Epicurus' Doctrine of the Soul

Author(s): G. B. Kerferd
Source: Phronesis, Vol. 16, No. 1 (1971), pp. 80-96
Published by: BRILL
Stable URL: http://www.jstor.org/stable/4181858 .
Accessed: 08/12/2013 15:10

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 [email protected].

BRILL is collaborating with JSTOR to digitize, preserve and extend access to Phronesis.

http://www.jstor.org

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
 Epicurus'doctrine
of thesoul
G. B. KERFERD

T HE discussionof the soul in Epicurus'LettertoHerodotus(Diogenes

Laertius X. 63-68) raises two major problems of interpretation,
that of the structure of the soul, and that of its distribution and
manner of functioning in relation to the body. It will be convenient
to begin with a preliminary survey of these problems in their various
aspects before suggesting some possible lines of solution.

I. The Structure o/ the Soul

Apart from one sentence the meaning of which is quite uncertain,
the whole treatment of the soul in the Letter suggest at first sight a
unitary view - the soul is treated as if it were a single substance.
This approach stands in contrast to later Epicurean traditions ac-
cording to which there was
(a) a Doctrine of Parts (so T6 XoyLM'rx6v and r6 &Xoyovin E ad. D. L. X. 66
fr. 311 Us., Aetius IV. 4. 6 = fr. 312 Us. = 140 Arr., Diog. Oen. fr. 37 Chil-
ton = 36 Grilli = 39 William, and animus and anima in Lucr. III. 136 ff.,
lamblichus ap. Stobaeus 363. 11 Wachsm. = Usener, Epicurea p. 353).
(b) a Doctrine of Elements according to which the soul is in some sense a
mixture made up of different constituents (so Lucr. III. 231 ff., Aetius
IV. 3. 11 = fr. 315 Us. = 139 Arr., Plutarch, Adv. Col. 1118 d = fr.
314 Us. 138 Arr., Alexander Aphr., De An. I. 8 = fr. 315 Us.) In Lu-
cretius these elements are four, namely breath, heat, air and the fourth,
nameless, element.
This discrepancy raises problems. But we are told in par. 35 that the
Letterto Herodotuswas a summary (either exclusively or inclusively)
for those who have already made some progress in the study of Epi-
curus' doctrine (troy 7rpoPePjx6toc 'xxvJr ?v 6- Tov 6X?av rl(XkeL)
and in par. 68 we are told that the summary has dealt with general
outlines - s'7rot - in such a way that the details can be worked out
from them subsequently, cf. 45 and 83. We are accordinglyconfronted
with the following questions. Do either or both (a) and (b) above
represent views held by Epicurus, and if so are they already implied
in what he says in the Letterto Herodotus?In that case what is the re-
lationship between the Doctrine of Parts and the Doctrine of Elements?
Or is one or both of views (a) and (b) a later development, later either

80

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
in the lifetime of Epicurus (suggested by Bignone, Epicuro, Opere
97 n. 2 and Atene e Roma (1940) 175) or after his death?

1. The problemof the first two sentencesof par. 63.

Mvst
8& raurio aZ ouvop&v&vocppovrcx &7LT'Xq oLaDaelLq xcxl '&7tawa1-
o( T yaPIp
a' n LOt7'O71 7rt=L9r)
e-OCLor 7cqtoL aT 'a'rt Xvmo?Lep?
MXp 6Xov 'ro &apoLtaL 7rwapeaTrAop[Levov,'rrpoGaEpepCaToctov U 7VrVLOeLC'
5 eV TO&rcpnpoa (pep, 7n ?
LTVXXpOcaLV 9XGVTL XOaL7tfl
9=L ae rTO [LXpO o7O?o7pV oapUcy v eWykp 'r Xro[epd'z xocx'ouiiv
-rou'&v, aup ? a? ro&9cp VXXO' xoL ro ?o7tn A&po'a?.
It is widely held that these sentences refer to a three-element view
of the soul (Diels' emendation of the text by inserting <y> implies
this. It is the view taken by Giussani, Studi Lucreziani 210, Bailey,
Epicurus p. 227, Greek Atomists 388-9, Commentary on Lucretius
Vol. II p. 1006, Diano, Giorn. Crit. della Filos. It. (1939) 105 ff.,
Arrighetti, Epicuro Operep. 470: la quarta nature priva di nome e la
terza della nostra epistola, Furley, Two Studies in the GreekAtomists
196-7.) On this view the soul consists of particles of (or like) heat
and breath and a third unnamedelement. We are left with two further
problems - why in the Letterto Herodotusis air not mentioned sepa-
rately from breath as it is by Lucretius and why is there no mention
of the distinction made in Lucretius between animus and anima?
But there are serious objections to this whole interpretation which
must first be discussed before we attempt to deal with what are essen-
tially consequential problems. The first difficulty is that the language
of the first sentence is so unitary in character that it does not seem
by itself to have anything to do with a doctrine of elements. The
soul is a asicpo e7%tLepeg and this suggests that it is a single body.
While the term Xero[Lep?no doubt does imply constituent particles
it does not imply or suggest different kinds of particles (although
of course it does not exclude such a possibility either). In the second
half of the sentence the unitary flavour is even stronger. We are not
told that some soul particles are like heat and others like breath,
but that the soul itself is like a blend of the two and that in some
respects it is like the one and in other respects it is like the other.
This comparison does not naturally suggest either different parts of
the soul like one or the other element or different constituent elements
of the soul, one like breath and the other like heat. What it does
suggest is that the soul as a whole has some resemblance to breath

81

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
and some resemblance to heat.' When we turn to the second sentence
there also we find that with the text as given in the manuscripts2
the natural reference is not to constituent particles and so to a doctrine
of elements in the soul as a whole, but to some doctrine of a part in
relation to the soul. Heinze (in his edition of Lucretius III pp. 35-6)
argued that the reference must be either to the whole soul seen as a
part - the quXzXov ,ue6poq - of the body, or to a part of the soul. He
preferred the first of these alternatives. At this stage, that of first
sight, it may seem that the second is the more likely. When later writers
wish to distinguish the two main parts of the Epicurean soul, the
rational and the irrational, it might be said, it is ,epoq and its deriva-
tives that they naturally use for this purpose (see references in (a)
above on p. 80). Accordingly we should ask whether the second sentence
should not be taken as referring to one of the two parts of the soul,
rather than to one of its elements. (Of course if one of the two parts
of the soul was equated either exclusively or by predominance with
one of the elements the result might be the same as the view here
rejected. But the meaning of the sentence would not be altered - it
would still re/er to a part of the soul rather than to a constituent
element as such.) If the reference is to such a part, to which would it
be? The reference to the superior fineness of its parts might lead us to
think of the animus, because of the superior speed normally ascribed
to intellectual processes, such speed being promoted by superior
fineness. But the reference to its power of "sympathy" (auJX7Tcx)3
and to the general 'powers' of the soul in the later part of par. 63 suggest
that it is rather to the anima that we should look in the first instance.
2. The evidence of the scholium in par. 66.
The scholium first states that 'elsewhere Epicurus says that the soul
1 How dangerous preconceived ideas may be can be seen from Bailey, Greek
Atomists 388, who actually says 'Breath and heat, then, or rather, as Epicurus
states it, with a scrupulous accuracy not imitated by his follower, 'particles
resembling those of breath and heat', are the first constituents of the soul.'
Unfortunately Bailey himself is not scrupulously accurate. The Greek does not
say 'particles resembling' - it is the whole body of the soul that does the re-
sembling.
a All the manuscripts read lat 8i Tb ipo4. The definite article is appropriate if
the part in question is one of two. Whether the reference is to part or to element
the mss. reading should be preferred to the emendation to 'L by Woltjer,
followed by Apelt, Von der Muhll and Arrighetti. As Bignone (Epicuro, p. 97,
n. 3) says 'Nel dubbio 6 periculoso toccare il testo manoscritto.'
a See what Bailey (who does not take this view) says on auF7ra$Om(Epicurus p. 227).

82

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
is composed of very smooth and very rounded atoms, extremely dif-
ferent from the atoms of fire.' On this Bignone (Epicuro 100 n. 4)
wrote: "1 questo un indizio che Epicuro, nel luogo a cui si riferisce
lo scoliasta, polemizzava contro Democrito, il quale faceva consistere
l'anima solo di atomi ignei." The polemic against Democritus, which
most scholars accept, is likely enough, although the reference could
equally well be to the traditional view of the soul as involving heat
for which cf. Aristotle De caelo 303 b 20. But we are not justified in
inferring that here Epicurus was positing other elements in addition
to fire which is what Bignone's words suggest. The Greek taken by
itself suggests that the soul consists of a different kind of atoms
altogether, excluding fire, and of a single kind of atom at that. This
is entirely consistent with the statement in par. 63 that the soul is a
a6.x Xenro?epec and it goes no further than that statement so far as
concerns any doctrine of a plurality of elements. The important
statement which follows in the scholium clearly atributes the basic
distinction between animus and anima to Epicurus, but it does so in
such a way as to suggest that they are differentiated by location rather
than by any difference in the constituent elements. Once again, we
are left with a unitary view of the structureof the soul. How this could
conceivably be reconciled with the undoubted later evidence of a
four-element view of the Epicurean soul is discussed below.

II. The Problemof theFunctioning of the Soul.

The soul for Epicurus while undoubtedly material and composed of
atoms, is also necessarily responsible for life, sensation and thought.
Clearly not all atoms occurring in nature involve all or any of these
three functions. What then is it that makes these particularatoms fulfil
the functions of a soul in this way? This is the problem which in
Italian has been called the problem of the PsichicitcLof the soul,
and for this I am tempted to coin an English equivalent, Psychicity.
Some important general statements bearing on this problem are given
in the Letterto Herodotus64-66, but the main evidence comes from
later sources, above all from Lucretius.

1. The evidenceof the Letterto Herodotus64-66.

The following points seem reasonably clear:
(a) The soul is the major cause of sensation but for this purpose it
must first have been contained within a body.

83

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
 (b) The body in such cases itself receives a share in the same
cnaurrcowpof sensation.
(c) This share does not cover the whole range of the soul's functions.
(d) This share of the body in sensation is derivative in a way that
is not the case with the soul's power of sensation. The soul came into
existence together with the body but when so brought into existence
acquired sensation in a more fundamental way than the body, because
there is a differencein the way in which a soul loses its acquired power
of sensation and the way in which the body loses it.
While I believe that (d) is a fair representation of what is said in
the Letter the details are often not clear and there is much that is
unexplained.For furtherelucidationwe must turn to the later tradition.

2. The doctrineof elementsin the later tradition.

The soul is described as a xpa,ptuExErr'c&pcov by Aetius IV. 3. 11-
315 Us. 139 Arr., the four elements being 7up683i, OCep6)r], veuFx-
'rtx6v and &xcxrovo6?Cea'rov.This description is found also in Plutarch
(Adv.Col. 1118 d- 314 Us. = 138 Arr.) and it is developed in detail
in Lucretius III. 231 ff. The fourth unnamed element is responsible
for sensation according to Aetius, while according to Plutarch it is
that by which the soul judges, remembers, loves and hates, and in
general its thinking and reasoning faculty spring from the unnamed
element. According to Lucretius it is the fourth element which starts
all the movements of sensation as the other elements are insufficient
without it. He may have added in III. 240 the statement that the
fourth nature is also necessary for thought, but this depends on an
uncertain reading and apparently he does not say this elsewhere.
Three different views were put forward in the nineteenth century
about the relationship between the four elements and the two parts
of the soul:
(a) The animus consists solely of the nameless element, while the anima
is made up from the other three elements without the nameless
element.
- Reisacker, Brieger (second view).
(b) The animus consists of the other three elements together with
the nameless element, while the anima is made up from the other
three elements without the nameless element.
- Woltjer, Eichner, Tohte, Brieger (earlier view, 1877, 1884),
Munro (Vol. II" p. 194)

84

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
(c) All four elements are found both in the animus and in the anima.
- Giussani, Studi 188 ff.
Heinze pp. 41-42.
Bailey, GreekAt. 392 and Appendix V.
(Brieger in 1893 seems to have conceded that this is correct
for Lucretius, cf. Giussani 192 n. 1)

I believe most would now agree that Giussani, Heinze and Bailey
have so conclusively proved their view against the previous two views
that it is perhaps not necessary now to rehearse their arguments.
3. The fourth elementand the problemof Psychicity
As the fourth element is necessary in order to make sensation possible
it is of interest to ask how the fourth element produces this result.
It is clear that sensation involves a whole series of movements - move-
ments of images, movements of the body, even movements of the other
elements in the soul. But not all of these movements are themselves
sensations - the presence of the fourth element is a prerequisite.
Indeed psychicity may fairly be said to be related directly to this
fourth element.
It will be convenient to distinguish logically some six or seven
different possibilities. The first four of these would assign psychicity
to the fourth element per se.
(a) It is a specific property of the separate individual atoms of the
fourth element.
It is hardly possible that this should have been Epicurus' view.
It violates the general principle that atoms are quality-free except
for shape, size and weight - par. 54 (though admittedly the pro-
hibition there applies to the assignment of perceptiblequalities
other than shape, size and weight) and the doctrine that atoms
cannot change and so are -awab par. 54-55.
But more important than these considerations are the emphatic
arguments of Lucretius (II. 865-930) that the power of sensation
arises from things which are insensible - ex insensilibus sensile gigni-
followed by a further argument (IL.933-43) to the effect that sensation
arises by a union of matter and not by a mutation of the original
particles.
(b) It arises when atoms of the fourth element are brought together
and so it is a property of their conciliumwhenever this occurs.

85

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
(c) It arises when atoms of the fourth element are brought together
in one or more particular patterns of arrangement - i.e. not
simply any concilium but in the right concilium.
Bailey's analysis of the genesis of colour might suggest that Bailey
favoured this view, but his referenceto the need for appropriatemove-
ment in a conciliummeans that it was not his considered view; cf.
GreekAtomists393-4. It could not be Epicurus' view because clearly
such a pattern could arise outside a body, yet sensation cannot.
(d) It arises from the appropriate movement in a conciliumof atoms
of the fourth element in the right positions, and arrangements.
This seems to be the view of Giussani (Studi, p. 189). But the
objection already mentioned under (c) could apply here also. Moreover
all of the possibilities so far mentioned involve treating sensation as
a property of the fourth element per se. This would surely make sen-
sation into a au,upp?xo6q. Yet in the Letter to Herodotuspar. 64 it
is spoken of as a asvust&oia both of the soul and of the body.
(e) It might be that sensation is a special kind of movement of fourth-
element soul-atoms, only found when these move within the pores
of the body-atoms. 'They are forced by the very minuteness
of their field of movement into the appropriate motions of sen-
sation' - Bailey, GreekAtomists398.
On this view it is the pores which alone make possible these sen-
sileri motus. It might be objected that any specified motion could
arise by chance in infinite space outside a body. But this objection
would apply also to the soul as a whole. Yet Bailey (I.c.) seems correct
in saying that for Lucretius the soul owes its very existence as an
aggregate body to its confinement within the body and there seems no
reason why sensation also should not have been regardedas a ,TLM
dependent on the special motions imposed by the internal configu-
ration of the body-atoms. Controversial support for the view here
suggested may come from the interpretation of Lucretius III. 262 pro-
posed below. But it seems also to be supported by the first sentence
of par. 64 in the Letterto Herodotus.
(f) It could be that the fourth element never itself constituted sen-
sation but produced it by acting as a catalyst. Epicurus, Letter
to Herodotus,par. 64-66 does seem to be saying in effect that body
and soul each exercise a sort of catalytic effect on the other so
far as sensation is concerned. But there is nothing to suggest that

86

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
 either acted as a true catalyst i.e. that they were initiating changes
in processes without themselves participating in them.
(g) The general view as given in Lucretius is strongly in favour of
the assignment of sensation to the whole soul and indeed to
the body also though it is admitted that there are insensitive
parts of the body. It is transmitted, admittedly from one atomic
complex to another, but if all have the power of sensation some-
how or other then psychicity must be a kind of motion.

4. The internaltreatmentby the soul o the e'&acoXo

derivedfromperception.

The et8Awcxof perception must leave a permanent record of themselves

in the soul to make memory possible and (probably) also to enable
nrpo?B4itto be produced.
It might be that the actual images themselves pass on through
the pores of the body into the mind - so e'q r-v 6+Lv ' 'rv &&votav in
par. 49 of the Letterto Herodotus(as Zeller held, Ph. d. Gr. III.1l.436,
cf. also Diano, GCFI (1939) 132) but it is more likely that the image of
sense-perception was 'reproduced' as a new image in the mind by
the sensiferi motus (so Bailey, GreekAtomists 418). More difficult is
the way in which they produce a permanent record.

(a) It might be that the images however received are physically

collected and stored up in the mind. So apparently Bailey,
GreekAtomists 245, 417-419. But see contra Furley, Two Studies
197: this would mean bringing e.g. iron atoms into the soul in
cases where iron is perceived. Yet the soul can only consist of its
own (four) kinds of atoms. This objection however would only
apply to the direct-entry theory of images, not to the telegraph
theory.
(b) It could be that the images produce changes of patterns of ar-
rangement in the soul atoms. To this it might be objected that
if specific arrangement of the soul-atoms is what gives the soul
its Psychicity any rearrangementwould endanger this or destroy
it altogether. This objection could be met by supposing that the
rearrangementswere of a subordinate kind, within an unchanged
overall pattern.
(c) It could be that a change of patternof movementof the soul-atoms
rather than a change in their arrangement in relation to each

87

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
 other is what is involved - so Furley, Two Studies 202, and cf.
Letterto Herodotuspar. 51.
&WXv 'trvm xLv7]aLvin
(d) Lucretius in Book IV 962-1036 suggests that memory is aroused
by fresh similar images from outside, and perhaps the best view
of all is that of Diano (GCFI (1939) 138-9) based in part on this
statement of Lucretius. According to Diano memory is neither
a change of pattern or of motion, but the capacity of repeating
the act of apprehensionto which the mind has become accustomed.
Exactly how such a capacity could be stored in the soul-atoms
is however not easy to understand.

The difficulty of deciding this question is acute because all the

evidence seems to relate either to individual acts of perception or
individual acts of recollection or individual acts of thought. In all
these cases movementis naturally involved. But what we want to know
is how memories are stored when not actually in use and there seems
to be no passage bearing directly on just this question.4

5. Thedoctrineof molecules
Individual atoms moving downwards through the void move all
with equal speed, and the same speed continues when deflexions
follow after collisions. This speed is the speed of thinking (Letterto
Herodoluspar. 61). Even within compound bodies all individual atoms
still move at the same speed (par. 62). Collisionscan however produce
the appearanceof slower movements (par.46 fin.) and Bailey(Epicurus
pp. 220-1) is likely to be right in supposing that this operates in two
ways: -
(a) because a given atom takes longer to travel from point a. to point b.
if it is deflected from its course from a. to b. (In this Bailey is following
Brieger (1893) pp. 7-9)5.
(b) In the movements of compound bodies which do vary in speed but only
because in the faster body more atoms are moving in the direction of the
whole body than is the case with the slower body.

' Cantarella, L'Antiquitd Classique (1936) 273 ff. supposed that there was a
reference to memory in Arrighetti fr. 32. 10, but see Arrighetti pp. 583-4.
5 There is no good evidence for Bignone's view (Epicuro, pp. 225-238) that there
is a pause at the moment of impact. Nor is there any likelihood of truth in the
attempt by Arrighetti pp. 467-8 to re-introduce differential velocities for light
and heavy atoms after impact while accepting that there is no differentiation
before impact. See Giussani, Studi 100 ff.

88

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
Giussani (Studi, p. 58) argued that the term 6yxog was used to re-
present what we call a molecule, i.e. the least particle of a substance
which possessesthe qualities distinctive of that substance, and he further
argued that in II. 454 Lucretius used the term glomeramento mean
'the molecule of liquids'. Bailey (GreekAtomists 342-4 and Appendix
IV) successfully refuted this view, but argued that in Lucretius the
term semen, while it sometimes refers to individual atoms, on other
occasions refers to "the already formed nuclei, specifically adapted
for the making of 'things'." But Bailey has to admit that in many
passages seminain Lucretiusrefersto individual atoms and it is probable
that this is true in all cases.
Is there then no doctrine of molecules in Epicirus? The answer
must surely be that there was. Once it is admitted that the properties
of individual atoms are limited and that the properties of things and
substances only arise in combination this inevitably carries with it a
doctrine of molecules.
6. What Lucretius says about the relation betweenthe elementsand the
soul in Book III.
The soul is not simplex (231) but is triplex (237) and there must also
be included a quarta natura (241). The relationship between these
constituents is difficult to express because of the patrii sermonisegestas
(260), the following points however seem clear:
They cannot be separated. (263-4).
Their capacities cannot be divided spatially. (264).
They are like the multiple powers of a single body. (265).
Why should he refer here to the poverty of the Latin language?
Bailey (CommentaryVol. II p. 1033) writes: 'Yet the difficulty clearly
was not so much that of expression in Latin, but rather that of the
explanation of an unusually subtle idea; this is one of the Graiorum
obscurareperta.'
But I want to suggest that the meaning is perhaps more specific
than this. In the other passage where Lucretiusrefers to Patrii sermonis
egestas (I. 832) it is because he has no word in Latin to express homoeo-
meria, a technical doctrine which he ascribes to Anaxagoras. Again
in I. 136-139 it is the absence of technical terms in Latin which is the
source of the trouble. Perhaps in Book III also it is a similar absence
of technical terms in Latin which is what he has primarily in mind.
In that case I would suggest that the kind of technical term for which
Lucretius is looking is one that will express the special Epicurean
doctrine of atomic blending.
89

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
 According to Aetius IV. 3. 11 = 315 Us. - 139 Arr. the soul for
Epicurus is a xp&t?v.6This is a not very common word for the product
of xp&at4,which term is used in par. 63, and it seems frequently to be
found in Epicurean contexts,7 so that it could even be that this is the
very word in question. The doctrine involved is explained in Alexander
of Aphrodisias De Mixtione 214. 28 - 215. 8 Bruns - fr. 290 Us.
Democritus had supposed that a xp&aL involves really only a juxta-
position of separate ingredients each preserving its own nature
as before the mixture, but no longer separately perceivable because
of the smallness of the amounts juxtaposed in any particular part.
But Epicurus supposed that in xpoc&atthe separate substances were
first broken up into their constituent atoms which were then re-
combined. So it was not so much a combination of two or more sub-
stances as a new combination of substance-forming atoms.
Two points are of great interest here. The first is that this distinc-
tion attributed to Epicurus clearly does imply a doctrine of molecules.
Plutarch (Adv. Col. 1110 b = fr. 60 Us. = 20. 2 Arr.) preserves an
example from wine: 'Often the wine does not possess the property
of heating or cooling as it enters the body. Rather the bodily mass is
so set in motion that the corpuscles shift their position: the heat-
producing atoms are at one time concentrated, becoming numerous
enough to impart warmth and heat to the body, but at another time
are driven out, producing a chill.' See for discussion Bailey, Epicurus
p. 389. The second is that it fits exactly with what Lucretius is trying
to say about the soul, as Heinze saw (p. 41-2). It is not a mixture of
four substances by juxtaposition, but a true Epicurean xp&,zx. This
explains the reference to semina in III. 127 - 8 and primordia in III.
236, and it probably explains the use of 7rotoi3in Aetius IV. 3. 11. That
it represents exactly the way Epicurus frequently thought is shown
by the reference to &'rot?otnupk &=oTe)ea'xo( in D.L. X. 115. More-
over I believe it enables us to approach with greater confidence the
well-known problem-passage in Lucretius III. 262 - 5:

inter enim cursant primordia principiorum

motibus inter se, nil ut secernier unum
possit nec spatio fieri divisa potestas,
sed quasi multae vis unius corporis exstant.

6 So Plut. Adv. Col. 1109 e, Alexander Aphr. De Mixtione 215. 11, 231. 28, 232. 28.
7It may be that Epicurus had a special preference for neuter nouns in -,. cf.
Cleomedes II. 1 given in Usener p. 89.

9O

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
On this Bailey argues as follows (CommentaryVol. II p. 1037): prin-
cipiorummust, as usual, be gen. of primordia;see i. 55 n. It is impossible
that primordiaprincipiorumshould go together, since first beginnings
cannot themselves have first beginnings. It must therefore go with
motibus (which otherwise would be left in the air) and mean 'with the
motions of first beginnings' i.e. suis motibus or propriis motibus,
'with the motions proper to them'. He translates:
'For the first-beginnings course to and fro among themselves with
the motions of first-beginnings, so that no single element can be put
apart, nor can its powers be set in play divided from others by empty
space, but they are, as it were, the many forces of a single body.'
In all the essentials of this view Bailey was preceded by Munro,
Giussani, Heinze, and Ernout and Robin (Commentaire),and this is
the interpretation followed by most translators. But Bailey saw clearly
enough that nil unum does not refer to single atoms but to the four
constituent elements in the soul (cf. also line 269 with its emphatic sic).
This however seems to need some specific referenceto the four elements,
in the same sentenceif possible, in order to make the statement intelli-
gible. This raises the possibility that principiorum may here after all
mean elements. This is actually proposed in a complicated way by
Leonard and Smith, while still keeping principiorum motibus as a
single phrase, meaning 'with the motions of the elements'. But the
best sense of aUlis given if we take together primordia principiorum -
with the meaning 'the atoms of the elements course to and fro among
themselves with such motions that no single element can be separated
...' This seems to be the interpretation actually reached by Emout
in the Bud6 translation - sans doute, les mouvements des corps premiers
de ces substances s'entre-croisent a ce point qu'il est impossible d'isoler
une d'entre elles et de localiser chacune de leurs facultes - and I have
little doubt that it is the correct one.8

7. The origin of the sensiferi motus in Lucretius.

According to Lucretius the fourth nature penitus prorsum latet and it
subestso that there is nothing in our body magis hac infra. (III. 273-4).
We have already been told that there can be no spatial separation of
one element from another (III. 264) and Bailey is presumably right
(CommentaryVol. II. 1038-9) in inferring that the reference must be
to the relative imperceptibility of the fourth nature rather than its

8 A similar translation is given by R. Waltz (1954) and M. F. Smith (1969).

91

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
separate location im particular parts of the body. But he might have
added that its obscurity is in relation to the body (cf. 274) rather than
in relation to the other elements - the quicquamin line 274 is quite
general and refers to everything whatsoever, not simply to the other
elements. Indeed if I am right about the distribution of individual
atoms in the soul-xpa&aa locational distinction would not be possible
even within a particular part of the soul. None the less we are clearly
told that it is the fourth element which is the source of the sensi/eri
motus(III. 245-6, 271-2).
Lucretius makes two points with some emphasis:
(a) the fourth element is stirred first and then the movements pass
to heat, wind and then air, and then everything including blood
and flesh (III. 245-251).
(b) but the body itself feels (III. 350-369) - neither body nor soul,
whether as anima or animus can feel when isolated from each
other (III. 331-336).

The fact that Lucretius refers to the fourth element as 'consisting'

of atoms of a particularkind, and that he refersto the other constituent
elements of the soul in similar terms is not inconsistent with the kind
of xpoas doctrine suggested above. Even when he explains differences
in temperament and character as due to a predominance of various
elements this need not mean that there were any areas where such
elements occurred in pure form i.e. without admixture of other soul
atoms. In each case we need have no more than &'ro[ot tup6qMc7roreXecnt-
xoEand so on.

III. The letterto Herodotus63-68.

I return now to the interpretation of the text of the Letterto Herodotus
63-68. In the light of the evidence already discussed the following
points may now be made.
1. The first paragraph does not describe different elements in the
soul. This is so because of the way in which the elements were com-
bined in a xpa,ua.The behaviour of the soul, however, is in various
ways like the behaviour of such elements in separation because it is
composed of atoms of the elements although re-combined in a new
way.
2. There remains the possibility that there is a reference to a doc-
trine of parts of the soul - or more strictly to the soul as a whole

92

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
and to one particular part within it, namely the animus. Heinze was
right in maintaining that there is (in Lucretius) no relationship be-
tween the doctrine of elements and the doctrine of parts in the soul.
This differentiation of the "part" is stated to be based on the fact
that it is iroXX'v7ropo)XXyrJv xoclaxr6v tou&tv.
sprd gc,rn Xe7-roR?epeLqc
This is usually interpretedas though noXA?vwere n),E meaning 'more',
and as though the phrase itoX?'v7rapV?XOyrkvZLX?Jyp Tq XCTCroPEpC
meant toX ? The reason for this interpretation
is the belief that the reference must be to the fourth nature.
But-
(a) This is a very elaborate and unnecessary periphrasis. Why not
say simply ?r'soCuepi=epov?
(b) The phrase does not, anyway, naturally have this meaning.
The commonest meaning of napcxUay4is 'change' - very commonly
change of position or movement, and often with the idea of
'interchange'. It also frequently refers to qualitative variation.
Moreover it would normally be accompanied by a genitive of
that which varies or is varied - so 7p.?Xy'v ,i.y&&Cv in par. 55
of the present letter. This is the meaning of xwcrmc 7cpaUcxyiv
Lpi5vin the Letterto Pythoclespar. 95 and again in -&c lkpouve'voc
7ropaoAXUxya&in par. 113. This applies, I would say, in all probability
also to the other cases listed under 7 xpaxXcx in Arrighetti'sindex.
A comparison after 7rxpocday would normally have np6s +
accusative, not a genitive.
For these reasons I now suggest that the currently received interpre-
tation of this sentence should be rejected. There is no question of
different degrees of lightness of parts, i.e. degrees of XvCTOPELOL.
What we are concerned with is a p.epo4- and I now translate - 'which
has acquired great mobility (or perhaps 'great capacity for change',
i.e. 'variability') as a result of the lightness of parts of just these things
(namely of breath and heat)'. It is by virtue of its power of variation
that it is able to undergo modifications jointly with (ausv- in au[L-
7rxO*q ... ) the rest of the structure.
This interpretation is not new - it is to be found in the Latin
translation attached to the Cobet's Greek text published in the Didot
series,9 which reads: "Est autem pars que multam accepit immutati-
o This translation was based on Hubner's corrections made in 1844 to the first
Latin translation of Diogenes Laertius completed in 1431 by St. Ambrose of
Camaldoli (Ambrogio Traversari) who had learnt Greek from the Byzantine
Manuel Chrysoloras.

93

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
onem exilitate partium etiam horum ipsorum, consentit autem huic
magis et reliquae congregationi." But we must now consider again
the question of the ue?poa which is the subject. Assuming that it refers
to a part of the soul, I have already mentioned the difficulty of de-
ciding whether this would be the animus or the anima. But the following
sentence suggests that it is neither, as the contrast seems to be to r6
XOL7OV 64pOLap. This suggests that the reference is to the ai-?L which
is the soul as a whole. This would be possible if we followed Heinze's
interpretation of tLipo4as referring to the soul as a part o/ the body.
This indeed is what I think must be the meaning, and if it is, the
whole passage begins to make sense - it is the relation of soul to body
which is being discussed, not of one part of the soul to another.
But difficulties remain. Cobet's translation treated gcTL as existential,
and this interpretation of grrL has also been built into the received
interpretation, so that we find translations such as 'there is also the
part which...' This will not do if the reference is to the soul as a part
of the body and Heinze, apparently not very confidently, (p. 34)
offered 'sie (sc. Die Seele - GBK) ist aber ein Theil, der, an Feinheit
selbst diese noch weit iibertreffend, eben dadurch mit dem Rest der
Anhaufung mehr mitempfindet'.
I wonder if we may not do better than that: if so-rt is not existential,
it could be taken with e'L?64 in the sense 'but the part has acquired. . . '
This periphrastic use of the verb 'to be' with a perfect participle would
have plenty of parallels. Linguistic features now begin to fall into
place. There is no xatxbefore r6 Vkpoq,which is something that ought
to have been there if the reference were to a part not already mentioned.
There is no need for a second article after pepo4 (i.e. r6... eLWX6;)
which is needed on the existential interpretation if the original article
is retained before 'poq. The soul while resembling breath mixed
with heat is not identical with them. But it does derive one quality
from the quality of breath and heat, namely variability resulting
from their XenroCeptoc.This it derives from the individual atoms,
and this is the basis of the LxwvqaLoctof the soul mentioned in the later
part of the same paragraph. It is this sympathetic ease of movement
which is the source of the soul's Psychicity.
On the basis of this hypothesis - of an essentially unitary soul built
up in a special way from highly mobile individual atoms - we may
now consider some problems in the succeeding chapters dealing with
the soul in the Letter to Herodotus, while recognising that at many
points we are necessarily dealing in conjecture rather than certainty.

94

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
 The soul contains the greatest part of the cause of sensation (par.
63 fin.), this being its power of sympathetic variation. It 'contains'
it because it contains matter which can move according to suitable
patterns. But this capacity of suitable movement is acquired, not
innate - so oU ,Vu E'L)<TL xv tocur'ivin par. 64 - and it is acquired
because it is enclosed in the whole structure of the organism function-
ing as a kind of vessel-
ui et ntUovOl XOMoi &i&poEatO4 earey&?oO 7rq.
This capacity is a a)VZTW0G and we should understandthat it is equaly
a aumCuall of the soul and of the body - to neither is it a permanent
property or V Heinze (p. 37) is however possibly correct
in regardingthis capacity as a auVex6q of a man.
But the body does not acquire a share in all the things which the
soul has obtained - oi) jLeVoL ntaivtv v ixeLVvx6x'nyat. This sentence
is obscure. Bailey took the reference to be to thought and visualiza-
tion (Epicurus p. 228). But this can hardly be correct in the case of
visualization if we accept as we surely ought that Lucretius III.
359 - 369 is evidence for Epicurus. Moreoverthe sentence that follows
seems clearly intended to be an explanation. This suggests that once
again perhaps Heinze is right, when he infers (p. 37) that the reference
is to a NUVocLL possessed by the soul but not in the same way by the
body, namely the 8iVOqLLq of sensation. The soul has the capacity for
sensation when enclosed in a body. But the body does not have the
capacity of sensation when enclosed in any other receptacle - it needs
something inside itself, namely a soul, to make sensation possible.
The remainder of paragraph 64 is clear in its general meaning
although certain details have seemed unclear. The subject of xocs-
axSuocevcan only be 'the body' (assuming that we keep the manuscript
reading and do not follow Usener's emendation, as Heinze does).
Consequentlythe 6 must be the soul. Divergent views have been taken
as to whether repLxuro6refers to the soul (Bignone, Giussani and
Bailey) or to the body (Arrighetti, translation, and note on p. 470).
As against Heinze I do not think that the reference is to the achieve-
ment of a particular act of perception but to the same iV(qtLqthat
has already been mentioned at the beginning of the sentence and this
must be the general capacity of sensation. This general capacity is
"accomplished" as a result of motion. As the motion of perception
occurs above all in the soul I incline to think that the reference here
is to the capacity of the soul to perceive and so I would follow Bailey
in taking 7repoc&'oas referringto the soul.
For paragraph 65 and the commencement of paragraph 66 it will

95

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions
 probably be sufficient now to offer an expanded paraphrasewith some
added comment:
'For this reason then it is also the case that as long as the soul is
present in the body it never loses the power of sensation although
some other part of the body than the soul be lost. (This is because the
soul has acquired the power of sensation, by being contained in the
body as a vessel as in this way, and only in this way, are the appro-
priate movements possible.) But in the case of a quantity of soul
being destroyed jointly (with a part of the body separated e.g. by
amputation), when the container has been broken either in its whole
or in some part, provided that the container survives, then the soul
that is being destroyed in it preservesits power of sensation.'
This last is certainly a difficult sentence. My interpretation would
take -r a'T?ymoovas the subject of &?taiv - I am not aware that this
has previously been proposed - and the reference as a whole to the
continuation of activity and so presumably sensation in severed limbs
and parts as described in Lucretius III. 634-699. It appears that this
passage in Lucretius has not previously been brought into conjunction
with what Epicurusis saying at this point.
So much for the effect of amputation or loss of parts, provided
the soul has not left the whole or the parts. 'But if the body survives
either in whole or in part, it loses sensation as soon as the number of
atoms has been lost, whatever that number may be that produces
the tension necessary for the constitution of the soul.' This statement
so far from being obscure seems in fact to be rather precise. It suggests
that a loss of part of the soul is sufficient to destroy sensation. This
part is described as constituted by the atoms that provide the tension
and it suggests a sort of loss similar to that of blood-pressurewhich is
sufficient to destroy the power of sensation. This is a much simpler
idea than Giussani's suggestion that Epicurus is arguing against
Aristoxenus. Nor is Bailey likely to be right in thinking that the ten-
sion is an attunement o/ the body into harmony with the soul (Epicurus
p. 232). 'Finally when the whole body is being broken up and destroyed
(reading 8La.Xuo1ovou)then the soul is dispersed and loses the capacity
for sensation since the container is necessary to enable the appropriate
motions to occur.'

University o/ Manchester

96

This content downloaded from 195.221.71.48 on Sun, 8 Dec 2013 15:10:14 PM

All use subject to JSTOR Terms and Conditions