The History of A Greek Proverb

WHEELER, EVERETT L.
, "Polla kena tou polemou" [Greek]: The History of a Greek Proverb ,

Greek, Roman and Byzantine Studies, 29:2 (1988:Summer) p.153
IToAAa KEVa TOt) 7ToAEP.OV:

The History of a Greek Proverb
Everett L. Wheeler
I N THE SPRING of 427 B.C. a Peloponnesian fleet under the nauarch

Alcidas lumbered across the Aegean to raise the Athenian siege of
Mytilene. i At Embaton in the Erythraea the fleet received confir-
mation of the report that Mytilene had already fallen; the Spartan
mission had failed. 2 But the Elean Teutiaplus, a figure otherwise
unknown, proposed to salvage the situation through stratagem: a sur-
prise attack by night would catch the Athenians unprepared, dis-
persed, and negligent in victory (Thuc. 3.30.1-3).
Teutiaplus justifies his plan by reference to TO KEVOV TOt) 7TOA€P.OV
(3.30.4), an obscure phrase, which then appears (always in the plural
and always in a military context) in other authors down to Cicero,
before eventually entering the collection of proverbs attributed to
Diogenianus. 3 Metaphorical use of the phrase (i.e., in a non-military
context) is first seen in Philo of Alexandria (Ebr. 76) and Plutarch
(Mor. 41B). Neither Thucydides nor Aristotle, however, recognized
the phrase as a proverb, of which Polybius gives the first explicit
attestation. The history of the phrase KEV'ovl KEVil. TOl! 7ToAEP.OV provides
a rare glimpse of how a Greek proverb originated, and allows us to
probe its relationship to ancient military thought and psychological
theory.
Although proverbs are often regarded as colloquial aphorisms of
'folk wisdom' originating at indeterminable points in the past, some
I The following will be cited by author's name: C. von CLAUSEWITZ, On War, tr. H.
Howard and P. Paret (Princeton 1976); P. BORGEAUD, Recherches sur Ie dieu Pan
(=Biblioteca Helvetica Romana 17 [Rome 1979]); R. B. KEBRIC, In the Shadow of
Macedon: Duris of Samos (=Historia Einzelschr. 29 [Wiesbaden 1977]); K. MEISTER,
Die sizilische Geschichte bei Diodor von den Anfangen bis zum Tad Agathokles
(diss.Munich 1967); W. K. PRITCHETT, The Greek State at War I-IV (Berkeley 1971-
85); F. WEHRLI, Die Schule des Aristoteles lIP Klearchos (Basel 1969).
2 For a reassessment of Spartan strategy in the Archidamian War, emphasizing
Spartan hopes for effective naval action, see T. Kelly, "Thucydides and Spartan
Strategy in the Archidamian War," AHR 87 (1982) 25-59.
3 Arist. Eth.Nic. 3.8.6; Polyb. 29.16.3 (=Suda s. v. ?ToMa KaLva TOV ?ToA£p.ov Adler);
Diod. 17.86.1; 20.30.1, 67.4; 21.2.3; Cic. Alt. 5.20.3; Diogenian. 7.80 (CPG 1300) =
Apost. 14.53 (CPG II 618).
153
WHEELER, EVERETT L., "Polla kena tou polemou" [Greek]: The History of a Greek Proverb ,
154 nOAAA KENA TOY nOAEMOY
do arise from datable historical events, 4 and it is only an assumption

that proverbs must derive from oral tradition. In fact, as I hope to
show, the proverb 7TOAACz. KEVCz. TOt) 7rOAEP.OV developed over time in
both its form and its associations. Speculation about the frequency of
this phrase in popular speech would be fruitless, since scholars can
work only from written sources. The most we can say with any
certainty is that the initial form of the phrase occurs in Thucydides; it
is first called a proverb by Polybius; and its metaphorical use appears
only in the first century (Philo and Plutarch). To claim it as a proverb
before Thucydides would be an argumentum e silentio, especially as
our first testimony to this effect comes almost three centuries later.
7rOAACz. KEVCz. TOt) 7rOAEP.OV thus offers an instance of a proverb deriving
from the literary tradition, and it is an exaggeration to think that all
proverbs collected in the Corpus paroemiographorum graecorum orig-
inated exclusively in colloquial use.
The study of proverbs presents numerous difficulties, not the least
of which is determining a proverb's correct meaning. 5 7rOAACz. KEVCz. TOt)
7rOAEP.OV is no exception. Throughout its history the phrase was associ-
ated sometimes with stratagem, the concept of trickery and deception
in war (its initial context in Thucydides), sometimes with panic, the
sudden terror seizing armies without rational cause, and sometimes
with both. 6 Yet KEVOV/ ICEVCz. TOt) 7rOAEP.OV is synonymous with neither
strategema nor panika. Indeed both the meaning of the phrase and its
text are at issue: its association with stratagem, especially as found in
Thucydides, has prompted arguments for accepting KaLvJv/ lCaLVa, the
reading in some manuscripts, or even ICOLVJV/ ICOLVa rather than ICEVJV/
KEVa. Does a textual problem really exist? In contrast, Cicero equates
the phrase with 7raVLlCa, an interpretation unsuitable for its original
Thucydidean context. How did these two apparently unrelated con-
cepts, stratagem and panic, gain association with KEVOV/ KEVCz. TOt) 7rO-
AEP.OV?
Previous scholarship has not addressed this question. Gomme,
although aware of the phrase in other authors, declined to draw upon
this material to explain Thuc. 3.30.4. His interpretation of the phrase,
as an error to be avoided yet exploited when seen in others, only
4 Cf John Patrick Lynch, "Hipparchos' Wall in the Academy at Athens: a Closer

Look at the Tradition," in Studies Presented to Sterling Dow (=GRBM 10 [Durham
1984]) 173-79.
5 See the introduction to Winfried Buhler, Zenobii Athoi proverbia (Gt>ttingen
1982).
6 Stratagem: Thuc. 3.30.4; panic: Arist. Eth.Nic. 3.8.6; Polyb. 29.16.3; Diod.
20.67.4; Cic. Alt. 5.20.3; both: Diod. 17.86.1,20.30.1,21.2.3; Diogenian. 7.80.
EVERETT L. WHEELER 155

partially grasps its meaning, and he failed to understand the universal
truism of war that Thucydides attempted to signify. Shackleton Bailey
in his commentary on Cic. Alt. 5.20.3 offers an excellent brief dis-
cussion; but his treatment is by no means exhaustive, omitting refer-
ences to the phrase in Polybius and Diodorus, and offering no ex-
planation of its association with both stratagem and panic. 7 A more
extensi ve treatment is therefore justified.
First, I shall argue that Kf.v6vlKf.va is the correct reading. Next, I
shall demonstrate that the original Thucydidean meaning of the
phrase is a truism signifying a general's problem in perceiving a
situation and deriving correct information in war, where the dilemma
of discerning reality from illusion has practical significance: illusions
and misperceptions are opportunities for stratagem. Thucydides'
phrase can be seen as an ancient version of what Clausewitz would
later call the 'fog of war'. Finally, I shall discuss how the phrase
became a proverb and account for its association with panic-a new
aspect of the phrase that began with its appropriation by Aristotle
from Thucydides. In the fourth century B.C. the practical problem of
army panics became a topic of both military theory and Peripatetic
psychological speculation. The key to the later history of the phrase
lies with the Peripatetic Clearchus of Soli, who perpetuated the
school's practice of collecting proverbs and combined the military
and psychological interests in panic in a treatise on the subject.
Clearchus, I believe, is the source of the proverb in the collection of
the Ps.-Diogenianus, and his association with Duris of Samos, anoth-
er Peripatetic, suggests that Duris may be Diodorus' (and perhaps
Cicero's) source for the proverb.
I
Although Kf.v6vlKwa has been the preferred reading in recent edi-
tions of Thucydides, Aristotle, Polybius,8 Diodorus, and Cicero,
Greeks from about the second century B.C., as Gomme noted (ad
Thuc. 3.30.4), did not distinguish kenos from kainos in pronuncia-
tion, and thus Gomme concluded that the manuscript tradition of
both Thucydides and other sources of the proverb had no value. Even
the scholiast (of unknown date) on Thuc. 3.30.4 did not know wheth-
7 A. W. Gomme, A Historical Commentary on Thucydides II (Oxford 1956) 292f.
D. R. Shackleton Bailey, Cicero's Letters to Atticus III (Cambridge 1968) 226f.

8 Buettner-Wobst records no manuscript variants for !C(Va at Polyb. 29.16.3 (Teub-
ner ed., Leipzig 1904), but Adler, Suidae Lexicon IV (Leipzig 1935) 156, has !CalVa.
The Suda is the only source for this fragment of Polybius.
156 I10AAA KENA TOY I10AEMOY
er Kt:VOV or KatVOV was correct and could not offer a single interpreta-
tion of the phrase. 9 The variant reading KatvovlKawa, however, is not
found for all occurrences of the phrase: it is lacking for Diod. 20.30.1
and 64.7. 10 Modem translations often skirt the textual issue. ll Some
scholars either prefer the readings of inferior manuscripts or reject the
manuscript tradition completely in suggesting TO KOtVOV (the impar-
tiality of war) or KatpOS (the opportunity ofwar).l2
Clarence Bill has argued for reading Kawa, although he considers
the proverb only in Thucydides, Aristotle, and Cicero. In his view
KEva TOV 7rOA(JJ.OV to an ancient Greek could only mean the "empty
fears of war" or the "futilities of war," neither of which, he thinks,
suits the contexts of the proverb in his three authors. Diogenes Laer-
tius 5.41 is alleged to prove this: TO O€ Kt:VOV TOV fJiov 7rA(OV TOV
CTVJJ.4>(POVTOS. Bill, favoring the reading KatVa, translates the proverb as
the "shifting nature of war" or the "contingencies of war." Further
9 1: ad Thuc. 3.30.4 Hude: TO lCaLVOV TOV 7TOA.€P.OV· TO lCaLVOV 0' p.€V aLa. lJufJ8oyyov
ypa"'avTf:s OilTC.tlS EV07JlTaV, TO 7Tap' EA.7Tlaa lCat 7Tapa. aofav 7TOtOVV EV TO'iS 7TOA.€P.OLS VLlCaV,
TOVT€ITTt TO E7rt7TflTf'iV ac/JVA.aICTOLS TO'iS EX8pO'is· 0' lJE aLa TOV (" "'LA.OV ypa"'avTfS oilTC.tlS'
£f7JYOVVTa', TO alalCfVOV lCai apyov TOV 7TOA.€P.OV. Of the two hands that contributed
scholia to Laurentianus gr. 69.2 (saec. X), Hude follows the more recent (C2) in
printing lCatvov.
10 See C. Th. Fischer's Teubner edition (Leipzig 1906). At Diod. 17.86.1 the variant
lCalVO'S appears only in R (Paris. gr. 1665, saec. XII) by a late hand.
11 E.g., Thucydides: Hobbes: "all stratagems of war whatsoever are no more but
such occasions" (R. Schlatter, ed., Hobbes' Thucydides [New Brunswick 1975] 192);
Crawley: "baseless panics common in war" (Thucydides, The Peloponnesian War
[Modern Library ed., New York 1951] 161); Smith (preferring the iectio lCaLva): "the
element of surprise in warfare" (Thucydides [Loeb ed., London 1920] II 47); Warner:
"an example of the unknown factor in warfare" (Thucydides, The History of the
Peioponnesian War [Penguin ed., Baltimore 1959] 209); Lateiner: "the opportunity of
war" (n.27 infra: 177 n.7); Aristotle: Rackham: "war (as the saying is) is full of false
alarms" (Aristotle, Nicomachean Ethics [Loeb ed., London 1945] 165); Diodorus:
Goukowsky ad 17.86.1: "par une ruse de guerre" (Diodore de Sicilie, Bibliotheque
Historique [Bude ed., Paris 1976] 120); Greer ad 20.30.1,67.4: "the empty alarms of
war" (Diodorus Siculus [Loeb ed., London 1954] X 223, 325; Walton ad 21.2.3: "the
futilities of war" (Diodorus Siculus [Loeb ed., London 1957] XI 9); Cicero: Shackle-
ton Bailey: "nerve warfare" (supra n.7: III 59).
12 1C0LVOV/ICOLva: Steup argues for 1C0lVOV in Thuc. 3.30.4 on the basis of Hom. I/.
18.309, Arist. Rh. 2.21.11, and Thuc. 5.102, but takes the untenable position that
ICwa should be read in the phrase in other authors. Classen prefers lCalvov, as do
Pop po and Stahl: J. Steup, Bemerkungen zu Thukydides (Freiberg n.d.) 259-62; J.
Classen and J. Steup, Thukydides 3 III (Berlin 1892) 54, 253f; J. Classen, Thukydides
III (Berlin 1867) 188f; E. F. Poppo and J. M. Stahl, Thucydidis De bello Pelopon-
nesiaco 11.1 (Leipzig 1875) 46f. R. Y. Tyrell and L. C. Purser, The Correspondence of
Cicero III (Dublin 1890) 116 suggest 1C0Lva at Cic. Au. 5.20, but fail to support their
suggestion with the reading KOLva found in the very poor digamma class of manu-
scripts, as noted in Shackleton Bailey's edition. lCalpoS': Hans-Jorg Schulz, "Zu Thu-
cydides 3,30,4," Hermes 85 (1957) 255f, written in ignorance of Bill's article (n.13
0'
infra) and citing Thuc. 1.142.1 as a parallel: TOV aE 7TOA.€P.OV lCaLpOt ov P.fVfTOl.
proof is adduced from an adscriptum to Au. 5.20.3 in the delta class of

manuscripts, where the proverb is given as nova belli. 13
There is less room to fault Bill's translation than his arguments. The
passage from Diogenes proves nothing, and the adscripta in the delta
manuscripts of Cicero's letters are generally unreliable, as Bill also
recognizes. Furthermore, the only variant reading in the manuscripts
for Att. 5.20.3 is KOWa, not KaWa. 14 Seeking a Latin solution for the
proper reading of the Greek proverb, however, has merit, since the
proverb also appears in Latin sources in contexts where the Roman
author has most probably either directly or through a Zwischenquelle
consulted a Greek text.
Curtius' account of Alexander's stratagem to capture the Sogdian
Rock offers the best example. When the local king refused to surren-
der, Alexander sent a band of Macedonian mountaineers to seize the
summit above the king's camp and then dispatched the king's son to
urge surrender. When the king saw Macedonians waving white flags
from the peak and the Macedonian camp below raised a shout, he
yielded:
iamque e Macedonum castris signorum concentus et totius exercitus
clamor audiebatur. ea res, sicut pleraque belli, vana et inanis bar-
baros ad deditionem traxit: quippe occupati metu paucitatem eorum,
qui a tergo erant, aestimare non poterant (Curt. 7.11.25)
Certainly Curtius is drawing on a Greek source, and Alexander's
stratagem of creating an illusion to induce his opponent to draw a
false conclusion in a state of panic fits a pattern of behavior re-
peatedly associated with the proverb, as we shall see. IS Curtius'
phrase, sicut pleraque belli, vana et inanis, can be a paraphrase only of
Kwa TOt) 7TOA€/J-OV. However much KaWa may appear in some cases a
more logical reading of the proverb, KEva TOt) 7TOA€/J-OV must be correct.
Curtius did not write nova in 7.11.25 or elsewhere.
In 4.13.5, part of Parmenio's speech urging a surprise attack by
night on the Persians at Gaugamela (the same kind of military action
that occasions the proverb in Teutiaplus' speech at Thuc. 3.30),
Curtius has Parmenio note the advantage not only of creating a noc-
13 Clarence P. Bill, "TA KAINA TOY TIOAEMOY," CP 32 (1937) 160f.

14 Cf. supra n.12. Shackleton Bailey (supra n.7) 227 also rejects Bill's arguments but
without refuting all of them.
15 Although names of the Sogdians differ in Curti us, Plutarch, and Arrian (Dio-
dorus' account is lost), Plutarch's abridged account (Alex. 58.4) seems to follow the
same source as Curtius, i.e., Alexander intimidated the king into surrender: TaVT1/V
(the Sogdian Rock) P.(V otJv (KcpoMCTas TOV r.LCTLP.{8p1Jv tAa/3f.v. Arrian (A nab. 4.21)
omits the stratagem and has Alexander induce surrender solely through negotiation.
158 IIOAAA KENA TOY IIOAEMOY
turnal panic but also of Macedonians avoiding the horrifying sight of

Scythians and Bactrians: vanis et inanibus militem magis quam iustis
jormidinis causis moveri. Although the passage is a less obvious and
less literal version of the proverb, the contrast is again between real
fear (iustis jormidinis causis) and imagined fear or false alarm (vanis
et inanibus). Nova/'caLva will not do.
Use of the Greek proverb in Latin, however, is not limited to
Curtius: Cato also knew it. When as consul in 195 B.C. Cato governed
Hispania Citerior during a major frontier crisis, the chieftain of the
Ilergetes, a Roman ally, demanded protection for his people. With too
few troops under his command to permit division of his forces in
event of an attack, Cato created an elaborate illusion of sending aid,
and actually embarked some troops, only to recall them. 16 His de-
liberation of the stratagem reveals not only a paraphrase of the Greek
proverb but also a theoretical perspective on the role of deceit in war
(Liv. 34.12.3f):
stat sententia non minuere copias, ne quid interim hostes in/erant
ignominiae; sociis spem pro re ostentandam censet; saepe vana pro
veris, maxime in bello, valuisse, et credentem se aliquid auxilii
habere, perinde atque haberet, ipsa juiucia et sperando atque auden-
do servatum.
Cato's stratagem against an ally conforms to a Greek view that a
general's deception of his own forces can be just (Xen. Mem. 4.2.15-
17), and the very use of stratagem by the arch-advocate of Roman
conservatism points up the hypocrisy of Roman propaganda (e.g.,
Liv. 42.47.4-9) that Romans never resorted to stratagem (Cato was a
skilled practitioner of military ruses-a theme that must be pursued
elsewhere).
The phrase saepe vana pro veris, maxime in bello, valuisse para-
phrases the substance of the Greek proverb. The immediate source is
Livy, but it is generally agreed that Livy draws upon Cato's extensive
account of his Spanish campaign in the Origines, although whether
directly or indirectly cannot be determined. 17
Furthermore, Cato's source for the proverb can even be posited.
Among extant sources in which the proverb appears only Thucydides
16 Liv. 34.12.2-13.2; Frontin. Str. 4.7.31. For a recent discussion of the Spanish
wars of the 190's see S. L. Dyson, The Creation of the Roman Frontier (Princeton
1985) 186-93.
17 A. E. Astin, Cato the Censor (Oxford 1978) 302-07; J. Briscoe, A Commentary
on Livy, Books XXXIV-XXXVII (Oxford 1981) 63-65; H. Trankle, "Cato in der
vierten und funften Dekade des Livius," AbhMainz 1971.4: a strong argument for
direct use.
and Aristotle antedate Cato}S Knowledge of Aristotle finds no trace

in Cato's own writings or biographies of him, but a connection with
Thucydides is openly asserted. Cicero twice couples Cato and Thu-
cydides as writers, but Plutarch offers more conclusive evidence.
Plutarch asserts that despite learning Greek late in life, Cato in his
oratory benefited from a knowledge of Thucydides and Demosthenes,
that he moderately embellished his writings with Greek sententiae
and stories, and that many of his aphorisms were literal translations
from Greek sources. 19 Cato's quotation of Themistocles and echoes of
Demosthenes confirm this testimony,20 just as Livy's saepe vana pro
veris, maxime in bello, valuisse is, I believe, a Catonian interpretation
of Thucydides' TO K€VOV TOV 7rOAEP.OV. The proverb does not occur in
Demosthenes. Nor does Cato's allegedly late knowledge of Greek pose
a problem. Cato's memoirs of the Spanish campaign need not be a
work of 195 B.C., and the Origines were still being composed close to
the time of Cato's death in 149. In fact Cato probably learned Greek
early in his career.21
Two other possible Latin occurrences of the proverb may be men-
tioned. Shackleton Bailey cites Tac. Hist. 2.69, Vitellius' dismissal of
troublesome Gallic auxilia in A.D. 69: reddita civitatibus Gal/arum
auxilia, ingens numerus et prima statim defectione inter inania belli
adsumptus. Of course inania belli could be equated with K€Va TOV
7rOAE/1-0V, implying a Thucydidean echo in Tacitus. Yet Tacitus'
phrase, which does not recur in his works, is obscure, and recent
scholarship is skeptical of any direct or significant influence of Thu-
cydides on Tacitus. 22 Wellesley translates the phrase: "as a form of
military window-dressing. "23 In Shackleton Bailey's view the Gallic
18 As I shall argue below, the proverb was popular in Peripatetic writers such as
Clearchus of Soli and Duris of Samos, both of whom antedate Cato, but Cato's
familiarity with these writers cannot be demonstrated. Polybius, Cato's contem-
porary, also seems unlikely: Cato might have been a source for Polybius on some
matters, but no instance of the converse is known.
19 Cic. Brut. 66,294; Pluto Cat.Mai. 2.5f, cf 12.5 and Cic. Sen. 26.
20 Pluto Cat.Mai. 8.4; Cato ORFl fr.20, cf. Dem. Phil. 1.30; fr.195, cf. Dem. Phil.
3.8, 17; Astin (supra n.17) 149.

21 Nep. Cat. 3.3: senex historias scribere instituit; Origines fr.49 (Book 2 written ca
168 B.C.), frr.106-09 (Sulpicius Galba affair of 149), HRR J2 68, 89-91; Astin (supra
n.17) 147, 149, 212f. Astin strongly doubts a Catonian interest in Greek rhetorical
theory.
22 Shackleton Bailey (supra n.7) 227; O. Luschnat, "Thukydides," RE Suppl. 12
(1970) 1298; H. Heubner, Publius Cornelius Tacitus, Die Historien II (Heidelberg

1968) 241, contributes nothing significant. The Gallic auxilia are probably those
provided by the Agrippinenses, the Treveri, and the Lingones immediately after Vi-
tellius' proclamation as emperor: Tac. Hist. 1.57.2.
23 K. Wellesley, tr., Tacitus, The Histories (Penguin ed., Baltimore 1964) 122.
auxilia constituted a useless addition except for their value "in ter-
rorem." This phrase, however, is not Tacitus', as Shackleton Bailey
implies, nor does Tacitus use inania belli when he does record mili-
tary panics: e.g. Ann. 1.66, Hist. 1.63.1. Certainly it could be argued
that the Gallic auxilia were a rabble unfit for combat and accepted for
service only to create the illusion that Vitellius' army was larger than
it really was-hence inania belli-a form of stratagem to induce con-
sternation in Vitellius' opponents. If this is really what Tacitus meant
to imply, then the association of the phrase with terror or panic is
most obscure. As a historian famous for the theme of dissimuiatio,
Tacitus is of course concerned with the problem of appearance vs.
reality. Inanis, nevertheless, is a frequent adjective in Tacitus,24 and a
Greek source does not come into question for the passage. Tacitus'
inania belli need not be a Latin rendering of the Greek proverb, but
rather could be Tacitus' own abstract phrase. The apparent correla-
tion in meaning between the two phrases may be entirely accidental.
Finally, for the sake of completeness we should note the Latin
translation of Byzantine scholia on Aristotle's Nicomachean Ethics by
Robert Grosseteste (d. 1253). For Eth.Nic. 3.8.6 (1116b) Kfva TOt)
7TOAEP.OV, Grosseteste has inania belli. 2s Grosseteste translates from an
anonymous continuation of the commentary on the Eth.Nic. by Eu-
stratius, metropolitan of Nicaea. Although G. Heylbut, editor of the
continuator, notes the KfVuiKaLVo. problem in his apparatus criticus of
the scholiast, the text used by Grosseteste obviously read Kfva, since
KaLVa cannot be translated inania. Two earlier commentators on Eth.
Nic. 3.8.6, Aspasius and Heliodorus, have only KfVo. without the vari-
ant KaLVo. in their manuscripts. 26
This survey of occurrences of the Greek proverb in Latin confirms
the correctness of recent editors' preference for reading Kfv6vlKfva. In
addition, Latin usage confirms the ambiguity of the proverb: in Cur-
tius and Cato (Tac. Hist. 2.69.1, I have argued, is not an occurrence)
the proverb appears in the context of stratagem involving the manipu-
lation of illusion, and both of Curtius' examples associate it with
panic. Examination of the proverb in its various Greek contexts can
bring the problem of its precise meaning into sharper focus.
24 cf A. Gerber and A. Greef, Lexicon Tadteum (Leipzig 1903) 610f s. v. Classen
(supra n.12: III 189) suggests emending inania to semina.
25 H. P. F. Mercken, The Greek Commentaries on the Nicomachean Ethics of
Aristotle in the Latin Translation of Robert Grosseteste. Bishop of Lincoln (=Corpus
latinum commentariorum in Aristotelem graecorum VI. 1 [Leiden 1973]) 284, 287.
26!. in EthNic. 3.11, Comm.Arist.gr. XX 165.18-20; Aspasius in EthNic. 3.11,
Comm.Arist.gr. XIX. 1 84.31-33; Heliodorus, Paraphr. in EthNic. 3.8, Comm.Arist.
gr. XIX.2 56.21-28.
II
Thucydides offers the initial and most obscure occurrence of the
phrase as part of Teutiaplus' speech (3.30):27
, AAKioa Kat I1EA07TOVV7Juiwv (;UOt 7TapEUjJ.EV l1.PXOVTES TfjS UTpaTtaS, EjJ.Ot
~jJ.as E7Tt MVTtA~V7JV 7TptV fK7TVUTOVS YEVEu()at, /JJU7TEP lxo-
OOKE'i 7TAE'iV
P.EV. KaT~ y~p Tb ElKbs tzvopwv VEWO"Tt ?T6ALV tX6VTWV 'lTOAt. Tb tzcpt,AaK-
TOV d)P~UOjJ.EV, KaTd. jJ.h OaAauuav Kat 7Tavv,
, '0" ~ I
EKE'iVOL TE iLVEA7TtUTOt
\.. I , (' .....
n
~, '\. ' I ,
E7TtYEVEU at av Ttva U",tut 7TOI\.EjJ.tOV Kat 7JjJ.WV 1] al\.K1] TVYXaVEt jJ.aAtUTa
1" ., \ ~ , \, ~, .,..... .,,, I.,\. ' ('
ovua' EtKOS uE Kat TO 7TE~OV aVTWV KaT OtKtas ajJ.EI\.EUTEPOV WS KEKpaT1]-
,
KOTWV
~ '0
utEU7Tap at.
,~ , ",f..
Et OVV 7TPOU7TEUOtjJ.EV a",vw TE KaL VVKTOS,
\ "\'~
EI\.7TL~W
\
jJ.ETa
TCi)V lvoov, d TtS l1.pa ~jJ.'iv fUTtV t17TOAOt7TOS Eiivovs, KaTaA1]cpOfjvaL ClV Tet
7TpaYjJ.aTa. KaL jJ.~ iL7TOKV~UWjJ.EV TOV KLVOVVOV, vOjJ.LuaVTES OVK l1.AAO Tt
1" , , .... \. ' '" ,.. (,\" ,,' ~ ....
ELVaL TO KEVOV TOV 7TOI\.EjJ.0V 1] TO TOtDVTOV, 0 Et TtS UTpaT1]yos EV TE aVT,!>
CPVAclUUOtTO Kal. TO'iS 7TOAEjJ.iOtS EVOpWV f7TLXEtPOL1], 7TAE'iUT' ClV Op()O'iTO.
Gomme interprets the phrase solely on the basis of Thuc. 3.30,

although he mentions the verbal parallels elsewhere in Thucydides
and use of the proverb in other authors. The meaning of TO KEVOV TOV
7TOA€/J-OV, in Gomme's view, depends on that of TO TOtOVTOV, which he,
like Classen and Poppo/Stahl, takes as a forward reference to the
following subordinate clause. He concludes that the proverb, which he
translates "the empty, or the fruitless thing in war," signifies an error
to be avoided in ourselves and to be exploited when discerned in
others. 28 Yet TO TOLOVTOV provides an antecedent only for the relative
pronoun C5, not for the entire clause. The concluding subordinate
clause serves only as a didactic flourish to end the speech and does not
define the proverb. TO TOLOVTOV actually summarizes preceding ma-
teria1. 29
Teutiaplus' speech advocates a surprise attack by night, but
Gomme correctly saw that TO KEVOV is not surprise attack, which
merely takes advantage of the error to be avoided. On the other hand,
Gomme rejected equation of TO KEVOV with TO cupvAaKTov, thinking that
this limited the meaning of the phrase too much. Certainly lack of
precaution is not the exclusive meaning of TO KEVOV TOV 7TOA€/J-OV, but it
27 For commentary see D. Lateiner, "The Speech of Teutiaplus," GRBS 16 (1975)
175-84. Lateiner essentially follows Gomme's interpretation of the proverb. Note
that TaoE is unique here in introducing a speech in Thucydides, who elsewhere uses
TouioE or TOLaVTa, implying invention, whereas to introduce documents he uses a
definite pronoun, signifying quotation. Hence Gomme argues that Teutiaplus' speech
is historical, i.e. not a Thucydidean invention: Essays in Greek History and Literature
(Oxford 1937) 166f, Commentary II 292f; disputed by S. Hornblower, Thucydides
(Baltimore 1987) 53. Conceivably TaoE at 3.29.2 is a scribal error for TOLaoE.
28 Gomme (supra n.7) 292; Classen and Poppa-Stahl (supra n.12).
29 Cf Steup (supra n.12), Bemerkungen 258f, Thukydides 3 54.
does represent one of its components. Thucydides' language is ob-

scure because his idea is abstract. TO KEVOV represents not a single idea,
but the abstraction of a group of ideas emphasized in 3.30: undis-
covered (7TpI.V EK7TVCTTOVS), lack of precaution (TO acpVAaKTOV), unex-
pected (avEA1J'LCTTOL), negligence (aIJ.EAECTTEpOV), and finally the conse-
quence of these-nocturnal surprise attack (7TpOCT1J'ECTOIJ.EV acpvw TE Kal.
VVKTOS).
Teutiaplus' speech should be compared with two other passages.
First, Brasidas' battle plan against Cleon at Amphipolis: the situation
is again a surprise attack on an unsuspecting opponent, and Thucydi-
des explicitly characterizes the plan as a stratagem. 30 The Spartans
intend to strike without prior exposure to the enemy (5.8.3, aVEv 7TpO-
O"'EWS), and the attack will be sudden (5.8.4, E7T&XnpELV alcpVLalws) on an
unsuspecting enemy in disorder and taking no precautions (5.9.3, OUK
av EA7TlCTavTEs, aTd.KTws, OAtYWPELV). Furthermore, the language and
idea of 3.30.4 are strongly echoed in 5.9.4f:
" ~ \, \ , , , .
ounS' uE TaS' TOtaVTas ap.apnas TWV Evavnwv /Cal\l\tuTa tuwV /Cat up.a
~, ~ ~ ,~ \ \ ~
7fPOS' T~V faVTOU avvap.tv T~V i7ftXElp1}UtV 7fOtELTat p.~ (17TO TOU 7fpoct>a-
~ ~~\ \ , 6''''
vovs p.al\l\ov /Cat aVn7fapaTax EVTOS' 1} E/c TOV 7fpOS TO 7fapOV f,;,VP.."EpOV-
~ \ \ \ ~ .... '
TOS,
~ ,"~ '6 ~
7fI\EtUT av op OtTO·
\ \ 'I.'
/Cat Ta /CI\Ep.p.aTa
~ ~\'
TaVTa /Cal\l\tuT1}V uOf,;,av EXEt a
~,~,'"
TOV
\ ,
~ , ,,,
7fOI\EP.tOV p.al\tuT av
\
nS'
, ,
a7faT1}uaS'
\ .... ,~
TOUS' ."tI\OVS' P.E"YtUT av W."EI\1}-
, ''''.... \ '
UEtEV.
It would thus appear that TO KEVOV is closely associated but not iden-
tical with KAElJ.lJ.aTa and could be equated with alJ.apTla. Certainly
error, like stratagem, is involved in the idea of TO KEVOV TOt) 1J'OAEP.OV,
but ap.apTta does not convey the proper nuance, as another Thucydi-
dean passage suggests.
When King Archidamas of Sparta addresses his army before the
first invasion of Attica, he warns against negligence and overconfi-
dence based on numbers, fearing the possibility of surprise attack
(2.11.3f):
" ,,, \~ ~ \'8"
oV/Covv XP1}, Et TC!> /Cat uOKOVP.EV 7f1\1} Et E7fLEVat /Cat au."al\Eta 7fOI\I\1} ELvat
\' .... ' \ 'I. 'I. \ ...
\" '\6 ~ \ , ,
p.1} av EI\ EtV TOVS Evavnovs
< ~ 51: \ ' ' ' ' '
1}p.tv uta p.aX1}S', TOVTWV EVE/Ca ap.EI\EuTEPOV
\' ,
, ~ ''I.~\ \ '). ., . , \
n 7fapaU/cEvaup.EVOVS XWPEtV, al\l\a /Cat 7fOI\EWS' E/CaUT1}S' 1}"YEp.ova /Cat
UTpanWT1}V TO /Ca
,,~~
,
au1}l\a "Yap Ta TWV
\ \
\
~
6' • \
).
7fOI\EP.WV,
,\
/Cat
\ ,~
E~
~
,~,
, 8
aVTOV atEt 7fPOUuEXEU at ES' /CtVuVVOV nva 1}f,;,EtV.
, \
Ol\tYOV Ta 7fOl\l\a /Cat ut
'
\
,
~ \
51:'
\ ~"
OP"Y1}S at
,,~
~ .
30 Thuc. 5.8.2 (T'XV71), 9.5 with schol.: /CA'p.p.aTa=(1'TpaT71i'~p.aTa. Cj Otto Luschnat,
Die Feldherrnreden im Geschichtswerk des Thukydides (=Philologus Suppl. 34.2 [Leip-
zig 1942]) 70 with n.l; E. L. Wheeler, Stratagem and the Vocabulary of Military
Trickery (Mnemosyne Suppl. 108 [Leiden 1988]) 28f, 32f. On the battle see N. Jones,
"The Topography and Strategy of the Battle of Amphipolis," CSCA 10 (1977) 71-
104.

, I I ~~I \,,~ ~~8 ~~\"
E7TLXUp7JCTU!; YLYVOVTaL' 7TO""aKL!; TE TO E"aCTCTOV 7T"7J O!; uEuLO!; ap.uvov
.qp.vvaTO TOU!; 7TA£OVa!; OLa TO KaTaq,pOVOVVTa!; a7TapaCTKEVOV!; YEV£CT8aL.
The themes of negligence, lack of preparation (a variation on the ideas

of lack of precaution and unexpectedness at 3.30, 5.8f), and sudden-
ness of attack are common to the speeches of Teutiaplus and Brasidas.
The key phrase, however, is 11.01JAa Ta TWV 7rOAEP.WV. 31 The "unseen
elements of wars" induce error (c[ app.aTia, 5.9.4) and can be ex-
ploited for stratagem, when avoided in one's own conduct and seen in
the enemy's.
Thucydides' Tb Kwbv TOV 7roAEP.OV thus represents the void of correct
information in war, produced by unseen, obscure, or poorly perceived
factors. Teutiaplus' appeal to the KEVbV of war summarizes his earlier
arguments: the Spartan fleet is yet an unseen element; the Athenians
revel in the false security of a victory, contributing to their lack of pre-
caution and their negligence; and the Spartans can pounce upon them
under cover of night.
Contrary to those who would emend the text, KEVbV is appropriate
here. Thucydides' "void of information" may be compared to Democ-
ritus' TO KEVbV for the void of space. "Idle boasts" (KEva dJyp.aTa)
appears in Homer, and fifth-century Greeks used KEVbS with EA7rls,
YV~fL7J, <ppovTloES, and <pb{30L.32 Xenophon describes panic as KEVOS
<pb{3os (Anab. 2.2.21). Most probably, Thucydides' Tb KEVbV is not an
ellipsis that requires a noun to be understood with KEVbV, but rather
simply an abstraction from substantive use of the adjective, sum-
marizing the thought of the preceding three sentences of the speech.
Thucydides' concept of TO KEVOV TOV 7roAEP.OV can be better under-
stood through comparison with the ideas of a modem classic on the
art and philosophy of war, Carl von Clausewitz's Vom Kriege. Al-
though it cannot be proved that Clausewitz ever read Thucydides,
both, as profound thinkers on the phenomenon of war, made chance
and the unexpected a motif of their works.33 In Clausewitz's view a
general can fully know only his own situation. Real knowledge about
31 Lateiner (supra n.27: 177 n.7) cites 2.11.4 as a parallel to 3.30.4 but does not
expound on its significance for understanding the proverb.
32 LSJ s. V. KEVOS' 1.1 f. Cf the proverb KEvol KEVCt AO)',(ovTaL: Suda s. v.
33 Thucydides: e.g. 1.78. If, 83.3f, 140.1; 4.18.3f, 62.4; 7.61.3; 8.24.5; cf L.
Edmunds, Chance and Intelligence in Thucydides (Cambridge [Mass.] 1975). Clause-
witz: e.g. 84-86,89,101-03,117, 139f, 156, 167, 193,580; cf K. L. Herbig, "Chance
and Uncertainty in On War," in M. Handel, ed., Clausewitz and Modern Strategy
(=]ournal of Strategic Studies 9.2-3 [1986]) 95-116. I do not accept a recent
argument that Thucydides and Clausewitz had totally different views and perspec-
tives. I hope to treat this matter elsewhere. See V. Ilari, Guerra e diritto nel mondo
antico I (Milan 1980) 123-25.
the enemy must be carefully derived by a sensitive and discriminating

judgment, skillful intelligence, determination, and presence of mind,
from an overwhelming mass of false information and rumors exag-
gerated by fear in an environment uncontrollably governed by chance
and the unexpected. 34 In brief, a general must operate in a constant
fog of uncertainty, in which illusion and reality can scarcely be dis-
cerned:
Der Krieg ist das Gebeit der Ungewissheit; drei Viertheile derjeni-
gen Dinge, auf welche das Handeln im Kriege gebaut wird, liegen
im Nebel einer mehr oder weniger grossen Ungewissheit .... End-
lich ist die grosse Ungewissheit aller Daten im Kriege eine eigen-
thtimliche Schwierigkeit, weil alles Handeln gewissermassen in ei-
nem blossen Dammerlicht verrichtet wird, das noch dazu nicht
selten wie eine Nebel- oder Mondscheinbeleuchtung den Dingen
einen tibertriebenen Umfang, ein groteskes Ansehen giebt.35
This "fog of war" constitutes a major source of the "friction of war,"
Clausewitz's metaphor for those factors complicating and interfering
with the execution of plans-the difference between war on paper and
in reality.36
Thucydides' KEVOV TOt) 7rOAEP.OV and Clausewitz's "fog of war" repre-
sent essentially the same idea, portraying the problem of a general's
uncertainty of information in war as a sort of existential dilemma.
The proverb, however, conceptually at least where error due to negli-
gence and the unexpected is involved, is also related to an exemplum
met with some frequency in Stoic writings and attributed variously to
Iphicrates, Fabius Maximus, and Scipio Aemilianus. The version of
Valerius Maximus (7.2.2) offers a point of departure:
Scipio vero Africanus turpe esse aiebat in re militari dicere "non
putarem," videlicet quia explorato et excusso consilio quae ferro
aguntur administrari oportere arbitrabatur. summa ratione: inemen-
dabilis est enim error, qui violentiae Martis committitur. 37
The exemplum emphasizes foresight and expecting the unexpected,
in other words, giving heed to the KEVOV TOt) 7rOAEP.OV. Scipio Aemili-
34 Clausewitz 84, 101, 103.

3S Vom Kriege 4 I (Berlin 1880) 50, 108 (=tr. Howard/Paret 101, 140).
36 Clausewitz 117, cf 119-21.
37 Cf Sen. De ira 2.31.4: turpissimam aiebat Fabius imperatori excusationem esse:
"Non putavi," ego turpissimam homini puto. omnia puta, expecta. That Valerius'
Scipio is Aemilianus is clear from Gell. 13.3.6 = Sempronius Asellio fr.5 (HRR P
181). Iphicrates: Plut. Mor. 187 A; Polyaen. 3.9.17. Non putavi as a general aphorism:
Cic. Lig. 30, Off. 1.81; Anon. Byz. De re strat. 20.1; Ps.-Maur. Strat. 8.1.26, 2.36
Dennis.
anus probably learned the aphorism from his father Aemilius Paulus
(c[ Gell. 13.3.6), who may in tum have heard it from Cato.38 As
argued above, Cato knew the proverb TO K£VOV TOU 7rOAEfJ-OV from Thu-
cydides, and this exemplum, eventually attributed to several famous
generals, could indicate Cato's further elaboration of Thucydides'
idea.
Tb Kwbv TOV 7TOA'P.OV originally signified the void of accurate infor-
mation in war, the problem of discerning reality from illusion and
expecting the unexpected. A connection of the phrase with panic finds
no trace in Thuc. 3.30, except for the logical inference that a noc-
turnal surprise attack would produce panic in the enemy. If it is cor-
rect to connect the phrase with the exemplum of "I didn't think,"
Cicero's association of this exemplum with courage and foresight (Off.
1.8.1) provides a hint of what direction use of the phrase would take
after Thucydides.
III
The subsequent history of the phrase begins with Aristotle. After
defining courage as the mean between boldness and fear (Eth.Nic.
3.7.13), Aristotle discusses different kinds of courage, first distin-
guishing the true courage of citizen soldiers from the courage of
soldiers compelled by fear of their commanders to be brave (3.8.1-5).
Next he discusses courage based on experience, exemplified in the
courage of mercenaries (3.8.6-8 [1116b]):
.. ~ .. \
uOKEL \ < , , < \" , .. '
uE Kat T/ E/J-7fELpta T/ 7fEpt EKaura avupELa
[
nS'
] l'
Elvat· "0 \ <
0 EV KaL 0
~ , "8' ,
,,-WKpaTT/S' ~T/ 1J E7fLUT1J/J-1JV
l' \'..' ~
Elvat r1Jv avupELav. TOtOVTOt uE aAAOt /J-EV EV
.. \ " \ \ \,
"'I. \ ' ~ \. ~ .. , < ~.. ~ \ l' \ \ \ \

aA"OLS', EV rOLS' 7fO"EiJ-LKOLS' u OL UTpanwraL· uOKEL yap ElVaL 7To""a KEva
TOt) 7TOA'/J-OV, a
/J-aAtUTa UVVEWpaKaUtV OVTOt· 4>a{VOVTat O~ aVOpEtOt ()n
OVK io-autV OL llAAOt ora funv. EtTa 7fotijuat Kat /J-~ 7faOEtv /J-aALUTa
.. ' ,~, ' .. ' ~ 8
uvvaVTaL EK T1JS EiJ-7TELpLas, uvvaiJ-EVOL XP1JU at TOLS' 07T"OLS Kat TOtaVTa
~"\. \ ~
"
EXOVTES'
~,...""",,...
07TOta av EL1J KaL 7TpOS TO 7TOt1Juat
\'"
Kat 7TpOS TO iJ-1J
8~
7Ta €LV Kpa-
,
"
TLuTa. WU7fEP 1 'aV07f"OtS'
OVV ,' \ < \ "
W7T"tU/J-EVOt iJ-aXOVTat Kat\ a'()\.A1JTaL\ '!It '
tutWTatS·
Kat yap fV TOtS' TOLOVTOtS' aYWULV Ovx OL aVOpW)TaTOt /J-axt/J-WTaTO{ ElutV,
'\.\'
a"" Ot<iJ-a"LUTa
'\. "
tUXVOVTES Kat\ Ta
\' "
UWiJ-aTa aptura "
EXOVTES.
Does Aristotle's understanding of the phrase differ from Thucydides'?

38 Cato in Veget. 1.13: deinde in a/Us rebus, sicut ait Calo, si quid erratum est, potest
postmodum corrigi; proeliorum delicta emendationem non recipiunt, cum poena statim
sequatur errorem; nam aut confestim pereunt qui ignave imperiteque pugnaverint aut
in fugam versi victoribus ultra pares esse non audent. It could also be argued that Cato
heard it from Fabius (Sen. De ira 2.31.4) or that the aphorism originated with
Iphicrates (cJ supra n.37), but I would judge both these attributions as inventions
common in the literature of exempla.
If allowance is made for Aristotle's transfer of the phrase from an

historical to a philosophical context, no difference in meaning occurs.
Aristotle says that mercenaries are experienced soldiers who under-
stand the many K(Va of war while others, presumably non-professional
citizen-soldiers, do not. Experience is the key element. 39 All three
passages in Thucydides (2.11.4, 3.30.4, 5.9.4f) are speeches of experi-
enced commanders who understand the fog of war and seek to exploit
it to advantage (Teutiaplus, Brasidas) or to avoid defeat because of it
(Archidamas). Certainly the element of experience is implied in Thu-
cydides' K(VOV TOU 7roAip.ov, and Aristotle uses the same contrast be-
tween the knowledgeable and the ignorant. Furthermore, Aristotle's
elaboration on this experience of the K(Va «(tTa taken consequentially,
'therefore', 'accordingly'), which he says facilitates offensive and de-
fensive operations, is his own reworking of the final subordinate
clause in Teutiaplus' speech (3.30.4, cf 5.9.4). Aristotle used Thucydi-
des for his Athenian Constitution and probably for his I.vva)'w)'~
T€XVWV.40 Eth.Nic. 3.8.6-8 now offers another instance of Thucydides
as a source for Aristotle.
Moreover, Aristotle contributes new elements significant in the
later history of the phrase. Thucydides' abstract K(VOV is changed to
the plural (7roAAa. K(Va), and all occurrences of the phrase after Ari-
stotle will also be plural. In addition, Aristotle associates the K(Va with
fear: experience of the K(Va promotes courage, but ignorance of the
K(Va brings terror. 41 Aristotle establishes the connection of the prov-
erb with panic found later in Diodorus, Cicero, and Diogenianus.
Polybius (29.16), the next writer to employ the phrase, also asso-
ciates it with an effect on morale. An eclipse of the moon before the
battle of Pydna in 168 B.C. prompted a rumor (q,7Jp.~) that the eclipse
portended the eclipse of a king. The rumor encouraged the Romans
and damaged Macedonian morale. Polybius comments OiJTOOS aA7J(Us
fun TO 7r(Plq,(POP.(VOV C>n 7rOAAa. K(Va. TOt) 7rOAEP.OV (29.16.3). The ration-
alist Polybius scoffs at this interpretation of a natural phenomenon,
although realizing the significance of the fog of war on the two armies.
His source for the phrase defies even speculation, but two facts should
be noted: first, Polybius specifically applies the K(Va to a rumor, a
strictly aural and unseen phenomenon, and rumor as a cause of panic
39 Scholiasts on the passage (supra n.26) merely paraphrase Aristotle without adding
anything of significance.
40 See Luschnat (supra n.22) 1284f, 1287.
41 Cf Aspasius in Eth.Nic. 3.11, Comm.Arist.gr. XIX.l 84.29-33. Rackham's
translation of Eth.Nic. 3.8.6 (supra n.ll) treating the phrase as a proverb is mislead-
ing, as Aristotle does not recognize it as such.
will be encountered again below; second, Polybius is the first author to

call 7TOAAa Kwa TOU 7TOA€/J-OV a proverb (TO 7TfPU/>fpO/J-WOV).42 Polybius'
use of the phrase does not conflict with the meaning in Thucydides or
Aristotle and indicates that the phrase had acquired a wider currency
by the second century H.C.
The proverb occurs four times in Diodorus-more than in any
other author, but still hardly a frequent phrase considering the vast
corpus of his work preserved, nor likely to be his own addition, as I
shall argue below. Our first concern is the meaning of the proverb in
Diodorus.
When Alexander's persistence in the siege of Aornus alarmed the
Indians, he removed his guard from a path down the rock, permitting
the Indians to escape by night. Diodorus adds (17.85.7-86.1):
, \ '" .... ~f3' f3 I}: \ \ .... , ( , I
TO jJ.EV OVV 7TPWTOV OL ap apoL uLa Ta!i' TWV T07TWV V7TEpOxa!i' 7TpOETEpOVV
, \\'" ~ ~ f3 ~, ~ 10' ,
KaL 7TOI\I\OV!i' avypovv TWV 7TP07TETW!i' La",ojJ.EVWV· TOV uE XWjJ.aTO!i' <TVVTE-
\ 8' ,~, l: f3 \ ~ \ ~ '~"\ \
I\ECT EVTO!i' KaL TWV o",v EI\WV KaTa7TEI\TWV KaL TWV al\l\wv opyavwv E7TL-
" ,
e' ,,,,, ~ f3
CTTa EVTWV. 7TP0!i' uE TOVTOL!i' TOV
\ ' "" ~ 8 ~ • ,
a<TLI\EWS" .."aVEpov Ka ECTTWTO!i' W!i' OVK
, I
a7TO<TT1JCTETaL ~
T1J!i' \ I OL"jJ.EV
7TOI\WpKLa!i' ' 1 1VuOL
0 ' KaTE7Tl\aY1JCTav,
" 0'l>"A,Il:
u I\E~av-
10
UpOS" ' "" I
EjJ..."OVW!i' •., , '
7TPOLu0jJ.EVO!i' "
TO "
jJ.EI\I\OV 'l:',
Et;,EI\L7TEV " EV Ty~ 7Tapou~
T1JV '" KaTaI\E-
\
AELjJ.jJ.EV1JV 4>vAaK~v, OLOOV!i' E'fOOOV TOL!i' f30VAOjJ.EVOL!i' EK TfjS" 7TETpa!i' &va-

~
XWpEW. '5>'
OL uE f3'
ap f3 apOL.."o 8' ,~
"" f3 1J EVTE!i' Ta!i' TE TWV M aKEuOVWV
"1/ ' , ,
apETaS" KaL,
T~V TOt! f3acnAiw!i' C/>LAonp./.av VVKTO!i' i,fAL7TOV T~V TrfTpav.
o 0' , AAEfavopoS" TOL!i' KEVOL!i' TOU 7TOAEjJ.OV KaTaCTTpaT1JY~<Ta!i' TOts
'15>' , 5>' "
VuOV!i' XWPL!i' KLVuVVOV EKVPLEV<TE
~,
T1J!i' 7TETpa!i'. KaL
,~,
T~
'\:0 I ,
jJ.EV ou1JY1J<TavTL Ta!i'
• ' , , ' , I" ' , ,,"~ l: , '~"I/ I
WjJ.OAoY1JjJ.EVa!i' uwpEaS" a7TEuWKEV, aVTO!i' uE aVE",Ev",E jJ.ETa T1J!i' uvvajJ.EW!i'.
The context is Alexander's use of the well-known stratagem of leaving

the enemy a way of escape, and further signalled by KaTaCTTpaTrr)l~CTa~,
a verb meaning to 'out-general through stratagem'.43 The Kfva in this
case, however, is Alexander's psychological ploy of demonstrating his
perseverance and determination, which in conjunction with Mace-
donian feats of military engineering already displayed at the siege
causes the Indians to fear.44 The Indians failed to see through the fog
of Alexander's illusion.
42 TO 7TEPHP£POP.£vov is certainly a more obscure and periphrastic way to quote a
proverb than lCaTa ICOLV~V 7TapoLp.{av as at Polyb. 23.14.4, but whether this periphrasis
means that 7TOAAa ICE va TOl) 7ToA€p.OV has not yet reached the state of a ICOLV~ 7TapoLp.ia is
open to debate. Stylistic variatio cannot be ruled out.
43 Way of escape: e.g. Thuc. 1.82.4; Xen. Hell. 6.4.33; Diod. 30.12; Onasander
38.2-6; Joseph. BJ 3.208-10; Pluto Lye. 22.5; ef Frontin. Sfr. 2.6. The verb: see
Wheeler (supra n.30) 9f with n.33.
44 Parallel accounts omit the proverb, but contain similar elements. In Arnan
(A nab. 4.30.2-4) the Indians, alarmed by Macedonian progress in the siege, arrange a
truce for negotiations which they intentionally prolong to cover their escape. When
Alexander realizes this, he permits their escape, seizes the deserted rock, and then
168 I10AAA KEN A TOY I10AEMOY
The next three occurrences in Diodorus all concern Agathocles. 45 In

309 B.C. the Carthaginian Hamilcar decided to make an attempt on
Syracuse, which was defended by a relatively small garrison as the
bulk of the Syracusan forces were in Africa with Agathocles. He
initially camped near the Olympeium and intended to occupy Eurya-
Ius on the Epipolae after traversing the Anapus River valley by night.
The Syracusans meanwhile divined the plan and occupied Euryalus
with a small force. Ignorant of the rough terrain and narrow roads, the
Carthaginian army stumbled on its way in the dark of night, con-
stantly thrown into disorder by its baggage train. When the Syra-
cusans at Euryalus perceived the Carthaginians in difficulty, they
immediately attacked from their superior position. Already in dis-
order and hampered by the darkness, the Carthaginians supposed that
a major Syracusan force had arrived, and panic added to their rout
(Diod. 20.29.2-11). Diodorus, in comparing this Syracusan victory to
Agathocles' defeat at the Himeras River in 311 B.C., marvels at how a
small force can defeat a larger (20.30.1):
, 'e avp.aCTtWTaTOV,
Kat TO ' I> 'I>
uWuEKa 'I> ~, ,,"
p.vptaua~ '7TE~WV Kat '7TEVTaKtuxtl\tOV~ t'7T-
'7TEL~ OALYO~ aptep.o~ '7TOAEP.LWV, '7TpouAa{36p.Evo~ a'7TclT7]V Ka't T6'7TOV, KaTa
'" " , al\1J
1JTT1JUEV, WUT
KpaTo~ '" e' E~ l'
Etvat ,,
TO ,,
I\EyOP.EVOV "
OTt " " " Ta KEva
'7TOl\l\a ~
, TOV
'7TOA£P.OV.
Once again the context of the proverb is a nocturnal surprise attack,

but in this case the attackers are surprised. Moreover, defeat of a
superior by an inferior force also figures here, just as in Thucydides'
three passages discussed above. 46 Diodorus attributes the victory to
terrain (T<~'7Tos) and deceit (a7T<tr1]), more aptly understood as self-
deception, since the a7TclT1] of 20.30.1 is the equivalent of the Cartha-
ginians' l.l:yvOLa in 20.29.9. Thus self-deception and ignorance consti-
tute the KEVcl of this incident and produce the panic.
Two years later in Africa after Agathocles had suffered a defeat, five
thousand Libyan mercenaries attempted to desert the Syracusan for
the Carthaginian army by night. As they approached the Carthaginian
camp, which was ablaze from an accidental fire, the Carthaginian
sentries supposed the Libyans to be a surprise attack by Agathocles.
The Carthaginian forces, already in an uproar from the fire, fell into
panic. The Libyan mercenaries decided to return to Agathocles'
attacks the retreating Indians, creating a panic. In Curtius (8.11.19-24) Alexander's

illusion of perseverance is countered by the Indians' illusion of remaining, and
Alexander is deceived. When the Indians' departure is discovered, the shout in
unison of the Macedonian army causes panic in the Indians.
45 For a recent discusssion of Agathocles, see K. Meister, "Agathocles," CAH2 VII. 1
(Cambridge 1984) 384-411, 524-77.
46 Thuc. 2.11.4, 5.9.2 (cJ 5.8.2f); implied in 3.30.2.
camp, where their approach was mistaken as a surprise attack by the

Carthaginians. Here panic likewise ensued (Diod. 20.65-67.3). Dio-
dorus summarizes the incident (20.67.4):
eJA1}V Of T~V VVKTa 7TaVTaxii OtaU7TELpOp,EVWV aVTWV Kat 7TavtK~ 80pV{3f{J
UVVEXOp,EVWV uvvE{31} 7TAE{OV~ TWV nTpaKLuXtA{WV fLvaLpE8fjvaL. f7TLYVW-
(TeE{(T7}~
OE llc5yt~ T~~ aA7}e£{a~ Ot ota(Tw8'vTf~ i7rav~A8ov £t~ 1'~V 7TapEp.-
/3oM,v. at p,£v ovv ovvcip,EL~ ap,cponpaL TOV Elp'TJp,EVOV 1'p07TOV ~TvX'TJCTav,
fta7TaTTJ8liCTaL KaTCz. T~V 7TapoLp,{av TO'~ KEVO'~ TOU 7TOAEp,OV.
The passage requires no commentary: the KE"Va are self-deception and

false expectation (cJ 30.67.1, OL' a7TaT7]V Kat 7TpOUOOK{av ",E"voij), and the
panics were the effects of the K€Va.
In 299/8 Agathocles sailed to the aid of Corcyra, which was be-
sieged by Cassander. The incident is only partially preserved in By-
zantine excerpts of Diodorus. Agathocles apparently surprised the
Macedonians and burned their fleet, but in ignorance of the panic
created he missed his chance for a more complete victory (Diod.
21.2.3):
" 'A yauoKI\1}~
on i1 \~ ,
EL
,
p'EV 'QQ'
a7To,...L,...aCTar; \
TTJV 5>'
uvvap,Lv , ,
E7TLKELp,EVOLr;
~"
TOLr; 7TOI\E-
,
p,LOLr; "i1
E7TEUETO, KaTEKo."EV av pautwr; TOvr; M aKEuova~'
' ./, " r l>' l> '\ , ,
ayvo1}CTar; l> \
uE
\
TTJV
,
yEyEVTJp,EVTJV \'
7TpOCTayyEI\Lav KaL\ TTJV
\ EK7TI\TJr"LV
" " t TWV ~ 'i1'
avupw7TwV "
TJpKECT 8TJ
T~V ovvap,Lv a7Tof3Lf3aCTar; Kat Tp01Tawv CTT~CTar; oLaAa{3liv aA'TJ8~ TOV
\'
I\oyov 0;-
ELVaL on
" \\'
7TOl\l\a KEva,
TOV~ 7TOI\Ep,OV.
\' "
ayvoLa " "
yap \
KaL a7TaTTJ 7T01\-
\ ' "\ ' '~~, ~ r ~, ,
l\aKLr; OVK El\aTTW Kanpya",ETaL T'TJr; EV TOLr; o7TAoLr; EVEpYELar;.
Just as in 20.29.9, 30.1, a7TaT7] and ayvoLa are the K€Va. This passage,
however, represents the only case in which the K€Va worked negatively
on a potentially major victory. Clausewitz realized that the fog of war
could also produce ill-timed inactionY
Cicero's fall campaign of 51 B.C. as proconsul of Cilicia48 provides
the context for the next occurrence of the proverb. In the aftermath of
Carrhae two years earlier, Parthian attacks on Syria and Cappadocia
were feared, and in October 51 a Parthian force actually approached
Antioch. Writing to Atticus on 19 December and describing his cam-
paign, Cicero attributes the Parthian withdrawal from Syria to the
fame of his military success in Cilicia (Alt. 5.20.3):
ibi (at Issus) dies quinque morati direpto et vastato Amano inde
discessimus. interim (sds enim did quaedam 7TavLKa, did item TCz.
KEVCz. TOU 7TOAEp,OV) rumore adventus nostri et Cassio, qui Antiocheia
tenebatur, animus accessit et Parthis timor iniectus est; itaque eos
cedentis ab oppido Cassius insecutus rem bene gessit. ...
Clausewitz 84.
47
For a recent treatment see Elizabeth Rawson, Cicero: A Portrait (Ithaca 1983)
48
164-82.
170 IIOAAA KEN A TOY IIOAEMOY
As should by now be clear from our examination of the proverb, the

proper equation should be KfVcl with rumore (cf Polyb. 29.16.2f), not
KfVcl with 7TavLKcl. Cicero boastfully attributes the Parthian withdrawal
to his own fame and presence on the Cilician-Syrian border,49 the
latter of which perhaps had some influence on the Parthians, although
the occurrence of a true panic seems most doubtful. Cicero's apparent
equation of the proverb with panic, which has misled scholars to
emend the text of the proverb, is a false interpretation ofthe proverb's
original meaning. Nevertheless, the association of the proverb with
panic, first seen in Aristotle and most prominent in Diodorus, per-
haps reaches its culmination in Cicero's apparent equation of the two
ideas. The association occurs also in the proverb's latest appearance.
In the collection of proverbs attributed to the Hadrianic gram-
marian Diogenianus the following entry occurs, later reproduced in
the collection by Michael Apostolius (d. ca 1480):50
~ ~ \
1TOAAa \
KEva ~
TOV ~ ,
1TOAEIJ.OV· "
1/TOL 'I> \
uLa \
TO ~ ~ \
1TOAAa Ka e'"V1TOVOLav "".."EpEW·
' .,
1/~ OTt
J\. aKEuaLIJ.oVLOL
'1>' \
KEvaLS "" 'I> '
(T .."EvUovaLS
~ E'f'
KaL\ VEvpaLS ',I. O.."OVV
' "" \"
1TpOS ~ ~
fK1T A7J'OLV
TWV 1TOAEIJ.lwv.
The first definition does not conflict with the original Thucydidean
meaning of the phrase: to be suspicious about everything is apt to the
general's dilemma in the void of true information in war. Further-
more, this definition also recalls the point of the "I didn't think"
exemplum-expecting the unexpected.
The second explanation of the proverb, however, perplexingly adds
a totally new element: a Spartan stratagem to panic the enemy. Asso-
ciation of the proverb with stratagem and panic had become tradi-
tional, and noise to terrify an enemy suggests nocturnal activity-
another traditional trait of the proverb. But identification of the
proverb as particularly Spartan is new. Two interpretations of this
49Cf AU. 5.20.3: erat in Syria nostrum nomen in gratia.

50Diogenian. 7.80 (CPG I 300)=Apostol. 14.53 (CPG II 618). Apostolius' quotation
of the proverb is corrupt: "ll'OAA<t KEV' 4"11'0 "ll'OA'~OV. No reference to a collection of
proverbs appears in the Suda entry on Diogenianus and opinio communis believes
Diogenianus' name was added to a collection of much later, although uncertain, date.
Diogenianus' authorship of a lexicon is genuine. On the problem see K. Rupprecht,
"Paroimiographoi," RE 11 (1949) 1770; Hans Gllrtner, "Diogenianus (2)," KI.Pauly 2
(1967) 48f; P. Strengel, "Diogenianus (4)," RE 5 (1903) 778-83; K. Latte, Hesychii
Alexandrini Lexicon I (Copenhagen 1953) x, xlii-xliv; F. Brachmann, Quaestiones
Pseudo-Diogenianeae (NJbb Suppl. 14 (1885)) 239-416. The attribution probably
came about from a supposed parallel with Zenobius, a Hadrianic sophist (Suda s.v.)
whose epitome in three books of the proverb collections of Didymus and Lucillius
Tarrhaeus is genuine.
second explanation seem possible: either this explanation preserves

the original occasion of the proverb, or it is a later invention to
account for the proverb's association with panic. The latter is more
probable.
Certainly, from one perspective, Thucydides' singular K£v6v could
be seen as an anomaly, since all other occurrences of the proverb have
K£Va. The context of the proverb in Thucydides is Spartan, just as it is
in Diogenianus, and this could indicate a pre-Thucydidean Spartan
origin. Strictly speaking, however, the speech of Thuc. 3.30 comes
from the mouth of Teutiaplus, an Elean. Furthermore, although Spar-
tans are usually seen as hoplites, Spartan stonethrowers appear in
Tyrtaeus, and these could correspond to the slingers in Diogeni-
anus-thus another argument for a pre-Thuycdidean origin. 51 Nor is
the use of stratagem implied in Diogenianus alien to the Spartan
character or the Spartan art of war: the Spartans had a reputation for
trickery and deceit. 52
Yet despite these possible arguments, a 'Spartan explanation' of the
proverb cannot be substantiated by other sources. 53 The proverb is
not identified as Spartan and does not occur in a Spartan context in
Aristotle, Polybius, Diodorus, and scholia, nor in its Latin use in
Cato, Cicero, and Curtius. In fact neither Thucydides nor Aristotle
recognizes the phrase as a proverb, of which the first attestation
occurs in Polybius, although the phrase may, as I argue below, have
achieved the status of a proverb by the late fourth or early third
century D.C. It is preferable to see the Spartan explanation of the
proverb as a later interpretation, which belongs to the same category
oflegends as the myth of the Spartan drillmasters. 54
I have argued that K£VOV TOt) 7TOAEJJ.OV first occurs in Thucydides as an
abstraction for the idea of the void of accurate information in war-
51 Tyrtaeus 11.35fWest, P.Oxy. XLVII 3316.14.

52 Thuc. 2.39.1; 5.42.2, 45, 105.4; Xen. Res. Lac. 2.6-9; Gell. 11.18.17, cf PI. Leg.
8.547E-548A; Isoc. 12.213f; Paus. 4.17; Pluto Marc. 22.3-5, Mor. 209B, 218F, 223A,
cf Cic. Off. 1.33, Hdt. 6.78f; Plut. Ages. 9.3; Polyaen. 6.6.2, cf Paus. 4.5.8; A. Ar-
naud, Quelques aspects des rapports de la ruse de la guerre dans Ie monde du VIlle au
ve siec/e (these 3me cycle, Paris 1971) 164-67; E. Heza, "Ruse de guerre. Trait carac-
teristique d'une tactique nouvelle dans l'oeuvre de Thucydide," Eos 62 (1974) 229f;
additional references in P. H. Epps, "Fear in Spartan Character," CP 28 (1933) 21. I
hope to treat the topic of Spartans and stratagems in detail elsewhere.
53 No trace of the Spartan sling players appears in Polyaenus, Pausanius, Athen-
aeus, Plutarch, or the fragments of Attic comedy-sources likely to tell such a story.
Unfortunately Plutarch's I1ws Ot." AalCwva JJ.ax(u8al (Lamp. Cat. 213) is lost.
54 See my "The Hoplomachoi and Vegetius' Spartan Drillmasters," Chiron 13
(1983) 1-20.
the dilemma of every general. The KEVOV signifies the problem of

discerning reality from illusion and finds a modem counterpart in
Clausewitz's notion of the fog of war. Aristotle borrowed the phrase
from Thucydides, but changed it to the plural, thereby emphasizing
the multiplicity of forms which the KEva could take. The initial context
of the phrase, the stratagem of nocturnal surprise attack, often reap-
pears in later authors. Likewise, the association of the proverb with
panic, first seen in Aristotle, was frequently repeated, although panic
is a result of the effects ofthe KEva and does not constitute the KEva per
se-the error of Cicero's apparent equation of the two. In the fourth
century B.C., however, panic as a military and psychological phenome-
non began to interest both military writers and philosophers. A brief
study of panic will highlight another aspect of the proverb's history.
IV
Panic, a phenomenon of mass psychology, particularly affected
armies. In fact, the concept, as expressed by 1TavELa, 1TavLKos, etc.,
initially has an exclusively military connotation, although based on
analogy with a natural phenomenon (see below). Thucydides notes
that large armies were especially susceptible to panic at night and
when in hostile territory or in proximity to the enemy. Aeneas Tacti-
cus adds that panics could occur in a camp or in a city, after a defeat,
by day, or especially by night. Heets could also suffer panic, in
Polybius' view. Panic, to be distinguished from fear in general, is a
sudden and irrational terror, which could tum an organized army into
a frenzied armed rabble, capable of killing its own members in its
madness. 55 Panic was recognized as a real problem, and the connec-
tion of sudden terror without actual cause made for the easy associa-
tion with the KEva of war, the manipulation of illusions.
The phenomenon of panic was by no means new in the fourth
century B.C., nor was its association with a god. But 1TaVELa, 1TavLKos,
etc., attributing the phenomenon to Pan, constituted an innovation.
In Homer the concept of panic is connected with Phobos (a real god,
55 Thuc. 7.80.3, cf 4.125.1; Aen. Tact. 27.1, 4; Suda s.v. navuei/i llf:t/J-an; Polyb,
5.110.1-9; Polyaenus (1.2) defines ?Tavueel as baseless (ICf:VOVS) nocturnal fears; panic
produces casualties or potential casualties: e.g. Aen. Tact. 27.4, 7-10; Diod. 15.24.3,
20.66f; Paus. 10.23.1-8; Polyaen. 2.2.10, 4.3.26. By the Roman period, if not earlier,
panic was no longer exclusively a military concept: Pluto Mor. 356D. Polybius
(5.110.1-9) does not give evidence that anyone in the fleet except Philip V panicked,
and the occurrence of a real panic, strictly defined, is open to doubt. Repeated
references to Philip's stupidity and gullibility to a false report are probably another
instance of Polybius' malignity to Philip.
not a personification), the son of Ares, who often contributes to the

flight of a demoralized army. By the fifth century Dionysus could also
cause panic, although neither Herodotus nor Thucydides attributes
the phenomenon of panic to a god. 56
Toward the middle of the fourth century, if not earlier, Pan
emerged as the chief divinity of the phenomenon that bears his name:
Aeneas Tacticus notes that 7TaV€La was a Peloponnesian and especially
an Arcadian word; Ephorus apparently spoke of 7TavLKo~ eopvf3o~; and
Epimenides' Cretica contained the word 7TavLKov. 57 Pan, the god of
herders and hunters, was thought responsible for the sudden, unex-
plained stampedes of domestic herds or of wild animals and the
mysterious echoes of mountains, glens, and caverns-Pan's flute play-
ing to the nymphs. 58 An echo, like the K€VOV of a false rumor, is
unseen, and a state or an army could be viewed as a flock under the
rule of its shepherd (cf Xen. Cyr. 1.1.2). Furthermore, Pan was the
son of Hermes, a god who could cause fright and surprise, and well
known for his stealthy activity. Pan, however, differed sharply from
Phobos, who participates in battle, causes flight, and symbolizes the
combatant's fear. Pan acts from a distance and, like the ruse general,
can achieve his goal of rendering the enemy hors de combat without
fighting. 59
The militarization of Pan in the fifth century is seen in both artistic
and literary sources. Vase paintings represent Pan as a warrior. 60 He
figures prominently in Herodotus' account of Marathon (6.105) and
possibly also has a role at Salamis, although in both cases his alleged
activity is most obscure, and his responsibility for Persian panic can-
not be proved. 61 With the possible exception of a passage from Rhesus
56 Eur. Bacch. 302-05 (military context); Hdt. 7.43.2; Thuc. 4.125.1, 7.80.3. On
Phobos see Ernst Bernet, "Phobos (1 )," RE 20 (1941) 309-17; Pritchett III, 45, 162f,
cf Epps (supra n.52) 1-29; H. Mitchell, Sparta (Cambridge 1952) 270-73.
57 Aen. Tact. 27.1; Ephorus FGrHist 70F208 (Diod. 14.32.3) with Jacoby's notes,
and E. Schwartz, "Diodorus (38)," RE 5 (1903) 679; Epimenides FGrHist 457F18=
824 D.-K.=Ps.-Eratos. Cat. 27. Harrison's attempt to derive 'In:tvna from 'lTavos (fire
signal, alarm) has been refuted: see J. E. Harrison, "Pan, Paneion, Panikon," CR 40
(1926) 68; Borgeaud 138 n.6. Greek expressions for panic are listed in Borgeaud 139
n.7.
58 Apollod. FGrHist 294FF135, 136a-b, cf 1: ad Theoc. 5.16.1; K. Wernicke,
"Pan," Roscher Lex. 3.1 (1901) 1388-90.
59 Borgeaud 142, 153, cf Ammonius' distinction of ¢ofJos from O(os: ¢ofJos is
immediate terror, but O(OS is a long-term suspicion of evil: cited in Bernet (supra
n.56) 309; on Hermes see N. O. Brown, Hermes the Thief(Madison 1947) 3-83.
60 Borgeaud 142,200-02, cf F. Brommer, "Pan," RE Suppl. 8 (1956) 969.
61 Borgeaud 146f; Brommer (supra n.60) 969; Harrison (supra n.57) 6f correctly re-
jects finding a Pan/panic connection in Herodotus and Aeschylus. Borgeaud's attempt
(146-49) to connect nearly all military panics in Greek sources down to Plutarch with
a local sanctuary or cult of Pan at the scene of the panic is not convincing.
(34-37), attributed to Euripides but whose authenticity and date have

long been debated, the association of Pan with the irrational fear of
armies dates from the fourth century B.C. 62
The real origins of the Pan/panic connection lie in Arcadia, where
Pan, widely thought of as Arcadian in origin, was zealously wor-
shipped. Aeneas Tacticus implies (27.1) that '1TaVn4 was an Arcadian
word. It would thus seem that propagation of the association of Pan
with military panic corresponds to the proliferation of Arcadian mer-
cenaries in the fifth and fourth centuries. 63 Perhaps Arcadian mer-
cenaries were less sophisticated than their Athenian or Spartan coun-
terparts and were more susceptible to panic. 64
The eventual significance of Pan's military function appears in
Polyaenus (1.2), who credits Pan with the invention of military de-
ployment, the term phalanx, and the use of panic fear as a stratagem.
The last is attributed to Pan as a general of Dionysus, presumably
during his Indian campaign, when Dionysus' army, camped in a
hollow glen and facing superior hostile forces, was ordered to raise its
war-cry by night, so that the echo would create the impression that the
army was larger than in reality. The echo aroused the enemy's fear
and caused their flight.
Although some believe that Polyaenus derives this account of Pan
from a source dating after Alexander the Great, possibly Megasthe-
nes, it has long been recognized that Ephorus was Polyaenus' main
source for his first book,65 and some elements of Polyaenus' anecdote
appear in another fourth-century source. Epimenides, who transfers
the seat of action to Crete, has Pan aid Zeus in his war against the
Titans. Pan both armed Zeus' forces and routed the Titans by the
62 Borgeaud 137. On Rhesus see W. Ritchie, The Authenticity of the Rhesus of

Euripides (Cambridge 1964), who argues for authenticity and a date of composition
445-440 B.C. Pritchett III 45 cites Diod. 14.32.3 (panic of the besiegers of Phyle in
401) as proof for the Pan/panic connection in the fifth century, but Diodorus' source
is Ephorus (supra n.57), and panic is not mentioned in parallel accounts: Xen. Hell.
2.4.1-4, Nep. Thrasyb. 2.1-5.
63 On the military influence of Arcadians in this period, see Wheeler (supra n.54) 6-
9.
64 It is easy to overestimate the degree of training and discipline in Greek armies of
the fifth and fourth centuries, but as late as Chaeronea in 338, one tradition holds,
the Athenian army was still untrained: Frontin. Str. 2.1.9; Polyaen. 4.2.7, cf Ps.-Xen.
Ath.Pol. 2.1.
65 Borgeaud 154; J. Melber, Ueber die Quellen und den Wert der Strategemensamm-
lung Polyilns (NJbb Suppl. 14 [1885]) 421-23; R. J. Phillips, The Sources and
Methods of Poiyaenus (diss.Harvard 1970) 120-38. My sincere thanks to Dr Phillips
for permitting me to see a chapter from her unpublished work.
echo of a blast from his conch shell. 66 Serious concern for panic as a
military problem begins with Thucydides. His two brief asides,
4.125.1 and 7.80.3 (summarized at the beginning of this section),
constitute the first theoretical observations on panic in Western mili-
tary thought. In both cases the panics are nocturnal, and in 4.125.1 a
false rumor occasions the panic. Thucydides does not connect these
passages with the KEVOV of war, although in his plan for the surprise
attack on Cleon at Amphipolis, Brasidas hopes his attack will cause
panic in the Athenians (5.9.7f).
Aeneas Tacticus, composing his Strategika between 360 and 346
B.C., knew Thucydides and apparently treated panic in some detail. 67
In the surviving long fragment on defense of cities, Aeneas promises a
discussion of panic in conjunction with guards, patrols, watchwords,
and counter-watchwords in his book on encampments (21.2, cf 25.1).
He later devotes an entire chapter (27) to panic, the lengthiest extant
discussion of the topic in ancient military theory. There is no need to
paraphrase this chapter, but it is significant that Aeneas, like subse-
quent historians and military writers, is more concerned with stop-
ping or preventing panic than with creating it in the enemy.68
Two of his exempla find echoes in the actions of other commanders
of the fourth and third centuries B.C. Aeneas records that Euphrates, a
Spartan harmost in Thrace (otherwise unknown), to stop frequent
panics in his forces, ordered his men when one occurred to sit up in
their beds with arms in hand, but anyone standing up would be
regarded as an enemy. Similar instructions are attributed to Clear-
chus, the mercenary commander in Xenophon's Anabasis; to Dercyli-
das, the Spartan general in Asia Minor in the 390's; and to Theodorus
of Rhodes, prominent in Antiochus I's victory over the Gauls ca 270
66 FGrHist 457 F 18, cf supra n.57. Romans attributed the panics of armies to
Faunus, who could be identified with Silvanus, Inus, and Pan. See Cic. Div. 1.10 I,
Nat.D. 2.6; Dion. Hal. Ant.Rom. 5.16.3; W. F. Otto, "Faunus," RE 6 (1909) 2054-73,
esp. 2054-62.
67 Bibliography on Aeneas in Wheeler (supra n.54) 8 n.40, to which should be
added: G. A. Lehmann, "Krise und innere Bedrohung der hellenischen Polis bei
Aeneas Tacticus," in W. Eck et al., edd., Studien zur antiken Sozialgeschichte:
Festschrift Friedrich Vittingho!(Ko/n.histAbh. 28 [Cologne 1980]) 71-86; Thucydides
and Aeneas: W. A. Oldfather's Loeb edition of Aeneas Tacticus, Asc/epiodotus,
Onasander (London 1923) 10; A. Dain and A.-M. Bon, Enee Ie Tacticien, Polior-
cetique (Bude ed., Paris 1967) xix; A. W. Lawrence, Greek Aims in Fortification
(Oxford 1967) 58; contra, S. Celato, "Enea Tattico: il problema dell' autore e il valore
dell' opera dal pun to di vista militare," Memorie Padua 80 (1967/8) 60 with n.35.
68 Preventing panic: Aen. Tact. 27.2-13; creating panic: 27.14, cf Polyb. 3.93f, Liv.
22.16.4-18.4; and see Onasander 41.2.
176 I10AAA KEN A TOY I10AEMOY
B.c. 69Another means to stop panic involved having a herald com-

mand silence and promising a gift of silver to whoever would report
the man responsible for turning his horse loose in the camp. In
Xenophon these are the orders of Clearchus to stop a nocturnal panic
after Cunaxa. A similar tale is told of Iphicrates, who trained his men
in how to expect panics and how to manage them.70
This discussion, though by no means exhaustive, will suffice to
demonstrate that panic was a real problem to Greek armies of the
fourth century, and that military thinkers and commanders sought
means to prevent it. Panics would continue to be a problem in
Hellenistic times, for the Romans, and for the Byzantines, as it does
for modem armies. 71 The complexity of the problem further emerges
from analysis of the causes of panic in the examples cited here.
Inexplicable or multiple causes characterize nine examples while false
TUmor or report appears only three times as an independent cause.
Yet perhaps most interesting is that despite the close association of
Pan and panic with echo, i.e., the aural sense, visual causes, the
misperception of something seen, account for six cases.72
If generals and military thinkers sought practical solutions to the
problem of panic, the topic also aroused philosophical interest. Ari-
stotle's student Clearchus of Soli composed a I1EPL TOV 7ravtlcov, of
which only a fragment survives: 73
69 Aen. Tact. 27.7-10; Clearchus: Polyaen. 2.2.10, cf Exc. Polyaen. 27; Dercylidas:
Jui. Afric. Cest. 7.1.11; Theodorus: see Wheeler (supra n.54) 15 n.77. Cf Aen. Tact.
27.2f, Polyaen. 4.3.26 (Alexander the Great).
70 Aen. Tact. 27.11; Xen. Anab. 2.2.21; Polyaen. 3.9.4, 10,32. Cf Tac. Ann. 1.66.3
71 Hellenistic: e.g. Polyb. 5.96.3, 110.1-9; 20.6.12; Paus. 10.23.1-8; Romans: Pluto
Pomp. 68.3, Caes. 43.6; Joseph. BJ 5.93, 291-95; Tac. Ann. 1.66, Hist. 1.63.1;
Byzantines: Anon. De obsidione toleranda 69 van den Berg, cf Exc. Polyaen. 3.2; Leo
Tact. 20.195; A. Dain, "Memorandum inedit sur la defense des places," REG 53
(1940) 124 (13), 125 (19); modern: F. Gambiez, "Etude historique des phenomenes
de panique," RevHistModContemp 20 (1973) 153-66. Antigonus Gonatas' victory
over the Gauls at Lysimacheia in 277 B.C., although won by stratagem (Just. 25.1.2-
2.7), did not involve panic. The head of Pan on coins of Gonatas has no relation to
the battle: Pan, a patron of the Macedonian house, appeared on the coins of earlier
kings, and Gonatas' Pan coins began ca 271, possibly occasioned by his defeat of
Pyrrhus. See G. Nachtergael, Les Galates en Grece et les Soteria de Delphes (=
MemAcRoyBelg 63.1 [Brussels 1977]) 177-80 with n.231; Pritchett III 33fwith n.93;
R. M. Matthison, ANSMN 26 (1981) 110-14.
72 Inexplicable/multiple causes: Thuc. 7.80.3; Xen. Anab. 2.2.19-21; Polyb. 5.96.3;
Diod. 20.29f, 21.2.3; Onas. 41.2; Pluto Pomp. 68.3 and Caes. 43.6 (the same occur-
rence); Tac. Hist. 1.63.1; Paus. 10.23.7; false rumor: Thuc. 4.125.1; Polyb. 5.110,
20.6.12; visual: Diod. 14.32.3, 17.85.7-86.1; Curt. 7.11.25; Onasander 6.5; Joseph.
BJ 5.91-93.
73 Ath. 9.389F = fr.36 Wehrli. Kroll, "Klearchos (11)," RE 11 (1921) 582, sees a
possible occasion for the work in Antigonus Gonatas' victory over the Gauls at
Lysimacheia, but a panic in this battle cannot be proved: see supra n.71. If the work
KAEapxo~ 0' fV T<!> TItPL TOV 7!'aVLKOV "or CTTpOV()ol," ct>7]CTl, "xol 7!'EpOL-
,/ ~, t: ') \ I 't: " .,( , '')'
Kf~, fTL uf Ot al\fKTpvOVfS Kat Ot OpTVyfS 7!'pOtfVTat T7/V YOV71V ov I-'0VOV
'5: f ~ f
' ( ) 71l\ua~, , \ \ ' " 'f , ~ '''' f f 5:'
LuOVTf~ Ta~ al\l\a Kav aKOVCTWCTLV aVTWV T71V .,.,WV71V. TOVTOV uf
" ~ ,I, ~ f '" f ,~, ~ '" f
aLTLOV 7J< TTl ."VXTl YW0l-'fV71 .,.,aVTaCTLa 7!'fPL TWV 7!'1\71CTLaCTI-'WV • .,.,aVfpWTa-
5:::' I , , ') , 'tf 't ' , , .... ()~ f
TOV uf YLVfTaL 7!'fPL Tas 0XfLaS, OTav f~ fVaVTLas aVTOLS TIS KaT07!'TpoV'
, 's:: \ , )! "" t: \ . ' , , .f. \
7!'pOCTTPfXOVTf~ yap uLa T71V fl-'.,.,aCTLV al\LCTKOVTaL Tf KaL 7!'pOLfVTaL TO
(T7d p p.a, 7TA~V TWV 6.AEKTpv6vwv. TOiYTOVS 0' ~ T71S tP.<t>d.CTEWS a~fCT87JCT'S
')'
HS l-'aX7Jv 7!'poaYfTaL I-'0VOV.
I '''.... \
TaVTa I-'fV 0
t: K'I\fapxo~.
f
This amusing passage on the libido of birds is not as irrelevant to

military panic as it may initially appear (and its biological accuracy is
not pertinent to my argument). It is unwise to draw from a single
fragment conclusions about the tone and content of a work, especially
when the excerptor is Athenaeus.
Besides various Platonic interests, unusual for a Peripatetic, Clear-
chus specialized in psychology. His I1fP't tJ7fVOV in at least two books
treated sleep as a Sonderexistenz associated with the soul's immor-
tality. His Erotica, perhaps influenced by Theophrastus' Eroticus,
concerned the phenomenology of love, especially in its pathological
forms in both men and animals (e.g. fr.27: a goose's love for a boy),
and the power of mirrors over quails is also alluded to in Clearchus'
commentary on Plato's Republic, a work that cannot have been so
silly as Ath. 389F might be taken to imply of the I1EPI. TOV 7raV'Kov. 74
Clearchus fr.36 must be placed in its proper context of Aristotelian
psychology. Clearchus states that during the mating season the male
birds' desire for copulation is so strong that both the sight and the call
of females causes the emission of semen, and even the bird's own
reflection in a mirror can produce this result. In his theory of sensa-
tion in De anima (2.12 [424a 25-34]), Aristotle distinguishes between
a sense organ and its capacity:
From this difference it is also clear why excess of objects perceived
destroy the sense organs; for if the movement imparted to the sense
was occasioned by a battle, the more likely candidate is the defeat of the Gauls at
Delphi in 279, although only Pausanias (10.23.1-8 and not in 1.4.3f) mentions a
panic. Cj Diod. 22.9; Just. 24.8. Timaeus, Hieronymus of Cardia, Menodotus of
Perinthus, and Agatharchides of Cnidus have all been proposed as Pausanias' source
in 10.23, but Hieronymus at least can now be eliminated: J. G. Frazer, Pausanias'
Description a/Greece V (London 1898) 341, cf J. Hornblower, Hieronymus o/Cardia
(Oxford 1981) 73; C. Habicht, Untersuchungen zur politischen Geschichte Athens im
3. Jahrhundert v.Chr. (=Vestigia 30 [Munich 1979]) 87-94, esp. 89 nn.8f; Nachtergael
(supra n.71) 15-99. Clearchus was probably born in the 340's, and his authorship of a
monograph on the Academic Arcesilaus attests a long life into the third century B.c.:
Kroll 580f; Wehrli 45.
74 Clearchus frr.3f, 27; Wehrli 47,54-56; Borgeaud 137 n.4.
organ is too strong, the form is destroyed; and this form is the
sensation; just as the harmony and musical note is destroyed when
the strings are struck hard. 75
If the pupil followed the master's teaching on this point, Clearchus'
argument can be reconstructed: during mating season the sight and
sound of females so overloads the males' sensory system that they
"panic," i.e., ejaculate prematurely. In all probability Clearchus draws
upon this example from nature as an analogy to the panic of armies: at
night or in enemy territory a strange noise or false rumor can produce
such fear as to overload the rational faculties and produce panic. This
view is paralleled somewhat in a modem theory of military incompe-
tence, in which generalship is seen in terms of information theory and
processing: 'noise' in the system and/or overloading the system's
capacity causes a breakdown and failure. 76 Another Aristotelian com-
ment relevant to panic appears in the Parva naturalia, where he states
that the impulse of a thought cannot be deterred from its object, and
that the impulse created by an outburst of fear will produce its own
reactions against movements to stop it.77 Hence a psychological ex-
planation why panics were so difficult to stop-they resisted reason.
Wehrli (58) sees the nEp' TOV '1Tavucov as an anecdotal-educational
work with emphasis on panic as a reaction to the stimuli of the senses
and with reference to animal psychology, as in the Erotica. This view
is safely conservative, given the content of the single extant fragment,
although probably overemphasizing a supposed similarity between
the Erotica and the IIEp' TOV '1TaV'Kov. The use of one example drawn
from nature need not indicate that the work contained many such
examples, especially since the fragment derives from Athenaeus, who
was prone to excerpt sensational and exotic anecdotes. Wehrli ignores
panic as a contemporary practical problem of Greek armies, and use
of '1Tavna, '1TaV'KoS', etc., in extant sources of the fourth century occurs
only in military contexts. The apparent natural phenomena, to which
by analogy military panic was compared, are echo and the unex-
plained sudden stampedes of animal herds, neither of which has the
least to do with the sexual behavior of birds. The extension of the
term 'panic' to describe such fowl conduct may result from Aristotle's
75 W. S. Hett tr. (Loeb ed., London 1935) 137, cf Wehrli 58.

76 See N. Dixon, On the Psychology of Military Incompetence (New York 1976) 27-
35.
77 Par.Nat. 453a23-26=Mem. 2. Cf Clausewitz 117: "In short, most intelligence is
false, and the effect of fear is to multiply lies and inaccuracies. As a rule most men
would rather believe bad news than good, and rather tend to exaggerate the bad news.
The dangers that are reported may soon like waves subside; but like waves they keep
recurring, without apparent reason."
research in natural science or may be Clearchus' own contribution.

However this may be, an interpretation different from Wehrli's can be
offered.
Aelianus Tacticus in his list of military writers notes a Clearchus
who authored a Tactica, and Arrian adds that this is not the Clearchus
of Xenophon's Anabasis. Since Aelian and Arrian give these writers in
chronological order (so far as it can be determined), Clearchus tacti-
cus must be the contemporary of Pyrrhus of Epirus and his son
Alexander in the first half of the third century B.C. Kochly and Rii-
stow, followed by Jahns, identified this Clearchus with the Peripatetic
from Soli, and nothing contradicts that a Peripatetic for whom the
titles of fifteen works are known, of which one treats military panic (as
I argue), could have written a Tactica. 78 Peripatetics were interested
in military affairs: Aristotle devoted sections of the Politics and the
Rhetoric to this topic; Demetrius of Phalerum wrote a Strategika; and
Phormio, who attempted to instruct Hannibal on the art of general-
ship at the court of Antiochus III, was a Peripatetic. 79 Some philoso-
phers considered tactics a branch of mathematics, and Clearchus'
commentary on Plato's Republic was a TIEpt TOOV (V Tn TIAaTwvo~
7TOALTE{~ IJ-a87JIJ-aTLICOOS EtP7JIJ-(Vwv. 80 Since neither Aelian nor Arrian
added a city of origin to the name of his Clearchus, doubt can always
intervene, but compelling evidence supports the probability of Koch-
ly's and Riistow's identification. The burden of the argument really
lies with denial of the attribution, essentially based on a modem
prejudice that ancient philosophers were too concerned with lofty
esoteric matters to consider the practical problems of war in their
society.
Clearchus' authorship of a Tactica adds further strength to the
probable military nature of the TIEpt TOV 7TavLKov, but nothing suggests
78 Ael. Tact. 1.2; Arr. Tact. 1.1; H. Kochly and W. Rtistow, Griechische Kriegs-
schriftsteller 11.1 (Leipzig 1855) 29f; M. Jahns, Geschichte der Kriegswissenschaften I
(Munich 1889) 47. Kbchly and Rtistow also considered the possibility of Clearchus,
tyrant of Heraclea in Pontus and a student of Plato and Isocrates, but he is not
known to have written anything, nor does he fit chronologically; see Lenschau,
"Klearchos (4)," RE 11 (1921) 577-79. Wehrli rejects the identification without citing
Kbchly and Riistow or Jahns and in his fr.I13 gives only Arrian's text, thereby
ignoring Aelian. Unaware of the chronological order of Aelian's and Arrian's lists, he
considers Arrian's reference too vague for a famous figure like Clearchus of Soli.
Kroll (supra n.73) 583 denies the identification without argument.
79 Arist. Pol. 1.8.12; 2.9.7f; 3.4.14f, 7.3f; 4.3.1-3,13.7-11; 6.7.1-3, 8.14f; 7.2.9-17,
5.3-6.8, 8.7,9.4-6, 14.13, 19-22; Rh. 1.3f; Demetrius: Diog. Laert. 5.80; Phormio:
Cic. De Or. 2.75f.
80 G. Aujac, Geminos, Introduction aux phenomenes (Paris 1975) 114f (fr.2), cf R.
von Scala, Die Studien des Polybios (Stuttgart 1890) 295 n.4; M. C. P. Schmidt,
"Philologische Beitrltge zu griechischen Mathematiker," Philologus 45 (1886) 74f:
Clearchus frr.3f Wehrli.
that the two works are identical: Arrian .in particular would not have
been so lax about a title. Kochly and Rustow also believe that the I1EPI.
TOV 71'aVLKOV belongs to the genre of Homeric tactica, which saw Homer
as the father of military science and used excerpts from the Iliad and
Odyssey to instruct in the military art. Conceivably panic was a
frequent theme of such works. 81 If this view is correct, Clearchus'
treatise would conform to a Peripatetic interest in Homeric commen-
tary. Aristotle wrote a Homeric Problems, and Demetrius of Ph ale rum
wrote commentaries on both the Iliad and the Odyssey, curiously
listed in Diogenes Laertius' catalogue immediately after his Strate-
gika. More significantly, however, some fragments of Aristotle's
Homeric work treat military affairs, and one (fr.159 Rose) explicitly
refers to the problem of nocturnal fears in armies. 82
Clearchus' I1EPI. TOV 71'aVLKOV thus appears a thoroughly Peripatetic
work, drawing on Aristotelian psychology and analogy from natural
science, concerned with exegesis of Homer, and bringing this knowl-
edge to bear on a contemporary practical problem, the panic of
armies. But what has all this to do with the history of the proverb
71'oAAtt KEVtt TOV 71'OAEP.OV? It is now possible to suggest (although not
conclusively prove) that Clearchus is the source of the proverb as
found in the collection attributed to Diogenianus.
Certainly the phenomenon of panic fascinated Clearchus, and his
interest has deep Aristotelian roots. The connection of the proverb
with panic, as already noted, begins with Aristotle, although he does
not explicitly call the phrase a proverb. Aristotle, however, founded
the genre of the collection of proverbs, which he considered a form of
pre-literary philosophy, and with which he embellished his later writ-
ing on politics and rhetoric. His Paroimiai in one book prompted
Theophrastus to compose a collection in one book, but Clearchus ex-
celled his predecessors with a collection in two books.83
81 KochlylRiistow (supra n.78) 30; Ael. Tact. 1.1f; Paus. 4.28.7f, cf Wheeler (supra
n.54) 17 n.85, 18 n.91.
82 Diog. Laert. 5.26=catal. Arist. pp.7 Rose (no. 118), 14 (no. 106), 16 (no. 147);
Arist. frr.152, l59f Rose; Demetrius: Diog. Laert. 5.80f. Cf Wheeler (supra n.54) 16
n.83. Homeric scholia underlie Pritchett's unsuccessful attempt (IV 7-93) to prove
use of the phalanx formation in Homer, although he fails inter alia to demonstrate
their historical accuracy for conditions of the eighth and seventh centuries. Cf TAPA
117 (1987) 169 n.54. To Pritchett's claim that I misrepresented his views (II 217) in
GRBS 23 (1982) 225 n.ll, I must note his misrepresentations of mine in IV 16 n.51
and 61 f n.189 and urge that readers judge the matter for themselves.
83 Diog. Laert. 5.22=catal. Arist. pp.8 Rose (no. 138), 15 (no. 127); Arist. fr.13
Rose; Theophrastus: Diog. Laert. 5.45; Clearchus frr.63-83 Wehrli; R. Pfeiffer,
History of Classical Scholarship (Oxford 1968) 83f; Rupprecht (supra n.50) 1736f;
Wehrli 68.
If, as argued above, Aristotle borrowed the phrase KEVbV TOU 7rOAEP.OV
from Thucydides and changed it to the plural, it is unlikely that
Aristotle designated the phrase a proverb, and the first evidence for its
proverbial use is Polyb. 29.16.3. Since the phrase in its proverbial
form postdates Eth.Nic. 3.8.6, and so appears in Diogenian. 7.80, Ari-
stotle and Polybius offer respective termini post quem and ante quem
for when the phrase became a proverb. Both Clearchus and Theo-
ph rastus fit these termini, but no connection of Theophrastus with
panic can be established, and too little is known of Theophrastus'
proverbs to consider his candidacy seriously. Conceivably either The-
ophrastus or Clearchus as the third generation to use the phrase
KEVbvIKEV'a TOU 7rOAEP.OV could have recognized it as a proverb.
But apart from the common interests of Clearchus and Aristotle in
panic and proverbs, the new element of the proverb as found in
Diogenianus, the 'Spartan explanation' connecting the proverb with
panic, can also point to Clearchus, whose work Bioi (lifestyles, not
biographies) included Spartans and stressed a preference for a moral
life over luxuriousness. 84 In fact two proverbs of Clearchus deal with
Spartans: both appear in Diogenianus, but one only in a variant
version. In all, six of Clearchus' twenty-one extant proverbs are found
in Diogenianus either wholly or in variants, and Diogenianus' seventh
book contains two variants of Clearchus' proverbs (7.13, 23) besides
the KEV'a TOU 7rOAEP.OV (7.80).85 The case for Clearchus as the Urquelle of
Diogenianus is not unquestionable, but the arguments of Quellenfor-
schung rarely are. Nevertheless, all things considered, Clearchus re-
mains a most likely candidate-perhaps the only one for whom a case
can be made at all.
v
As a final note on 7rOAA'a KEV'a TOU 7rOAEP.OV, its repeated use in
Diodorus permits a suggestion about his source. The proverb occurs
84 Clearchus frr.37-62 Wehrli, esp. fr.39; Kroll (supra n.73) 581. The Spartan
element permits elimination of the Atthidographer Demon (jl. ca 300 B.C.), whose
collection of proverbs contained forty books. No references to Sparta or Spartans
appear in Demon's extant fragments, and none of his proverbs is found in Dio-
genianus. It is also unlikely that Demon was a Peripatetic: see Felix Jacoby, Atthis
(Oxford 1949) 78, and ad FGrHist 327 (pp.201f); Schwartz, "Demon (6)," RE 5
(1903) 142f.
85 Spartan proverbs: fr. 74=Diogenian. 1.83, fr.73 cf. Diogenian. 5.1; Clearchus and
Diogenianus: fr.65 cf. Diogenian. 7.23, fr.66a-c cf. Diogenian. 7.13, fr.73 cf. Dio-
genian. 5.1, fr.74=Diogenian. 1.83, fr.77=Diogenian. 8.62, fr.81 cf. Diogenian. 2.41.
In comparison, five proverbs of Clearchus or their variants occur in Zenobius: frr.65,
66b, 67, 78, 81 cf. Zen. 2.32. Three are common to both Diogenianus and Zenobius:
frr.66b, 67, 81.
here four times, more than in any other author, though it hardly
qualifies as a standard motif given the vast extent of Diodorus' extant
work. Three of the four occurrences concern events during Agatho-
cles' tyranny, and one is associated with Alexander's stratagem to
capture Aornus. 86 This distribution suggests a source who treated
both Alexander and Agathocles. Duris of Samos has long been recog-
nized as either the Haupt- or Nebenquelle of Diodorus for Agathocles,
and Duris also devoted three books of his Macedonica to Alexander. 87
A plausible case for Duris as Diodorus' source for the proverb can be
constructed.
First, Duris was a Peripatetic: he studied with Theophrastus in
Athens ca 304 B.C. and wrote a Homeric Problems besides his various
histories and other works. 88 Like Clearchus of Soli, with his tale of a
goose and a youth, Duris could tell a love story about a dolphin and a
boy, and his work was moralizing, preaching rejection of luxury.
Indeed Kebric has catalogued so many parallels between the frag-
ments of Duris and Clearchus that a direct relationship seems most
probable. 89 Furthermore, Duris liked proverbs, of which many sur-
vive in Zenobius but none in Diogenianus. 9o
Second, the proverb 7roAAa KEva TOt) 7rOA€p.ov in Diodorus suggests a
criterion for determining his source. Meister for other reasons assigns
Diodorus 20.60-67 (the proverb in 20.67.4) to Duris, but 20.29f (the
proverb in 20.30.1) to Timaeus, and he denies a discernible source to
21.2 (the proverb in 21.2.3).91 Since Aristotle, however, all occur-
rences of the proverb are associated with panic, and this holds true for
Diodorus, although only in 20.67.4 is the proverb associated with the
word 7raVLK(>s. Other instances of panic in Diodorus explicitly indi-
cated by 7raVLKOS derive from Ephorus and Timaeus, but knowledge of
the proverb cannot be demonstrated for either of these authors. 92
Timaeus certainly did not treat Alexander. It is not in dispute that
Duris was a source for Diodorus' account of Agathocles. His Peripa-
86 Diod. 17.86.1; 20.30.1, 67.4; 21.2.3.

87 See Meister 133f, cf Melber (supra n.65) 504-10: Polyaen. 5.4.1-8 on Agatho-
cles=Timaeus, Diodorus on Agathocles=Duris; Kebric 49f.
88 FGrHist 76; Homeric Problems: FF30, 88?, 92?, 89-91; Kebric 5f.
89 Dolphin: F7, cf Clearchus fr.27 Wehrli (goose and boy); Kebric 28-31, cf Wehrli
56, 61. Kebrie's monograph is an attempt to rehabilitate Duris' reputation, arguing
that he was a moralist and that the sensationalistic fragments extant represent things
Duris refuted.
90 FF62f, 85, 95f, cf 84, 93 with Jacoby's commentary.
91 Meister 149, 158, 164. Kebric rejects Meister's analysis, because Meister claims
moral criteria have no role in Duris, and Kebric denies that anything conclusive can
be gained from studies of Diodorus' sources: 75-77 with n.72.
92 Diod. 14.32.3=Ephorus: see supra n.57; Diod. 15.24.3=Timaeus: Meister 103.
tetic background, his apparent relationship to the scholar of panic

Clearchus of Soli, and occurrence of three of Diodorus' four instances
of the proverb in the account of Agathocles all speak for Duris as
Diodorus'source.
If this argument is correct, then Duris must also be the source of the
proverb in Diod. 17.86.1. Duris treated Alexander in Books 6-9 of his
Macedonica. Recent views on the problem prefer Cleitarchus as Dio-
dorus' source, but Cleitarchus was also a source for Duris and both
reflected an anti-Macedonian attitude. Use of Duris at Diod. 17.86.1
need not indicate that Diodorus used Cleitarchus exclusively through
Duris. 93 In fact Diodorus' account differs slightly from both that of
Arrian and that of Curtius, with whom Diodorus is alleged always to
agree. If Cleitarchus is the sole source of Diod. 17.86.1, then the
proverb should also appear in Curt. 8.11.19-24. But it is lacking. 94
Perhaps Duris deserves better treatment by students of Alexander.
Furthermore, Duris with some probability can also be posited as
Cicero's source for the proverb. Cicero had read Duris and considered
him homo in historia diligens (AU. 6.1.18). When Cicero set out to
assume his Cilician command, he tried to compensate for his lack of
military experience by studying Pyrrhus' Tactica and Cineas' epitome
of Aeneas Tacticus. 95 Perhaps Cicero chose Duris as his vade mecum
for Alexander's campaigns. Cicero's use of the proverb (AU. 5.20.3)
occurs just after he reports his activity while camped at Issus-an
account embellished with a reference to Alexander's battle there with
Darius. Indeed Cicero's praise of Duris in Au. 6.1.18 (20 Feb. 50)
comes only two months after his use of the proverb in 5.20.3 (19 Dec.
51). Cicero's citation of a proverb prominent in Duris and used in the
latter's account of Alexander, when juxtaposed with AU. 6.1.18, sug-
gests more than a passing familiarity with the historian on Cicero's
part during his governorship in Cilicia.
Originally a Thucydidean abstraction for a universal truism of war,
the phrase 7rOAAa KWa TOt) 7rOA'fJ-0V thus became a Peripatetic proverb
93 N. G. L. Hammond, Three Historians of Alexander the Great (Cambridge 1983)

79, 84f; Kebric 9,46,49,65. Cf J. E. Atkinson, A Commentary ofQ. Curtius Rufus'
Historiae Alexandri Magni Books 3 and 4 (=LondStudClassPhil4 [Amsterdam 1980])
64-67. Goukowsky (supra n.11) 120 attributes the proverb to a collection of
stratagems. But neither of the two extant examples of this genre, Frontinus and
Polyaenus, includes the proverb.
94 See supra n.44. Duris could also be responsible for the proverb in Curt. 7.11.25
and its possible occurrence in 4.13.5. Atkinson (supra n.93) rejects use of Duris in
Curtius Books 3 and 4.
95 See Wheeler (supra n.30) 13f.
through its association with panic, and passed to the Romans in both
its Thucydidean (Cato) and its Aristotelian forms (Cicero, Curtius).
Few Greek proverbs have such a traceable history.96
DUKE UNIVERSITY
June, 1988
96 My research has been generously supported by the Harry Frank Guggenheim

Foundation.

The History of A Greek Proverb

Uploaded by

Copyright:

Available Formats

The History of A Greek Proverb

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The History of A Greek Proverb

Uploaded by

Copyright:

Available Formats

WHEELER, EVERETT L.

, "Polla kena tou polemou" [Greek]: The History of a Greek Proverb ,

IToAAa KEVa TOt) 7ToAEP.OV:

I N THE SPRING of 427 B.C. a Peloponnesian fleet under the nauarch

154 nOAAA KENA TOY nOAEMOY

do arise from datable historical events, 4 and it is only an assumption

4 Cf John Patrick Lynch, "Hipparchos' Wall in the Academy at Athens: a Closer

EVERETT L. WHEELER 155

7 A. W. Gomme, A Historical Commentary on Thucydides II (Oxford 1956) 292f.

D. R. Shackleton Bailey, Cicero's Letters to Atticus III (Cambridge 1968) 226f.

156 I10AAA KENA TOY I10AEMOY

EVERETT L. WHEELER 157

proof is adduced from an adscriptum to Au. 5.20.3 in the delta class of

13 Clarence P. Bill, "TA KAINA TOY TIOAEMOY," CP 32 (1937) 160f.

158 IIOAAA KENA TOY IIOAEMOY

turnal panic but also of Macedonians avoiding the horrifying sight of

EVERETT L. WHEELER 159

and Aristotle antedate Cato}S Knowledge of Aristotle finds no trace

3.8, 17; Astin (supra n.17) 149.

(1970) 1298; H. Heubner, Publius Cornelius Tacitus, Die Historien II (Heidelberg

160 I10AAA KENA TOY I10AEMOY

EVERETT L. WHEELER 161

Gomme interprets the phrase solely on the basis of Thuc. 3.30,

162 I10AAA KENA TOY I10AEMOY

does represent one of its components. Thucydides' language is ob-

EVERETT L. WHEELER 163

The themes of negligence, lack of preparation (a variation on the ideas

164 I10AAA KENA TOY I10AEMOY

the enemy must be carefully derived by a sensitive and discriminating

34 Clausewitz 84, 101, 103.

EVERETT L. WHEELER 165

"'I. \ ' ~ \. ~ .. , < ~.. ~ \ l' \ \ \ \

Does Aristotle's understanding of the phrase differ from Thucydides'?

166 nOAAA KENA TOY nOAEMOY

If allowance is made for Aristotle's transfer of the phrase from an

EVERETT L. WHEELER 167

will be encountered again below; second, Polybius is the first author to

AELjJ.jJ.EV1JV 4>vAaK~v, OLOOV!i' E'fOOOV TOL!i' f30VAOjJ.EVOL!i' EK TfjS" 7TETpa!i' &va-

The context is Alexander's use of the well-known stratagem of leaving

168 I10AAA KEN A TOY I10AEMOY

The next three occurrences in Diodorus all concern Agathocles. 45 In

Once again the context of the proverb is a nocturnal surprise attack,

attacks the retreating Indians, creating a panic. In Curtius (8.11.19-24) Alexander's

EVERETT L. WHEELER 169

camp, where their approach was mistaken as a surprise attack by the

The passage requires no commentary: the KE"Va are self-deception and

170 IIOAAA KEN A TOY IIOAEMOY

As should by now be clear from our examination of the proverb, the

49Cf AU. 5.20.3: erat in Syria nostrum nomen in gratia.

EVERETT L. WHEELER 171

second explanation seem possible: either this explanation preserves

51 Tyrtaeus 11.35fWest, P.Oxy. XLVII 3316.14.

172 nOAAA KENA TOY nOAEMOY

the dilemma of every general. The KEVOV signifies the problem of

EVERETT L. WHEELER 173

not a personification), the son of Ares, who often contributes to the

174 nOAAA KENA TOY nOAEMOY

(34-37), attributed to Euripides but whose authenticity and date have