The "Supplices" of Euripides Diggle, James Greek, Roman and Byzantine Studies; Fall 1973; 14, 3; ProQuest pg. 241
42
42/43
44
44/45
46/47
The Supplices of Euripides
James Diggle
I
., ,
LK€T€VW C€, Y€pCl.UX,
......, , \
Y€pCI.LWV €t( CTOfLCl.TWV, TTpOC
" \ I yovv TTLTTTOVCCI. TO COV·
tavofLoL T'KVCI. AVCCl.L
'/"0' , .. 't' LfL€VWV V€KUWV ° L
KCl.TCl.A€tTTOVCL fL'AYJ
OCl.VaTC[) AUCLfL€A€L OYJpdv OP€tOLCL j3opav.
COMMENTATORS and emendators, with few exceptions, find the
antecedent of the relative Ot in V€KVWV in line 44/45:
" ... corpses which leave behind their limbs as a prey to
beasts." The gibbering tjJvX~' knocking in vain at the gates of Hell, may
have left its limbs behind as carrion. A corpse on the battlefield has
abdicated control over its limbs: it does not enjoy the privilege of be
queathing them to anybody. The conjectures of the interpreters in
line 44 are not such as to redeem the improbability of their interpre
tation: alla fLOL T'KVCI. AVCCl.L cfoOLfLEVWV V€KVWIl ed. Brubachiana and the
early editors, rendered as "ut redimas mihi filiorum extinctorum
cadauera" or "ut eximas me os liberos ex cadaueribus defunctorum,"
and modified by Brodaeus and Markland to avCI. fLOL KTA., "surge mihi,
redime filios meos, etc."; alla A€LtjJCl.vCI. AVCCl.L Kirchhoff, ava fLOL CTtXCI.
AvcCI.L Musgrave, a7T(~ CWfLCl.TCI. AVCCl.L Wecklein,1 avofL' Cl.LCX€CI. AVCCl.L
Bruhn apud Murray.
A few have tried a different path. Reiske and Markland find the
antecedent of Ot in TEKVCI., and Markland offers a choice of three con
structions for the phrase cfoOLfLEIlWV V€KVWV: (i) "ex cadaueribus defunc-
1 Ed. maior (Leipzig 1898) and small annotated edition (Leipzig 1912). The conjecture is
accepted by the latest editor, G. Italie (Groningen 1951), who also changes ¢>8tp'€vwv V€KVWV
to V€KVWV ¢>8tp'€vwv. His laconic reason for this change (" zie antistr." is all he says) I take to
mean that ¢>8tp'€vwv in the strophe now occupies the same position as ¢>8LfLEVOVC in the anti
strophe. Such correspondences occasionally occur in Euripidean lyrics, but they are not to
be introduced by this sort of 7T€pt€pyla.
241
242 THE SUPPLICES OF EURIPIDES
torum,"2 (li) genitive absolute, "cadaueribus tabescentibus," (iii) de
pendent on P.lA1] (so Reiske). "Haec nobis incerta sunt, quia ea pronun
data non audiuimus: un de nescimus ueram distinctionem"; but it
will take more than punctuation and pronunciation to turn Mark
land's text into intelligible Greek. Gregoire3 writes «va P.Ot ·rlKva
AVCat. cflhp.lvwv V€KVWV p.~ KaTaA€l7TOVCa p.IA1] KTA., which Professor Page
once called-perhaps a trifle indulgently-"pretty, though ... un
convincing.'" Murray produced a text of fits and starts, or, as he
called it, of "clamores confusos precantium": «voP.Ot •• • -'TtKVa AVCat.
-O'P.EVWV KTA., "Impii Thebani" ... "Reddere6 filios!" ... "qui mor
tuos feris relinquunt." This division of speakers was exploded by
Page, who himself conjectured «va P.Ot V€KPa. Aikat rfo0tP.EVWV T€KEWV.6
But Murray has spotted what seems to me to be an obvious truth:
that the only party which may be described as leaving limbs as a prey
to beasts are the Thebans, who are refusing burial to the corpses.7
And this view seems to be shared by the author of the most recent
conjecture known to me: A. Y. Campbell8 conjectures, without ex
planation, avop.ovS' KaTcX7TaVCat, "put a stop to the lawless men
who ... ," in which the sense is more plausible than the alleged cor
ruption.
In listing the conjectures which take V€KVWV as the antecedent of at,
I omitted to record two conjectures which are simpler and better than
the rest. O. Ribbeck9 proposed a7TO p.o, for «voP.Ot, with the construc
tion a7To p.ot TEKva Avcat rfo0'P.EVWV V€KVWV, "release for us our children
from the dead corpses." Tmesis is common in Euripides' lyrics;lO
tmesis of the same verb, in a similar construction, occurs at Hom. Od.
B Defenders of this construction quote no parallel. so I offer them Aesch. Ag. 1023 TWV
~8,p.lvwv ava"nv ("bring up from the dead").
a Bude ed. (Paris 1923).
, CQ 31 {1937} 96.
, 'Redime' in fact. since Jn80ika in 48 shows that '\ikaL is middle imperative and not
aorist infinitive. But that would make Murray's text even less coherent.
I Ix.cit. (supra n.4).
7 For KaTa.\£l1r£&V used ofleaving corpses on a battlefield see II. 12.226-271To,uovc yap TpdJwv
KaTa).(lI/Jop.£v, ove K£V • AXa,oll Xa.\Kip 8uwewcw. Before Murray the only note of disquiet I can
find is A. Matthiae. Obseruationes criticae (Gottingen 1789) 14: "KaTa.\€l1Tf.&V P.£'\11 de moriente
uix bene dicitur; et hoc loco esse saltem deberet Ka'T€'\£LI/Jav." I will not repeat his conjec
ture, which he withdrew in his edition (text 1814. commentary 1823).
8 In his edition of Helen (Liverpool 1950) 123.
• RhM N.F. 31 (1876) 614.
10 KUhner-Gerth I 534-35. W. Breitenbach. Untersuchungen {ur Sprache der EUripideischen
Lyrik (Stuttgart 1934) 266.
JAMES DIGGLE 243
12.420-21 cbr6 TotxoVC I AUCE" KAJ8wv Tp6moc ("the wave removed the
sides from the keel").H But I doubt if (bro is the preposition we want
in this context; EK is what we should expect, and EK was neatly sup
plied by a second conjecture of Professor Page, published but reduced
to unintelligibility by misprinting in A. S. Owen's edition of Ion (Ox
ford 1939) p.117: ava Jim 'r€Kva AVCat 'K cpfltfL€vwv. For the preposition
see 346 V€KPOVC EKAvcOfLat, and for fLOt see 168 cwcov V€KPOVC fLOt.
"There are no certain cases of prodelision after at in tragedy," says
Platnauer.12 But he is wrong: there is at least one. At Soph. OC 1608
1T€covcat 'KAaZov (Heath, KAaZov codd.) the manuscript reading is not
to be defended by the plea that the syllabic augment may be
omitted in messenger speeches: for the conditions under which such
omissions are permitted see Page on Med. 1141. Of the four remaining
possible instances, I have already shown that Platnauer's doubts
about two of them are justified;13 but two instances which are prob
ably to be accepted are Ret 953 a;'p~cOfLat 'ycfJ (Porson, a;'p~cofLat TO L),
and IA 1396 Y€V~COfLCXt 'yw (Reiske, y€~COfL' EYcfJ L). Platnauer has
shown that there are seven instances of the prodelision -fLat 'ycfJ in
Aristophanes.14
There is only one drawback to accepting this conjecture: the ante
cedent of ot is still V€KVWV. But change the case of the relative and all
will be well:
)/ I \..... ,
ava fLOt TE"KVa I\VCCU K
.I..(}' '''' '(J 'fL€VWV V€KVWV wv
KaTaA€{1ToVct fL€ATJ • • •
"arise, and release for us our children from the corpses whose limbs
11 The same construction (as a1TO'\V€tv 7l Ttvoc) is used with the uncompounded verb at
470 '\vcaVTa C€JLv& CT€JLJLaTwv JLVCn7pta, "release the JLvcn7pta (Demeter's temple: cf. 173)
from the suppliants' garlands." So the passage is rightly explained by B. Lavagnini, AJP 68
(1947) 84-86. Commentators join C7€JLp.aTwv JLvcn7pta or, since that is an impossible phrase,
accept Nauck's l~pLa. And there is one more place in the play where the verb a1TOAQ€t"
ought probably to be restored. At 638-39 the messenger, announcing victory, declares to the
chorus '\oyov 8' C€ I JLaKpov a1To1Tavcw, which, one would suppose, means "I shall stop you
from making a long speech," than which no remark could be less apposite. H. van Her
werden, Mnemcsyne N.S. 5 (1877) 36, conjectured a1To'\vcw, "I shall relieve you of a long
speech" (i.e. I shall speak briefly). Compare Hec. 918, where Murray very plausibly conjec
tures KaTa},vcac for KaTa1Tavcac.
12 CQ N.S. 10 (1960) 141-
11 Heracl. 999, IA 143;: see CQ N.S. 22 (1972) 244.
U And let me add another instance for consideration: Blaydes' .pa{vETat (' K > nEW" at Aesch.
Pers. 604, which is accepted by Page (OCT. Oxford 1972).
244 THE SUPPLICES OF EURIPIDES
they are leaving for the beasts." ava is an invitation to Aithra to leave
the altar and approach her son on the chorus' behalf (cf. Ale. 276 aAA'
ava TOAfLa, Tro. 98-99 ava, 8vc8atfLov, 7T€8oO€v K€cpaA~v (P, K€cpaAa V), I
E7Ta€tp€ 8lp'T}v,15 Soph. Aj. 192 aAA' ava Eg e8pavwv). The subject of
KaTaA€{7ToVCt is left unexpressed, as well it may be: the identity of the
subject is not in doubt, for this reprehensible behaviour of the The
bans in refusing burial was described only a few lines earlier. For the
idea 'release from the corpses' (i.e. 'from the corpse-strewn battle
field') see 762 Olpa7T€c 1jyov €K cpovov, "the servants brought (the
corpses) out of the carnage." The corruption of ciJv to ot is easy enough:
either ciJv was lost by haplography after V€KVWV and ot was supplied as
a subject for KaTaA€{7ToVCt, or, more likely, ciJv was simply assimilated
to the case of the subject of the following verb. The repetition of the
same sound in the adjacent syllables-wv ciJv causes no offence: see the
passages I have cited in ProcCambPhilSoc 194 (1969) 59.
346
II
"" ''>'' l' , " \' opacw Tao' €tfLL KaL V€KPOVC EKI\VcofLaL
\' '0' "" 'f3' '" ' I\OYOLCL 7T€L WV' €t O€ fLYJ, LCf oopOC
"'>' ''>'''' ", .1..0' 0 ~ YJOYJ TOO ECTaL KOVXL CVV 'f' Ovcp €wv.
346 8pacw Kirchhoff, 8pacwv L. 347 7T€tOwV Nauck,
7T€tcwV L.
"I shall do this. I shall go and redeem the corpses by using persuasion;
failing that, it will be done 7f8'T} by armed force and without divine
disp leasure."
The word 7f8'T} harbours a problem generally ignored. "Failing that,
it will be done 7f8YJ by armed force." While 7f87J commonly refers to
what is to happen in the immediate future ('now at once'), it is not
clear that it may legitimately be used in the apodosis of a conditional
sentence to refer to an event which, so far from being immediate, is
15 I have seen no convincing treatment of these lines. Some, with no warrant, give ava the
force of a transitive verb: "ava construendum uidetur cum K€cf)(I)..~V, i.e. aV€X€ K€CPaA~V"
(Hermann), "ava for ava€£p€, as John Milton (ap. Barnes) rightly took it" (Paley); Murray
prints K€CPaA~, with an impossible change of addressee to follow; Parmentier punctuates
K€CPaA~V IllT(f€£p€, SlpT)v, which is abominable style; only Musgrave's addition of T€ after
8lpT)v deserves consideration (K€CPaATJV I €7T(X€£P€ 8lPTJv T'). But I wonder whether 'lT€868£v
K€CPaA~V, €'lTa€£p€ 8lpT)v does not belong to that species of the «'lTO KO£VOV construction illus
trated by Soph. El. I05-{)6 ECT' av 'lTafLcp€yy€ic aCTpwv I p£1rac, A€1JccW 8£ T68' -?fLCXP, Hor. Carm.
1.30.5-6 solutis I Gratiae ~onis properentque Nymphae, though in these and all other instances
known to me there is a copula and not asyndeton.
JAMES DIGGLE 245
contingent upon a future event of uncertain time. Or, to formulate
the problem bluntly, can ~oYJ mean TOTE? No evidence has been
shown that it can.
Nauck16 deleted 348 and left El OE J.t~, f3tq, oop6c effective enough in
its abruptness; and unless it can be bettered by conjecture, the dele
tion deserves serious consideration,17 Such a conjecture is not Wila
mowitz's Kat. o~, for it gives to the line a frigid aimlessness which, had
the manuscript presented this reading, would probably have been
despatched by Nauck with the same remedy. That Beck conjectured
7j o~ I mention only because it is creditable to make even a bad con
jecture when others are asleep.
If the text is unsound, then perhaps the replacement of TOO' by TOT'
will mend it:
, ~ \ , f3' ~ \
EG OE J.tYJ, Lq, OopOC
,,~ ,." ,\ \ ,J..8 I 8-YJoYJ TOT ECTaL KOVXG CVV 'f' OVGcf3YJT~COJ.tEV. See also Aesch. PV 910-11 (Zeus will
b h ) \~, , \ I T7 I I , ,,~ \ - 8 ' e overt rown 7TaTpoc 0 apa npovoV TOT YJOYJ 7TaVTEI\Wr; Kpav YJCETCXG,
A 970 71 " ~ \ , Z \ , ", ,J.. - I" I '''~ .f.-g. - OTaJl OE n:vxYl EVC a1T of1-'f'aKoc 1TLKpac 0 LJlOJl, TOT YJOYJ 'f'vXOC
€JI 06J.toLC 1TEAEL, Cho. 819 Kat. TOT' ~o1J (Blomfield, TOTE o~ M) ... J.tE8+
COJ.tEV, Soph. OC 437-41 Xpov8ovEPot TLC, alcxvVYJc aTEp I ECTW, Ag. 217 EO yap
E/''Yj, Cho. 868 eL'Yj 8' E7T~ ""{KT/, Eur. Med. 89 EO yap ECTCXG, HF 1292-93 cP o·
aEi KaKWC I ECT', Hei. 1273, Or. 1106.
III
365 C <'f3 "A '1" ."~ HO. L1T1TO OTOJI pyoc, W 1TaTpLOJI EJ.tOJl 7TEOOJl, STR.
EKAVETE Ta8', EKAVETE
JI rl '8 \ avaKTOC OCLa 1TEpL EOVC
16 BullAcImpSt.Petersburg 22 (1877) 92.
17 It is accepted by Wecklein in 1898 but not in 1912.
246 THE SUPPLICES OF EURIPIDES
370 , , '" \, " ,- -Et yap £7Tt 7'£PILa Kat 7'0 7T1\£OV £ILWV KaKWV ANT.
375 __ \' ~'" __ \ 1\ 'Q' I KaI\OV 0 ayal\/-La 7TOI\£CLV £VC£fJ7JC 7TOVOC STR.
, ,,, \, "
xapw 7' £X£L 7'av £C aL€L.
I have transcribed the first strophe in order to present it with a
question-mark at the end. All the editors I have seen, with the excep
tion only ofItalie, make the sentence a statement: H Argos, you heard
this good news from king Theseus." If Argos heard it, fifty miles
away, then Athens is ruled by king Stentor. I translate the remainder.
HMay he, in going as far as the ending of my miseries and still fur
ther,18 remove the bloody ayc:tAw:t. of a mother and make the land of
Inachus friendly to himself by doing it service. Labour undertaken in
a pious cause is a fine &yaA/-La for cities and wins everlasting gratitude."
The Hbloody a:yc:tA/-La of a mother" is taken to mean the bloody corpses
of their fallen sons, and there is no reason why the words /-La7'/poc
&yc:tA/-La t/>IJVLOV should not have that meaning: see 631-327'0 COl' ayc:tAw:t.,
7'0 COl' t8pv/-Lu I7TOA€OC ("the glory, the stay, of your city," meaning
these same dead heroes), 1163-64 OVKl7'L t/>{).ov I t/>{).ac ayc:tA/-L' t)"'olLul C£
ILU7'POC ("no longer shall I see you, dearly beloved delight of a loving
mother"), IT 273, Aesch. Ag. 208, Soph. Ant. 1115, tr. fro adesp. 126.3.
And for the adjective t/>OVLOV see 812 CW/-Lu(J' ui/-Lu7'oc7'ayfj. But there are
two difficulties. First, the recurrence of Ci.yc:tA/-Lu only eleven words later
and with a different connotation betrays clumsiness to a high degree.
Negligent repetition within a short space of common and colourless
words is a well-known feature of tragic style: the word &yc:tA/-Lu is
neither colourless nor common. Second, the failure to define the verb
.ucac ~1Tl TO
1TAlOV Uce-o p.oleac by Wilamowitz, Analecta Euripidea (Berlin 1875) 94. And for TO 1r>'EOV see
also 158 TO (Musgrave, TtL) 8~ 1r>'EOV 'further than that', 'moreover', a certain conjecture, in
spite ofG. Zuntz, The Political Plays of Euripides (Manchester 1955) 69 n.5.
JAMES DIGGLE 247
I will add that cP0VLOV in this context is not apposite, although I am
well aware that others will find it a choice epithet.
In place of a:yaA/-ux I suggest a/-wYfJ-a: "may he put an end to a
mother's bloody cheek-tearing." This clause defines -rlpfJ-a 'TWV ll-'wv
KaI(WV in the same way that'T6 7TMov €n is defined by yaJ! •• • oV1]eae.
The noun appears at Andr. 826-27 QVVXWV 'TE oa;;' afJ-vYfJ-a'Ta (h]COl-'aL,
Soph. Aj. 634 7TOALaC al-'vYfJ-a xat'Tac, and its cognates at Eur. fr.9Z5a
Snell yEvVwV 'T' al-'vxac, Aesch. Cho. 24 1rP€7TEL 7TapiJe CPOtVLCC' aI-'VYJ-LoLC
(text doubtful, 7Tap"1tc CPOLVtOLC afJ-VYfJ-OLC Stanley, 1rapnCL cpowta£c al-'VYl-'6c
Conington). ,19 Hec. 655-56,
Ret 373-74, 1089, Or. 961-62. For the verb €~aLpELv in the sense 'put an
end to', with an inanimate object, see Phoen. 991 7Ta'Tp6C E~Ei>..OV cp6{3ov,
Med. 904 VELKOC 7Ta'Tp6C EgaLpovl-'€V'T} , PI. Resp. 387n Ka;, 'Tove d8vpJ-Lovc
apa l~aLp~COfJ-EV Ka~ 'TOVC OiK'TOVC TWV E>">"Oytl-'wv avopwv, 387B, Isoc. 12.165
E~aLpEiv ••• 'TaC owcpopac, LS] S.u. III init.20 Comparable to the whole
expression is Pluto Sol. 21.6 al-'VXaC 8~ K07T'TOfJ-'VWV .•• acpEi>..Ev (Solon
"put an end to the cheek-tearing of mourners").
IV
476 '.1. ~ , , , A' A e ' CKEo.raL OE, Kat fJ-"1 'TOLC Ef-LO£C Vt-L0Vf-LEVOC
\ , r ~ , '\ .\ e ' " I\OYOLCLV, wc 0"1 1rOI\W EI\EV Epav EXWV,
../... A ,. '.1. AO 'Q ' C't'PLYWV'T CtfJ-EL't'TI fJ-V OV EK t'paXLOVWV.
'\ \ ", ~ &\ \ \ ' 1\ EI\7TLC yap EC'T Ct7TLC'TOV, "1 1rOl\l\ac 7TOI\ELC
A.I.' " e" r {3 \' 21 CVV"1't' , Ctyovca VI-'OJ! EtC V7TEP ol\ac.
19 For the text of these and the corresponding lines in the antistrophe see G. Zuntz, An
Inquiry into the Transmission of the Plays of Euripides (Cambridge 1965) 65-67; A. M. Dale,
The Lyric Metres of Greek Drama2 (Cambridge 1968) 75 n.1. I have two points to add: (i) the
expression '\t"VKOV I a[/LaToVrt" xpwTa cp6vtov, where X.pwTa is qualified by two epithets, the one
descriptive, the other proleptic ("bloody the white flesh gory"), is so inSipid that one of the
adjectives must be altered. Since CPOVtWL ("with gory nail": so Hec. 657 o{aL/LOV ~vvxa, Hel.
1089 ~vvxa rp6VLOV) requires an improbable correption (Zuntz's objection to "the separation,
excessively wide. of noun and adjective" is unwarranted: see Breitenbach, op.cit. [supra
n.10] 243fT), perhaps we should consider "t"VKctC (Page): see Med. 923, 1148, EI. 1023; (ii) the
credit for first adding OVT EJLOV OVOE C EVOVTa JLU,:>OV. a ey s CIta-
tion of Aesch. Supp. 357 Eg &€A7T7'WV is less apposite than the following
passages: Herael. 148-49 K{VOVVOV J~ &JLYJX&vwv I p{7TTOVTEC ("hazarding
a risk in a desperate situation"), Soph. Traeh. 1109 XEtpJJcOJLat K&K
TWVO€ CI shall destroy her even in my present state of health"), Phil.
91-92 OU yap E~ Evoe 7Toooe I ... XEtpJJe€Tat ('he will not defeat us with
only one leg to stand on"); see also Eur. Med. 459 and Hipp. 705 K&K
TWVO€, Aesch. Ag. 1423 EK TWV 01.LO{wv, Soph. El. 455 J~ {17T€pT€pae x€p6e,
DT 528 J~ OJLJL&TWV op8wv o~ K&~ op8fje cpp€v6c, Traeh. 875 J~ &Ktvr]TOV
77'006c, DC 807 Jg a7TaVTOc. And very similar is Thuc. 5.103.1 JA7TLc •••
\ \ , , '(Hf b d f " Hf TOVC JLEV a7TO 1TEptoVCtac rom a supera un ance 0 resources, rom
26 Nor a superlative {Jpaxv-ra-roc. Sophocles twice has {JpaXLe-roe, the form which presup
poses a comparative {JpaXlwv.
27 The citations by F. H. M. Blaydes, Spicilegium tragicum (Halle 1902) 242, of Hermippus
fr.58 Kock c..>..&.aoc KptOljcETaL.
- aLa aopOC El7Tac 7} A6ywv gvva>..>..aya'ic;
, ,.. , ~ '1:'" 'A. - YEVOLT av KEpOOC' EL 0 apEL't'aToL
A. " A .,' '\ 't'0VOL p,aXaL CTEPVO'TV7TELC T ava 7TTOI\W
I ' A.' '\ '\ , 605 606 KTV7TOL 't'aV7]COVTaL. Tal\aLva. Ttva I\oyov,
, ... A~' ., \'{J 607 TLV av TWVO aLTtaV I\a Otp,t;
'\', \ , I \ ,,, f ""' - al\l\a TOV EVTVXL~ l\ap,7TpOV av TLC aLpoL ANT.
p,o'ipa 7T&'AW' T68E p,OL Op&.coc ap,~L{JalvEL.
599 8&cCEL Murray, Tap&ccEL L. 604 T' ava 7T'T6ALV Murray, y' ava T07TOV 7T&>..W
L, iam T' ava 7TTO'\'W 7T&>..W Markland. 606 T&>"aLva Hermann, c1J T&>"aLva L.
608 EVTvXtq. Markland, EV-rVXij L; alpo, Matthiae, alpij L.
This, for the most part, is Murray's text. I differ from him in two
respects. First, I have followed Dale's29 colome try in 605-07", 615-17.
And, second, I have restored alTlav at 607, where almost all accept
Hermann's alTla, since I cannot believe that the iteration Tlva A6yov
Tlv' has any but an enervating effect in this context.30 The meaning is
«What word of reproach, what blame would I receive 1" For >..6yov
Aa{JE'iv see Heracl. 165-66 KaKov >"6yov I K-n}CrJ 7TPOC aCTwv,31 and for
alTlav Aa{3E'iv see Thuc. 2.18.3 alTtav TE OVK J.AaXtc'T7]v • Apxt8ap,oc
lAa{3Ev a,7T' aVTov, 6.60.1. The two emendations of Murray himself, at
599 and 604, are admirable. At 599 v~' i}7TaTt ••• 8E'ip,a O&CCEt may be
29 "Metrical Analyses of Tragic Choruses," Brcs Suppl. 21 i (1971) 78.
30 aZ-r[av is also retained by W. Headlam, CR 15 (1901) 19. and by Gregoire, and approved
by Zuntz, op.at. (supra n.19) 73.
81 AOYOC cannot by itself mean 'word of reproach', and it is no good supposing that such a
meaning is possible at 565 'lToAAovc Inr€KI.OV (Jp6vov, Bum. 517-19 ;c(J' 01TOV'T6 OEtV6V EO I KlXl
CPPEVWV €1TLeK01Tov I oet piVEtV KIX(J~J.tEVOV, Eur. Ale. 604 '1TP6C €J.tfi. t/Jvxfi.
Oapcoc 7}C'TCu. If awpL{3alvEL is the right reading at 609 (and there is no
good reason to doubt it), then I cannot see how correspondence be
tween that word and the end of line 599 will be achieved more neatly
than it is by this conjecture; and I should hope that nobody will re
turn to Markland's OpaceEt.32 Equally, in 604 ava '1T'T6'\w very simply
restores correspondence with the antistrophe at a point where the
text of the antistrophe seems unimpeachable.
But Murray, while curing one fault of responsion in 599, has left
another fault in the line unmended. In 609 the third dactyl of the
praxillean (J.t0L Opacoc) cannot be answered by a spondee in 599
(X'\wp6v). Further, the break between x'\wp6v and oELJ.ta presents Ha
unique example of word-end after long biceps."33 Most of the solu
tions offered are either improbably violent (oetJ.t1X xo'\~v 'TlXpaCCeL
Camper, oeLJ.ta cpp€vae 'Tapacce, Hermann) or metrically unsafe (oeLJ.t1X
x'\oepov 'TlXpaCCH P '" J.tOL <'TO> [Musurus] or J.tol [Musgrave] or 'Tol J.tE
[BlaydesJ34 Opaeoe alLq>L{3atvet). The simplest solution yet proposed is
Hartung's XAwp6v oetJ.t1X ,.... J.to, (Japeoe aJ.tcpL-, giving a hemiepes and
ithyphallic, as in the previous line, but it inspires little confidence. In
Dale's35 modification of this, X'\WP6V <'TO> oeiJ.tlX, the article is stylistic
ally abhorrent.
HTutissima proinde corrigendi ratio est, uocularum, si opus est,
transpositio," said Porson.36 And here, by shifting the position of
xAwp6v, and changing it in the process to x'\oep6v,37 we may achieve
the praxillean c1Je x'\oep6v J.to, 15CP' .:}'1Tan oetJ.ta (JacceL. The displacement
of an adjective so that it may occupy a position next to its noun, or of a
32 But the corruption may well have arisen by way of 8pacm: cf Hesych. 8paTT€W EVOX·
>"€tV, TapaTT€LV; Suda 8paTT€w' TapaCC£LV.
33 L. P. E. Parker, CQ N.S. 16 (1966) 24.
34 Aduersaria critica in Euripidem (Halle 1901) 539.
36 loc.cit. (supra n.29).
38 Two leading advocates of this method of correction are in the habit of misquoting this
remark with uocabulorum for uocularum: Headlam, CR 16 (1902) 243; G. Thomson, CQ N.S.
15 (1965) 164, and Oresteia (1966) I 71.
37 The two words are confused at [Hom.] Batrach. 162, Philox.Leuc. PMG 836 (b) 17,
Theoc. 13.41. Since they are semantically akin, and since X>..wp6c is very much commoner
than x>"0€p6c, it can be of no consequence that only x>..wp6c is attested in application to such
nouns as 8€tp.a, 8'oc. For discussion of the shades of meaning of XAwp6c see Jebb on Bacchyl.
5.172 (Appendix 473-74), Page on Med. 906.
252 THE SUPPLICES OF EURIPIDES
noun so that it may stand next to its adjective, is a common error, and
illustration exists in abundance.3s The same type of error has also
been detected at 274 OVC IJ7TO TEtXECt Ka8p.EtoLcLV a7Tw'\'Eca KOVpOVC, where
Hermann (in notis mscrpt." Wilamowitz, Analecta Euripidea) and
independently Nauck restore the caesura and the rhythm by writing
ovc IJ7TO Ka8p.EloLCW am:VAEca TElXECt KOVpovc.39 This transposition must
stand or fall with the decision on an equally anomalous hexameter in
the same passage, 282 p.~8' aT&.~ovc T€KVOV EV X{Jov£ K&.8p.ov x&.pp.aTa
{J'YJpwv, where Wilamowitz proposed EV K&.8p.ov x{Jovl. Dale 40 claims
that these are "irregularities which should not be emended away." but
does not explain why not. Fraenke141 cites 274 in defence of Ag. 111
, i:. \ <:- \ \ \ 1 {J 1 " h h '11 C 7TEP.7TEL sVV oOpt Kat XEpL 7TpaKTopL OvpWC 0PVLC, were ot ers WI preler
the colometry of Murray and Page; he also cites Ag. 156 TOL&.8E K&.'\'xac
guv P.Ey&''\'OLC aya{Jotc a7T€K'\'aygEv, where the strong caesura in the
fourth foot makes all the difference, and the textually dubious Eum.
365 ZEUC [yap] alp.ocTay€C (alp.aTocTay€C codd.) agLC)p.LcOV ;{JVOC To8E
Mcxac. As to 282, L. P. E. Parker42 shows, in favour ofWilamowitz's
transposition, that word-end after the spondaic fourth foot is an
equally grave anomaly. I therefore conclude that Hermann's and
Nauck's transposition in 274 is to be accepted. And I shall soon be
suggesting a further transposition for which a similar desire to
simplify the word-order may be given as a cause: see on 699, infra
p.264.
VI
\ \ \ , \ <\1 \-1..' 650 MESS. l\ap.7Tpa P.EV Ct.K'TtC 'YJI\LOV, Kavwv cCt.'I''YJc,
€f3a'\'\E yatav' ap.~;' 8' , HMKTpac 7Tv'\ac
" {J \' , -\f3' ECT'YJV Ea'T'Y]C 7TVpyOV EVCt.YYJ I\a WV.
• - ~ \ -I.. -\ ,- , opw OE 'l'VI\Ct. TpLCt. TptWV CTpCt.TEVP.Ct.TWV·
-1..' \ \ \ , , '" TEVXEC'I'0POV P.EV I\aov EKTEWOVT avw
'T' \"{J' \,J: \' 655 Jcp.'YJVLOV 7TpOC oX OV, WC P.EV 'IV I\oyoc,
" '" -'>' \ \ A ' , aVTOV T avaKTa, 7Tawa KI\ELVOV LYEWC,
38 Headlam, op.cit. (supra n.36) 243-56; J. Jackson. Marginalia Scaenica (Oxford 1955) 228-
231; E. W. Whittle, CIMed 29 (1972) 14.
39 The same adjective has suffered transposition at Soph. Ant. 1115 Ka3fLdac ayaAfLa
vVfL'Pac Nauck. Ka3p.£lac vvp..pac aya).p.a codd., and so too has the adjective XAWPOC at Ar.
Lys. 255 papac XAwpac .p,pwv €AaaC Bentley, .p'pwv papac XAwpac €AaaC codd.
,oop.cit. (supra n.19) 29.
U Aeschylus. Agamemnon II (Oxford 1950) 57-58.
u loc.cit. (supra n.33) 21.
660
665
JAMES DIGGLE
, "'-<:'c' , Kca Tove cvv aVTcp DES tOV TETaYftEVOVC
, \-T.T ",
KEpac, 7Tal\atac n.EKP07TtaC OLKT)Topac,
t ' , ~'n' \ ' \ , ~, aVTOV DE apal\ov ECTOI\LCfLEVOV oOpt
I '" '''A f I ~) J! \. KpT)VYJV 7Tap aVTT)V pEDe L7T1TOTT)V <0 > 0XI\OV
"~ ,~ ,
7TpOC KpaC1TEOOLCL CTpaT01TEOOV TETaYfLEvov,
LCovc apLefLov' apfLaTwv 0' dX~fLaTa
EVEpeE CEftVWV ftVYJfLaTwv ' AfLCPlovoc.
TT , r;:, r;:,' \ ' 'I'' , e ' n.aoftOV DE I\aoc TjCTO 1TpOC E TELXEWV
, " 8 8' "')" , I VEKpOVC 07TLC EV EftEVOC, WV EKELT aywv.
• -~,. - l' '8 \ , L1T7TEVCL 0 L7T7TT)C T)cav av W7TI\LCfLEVOL
, "'" ~ e' oilqTopac (" distinguun
tur Thesei comites et indigenae Cecropii"), where the T€ is anomalously placed: see J. D.
Denniston, Greek Particles (Oxford 1954) 517, Fraenkel, op.cit. (supra nAl) 130-31.
254 THE SUPPLICES OF EURIPIDES
and a left wing, possibly led by Paralos (659-60). This is quite clear:
see Mr Collard, especially 179 n.3 and 181. The messenger, whose
vantage-point is a tower near the Electran gate (651-52), defines the
position of the three detachments by reference to three distinct land
marks. The right wing of the infantry stretches • ICfJ,~V'OV 7TPOC ()X(Jov
(655), "towards the Ismenian hill," whose location is at the southeast
of the walls.46 The left wing is drawn up KP~V1JV 7Tap' av~v "Ap€oc
(660), "alongside the fountain of Ares," which is located at the south
west of the walls.47 The infantry, therefore, invests the whole south
ern circuit of the walls.48
Before proceeding to the third landmark, I will speak of the text
and interpretation of line 659, which I have marked as corrupt. Our
chief uncertainty attaches to the noun llapaAov: is this name intended
to signify the hero Paralos, or the Paraloi, the inhabitants of the
Paralia named after him ? First, let me explain why this name is intro
duced here at all. In old king Cecrops' day Attica was a conglomera-
. f ' \ \ TTl \ - , Q '\ I • 'A \' non 0 towns: E7T£ yap fl.EKP07TOC Ka£ TWV 7TPWTWV ~aClI\EWV TJ TnKTJ EC
£:\ I , \ \ 1,\ , A \ • I , ~ I , l: _
ICITJCEa a£££ KaTa 7TOI\££C CPKE£TO ••• Kat 07TOTE /1.TJ n oE£CE£av OV ~VVTlcav
Q '\ I • \ Q '\ , ''\ '\' , \., ''\ I \ 'Q
~OVI\EVC0/1.EVO' wc TOV ~aClI\Ea al\l\ aVTOt EKaCTOt E7TOl\tTEVOV Kat E~OV-
AEVOVTO ••• €7TE£8~ 8~ @TJCEVC €fJacl>"EVCE ••• (Thuc. 2.15). Cecrops lived
in Athens. But there are texts which hint at an even more specific loca
tion for 'Cecropia': Ion 936-37 KEKpo7Tlac 7TI.Tpac I 7Tp6cfJoppov aVTpov
(Acropolis); Mel.Soph. 10-11 (JvyaTTJp , EpEX(Jl.wc KEKpo7Tlac €7T' avXl.vt I
"Iwv' ETtKTEV Con the side of the Acropolis," unless the traditional
site of Ion's birthplace has been changed); and the Acropolis again
looms large in EI. 1289 ()X(Jov (Valckenaer, OlKOV L) KEKpo7Tlac Ev8al
/1.0va. These are the only tragic passages specific enough to help in the
location of 'Cecropia'; they suggest that Cecrops was imagined as
having his palace on the Acropolis. I do not say that Cecropia was felt
to be synonymous with the Acropolis; but it does seem likely that
Cecropia was felt to be limited to Athens, as centred on the Acropolis.
Now, the Paralia is not a part of Athens. It is a part of Attica, and so it
II See the plan in Schober, Zoc.cit. (supra n.44) 1426. In Mr Collard's diagram (reproduced
below) it has strayed a little too far north. Mr Collard rightly commends Murray's inter
pretation of the words cfJc p.~v tjv '\o')'oc in 655; but the credit for this interpretation should
go to P. P. Dobree, Aduersaria II 81.
&7 The evidence for this location, and against the location given by Pausanias, is decisive:
Wilamowitz, Hermes 26 (1891) 241-42, Pindaros (Berlin 1922) 32 n.1; Schober, Ioc.cit. (supra
n.44) 1426. In Mr Collard's diagram it has strayed a little too far south.
" Mr Collard's diagram unaccountably shows the infantry investing only the southeast.
JAMES DIGGLE 255
would not be under the direct control of Cecrops. Theseus unified
Attica, and Theseus therefore would be (or so Euripides might reason)
the first king who might appropriately be described as leading an
Attic, as opposed to an Athenian, army. It is proper that Theseus him
self should lead the "inhabitants of old Cecropia," while someone
else leads the outsiders. The epithet 7TaAaLiic reinforces the distinction
between the past and present states of Athens-Attica.
At the beginning of 659 aVT6v cannot be tolerated: it is one thing to
sayavT6v T' d.vaKTa, "the commander himself," quite another to say
av'TOV 8~ IIapaAov, "Paralos himself," when this is the first we have
heard of him. Furthermore, aVT6v T' d.vaK'Ta •.• aVTov 8~ IIapaAov •..
KP~V7JV 7Tap' aVT~v is very clumsy writing. It looks as if the second aVTOV
has intruded under the influence of aVT6V overhead at 656 (for a simi
lar intrusion from above see on 1090, infra p.266). We therefore have
a free hand to replace ath6v with anything suitable. We might bring
in the Paraloi simply enough by writing Aaov 8E IIapaAwv (Aaov Jacobs,
also Dobree, Aduersaria II8l, IIapaAwv Kirchhoff). But what one would
expect to be told, after hearing that Theseus and the Cecropids are on
the right, is that Paralos and/or the Paraloi are on the left. Reiske's
Aa£6V for av'Tov, whether interpreted as masculine in agreement with
IIapaAov or as neuter in agreement with the noun in the phrase Klpac
T€Tayp.Evov (mentally supplied), is unsatisfactory for the reasons given
by Mr Collard, p.lS!. I can see no economical way of introducing a
reference to the left-wing position of a plurality of Paraloi; but we
may specify such a position for Paralos by writing Aa£ctJ 8~ IIapaAov
ECTOA£cp.lvov 8opt, where 8opl, as often, has a collective sense. Paralos
may stand as eponymous part for the whole of the folk named after
him. He was a sufficiently celebrated hero, with a portrait in the
Propylaea and a shrine at Piraeus.49 Wilamowitz's claim50 that Para
los would have been named only "wenn er in der Schlacht etwas
thate" is unconvincing. Apart from the deliberate contrast which
Euripides is exploiting between Cecropids and Paraloi and which
alone would justify the introduction of the eponymous hero, the
commander of the cavalry is later given a name (680 Phorbas), but for
no conspicuous achievement.
A final point. I think that we must accept Murray's T€ in place of 8£,
for these reasons: (i) 81 interrupts the essential triple division T€VX€C-
u RE 18 (1949) 1208-09 s.u. PARALOS 3.
60 loc.cit. (supra n.47) 233 n.l.
256 THE SUPPLICES OF EURIPIDES
..L ' '\' ., ~'" \ ., ~'" (") • 0XI\OV ... app,aTwv 0 0X"lp,aTa, 11 In
" ,., t ' , \;" n' \ h fi d .., , aVTOV T avaKTa... aVTOV OE apal\ov t erst TE oes not JOIn aVTOV
avaKTa to what precedes but rather looks forward to a corresponsive
particle; a second TE is needed to correspond to the first, so that the
two wings may be seen to be coordinated as subdivisions of the TEVXEC
q;6poc Aa6c. It may be argued in reply to (i) that the secondary con
trast thus interposed is so straightforward that it does not confuse the
picture, and to (li) that examples of corresponsive TE ••• DE are offered
by Denniston, Particles 513 (the Euripidean instances are an unhappy
and precarious collection). But we ought not to scruple to make a
change which could be documented a thousand times over when the
gain in lucidity is substantial.
We may now proceed to the third landmark, the "sacred monu
ment of Amphion" (663), that is, the tomb of Amphion and Zethus.
It is the location of this monument that I must make the beginning of
my contention. Aeschylus at Sept. 527-28 (quoted by Mr Collard, 180
n.2) speaks of Parthenopaeus as 'TTpocTaxfNvTa Boppatatc 'TTtJAatc I TVp,f30V
KaT' aUTOV LJ tOYEVof}C 'Ap,q;tovoc. The Borraean gate can be situated no
where except on the northern side of the city; Aeschylus says that
Amphion's tomb is situated near that gate. This is significant evidence.
Archaeologists have attempted to identify this TUp,f30C or p,vijp,a with a
hill directly north of the Cadmea. It has been replied that this hill is
too large to permit such an identification, since Pausanias (9.17.4)
describes the monument as xwp,a ou p,Eya. A further piece of evidence
has been adduced: that Pausanias locates the tomb vaguely in the
neighbourhood of the Proitidian gate, which is believed to have been
situated in the northeast of the wall. But this evidence must be treated
with caution. Pausanias locates his buildings and his sites by reference
to three gates only: the Electran, the Neistan (believed to be in the
northwest) and the Proitidian. The sites which he mentions as being
near the Proitidian gate he locates in the vaguest terms, and they are
probably scattered over a wide area. We need not therefore suppose
that there is any contradiction between Aeschylus and Pausanias. The
tomb of Amphion may safely be located in the north or northeast.
But Mr Collard follows neither Aeschylus nor Pausanias. Instead he
locates the tomb due east.51 Why he does so I do not know, since he is
able to offer no evidence in support. He does, indeed, say that Eurip
ides "seems to bring the tomb of Amphion a little nearer the gate of
61 Mesk, loe.cit. (supra n.44), who does not give a plan. seems to require the same location.
JAMES DIGGLE 257
Electra than its northeast location (confirmed by archaeological evi
dence) strictly requires." I think that "a little nearer" is perhaps an
understatement for what is a movement through forty-five degrees,
from northeast to due east. But let that go. I ask only, what is the
archaeological evidence which confirms the location of the tomb in
the northeast? Mr Collard quotes none, and none is quoted by
Schober 1446. The archaeological evidence which Schober does quote
supports the location of the tomb due north of the walls. Further
more, Euripides says that the chariots were disposed "beneath" the
tomb. Mr Collard is obliged to dispose them between the tomb and
the Ismenian hill. He says that "the chariots would seem to be below
it (;VEp()E 663) to an observer looking northward from a tower at the
gate of Electra." If all else were in favour of Mr Collard's interpreta
tion, this further geographical imprecision might be overlooked. As
it is, it must be accounted as another difficulty created by his interpre
tation.
And so the position which we have reached is this: Mr Collard,
following unspecified archaeological evidence which locates the tomb
in the northeast, locates it due east; I, following Aeschylus and the
archaeological evidence presented by Schober 1446, which mayor
may not be relevant (for I have no competence to assess it), locate the
tomb north of the Borraean gate. I shall therefore locate the chariots
at the north of the city in order to see what effect this has on the re
mainder of the narrative.
I now come to the disposition of the cavalry: 660-62 i1T1TOT1JV S'
0XAOV l1Tp6C Kpac1TESOtCt cTpaTo1TESoV TETaYlkEvov' I LCOVC apt ()lkov. The
old interpreters took these words to mean that the cavalry was dis
posed on the edges of the army in two detachments of equal number.
Mr Collard (p.180) rejects this interpretation for the following reason:
"In 680ff. the cavalry come to the immediate aid of the chariots, which
began the battle (674f.) and are stationed ;VEp()E CElkvWV lkV1JlkaTWV
'AlkCP{ovoc (663). The cavalry therefore was on one flank of the army
only."52 It all depends on where you place the chariots. If you place
the chariots where Mr Collard places them, then the only way to
make sense of the succeeding narrative is to place the cavalry where
he places it. If you place the chariots elsewhere, then you can think
again about the disposition of the cavalry. But, before we do think
again, consider this. Euripides says that the cavalry was drawn up
62 Similarly Mesk, loc.cit. (supra n.44) 52.
258 THE SUPPLICES OF EURIPIDES
'1TpOC KpCtc'1T/8oLCL CTPCtTO'1T/8ov. It is essential to Mr Collard's interpreta
tion that this should refer to one edge only; so, for the moment, let us
allow that KpCtc'1T/8oLCL refers to a single edge. Here will be the pro
gress of the messenger's description: (i) the infanty is disposed be
tween the Ismenian hill and the Fountain of Ares, (ii) the cavalry is
disposed "on the edge of the CTPCtT6'1T£8ov." Immediately the reader
asks himself two questions: which edge, left or right? and whose edge,
that of the infantry, or some other edge? To the first question-left or
right ?-the reader must answer "I cannot tell." And if Mr Collard
denies this by reminding us that "in 680ff. the cavalry come to the
immediate aid of the chariots," who according to his arrangement are
on the right, I reply that 680ff are twenty lines away and that we
cannot be kept waiting until the fighting is nearly over before we dis
cover what position the combatants were occupying before the fight
ing began.53 To the second question-whose edge ?-his reaction will,
I think, go somewhat like this: «We are told that the cavalry is drawn
up on the edge of the CTPCtT6'1T£8ov. It must therefore be contiguous
with the infantry, for no other edge has been mentioned, nor do I
know that any other edge is going to be mentioned. I cannot possibly
retain an open mind about the identity of the edge, because I have no
means of divining that in the next line but one the poet is going to
mention the chariots and so create another edge or two." No, it will
not work: if you are to retain your reader's comprehension, you can
not define the position of X by reference to Y, when Y does not yet
exist. Moreover, if Mr Collard were correct in locating the cavalry on
the right edge not of the infantry but of the combined forces of the
infantry and the chariots, then Euripides has chosen a very odd way of
defining the position of the chariots. Why did he locate them beneath
that problematic monument, the tomb of Amphion, when he could
have avoided all ambiguity and imprecision by simply telling us that
they were located between the infantry and the cavalry?
And now let us see what happens when the words '1TpOC KpCtc'1T/8oLC'
CTPCtTO'1T/8ov T£TCtyf.L/vov are interpreted to mean "drawn up on the
edges of the infantry." Everything will be found to fall into place. The
63 In fact, the statement that "the cavalry come to the immediate aid of the chariots" is
a misunderstanding of 680ff. EUripides says that when the Athenian and Theban cavalry
saw that the chariots had begun fighting, they ~"'av a.\~v K&Kp&-rovv ~CCWVTO TE (683). The
words ~"'av aNolv mean the same as cvvijI/Jav p4}(TJV "they joined battle (with each
other)," not "they went to aid (the chariots)." The mistake goes back to Wilamowitz,
Analecta Buripidea 106.
JAMES DIGGLE 259
cavalry, in two detachments of equal number (icovc apdJ}L6v), invests
the east and west ;54 the infantry invests the south; the chariots invest
the north. The city is surrounded, as it was when the Septem attacked
it. And now consider a consequence of that earlier investment. Since
the Septem and their contingents attacked the complete circuit of the
walls, the dead, whose bodies have not been moved from the places
where they fell, must be assumed to be lying at intervals around the
whole circuit. If the Athenians are to invest only the southeast, as Mr
Collard wishes, then, since we are explicitly told that the Thebans
stationed themselves directly opposite the Athenians contingent for
contingent (666-67), Theseus must be severely faulted for his general
ship if he failed to despatch Para los or a handful of the Paraloi to col
lect at least those corpses which lay unguarded around the northern
and western sides. Reason conspires with the indications of the text to
suggest that, when the Thebans took their stand Hin front of the
corpses, for whose possession the battle was being fought" (665), they
stood in front of them all, not half of them.
I give overleaf a sketch of Mr Collard's battle plan and a sketch of
my own. I have tried to reproduce Mr Collard's plan as accurately
as possible; my own sketch of the walls describes a circle, for reasons
which will become clear later.
It remains to consider two possible objections to my arrangement.
First, offence has been taken at the words icovc apdJ}L6v, placed in
apposition to ox'\ov. Mr Collard says on p.180 that they are Hvery
difficult to explain and to substantiate grammatically," and again on
p.182 that they are Hsupposedly constructed 'Ka:ra CVV€Cw' with oX'\ov
in 660, but they are unclear in meaning ... nor does the account else
where state that they [the cavalry] were placed equally on both sides
of the army; I doubt if the Greek will bear that sense." If the words
• , ~'N' I"~ ,~ , d L1T1TOT'Y}V 0 0X"OV 1TpOC KpaC1T€OOLCL CTpaT01T€OOV T€TaYfL€VOV are un er-
stood, as I have argued that they are most naturally understood, to
mean that the i1T1T6T'Y]c ox'\oc was arranged on both edges of the in
fantry, then we already have a mental subdivision of the oX'\oc into
two parts. To append the phrase icovc apLOfL6v to that now plural con-
6& The two wings of the infantry were the regular station for the cavalry in fifth-century
warfare: Thuc. 4.93.4, 99.1, 96.5 (Oelium), 5.67.1-2, 73.1 (Mantinea); A. W. Gomme,
Historical Commentary on Thucydides I (Oxford 1956) 15. For KpaC1T€8a used of the wings of an
army see Xen. Hell. 3.2.16 TOVC 8e 1T€ATaC'T(XC E1Tt TeX KpaC7Tf:8a iKaTtpw9€1I Ka9lcTa.c9a£ Kat Tove
l1T1Ttac, quoted by Markland.
260 THE SUPPLICES OF EURIPIDES
Borraean
Gate
G~
CADMEA ~ 0 ::,~:"
j'Tombof\
~ Amphion f
Gate of fj D "{"II '" "',,,,,,--
Electra I1 Chariots
N
t
Fount
of Ares ~
"\\I\UUlllllllr,,,
~
~ Ismenion § - ~
'I " /11111111111"'\
Fountain X
of Ares
o Athenians
Infantry IZI
Thebans
PLAN OF BATTLE ACCORDING TO COLLARD
Tomb of X
Amphion
Infantry
PLAN OF BATTLE ACCORDING TO DIGGLE
o Athenians
f2LI Thebans
Cavalry
X Hill of
Ismene
JAMES DIGGLE 261
cept creates no difficulty either logical or linguistic.55 Mr Collard's
own solution is to emend line 662 as follows: icwv <8' > apdJp,ov app,a:rwv
[8'] oX~p,CtTCt, "the chariots, equal in number (to the cavalry)." I find
tcwv ap,ep,ov a rather ponderous attribute to be borne by app,cXrwv when
this is dependent on OX~p,ClTCl, and I should rather have expected that
style would have dictated the attachment of such an attributive
phrase not to the genitive but to the governing noun. "Es ist im
Griechischen ein vollig zu Recht bestehender Sprachgebrauch, dass
ein Adjektiv zu dem regierenden Substantiv tritt, auch wenn es dem
Sinne nach eigentlich nur zu einem von jenem abhangigen Genetiv
gehort," says Wilamowitz on HF 468; see also Jebb on Soph. Ant. 794
and Fraenkel on Ag. 504. I will quote only one example of such enal
lage: Soph. Trach. 656 7ToI..VKW7TOV 0XYJp,Ct vCtoc. But it is not a necessary
part of my case to invalidate Mr Collard's conjecture.
The second possible objection to my arrangement is that the
messenger, immediately after describing how first the chariots and
then the cavalry joined battle, proceeds: 684-88 I..EVecWV 8~ TCtVTCt KOV
KI..Vwv ( EKE£ yap r; I Eve' app,ClT' ~ywvt~Ee' o'l T' €7TEp,{3aTCtL) I TaKE£ 7TCtpoVTCt
'\'\' , , . " I' ~ " , ", \ I ' 7TON\Cl 7TT)fLClT , OVK €XW TL 7TPWTOV Et7TW, 7TOT€PCl TT)V €C OVpClVOV KOVLV
7TpoeCtvT€>">"ovcav, we 170>">"0 7TCtp~V, KTI... 56 The messenger was on a tower
near the Electran gate, which, it is believed, was situated at the south
east of the wall.57 He now claims that he was on the spot where the
app,ClTCl and the €7TEfL{3aTClL fought. There is a preliminary problem to
be considered: are the E7TEp,{3aTClL the charioteers or the cavalry? In
spite of 585 7TavT' avSp' Cml..tTTjv app,aTwv T' €7TEp,{3aTTjv, I think it more
likely that they are the cavalry. There is no reason why €7TEp,{3aTT)C
should not mean a cavalryman (so Bacch. 782 i7T7TWV ••• €7TEp,{3aTCtc),
and the sequence of thought favours a mention of the cavalry. In
674-79 the messenger has described the clash of the chariots; in 680-83
he describes the engagement of the cavalry; when he proceeds
I..dccwv S€ TClVTCt KOV KI..VWV-€KE£ yap r; I Eve', we expect him to say not
"where the chariots and the charioteers fought" but "where the
Iili In the same way reo£ is applied to two groups, equally distributed on right and left, at
He!. 1573 llio£ o~ Tolxove of[£Ove "a£Ove T' iem.
6tI Line 686 TaKf' 7Tapov-ra 7ToAAa 7T~/J-aT', OVK EXW was deleted by Herwerden, Mnemosyne
N.S. 5 (1877) 37, not without reason: "uide quam inuenuste interpolator usus sit uocabulis
7Tapov-ra 7ToMcf, sequente tertia post uersu 7ToM~ 7Tapfjv, quamque ridicule is qui 7ToMa
7T~/J-aTa narraturus est, primo loco memoret puluerem. rem minime iucundam esse experti
nouimus. sed quis tamen puluerem serio 7Tij/J-a uocauerit r'
i7 Schober.loc.cit. (supra n.44) 1430. See also Wilamowitz, loc.cit. (supra n.47) 210-11; J. G.
Frazer. Pausanias V (London 1898) 36.
262 THE SUPPLICES OF EURIPIDES
chariots and the cavalry fought."58 But let us leave the answer to that
difficulty in suspense. The charioteers at least cannot be brought any
closer to the Electran gate than northeast. Therefore, if the Electran
gate is situated in the southeast, why does the messenger claim to have
been on the actual spot where the chariot fight took place? Paley
faces the question with blunt common sense: "the Athenian chariot
eers had advanced from below the tomb of Amphion (v.663), i.e. from
near the 1Tv,\at IIpotTlDEc, to the 1Tv,\at "H'\EKTpat (v.651)." Wilamowitz
damned the messenger as a liar, who speaks "mit offenbarer Rlick
beziehung."59 Gregoire damned him as a braggart: "L'Argien un
peu hableur exagere visiblement en decrivant cette melee homerique.
En lui faisant developper la formule courante '\EVCCWV KOO K'\VWV 60
jusqu'a lui faire dire un mensonge evident pour quiconque connaissait
un peu la ville de Thebes, Euripide a voulu nous montrer que les
redts de bataille des temoins civils n'etaient pas toujours plus stirs que
ceux des combattants." But there is a simpler solution: imagine that
the position of the Electran gate, for the purpose of this narrative, is
higher up the eastern wall.
We must consider what sort of picture of Theban topography
Euripides was trying to implant in his listener's mind and what sort
of picture an Athenian mind was capable of apprehending from such
a verbal narration. fJtfJ'\lov T' EXWV £KacToc p,av(J&'vEt TU DEtt&.: the
fJtfJ.\lov with which Mr Collard must equip his spectator is, I fear, a
publication by Bartholomew & Co. The picture at which Euripides
was aiming was a picture of broad outlines: he could not aspire to
anything more precise. His listeners had never seen a ground plan of
Thebes. Few of them knew where the Electran gate lay, and the Is
menian hill, the fountain of Ares and the tomb of Amphion were no
more than names to them. Euripides knew the location of those land
marks well enough; and he was bound to construct a narrative which
was not inconsistent with that location. He chose to mention the Is
menian hill, the fountain of Ares and the tomb of Amphion because
the names are decorative and contribute an air of precision and veri
similitude. He located the Athenian forces by reference to those land
marks, and he placed the Theban forces contingent for contingent
68 I exclude the pOSSibility that the lTT£JLPaTa, are the 1TapmpaTa' mentioned in 677 and 679.
61 lac.cit. (supra n.47) 234.
60 Cf Aesch. PV 266 Kai JL¥ 1TapWlJ y£ KOV A6yovc &AAwv KAvwv, Soph. Tracll. 747, Eur. IT
901. See also Theseus' speech at 846-56.
JAMES DIGGLE 263
facing the Athenians. That the Thebans, thus deployed, are able to
protect the corpses which lie around the circuit of the walls is the
strongest indication to the listener that the whole circuit of the walls
is invested. The listener does not need to know the precise location of
each individual landmark; for, even if he did know it, he would not
have time during the messenger's narrative to piece together the
facts which might complete in his mind a consistent picture of the
topography and of the fighting. The plan of battle which I have
sketched is therefore the plan which I believe Euripides would have
sketched ifhe had been called upon to explain his narrative. He would
have claimed that he had envisaged the three landmarks as occupying
equidistant points around the circumference of the Cad mea and that
he had envisaged the Electran gate as located somewhere on the
eastern circuit of the walls. Not even Meton himself would have found
fault with such an explanation.
VII
\ '/:. " \ K(U cvp.1TaTa<:;aVT€C p.€COV 1TaVTa CTpaTOV
700 " " €K'T€LVOV EK'T€LVOV'TO •••
Of Euripidean lines which have been alleged to lack a caesura few
emerge from scrutiny with their claim untarnished: 303 CcpaAfjc yap EV
TOtJTCfJ p.6vCfJ TaAA' €o cppovwv is emended with certainty by Marchant;
Hec. 355 ytJVCtL~;' 1TapOlvoLc cm6j3A€1TTOC p.l'Ta and El. 546 EKetpaT' ~ 'TfjCOE
CK01TOVC Aaj3wv xOov6c have other faults and are marked as corrupt by
Murray; Andr. 397 (hap 'Tt'Tav'T' ootJPO(.LaL 'Ta 0' €V 7TOcLV requires little
ayxlvoLa to set right; Hei. 86 a'Tap 'Tlc Et 7T60EV; Tlvoc 0' avo&v C€ Xp~;
ought never to have been cited, since 0' avo&v is merely a bad conjec
ture for €gcxvo&v; IA 630 Kcxt O€VPO o~ 7TaTlpa 7Tp6cEL1T€ cov cplAOV is prob
ably not by Euripides, but, if it is, it is easily emended. The only three
serious claimants are Hec. 1159 ylvoLvTo oLaooxa'i:c ap.€lf3ovcaL XEpWV,
fr.495.6 (=Page, GLP 13.31) opOocTaoov A6YXaLc €7TElyov'T€c cp6vov, and
perhaps Bacch. 1125 Aaj30vca 0' wMvaLc (wAEv[a]fCf II) aPLC'TEpaV xlpa,
though this verse is troublesome on other counts. Some6! would
create a caesura in all three places (the papyrus has already done so in
one) by writing -aLc' for aLC. Dodds on Bacch. 1125 calls this "a rather
61 See P. Maas. Greek Metre (Oxford 1962) § 103.
264 THE SUPPLICES OF EURIPIDES
artificial device." Perhaps it is; but it is a well-nigh miraculous coinci
dence that the same three lines in which alone we have any justifica
tion for suspecting that Euripides may have dispensed with the
caesura also happen to contain a word ending in -ate immediately
before the division of the verse.
In 699 only two conjectures are known to me. The conjecture
CVfL7Ta'Tagav'T' €C is ascribed by editors to Blomfie1d, Museum Criticum 1
(1826) 184, but it should more correctly be assigned to Dobree,
Aduersaria II (1831) 81, for Dobree died in 1825. The authors of this
conjecture call it a nominative absolute Ci.e. CVfL7Ta'T~&vroLv 'Toiv
fiaaMoLV," "Creon sc. ac Theseus. Est nominatiuus, quod aiunt, pen
dens"), and the conjecture is accepted by almost everybody. To me it
seems that the economy of this solution is an inadequate recompense
for the hispidity of the construction and style. Murray's reshuffle
c'Tpa'Tov OE: 7T<:£V'Ta cVfJ-7Ta'Tagavr€c fJ-l.cov is a more hopeful approach, but
his introduction of OE: for Kat lessens its probability. I think that trans
position may well be the solution; and perhaps we have here another
example of that scribal habit which I discussed on 599 (supra p.251).
the habit of bringing closer together words in agreement with each
other. If we alter the order of words and add one letter, we shall have
\ '" I C I 62 Kat fJ-€COV 7TaV'Ta cVfJ-7Ta'Tar"av'T€c c'Tpa'Tov.
The meaning will be "they dashed together the whole of the centre
of the army." The centre of the army will be the infantry; and per
haps this is a reasonable way of designating the infantry, since the
conventional placing of the cavalry on the infantry'S wings63 suggests
that the infantry, led as it is by the commander-in-chief, may be
looked upon as the army's centre. But perhaps a further improve
ment is desirable; for I fancy that, if the manuscript had presented
the verse in the form in which I have given it, the verse would have
attracted a further very slight change, that of Kac for Kat:
, I r! 'i: ' Kac fL€COV a7Tavra CVfJ-7Ta'Tar" avr€c C'Tpa'TOv.
The infantry is now designated by c'Tpa'Toc, as it was by c'Tpa'TcYrr€8ov
62 A similar transposition will solve a metrical problem in the tragedian Ezechiel
(B. Snell, Tragicorum Graecorum Fragmenta [Gottingen 1971] I 128, pp.288ff). Line 164 reads
CK£Vr] tKOqWV T£ '7TIfvO' ~v avOpw1ToC 4>lp£t, and Snell suggests CK£Vr] T£ KOCp.oV 0' OAOV. Simpler
is 1T&VTa CK£Vr] KOCp.oV 0', where the scansion of CK£Vr] is defended by 209 ~V71 T£ 1ToMa Ka~
8oJ.£wv &1TOCK£tn7.
63 Supra n.54.
JAMES DIGGLE 265
at 661. For €c ,.dcov used of combatants advandng to meet each other,
see II. 23.814 EC p.'cov eXp.cP0'dpw CVVLT7]V P.€P.tXWT€ P.aX€C(}tXL, Soph. Trach.
513-14 (Heracles and Achelous) oi TOT' eXOi\i\€LC I iCtXV EC picov, Eur.
Phoen. 1361 Ei\()OVT' EC P.ECOV p.€TaLxp.wv. For word-end after initial
dactyl see 93 p:YJT€pa; it is found in plays produced before or about the
same time as our play at Aesch. Ag. 7 eXcTEpac, Cho. 216 Kat TLva, 986
iji\wc, Soph. Aj. 846 iji\L€, Eur. Ale. 802 ov {3{oc, Telephus (c. Austin,
Noua fragmenta Euripidea [Berlin 1968J 102.10) p.7]TEpa, Ereetheus (65.56
Austin) 7TOVTt€.
VIII
811 I ~ I 7TpocaY€T€ < > OVC7TOTP.WV
cwp.a(}' atp.tXTocTtXyfj.
7TpocaY€T€ < > OVC7TOTP.WV '" 798 CT€VtXyp.ov Jj IWTEP€C. Hermann's
7TpocaY€T€ restores responsion, but 7Tpoc aY€T€ does so
more stylishly and shows why the loss occurred. Similarly Ale. 400
{maKovcov aKovcov, Hipp. 1374 7TPOCtX7TO;"i\VTE p.' Oi\AVT€ (7TPOCtX7Toi\AVT'
eX7T(~i\AVT€ Wilamowitz), Hee. 167 eX7TwAECtXT' wAEcaT', Or. 181 OLOLX0p.dP
olxop.E8a, Baeeh. 1065 KaTfjy€V ?}y€V ?}y€v. The same corruption is found
at Med. 1252 KaT{O€T' iO€T€ OL, KaT{O€T€ AVBP, and at Or. 1465
eXvLaX€V iaX€V, where one manuscript has eXv,ax€v alone.64
IX
(I ) , \ '''i\i\ ,- I PHIS EyW ytxp tx OVC €LCOpWV TEKVOVP.€VOVC
I~ , \ 'i' '() " i\i\ I 7Tawwv EpaCT7]C 7] 7T0 <.p T a7TW vJL7]v.
tEl 8' EC T08' 7]1..o0V KeXg€7TEtpa()7]v TEKVWV
l' I () I I I 1090 OWV CTEP€C aL 7TtXT€pa YLyvETaL TEKVWV,
,,, " I~' 'i'i\() '" - I t OVK txv 7TOT EC TOO 7] OV EtC 0 VVV KaKOV·
U ,/..,' \ I , OCTtC ,/"VTEVCtXC KtXL VEaVLav T€KWV
" 'I' _~ - I
tXPLCTOV etTa TOVOE VVV CTEP'CKOP.tXL.
This is the text ofL and Murray; and Murray's are the obeli. I para
phrase Iphis' speech (1080-93). "Why is it not possible to become
64 On the habit of following a compound verb with a simple verb in which the force of
the compound is maintained, see the works cited by Fraenkel, op.cit. (supra 11.41) II 17511.3;
to which may be added C. Watkins, HSCP 71 (1966) 116-19; R. Renehan, Greek Textual
Criticism: A Reader (Harvard 1969) 78-85; E. J. Kenney 011 Lucr. 3.261 (Cambridge 1971).
266 THE SUPPLICES OF EURIPIDES
young again and live one's life afresh? In matters of domestic manage
ment65 if something goes wrong it can be set right by a change of
plan. But mistakes concerning one's life cannot be set right in this life
time. And yet if we had our lives to live again we should avoid making
the same mistakes twice. When I was young I wanted children. But if
I had realised what it means for a father to lose his children, I should have
had none, and so I should have avoided my present plight. For I fathered a
fine son, and now I have lost him."
The words in italics represent the evident sense which must be
borne by the obelized lines. The repeated TEKVWV (1089-90) betrays
corruption, and some have also found offence in the repetition of EC
TOO' tj>..Oov (1089, 1091). Canter proposed to replace TEKVWV by TEKdJV:
«ifI had come to this and had experienced, by having fathered a child,
what it is for a father to lose his children ... " Gregoire claims the
conjecture as his own, which is surprising, since it had received the
approval of Reiske, Heath, Markland, Musgrave, Porson, Hermann,
Dindorf and Paley; and it is also accepted by the most recent editor,
Italie.66 That is an impressive crowd of supporters; but now listen to
Elmsley. "This is an emendation, of which we may say, in the lan
guage of Mr Wakefield, friget, uehementer friget. When two contiguous
verses end with the same word, and there is reason to suspect that
word to be erroneous in one instance, the critic may be allowed to
take a greater latitude of conjectural emendation, than has been
taken in the passage before us ... In our passage the reader is at
liberty to replace the first TEKVWV by any word which appears to him
to improve the sense."67 Availing himself of this liberty Elmsley con
jectured 'mxpoc and TOTE. And the following disyllables have been
65 For III a6p.o,c in 1082 Nauck, BullAcImpSt.Petersburg 9 (1866) 390-91, proposed III 116p.o,c,
which is perhaps an improvement, though not a necessary one.
66 And at 479 Gregoire claims for himself a conjecture by Musgrave which went out of
fashion long ago. While I am defending Canter's conjecture against misappropriation, let me
take from Canter a conjecture to which, through no fault of his own, his name has become
wrongly attached. At 174 Markland writes: "Canterus uertit quas ipsas, quasi legisset /lc
av.rac." The version to which Markland refers is by Portus, not Canter. "Insigni sane fraude
Porti uersionem, obscuri scilicet hominis, et cuius nomen libro praefixum parum gratiae
conciliaturum esset, sub illustriore Canteri nomine uenditauit Paulus Stephanus," Elms
ley, preface to Heracleidae (cf Quarterly Review 7 [1812] 454-55). And yet the conjecture
should not be ascribed to Portus either, for quas ipsas is already the rendering of Melanch
thon, the second edition of whose translation (I have not seen the first) was published at
Frankfurt in 1562. Indeed, Portus' version is merely a revision of Melanchthon's, just as
later Latin versions are revisions of Portus'.
67 Classical Journal 9 (1814) 60.
JAMES DIGGLE 267
offered in emulation: ccupwc Hartung, iyw Hirzel, KCXAWC Holzner,68
luxOdJv Nauck, -ropwc Prinz, a7Tag Fritzsche and later Hartman,69 mxAat
Naber,70 7TaOdJv Paley and later Blaydes.71 Not one of these conjectures
amends the line, for there is still a fault to be found in the words El 0'
€C -roo' ~A80v. "If I had come to this"-come to what? There is nothing
to which -rOOE may refer. Editors appear to suppose that it refers
either to the acquisition of a second youth or to the fathering of
children in the first youth. But it can refer to nothing of the sort. No
good is therefore served by repunctuating with a comma after KagE-
7T€£paOT}v and writing KaKov for rEKVWV in 1091: rEKvwv I ofov crEpECOat
7Tar€pa y{yvErat KaKOv. 72 And it is a desperate man who will consider
replacing El 0' EC -roo' ~Aeov by any of the following proposals: El 0'
, 1>" 'I>" H ' 1>" , 8' 'c'8'I>' H . h' E£CLOWV roo artung, Et 0 1JC 0/LYjV -rE Kas E7TEtpa YjV rooE elmsoet, €t
"" l' '1>"" H' 1 'I>" 1''' ~ 'c '8 ' S h k1 ''''' o EU -roo EyVWV lrze, €t 0 YjC EXW VUV ES E7TELpa YjV -rUXT}C C en ,€t 0
auroc €/LaOov KagE7TELpa8Yjv rvxYjc Wecklein olim,73 El 0' ~V 7TpoOYjAov
Holzner, El 0' au-roc tioYj Nauck, Elo' EV -roo' iioYj Haupt74 (accepted by
Wilamowitz in 1875 and by Wecklein in 1912). I forbear to transcribe
the verses of H. G. Viljoen, Acta Classica 5 (1962) 12-13.
The only fault in the words El 0' EC -roo' ?jA80v is that -rOOE has noth
ing to refer to either before or after it; but if it is to be retained, it
must be made to refer one way or the other. In fact, the problems of
rODE and rEKVWV are not two but one. To prove that, I must set out the
evidence for the various uses of the locution EC roo' €PXO/LaL and cog
nate locutions. These uses may be distributed into four classes: (i) -rOOE"
refers back to a clear conception expressed immediately before:
Soph. Aj. 554-56 EV -r0 cPPOVELV yap /LYjo~v ;;OLC-rOC {Hoc . .. 8-rav 0' ZK1J
7TPOC -roilro (i.e. ro cPPOVELV), OT 1157, OC 548, 981,75 Eur. Ion 1411, Tro.
401, Bacch. 1380, IA 1368; (ii) -rOOE refers forward and is picked up by
an epexegetical clause: Hipp. 1298-99 aM' Ec -roo' ?jAeov, 7TaLooc EKoELgaL
cPPEva I roil coil oLKa{av, HF 1356, Phoen. 1328, Antiope (Page, GLP 10) 7;
(iii) rooE" is qualified by a noun in the genitive, and this phrase refers
68 Studien zu Euripides (Vienna 1895) 83.
69 Mnemosyne N.S. 10 (1882) 313.
70 Mnemosyne N.S. 10 (1882) 155.
71 paley in his school edition of Supp. (Cambridge 1888), a reference lowe to Mr Collard;
Blaydes, loc.cit. (supra n.34) 152.
72 KaKov is Toup's conjecture, the punctuation is Lennep's.
73 JahrbCLPh, Supplbd. 7 (1874) 331.
74 Hermes 8 (1874) 4=Opuscula III (1876) 606.
75 Editors punctuate as if 708' agreed with avocLOV c70/La. They are corrected by Housman,
AJP 13 (1892) 156-57=Classical Papers 196, approved by Jackson, op.cit. (supra n.38) 194.
268 THE SUPPLICES OF EURIPIDES
forwards and is picked up by an epexegetical clause: Med. 56-57 €yc1
yap €C 'TOU'T' €K{Nf3'Y'JIC' &,\Y'Y'J86voc I c;iC'TE, Hipp. 1332, Andr. 170, HF 1281,
1294, El. 918, Phoen. 963, Or. 566; (iv) 'TOSE is qualified by a noun in the
genitive, and this phrase refers backwards: Soph. OT 124-25 7TWC .•• €C
'TOS' av 'To,\p.'Y'Jc ;f3'Y'Jc;, Eur. Ion. 244.
It should be clear, then, that El S' €C 'TOS' ~,\(Jov, if sound, requires
some qualification. And it is likely that this qualification will take the
form of a noun in the genitive, whose place has been usurped by
'TI.KVWV. The choicest noun available is 7T(XOOVC:
EX S' €C 'TOS' ~'\OOV KeXgE7TEtpaO'Y'JV (7TaOovc >.
.. 'lJ " , otOV C'TEPECU(X£ 7Ta'TEpa Y'YVE'Tat 'TEKVWV,
.,,, ,., I~' 'J'\ lJ ., ~..... ,
OVK av 7TO'T EC 'TOO 'Y'JI\UOV ELC 0 VVV KaKOV.
The noun 7TCXOOVC is governed jointly by both €C 'TOS' ~'\Oov and €~E-
7TELpaO'Y'Jv, and the expressions €C 'TOS' -lj,\Oov 7TaOovc and €~E7TELpaO'Y'Jv
7TaOovc may be said to form in combination a single unit which paves
the way for the epexegesis in the following line: "if I had come to
such a length of suffering and had experienced it-what it is like for a
father to lose his children-I should not have come into this my
present misery." For a similar turn of phrase see Med. 34-35 lYVWKE S'
t 1\ .l. A" I l' "'" n n' 'Y'J 'Tal\aLVa cvp.'f'opac V7TO OtOV 7Ta'Tpcpac p.'Y'J a7TOI\EL7TECuaL XUOVOC.
I have chosen the noun 7TaOovc because of its similar application in
the following passages: 11-13 ... 7TaOoc 7TaOOUCat 8EWOV' &p.cP~ yap mJ'\ac I
v /<;: n '" I I I" <;: I • 83 85 ' , n.aop.ov uavoV'TWV E7T'Ta YEvvaLWV 'TEKVWV a7TaWEC ELctV, - 'TO yap
n / I I " I 'A I' I , n '.l. 76 uavOV'TWV'TEKVWV E7TL7TOVOV'Tt Ka'Ta yvvatKaC EC YOOVC 7TaUOC 7TE'f'VKEV,
786-93 " , ,,, <;: A' > , I v , " 'I ".l. ,. t A ayap.ov p. E'Tt OEVP aE' ApOVOC 7Tal\atOC 7Ta'T'Y]p W'f'EI\ ap.Epav
(Porson, ap.I.pa L) K'Tlcat. I 'Tt yap p.' ;SEt (Markland, P.E SEt L) 7Tat8wv j I 'Tl
(Nauck, 'Td L) P.€V yctp iJAm'ov av 7TE7TovO'Vat I 7TaOoc 7TEptcCOV. EX yap.wv
&7TE~VY'Y'JV; I vuv S' Jpw cacPI.C'Ta'TOV I KaKOV. 'TI.KVWV cP''\'Ta'TWV C'TEpEtca
(Markland et fortasse L, C'TEpEtdJat I, C'T'pECOat Blomfield), 1120-22 'Tl
, .. Ay A<;:'" n A l'n 'l: 1 I" 1 n / , yap av p.Et<;,OV 'TOVO E'Tt UV'Y]'TOLC 7TaUOC EC;;EVpOtC 'Y'J 'TEICVa uaVOV'T
€ctS'cOat; .
No further change is needed. The variation El S' €C 'TOS' ~'\OOV 7TaOovc
••• OUK '8fC tlc T63f 8pacovct. Porson supplied £C T63' 8pacovc.
78 Wecklein suggested in 1898 a lacuna after rpvTfvcac, which in 1912 he filled with
; Schenkl suggested a lacuna after 1093; w. Gilbert,
ActaSocPhilLips 6 (1876) 337, proposed OCT" rpvTdcac Ka~ vfavtav [TfKclw] I aptCTOv flTa
nov3 .. (Bothe) vVv CT., or aene rpvTfvcac Kat vfavtav I TfKclw aptCTOv £ITa [Toil3f] vi/v
CT. ; H. G. Viljoen, Acta Classica 5 (1962) 12-13, acnc 8vyaT'pa Kat vfavtav TfKWV I aptCTOV flTa
Toiv3f (Camper on El. 333 [Lugd. Bat. 1831], a reference lowe to Mr Collard) vVv (or Toiv
8voiv) CT. The tautology rpVTfvcac Kat . •• TEKWV is defended by Wilamowitz on HF 1367,
though he proposes to read rpVTEvcac Kat TfKWV VEaVtaV KT'>'. The remarks directed against
Wecklein's and Gilbert's proposals by G. Kiefner, Die Versparung (Wiesbaden 1964) 97,
show that he had no inkling of the problem which they were tackling.
79 I am indebted to Professor Sir Denys Page for invaluable criticism and to Mr Christo
pher Collard for the loan of copies of Ammendola's and Italie's editions and for further
helpful discussion.