This article was downloaded by: [Moskow State Univ Bibliote]

On: 01 December 2013, At: 09:41

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Symbolae Osloenses: Norwegian Journal of Greek and

Latin Studies
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/sosl20

On the date of democritus

John Ferguson
Published online: 22 Jul 2008.

To cite this article: John Ferguson (1965) On the date of democritus, Symbolae Osloenses: Norwegian Journal of Greek and
Latin Studies, 40:1, 17-26, DOI: 10.1080/00397676508590564

To link to this article: http://dx.doi.org/10.1080/00397676508590564

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the
publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations
or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any
opinions and views expressed in this publication are the opinions and views of the authors, and are not the
views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be
independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses,
actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever
caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 ON THE DATE OF DEMOCRITUS
BY

JOHN FERGUSON
Downloaded by [Moskow State Univ Bibliote] at 09:41 01 December 2013

I
It is notorious that the doxographic tradition for the dates of the pre-
Platonic philosophers is schematic and unreliable.
1. Dates are hung on to the peg of some convenient historical event.
Thus the floruit of Xenophanes is linked to the foundation of Elea;
so is the birth of Parmenides; the floruit of Empedocles to the founda-
tion of Thurii; and, if we accept easy and necessary emendations, the
floruit of Thaïes to the eclipse of 585 and that of Anaximenes to the
fall of Sardis.
2. It is continually said that one philosopher 'heard' or was a
disciple of another. Anaximander was a pupil of Thaïes. Anaximenes
'heard' Anaximander. Pythagoras was associated with Pherecydes
and buried him. Empedocles, Epicharmus and Alcmaeon 'heard'
Pythagoras. Xenophanes 'heard' Anaximander; so did Parmenides;
Parmenides 'heard' Xenophanes; so did Heraclitus; Zeno was a dis-
ciple of Xenophanes or Parmenides; Melissus 'heard' Parmenides.
Empedocles was an associate of Parmenides. Anaxagoras 'heard'
Anaximenes; so did Diogenes of Apollonia. Archelaus was a pupil of
Anaxagoras and teacher of Socrates. Gorgias was a disciple of Empe-
docles. Leucippus 'heard' Zeno. Democritus fell in with Leucippus
and Anaxagoras. Protagoras 'heard' Democritus. The list could be
extended. Some of the collocations are quite fantastic, and few are
based on evidence beyond the sort of deduction from an author's
thought which we can make for ourselves.
3. A man's floruit is assumed to occur at the age of 40 and a forty-
year cycle established from master to disciple. Thus Parmenides is
placed forty years after Xenophanes, and Zeno forty years after
Parmenides.
4. It has not always been noticed that in the schema of the doxo-
2 - 644133 Symbolae Osloensea
 18 JOHN FEBGTTSON

graphers Socrates represented a new beginning (cf. D.L. 2,47). We

do the same today. Kirk and Raven call their book The Pre-Socratic
Philosophers while accepting a date for Democritus which makes him
ten years younger than Socrates. We habitually teach the 'Pre-
Socratics' before going on to Socrates, the Socratic schools, Plato
and Aristotle. We assume that physical philosophy had come to its
conclusion before Socrates struck out along a new path. So did the
ancients. Hence Thrasyllus's date for the birth of Democritus in 01.
Downloaded by [Moskow State Univ Bibliote] at 09:41 01 December 2013

77,3—one year before that of Socrates.

It follows that in dating Democritus or any other of the pre-
Platonics, unless there are firm dates in the tradition such as Thales's
eclipse or Protagoras's association with the constitution of Thurii,
we have to rely on a fresh evaluation of the succession, combined with
such external evidence as may be available.

II
We must next seek to establish the dates of those thinkers whose
chronology is intertwined with that of Democritus. Such an examina-
tion must perforce be summary.
PABMENIDES. Diogenes Laertius, no doubt following Apollodorus,
gives his floruit as 01. 69 (504-0); this is forty years after the founda-
tion of Elea and valueless as evidence. An alternative scheme is
found in Plato's Parmenides. Despite the scepticism of Athenaeus
(11,505F), there is no reason to doubt that Socrates met and conversed
with Parmenides—Plato more than once insists on it (Theaet. 183A;
Soph. 217C)—though there is excellent reason to doubt the historicity
of the particular conversation reported. The meeting, if Socrates was
'pretty young', must have been in the early 440s. Plato goes out of
his way to record that Parmenides was about 65 and Zeno nearing
40 (Parm. 127B); the information is gratuitous and therefore prob-
ably reliable. Parmenides then was born about 512, and probably did
his best work between about 480 and 460.
EMPEDOCIJES. Empedocles's grandfather won a victory at Olympia
in 01. 71 (i.e. 496). We have a definite statement also that the philo-
sopher visited Thurii shortly after its foundation in 443 (D.L. 8,62);
this comes from Glaucus, an early authority, via Apollodorus. This
is all we know; his floruit in 01. 84 (444-1) (D.L. 8,74) is based on this,
and no credence can be given to the figures for his eventual age, whe-
 On the Date of Democritus 19

ther 60 or 109. For the rest we must be content with establishing his
place in the succession. A description in his work is generally taken
as referring to Pythagoras, though some have thought it Parmenides
(fr. 129); in any case his thought is clearly later than either. Anaxa-
• goras, we are told (Arist. Met. A3, 984 a 11) was his senior in years,
but his junior in philosophical activity. We are also told that Gorgias
was his pupil in rhetoric (D.L. 8,58). Probably Apollodorus's date is
on the late side; we cannot say more for certain than that Empedocles
Downloaded by [Moskow State Univ Bibliote] at 09:41 01 December 2013

was active in the mid-fifth century.

ZENO. Zeno's dates we have already established in our discussion
of Pannenides. He was nearing 40 in the early 440s and was therefore
born about 486; further, his book was published about 450, so that
we have a clear date for the man and his work. Apollodorus gives his
floruit in 01. 79 (464-1) (D.L. 9,29) but this is based simply on his
Eleatic cycle. Eusebius, who put him in 01. 81,1-3 (456-4) (DK: A3),
was nearer the mark, though his giving Heraclitus identical dates
impairs his judgment.
PEOTAGOBAS. Protagoras has received pellucid treatment by
Morrison (CQ X X X V (1941) pp. 2-7) and a re-examination by Davi-
son (CQ XLVII"(1953) pp. 33-38). We have a firm date in 443 when
Thurii was founded and Protagoras responsible for its constitution;
this shows that he had an established reputation by that date, and
Apollodorus characteristically places his floruit there (D.L. 9,56).
The fact that Amipsias in the Connus of 423, a satire on contemporary
philosophy, did not introduce Protagoras, but Eupolis in the Gdlaces
of 421 portrayed him as resident in Athens, suggests that he was not
in Athens in 424-3 but was in 422-1 (Ath. 5,2180). Morrison has given
good reason to suppose that Euripides in Palamedes (fr. 588) was
referring to the death of Protagoras; Palameies can be dated to 415 and
this is not incompatible with the statement that he had been prose-
cuted by Pythodorus 'one of the Four Hundred' (D.L. 9,54), since
'one of the Four Hundred' identifies the particular Pythodorus, not
the date of the prosecution. It is tempting to suppose, knowing as we
do that prosecutions on religious grounds in Athens invariably arise
from some political issue, and that Euripides was a friend of Prota-
goras and produced his Troades with Palamedes, that criticism of
Athenian imperial ambitions and Machtpolitik at Melos was the real
issue. Now, Plato tells us (Meno 91E) that Protagoras started his
 20 JOHN FERGUSON

professional career at the age of 30 and lived to be 70, and Apollo-

dorus accepts this 70. Some authorities give 90 (D.L. 9,55 cf. Hesy-
chius in D E : A3), but he does not appear in the Lucianic essay on
longevity, and the evidence of Meno is decisive; the extension is due
to an attempt to rescue Plato's chronology in Protagoras (so Davison
against Jacoby's suggestion of textual corruption). This gives us dates
something like 486-416, with his teaching career starting in the mid-
450s; it is possible that his characteristic sceptical humanism was not
Downloaded by [Moskow State Univ Bibliote] at 09:41 01 December 2013

formulated till after 443. Three other factors in the tradition remain to
be examined. In Plato's Protagoras he describes himself as having
spent many years as a teacher and lived many years and as being old
enough to be the father of any of those present. The absolute date is
impossible to determine; Plato was not nicely consistent about his
dramatic dates. The presence of Pericles's sons implies a date before
429; the presentation of Pherecrates's Wise Men 'last year' implies
419. Even if we accept a dramatic date of 432 (but we must stress
Plato's inconsistency about dramatic dates: note also the established
reputation of Prodicus) Protagoras's words are quite compatible
with twenty-five years' teaching and an age in the mid-fifties. As to
the relative age, these figures make him seventeen years older than
Socrates, and the words 'I'm old enough to be your father' are, with
all respect to Davison, used very loosely and with an element of exag-
geration. The words as they stand are in fact physiologically possible
and psychologically plausible; to someone thirty years younger I do
not say 'I'm old enough to be your father' but 'I'm old enough to be
your grandfather'. A second factor is Philostratus's statement that
he was educated by the Magi who accompanied Xerxes to Greece in
480 (DK: A2). A similar story is told of Democritus. Davison makes
much of this, but it is not really very likely. I t may represent a genuine
tradition of Persian influence, conflated with Herodotus's record of a
Persian stop at Abdera (7,109; 8,120); it is possible that some Magi
were left behind in Abdera. Thirdly, the tradition that Protagoras
was a pupil of Democritus is impossible. Democritus cannot have been
teaching between 465 and 455, let alone earlier. There is a tendency
to confuse the traditions about the two thinkers (e.g. D.L. 9,50 with
Clem. Al. Strom. 6,32; Suid. s.v. Δημόκριτος).
ANAXAGORAS. Anaxagoras raises the most acute chronological prob-
lems of all. Again we have a firm date: the Aegospotami meteorite
 On the Date of Democritus 21

of 467 gave him his views on the sun. He was a friend of Pericles;
Plutarch's celebrated anecdote of the one-horned ram refers to the
year 443 or just before (Plut. Per. 6). Pericles is generally called his
pupil (Isoc. 15,235; D.L. 2,12; D.S. 12,39; etc.), but this need not be
taken too seriously, and the language of Plato rather implies that he
was not in any real sense a pupil (Phaedr. 270A cf. 1 AL·. 118C).
Pericles was still learning from Damon in his old age, and there is no
need to assume that Anaxagoras was teaching in Athens when Pericles
Downloaded by [Moskow State Univ Bibliote] at 09:41 01 December 2013

was a youngster. Diogenes Laertius (2,7) is exceptionally muddled,

even for him: 'He is said to have been twenty years old at the time
of Xerxes's crossing' (i.e. 480) 'and to have lived to be 72.' Apollo-
dorus in his Chronicles says that 'he was born' or possibly 'flourished'
(see Davison CQ XCVII (1953) 40) 'in 01. 70' (i.e. 500-497) 'and
died in 01. 78,1' (i.e. 468-7, but Scaliger emends plausibly to '01.
88,1' (i.e. 428-7). 'He began to philosophize at Athens at the age of
twenty, when Callias was archon' (i.e. 456-5, though some editors
emend Callias to Calliades i.e. 480-79) 'as Demetrius of Phalerum
tells us in his register of archons, and they say he spent thirty years
there.' Now, Anaxagoras's philosophical activity must be later than
Parmenides and Zeno, and if we accept Aristotle's statement at its
face-value (Met. A3.984 a 11; the meaning is not absolutely certain),
he was older than Empedocles, but later in philosophical activity.
The most reasonable harmony seems to be birth about 500; this is
supported for what it is worth, which is not much, by Eusebius (DK:
A4), who lumps Heraclitus and Democritus together with Anaxagoras.
Then there would be a late start to philosophical activity in 456-5,
perhaps with his distinctive contribution emerging only in the 440s
and 430s, and his death in 428-7. He was, we are told, a 'one book'
man, and there is no reason against, and much for the publication of
that book about 445-0. There remains the question of his trial or
trials. This has been examined by Taylor (CQ X I (1917) 81-7), Wade-
Gery (JHS LII (1932) 220) and Davison (CQ XLVII (1953) 39-45).
There are two traditions of the trial (D.L. 2,12), one from Satyrus that
he was prosecuted by Thucydides for impiety and medism, one from
Sotion that he was prosecuted by Cleon for impiety. Taylor argued
for the early date, but Wade-Gery's examination weakens his case.
Davison suggested that both trials are authentic. It is perhaps signi-
ficant that tradition, which usually has no difficulty in swallowing
 22 JOHN FERGUSON

'doublets', is here sceptical. If the date here suggested for Anaxa-

goras's philosophical activity is right, the earlier prosecution is most
unlikely. However, unless it could be proved that the earlier trial was
historical and therefore Anaxagoras's philosophy must be pushed
back earlier, the prosecutions, interesting as they are historically, do
not affect our evaluation of Anaxagoras's place in philosophy. Strabo's
statement that Anaxagoras was an associate of Anaximenes will not
stand, but it is perfectly possible that Euripides sat at his feet, and
Downloaded by [Moskow State Univ Bibliote] at 09:41 01 December 2013

that Archelaus also learned from him and succeeded him at Lampsa-
cus (where he spent the last years of his life) (DK: A7).
MEUSSTJS. With Melissus we have one of our rare firm dates: he
was the Saraian admiral who defeated Pericles in 441-0 (Plut. Per.
26). This is no doubt why Apollodorus gives his floruit as 01. 84
(444-1) (D.L. 9,24). We are told that he studied under Parmenides, and
this may well be true: it would be compatible with a birth in the 480s.
He returned to Samos and became a statesman of repute, and then
a naval commander. All this leaves little time for constructive philo-
sophy. There is reason to believe that Melissus's work is considerably
later than that of Zeno (Kirk-Raven, The Pre-Socratic Philosophers,
298 ff.) and indeed that some of his arguments are directed against
Anaxagoras (Burnet, Early Greek Philosophy, 328) and Ion of Chios
(Kirk-Raven, op. cit.); we are justified in seeing Melissus as a man who
returned to philosophy as an elder statesman, and in dating his one
book to the 430s.
SOCKATES. The dates of Socrates are not in doubt; he was executed
in 399 at the age of 70, which gives a birth-date of 469. Apollodorus
gives a birth-date of 01. 77,4 (469-8) (D.L. 2,44). I shall argue else-
where that his interests were predominantly scientific until about 421,
so that his ethical prominence is likely to be later.
LEtrciPPtrs. We now come to thornier ground. Leucippus, says
Kathleen Freeman justly, is the most shadowy figure of early Greek
philosophy. Epicurus (D.L. 10,13) appears to have asserted that he
did not exist, but this may be based on a misunderstanding. (See
Burnet, Early Greek Philosophy, 330 n. 2; Kirk-Raven, The Pre-
Socratic Philosophers, 402.) Exist he must have done; there is too
much fourth century evidence to deny it. Aristotle asserts very rea-
sonably that Leucippus devised the atomic theory as an answer to the
paradox of the Eleatics (De Gen. et Corr. Δ8, 325 a 23). Hence the
 On the Date of Democritus 23

doxographic tradition makes him a pupil of Zeno (D.L. 9,30 cf. DK:
A4; 5) or Melissus (DK: A5). There is no special reason to believe this.
Logic compels us to believe that hie theory stands in line of succession
to those of Empedocles and Anaxagoras, who faced the same chal-
lenge with less radical solutions. Zeno's book can be dated to about
450, Empedocles was active at about the same time, the creative work
of Anaxagoras must therefore be dated later, to the 440s or even the
430s, that of Melissus belongs to the 430s, and Leucippus must be later
still. "His influence on Diogenes of Apollonia is sometimes assumed
Downloaded by [Moskow State Univ Bibliote] at 09:41 01 December 2013

to provide a terminus ad quern of 423 (when Diogenes's views are

represented in the Clouds). But that eclectic could have swallowed
new ideas at almost any stage of his life. In any case I agree with
Diïler (Hermes 76 (1941) 359 ff.) that there is no Leucippan element
in Diogenes worth speaking of, though I cannot accept his chrono-
logical conclusions, which put both Diogenes and Leucippus before
Melissus. The further problems of dating Leucippus are best linked to
Democritus.
m
DEMOCBITUS. Suppose that for a moment we forget the doxographic
tradition. We are left with the following points:
1. There is no shadow of the atomic theory in Aristophanes or in
the fragments of fifth century comedy. In the Clouds the theories of
Anaxagoras and Diogenes are the latest thing. If the above account
of Socrates is right, this is confirmed by the Phaedo. Even in the
Apology (26D) Anaxagoras appears to be the last word in materialistic
atheism.
2. There is no clear allusion to the atomic theory in Plato before
the Sophist where the Atomists are almost certainly included and
probably primarily intended under the heading of the materialist
Giants (246A ff.), though even this has been questioned (e.g. Bumet,
Greek Philosophy I 279; Taylor Plato 384).
3. After Democritus there is a gulf in the history of atomism till
Epicurus at the end of the fourth century, peopled by the shadowy
figures of Nessas, Metrodorus, Diogenes of Smyrna, Anaxarchus and
Nausiphanes.
This is the negative evidence and it by itself would point to the
conclusion that atomism belonged to the late fifth or fourth century.
 24 JOHN FEBGTJSON

4. Democritus's theory of the subjectivity of taste and feeling (e.g.

fr. 11) seems to be clearly post-Protagoras. The saying that taste is
subjective, but truth and morality objective (fr. 69) is surely an an-
swer to Protagoras's doctrine that truth is relative (cf. Plat. Theaet.,
166D ff.). Plutarch (1108F) shows us Democritus contradicting a
view of Protagoras. Heath (Greek Mathematics, I 179) suggested that
Democritus's work On the Contact of a Circle and a Sphere was also
written against Protagoras (cf. Arist. Met. B2, 998 a 3).
Downloaded by [Moskow State Univ Bibliote] at 09:41 01 December 2013

5. Democritus's political thought, conveniently set out by Havelock

with a highly controversial interpretation, a wholly unhistorical
linking of Democritus to Athens, and a far too early date (The
Liberal Temper in Greek Politics, 125-54; the important fragments
are DK 245 ff.), contains a very mature version of the social contract
which suggests that it comes late in the sophistic period.
6. Some aspects of his ethical thought clearly owe a debt to Socrates.
The most striking example is a saying 'It is worse to do wrong than
to be wronged' (fr. 45, admittedly a saying of 'Démocrates', but
there are not now many separatists). So too 'Ignorance of the better
is the cause of wrong-doing' (fr. 83) must surely be post-Socratic.
Another saying.(/r. 49) that it is not easy to be governed by someone
worse than yourself is close to Socrates's words in Rep. I 347C, but
these are less certainly Socratic, and the fact that Thrasymachus's
attitude to justice in the Republic is unlike a fragment of the historical
Thrasymachus (fr. 8) and that Democritus (fr. 267) declares that it is
natural for the stronger to rule leads one to think that Plato may
have had Democritus in mind in composing Republic I. Socratic
however is the emphasis on the psyche (fr. 37; 170; 171), and the
general mood of many of the other ethical maxims seems Socratic.
7. Democritus's vocabulary is very remarkable. It is Ionic in form,
and contains Herodotean words; it is poetic and draws on Homer
and Aeschylus; and it contains words which he patently coined him-
self. Apart from these there are a significant number of words which
are acclimatized to the fourth century, μεγαλοψυχία for instance
(fr. 46), first in Isocrates, Demosthenes and Plato; and those fourth-
century key-words ομόνοια (fr. 250), αυτάρκεια (246) and ευεργετεϊν
(248). ομόνοια is found in late Thucydides (8,93) and Andocides
(1,140), it is true, as well as Antiphon the sophist, but αυτάρκεια can
only claim a Hippocratic epistle before Plato, and ευεργετεϊν, a
 On the Date of Democritus 25

good verse form, appears in prose in Xenophon, Plato, Demosthenes

and Lycurgus. φιλαργυρία (222) is also in the Hippocratic letters;
also Isocrates and Demosthenes, εύθυμειν (3), another poetic word,
is first found in prose in Xenophon; ευγνώμων (231) is first in Ando-
cides and Xenophon. This vocabulary points to a date in the early
fourth, or perhaps very late fifth century. Unfortunately the direct
quotations from Democritus pertain largely to his ethical and general
writings; for the physics we rely almost exclusively on second-hand
accounts. One direct quotation inculcating the principle of 'like to
Downloaded by [Moskow State Univ Bibliote] at 09:41 01 December 2013

like' may come from the physics (164): amongst the Ionic words and
peculiar coinings are found διακριτικώς (the adjective appears first in
Plato), ομογενής (in prose first in Plato), περιφερής (Hippocratic
corpus and Plato), συναγωγός (Plato).
The natural conclusion of this is to place the ethical works in the
fourth century, and the atomic theory at the very end of the fifth, and
in no circumstances prior to 420. W. Kranz saw this fifty years ago,
a sure sign of his grasp of essentials {Hermes 47 (1912) 42).
We must now examine the doxographic tradition. That his birth
fell in 01. 70 (500-497) (DK: A4) is due to an attempt to make him a
pupil of the Magi in 480 (the same story is told as with Protagoras)
(D.L. 9,34); the date 494 (DK: A5) is the fall of Miletus; ThrasyUus's
date of 01. 77,3 (470-469) merely puts him one year above Socrates
(D.L. 9,41 cf. DK: A2); so too Aulus Gellius (17,21); 459 (DK: A2,4)
has more to commend it, and may be not far out; so also Apollodorus
(D.L. 9,41). We have his own words (jr. 5) to tell us that he was forty
years younger than Anaxagoras: the number need not be exact, but
the dates happen to fit. By most of the traditions he was long-lived:
no one supposes he died before 404 (DK: A5), and some extend his
life to 359 (DK: A4). A date for his death between about 377 and 349
is suggested by Anaxandrides's description of a girl as γελασίνη {fr.
25 Edmonds); the only other occurrence of the word at this period is
in the familiar epithet of Democritus (Ael. V.H. 4,20); it is likely
therefore to belong to his lifetime or even to mark his death.
These limits for his life (b. 459 and lived well into the fourth cen-
tury) would be widely agreed. The question is—at what point in his
life did he do his constructive work? We have an explicit statement
of his to the effect that he wrote The Little World-Order, which we
may assume was his most important work in physics, 730 years after
 26 JOHN FEBGT7SON

the capture of Troy (D.L. 9,41). The most widely accepted date for
this event was 1183, which seems to have been the view of Ctesias
and which was canonized by Eratosthenes. But this was only one of
many chronologies (see, conveniently, Forsdyke, Greece Before Homer,
62 ff.). 453 is too early for Democritus's book, with Leucippus to
precede him, by almost any standards. We can discount all views,
which include that of Herodotus, which place the fall of Troy earlier
than 1183. We are left with Sosibius's date of 1171, giving 441 for The
Little World-Order, which is still too early, and Ephorus's date of
Downloaded by [Moskow State Univ Bibliote] at 09:41 01 December 2013

1135. If we accept the dates given above for Anaxagoras, 405 is the
only possible date for The Little World-Order consistent with what
Democritus himself tells us.
We conclude therefore that Leucippus was propounding his views
sometime between 430 at earliest (if he influenced Diogenes of Apol-
lonia before the Clouds) and 410 at latest, that Democritus offered his
major contribution to the atomic theory in 405, and in the early
fourth century turned increasingly to ethical problems. The list of his
works is large, and it is possible and indeed likely that from 430-410
he was engaged in mathematical and literary studies and in his no-
table biological observations.