroughconditioning,or have deep biolo-
gical foundation,is anotherissue that is not discussedhere.
to note that Smith's theoryallows us to conceiveal-
It is sufficient
truismas an elasticbehavior.It varieswiththevariationofsocial proximity;
it is not limitedby fixedgeneticproximity. To accountforsocial proximity,
whichpartiallydeterminesthe altruisticact, we need an operativemorpho-
logyof the stationof the actor in relationto the recipient's.

1.2 Sympathyas Station Switching

As stated above, Smith's principleof sympathyentailscontinuitybetween
the pursuitsofself-interest (what Smithcalls the "virtueofprudence") and
other-interest(the "virtueof beneficence" ). He groundedthe commensura-
bilitybetweenself-and other-interest on the "self acting froma separate
station-a stationthat impartiallyadjudicates betweenthe needs of the self
and the needs of the other:
We can neversurveyourownsentiments and motives...unlesswe remove
ourselves,as it were,fromour ownnaturalstationand endeavorto view
themas at a certaindistancefromus. ... We endeavorto examineourown
conductas weimagineanyotherfairandimpartial spectatorwouldexamine
it (Smith(1976) p. 110) (emphasisadded).
WhenI endeavorto examinemyownconduct...it is evidentthat...I divide
myselfintotwopersons;andthatI, theexaminer andthejudge,representa
different
character fromthatotherI, thepersonwhoseconductis examined
intothejudgedof (Smith(1976) p. 113).
For Smith, humans are capable of judging themselvesbecause the
principletheyuse to judge othersis readilyavailable:
bywhichwenaturally
The principle ofourown
eitherapproveordisapprove
conduct,seemsto be altogether the same withthatby whichwe exercise
thelikejudgements concerningtheconductofotherpeople(Smith(1976)
p. 107).

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
EliasL. Khalil 425

However, thereis an irregularity thatSmithoverlooks. Agentsoften

applydoublestandards : In manycases,agentstendto do whattheywould
condemnin others.At firstexamination at least,thereis a difference bet-
weentheagent-as-actor and theagent-as-judge. in
Agents many occasions
do notswitchstationswhentheyjudgetheirownactionsas fairly as when
theyjudge the actions of others.Although Smith discusses many other ir-
regularities stemming from the weakness of character, such as false pride
and self-aggrandizement (Khalil(1996)),he does notdiscussthisirregula-
rityexplicitly. Smithdoesdiscusssimilarself-biases, suchas self-deception,
and reasonsthat"generalrules", or whatmoderneconomists call "institu-
tions", ariseto correctthem.It wouldbe outsidethescopeofthisessayto
detailSmith'sdiscussion oftheoriginofgeneralrules(see Khalil(2000)pp.
381-384).However, following Smith'sorderof theoretical approximations,
it wouldbe consistent to assume,at firstlevelofgenerality, thatagentsdo
notsuffer fromdoublestandards, and thenaccountforanomaliessuchas
doublestandards, andcorresponding remedies, at secondand thirdlevelsof
generalities.In this light, double standards are notfatalanomaliesbecause
theycan explaintheoriginofgeneralrules.
Disregarding doublestandards, self-judgment, forSmith,is possible
becausenaturehasendowed theagentwiththequestto be "whathehimself
approves ofothermen", andto "dreadtheverythought ofresembling" what
he hatesand despisesin otherpeople(Ibid.p. 117). Thus,whentheagent
examines themeritofhisaction,he adoptstheviewofa would-be impartial
spectatoror at leasthe is supposedto adopttheviewoftheimpartial spec-
tator.It is truethatalmostall people,at leastin a fewoccasions,failto do
so. But thisis a problemforsecondand thirdtheoretical approximations
as just noted.
Smith'stheory ofsentiments resembles, at theformal levelonlyhowe-
ver,Becker'segocentric approach.It differs from Becker'sat thesubstantive
level.To reconstruct Smith'stheory, we haveat handthestationoftheac-
tingself,5, the stationof the recipient other,O, and the stationof the
would-be impartial spectator occupiedbytheselfreflecting on itself,coined
here"spectator 5s.
self," Figure 1 illustratessuch a three-station scenario,
where5s examinestheutilities ofO and 5s froma detached,thirdstation
whoselocationis determined by thedegreeoffamiliarity. The occupation
ofthethirdstationoftheimpartial spectatorbytheactorhimself has two
clearimplications. First,whentheactorempathizes withthesuffering ofthe
other,it is notbyimagining suchsuffering as happening to hisownperson
as supposedbyBecker'segocentric view.Second,thejudgeofthepotential
actionis not an actualspectatorfortworeasons.The firstreasonis that
thejudgeis nota disinterested observer according to whoseopinionagentS
triesto conform. In otherwords,wedo nothaveherea socialization process
whereS triesto appeasethepublicand gainitsapplause.Rather,S adopts
theviewofSs - theimpartial spectator whoemerges whichS examineshis
ownact froma distance.Sucha Smithian approachmeansthat,at firstap-
proximation, the "self precedespublicopinionor thesociocultural milieu.

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
426 RecherchesÉconomiquesde Louvain- LouvainEconomieReview67(4), 2001

1stStationS ^ ^ 2rdStationO

'/
2ndStationO
ScenarioofAltruism
Figure 1 : Three-Station

But,forSmith,themilieuis nottotallydisregarded. It playsa rolewhenit

helps the formation ofthe selfbyallowing agent reflect
the to on hisaction
as he does on theactionofothers.Withoutsociety, theagentcannottake
a lookat hisactionfroma distance.So, societyacts as a reflection mirror
thatclarifies S's viewofhimself.
and intensifies ForSmith,whilesocietyis
essentialforthedevelopment oftheself,it does notact as a construction
engineer à la Mead.
The secondreasonis thatthejudgeoftheactionis notO as entailed
by the Mead/Frank altercentricview.Rather,it is 5s whojudges,given
the degreeof familiarity, whether the benefitof O is worthsatisfying at
theexpenseofS's interest. Thisentailsthat,besidesfamiliarity, onehas to
introduce theissueofcomparative benefit inorderto providea determinate
judgmentifsomeoneshouldhelpa potentialbeneficiary. It is insufficient
thatthe otheris a close associate.The benefactor maypreferto donate
fundsto a charity thathelpsabusedchildren thanto helpa lessneedybut
a closeracquaintance. Smithdidnotdiscusstheissueofcomparative benefit
explicitly.However, Smith succeeded in the of
locating question familiarity
or sympathy as theproperentrypointto thestudyofaltruism.

1.3 The Advantages of Smith's Theory

Smith'snotionofsympathy forseveralreasons.First,Smithuses
is fruitful
theconcept"sympathy" ina sensemuchbroaderthanaltruism. He employs
ittoexplainetiquettes,thepropriety ofemotions in the
public, judgement of
respect, of
theexpression admiration, and so on (Khalil(1990),(1996)). Se-
cond,Smith'snotionofsympathy allowsoneto commence withtheinterac-
tionofrationalagents-where agents make decisions inlightofpreferencesof
theselfand cared-about other,constraints, and technology. Smith'snotion
ofsympathy accountsforaltruism without appealingto specialpreferences.
In thissense,Smith'snotionaccordswithone majorthrustof economic
theory, ofpreferences
i.e.,thestability : Do notstarttheorizing at theabs-
tractlevelby tracing differences in decisionsto differencesin preferences
(Stigler& Becker(1977))ranging fromGod's commandments, socialvalues
and commitments, genetic/biologicalfactors,to culturalinstitutions.The
commencement withpreferences andvariation acrosscultures begsa bigger

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
 EliasL. Khalil 427

question: whydo preferences differ? In fact,one mayuse Smith'snotion

ofsympathy to uncovertheoriginofmanyculturalinstitutions and social
norms.
Third,Smith'sapproachlocatesthe explaining variable,sympathy,
at the same levelas the explaineditem,the genuineconcernabout the
recipient'swelfare.Fourth,Smith'snotionallowsus to avoidthearbitrary
divisionofagentsintothealtruistandegoisttypes(e.g.,Becker(1976)).To
determine whois thealtruistand whois thebeneficiary, onedoes noteven
needto assumedifferent intensitiesofsympathetic sentimentsor appealto
tastes.The onlydeterminant can be relativeincomesor endowments.
Fifth,Smith'snotionpermits us to use theprinciple
ofrationalityin
the senseof the maximization modelof standardeconomics; i.e., thereis
no needfortheevolutionary Darwinianmodelor,in specific, theinclusive
fitnesshypothesis.Thisis an advantage sincetheevolutionarymodelcannot
explaineasilyaltruism whenthereis no geneticproximity.1
Sixth,Smith's startingpointwithsympathy theimportance
highlights
of social proximity (Smith(1976) pp. 219-221).Familiarityprovidesthe
operative morphology ofthestationoftheactorinrelationto thestationof
thepotentialbeneficiary. To quoteSmithon theimportance offamiliarity,
as muchas a manofhumanity in Europesympathizes withthevictimsofa
calamityin China,he is, "provided he neversaw them",morestressedfor
losinghislittlefinger:
Let us supposethatthegreatempireofChina,withall itsmyriads ofinhabi-
tants,wassuddenly swallowedup byan earthquake, and letus considerhow
a manofhumanity in Europe,whohad no sortofconnexion withthatpart
oftheworld,wouldbe affected uponreceiving intelligence ofthisdreadful
calamity.He would,I imagine,firstofall, expressverystrongly hissorrow
forthemisfortune ofthatunhappypeople,hewouldmakemanymelancholy
reflections
upontheprecariousness ofhumanlife,and thevanityofall the
laboursofman,whichcouldthusbe annihilated in a moment. ... Andwhen
all thisfinephilosophy was over,whenall thesehumanesentiments had
been once fairlyexpressed,he wouldpursuehis businessor his pleasure,
takehis reposeor his diversion,
withthe same ease and tranquillity, as if
no suchaccidenthad happened.The mostfrivolous disasterwhichcould
befalhimself wouldoccasiona morerealdisturbance. If he was to lose his
littlefinger
to-morrow, he wouldnotsleepto-night; but,providedhe never
saw them,he willsnorewiththe mostprofound security overthe ruinof
a hundredmillionsof his brethren, and the destruction of that immense
multitude seemsplainlyan objectless interestingto thanthispaltry
misfortune ofhisown(Smith(1976) pp. 136-137). 2 him,

1 The
suggestion that evolutionary theory is somewhat irrelevant goes against the work of Frank (1988),
Margolis (1982), Simon (1990), Bergstrom (1995), Getty (1989), and many others who tryto trace altruism
to some genetic foundation. Even ifsympathy has some biological foundation, one cannot explain the variety
of choices by appealing to genes. Also, ifsympathy has a biological foundation, it does not mean it is the
product of natural selection. In fact, natural selection cannot explain the origin of sympathy; it can only
explain its diffusion given that itexists. The inclination to invoke natural selection as soon as one appeals to
the relevance of biology with respect to behavior probably stems from the misidentification of organization
(physiology) with evolutionary diffusion (evolutionary biology) (see passim Boyd & Noble, (1993); Rosen,
(1991); Khalil, (1999)).

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
428 Recherches
Économiques - Louvain
de Louvain Economie
Review
67(4),2001

Seventh,Smith'snotionof sympathy allowsus to modelself-and

other-interest(whathe calls thevirtuesof "prudence"and "beneficence")
as lyingalonga continuum. In thisregard,
Smith'sconceptionis congruent,
as suggestedearlier,withthe modernegocentric view,wherealtruismis
regarded as a motivein theutilityfunctionthatis smoothlysubstitutable
withself-interested
motives.

2 Smith's Critique ofThree Theories ofAltruismof

His Time

Although he referred to thembydifferent names,Smithexplicitly and em-

phaticallycensured whatis calledhereAxelrod'segoistic, Becker'segocen-
tric,and Mead/Frank's altercentrictheoriesof altruism.Withregardto
theegoisticapproach, whichstresses as almostthesolemotive,
self-interest
Smithcastigatedits representative of his day,viz., BernardMandeville.
Smith((1976) pp. 308-313)pointedout that Mandeville 's viewcouldn't
evendistinguish selfishness fromlegitimate or whatSmithcal-
self-interest
led the "virtueofprudence."WhileMandevilleequatedself-interest with
vice,Smithregarded thepursuitofself-interest as a legitimate, moraltask
entrusted by "Nature"in us : "The preservation and healthful stateofthe
bodyseemto be theobjectswhichNaturefirstrecommends to thecareof
everyindividual" (Ibid.,p. 212).
Moreover, Smithchidedtheexemplary oftheegocentric explanation
ofhis time,viz.,ThomasHobbes,whichself-centric, vicariouspleasureas
thesolemotive. Although Hobbesdidnotdevelopfully histheory ofpolitical
psychology, it entailedthattheact ofaltruismis aimedat enhancing the
donor'sutility byimagining theconditions ofthe as
recipient happening to
thedonor'sownstation.A famousvignette toldbyJohnAubreyillustrates
Hobbes'view:
Onetime,I remember, goeingintheStrand, a poorandinfirme oldman
cravedhis(Hobbes')almes.He,beholding himwitheiesofpittyandcom-
passion,putthishandinhispocket, andgavehim6d.Sayda divine(seil.
Dr. JasparMayne)thatstoodby-'Would youhavedonnethis,ifit had
notbeenChrist's command? '-'Yea',sayd he.-'Why?'quoththeother.-
'Because',saydhe,'I wasin paineto consider themiserable conditionof
theoldman;andnowmyalmes, givinghimsomereliefe,dothalsoeaseme'
(Aubrey (1898)p. 352).
Thatis,thereasonbehindthedonation is thealleviationofthedonor's
discomfortat thesightofa beggar.The improvement ofthebeggar'sincome
is the donor'sattemptto enhancehis vicariouspleasure.In responseto
Hobbes'egocentricism, Smithstressedthat the sympathetic act involves
putting one's selfin the other's stationrather than judgingthe other's
pleasurefrom one'sown station.
Sympathy would be a "selfish
sympathy"

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
 Elias L. Khalil 429

ifit entailsimaginingthe other'sconditionsas happeningto one's station,

i.e., one's "ownpersonand character":
Sympathy ... cannot,in anysense,be regardedas a selfish principle. When
I sympathize withyoursorrowor yourindignation, it maybe pretended,
indeed,thatmyemotionis foundedin self-love, becauseit arisesfrombrin-
gingyourcase hometo myself, fromputtingmyself in yoursituation,and
henceconceiving whatI shouldfeelin the likecircumstances. But though
sympathy is veryproperly said to arisefromimaginary changeofsituations
withthepersonprincipally concerned, yetthisimaginary changeis notsup-
posedto happento me in myownpersonand character, but in thatofthe
personwithwhomI sympathize. WhenI condolewithyou forthe loss of
yourson,in orderto enterintoyourgriefI do notconsiderwhatI, a person
ofsuchcharacter and profession, shouldsuffer,ifI had a son,and ifthatson
was unfortunately to die: butI considerwhatI shouldsuffer ifI was really
you,and I not onlychangecircumstances withyou,but I changepersons
and characters. My grief, therefore, uponyouraccount,and not
is entirely
in theleastuponmyown.It is not,therefore, in theleastselfish.Howcan
thatbe regardedas a selfishpassion,whichdoes not ariseevenfromthe
imagination ofanythingthathas befallen, or thatrelatesto myself, in my
ownproperpersonandcharacter, butwhichis entirely occupiedaboutwhat
relatesto you? A manmaysympathize witha womanin child-bed ; though
it is impossible thathe shouldconceivehimself as sufferingherpainsin his
ownproperpersonand character(Smith(1976) p. 317).
In addition,Smithattackedthe altercentric positionof his age, which
appeals to moral diet as
urns, expressed in the moral philosophyof Francis
Hutchesonand otherfiguresof the ScottishEnlightenment. Smithconside-
red themas leftoversfromthe medievalphilosophyof the Christianchurch.
Smith ((1976) pp. 139-140,300-304) distanced himselffromthe "whining
and melancholymoralists"such as Hutchesonwho identified virtueexclusi-
velywithaltruism.3This identification impliesthat self-interest is basically
selfishand amoral-a positionwhichis ironically,as Smithnotes,similarto
Mandeville's-whilevirtueis basicallyabout the adoptionof the interestof
othersor the communityat large. For such an altercentricview, thereis
little room leftforself-interest. The negationof self-interest occurs when
the self transportsitselftotally fromits station to the station of others.
This means that one adopts the needs of otherswhiledismissesone's own
needs and wants as illegitimate,immoral,and selfish.
Such an altercentric stand impliesthat altruismstemsfromstrictmo-
ral dictumsnot different fromhonestyor obligatorycommitment in general.
Smith ((1976) pp. 78-82) explicitlycriticizedsuch an implicationwhen he
distinguishedbetweensympathy,the originof altruismor what he called
"beneficence",and honesty,which stems fromwhat he called "justice".4
3 Such identification is to some extent echoed in recent
philosophical literature. For example, Sagoff (1986)
opposes "self-interest" and "public values" on the basis that the later are reflective. Likewise, Postema (1 987)
contrasts "personal values" and "collective values." Goodin (1980) argues that cost-benefit calculations
should be totally differentiated from ethical considerations. Likewise, Walzer (1983) distinguishes ethical
tastes (what he calls "obligations") on the basis of being "sacred" fromeconomic sphere of efficientrationality.
Therefore, itis surprising that Geoffrey Brennan and Lomasky (1 985) appeal to Smith's concept of sympathy
to argue that voting behavior is best understood as stemming from obligatory commitment rather than from

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
430 Recherches
Économiques - Louvain
de Louvain Economie
Review
67(4),2001

The conflation ofthetwois thecoreoftheconception ofaltruism as sub-

ject to rules notdifferent
from the rulesof Smith
justice. ((1976) partVII)
dedicatedthelast partofThe TheoryofMoralSentiments to thecritique
of medievalmoralphilosophers, whomhe calledthe "casuists",fortrea-
tingvoluntary actionsuchas altruismas subjectto strictrulesofjustice.
Casuistsactedas prigsor pedants,according to Smith.Theyfailedto ob-
servethe arenaof beneficence as separatefromobligatory commitments.
For Smith((1976) pp. 78-82),beneficence, unlikejusticeor rulesagainst
cheating, is similarto prudenceand,hence,is subjectto evaluation in light
ofcircumstances.

3 The Modern Reincarnationofthe Three Theories

Insofaras sympathy is thespringofaltruism, it allowsagentsto act notac-

cording to strategiccalculation butratheraccording to theconcern overthe
welfare ofothers-contrary to theegoisticview.It also allowsagentsto act
notoutofself-indulgence in vicariously experiencing thepleasureofothers
but ratherout ofgenuineconcern-contrary to theegocentric approach.In
addition, it allowsagents to act not to
according obligatory behavior butra-
theroutoffamiliarly andrelative circumstances-contrary to the altercentric
view.
Smith'sviewchallenges thethreeapproaches represented byAxelrod,
Becker,and Mead/Etzioni/Frank/ Simonin thesamewaysSmithconfron-
tedtheirparallelrepresentations ofhistime.Concerning themodernegois-
tic perspective, epitomized by Axelrod's"tit-for-tat" strategy, it was not
designedoriginally to explainaltruism, but ratherto explaincooperative
behaviorwithregardto non-cheating strategies.Insofaras cooperative be-
havioris notconfused withaltruism, Axelrod'segoisticapproachhas many
insights to learnfrom.However, insofaras altruismappearsto be a non-
selfishact, as evidentially the case in single-spot encounters, the egoistic
approachis problematic ifit extendsitselfto explainsuchaltruistic acts.
By extending itselfto the phenomenon of altruism, the egoisticap-
proachmaintains thatagentsassistothersstrategically, i.e.,to inducethem
to reciprocate favors.That is, all actionsare motivated ultimately byself-
interest.As such,it suffers fromthesame shortcoming thatSmithfound
in Mandeville'sapproach.It cannotdifferentiate betweenselfishness and
legitimate self-interest.
Concerning Becker'segocentric perspective,it maintains thatbene-
factorsincreasetheirutilityby imagining howrecipients are enjoying the

self-interest. While voting behavior may stem from obligatory commitment, Smith's principle of sympathy
cannot account for such a commitment. While Smith did not discuss explicitly obligatory commitment, it
might be derived from his discussion of the virtue of justice that he explicitly distanced from sympathy
(Khalil (1998)).

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
Elias L. Khalil 431

donatedgoods.Thisegocentric explanation canaccountforsingle-spot tran-

sactions,wheretheegoisticapproachfails.In fact,theegocentric explana-
tion,as mentioned earlier,resembles, formally, Smith'sexplanation.Ho-
wever,theegocentric explanation basicallycannotdistinguish benefactors
frommasochists whoenduredisutility whentheydonateresources in the
of an
hope attaining offsetting levelof arousal. In other words, maso-
the
chist, similar to any rational agent, suffersfrom pain (disutility) whenhe
donatesresources. But he expectsto be compensated bya greater vicarious
pleasure, experienced fromhisownstation,whenhewatchesthepleasureof
therecipient. Becker'smodelentailsthataltruists shouldfeeljoyfuloverthe
prospectofthemiseries ofothersbecausesuchmiseries occasionforthem
theopportunity to be aroused.In fact,Beckerrecognizes thisstrangeimpli-
cation,notedearlierbyHobbes,whenhe approvingly quotesthestatement
madebya beneficiary to hisbenefactors in CharlesDickens 's BleakHouse:
It'sonlyyou,thegenerous creatures,whom I envy. ... I envyyouyourpower
ofdoingwhatyoudo. It is whatI shouldrevelin,myself. I don'tfeelany
vulgargratitude to you.I almostfeelas ifyououghtto be grateful to me,
forgiving youtheopportunity ofenjoying theluxury ofgenerosity. ... I may
havecomeintotheworld expressly forthepurpose ofincreasing yourstock
ofhappiness. I mayhavebeenborntobe a benefactor toyoubysometimes
giving youanopportunity ofassisting meinmylittleperplexities (citedby
Becker (1981)p. 13,n. 2).
The inability to distinguish themasochist fromthealtruistis a direct
outcomeofthewayBeckermodelscharity orcaring,whichSmithhas found
deficient in Hobbes'theory. Sureenough,thereare manyacts ofresource
sharing that are motivated byself-centered indulgences thatagentsseekto
satisfy.However, hardlyanythinker wouldnotbe disturbed bytheprospect
ofidentifying all actsofaltruism as merely variantsofmasochism.
Concerning Mead/Frank's altercentric view,it avoidsthe failings of
theegoisticand theegocentric perspectives : The altercentric viewcan ac-
countforresource-sharing wheretheagentis involvedin single-spot tran-
sactionsand does notstandto extractvicarioussatisfaction. Thisis possi-
ble becausethealtercentric viewregardsaltruism as stemming frommoral
dictumsand,hence,can accountforsingle-spot transfers stemming out of
genuineconcerns. However, thealtercentric viewimpliesthataltruism does
notdiffer fromstrictrulesofjustice- suchas obligatory commitments not
to cheatorviolatetheproperty rightsofothers.To be accurate,it mightbe
thecase thatmanyinstances ofresource-sharing stemfromfairness and the
desirenotto appearas a "free-rider." This desireexplainsvolunteering to
support the local firedepartment, with
complying pollution-control custo-
maryrules,or supporting commonly sharedresources. However, thereis a
domainofresource sharingthatdoesnotarisefrom obligatory commitment.
Insofaras thealtercentric approachover-extends itselfto explainaltruism,
it raisessomequestions.
Forinstance,thecriticalfactorin altruismis sympathy, whichplays
a littlerolein thealtercentric view.Smith'sprinciple ofsympathy signifies

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
432 Recherches
Économiques - Louvain
de Louvain Economie
Review
67(4),2001

Egoism Egocentricism Altercentricism Altruism

Resource Resource Resource Resource
sharing sharing sharing sharing
ensures captures expresses acts on
cooperative self-centered canons sympathetic
responses sentiments ofmorality sentiments
Figure 2 : DomainsofResource-Sharing

thattheconcern overother-interest arisesfromparticular stationswitching

and,hence,voluntary, whilemoraldictums(or "moralgene"according to
biologicalversionsof the altercentricview(Frank(1988))) dictateaction
withusuallylittleregardto theparticular circumstances or personsinvol-
ved.Ifone advancestheprinciple ofsympathy to explainvoluntaryactsof
altruismà la Smith,onewouldsparehimself theneedto introducetheques-
tionofcommitment via Sen ((1977),(1985); cf.Khalil(1999))andFrankfurt
In
(1971). fact, one would notneedto invokethemultiple-self framework à
la Etzioni((1986);see Sen, (1980/1981))as well.Put differently,as Smith
notedin hiscritiqueofHucheson, thereare manyacts ofresourcesharing
whichdo not fallunderstrictrulesofjustice.Such acts are rathernon-
obligatoryand contingent on particular circumstances.

4 Conclusion

Monroe((1994),(1996),chs.6-9)reviews differenttheories ofaltruismalong

disciplinarylinesseparating sociology,economics, biology, and psychology.
In comparison, thispaperprovesthatwhatmattersis theconceptual core,
whichtranscends the somewhatartificial divisionsamongdisciplines and
evenlinguistic and intellectualapparatuses separatedby centuries.
The paperfindsthreemajortheories ofaltruism thatcut acrossthe
socialsciencesand intellectual milieus: theegoistic,egocentric, and alter-
centric The paperarguesthatneither
perspectives. Axelrod'segoisticview,
Becker'segocentric approach, norMead/Frank's altercentric perspectiveco-
versaltruisticresource sharing as understood by Smith. The receivedthree
theoriesare ratheraboutstrategic, masochist, or obligatory resourcesha-
ring.As summarized in Figure1, thethreemoderntheories do notexplain
altruism as stemming fromgenuinesentiments aboutthewelfare ofothers.
The egoisticapproachexplainsinsteadreciprocity as a strategicact to en-
hancefuturebenefit in infinitelyrepeatedgames.The egocentric viewre-
gardsresource-sharing as no from
different a utility-arousing masochist act:
theactorinflicts painon himself(donatesresources) to allow hisown person
- i.e.,withoutswitching stations- to experience enjoyment via thestimu-
lus ofwatching therecipient's pleasure.The altercentric agendais actually

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
EliasL. Khalil 433

aboutfixedcanonsand,hence,cannotexpressthevoluntary andvariedcha-
racterofaltruism. Insofaras altruismis a non-strategic, non-masochistic,
and non-obligatory action,thesetheories are generally deficient at firstap-
proximation.
Smith'stheoryofsympathy offersa betterstarting pointto unders-
tandnon-strategic, and
non-masochist, non-obligatory transfers ofresour-
ces. Ratherthanstartingwiththe egoist,ego-centered, or alter-centered
agent,Smithcommences witha normalagentwhois capableoftruesym-
pathyin thesensethathe can distancehimself fromhisownstation,but
withouttotallydisregarding hisownself-interest. Suchan agentmayshare
hisresources withothersevenin single-spot transactions, whichtheegoist
wouldnot.Suchan agentmay"empathize" withothersbytransferring him-
selfto theirstation,whichtheego-centered wouldnot.Andsuchan agent
maymakehisgrantcontingent onhissympathy andcost-benefit calculation,
whichthealter-centered prig would not.
Put in simpleterms,Smith'sapproachshowsthatthealtruistic sense
ofresourcesharingis neithermadefromthe firststationof the actor,as
maintained bytheegoisticand egocentric approaches, norfromthesecond
stationoftherecipient, as postulatedbythealtercentric approach.Rather,
followingSmith,thedecisionmakeris situatedexternally to theactorand
therecipient, viz.,in an imagined thirdstation. For thetheorist todetermine
theextentofassistance, thetheorist needsto specify theextentoffamilia-
rity.Moreover, thetheorist needsto specify thebenefactor's lossrelativeto
thebeneficiary's valuationoftheassistance.
AlthoughSmithdid notdiscussthequestionofcomparative utility,
he brought to ourattention thecentrality ofstationswitching andjudging
froma distance(i.e., sympathy) forthe understanding of altruism.One
implicationis that human societyis notheld together solelyon theground
of self-interestedand self-indulging passions.Also, for humansocietyto
subsist,thereis no needforan authority empowered by externalsystem
an
ofethics.Humanfraternity arisesfromtheevery-day interaction ofagents
thatnurtures familiarity. What is uniqueabout Smith'sapproachis that
the principle, whichoriginates fraternity,i.e., sympathy, is the same one
thatmakesthe agentinterested in promoting his ownself-interest. In his
critiqueofMandeville, Smithwasexplicitthatthepursuitofself-interest is
notvicebutrathera virtue.

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
434 Recherches
Économiques - Louvain
de Louvain Economie
Review
67(4),2001

References

Aubrey, J. (1898),BriefLives,chiefly ofcontemporaries, set downbyJohn

Aubrey, between the Years 1669 & 1696, vol. 1 (A-H), ed. Andrew
Clark,Oxford, ClarendonPress.
Axelrod,R. (1984),TheEvolutionofCooperation, NewYork,Basic Books.
Becker,G.S. (1976), "Altruism, Egoism,and GeneticFitness: Economics
and Sociobiology" , Journalof EconomicLiterature, 4, pp. 817-26.
(Reprinted in Zamagni,1995.)
Becker,G.S. (1981),"Altruism in theFamilyand Selfishness in theMarket
Place",Economica, 48,pp. 1-15.(Reprinted in Zamagni,1995.)
Bergstrom, T.C. (1995), "On theEvolutionofAltruistic EthicalRulesfor
Siblings", American Economic Review,85,pp. 58-81.
Boyd,C.A.R. and D. Noble(eds) (1993), TheLogicofLife: The Challenge
ofIntegrative Physiology, Oxford, OxfordUniversity Press.
Brennan, G. and L. Lomasky(1985), "The ImpartialSpectatorGoes to
Washington : Toward a Smithian TheoryofElectoralBehavior" , Eco-
nomicsand Philosophy, 1, pp. 189-211.
Collard,D. (1978),Altruism and Economy : A Studyin Non-Selfish Econo-
mics,Oxford, Martin Robertson.
Etzioni,A. (1986), "The Case fora Multiple-Utility Conception", Econo-
micsand Philosophy, 2, pp. 159-183.
Frank,R.H. (1988), Passions WithinReason:The StrategicRole of the
Emotions, NewYork,W.W. Norton.
Frankfurt, (1971),"Freedom
H.M. oftheWillandtheConceptofa Person",
Philosophical Review, 68, pp. 5-20.
Getty,G. (1989),"TheHuntforr : One-Factor and Transfer Theories",So-
cial ScienceInformation, 28,pp. 385-428.
Goodin,R.E. (1980), "MakingMoralIncentives Pay", PolicySciences,12,
pp. 131-145.
Haakonssen, K. (1981), TheScienceofa Legislator:TheNaturalJurispru-
denceofDavidHumanandAdamSmith, Cambridge University Press.
Hamilton, W.D. (1964), "The GeneticalEvolutionof Social Behavior",I,
II, Journalof Theoretical Biology, 7, pp. 1-16,17-52.
Khalil,E.L. (1990),"BeyondSelf-Interest and Altruism: A Reconstruction
ofAdamSmith'sTheoryofHumanConduct",Economicsand Philo-
sophy, 6, pp. 255-273.(Reprinted in Zamagni,1995.)
Khalil,E.L. (1996), "Respect,Admiration, Aggrandizement : AdamSmith
as EconomicPsychologist", Journalof EconomicPsychology, 17,
pp. 555-577.
Khalil,E.L. (1998), "Is Justicethe PrimaryFeatureofthe State? Adam
Smith'sCritiqueofSocialContract Theory" , EuropeanJournal ofLaw
and Economics, 6, pp. 215-230.

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions
Elias L. Khalil 435

Khalil,E.L. (1999),"Sentimental Fools: A CritiqueofAmartya Sen'sNotion

of Commitment" , JournalofEconomicBehaviorand Organization,
40, pp. 373-386.
(URL : www.sciencedirect.com/science/journal/01672681)
Khalil,E.L. (2000), "BeyondNaturalSelectionand DivineIntervention:
The Lamarckian Implication ofAdamSmith'sInvisibleHand", Jour-
nal ofEvolutionary Economics,10,pp. 373-393.
Mackie,J.(1980),Hume'sMoralTheory, London,Routledge & KeganPaul.
H.
Margolis, (1982),Selfishness, Altruism and : A
Rationality TheoryofSo-
cial Choice,Cambridge University Press.
Mead,G.H. (1934),Mind,SelfandSociety, editedbyC.W. Morris, Chicago,
University ofChicagoPress.
Monroe,K.R. (1994),"A Fat LadyinA Corset: Altruism inSocialTheory" ,
AmericanJournalofPoliticalScience,38, pp. 861-893.
Monroe,K.R. (1996), The Heartof Altruism : Perceptions of a Common
Humanity, Princeton,NJ,Princeton University Press.
Postema,G.J. (1987), "Collective Evils,Harms,and theLaw" Ethics,97,
pp. 414-440.
Rosen,R. (1991), Life Itself: A Comprehensive Inquiryinto theNature,
Origin,andFabrication ofLife,NewYork,ColumbiaUniversity Press.
Sagoff,M. (1986), "Valuesand Preferences", Ethics,96, pp. 301-316.
Sen,A.K. (1977), "RationalFools : A Critique theBehavioral
of Foundations
ofEconomicTheory",Philosophy & PublicAffairs, 6, pp. 317-344.
Sen,A.K. (1980/1981),"PluralUtility", Proceedings oftheAristotelian So-
ciety,80, pp. 193-215.
Sen, A.K. (1985), "Goals,Commitment, and Identity",Journalof Law,
Economicsand Organization, 1, pp. 341-355. (Reprinted in Zamagni,
1995.)
Simon,H.A. (1990), "A Mechanism forSocial Selectionand Successful Al-
truism", Science,250,pp. 1665-1668. (Reprinted in Zamagni,1995.)
Smith,A. (1976),The Theory ofMoralSentiments, editedbyD.D. Raphael
& A.L. Macfie,Oxford, ClarendonPress.
Walzer,M. (1983),SpheresofJustice : A DefenseofPluralism andEquality,
NewYork,Basic Books.
Wilson,E.O. (1975), Sociobiology : The New Synthesis, Cambridge, MA,
HarvardUniversity Press.
Zamagni,S. (ed.) (1995), TheEconomicsofAltruism, Aldershot, UK, Ed-
wardElgar.

This content downloaded from 27.251.83.10 on Thu, 14 Nov 2013 04:32:11 AM

All use subject to JSTOR Terms and Conditions