Pleasure: 360-300 B. C.

Author(s): John M. Rist

Source: Phoenix, Vol. 28, No. 2 (Summer, 1974), pp. 167-179
Published by: Classical Association of Canada
Stable URL: http://www.jstor.org/stable/1087416 .
Accessed: 12/06/2014 11:25
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.

Classical Association of Canada is collaborating with JSTOR to digitize, preserve and extend access to
Phoenix.

http://www.jstor.org

This content downloaded from 134.48.233.114 on Thu, 12 Jun 2014 11:25:56 AM

All use subject to JSTOR Terms and Conditions

PLEASURE: 360-300 B.C.

JOHN M. RIST

STOIC AND EPICUREAN theories and attitudes about pleasure vary

considerably;all, however,are a product of the reneweddebates of the
second half of the fourthcentury,of which our firstdocumentaryevidence is presumablyPlato's Philebus.A glance at the historyofthe debate
will not only shed lighton why Epicurus and Zeno said what theysaidand much of this groundis well-trodden-but it may help us to get to
gripswitha small part of the complicatedproblemof the developmentof
the thoughtof Aristotle.This area has oftenbeen tackled partially: some
have looked at the relation between Aristotle'stheoriesof pleasure and
those of Plato; other scholars have consideredthe impact of Aristotle
on Epicurus,' and, less frequently,
on the Stoics.2Rarely has the attempt
been made to co-ordinatethese inquiries,though the argumentsof the
late fourthcenturyabout pleasureaffordus theopportunityto make such
a co-ordination.We shall approach our topic primarilythroughdescriptionsand definitions(or what look like definitions)of pleasure.
The firstadherentof pleasure in the Philebusis Philebus himself,but
his defenceapparentlydoes not rise to the level of philosophy.He may
representthe "ordinary" man,3or perhaps the views of Aristippus.But
when we get to the argumentbetween Socrates and Protarchus,somethingthat looks like an attempteddefinitionappears: the restorationof
an organ "to its own nature," apparentlyto its own properand natural
condition (EIs ri-vabrcrwov
42d 5), is pleasure (or a pleasure). This
-4oL,
natural conditionis describedby Socrates as an attunement(of what is
not clear, 31d 4). When this attunementundergoesdisruption
(Xvo~ivris),
we simultaneouslyfeel"pain"-Socrates says that thereis a coming-to-be
des(~Y}YELts) of pain. The two passages seem to offerslightlydifferent
criptionsofpleasureand pain, and presumablyPlato has notdistinguished
these descriptions (or at least has not recognized any distinctionas
important).In 31d it looks as thoughpleasure and pain are generatedas
or emptying
processesat the same time as the processesof replenishment
-the compositionor disruptionof harmony-in the organism,but even
1E.g., E. Bignone, L'Aristoteleperdutoe la formazionefilosoficadi Epicuro (Florence
1936); C. Diano, "La psicologia d'Epicuro e la teoria delle passioni," GCFI 20 (1939)
105-145; 21 (1940) 151-165; 22 (1941) 5-34; 23 (1942) 5-49, 121-150.
2E.g., A. A. Long, "Aristotle's Legacy to Stoic Ethics," BICS 15 (1968) 72-85;
R. P. Haynes, "The Theory of Pleasure of the Old Stoa," A7P 83 (1962) 412-419.
'As suggested at 66e.

167
PHOENIx, Vol. 28 (1974) 2.

This content downloaded from 134.48.233.114 on Thu, 12 Jun 2014 11:25:56 AM

All use subject to JSTOR Terms and Conditions

168

PHOENIX

in the nextsection(31e-32a) Socrates appears to be sayingthat the actual

movementof the bodilycomponentsis itselfpleasureor pain (7 Kar& 4bTwv
b6bs)or a pleasure or pain. Thus it seems to be implied that it makes no
difference
whetherwe say that pleasure (or some pleasure) whichis itself
a process, is identical with the process of replenishment,or whetherit
accompaniesreplenishment.'But the questionof whetherthisindicatesa
confusionin Plato's thinkingshould be appraised cautiously. It would
help to know whetherhe is offeringa definitionof pleasure or whether
he is merelysaying that somethingis an example of pleasure: We may
say "Swimmingis fun" whileat the same timeagreeingthat swimmingis
also a means of propellingourselves throughthe water in an enjoyable
manner. It would be strange, however, if we were inclined to define
swimmingas fun,or even as some kindof fun.But in partsof the Philebus
Plato mightbe supposed to have lapsed intosayingthingslike "Swimming
is fun" ratherthan "Swimmingis an example of fun." In 42d, it has been
claimed, it is the restoration (Karwrao-ts) of a natural state which is said

to be not pleasant but pleasure. If Socrates had said that thisrestoration

is pleasant-or, as in the Timaeus (64d) that the experienceof returning
to a natural condition,is pleasant-we should not have worried.That
would have been like saying "Swimming is pleasurable" rather than
"Swimmingis fun." But the Greek is ambiguous: could Socrates mean
"restorationis a pleasure, i.e., an example of a pleasure"? Probably so,
for he recognizespleasures that are not restorationsat 51d-e.5 But we
should notice that to say that restorationis an example of pleasure is
not merely to say that it is pleasant. For "restorationis a pleasure"
could mean either"Restorationis identicalwitha pleasure," or "Restoration is accompanied by a pleasure."
Presumablythe Philebuswas writtenin the late 350's. The nextstage
(or possiblyanotherversionof the same stage) in the historyappears to
be representedby our version of Aristotle'sRhetoric1. It is impossible
to be certainwhetherAristotleis speakingherein his own person,but at
least he takes it as a workinghypothesisthat pleasure is-and this time
it does look like a seriousdefinition-a movementof the soul and at the
same time "a simultaneousand perceptiblerestorationto the underlying
els ripvlraLpXovUav
Op6avKai alO7rOy-V
(KarraTTaULJv
'At Phlb. 53c 4-7 it is suggested that the identificationof pleasure as a 7yveotasis a
to which, by implication,Socrates is not committed.Thus it looks
theoryof the
as though theKO.,4oL
path may be open for the suggestion that only one type of pleasure is a
but whetherthe path is intended to be open or not, Plato does not go along it.
yevts;
Perhaps there is some furtherindication that in the Philebus Plato is not wholly
committedto the view of the KOJA1OL
in that when earlier,in the Republic, the identification of pleasure and pain with "some kind of motion" comes up (583c 9) Socrates proposes it unambiguously as his own view.

(or basic) natural state"

5Cf.Resp. 584b.

This content downloaded from 134.48.233.114 on Thu, 12 Jun 2014 11:25:56 AM

All use subject to JSTOR Terms and Conditions

PLEASURE: 360-300 B.C.

169

obGtv,1369 b 33-35).6 That this formulahas certainsimilaritieswith the

language of the Philebus (as well as the Republic and the Timaeus) has
not been missed. Pleasure is described as some kind of movement,but
the class of pleasures which are not restorationshas disappeared. The
movement (Kvracls) is clearly in the soul,' while the "restoration"
is said to be simultaneous and perceptible.We need not
(KarTo-aams)
spend time on "perceptible,"except that we must assume that the word
should not be pressed to referonly to sense-perception.Aristotlecan
hardly have denied that there are many pleasures which are not recognized by the particularsenses,i.e., not recognizableas sensations.(What
would he say about mental pleasures?) But pleasure (apparently all
pleasure) is now a movementofthesoul and at thesame timea restoration
of the natural state. How is this to be understood?If it is a movement
of the soul, it must accompany the restorationof the natural state; yet
ifit is definedasa restorationofthe naturalstate,it cannotbe a movement
of the soul, at least in the Platonic understandingof soul. (Of course,if
the soul is the formof the body, thenthe double description-both movement and restoration-becomes possible.) But in fact the explanation is
probably simpler: "Pleasure is the restorationof the natural state"
may be the same kind of propositionas the travel agent's "Happiness is
Europe '72."
Aristotle's account of pleasure in the Rhetoricis not identical with
Plato's in the Philebus; indeed the notion that all pleasures are restoraBut in many respects it is
tions looks like a sloppy oversimplification.
Above all, all pleasureis apparentlystillsome
not substantiallydifferent.
kind of process or at the least associated with a process. If at this time
Aristotlehad developed any more sophisticatedaccount of pleasure, or
any furtherrefiningamong differentvarieties of pleasure, we are not
aware of it. The use of the term&Bpbacdoes not seem to look forwardto
futuredevelopments,at least not to the abandoning of the notion that
pleasure is a process,and the advocacy of an alternativeview thatit may
be completein the instant; the functionof 60p6ac seems to be to indicate
that the restorationof the bodily organs is significantenough to be
noticed.There may be an echo of Timaeus 64d. The translation"intense"
is too strong;8I should prefersomethinglike "significantlyperceptible"
or "noticeable." But there is no suggestionthat the process-theoryhas
been dropped.?
WForbir&tpXwsee G. Lieberg, Die Lehre von der Lust in den Ethiken des Aristoteles
(Zetemata 19, Munich 1958) 29-30.

7Thoughthisis alreadythecase in Resp. 583c9.

*So Hardie, Aristotle'sEthical Theory(Oxford 1968) 302.

OItseemsto be one ofthefewprobablycertainbeliefsofAristippus(theSocratic),as

Mannebach says, that pleasure is a motion (Aristippi et CyrenaicorumFragmenta,ed. E.

Mannebach[Leiden 1961]). See D. L. 2.85 = Mannebach193, withnote on p. 95,

This content downloaded from 134.48.233.114 on Thu, 12 Jun 2014 11:25:56 AM

All use subject to JSTOR Terms and Conditions

170

PHOENIX

The next item in our account is representedby Book 7 of the NicomacheanEthics'o-but chronologicallythis may be nearlycontemporary
with our versionof Rhetoric1. At 1152 b 13 Aristotlementionsa theory
that all pleasure is a perceptible
process to a natural state
(alOyrO-7)
the theory
that he had at least used as a
~orTL eis k~oLv ailaorqr7),
(yOveoLs
workinghypothesisin the Rhetoric." But he now argues against the
theory,though without mentioningthe names of its supporters.What
objections does he adduce? The crucial sentence at 1152 b 33 ff. (Kar-a
V
LK'V
al KaOLUrTaaLE6~
T7lv4VO

'

LV76CiTEaL
a E0LV) seems to mean that

Av#uppl3E7K6
the
processofrestorationto a naturalstate is onlyaccidentallypleasant.12
Thus pleasure cannot now be definedas a restoration.And Aristotlegoes
further.The "activity" that is goingon when we desiresuch restoration

is the activityof an unimpaired

Thus we
partofourbodilystructure.'3
feelpleasurenotin thepartsbeing"restored,"but at therestoration
in
thepartsthatarealreadyin a naturalcondition.
The proofthatAristotle
forthislastpointis thatsomepleasures(e.g.,thatofcontemplation)
offers
do not involvepain or desire;he seemsto thinkthat thisimpliesthat
since pleasuremust be physiologically
explainablesimilarlyin all its
ifsomepleasuresare necessarily
manifestations,
onlyfeltin thoseparts
of thebodythatare devoidofpain,desire,and deficiency,
therefore
all
pleasuresmustbe feltin such parts. (This of courseimpliesthat the
nature of pleasure can be best determinednot fromthe "bodily"
but fromthepainlesspleasuresofthemind.)In
pleasuresofrestoration,
S'
Lava&&okvPfv. Mannebach's view is
adreaLatve7HtvXeIav KLVrlOLV els al'OaqvO
rkXos
endorsed
by Guthrie, HGP 3 (Cambridge 1969) 494. I suspect that Cyrenaic talk of
pleasure KaTa&
KIzV7aL (as Mannebach 195 = Ath. 12.546e) is of later date (i.e., of the
younger Aristippus or other Cyrenaics). This would follow if, as I argue below, the
KmLzVV language derives fromAristotle.
Kar shall refer
to this book as Book 7 of the Nicomachean Ethics, though it is also
10I
generally treated as Book 6 of the Eudemian Ethics. I shall only touch on the general
question of the relation between Eth. Eud. and Eth. Nic. whereit concernsthe discussion
of pleasure. A recent (and sane) account of the problem and its history is given by
C. G. Rowe, "The Eudemian and Nicomachean Ethics: A study in the development of
Aristotle's thought," PCPhS, Supplement 3 (Cambridge 1971). I shall referto Book 7's
discussion of pleasure as (A) and Book 10's as (B), followingFestugikre.
"Notice that the theoryin the Philebus and in (A) identifiespleasure as a
not
Yater,recurs
a KWlvrl7s.
In his Republic (583e 10) Plato suggested that it is a
Kivr)at.
Klvriat
in the same sense in the Rhetoric(1369 b 33), in (B) (1174 a 19), and, as
we shall see,
with the Stoics.
"We should notice that the wordKr&xTaracs (used in the Philebus and Rhetoric)does
not occur in (A); v
and -Tyeots do. And, as we shall see, it is C'
not a
KPrVELt
KY?V7l,
form of
that passes over to Epicurus. Of course the verbal
cognate of
KaT&aTaaLS,
is found at 1152b 34. Curiously enough,
(and
s)
Kar&TaUa
Kr&a4TrLs
&7roKaraTR7r
recur
in the Magna Moralia (1205 a 4, 1205 b 6, 1205
b 12.)
'"This is noticed by G. E. L. Owen in his helpful "Aristotelian Pleasures," Proc.
Atrist.Soc. (1972) 144.

This content downloaded from 134.48.233.114 on Thu, 12 Jun 2014 11:25:56 AM

All use subject to JSTOR Terms and Conditions

PLEASURE: 360-300B.C.

171

the processof restoration,therefore,

pleasure will arise accidentally(and
presumablysimultaneously,as in the Rhetoric).Notice, however, that
pleasure itselfis not an "accident," thoughit arises "accidentally." The
processesof restorationmay happen to producepleasure,but theyare no
guide to what the natureof pleasure is. (It may be an accident of this
table to be in my room,but beingin my room tells me nothingabout the
natureof the table.) Aristotlethen goes on to show how what is "naturally pleasant" (Obat'rj6) or "absolutely pleasant" can be distinguished
fromwhat is "accidentally pleasant." Things accidentallypleasant only
seem pleasant when some deficiencyfroma healthyor normalcondition
is being experienced.
Aristotlenext turnsto the natureof pleasure itself.It is not a process,
nor does it necessarilyoccur in connectionwith a process. Rather it is
an end and an activity.Here Aristotleis using his own technicaldistinction betweena process and an activity,a distinctionin accordance with
which he distinguishesan activityas "complete" in the instant,even if
also extendedover a considerablelengthof time. Thus if "I see" entails
(or is compatiblewith) "I have seen," seeingis an activity,not a process.'4
In this sense pleasure too is an activityand completein itself.Hence the
definition(proposedin theRhetoric)ofpleasure as a consciousprocesshas
to be rejected. Rather pleasure is an unimpeded activityof the natural
state, that is, of those parts of the organismwhich are in theirnatural
condition.
The most importantand most misunderstoodsection of the narrative
of (A) beginsat 1153 b 7. The firstargumentrunsas follows:just as some
kind of knowledgemay be the best thing(7&ptTro') even if some formsof
knowledgeare bad, so the fact that some pleasures are bad in no way
lessens the possibilitythat some pleasure may be the best thing.Aristotle
then goes on to argue that since some kind of unimpededactivity(which
is pleasure) is the thingmost to be desired,so some kind of pleasure (i.e.,
unimpeded activity) will be the thingmost to be desired.Yet this argument does not show that pleasure is to be definedas the supremegood,
but that some unimpeded activity, which is pleasure, but not only
pleasure (for theremust also be the activityof the facultyqua faculty,
unless we have facultieswhichdo nothingelse than generatepleasure) is
the best thing (or supremegood, if you will). But even if the argument
only shows that much, does Aristotleunderstand it to show more?"5
Clearly not, forthe purpose of the argumentis indicated at 1153 b 25 ff.
"Cf. J. L. Ackrill,"Aristotle's Distinction Between ENERGEIA and KINESIS,"
New Essays On Plato and Aristotle,ed. R. Bambrough (London 1965) 121-141.
"1As (e.g.) P. Merlan observes (Studies in Epicurus and Aristotle[Wiesbaden 1960]
28). Aspasius was the firstto suggest that an identificationof pleasure and the good
mightindicate that (A) is the work of Eudoxus (CAG 191,p. 151, 24 ff.Heylbut).

This content downloaded from 134.48.233.114 on Thu, 12 Jun 2014 11:25:56 AM

All use subject to JSTOR Terms and Conditions

172

PHOENIX

Here the fact that all beasts and men pursue pleasure is held to be an
indicationthat in some sense (iros) pleasure is the best thing."1The "in
some sense" cannot be neglected; and it fitsour interpretationof the
previous section. "The best thing is in some sense pleasure," seems to
mean that the best thingis pleasurable, or even, the most pleasurable,
not that the best thingis to be simplyidentifiedwith pleasure.
What we have then is not an identificationof pleasure as the supreme
good, but a statement about the principal object of desire which is a
pleasure.But, an objectionmightrun,all beasts and menpursuepleasure,
but it is not the same pleasure whichtheyall pursue. Aristotle,however,
will not have this-and his commentson it have rightlybeen seen as a
particularlystrikingfeatureof the discussionin (A). Perhaps,he explains,
it is reallythe same pleasure that is pursued,even if the pursuersdo not
recognizethe fact. But why should this be so? Because beasts and men
share somethingdivine by nature, and-Aristotle apparentlywishes to
say-this fact is indicated in the universal search fora pleasure, which
must thereforebe an identical pleasure.
For our own purposesthereare a fewspecificpointsin the later part of
(A) to whichwe must draw particularattention: the notion that a man
can be happy on the wheel is rejected at 1153 b 19; and it is argued at
1154 b 6 ff.,that thereare people so constitutedthat forthemthereis no
neutralstate betweenpleasure and pain. Hence both absence of pleasure
and positive pain are to them painful. Thus pain would appear to be
definable as an absence of pleasure. Finally the whole discussion of
pleasure in (A) ends witha theorythat,since our natureis not "simple,"
we cannot enjoy a singlepermanentpleasure as God can. Our pleasures,
because of some weakness
TL), come and go. Hence we cannot
(romptavw which God
enjoy that pleasure in rest
enjoys; we need also
(bv 7)pei4l)
pleasure in movement (bv Kwi(fTe). Both pleasures are, of course, still
activitiesin the Aristoteliansense, forthereis an activityof movementas
well as an activityof immobility.17
From this briefaccount of some of the relevantparts of (A) let us now
extracta seriesof propositions:
1. Pleasure is not a process
but may be an accompanimentof a
(ybvEs)
in
those
or
process occurring
organs parts of organswhichdo not need
"restoration."
2. Pleasure is an activityand an end.
3. There are two kinds of pleasure, pleasure in movement
and
(KlVtIOS)
pleasure in immobility.Both are (as 2 above) activities.
'"The point was made in general by A. Barbieri, "Aristotele e l'edonismo di Eudosso,"

GCFI 33 (1954) 525.

j pj
"Cf. Eth. Eud. 1249 a 20 (ob ylverat 5~b~5ovt)

v rpde).

This content downloaded from 134.48.233.114 on Thu, 12 Jun 2014 11:25:56 AM

All use subject to JSTOR Terms and Conditions

PLEASURE: 360-300 B.C.

173

4. Pleasure is thereforecompletein the instant,thoughit may of course

have longerduration.
5. Pleasure is an unimpedednatural activity.
6. In some sense (leftunclear) pleasure is "the best thing."
7. All men and animals probablypursue the same pleasure (even though
oftenunwittingly).
To these we may add two more peripheralpoints:
8. It is nonsenseto say that a good man is happy on the wheel.
9. For some people at least, absence of pleasure is pain, i.e., thereis no
intermediatestate between pleasure and pain. But the sensible man
pursues absence of the pain (&Xvrta)that arises fromthe desire forrestorativepleasures (1153 a 31).18
The nextmaterialwe should consideris not the tenthbook oftheEthics,
(B), but thephilosophicalwritingsof Epicurus. We may noticethe following points in particular:
1. Pleasure is never identified as a restoration (Kard7uracus)or as a process
but some pleasure (the highest) is r6
capKbS KaraiLT7a. .19
2. There are two kinds of
which are
thoseebraOhs
(,yineas);

pleasures,

katastematic

(D.L. 10.136),20 and those which are Karc KlU7VV or EVKUi7El. This appears
to be the Aristotelian distinction. Even the phrase 'V KtV'7YO occurs in (A)

(1154 b 28); the wordKV77TWK?'is not used by Epicurus.21

3. All "kinetic" pleasures vary katastematic pleasures in the same
organs.22
4. Xap& and Eb&poabV'7
KaTa Kivr7atL
VEp7YELL
i.e., even kinetic
3.#XroVTrac,

pleasures are activities (D.L. 10.136).

5. Pleasure is completein the instant.23
6. Pleasure is the end of life.
7. All creaturespursue "basic" pleasure,i.e., the same pleasure (whichis
in fact absence of pain).
8. The wise man is happy on the rack (D.L. 10.118).
9. The highestpleasure is absence of pain: thereis no intermediatestate
betweenpleasure and pain.
ofRhodes
18'AXvrta in generalbecomesthegoal ofthelaterPeripatetic
Hieronymus
frs.8-18).
(F. Wehrli,Die Schuledes Aristoteles,
19Plut.Non posse,1089d(Us. 68).
in Phil., p. 274 Stallbaum= Us. 416.
2oCf.Olympiodorus
2lCf. J. M. Rist,Epicurus(Cambridge1972) 102.
22Cf. mostrecently,
J. M. Rist (above,note21) 108-111,170-172.
VS 42 (and KD 19); cf. C. Diano, "La psicologiad'Epicuroe la teoria
23Epicurus,
dellepassioni,"GCFI 21 (1940) 159.

This content downloaded from 134.48.233.114 on Thu, 12 Jun 2014 11:25:56 AM

All use subject to JSTOR Terms and Conditions

PHOENIX

174

What I need to indicate is that these nine featuresof Epicureanismall

findan analogue or point of comparisonin (A). CertainlyAristotleand
Epicurus do not agree on everydetail, thoughthereare strikingpointsof
coincidence,such as the ideas that all pleasure is activityand that there
are two specifickindsof pleasure. Aristotlementionsa view that absence
of pleasure is pain; Epicurus holds that absence of pain is pleasure.'4
Both indicatethat theproblemofa possibleintermediatestate is notdead.
Epicurus talks about whetherthe wise man is happy on therack,Aristotle
thinksof the wheel. Again the problemis the same. Our next objective
should be to considerhow many of these specificEpicurean pointsin (A)
occur in the tenthbook of the Ethics (B).
First comes the notion of pleasure as an activity, accepted in (A);
it is not mentionedin (B). A different
formulationis offered.Pleasure is
certainly still not a process (KL'VrlOL KalcTevLP, 1173 a 31 ff.); it is indeed
something "in accordance with which" we are active ('cepyETy, 1173 b 3).

It is completein the instant (1174 b 5-6) but accompanies the activities

of various bodilyor mentalorgans.However it is describedas completing
the activityofthe organ,but not qua the actual functioning
of that organ.
Rather it is a superveningend, like the bloom of health on those in the

prime of life (arLCy

r'XoS,
osov

lotpa)-and
roiT
&KosatoLs

the same

7rL-y'ELbv
be foundin Book 2.25 And thereasonwe do not continue
attitudeseemsto
than
feelingpleasure indefinitelyis also explained somewhat differently
in account (A). In (B) the somewhatsimplisticremarkthat we are not
simple is given more body. Pleasure depends on the state of each organ,
and tirednesswill lead to a declinein pleasure,just as it leads to a decline
in the successfuluse of the organ itself(1175 a 4-10).
In brief,although(A), (B),26 and Epicurus agree that pleasure is not a
processand that it is therefore
completein the instant,in several respects
where(A) and Epicurus come together,(B) remainsapart. For Epicurus
and (A) pleasure is the end, or an end. (B) qualifies this: pleasure is
specificallya supervenientend. And if (A) and (B) are nearerthan they
look, the wordingof (A) is much nearerto that of Epicurus. Above all,
where (A) emphasisesthe divine in the search forpleasure and brackets
men with animals,27 thus pointingto Epicurus, (B) emphasisesthe ques24As we shall see, the Magna Moralia commentsthat a life freefrompain is "near" to
pleasure (1204 a 24).
25Cf. Eth. Nic. 2, 1104 b 4-6, 32tIEov 4 ~El 7rocElOaL 7Cov
r@wv yOe 7 PLv
L/olP
Xbrtv
T
and 1104 b 34,
L
Kal
rEaL
S,
rosl
Epyoys
roL
KOLWvT
re yd&paIryTTolS
60ovV
b~rb
r7ry alppeovLrapaKOXoovOe. We should notice that the nearest passage in the parallel
section of Eth. Eud. does not contain formsof ArTyLIyvcrOat
or irapaKoXoOEETV
(Eth. Eud.
1220 b 13-14, ols rera~ ... .7 aI0crT7TLKI)
KaO' ab-r; cf. Eth. Nic. 10,
'ovl) ' Xv7r7
1175 a 6).
the relation between (A) and (B) see now Owen (above, note 13).
260On
27Cf.Merlan (above, note 15) 20, 22.

This content downloaded from 134.48.233.114 on Thu, 12 Jun 2014 11:25:56 AM

All use subject to JSTOR Terms and Conditions

PLEASURE:

360-300 B.C.

175

tion of intellectand separates men fromanimals, thus pointingto the

Stoics.
There is one more quasi-Aristoteliandocument to which we should
brieflyturn: the Magna Moralia. I do not intendto go into the question
of authorship,but merelyto express an opinion against the view that
this is an Aristoteliandocument.28 But clearly it is at least heavily indebted to Aristotle,and it may be helpfulto us to see on which of the
Aristotelian discussions of pleasure it chieflyrelies. Fortunately the
answeris clear; it is discussion(A). It is true that the most significantof
all the special proposalsof (A) is omitted: the proposal that the factthat
men and animals perhaps pursue a single pleasure is to be explained by
their common possession of "somethingdivine." But this must be set
against a varietyof otherconsiderations.Pleasure in the Magna Moralia
is not a process, not a restoration(the formulationof this at 1205 b 7,
with the phrase rvvabro3, looks back to the Philebus,42d 5); there are
pleasures of restorationand pleasures of the restoredstate, but both are
activities(1205 b 21). There is indeed an argumentabout the claim (but
perhapsof theRhetoric[1362 b 5]) that all thingsdesirepleasure; but the
suggestionis rejected that it is neverthelessthe same pleasure (1204 a
38 ff.,1205 b 9 ff.).And we have already noticed how forthe author of
the Magna Moralia absence of pain is near to pleasure.29 The only
phraseologywhichseems to suggest(B) (and Eth. Nic. 2) ratherthan (A)
is J YdapXv'rtl rtrolTs
W' VanyKl7V
(1206a 15); the~irtperhapsreminds
OrTLv
us of the 7rvvLyv6EP6v6
rL r'XoSof 1174 b 36. But furtherdetails are unnecessary.No one can (or does) deny that the Magna Moralia is closerto
(A) than to (B).
Let us now look at the writingof the older Stoics. Pleasure is not their
most frequentsubject of discussion,but a fewpassages are informative.
The Stoics,or some of them,distinguish'boval(pleasuresof a "perverted"
reason) fromxapal (emotional states associated with a healthyreason),
but forour presentinvestigationsthis is unimportant:what mattersis
that both lbovaLand Xapal are by-products(brLyevPv1para).
That this
looks somethinglike Aristotle'saccount in (B)-or Book 2 of the Ethicshas been noticed,s0but the question has not been considered closely
enough. For the Stoics thereis no indicationthat pleasure is a -rXos,so
"2For some effectivecomments on Dirlmeier's more recent idea that the Magna
Moralia is authentic see D. J. Allan, 7HS 77 (1957) 7-11, and his review of Dirlmeier's
Eudemische Ethik (Gnomon38 [1966] 142-144).
"2Does this echo (A), 1153 a 31, or is there an influenceof the later Peripatetic Hieronymus, or of Epicurus himself?
0oD.L. 7.86 (= SVF 3.178) forJbjovj; 7.94 (= SVF 3.76) forXap& and ebLpooavrq.
Cf. Haynes (above, note 2) 414; Long (above, note 2) 80; Rist, Stoic Philosophy (Camsee especially Stoic Philosophy 48, note 5 (with
bridge 1969) 46. For e'rLtye-vv-PCara
Marcus Aurelius 6.36).

This content downloaded from 134.48.233.114 on Thu, 12 Jun 2014 11:25:56 AM

All use subject to JSTOR Terms and Conditions

176

PHOENIX

that here at least Aristotleis at best silentlycorrected.More interesting,

however, is the question of whetherpleasure is "natural." Aristotle's
account (A), ofcourse,assumesthatit is, as does Epicurus. But the Stoics,
when theymake theirdistinctionbetweenthe pleasuresof a healthyand
a perverted"reason" are in doubt. 'HSovi is a pathos,and a pathosis "an
irrational and unnatural movementof the soul" (D.L. 7.110).11 The
definitionas a movementseems to suggesta neglectof the long tradition
of analysis of the relation between pleasure and movement fromthe
Philebuson-or a totallydifferent
approach whichdoes withouta bodysoul dichotomy.32The Stoics may have gone back to Plato's Laws,"3
but handled him in a novel manner.
But in one importantrespect the Stoics did not go back to Plato;
they were liable to call pleasure (or at least jbov') unnatural: "an irrarational and unnatural movementof the soul," reportsDiogenes.34We
must try to inspect the theories of the earliest Stoic leaders in more
detail: forZeno, we have no usefulevidence,but Cleanthesseems to have
thought that even apparently innocuous (and perhaps unavoidable)
pleasures are non-natural (Arj7 Kar&
aIwoLt). Cleanthes holds the "Aris-

totelian" view ofpleasureas an aLrvyC~rvva,

but his explanationofit is that
pleasureis thus like a cosmetic."5We noticethat he does not claim that it

is unnatural (7rap&4b~rv): his view that it is non-natural is presumably a

gloss on its being a by-product.We notice that the Stoics apparently

used the word 80ov7in different
senses: sometimes,as we have seen, it is
opposed to xapL, as "irrational" pleasure; at other times it is a general
eitherto "rational" or to "irrational" pleasure. It must be
termreferring
this latter sense which is employed by Cleanthes. Doubtless he thinks
some pleasures are vicious-and hence unnatural. The rest are nonnatural.Is therethenno "natural" pleasure?The evidencewe have would
suggest not. Perhaps no "by-product" is natural; and xap& is a "byproduct," as we have seen. Pleasures are eitherunnaturaland associated
with an unhealthy"reason," or non-naturaland associated with a sound
"reason." This curious approach is correctedby Archedemuswho proposed that all pleasureis natural (Kara&
'4voL), and by Panaetius who took
the obvious way out and said that some pleasure is natural (KarT&bLtv),
the rest unnatural (7rapd&
avba).36 We need not spend time, however,on
"'Cf. D. L. 7.114 (= SVF 3.400).
"*Cf. Rist (above, note 30) 22-36.
"3Laws 897a makes pleasure and pain movements of the soul.
34The Stoic descriptionof pleasure as a KlmV7os,ratherthan a T'veoLs, remindsus of
Aristotle's account (B) rather than (A) (see note 11 above), though again we may be
going right back to Plato. But if to Plato, then to the Republic (583e 10), not the

Philebus.
"*Sext.Emp. Math. 11.73 = SVF 3.155.
"*Ibid.

This content downloaded from 134.48.233.114 on Thu, 12 Jun 2014 11:25:56 AM

All use subject to JSTOR Terms and Conditions

PLEASURE: 360-300 B.C.

177

Stoicarguments;
theinternal
ononepointall theStoicsagree:pleasure
is at besta by-product,
a cosmetic
forCleanthes,
likehairunderthe
for
Archedemus.
More
K
is thephraseTrXLKY
armpits
important
,yaOv,
Wedonotknowwhosephrase
itwas.Again,
appliedtoxapdbyStobaeus.
it
be
an
of
echo
a
one."'Butagain,
Aristotle,
perhaps, may
thoughdistant
notan echoof(A).
certainly
Let us tryto pull a fewthreadstogether.
on pleasure
Discussion
flaresup againin the late 350's,and Plato's Philebusis a witness.
Aristotle's
Rhetoric
and(A) indicate
further
(andpossibly
contemporaneof
this
and
intowhichthe
ous)fragments discussion,38 thisis thecontext
workofEpicurus
fits.Oddlyenough,
theMagnaMoraliatoohassimilaritieswiththissetofwritings,
thisprobably
though
onlymeansthat
sourceforthediscussion
ofpleasure.
Ontheotherhand
(A) isitsprimary
on
(B) marksa different
phase-a phaseofwhichtheStoiccomments
seem
to
be
a
distant
echo.WheretheStoicsgobackbehind(B)
pleasure
tobelooking
attheoriginal
ofPlato'sPhilebus,
themes
theyseemperhaps
orLaws,or evenRepublic,
rather
thanto otherAristotelian
discussion.
WhileEpicurusrepresents
whichbeganin the
partof theargument
seemtobe ofan agewhere
theissueis nolonger
350's,theStoicwritings

live.
Is it possible to put dates to the Epicurean and Stoic discussions?
Epicurus arrivedin Athensin 323 at the age of 18. Althoughnot without
previous exposure to philosophy, he probably firstmet the ideas of
Aristotle,howeverbriefly,at this time. Demetriusof Magnesia, however,
only tells us that he listened to the Platonist Xenocrates (D.L. 10.13).
But in any case Aristotledied in the next year,322, so that both (A) and
(B) musthave existedat the timeof Epicurus' visit,iftheyare Aristotelian
at all. But of course we do not know in what formthey existed.
Diiring has proposed a relative chronologyof Aristotle'sworkswhich
makes a good deal of sense.39He puts an originalversionofRhetoric1 and
2 between 360 and 355, and the Eudemian Ethics, togetherwith the
Philebus,between355 and Plato's death in 347. That means that if (A)
was originallypart oftheEudemianEthics,or derived froman account in
the Eudemian Ethics,it must,along withthe Philebus,be a contribution
to the ongoing debate about pleasure. Diiring dates the Nicomachean
Ethics (and thereforecertainlyaccount [B]) to the period afterPlato's

7"Stob. Ec. 2, p. 71, 15 = SJVF3.106. Aristotlecertainlythought that some "goods"

are associated with means and some with ends.
8sScholarshave often noticed that while (B) mentions Eudoxus by name as the upholder of a theoryfromwhich Aristotledissociates himself,(A) does not mentionhimand some have assumed this to mean that (A) is more or less straightEudoxus. We can
certainly agree that in (A) Aristotleis more powerfullyinfluencedby Eudoxus, and that
this is one of the pointersto its earlier date (in some form,though not necessarilyours).
(Heidelberg 1966) 48-52.
3A4ristoteles

This content downloaded from 134.48.233.114 on Thu, 12 Jun 2014 11:25:56 AM

All use subject to JSTOR Terms and Conditions

178

PHOENIX

death and afterAristotle'sreturnto Athens,that is, between334 and 322.

Yet, althoughsome versionof the NicomacheanEthicsmust have existed
by the timeEpicurus firstreachedAthens,he knowsonly the ideas about
pleasure representedby the earlierversion(A). Indeed, if therewere two
versionsof (A) in circulation,one originallydesigned forthe Eudemian
Ethics,and one revised forthe NicomacheanEthics,Epicurus must have
been familiaronly with the earlierversion-or he would have known(B)
as well! What this suggestsis a theorythat the textof the Nicomachean
Ethics (or at least of Book 10) was not widely available (even among
those interestedin philosophy,and in pleasure in particular)until after
Aristotle's death. We can easily understand that Epicurus might not
have heardAristotletalkingthe language of (B); it is harderto assume
he would not have read it, if it was in the public domain.
We may speculate further.CertainlyEpicurus returnedto Athens in
306, by whichtime the NicomacheanEthics was presumablymorewidely
known.Why did Epicurus not make use ofit? Perhaps by the timehe had
set himselfup in Athens as a teacher,he had no interestin learning;he
already had a philosophy!But do we need to followthisunflattering
line,
characteristicof Epicurus thoughit is? Do we knowthat theNicomachean
Ethicswas available in Athenseven in 306? Not forcertain.We do know,
as we have argued, that certain Stoics are aware of some of the ideas to
be foundthere,thoughthe debate about pleasurehad long since subsided
and no longeroccupied the centreof the philosophicalstage. But we do
not know whetherZeno, the founderof Stoicism,is responsibleforthe
Stoic use of the ideas of NicomacheanEthics 10 (and 2?). For Cleanthes,
as we have seen, the concept of pleasure as an rL'rY}r7ima
is certain,but
Cleanthes is a mid-thirdcenturyfigure.40Even thoughhe probablymade
few startlinglynew approaches in ethics, we cannot assume that his
account of pleasure exactly correspondswith that of Zeno. So it still
remainspossible that even in 306 the NicomacheanEthicswas not in wide
circulation.And in generalit mustbe admittedthat althoughStoic ethics
frequentlybuilds on foundationsprepared by Aristotle,we cannot pin
down any of the relevant texts to Zeno ratherthan to later Stoics. We
must remain agnostic, therefore,as to when the influenceof the NicomacheanEthics (and therefore,of [B]) penetratedto the Stoic schooland whetherEpicurus could have read it by 306.
There is a featureof the comparison between Stoic and Epicurean
reaction to Aristotle's doctrine of pleasure to which we must return.
Epicurus, like Eudoxus, and perhaps even the Aristotleof (A), makes
pleasure the highestgood: Epicurus is joining in an as yet unresolved
debate at the time when Aristotleat least had passed it by. By the time
"'Cf. H. von Arnim,"Kleanthes,"RE 111.558-574.

This content downloaded from 134.48.233.114 on Thu, 12 Jun 2014 11:25:56 AM

All use subject to JSTOR Terms and Conditions

PLEASURE: 360-300 B.C.

179

the Stoics come across Aristotle'swritings,the struggleis over; hence

perhaps the more casual and vaguer Stoic referencesto (B) when compared with the morespecificand close attentionEpicurus pays to (A).
An anxietyremains.If Epicurus was so interestedin (A), why did he
not take more interest,even in 323, in what Aristotle was doing, for
Aristotlemust at least have been talkingabout (B)? Or did he not learn
anythingat all about pleasure in 323? But if not, thenwhen did he come
across (A) while failingto hear of (B)? But do we have to believe that
Aristotlewas totallyobscurein 323 and that Epicurus knewlittleof him?
Some mightlike to followthis road, but at the least it must be admitted
that Aristotlehad been a prominentmemberof the Academy. Later on,
as we know,Epicurus despised Aristotle.41
Did he then come late in life,
when his own views were largelydeveloped, to pay so much attentionto
(A), the fairlyearly workof a man whom he despised?Or is the solution
afterall that what we findin (A) is basically Eudoxus (and knownto be
so by Epicurus), onlytranscribedby Aristotlewitha minimumof change?
Did Epicurus pick up his theoryof pleasure fromthe famousastronomer
ratherthan the obscurepolymath?It mightlook that way. But thereis a
serious, indeed probably fatal, objection: the author of the end of (A)
wields Aristotle's own account of KLvionts
and vpyLa--and Epicurus
knowsthat distinction.There is a primafacie case that Epicurus met (A)
in 323, in writtenform,and that he knew it to be Aristotelian.The
corollary,as we have suggested,is that (B) did not exist in writtenform

in323.42

UNIVERSITY

COLLEGE, TORONTO

41Cf. Usener 171, and I. Diiring, Aristotlein the Biographical Tradition (G5teborg
1957) 385-386.
42Atvarious stages this paper has been read by Professor T. M. Robinson, Dr A. A.
Long, and ProfessorF. H. Sandbach, and I should like to thank them fortheirconstructive criticism.

This content downloaded from 134.48.233.114 on Thu, 12 Jun 2014 11:25:56 AM

All use subject to JSTOR Terms and Conditions