L'antiquité classique

Athens' Difficulty Sparta's Opportunity : Causation and the

Peloponnesian War
C. A. Powell

Résumé
Entre 465 et 404, les Spartiates ont agi envers les Athéniens avec un opportunisme politique d'une rare intelligence. Les faits de
cette période - la portée n'en a pas été suffisamment soulignée - montrent comment Sparte s'entendait à exploiter la moindre
faiblesse d'Athènes, mais aussi à devenir prudente sitôt qu'elle avait lieu d'appréhender une nette résistance de sa part. Son
comportement n'était pas différent à l'endroit de certaines autres cités. L'attitude qu'elle prit en 431 résulta des difficultés que les
Athéniens connaissaient avec Potidée et avec ses voisins. Une analyse des événements d'alors, si elle tient compte de la
doctrine générale de la limitation des causes, conduit à élucider les motivations profondes des Spartiates.

Citer ce document / Cite this document :

Powell C. A. Athens' Difficulty Sparta's Opportunity : Causation and the Peloponnesian War. In: L'antiquité classique, Tome 49,
1980. pp. 87-114;

doi : https://doi.org/10.3406/antiq.1980.1966

https://www.persee.fr/doc/antiq_0770-2817_1980_num_49_1_1966

Fichier pdf généré le 18/12/2018

 M

ATHENS'
CAUSATION
DIFFICULTY,
AND THESPARTA'S
PELOPONNESIAN
OPPORTUNITY
WAR

Controversy over the origin of the Peloponnesian War bids to

become a fixed part of the inheritance of Ancient Historians and itself
almost a "possession for ever", in spite of Thucydides' attempt to settle
the matter l. That the developments of the 430s were not of great
importance for the causation of the war has recently been argued with
unusual cogency, by de Ste. Croix 2. He sees the Spartan decision of
432 to begin hostilities as belonging to a pattern of aggression against
Athens which extended over the previous decades 3. On the other
hands, Meiggs is not alone among modern scholars in assigning rather
more significance, for the outbreak of the war, to the expansion of
Athenian power evident in the 430s 4. Kagan writes that "... it was not
the underlying causes but the immediate crisis [i.e. of the late 430s] that
produced the war" 5, and Sealey, in a work revised in the light of de
Ste. Croix's, nevertheless holds that "... the reasons why the Spartans
went to war in 431 should be sought, not in an outlook prevalent as
early as 440, but in the actions taken by the Spartans in the years
immediately preceding the outbreak of the war" 6. It is hoped to show
here that there is, during the last 60 years of Athenian greatness, a
remarkable though seldom-noticed consistency in the foreign policy of
Sparta ; that her aggression towards Athens was both more profound
and more intelligently applied than has usually been realised. It will be

1 I, 23, 5. (References are to Thucydides, save where otherwise indicated.)

2 G. E. M. de Ste. Croix, The Origins of the Peloponnesian War, passim a work
which, as a contribution to Greek historiography, perhaps has not been surpassed in
English since the time of Grote. Debts to this work will frequently be apparent in what
follows.
3 Op. cit., pp. 3, 50, 180.
4 R. Meiggs, The Athenian Empire, pp. 203 f.
5 D. Kagan, The Outbreak of the Peloponnesian War, p. 346 ; cf. p. 300.
6 R. Sealey, in Class. Phil., 70, 2 (1975), p. 107.
öö C. ?. POWELL

argued that the revolts of Potidaea and her neighbours in 432 were
probably necessary for the outbreak of war in the following year 7. In
addition, an attempt will be made to elucidate the origin of the war by
examining an aspect of the theory of causation which is of general
importance for Ancient Historians.
Two distinct questions are thematic in what follows. "Did the
Spartans, from 465 to 405, always or usually wish to attack Athenian
interests when some weakness of Athens gave them a good opportunity
for doing so?" And, "Did Sparta consistently refrain from new
aggression except when there existed a special Athenian weakness?".
On the first question, several scholars have noticed Sparta's desire to
exploit various Athenian crises of the Pentecontaetia. Thus Ed. Meyer
states that the Spartans "did not pass up the opportunities when it
seemed possible to weaken Athens, like Thasos' request for aid" 8.
Andrewes points out the connection between Athenian difficulties and
the Spartan campaigns to Tanagra and (in 446) against Attica 9, while
de Ste. Croix notes hypothetically of the period after 432 that "the
Spartans could renew hostilities and launch an invasion of Attica at any
time that seemed favourable to them the revolt of a major Athenian
ally for instance. They had intended to break the Thirty Years Peace in
440, and they did break it in 432/ 1 ; they could hardly be relied upon
to resist a really tempting opportunity" 10. However, these
although accurate, do not fully answer our question. It is not easy
to find scholarly comment which deals squarely with this question, or
with the second one, concerning Spartan restraint. To prove an
affirmative answer in the latter case that Sparta in our period never
began aggression except at times of Athenian weakness would be to

7 "Necessary" in the sense that had they, or some other similar event, not occurred,
the war would not have broken out when it did. The sense of 'similar' will become
apparent from what follows.
8 Forschungen zur alten Geschichte, II, p. 311 ; cf. Meyer, Geschichte des
Altertums, IV, 1, p. 713, on Sparta's "incomparable opportunity ... to make a sudden
attack on Athens while her naval power was seriously engaged", during the Samian
revolt of 440-39. Cf. also D. MacDowell, in MS, 80 (1960), p. 121.
9 A. Andrewes, in CQ (n.s.), 9 (1959), p. 235 ; cf. A. H. M. Jones, in Proc. Camb.
Phil. Soc, (n.s.) 2 (1952-3), pp. 43 f. (quoted below, p. 1 12) ; P. A. Brunt, in Phoenix,
19 (1965), p. 258 (on the revolt of Megara from Athens) ; Meiggs, op. cit., p. 99 (on the
Tanagra campaign).
10 Op. cit., p. 209; cf. pp. 169, 203.
 ATHENS'
DIFFICULTY, SPARTA 's OPPORTUNITY 89

show that the Spartans in 431 probably would not have gone to war
save for the special Athenian difficulty then. It would also (as will be
argued) undermine traditional ideas about the significance of Athenian
growth in power in the 440s and 430s. If the question concerning
Spartan restraint has been neglected, it is not because of its
unimportance, or because the answer is too well known for scholars to
refer to in print. A degree of unoriginality is common in accounts of the
outbreak of the Peloponnesian War. It is now contended that the
familiar emphasis on Athenian expansion in the 430s is perverse, and
Athens'
that the timing of Sparta's attack then is to be explained not by
new strength but by her new weakness.
Below are listed synoptically two kinds of occasion within the period
465-405. A detailed account of principles used in compiling this, and
the subsequent, table will be found below ; in brief, however :
Column I lists those occasions which, after at least three years
without extra-Peloponnesian aggression against Athenian interests by
Sparta, the Spartans may have identified as their best opportunities for
seriously damaging the power of Athens. Such occasions, coming
when the Spartans thought themselves free, or possibly free, to act,
included the revolts or imminent revolts of major Athenian allies,
revolution and treachery (actual or threatened) within Athens, and the
withdrawal from Attica of very large numbers of troops in such a way
that they could not quickly return to help Athens, even if needed.
Column II contains the instances of Sparta's having hoped or decided
to open hostilities with Athens outside the Péloponnèse, again after a
period of at least three years without such conflict.
I II
465, revolt of Thasos n. 465-4, the Spartans promised the
Thasians to invade Attica, and
"were likely to do so" n.

11 I, 100, 2. References to events included in the following lists will in most cases
involve only the first notice of them by Thucydides.
12 I, 101, 2. It seems wrong to question Thucydides' report of Sparta's planned
aggression on the grounds that the Athenians perhaps heard about the plan and would
not shortly afterwards have sent their help against the helots if they had heard. In 42 1 ,
after 1 0 years of violent hostility, Athens swore to give Sparta similar help, and at a
time when, because of helot unrest, it could have seemed likely that her good faith
would shortly be tested (V, 23, 3 ; IV, 41 , 3 ; cf. 55, 1 ; 80, 2 ff.). It seems possible that
90 CA. POWELL

c. 460-455, Athenian expedition to 458-7, Spartan campaign to

Cyprus and Egypt 13 ; c. 459-7, gra 16.
Athens' war with Aegina 14 ; 458-7,
threat of treachery at Athens 15.
446, defections of Euboea and 446, Spartans invaded Attica 18.
n.
440-39, revolts of Samos and 440-439, conference of
19. league voted on whether to aid
the Samians 20.
445-435 (?), proposed revolt of

432, revolts of Potidaea, the Bot- 432, Spartan decision in favour of

tiaeans and Chalcidians 22. war 23.
415-3, Sicilian expedition. 415/4, Sparta arranged help for
Syracuse, and planned to garrison
Decelea 24.

the Athenians did learn of Sparta's promise to Thasos shortly after it was made, that as
a result they strongly desired to conciliate their former "yoke fellow", and that this
desire was expressed in the sending of Cimon's expedition. The abrupt reversal of
Athenian policy involved in making alliances with Argos and Thessaly (for long
enemies of Sparta but also inconvenient allies for the mistress of the Delian League)
could in this case have been caused by the memory of Sparta's promise to Thasos,
which would have made war seem likely, once the attempt to conciliate Sparta had
failed.
13 I, 1 04, 1 f. If the helot revolt ended in "the tenth" rather than "the fourth" year, it
may be that Sparta did not identify Athens' Egyptian and Aeginetan distractions as an
opportunity for aggression until the time of the Tanagra campaign. However, see de
Ste. Croix, op. cit., p. 188, on the possibility of active warfare between Sparta and
Athens orior to Tanagra and during an early stage of the Egyptian campaign (I, 105,
1 ff.).
14 I, 105, 2 ff.
15 I, 107,4; 6.
16 I, 107,2-4; 108, 1.
17 I, 114, 1.
I, 114, 1 f.
I, 115, 2 ff.
I, 40, 5; 41, 2 ; Jones, loc. cit., pp. 43 f. and esp. de Ste. Croix, op. cit., pp. 200-
203.
III, 2, 1 13, 1.
I, 58, 1.
I, 87, 3 ; 81
VI, 93, 2 f.
 ATHENS'
DIFFICULTY, SPARTA 's OPPORTUNITY 91

There is, at least at first sight, an extensive, close and suggestive

correlation between Sparta's best opportunities to begin a war of
aggression against Athens and the occasions on which she actually
wished to do so. But may the correlation nevertheless be, to an
important extent, fortuitous? Evidence may be gained by examining
Spartan behaviour when similar, exploitable, Athenian difficulties
occurred in wartime or, more precisely, at times when extra-
Peloponnesian aggression by Sparta had taken place within the
previous three years. If Sparta's decisions to begin aggression were
governed by Athenian distractions, one might expect the Spartans to
have made special efforts, at least occasionally, to exploit similar
distractions when they occurred at times of existing hostility. And if a
close and extensive correlation could be shown for such times, between
Athens' greatest difficulties and special Spartan aggression, that would
almost eliminate the chance that our previous correlation was largely a
matter of fortuitous coincidence. The following table relates to times of
pre-existing hostility :
Column III collects those occasions (including the occurrences of the
plague 25) which may have seemed to the Spartans the most opportune
for damaging severely the interests of Athens.
Column IV lists the instances of Sparta's having opened an extra-
Peloponnesian campaign against those interests in a new region 26.
Ill IV
430-28, plague at Athens 27. 430, Spartan expedition against Za-
cynthos 28. 429, Spartan-led
against Acarnania 29.

25 Cf. V, 41, 2 and infra, p. 95.

26 Any definition of a new region is bound to be somewhat arbitrary, in point both
of space and time. "New" here is applied to areas in which Sparta had not had armed
forces during the previous three years. Alternative notions of "new region" would
probably be such as to reduce, rather than increase, the number of cases considered as
new Spartan aggression : such redefinition would yield a correlation which, while less
extensive than the one presented here, would still suggest that Sparta depended on
special opportunities for her aggressive initiatives.
27 II, 47, 3.
28 II, 66.
29 II, 80, 1 ff.
92 CA. POWELL

428-7, revolt of Lesbos 30. 427, Spartan-led fleet sent to E.

Aegean to help Lesbos 31.
427, revolution at Corcyra 32. 427, Sparta sent fleet to Corcyra 33.
427/6-6/5, recurrence of plague at 426, Sparta sent force against
Athens 34 ; (special vulnerability of 36 ; battle of Olpae (426/5) 37.
Naupactus 35).
425/4, allies and former allies of 424, expedition under Brasidas sent
Athens appealed for Spartan aid to to Thraceward region 39.
Thraceward region 38.
413-2, aftermath of Sicilian 412, Sparta assembled large fleet in
: revolts (proposed or E. Aegean ; aided revolts of Chios,
of Chios, Erythrae, Lesbos, Erythrae, Lesbos, Cnidos,
Euboea, Cnidos, Rhodes 40.
412-1, Athenian siege of Chios 42 ; Spring 411, first Spartan intrusion
Spring 411, rule of the Four into Hellespontine area44.
at Athens imminent or
43.

30 III, 2, 1.
31 III, 16, 3; 26, 1.
32 III, 69, 2 ; 70, 1 ff.
33 III, 69, 2 ; 76.
34 III, 87, 1.
35 III, 102, 3 f., referring to the reluctance now of the Acarnanians in defending
Naupactus. Cf. Ill, 98 for the previous Athenian defeat by the Aetolians, with the
consequent loss of prestige in the area.
36 III, lOOff.
37 III, 106 ff.
38 IV, 79, 2. Compare the further appeal to Brasidas in 424/3 of (unidentified)
Athenian allies wishing to revolt (IV, 108, 3 ff), causing him to appeal in turn to
Sparta for additional forces (IV, 108, 6). After an initial Spartan reluctance (IV, 108, 7),
armies were sent to reinforce him (IV, 132, 2 f. ; V, 12 f.).
39 IV, 70, 1 ; 78.
40 VIII, 5, 4 (Chios and Erythrae) ; 5, 1 (Euboea) ; 5, 2 (Lesbos) ; 35, 1 (Cnidos) ; 44
(Rhodes).
41 VIII, 12, 3 ; 23, 1, 5 ; 26, 1 ; 35, 1 ; 39, 1 (assembly of large fleet) ; VIII, 14 ; 22-
23, 4 ; 24, 6 ; 35, 1 f. ; 44, 1 f. (aid to revolts).
42 VIII, 24.
43 VIII, 63, 3.
44 VIII, 62, 1 .
 ATHENS'
DIFFICULTY, SPARTA 's OPPORTUNITY 93

411, rule of the Four Hundred ; 411, Peloponnesian ships promoted

opposition from Athenian fleet at revolts of Byzantium47 and Eu-
Samos ; Byzantium offered to re- boea 48 ; Peloponnesians overran
volt 45 ; secession of Thasos 46. Aegina 49.
405, Athenian disaster at Aegos- 405-4, first Peloponnesian naval
potami. blockade of Athens 50.
During wartime Sparta appears never to have been willing to extend
hostilities against Athens to a new area without there existing first
an especially good opportunity for doing so. And when such an
existed, Sparta always, or almost always, exploited it by
campaigning in a new area. A parallel with Spartan behaviour at other
times is now apparent : the correlation between Athenian difficulties
and the undertaking of aggressive wars by Sparta seems not to be a
matter of mere coincidence. The opening of a war, and the extension to
a new area of an existing war, had in common two features which
distinguished them from the continuation of hostilities in territory
where conflict had taken place very recently. Both kinds of new
aggression occurred when confidence could not rest on the most
recent, direct, experience ; and both involved a new commitment to
persist until successful in the aggression thus begun, on pain of losing
important prestige 51. One might then in any case have expected to find
rather similar conditions, whether pf unusual opportunity for
aggression or of unusual pressure, tending to precede both kinds of
initiative by Sparta. But it may now appear to have been special
opportunity rather than special pressure which determined the timing
of her assaults upon Athenian interests. The patterns outlined may hold
a key to 60 years of Spartan foreign policy. Before we consider more
fully the method used to compile the above lists, is there any direct
evidence that the kind of strategic thinking suggested by them actually
was practised in our period ?

45 VIII, 80, 2.
46 VIII, 64, 3 f. ; cf. 64, 5 for other allied secessions at this time.
47 VIII, 80, 3.
48 VIII, 95.
49 VIII, 92, 3.
50 Xen., Hell., II, 2, 9.
51 Cf. I, 81,5.
94 CA. POWELL

Thucydides and Aristotle show that many Greeks were indeed

familiar with the principle of making unusually great efforts to exploit
the unusual weakness of an opponent. The invasion of the Megarid in
c. 459 was led by the Corinthians, "thinking that the Athenians would
be unable to come to the help of the Megarians, since they had large
armies away in Egypt and Aegina ; and that if the Athenians did in fact
come to help, they would have to abandon Aegina" 52. Mytilenean
envoys, appealing in 428 for Spartan and allied help against Athens, are
represented as arguing thus : "You have an opportunity such as you
never had before. Disease and expenditure have wasted the Athenians :
their ships are either cruising round your coasts or engaged in
blockading us ; and it is not probable that they will have any to spare, if
you invade them a second time this summer by sea and land ; but they
will either offer no resistance to your vessels, or withdraw from both
our shores" ". The appeal of the Mytileneans was followed by an
extraordinary Spartan effort against Athens. Not only was a fleet
despatched to the eastern Aegean (so far as is known the first to have
operated there under Spartan command for 50 years), but an attempt
was made to mount a second invasion of Attica within the year 428,
which only failed when the allied contingents proved slower than that
of Sparta herself to arrive at the Isthmus 54. According to Thucydides,
the Athenians perceived (αίσθόμενού that the enemy's preparations at
this time were made because of contempt for Athens' weakness (Sea
χαχάγνωσιν ασθενείας σφών) 5S. The Sicilian disaster, and the prospect of
revolt by Athenian allies, are among the causes mentioned by
Thucydides for the optimism with which the Spartans in 41 3/2 decided
απροφασίστως απτεσθαι . . . τοϋ πολέμου 56. Various attempts by Sparta to
exploit the Athenian revolution of 411 have been noticed above57.
Especially striking, in addition, are the actions now of King Agis, who
"sending for large reinforcements from Péloponnèse, not long
afterwards, with these and his garrison from Decelea, descended to the

52 I, 105, 3.
" III, 13, 3 f. (Crawley's translation).
54 For the fleet, see n. 31 supra ; for the attempted second invasion of 428, III, 15,
1 f. ; 16, 2.
55 III, 16, 1.
56 VIII, 2, 1-4.
57 Supra, nn. 41, 47-49.
 ATHENS'
DIFFICULTY, SPARTA 's OPPORTUNITY 95

very walls of Athens ; hoping either that civil disturbances might help
to subdue the Athenians to his terms, or that, in the confusion to be
expected within and without the city, they might even surrender
without a blow being struck (αύτοβοεί) ; at all events he thought he
would succeed in seizing the Long Walls, bared of their defenders" 58.
The times of an opponent's weakness were also used, and expected
to be used, for the opening or re-opening of war. It was through fear of
such opportunism that Sparta and Argos, when arranging in 420 to
decide their rival claims to the Cynurian land by battle at some future
point, stipulated that the challenge to battle should not be issued when
either state was distracted by a separate war or by plague 59. Numerous
members of the Athenian Empire were willing to revolt in the
aftermath of the Sicilian disaster 60. Nicias is shown insisting, before the
Sicilian campaign, that the failure or division of Athenian forces would
be followed swiftly by a Peloponnesian attack on Athens 61. And when
the Spartans did recommence their attacks on Attica in 413, they are
said by Thucydides to have been chiefly fortified by two
: that the Athenians had been the first to break the peace treaty,
and that Athens would be the easier to reduce because involved in a
double war, against Sparta and in Sicily 62. Thucydides implies a fear
on the part of Sparta herself in 42 1 that the helots would exploit her
special weakness at the time63. And Aristotle, in referring to the
frequent risings of the helots, notes that "they lie in wait, as it were, to
take advantage of the Spartans' misfortunes"64. The Spartans
themselves, it now appears, lay in wait, as it were, to exploit the
misfortunes and distractions of Athens. Helot and master evidently
shared more than the Greek language.

58 VIII, 71,1 (Crawley's translation). Thucydides, commenting on the failure of

the Spartans to sail against Piraeus immediately after their capture of Euboea in 4 1 1 ,
claims that "here, as on so many other occasions, they proved the most convenient
people in the world for the Athenians to be at war with" (VIII, 96, 5). Surely this is an
exaggeration. Sparta at about this time began to operate with a large fleet on Athens'
vital corn route through the Hellespont (VIII, 99 ; 103, 1 ; 107, 2) ; we find that the
Spartan fleet which won Euboea is shortly afterwards summoned to that area —
hardly the action of the most convenient enemy in the world (VIII, 107, 2).
59 V, 41, 2.
60 VIII, 2, 1 f.
61 VI, 10, 2; 4; 11,4.
62 VII, 18, 2.
63 V, 14, 3.
64 Pol., 1269a38 f.
96 C. A. POWELL

Moving now to the principles used in compiling the tables above : in

the identification of Sparta's best opportunities for aggression,
allowance has been made for the varying circumstances of Athens.
Thus Athens was far weaker during the period 410-06, which has not
been counted as one of special opportunity for Sparta, than in (for
example) 432-1, a period which has been so identified. No
inconsistency need be involved here ; by the standards of the time
Sparta did not have an unusually good chance to begin fruitful
aggression in 410-06, whereas she did at the earlier period. An analogy
may illuminate. A speculator need not be inconsistent in being
unwilling to buy at £ 1 shares of a kind which he has previously
bought at £ 2. For one thing, when the share is at £ 1 the speculator
may reasonably hope for a further fall in value, whereas when earlier
the share was at £ 2 there were perhaps no obvious grounds for
predicting a fall. The desire of the speculator to buy as cheaply as
possible is matched by the need of Sparta, with her chronic manpower
shortage, to conquer Athens as cheaply as possible. Sparta in 432-1
could not reasonably predict that Athens would decline much in power
unless there was new intervention against her by the Peloponnesian
league. Indeed, a long-term increase in Athenian power would have
seemed more likely. Nor did Athenian difficulties tend to follow each
other so quickly that a second distraction could have been expected to
arise while the Thraceward embarrassment lasted. The affair involving
Potidaea must have seemed to the Spartans at the time to offer one of
their best opportunities for aggression 6S. In the period 410-06 Athens
was much weaker, but the Spartans then had reason to suspect that she
would shortly become weaker still, through some new revolt, or defeat,
or internal dissension ; such distractions for the Athenians were
common enough for the Spartans deliberately to have waited for one to
occur, even though that might have meant waiting for several years. As
Athens declined, Sparta's estimate of a good opportunity is likely to
have changed ; she did not wish to attack when Athens was merely
weak, but when Athens was as weak as possible.
Have any episodes been omitted from the lists of Spartan
opportunities while being in fact worthier of inclusion than others
which are noted there? To compare in detail every included episode

65 On the Potidaea episode in detail, see below.

 ATHENS'
DIFFICULTY, SPARTA 's OPPORTUNITY 97

with every other conceivable candidate for admission is beyond the

scope of the present paper. It will perhaps suffice, instead, to compare
what may be thought the two strongest of the omitted candidates and
the least impressive of the admitted. The aftermath of Athens' Egyptian
disaster of the mid 450s probably included the withholding of payment
to the Delian League by Miletus, Erythrae and other places 66 :
Athenian morale, if not actual control of the Aegean, will have been
severely shaken. A few years later Cimon's expedition to the distant
island of Cyprus engrossed 200 ships and involved a siege (of Cition).
Did not both of these episodes present the Spartans with a better
opportunity for aggression than the occasion of the revolt of Potidaea
— not normally considered one of Athens' more important allies?
The strategy by which Sparta chose most often to exploit the
distractions of Athens consisted in the invasion of Attica 67. But if the
Spartans did think it possible to transfer a land army to Attica in the
aftermath of Athens' defeat in Egypt, they might have expected to
encounter a rather similar number of hoplites to that which had
resisted them strongly at the time of Tanagra. Athens in the meanwhile
had lost some hoplites on the luckless expedition of 50 triremes to
Egypt, but had gained through no longer having "a large army" 68
committed to the siege of Aegina and through having the control of
Boeotia. Nor need the Spartans in 455-4 have supposed that they would
find many Athenian troops tied down outside Attica in a struggle
against disaffected former allies. No Athenian military campaign from
these years, against an erstwhile member of the Delian League, was
long or difficult enough to have left any clear trace in our literary
sources 69. Of the states which did defect around this time, those about

66 Meiggs, op. cit., pp. 109 ff.

67 Such a strategy was executed or planned in (for example) 465-4, 458-7, 446, 43 1 ,
428,413,411.
68 I, 105, 3, assuming that the Athenian force at Aegina was not much reduced
between the Corinthian-led invasion of the Megarid and the battle of Tanagra.
69 For a possible Athenian naval campaign in Caria before 452, see Meiggs, op. cit. ,
p. 118. Even if historical, however, the campaign need have been neither long nor
difficult. Meiggs cites as a possible indication of Athenian police action in the Aegean
the statement of Plutarch that Cimon, after the making of the 5 Years' Truce, led the
Athenians to Cyprus "in order that they should not stir up trouble in Greece or by
expeditions round the islands or the Péloponnèse draw on the city responsibility for
wars against fellow Greeks and lay the foundations for protests from their allies"
98 C. A. POWELL

which we are least badly informed are Miletus and Erythrae 70 : in both
the anti-Athenian movement seems to have been far from
71 — a consideration which could have helped to repress any
Spartan thoughts of intervening on their behalf in the eastern Aegean.
Similar, debilitating, divisions no doubt existed in other defecting states.
Moreover, at least in the case of Erythrae, which while in revolt may
well have had dealings with the Persians 72, it is by no means certain
that such Spartan intervention would have been welcomed by those
seeking freedom from Athens 73.
More importantly perhaps, would Attica itself have seemed
either in 455-4 or when, later, Athenian morale and resources
had recovered sufficiently to permit Cimon's expedition to Cyprus 74 ?
It now seems to be widely accepted that between 457 and 446 invasion
of Attica by land was thought by the Spartans to be highly dangerous

(Ρι.υτ., Cim., 18, 1, cited at Meiggs, op. cit., p. 120). This reference could, however,
have arisen from a mere proposal of police action before 451 . Meiggs is surely right to
date Athens' intense activity in founding cleruchies to the period after Cimon's final,
and victorious, campaign against the Perians (op. cit., pp. 121-125). In particular, it
seems unlikely that Athens would have scattered small groups of her own citizens
around the Aegean at a time when they would have been exposed both to attack by
disgruntled former allies and to possible Persian intervention. We do not, therefore,
have to reckon that Athens was distracted (and Sparta encouraged) by elaborate
colonising expeditions in the Aegean in the years immediately following the defeat in
Egypt.
70 Meiggs, op. cit., pp. 112 ff.
71 Erythrae, which with its syntely paid 8.33 talents in 449 and on its own 7 talents
in 443 and 439, may have lost almost half of its citizens to Boutheia while in revolt in
453/2 .ATL, I, list 2 (p. 129), col. X, 1. 5 shows Boutheia at this time paying 3 talents,
in contrast to the mere 1 ,000 drachmae which it was to pay later. The special notice
given in the list of 453 to payments from "Milesians from Leros" and "Milesians from
Teichioussa" (ATL, I, list 1 [p. 129], col. VI, 11. 19-22) may mark a similar, substantial
exodus of Delian League supporters from a city in revolt (ATL, III, pp. 252 f. ; Meiggs,
op. cit., p. 112).
72 The Erythrae Decree refers to "the exiles who have taken refuge with the
Medes" (R. Meiggs & D. Lewis, Greek Historical inscriptions, no. 40, 11. 26 f.).
73 Any disgruntled members of the Delian League who nevertheless remained anti-
Persian might still remember the conduct of Pausanias, who had unintentionally
helped to create the League in the first place (I, 96, 1), and also the Spartans' attempt
shortly after Mycale, to abandon Ionia to the Medes after procuring a mass emigration
from the area (Hdt., IX, 106).
74 The making of the Five Years Truce, shortly before Cimon's expedition (I, 112,
1 f.), need not have been sufficient to restrain Sparta from attacking during the period
of truce, if opportunity were thought to exist ; see below, p. 1 1 2 f.
 ATHENS'
DIFFICULTY, SPARTA 's OPPORTUNITY 99

or impossible 75. De Ste. Croix in particular has impressively

emphasised the strength of Athens' defensive position in the Megarid, a
position which is known to have deterred the Spartans from attempting
to march through that district on one occasion in the early 450s 76. The
Spartan force which, shortly after 450, intervened at Delphi in the
Sacred War may only have been able to reach Central Greece because
unaccompanied by many allies 77. And for the Spartans without
numerous allies to have confronted the Athenians in Attica would have
been unthinkable.
Sparta's opportunity to damage Athenian interests in 432-1 contrasts
strikingly with her chances of doing so in the decade following
Oenophyta. At the later period, with the Megarid in allied hands, the
Spartans could invade Attica at will. Such invasion must have seemed
especially promising in 432-1 because of the large number of Athenian
hoplites and other troops predictably withdrawn for the siege of
Potidaea. (It seems that at no stage did the number of Athenian hoplites
engaged in the siege drop below 3,000 78 ; for much of the time the total
approached 4,600 79, and at one point almost 7,000 hoplites were
employed at, or close to, Potidaea 80, more than half of Athens' total 81.)
Unlike Miletus and Erythrae in the 450s, the population of Potidaea
does not appear to have been depleted during her revolt : instead it was
reinforced by a garrison from outside of some 1 ,600 hoplites and 400
light-armed troops 82. Moreover, Potidaea was not the only strong
opponent of Athens in the Thrace ward region during 432-1. An
important merit of Meiggs' recent work on the Athenian Empire is his

75 Brunt, Ioc. cit., p. 258 ; de Ste. Croix, op. cit., pp. 190-195 ; Meiggs, op. cit.,
p. 111.
76 Op. cit., pp. 190-195.
77 Thucydides (I, 112, 5) mentions only "the Lacedaemonians", and not any allies
of theirs in this connection. Contrast the Spartan campaigns to Central Greece during
458-7 and 446, both of which are shown by Thucydides to have involved numerous
allies (I, 107, 2 ; 108, 1 ; 114, 1 f.). De Ste. Croix has pointed out that the mobilisation
of the Spartan alliance must have given the Athenians ample warning of impending
attack (op. cit., p. 191) : the Spartans may have been willing to attempt the penetration
of Central Greece only if such warning were not given.
78 1,61 f.; II, 31, 2; III, 17,4.
79 I, 64, 2f. ; III, 17, 4.
80 II, 58, 1 with 56, 2 ; cf. 58, 2 on Phormion's force of 1,600.
81 II, 13, 6; 31, 1 f.
82 I, 60.
100 CA. POWELL

emphasis on the power of Potidaea's companions in revolt, the

Chalcidian and Bottiaean associations (cf. The Athenian Empire, pp.
210 f., 309 f., on the new Olynthus' becoming "a formidable force").
He points out that the towns in the area of Chalcidice which were
apparently in revolt in 43 1 had been assessed collectively at some 40
talents 83 : Athens never succeeded in crushing the two associations by
military means, and was to suffer serious defeats at their hands in 429
and 425 84. The Athenians had come to fear a general revolt in the
Thraceward region by the time the Spartans promised that, if Potidaea
were attacked by Athens, they would invade Attica : grounds for this
fear had no doubt been communicated promptly to Sparta by King
Perdiccas 85. And when Sparta made her formal decision that Athens
had broken the Thirty Years Peace, the revolts of Potidaea and her
allies were already in progress 86. The Corinthians, leading their
invasion of the Megarid in c. 459, had believed that Athens either
would lack the force to oppose them or would need to withdraw her
troops from the siege of Aegina in order to do so 87. The Spartans in
432-1 probably trusted likewise in a twofold possibility ; they could
hope for a cheap victory in Attica if Athens were to keep a large
proportion of her hoplites at Potidaea, and if those hoplites were
brought back (as Athens had brought back her army from Euboea
during the last invasion of Attica), there would be a prospect not only
of victory in Attica 88 but that the flourishing revolts in Athens' Empire
would prove contagious. The opportunity for Spartan intervention
against Athens at this period was better than has often been realised :
no such opportunity can be shown to have existed in 455-454 or during
Cimon's expedition to Cyprus.
The proposed revolt of Lesbos appears as the only case of a possible
opportunity for the Spartans which did not in the event arouse their
hopes of attacking Athens. The significance of Sparta's refusal to help
the Lesbians cannot, however, be firmly assessed, because of the

83 I,II,
84
85
86
87 Meiggs,
56,
58,
105,
79 2;1 3.cf.
; op.
57,VI,
61, cit.,
61 10,
;- 87-8.
p.
58,5.310.
1 ; 57, 4 f. (Perdiccas).

88 This prospect would arise from reflection on the Spartan victory at Tanagra and
on the Athenians' apparently having preferred bribery to military resistance when
Sparta invaded Attica in 446.
 ATHENS'
DIFFICULTY, SPARTA 'S OPPORTUNITY 101

uncertainty as to when it occurred. Thucydides shows that it happened

before the Ten Years War and in peacetime : a date between 445 and
435 is likeliest 89. If the proposal to revolt was made between 445 and
441, Spartan aggression at times of great and exploitable Athenian
difficulty would have to be seen not as invariable, but merely as
normal. However, a date between 439 and 435 would not have this
effect, because a Spartan refusal then could have been caused by a
belief that there was in reality no good opportunity for aggression. The
earlier conference to discuss the Samian revolt, at which the
Peloponnesians had been "divided in their vote", would have made
clear Sparta's probable inability either to collect the naval help which
Lesbos required or to mount a land invasion of Attica on the scale
which the close-fought battle of Tanagra had shown to be necessary 90.
While this case must be acknowledged as a possible exception to the
pattern of Spartan aggression, the regularity of that aggression at other
times may well suggest that Lesbos' proposal was indeed made in the
unpromising period 439-5, rather than earlier.
The tables compiled above depend very largely on Thucydides. May
the historian perhaps have kept from us information which would have
shown that exploitable Athenian difficulties corresponded with Spartan
aggression much less thoroughly than now appears? In the first place,
major Athenian distractions and consummated Spartan aggression
against Athens are obviously among the classes of events least likely to
be hidden from our view, even were Thucydides to have omitted them.
With unconsummated aggression, such as occurred in 465-4 and 440-
39, an argument from silence is less strong. Diodorus does seem to
preserve information about one impulse of unfulfilled aggression at
Sparta which is not noticed by Thucydides, that which aroused the
successful opposition of Hetoemaridas and is recorded s.a. 475/4 91.
And Lewis has shown that here, as elsewhere, Diodorus probably
drew on a source well-informed about Peloponnesian history 92. But
even on this topic Diodorus is far from thorough ; he omits completely
the conference of the Peloponnesian League which voted concerning
intervention against Athens in 440-39. It might be reasonable to suspect

89 III, 2, 1 ; 13, 1 ; cf. de Ste. Croix, op. cit., pp. 204 f. ; Meiggs, op. cit., p. 194.
90 Cf. de Ste. Croix, ibid.
91 Diod., XI, 50.
92 D. M. Lewis, in Historia, 2 (1953-54), pp. 413-415.
102 C. A. POWELL

that Thucydides gave a distorted impression of events, if it could be

established that he was aware of the correspondence between Athens'
difficulties and Spartan aggression as it stands in his work, and that he
might have emphasised it in the furtherance of some thesis. However,
neither point can be established. Thucydides was aware that the
Spartans and other opponents of Athens on several occasions
deliberately chose to attack when she was already burdened with
difficulties. But even a good historian may easily be so aware, while at
the same time being unaware of how thoroughly and significantly
attacks on Athens match special Athenian distractions in the 60 years
from the revolt of Thasos ; several modern works on the period
inadvertently prove as much. As to whether Thucydides could have
valued the correlations as confirming argument of his : the correlations
(to anticipate somewhat my conclusions on their significance) strongly
support the view that Athenian aggression or increase of power in the
late 440s and 430s was not necessary for the outbreak of war in 43 1 .
They would thus support a "polemic" 93 by Thucydides against the idea
that the Megarian decree was important in this connection, but might
seem to conflict with his statement that the Athenians, "growing great"
(present tense), forced the Spartans into war 94. Further, the correlations
tend to subvert Thucydides' own view on the individuality of the
Peloponnesian War. The regular aggression of the Spartans at every
seemingly exploitable Athenian crisis suggests that their quiescence,
between 446/5 and 441/0, and 439/8 and 433/2, was scarcely more
secure than the Peace of Nicias : it appears hardly more reasonable to
designate as a single war the episodes of hostility between 43 1 and 404
than to connect similarly those of the entire period c. 460-404 95. Yet to
have included in his war events of the 440s and earlier would have
been unattractive to Thucydides, because of the impossibility, which he
admits %, of establishing what happened then 97. If one is to allow any
suggestion of tendentious writing by Thucydides in this connection, it
might be enquired rather whether he is not likely to have underplayed

93 Meyer, op. cit.. p. 297.

94 I, 23, 6.
95 Cf. de Ste. Croix, op. cit.. pp. 3, 50, and esp. 180.
96 I, I, 3.
97 De Ste. Croix, op. cit., p. 180.
 ATHENS'
DIFFICULTY, SPARTA 'S OPPORTUNITY 103

the correlations. Is this perhaps why the highly significant Spartan

aggression of 440-39 is mentioned neither suo loco nor explicitly?
In interpreting now the patterns of correspondence traced above,
two possible theories may be quickly dealt with. First, is the timing of
revolts by Athens' allies determined generally by Sparta's attitude
towards secession, and do the occasions of these revolts mark little
more than that Sparta has already for other reasons decided to make
war then ? Sparta did of course have considerable power to check or
encourage the impulse of an Athenian ally to revolt. Her attitude seems
to have been decisive when Lesbos wished to secede before the Ten
Years War98. Also, Potidaea in 432, Lesbos again in 428 and 413-2,
and Chios and Euboea at the last date sought promises of Spartan aid
before committing themselves irrevocably to revolt ". But if Sparta did
choose to set off the revolts because she had already decided for other
reasons to go to war when she did, we should expect to find that all or
most revolts were begun by Sparta's taking the initiative and inviting
secession. We do not find this, but rather the reverse. Wherever
negotiation as a preliminary to revolt is known to have occurred
between an important Athenian ally and Sparta, the negotiation was
probably or certainly opened by an approach to Sparta from within the
allied state. Moreover, when Thucydides explains the desire to revolt of
Thasos (in 465) and Samos, and the initial impulse to secede of Potidaea
and of the various allies in 413-2, he does not adduce as a cause any
reliance on special help from Sparta 10°. It appears, then, that Sparta
normally exploited, but did not elicit, the initial desire to revolt on the
part of Athens' allies : these desires caused, rather than resulted from,
Spartan decisions at the time to attack Athenian interests.
Alternatively, it might be suspected that Sparta, on the occasions
when she wished to begin war, was influenced not by special Athenian
weakness but by Athenian attempts at expansion, which were
normally going on at those times. The case of Corcyra should weaken
any such suspicion. When in 433 Athens made a defensive alliance

98 in, 2, ι ; 13, l.
99 I, 58, 1 (Potidaea) ; III, 4, 4 ff. (Lesbos in 428) ; VIII, 5, 2 ; 32, 1 (Lesbos in 413-
2) ; VIII, 5, 4 (Chios and Erythrae) ; VIII, 5, 1 (Euboea) ; cf. VIII, 44, 1 (Rhodes) ; VIII,
80, 2 f. (Byzantium).
100 por fragmentary evidence of a possible appeal by Samos for Peloponnesian help
in 440-39, see Meiggs & Lewis, op. cit., no. 56, 1. 12 ; cf. Meiggs, op. cit. p. 190.
104 CA. POWELL

with this great naval power, she may have seemed to increase her own
strength more than by any other single step in the period 445-432. Yet
Sparta did not react by attacking Athens, although the route to Attica
lay open, and not only Corinth but probably the majority of her other
allies would have consented to war 101. On this occasion it seems to
have mattered more for the Spartans that Athens had no unusual
weakness than that she had just perpetrated an act of major expansion.
Later, the great Sicilian expedition threatened vast Athenian
and also presented a good opportunity for attacking Attica.
Sparta did duly attack. But when, with the expedition annihilated and
numerous Athenian allies ready to revolt, the threat to Sparta became
far less and her chances of defeating Athens considerably better,
Spartan aggression did not decline with the threat : it grew with the
opportunity. More generally, Sparta's permanent domestic distraction,
the helots, and her fear of misbehaviour by military commanders
abroad, enjoined a great caution about embarking on foreign
expeditions. If Spartan aggression had been motivated only by
Athenian aggrandisement, we should expect that Sparta would react to
promising defections, such as those of Megara and Euboea in 446 and
Samos in 440-39, by waiting to see whether they would succeed and
thus whether Athenian aggrandisement would indeed take place. To
expect such a policy may indeed seem unrealistic ; Sparta, it may be
thought, must have hoped to defeat Athens before each act of
aggrandisement was complete, and before the Athenians could bring
their newly acquired power to bear against her. But if Sparta was
influenced by reflection upon the difference between Athens' power
before and after the attempts at aggrandisement, she must also have
been influenced by Athens' temporary reduction in power when those
attempts were in progress. For Athenian enfeeblement in Attica at such
times was normally much more marked than the aggrandisement
which might immediately have followed.
It may be well to re-emphasise now the distinctness of the two
general propositions argued for: (1) that Sparta always, or usually,
wished to open an attack on Athenian interests when given a good

101 Most (t? p?????) of Sparta's allies voted for war in the following year (I, 125,
1); in 440-39, even without Corinth's vote or pressure, probably something
approaching half of the allies did likewise ; I, 40, 5 ; 41, 2.
 ATHENS'
DIFFICULTY, SPARTA 's OPPORTUNITY 105

opportunity by the weakness of Athens ; (2) that she was never willing
to do so without such an opportunity. Even if, as cannot be proved, the
correct adverb in the first case is not 'always' but 'usually', and Sparta
did neglect good opportunities in 455-4 and at the time of the proposed
Lesbian secession, the second proposition would scarcely be affected.
On present evidence it is hard to see how this proposition could be
assailed. Any account of Spartan hostility in 432-1 which concentrates
only on Athenian expansion as its cause is, in all probability,
inadequate. The insecurity of such an account could in any case be
shown very simply, by asking a question which is helpful in general for
identifying the status of explanations, but which all too rarely appears
to be put. "If one reflects on all the general rules of behaviour and the
specific conditions which one uses, whether explicitly or implicitly, to
explain an event, is it possible from such reflection alone confidently to
infer the existence of the event?" 102. The answer to that question,
when applied to Spartan hostility in 432-1 and to explanations of it
which notice explicitly as specific conditions only Athenian
and the threat to Sparta, is probably "no". In contrast, on the
strength of the above correlations we may infer with great confidence
that Sparta would not have gone to war in 43 1 without some unusual
opportunity of the kind presented by the revolts of Potidaea and her
Thraceward allies. It is not easy to find this conclusion expressed in
print, although it follows easily enough from the correlations. Probably
these correlations themselves have not normally been identified.
The regularity of the Spartans' behaviour also allows the confident
inference that they would seek by war to exploit the revolts which
began in 432. However, we have not explained why the Spartans
wished in the first place to damage Athens : their "motive" 102a rather

102 Perhaps the best-known assertion of this criterion of explanations is that by C.

G. Hempel (The Function of General Laws in History, in Journal of Philosophy, 39
[1942], pp. 35 ff.). For one attack on Hempel's position, see M. Scriven, in Theories of
History (ed. P. Gardiner), pp. 443 ff. Scriven's paper, like Hempel's, is seriously
marred by failure to distinguish explanations which advance only necessary causes
from those where sufficient causes are involved. See M. Brodbeck, in Minnesota
Studies in the Philosophy of Science, HI (ed. H. Feigl and G. Maxwell), pp. 231 ff.
102a The distinction here intended between 'motivéis)' and 'sense of opportunity' is
slightly different from that conventionally used in the analysis of crime. By 'motivéis)'
is meant not an attitude which might have given rise to a desire in principle for an
action, but one which actually did so.
106 CA. POWELL

than their sense of opportunity. There seems no reason to dispute the

opinion of Thucydides and of modern scholars, that fear of Athenian
expansion motivated the Peloponnesian War. We should, however,
now proceed to qualify this opinion, in the light of Sparta's dependence
upon special opportunity. Had the Athenian expansion reached a point
long before 432 such that the Spartans would have retained their
influential fear into the late 430s even without any further
by their rival? The opposite view is encouraged by
Thucydides, who describes the Spartans in the late 430s as being forced
into war by a process of Athenian aggrandisement which was
still going on 103. Further encouragement may be drawn from what is
perhaps a widespread modern notion concerning economy of causes,
to the effect that agents are unlikely to have more incentives for an
action than the minimum needed to bring the action about. This notion
deserves close examination, which cannot yet safely be left to the
journals of philosophy 104. It bears on Spartan policy in 432-1 , but also
has a neglected, and fundamental, significance for students of Antiquity
in general. It will be contended that the notion is wrong, and liable to
distort the day-to-day judgements of historians and others on causes
and effects.

The Peloponnesian War

and the theory of overdetermination

When historians use causal language and metaphor, they commonly

give no indication of whether the causes referred to are of the
sufficient, the necessary or some other kind. It is not quite certain that
historians themselves at such times always have a clear idea on the
point. How often one should expect to attain to explanations of various
kinds is dependent, however, on how frequently overdetermination
occurs. (By 'overdetermination' here is meant the occurrence of more
conditions conducive to an event than the minimum sufficient to bring
it about.) If overdetermination is rare, when one discovers a condition
probably favouring the occurrence of the explanandum, one may

103 I, 23, 6 ; cf. J. de Romilly, Thucydides and Athenian Imperialism, p. 20, n. 1.

104 For an account of overdetermination with different interests from the present
one, see J. L. Mackie, The Cement of the Universe .· A Study in Causation, pp. 43-47.
 ATHENS'
DIFFICULTY, SPARTA 's OPPORTUNITY 107

normally expect to have discovered arnecessary cause ; the commoner

overdetermination is, the greater is the chance that one may
hypothetically subtract one or a few favouring conditions while still
being left with the event. But if overdetermination is rare, in most cases
ignorance of just one (relevant) cause is enough to prevent the historian
from perceiving the sufficient causéis) of an event. On few occasions
can an event and its antecedents) be definitely identified as instantiating
some sufficient-causal law 105. Far more frequently, when the working
of no such law can be detected, doubt as to whether every relevant
cause is known would properly prevent the historian from assuming
that the sufficient cause(s) can be identified. It seems that historians
generally, to have a coherent view of their own explanatory practice,
must hold an opinion about the frequency of overdetermination.
Ancient Historians should see a particular importance in the subject.
Dealing on the whole with much sparser information than their
Modern counterparts, they should find it much more difficult to give
sufficient-cause explanations. Yet, like all historians, they can only
hope to tell an attractive story if they can identify with confidence
many causal connections. Assuming that a necessary cause is of more
interest than a superfluous favouring factor (or "redundant cause"), the
Ancient Historian is particularly dependent on that view of over-
determination which encourages the search for necessary causes even
when information is fragmentary the view that overdetermination is
rare. But would such a view be correct?
A belief in the rarity of overdetermination may rest on two
assumptions. The first is that Occam's Razor, when applied to human
history, dictates that the number of causes postulated for an event
should not be multiplied beyond the minimum sufficient for the event
to occur ; the second, and tacit, assumption derives from the metaphors
of breaking point and the last straw. An examination of these
metaphors may help to indicate how Occam's Razor should be applied
to human history and what, in this field, is the necessity beyond which
hypotheses should not be multiplied. The straw was originally chosen,
for the figure involving the camel's back, no doubt because its weight is
thought in most circumstances to be trivial. The difference between the

105 This is the case even if 'sufficient-causal law' is applied not only to universal
regularities but to the regularities existing in the circumstances of a particular period.
108 C A. POWELL

weight of one straw and another may be thought of as even more

trivial : more likely it is not thought of at all. Thus the metaphor, if it
influences our idea of causation, may help to smuggle in the
assumption that motives and motivating conditions are, like straws,
virtually identical in the force which they exert. But such an
assumption would be quite incorrect ; Spartan fear, for example, must
have been aroused much more by the great Sicilian expedition than by
the colonisation of Thouria. The metaphor may also suggest that
causes, like the straws, accumulate one by one. In practice, causes often
do not behave thus ; a state may, for example, at one and the same time
perpetrate more than one provocative or frightening action, or it may
perform a single action which is influential in more than one way.
(Thus Athens' attack on Aegina in c. 459 threatened not only to add to
the Athenians' dominion but to degrade Sparta in the eyes of her
remaining allies.)
A more illuminating metaphor for human decisions might involve a
set of scales, loaded at random with stones, and bundles of stones, of
obviously diverse weight : the point of decision is here represented by
the point at which the needle of the scales reaches or passes 1 cwt.
Sometimes a single stone is found to weigh 1 cwt. or more. Far more
often several stones are needed. When scales are gradually loaded thus,
very occasionally the needle will point exactly to 1 cwt. Much more
frequently it will pass the mark by some distance. Overdetermination in
our sense occurs when the extent of that distance is greater than the
distance accounted for by one or more of the weights imposed before
the needle passed 1 cwt. Now when the last weight brings the needle
past the mark, we can be sure overdetermination has not occurred if
the last weight is the lightest of the load. If, on the other hand, the last
weight is one of the heaviest, many of the lightest may well fail to be
necessary causes of the needle's reaching 1 cwt. when it does. Over-
determination in the case of the scales would surely be common. There
seems no reason to suppose that overdetermination is much less
common where human decisions are concerned.
There is, however, a reason for thinking that overdetermination in
the case of human action may be much more common than with the
loading of scales. The metaphors of the camel's back and the scales
suggest motivation which is cumulative but of a single type, and such
metaphor may even have encouraged the search by historians for
motivation of this kind. But motives of a more complex kind must also
 ATHENS'
DIFFICULTY, SPARTA S OPPORTUNITY 109

be taken into account. Where it is felt by a would-be agent that a special

opportunity is required before an action can, or should, be undertaken,
conditions intensifying the wish for the action may accumulate
abundantly while the opportunity is waited for. "Motive" in the great
majority of such cases may well be overdetermined by the time
opportunity is identified. Moreover, actions which historians are
especially interested in explaining, such as attempts at revolution, the
opening of war or revolt, and the conclusion of inter-state agreements,
are especially likely to fall into this class. The most important actions in
the eyes of an historian tend to have been unusually important also in
the eyes of those taking part in them : unusual care must often have
been taken to find a good opportunity for such actions, and thus
unusual scope given for the occurrence of superfluous motivating
conditions. The timing of Sparta's various attempts to begin war with
Athens, over the period 465-41 3, shows that she consistently depended
upon an unusually favourable opportunity for aggression. Sparta's
quiescence is not like a camel's back ; more than a burden is required to
destroy it. Perhaps it is a failure to appreciate this principle of Spartan
strategy which has led Thucydides, and good historians of recent times,
to suggest that Athenian aggrandisement in the 430s was a necessary
cause of Sparta's decision for war 106. If Athenian growth is mistakenly
accepted as the one true cause of Sparta's desire for war in 432-1 , then,
since that growth was, even as late as 433, plainly insufficient to
produce war, it will seem that further aggrandisement, or the threat of
it, was needed to spur Sparta into action. On this reasoning, a large
measure of overdetermination must appear out of the question, and
Athenian growth during the 430s, at least if taken as a whole, will seem
a necessary cause of the war's breaking out when it did. However, once
Sparta's dependence on special opportunity is understood, it becomes
clear that to rule out overdetermination is to demand a coincidence,
whereby "motive" only reached the level needed for action when
opportunity was identified. And the historian has no right to expect
such a coincidence.

106 Compare the view of Andrewes that the year 432 contained "the crucial
moment when it was realised that Athens' use of her power was intolerable" (loc. cit. ,
p. 226, n. 3).
110 CA. POWELL

To conclude on overdetermination : if this process is very common,

and especially so in those cases which historians find the most
interesting, the prospect of happening upon causes of the sufficient type
is somewhat improved. Our evidence, though incomplete, will often
reveal causes which, even if they had not been accompanied by others
now unidentifiable, would still have been sufficient to bring about the
events which we are concerned to explain. However, in the much
more common case where the evidence is so scanty as to encourage
only the hope of making a necessary-causal statement, the incidence of
overdetermination reduces the chance of actually doing so. Redundant
causes should be recognised as important objects of historical enquiry.
How then can we claim in the above to have discovered necessary
causes of many instances of Spartan aggression? These necessary
causes consist in opportunities not in "motives", and so far it is with the
overaccumulation of "motive" that we have been concerned.
Opportunity itself may of course sometimes be overdetermined, as, in
Sparta's case, during 458-7 and 412-1. It has, however, by now been
shown that in our period at least one impressive Athenian distraction
was needed for the occurrence of certain kinds of new Spartan
aggression. If, at the times of such aggression, there was usually some
farther impressive opportunity, now unrecorded, which made causally
unnecessary that opportunity which we do hear about, then the lack of
record of the former opportunity would, because of that very
impressiveness, be surprising. And if our sources were so fallible as
regularly to have omitted such impressive opportunities, we might
expect them in connection with some cases of Spartan aggression to
have omitted all reference to special opportunity. This does not happen.
The argument from silence against major, unrecorded, Athenian
distractions is indeed strong. Where only one special Athenian
distraction is known, it is probable that without that distraction Sparta
would not have attacked 107. And in cases such as 458-7 and 412-1,
where more than one major distraction is known and where
of opportunity probably occurred, we may still usefully claim
that, had none of the known distractions occurred, there would
probably have been no Spartan aggression.

107 Assuming of course that in the counterfactual scheme thus created the removed
distraction is not replaced by another like it.
 athens'
difficulty, sparta 's opportunity 111

The Athenians, the Spartans and their modern critics

In analysing the origins of the Peloponnesian War, scholars have

tended to form two schools : some, impressed by Sparta's long periods
of quiescence, believe her to have been dominated for long by a peace
party ; others are persuaded by her repeated aggression that she was
fundamentally hostile towards Athens. It now seems possible to
emphasise at the same time the evidence which each school has found
most impressive and in such a way that we arrive not at a quandary
but at a belief in a more consistent Spartan aggression than either has
tended to assume. The claim attributed by Thucydides to the
Corinthians in 432, that the Spartans are defensive in attitude and wish
to avoid hurting others, is dealt with briefly by de Ste. Croix as "a
curious piece of special pleading" 108. The very strangeness of the claim
(and of others like it in the same speech 109), as coming from members
of a state which had in 440-39 opposed, and very likely prevented,
Spartan aggression against Athens, makes it probable that the claim had
some substance, so as not to expose those who made it to a crushing
retort. De Ste. Croix's suggestion, that the Corinthians had in mind
Sparta's lack of aggression during the Corcyraean episode uo, is surely
correct. However, Sparta's behaviour during that episode, like her
earlier prolonged quiescence, must be squarely faced if, like de Ste.
Croix, we believe that she was profoundly hostile towards Athens. By
attending to Sparta's best opportunities for aggression we are able to
account for all such quiescence without assuming any deeply peaceful
disposition among the majority of those influential at Sparta. Sealey
urges against the idea that Sparta was fundamentally hostile to Athens
in the period 445-433 (or even in 440-39) the question, "If most
Spartans were willing to go to war so early (se. as 440-39), why did
they not go to war until 43 1 ?" 1U. Such an objection shows once more
that Spartan strategy has not been generally understood. Sealey's
rhetorical question may now be answered, by referring to Sparta's lack
of special opportunity for aggression between 439 and 432.

108 I, 71, 1 ; de Ste. Croix, op. cit., p. 95.

109 I, 68, 2 ; 69, 4.
110 Op. cit., p. 1 1 1 , n. 59 ; cf. Meyer, Forschungen zur alten Geschichte, II, p. 315.
1,1 Loc. cit., p. 106.
112 CA. POWELL

The instability of Spartan foreign policy, which is well shown by de

Ste. Croix in other connections, did not extend to decisions of the
Spartan state, made at times of great and exploitable Athenian
distraction, as to whether to attempt aggression 112. ?. H. M. Jones
wrote that the attempted Spartan aggression of 440-39 demonstrated
"the persistence at Sparta of a strong war party, eager to strike at
Athens, in defiance of treaties, whenever she was embarrassed" 113.
However, it now appears that at moments of serious and exploitable
Athenian embarrassment there was not merely an eager war party but
a regular bellicose majority among Sparta's decision-makers. Well-
placed Spartans could sometimes hope that a peace party might
predominate at such times ; Archidamus probably hoped as much in
432, while Pleistoanax and Cleandridas must have expected that their
failure to press home Sparta's advantage in 446 would be tolerated by
the Spartan state in general 114. All three men, however, were
disappointed, and we should not repeat their mistake. But perhaps
Sparta's willingness to agree to the Thirty Years Peace shows that her
objection in 446/5 was not to a peace policy so much as to the
supposedly corrupt motives of Pleistoanax and his adviser? The Peace
was made after the revolt of Euboea had been crushed 115, when
Sparta's best opportunity for aggression had clearly passed. It is
unknown with what expectations the Spartans agreed to the treaty. On
other occasions their behaviour was strongly influenced by religion 116 :
probably the Spartans in 446/5 would not have arranged and sworn
oaths which they believed likely to be broken by their own state. It is
just possible that they expected Athens to break the treaty before very
long, and thus to restore their own freedom of action. Far more likely,
however, many of the Spartans in 446/5, rather like penitents at
confessional, were perfectly sincere, and simply overestimated the
influence their own religiosity would have when temptation recurred.
In any case, Sparta was willing once more to attack Athens in 440-39,
although the power of the latter had not greatly increased and was

112 De Ste. Croix, op. cit., pp. 151 ff.

113 Loc. cit., pp. 43 f. ; cf. de Ste. Croix, op. cit., p. 169.
114 I, 80-5 ; II, 18, 3-5 (cf. de Ste. Croix, op. cit., pp. 142 f.) (Archidamus) ; I, 114,
Plut., Peric, 22, 2 (Pleistoanax and Cleandridas).
115 I, 114, 3 - 115, 1.
116 E.g. I, 103, 1 f. ; V, 54, 2 ; VII, 18, 2 ; Hdt., VI, 106 ; 120.
 ATHENS'
DIFFICULTY, SPARTA 's OPPORTUNITY 1 13

indeed severely threatened by the revolt of Samos : Sparta's tolerance

during an Athenian crisis had whatever the Spartans' own previous
predictions on the matter not grown greatly since 446. There is no
reason to suppose that it grew between 439 and 432, and thus that acts
of Athenian aggrandisement in that period were necessary for causing
the Peloponnesian War. Rather, from c. 465 very little Athenian
aggrandisement may have been needed to keep Sparta chronically
liable to attack Athens when opportunity offered.
This paper has so far been widely critical of modern theories on the
causation of the Peloponnesian War. It may be well to end by noticing
what can now be seen as the wisdom of many Athenians and Spartans.
Those Athenians who believed in 433 that a war with the
Peloponnesians would come, whatever they themselves did in the
matter of Corcyra, were justified since it was highly likely that some
major and exploitable Athenian difficulty would eventually have
presented itself to the Spartans. Cleon, Cleophon and the others who
rejected Spartan offers of peace after Athens' successes at Sphacteria,
Cyzicus (and perhaps Arginusae) again were probably acting
reasonably 117 in spite of occasional judgements to the contrary m.
Since Sparta was very likely to reopen war, and at a time highly
unfavourable to the Athenians, it scarcely made sense to grant her
peace when Athens had an advantage to exploit 119. Nor need one
consider entirely foolish the action of the Athenians in 415 when,
during the alarm over the supposed revolutionary conspiracy of the
Hermokopidai, they spent a night under arms in fear of a small Spartan
force which "happened to have advanced to the Isthmus in pursuance
of some scheme with the Boeotians" ,2°. Had there indeed been a
revolutionary plot, the only sane course would have been to guard

117 IV, 21, 2 ; 41, 3 f. (425 B.C.) ; Diod., XIII, 52 (410 B.C.) ; Arist., ap. schol. Ar.,
Ran., 1532 (cf. G. Grote, History of Greece, 2nd ed. [1851], VIII, p. 286) (406 B.C.).
118 The height of modern incomprehension seems to have been reached by W. S.
Ferguson, who wrote, concerning Athens' supposed refusal to make peace after the
victory of Arginusae : "the fol)y of the Athenians is explicable only on their own
theory that those whom the Gods would destroy they first make mad" (Cambridge
Ancient History, V, p. 359). Cf. J. B. Bury, History ofGreecei, p. 497, on the aftermath
of Cyzicus, in which connection Meiggs notes of Cleophon : "He was not mad" (op.
cit., p. 372).
119 Cf. Grote, op. cit., VIII, pp. 166-169.
120 VI, 61,2.
1 14 CA. POWELL

against a combination between the plotters and the approaching

Spartan force. Athenian conspirators had attempted such a
in 458-7 m ; something similar was to be tried in 41 1 m, while
earlier in that year Spartan troops, as we have seen, endeavoured to
exploit sedition within Athens by approaching the city.
As for Sparta, her very survival for centuries as leader in the
Péloponnèse argues a considerable intelligence in policy. To have risen
to dominate mainland Greece by 404, and to have maintained that
position, or one closely approaching it, until the number of her citizen
hoplites had dwindled to little more than 1 ,000, suggests skill of a very
high order. One form which that skill took is now apparent.
Thucydides emphasises that wishful thinking frequently influenced the
making of important decisions by Greek states 123. At Sparta, however,
in spite of the strong and long-lasting desire to crush Athens, wishful
thinking never in our period produced a decision to attack without
there existing first an unusually good opportunity. It might be difficult
to find in any period of history another state which for so long and so
systematically began, extended, and refrained from aggression against
another. In her attempts to reduce the city which had become the
School of Hellas, Sparta's policy had a classical elegance and economy
of its own.

Institute of Classical Studies, C. A. Powell.

31/4 Gordon Square,
London WC1 (England).

121 I, 107, 4.
122 VIII, 91, 2 f.
123 III, 3, 1 ; IV, 108, 4; VIII, 2, 2.