Interpolation in Greek Tragedy, II Reeve, Michael D Greek, Roman and Byzantine Studies; Winter 1972; 13, 4; ProQuest pg. 451

Interpolation in Greek Tragedy, II 
Michael D. Reeve 

THE FIRST PART of this trilogy (GRBS 13, 247-65) was a review of 
the only systematic attempt yet made to discredit deletion in 
the textual criticism of tragedy, J. Baumert's dissertation 

ENIOI ABETOYEIN. When he comes to Baumert's Lebenslauf, the 
reader discovers that behind Baumert there looms a more venerable 
figure: "Die vorliegende Arbeit regte Herr Professor Erbse an und 
betreute sie." The tenor of Baumert's dissertation is not unlike that 
of Erbse's articles on the Midiana1 and Iliad 202 and above all on 
Euripides' Phoenissae;3 for out of 181 lines in Phoenissae whose authen­
ticity he examines, Erbse acquits 180.4 

Now that Baumert has shown how unconvincing the prosecution 
can be, Erbse's arguments will be used to show that they are matched 
step for step by the defence. For their true example, however, the 
following pages are indebted not to Baumert but to the scholar whose 
work Erbse is bent on undoing, Eduard Fraenkel. Fraenkel brought 
to the study of Greek tragedy four virtues without which scholars 
might as well not study it at all: a logical mind, a respect for the 
idiom of each author, an awareness both of rhetorical structure and 
of dramatic convention, and the ability to discriminate between 
poetry on the one hand and doggerel and rant on the other. Nowhere 
are these virtues exercised with happier results than in his article on 
Phoenissae.5 It would be regrettable if Erbse's mere authority, unsup­
ported by compelling argument, were allowed to reverse such an 
advance in Euripidean scholarship.6 

1 Hermes 84 (1956) 135-51. 
2 RhM 110 (1967) 1-25. He has curious views on how magic horses cross the sea (pp.I4-15). 
3 Philologus 110 (1966) 1-34. 
'His arguments against the 1BIst (1634) are in fact unsounder than some of his argu­

ments in favour of the other 180 (see infra n.19). 180 might have been 150 ifhe had disclosed 
what he really thinks about 1737-66. 

5 SBMunchen 1963, Heft 1, 3-120. 
6 The approval of Erbse's article expressed by Elizabeth Rawson, GRBS 11 (1970) 109 n.2, 

is only one symptom of her indifference to critical problems, the main blemish in an un­
ambitious but level-headed essay on the theme of the play. Diller's reply to Fraenkel in 
Gnomon 36 (1964) 641-50 is altogether more discriminating than Erbse's. 

451 



452 INTERPOLATION IN GREEK TRAGEDY, II 

References are confined to the articles of Fraenkel and Erbse. which 
give access to the other literature. 

435 

438 

aM' EC C€ 'n:LV€£ TWVS€ SufAVC£C KCXKWV, 

p:ijT€P, Sux,.\,\&gcxccxv 0f-LOY€V€'ic <plAOVC 
...., I \... 1\ 

7TCXVCCX£ 7TOVWV f-L€ KCX£ C€ KCX£ 7TCXCCXV 7TO/UV. 

'\ ' l' r 0' .\\'., • A 
7TCXI\CXL f-L€V OVV Vf-LVTJ €V, CXI\I\ 0f-LWC €PW' 

\ I "0' I TCX XPTJf-LCXT CXV PW7TOtCL TLf-LLWTCXTCX 

~" \' A"O' " UVVCXf-LI.V TE 7TI\El.CTTJV TWV EV CXV PW7TOLC EXEL. 

ayJ, f-LEOfJKW S€VPO f-LVplCXV aywv 

AOYXTJV" 7TEVTJC yap ovSev EVyEvY]C avfJp. 

Fraenkel (pp.25-27) gave three reasons for deleting 438-42: (1) Poly­
nices is not concerned with money; (2) SLaM&gCXCCXV in 436 is picked up 
by SLaMay&C in the first line spoken by the chorus after his speech 
(443), and 8LaMCXKn]C also comes at the end of Iocaste's next speech 
( 468); (3) 7TCXVCCXL 7TOVWV f-LE Kcxl CE KCXt. mxccxv 7TOALV makes a suitable end 
to the speech, not least because Eteocles' final speech ends rfi S' 
EVACXf1El~ ••• 1T'POCEVX0f-LECOCX n]VSE 8LCXC<[JCCXL 1T'OAW (782-83). HMit_ 
angeregt sein mag die Interpolation von 438-442 dadurch, dass Poly­
neikes in dem Zwiegesprach mit seiner Mutter auf ihre Fragen, also 
in ganz anderem Zusammenhange, geantwortet hat (403) €J 1T'paccE' Ta 

.l.!\ ~" ~ ,., t' A d (405) , , \., \' 
<pl.l\WV U OVO€V, TJV TLC OVCTVXTJ un KCXKOV TO f-LTJ EXELV' TO y€VOC 

OVK lf10CKE f-LE" (italics editorial here and hereafter). 
Erbse (pp.3-4) contests (1), the decisive argument,7 by trying to 

show, in opposition to the italicized words in the last sentence, H dass 
der erste Teil der Stichomythie (383-407) die Voraussetzung fUr die 
Aussage 438-42 bildet" (p.4 n.3). HIn dem Gesprach, das er mit seiner 
Mutter fiber die Note der Verbannung ffihrt (383-407), geht es ihm 
nicht urn Sentimentalitaten, sondern urn sehr materielle Dinge. Das 
Leben draussen, so horen wir, enthalt Armut und obendrein Unfrei­
heit, da man auf fremde Unterstfitzung angewiesen ist. Der Zuschauer 
erkennt leicht, dass die nackte Not hinter dem Untemehmen des ver­
triebenen Konigssohnes steht" (p.3). Erbse might in fairness go back 
beyond 383-407 to 358, where Polynices speaks the following words: 

358 avcxYKcxlwc lXEL 

't' 'A" t\ t" '" \ \ \ , 
7TCXTPLOOC €PCXV CX7TCXVTCXC- OC 0 CXI\I\WC l\€yE£ 
\' , ,~, _ , ..... , tI 

I\OYOLCL XCX'P€£, TOV UE VOVV €KHC EXEL. 

7 (2) and (3) may not be decisive, but they are welcome confirmation. Erbse calls (2) "ein 
psychologisches Argument" (p.4). 



MICHAEL D. REEVE 453 

366 7TOAvSaKpvc S' CtrPLKOf1:YJV, 
1 ,~ \ 1\ e \ f3 \ e ~ XPOVLOC LOWV /LE/\a pa Kat W/LOVC EWV 

1 1 Q' l' ",/.. A 1 Q' fiC1' 
YV/LvaCLa U OLCLV EV€,Tpa'f'7JV ~LPK7JC U VOwp' 
l' ,~ 1 , \ Q' I:. 1 1\ 

WV OV ULKatWe a7TE/\avELC ~ EVTJV 7TO/\tV 

vatw, o~' OCCWV vaf.k' txwv oaKpvppoOv. 

Is this not the Polynices who says 

430 \\ , C1" A ~ , M '" 7TOI\I\Ot OE ~avawv Kat VK7JvatWV aKpoL 
I \ \ I , '~I 

7TapELCL, /\V7Tpav XapLV, avaYKataV UE /LOt 
~~, , \ \ , , , I 

OLOOVTEC' €'m yap T7JV E/L7JV c-rpaTEvo/Lat 
1\ Q , ~'" ,.' , 7TO/\LV. UEOVC 0 E7TW/LOC we aKoveLWC 

TOLe rPLATaToLC €KOVCtv ~pa/L7Jv Sopv 

and who ends his plea before Iocaste with the words 

491 /LapTVpac SE TWVOE oat/Lovac KaAW 
f' I ,~,~, " 

wc 7TaVTa 7Tpaccwv cvv otK'[J OLK7JC aTEp 
, ... I~' I ? 

a7TOCTEpOV/LaL 7TaTpLOOC CXVOCLWTCXTCX. 

In elevating 383-407 to the stature of a 'Voraussetzung' for 438-42, 
Erbse not only fails to see that they are an incidental element in the 
structure of the whole scene but also misjudges the function they 
fulfil in themselves, which is not to make the audience think «dass 
die nackte Not hinter dem Unternehmen des vertriebenen Konigs­
sohnes steht" but simply to create sympathy for him (418 10. TtC 

OOTOC; we ap' a8ALOe KaKELvoe ~v; cf 86-87, 163-67, and 301-54, especi­
ally 317-21).8 Once this latter point is recognized (and how many 
people are there who have not recognized it ?), the fate of 438-42 is 
sealed; for their effect on the audience could hardly have differed 
much from the effect attributed by Seneca (Ep. 115.14-15) to a more 
famous passage on the same theme (pecunia, ingens generis humani 
bonum, etc. = Nauck fr.324): cum hi novissimi versus in tragoedia Euripidis 
pronuntiati essent, totus populus ad eiciendum et actorem et carmen con­
surrexit uno impetu. 

753 ,'\' ."" "" ... , CX/\/\ EL/L, 07TWC cxv /L7J KCXTapYW/LEV XEpa' 

Kat /LOt YEVOLT' aOEAcpOv av'T~P'TJ ACXf3ELV 

KCX!. gVC'TCX8EV'TCX OUX /LaX'YJc €AELV Oop!. 
~ Q' " l' \ Q I~ Q , " KTCXVELV V , OC 7J/\VE 7TCXTpwa 7TOpV7JCWV E/L7JV. 

8 In addition, as Mr Barrett points out, Erbse forgets that Polynices only went in need 
7rP~V yap.otc €~P€'iv {Jlop (400). 



454 INTERPOLATION IN GREEK TRAGEDY, II 

If 753 goes, so do 754-56, but the converse does not hold. About 753 
Fraenkel argues as follows (pp.29-30): "Wenn in einer Tragodie eine 
Person aM' E1/Lt sagt, so verHisst sie nach wenigen Versen die Buhne. 
[Footnote: "Ich habe hier nur den Gebrauch von absolutem aM' tILt, 

<aber ich gehe', im Auge, also nicht die Falle, in denen jemand sagt 
<aber ich will da und da hingehen', wie ... "] Das gilt fUr Euripides 
und ebenso fur Aeschylus und Sophokles. Zwei scheinbare Ausnah­
men bestatigen in Wahrheit die Regel ... In beiden Fallen wiirde es 
nach dem aM' Et/Lt zu einem sofortigen Abgang des sprechenden 
kommen, wenn das nicht von einer anderen Person verhindert wiirde. 
Hingegen findet in der Phoenissenszene kein Dazwischentreten statt; 
ohne dass eine andere Person eingreift, redet Eteokles noch eine 
lange Zeit weiter, als hatte er garnicht gesagt aM' Et/Lt." Erbse ignores 
the footnote and appeals to four irrelevant passages9 along with two 
relevant ones cited by Fraenkel, El. 1132 and IT 636. He speaks 
throughout as though the distance of aM' E1/Lt from the end of the 
speech were what mattered and not rather the content of the in­
tervening passage. A mere glance at El. 1132 suffices to show that 
it is no exception to Fraenkel's rule: 

1132 '\\',. 1;.'\' () \. \,J..' al\l\ €t/Lt, 7TataOC apt /LOV wc 'TEI\EC'f'0POV 

() ' () ~ '1;."" 'e' vcw EOLCt· COL 0 O'Tav 7Tpa':,w xapw 
'I;." ,."" , 'I' , () \ ~ 'T'YJVO , €t/L E7T aypov OV 7TOCtC VTJ7TOI\EL 

AT' ,J.. '\ \ , , 1;." " " lYV/L'f'atCtV. al\l\a 'TOVCO OXOVC, 07TaOVEC. 

,J..'" '(}(}"""I;." 'f'a'TvatC ayOV'TEC 7TpOC E • 'YJVtK av OE /LE 

OOJCij'TE (}vdac 'Tfjco' a7T1]MaX(}at (}EO'iC. 
, I;.'~ \ \, 1;.'- , 

7TapEC'TE' O€t yap Kat 7TOCEt OOVvat Xaptv. 

IT 636 is not so straightforward, and Erbse could have made his objec­
tion more convincing if he had confined himself to this one passage: 

636 , \ \ , ,. I;.' '\ , , () -, , 
al\l\ €t/Lt OEI\'TOV 'T EK Eac aVaK'TOpwV 

" \, I;.' , , t A 'Q t OKW' 'TO /LEV'Tot OVC/LEVEC /L'YJ /LOV afJTlC • 
,J..A' ", 'A I;.' -" 'f'V aCCE'T av'TOvc. 7TPOC7TO ot. OEC/LWV a'TEp • 
., " \ -, - ,J.. '\ , tCWC aEI\7T'Ta 'TWV E/LWV 'f'tI\WV 'TtVt 

, ,I. \"A ", \ "\,J.. \ -7TE/L'f'W 7TpOC pyOC. ov /Lal\tC'T EyW 'f'tI\W. 

Kal. OEA'TOC av'T<p {wV'Tac OVC OOKE'i (}avE'iv 

A' ", (I;.' \, A~ 
EYOVC a7TtC'TOVC 'YJoovac a7TaYYE Et. 

8 «Naturlich denkt man auch an Med. 1067 (ill' elp.!. yap 8~ TA.1J/LOVfCTclT'f/V oSOv) ..• " 
Does one also think of the syntax? 



MICHAEL D. REEVE 455 

As Iphigenia turns towards the temple, she orders the attendants to 
keep Orestes and Pylades under guard; the order is made necessary 
by her departure. Likewise at Phoen. 753-83, Erbse might have argued, 
Eteocles' instructions about his family are made necessary by his 
departure, and he naturally completes them, regardless of their 
length, before he leaves the stage altogether. 

This defence, however, has an unfortunate consequence: the longer 
a passage is, the less plausible it sounds as an afterthought. Further­
more, the dispositions Eteocles makes in 757-65 are no afterthought 
but have been in his mind since the beginning of the scene: 

690 I \ , I r '11K I XWPEt CV Kat KOp,t":,E TOV l .. .LEVotK€WC 
TT I • '~\.I..' ,. T I .-
n.pEOJIT , aO€I\,+,ov p''Y}TpOC .LOKacT'Y}C Ep''Y}C, 
\ I I~'" ~ \ \ () , I\EyWV Tao, WC OtKEta Kat KotVa X ovoc 

(}l"w 7TpOC aVTOV cvp,{3a"E'iv {3ovAEvp,aTa, 
\, I \ <;' , 't. \ ~ 7TpW €C p,ax'Y}v TE KaL oopOC Ta~ W p,OI\€LV. 

At that point Creon arrives, and they thrash out between them the 
KOtVa {3ov"Evp,aTa. That Eteocles should then turn on his heel before 
broaching the olKE'ia {3ovAEvp,aTa is at the very least surprising. Again, 
however, a defence is available: in view of Creon's report 07T"OLC 

EAtgEwavTtK' 'ApYEtwV cTpaT6v (711), Eteocles is taking the military 
preparations in hand with due urgency (753 07TWC av p,~ KaTapywp,Ev 

xlpa), so that the oiKE'ia {3ov"Evp,aTa become secondary. 
It could be, then, that 753 is genuine. What about 754-56 ?10 Paley 

felt that they were out of place in a speech so businesslike. Further­
more, Eteocles has already expressed the same wish: 

621 P .... , \' E f "r .... I~ 
OL. 7TOV 7TOTE CT'Y}CT/ 7TPO 7TVpywV; T. we n p, LCTOPE LC TOOE; 

P , 't. - E ' \ -<1"" " OL. aJITLTa~op.at KTEVWV ce. T. Kap.E TOVO epwc eXEt. 

Finally, Iocaste's reaction of horror there (623-24) is the same as the 
messenger's later (1219-22) and the chorus's at a similar point in 
Septem (677-719); Creon ought therefore to be horrified by the prayer 
in 754-56, but he lets it pass. Which, then, is likelier, that the lines 

10 On 756 (= 1376) Erbse derides Valckenaer's observation that "suam ipsius patriam 
vastaturus" would have been more to the point (p.6 n.1) but nevertheless says (p.6) "es 
kann ... kaum belanglos sein, dass der Zug gegen die Heimat als Grund fiir die gewiinschte 
Totung genannt wird." To quote Erbse's criticism of Valckenaer, "der Sinn dessen, was 
iiberliefert ist, bleibt ... ganz W1beachtet." 



456 INTERPOLATION IN GREEK TRAGEDY, II 

were written by the poet or by the man who elsewhere anticipates 
the 'Wechselmord'?l1 

766 

774 

" ~" \ r - , I " () I .J.. EV 0 ECTW 1J1.LLV apyov, EL TL EC't'aTov 
, I ", I " .J.. I oLwvol-'aV'rLc .L ELpECLaC EXEL 't'pacaL, 

TOVO' £K1TVOlc(}aL TavT' ••. 
I, ~ \ \ \ ~" I 'CT I 7TOIlEL oE Kat COt Taw E7TLCK1J7TTW, npEOV· 

111 '" n' I I 1JV7TEp KpaT1JCTJ Tal-'a, OIlVVELKOVC VEKVV 
1-'~7TOTE TacpfjvaL rfjOE €!}1Jf3alq. X(}ovl, 
(}vriCKEW 8€ TOV (}OOfoaVTa, KaV cplAWV TLC V. 

"Was sollte denn den Dichter zu der eindeutigen Formulierung ~V 8' 
£CTtV ~I-'iv apyov veranlassen, wenn er den Sprecher hinterher doch 
noch zu einer vollig unerwarteten und tiberaus gewichtigen Mittei­
lung tibergehen lassen wollte ?" (Fraenkel p.35). Erbse rightly points 
out (pp.7-8) that EV 0' JeT tV ~I-'iv apyov is not the same as EV ou MAEKTaL 
TWV £I-'wv (Hipp. 1021) but means <there is one thing I have not done'; 
yet why should the one thing that Eteocles has not done be sand­
wiched between his other instructions? and why should the one part 
of his speech that looks forward to the next scene be buried away in 
the middle? 

Another argument of Erbse's (p.8) is that 774-77 are the KOLVO: 

f3ovAEVl-'aTa referred to in 692, as 757-65 are the olKEl:a; "die Aufforde­
rung, den Seher anzuhoren (766-73), steht folgerichtig in der Mitte, da 

11 765, 880, lZ63, 1269. Erbse cannot see why it should not be anticipated (p.lZ): "Der von 
der Sage gebotene, bereits von Aischylos behandelte Wechselmord enthielt fur den 
athenischen Zuschauer uberhaupt kein Spannungsmoment. Ihn interessierten nur die 
Voraussetzungen (wie es dazu kam) und die Wirkungen (das Leid, das der grauenhaften 
Tat folgte)." In other words, anything traditional can be mentioned at any point in the 
play, and the characters are allowed to know everything the audience knows. The applica­
tion of this dramaturgically ruinous doctrine to Phoenissae had been forestalled by Fraenkel 
(pAZ, on 880): "Wie vertraut auch immer das athenische Publikum mit der Geschichte 
vom Ende der beiden Bruder gewesen sein mag, der Dramatiker, in dessen Plan die 
spannungsreichen Botenberichte lZI9ff. und 1356fI. liegen, wird sich seine Wirkung nicht 
von vornherein dadurch verderben, dass er den Tod beider Bruder nebenher in Reden 
erwahnen lasst, deren Aufbau und wahrem Gehalt eine solche Erwahnung ganzlich 
widerstrebt." 

The anticipation in 880 is defensible from the lips of Tiresias, but it occurs in a spurious 
passage (see later on 834-1018). On 765, the anticipation that most jars, Erbse writes (p.7): 
"Der Vers 765 ... nimmt zwar die Katastrophe der Bruder voraus ... Aber man versteht 
nicht, inwiefern dieser Gedanke eine Athetese rechtfertigen solI; denn Eteokles geht ja in 
dem ganzen Abschnitt von der Moglichkeit aus, dass er nicht zuruckkehren konne." 
Apparently Erbse sees no difference between one death and two. 



MICHAEL D. REEVE 457 

Eteokles den Inhalt seiner Spruche nicht kennen kann." Even if 706-52 

do not constitute the KOLVa fjovAEvl-'a:rCt., the consultation of Tiresias un­
questionably falls into that category. Why after all does Eteocles want 
him consulted? Not because he may be able to offer some personal 
advice but for the reason given by Creon: 

863 , \~'" '" 8 ~ '8 ' EI-'0L 0 E1TEC'TCCI\K EKI-'CC EW CE Ell 1TCCpCC 

'~A "" , " Tt OpWIITEC CCII 1-'Ct.l\tCTCt. CCfJC(xtI-'EV 1TOl\tV. 

On 775-77 Fraenkel repeated an observation made by Polle, that if 
Eteocles survives (775 ijV1TEP KP(xT~C'[J TeXl-'a) he can see to it himself that 
Polynices is not buried ;12 in a footnote he added references to three 
other places in Euripides where TeXl-'a means no more than JycfJ. Erbse 
replies (p.7): "Die inhaltliche Nuance ist indessen auch in den von 
FRAENKEL genannten Beispielen sichtbar, in unserem Fall ist sie ent­
scheidend: TeXl-'a bezeichnet die Partei, zu der ich gehore." Erbse 
cannot have it both ways: if the victory of TeXl-'a need not include the 
survival of 'me', the defeat of the enemy need not include the death 
of Polynices. 

778 co, I-'€V Ta8' el1TOV' 1TPOC1T()AO£C 8' JI-'oic Myw' 
, ,/...' I I ""'/'" Q' I 

EK't'EpETE TEV')('J 1T(XII01TI\(X T CCI-''t'tt'I\'Y}I-'(xT(x. 

"FRAENKEL mochte ihn [778] ... tilgen, einer selbstgeschaffenen Regel 
folgend, wonach in dieser Formulierung nicht 'irgendwelche unter­
geordnete Diener' angesprochen werden durften. Die vier sprachlich 
verwandten Verse scheinen ihm Recht zu geben ... Aber es begegnen 
zahlreiche Beispiele darur, dass sich ein Darsteller vom Partner ab 
und zu den Dienern hinwendet (vgl. Her. 722-5, Ion 661-7, auch Med. 
819-23), und im vorliegenden Fall besitzt ja gerade die Aktion der 
Wappnung ein solches Gewicht, dass sie durch die feierliche Zasur 
yom Vorangehenden abgehoben und dadurch in ihrer Bedeutung 
kenntlich gemacht werden darf" (Erbse p.8). 

Two of Erbse's passages are nice parallels for the text as Fraenkel 
restores it: 

Her. 722 t A , ~ , , I~'" , , e I 
'Y}I-'EtC, E1TEW'Y} COL TOO ECT EV VI-'LOV, 

• ~ , "e. e' I Ot OEtl-'(xTWV ES"W EV EK1TOPEVC0I-'EV 

\ \ ~'" '" A ,., e '\ 
CVV I-''Y}TpL 1T(xLO(xC. OEVP E1TEC E, 1TPOC1TOI\Ot, 

f '1\ \' \' '" 1 
WC (XV CXOI\'Y}V I\VCWI-'EV Ct.CI-'EVOt 1TOVWV. 

111 "Das gleiche Argument wurde schon 1890 von POLLE vorgebracht. FRAENKEL zitiert es 
a.O. 34,4 und bezeichnet es als amiisant" (Erbse p.7 n.3). Erbse might accord Fraenkel the 



458 INTERPOLATION IN GREEK TRAGEDY, II 

Med. 819 JI " 1'" \1 l.TW· TT£PLCCOl. TTaVT£C ovv p,£cqJ l\oyOl.. 
-'.",,., \' Y' '1' 
W\I\ £ta XWp£L Kat. KOp,Lr:, .lacova· 
" ,~, \, \ I (j 

£C TTaVTa yap o7J COL Ta TTLCTa Xpwp,£ a. 

In the third passage a simple experiment can be performed: 

\ , ,~,,. ~ , ~" ~ " 
Cot p,£v TaO HTTOV· Op,WLCLV 0 £p,al.C l\£yW· 

-, "~ (j ~, '" " '(j ~, CLYCXT . OI\HC £ 0 TJV £/-'7J /-,CX TJ ocx/-,ap. . ~ ~ , - ~ ,~ , , ,~ 

V/-,l.V o£ Cl.yav, op,Wl.o£C, I\£yw TaO£, 
"'(j' " 'to' ", 7J avaTOV £l.TTOVcaLCL TTpOC OCX/-,CXpT £p'7JV. 

Which pair of lines stands at Ion 666-67 ? 
"Der Eteokles des Euripides gab seinen Dienem ohne pompose 

Einleitung nur den Befehl ~KcP'P£T£ T£VX7J 7TaVOTTAa T' &/-'cPl.f1A~p.aTa 
KTA., offenbar in beabsichtigtem Anschluss an die Worte, mit denen 
der Eteokles des Aeschylus seine tragischste Rede schliesst (Sept. 
675f.), cP€P' wc TaXOC KV7Jp,tSaC, alxp,fjc Kat TT€TPWV TTpof3A~p,aTa» (Fraen­
kel p.36). Erbse could have inferred from that whether a 'feierliche 
Zasur' is required to confer 'Gewicht' on €KcP€P£T£ T£VX7J. 

Of the Tiresias episode (834-1018), Erbse says (p.9): "Die entschei­
dende Partie der Szene ist die erste Teiresiasrede." Unless Erbse 
attaches some private sense to the word 'entscheidende', this state­
ment is flagrantly false: the whole scene revolves round 911-14: 

" ~ , (j'/" , ,-.~ , aKOV£ 07J vvv £C-rCXTWV £/-,WV ooov· 
,/..L.l: M ' ,~~ ~ , ., I C-r""bal. £VOLK£CX TOVO£ OH C VTT£P TTaTpac 
, ~~" ~ , " ,\ _., ~ 

cov TTaLO , €TT£to7J 77JV TVX7JV aVTOC KCXI\£tC. 

Everything before these lines leads up to them, and everything after 
them is consequent upon them. According to Erbse, however, "die 
Worte des Teiresias [865-95] sind ... ein Bindeglied, eine etwas 
steife, aber sehr geschickte Gelenkkomposition, welche die beiden 
Hauptthemata des Dramas vereinigt, namlich das Schicksal des Oidi­
pus und seiner Sohne mit dem des Menoikeus. In der Konzeption des 
Dichters ist dieses die Voraussetzung von jenem. In seiner Darstellung 
aber wird das Opfer des Menoikeus nur erforderlich, wei! es dem 
Seher nicht gelungen ist, die hadernden Labdakiden zu versohnen." 

courtesy of reading him carefully and reporting him carefully: "Amiisant ist die Bemer­
kung von F. Polle ... : '774ff. Diese Verse sind auffallig, denn wenn Eteokles siegt, kann er 
ja selbst dafiir sorgen, dass Polyneikes unbegraben bleibt. Das ist ein Fehler des Dichters'." 



MICHAEL D. REEVE 459 

Why the sacrifice of Menoeceus is necessary the poet explains with 
perfect clarity, and his explanation differs from Erbse's: 

931 "'~ ''''B\' 'f''''' • \ OE t TOVOE aMXI-LCXLC ov 0paKWV 0 ')I'YJYEVYJC 
" LI' ,,, , 

EYEVETO LpK'fJC val-UXTWV E7TLCK07TOC 
J.. ' J..' to ~ I.' ~ \ C'f'aY€VTa 'f'0VtOV aL!-,a YTJ oovvaL xoac 

KaSf-Lov, 7TaAaLWV "ApEoc EK f-LTJvLf-LaTwv, 

8c YTJYEVEL SpaKovn TLf-LWpEL cp6vov' 
, ~ I.' ~, , B' "A Kat TaVTa OpWVTEC cVf-Lf-Laxov KTTJCEC PTJ. 

B \ "'", ~ ", 0' " X wv ° aVTL Kap7TOV Kap7TOV aVTL atf-LaTOC 
f' ,,, \ 'Q Q ' <li:" ~ 

atf-L av l\atJrJ tJpOTEWV, E~ ET EVf-LEVYJ 

y~v, .;; Trof}, TJf-LLV xaAKo7T1JATJKa cTaXvv 

E7TapTwv avfjKEv. EK YEVOVC Se SEL BavELv 
-1."" I.' , ", J.. ~ 

TOVO oc 0paKOVTOC YEVVOC EK7TE'f'VKE TraLC. 
, "" , B ,\.', • ~ \ ' .,. ~ ~ , cv 0 EV ao TJf-LLV I\OL7TOC EL ",,7TapTWV YEVOVC 

aKEpawc, EK TE f-LTJTPOC apcEvwv T' a7To, 
., ~I.' Aft ,.,., 

Ot COt TE TraLOEC. Lf-LOVOC !-'EV ovv ya!-'ot 

J.. ", "" "B c'f'ayac a7TELpyovC • ov yap ECTLV TJ EOC' 
'f' I.' \ ~ \ ~I.'" , '\ 

OVTOC OE 7TWI\OC TTJO aVEtf-LEVOC 7TOI\Et 

(J ' I .... ' I tf avwv 7TaTpc.pav yaLav eKCWCELEV avo 

For once, however, Erbse's error is understandable. The fact is that the 
text offers two conflicting explanations of why Menoeceus must be 
sacrificed, one in 865-95 and the other in 931-48. Fraenkel deleted the 
greater part of the first (869-80, 886-90), but not 868 E~ ov 'TEKVW(JTj 

Aawc f3lCf (JEWV, on which it hinges. His deletion of 869-80 is confirmed 
by a stylistic detail that he misses: f-LEAEOC (869) is an intruder from 
lyric.13 If 868 is deleted as well, one of the conflicting explanations has 
vanished and Tiresias' first speech has become a vague and sinister 
foretaste of his second. It is no accident that the same language recurs: 
VOCEL ••• ';;SE y~ 7TaAaL (867), and Menoeceus must ccpaY€VTa cp6VLOV 

f' ~ I.' ~ \ T.T'I.' \ ~ "A' , (J' 1." aLf-La YrJ OovvaL x oac naof-Lov, 7Tal\atWV PEOC EK f-LTJVLf-LaTWV ••• X WV 0 

••• , E~ET' EVf-LEvf} y~v (933-34, 937-39). The sins of Laius and his 

descendants are beside the point. 

13 In Euripidean lyrics it occurs something like 70 times. In trimeters it occurs four times 
in OJ p.eAttoc (Or. 90, 447, 67l, 1029) and once in an etymology of Meleager's name (fr.517). 
Mr Barrett draws attention to Soph. fr.443.5, where Pearson thought that" <p.)ttAmt indi­
cates lyrics"; he himself reads the lines as trimeters and does not find the space too wide for 
nAttat. 



460 INTERPOLATION IN GREEK TRAGEDY, II 

1009 '\\' 1" "'I!:' ',\I!: " alV\ E tjLL Kat C'Tac E~ E1Ta ~ EWV aKpwv 

ccpa~ac EjLaVTOV CTJKOV EC jLE,\ajLfJaOij 

~ , "0'" 'I!:' OpaKOV'TOC, EV 0 jLavnc E~ TJ'Y'YJcaTO, 

',\ 0 I A " A' E EV EpWCW yaLav. ELpTJTaL 0YOC. 

1013 I 1:" 0 I I:'A ", ',\ CTELXW OE avaTOV OwpOV OVK aLCXpov 1T0 EL 

~ I I ~ , I~" \ \..!I!: 0 I owcwv, VOCOV OE TTJVO a1TalV\~W X ova. 
, , A fl'" ., 1:'1 I EL yap a wv EKaCTOC OTL OvvaLTO TLC 

, ~ 1\0 A , '.I.. I 
XPTJC'TOV oLEI\ OL 'TOV'TO Kac KOLVOV 'f'EPOL 

I~ A '" • 1\ .\ 1 
1Ta'TpWL. KaKWv av at. 1TOI\ELC El\aCCOVWv 

, \\', """ 1TELpWjLEvat. TO 1\0L1TOV EVTVXOLEV avo 

"Wenn jemand soeben (1009ff.) gesagt hat aM' EljLL Kat • •• ccpa~ac 

EjLaVTOV ••• lAwOEpcfJCW yaLav und dies en Satz mit einem dixi, EtpTJTa& 

A6yOC, abgeschlossen hat, so kann er wirklich nicht, jedenfalls nicht, 
wenn er das Geschopf eines die Redetechnik beherrschenden Dichters 
ist, wieder ganz von neuem einsetzen ... " (Fraenkel p.52). Erbse 
musters two exceptions, Med. 355-56 (del. Nauck and Fraenkel 
amongst others) and Soph. Phil. 389-90: 

A I ,\ 1\ A' 1:" 'A II:' A OYOC EI\EKTat. 1Tac. 0 0 TpELOac cTVyWV 
, I 0" I '0 A " .I.. 1\ EjLOt 0jLOLWC Kat. EOLC ELTJ 'f't.I\oc. 

Anyone who has been listening since 341 can see that AOYOC ,\EAEK'TaL 

1Tac in 389 means 'my story is finished': 

341 P - \ \ J.. ' H. TOLyapovv TO COY 'f'pacov 
1"0 '\ A'''' , ~Q av LC 1TCXI\LV JLOt 1TpayjL • 01TWC C EVVfJpLCCXV. 

That a brief imprecation should be appended to the story affords no 
support for Phoen. 1013-18, which are merely a sententious regurgita­
tion of 997-1012.14 

Erbse passes on to 1584, the beginning of the last episode: 

1584 C " '''~ \ 1 0' .., , .I.. R. 0 t.KTWV JLEV TJoTJ I\'YJYE ,WC wpa Ta'f'ov 

/Lvr]/LTJv TtOEcOaL 

Fraenkel (pp.74-76) deduced from a comparison of this opening with 
a number of similar ones that Creon has just entered the stage. <fAus 
der Hille des von FRAENKEL nicht beriicksichtigten Materials seien 
hier wenigstens zwei Stell en genannt, I.A. 496-7 (Menelaos gegen Ende 

14 Erbse is happy to translate 1015-17 'if everyone were to take what he has of value and 
go through it, and place it at the disposal of his country .. .' (p.15). 



MICHAEL D. REEVE 461 

des Gespraches mit Agamemnon ... ) und Hipp. 473-4 (die Amme 
schliesst ihre Rede an Phaedra mahnend ab)" (Erbse p.16). Ion 650, which 
Erbse relegates to a footnote, is more to his purpose: after a long 
speech by Ion and two lines from the chorus, Xuthus takes over with 
1TCxi)ca, '\6ywv TWVO', Nevertheless, since the majority of such openings 
occur when a character enters the stage, and many of them after a 
lyric ode, there must be a strong presumption at Phoen. 1584-85 that 
Creon too has just entered the stage. On its own this presumption 
would count for nothing, but in the presence of other anomalies it 
cannot be waved aside. 

The most important of the other anomalies here is that Creon re­
mains on the stage in silence from 1356 to 1584, even though the news 
that the messenger brings (1356-1424, 1427-79) is addressed to him 
(1335-55). Erbse is not at a loss (p.17): HEine vollkommene Parallele zu 
dieser Anordnung befindet sich im letzten Epeisodion der 'Andro­
mache': Peleus fragt den Boten, wie sein Enkel getotet wurde (1083-
4), und dieser berichtet (1085 bis 1165). Sobald er geendet hat, setzen 
die Anapaste der Frauen ein (1166fI.: Kat fL~V 00' avag if 07] cPopao7]v I 
LlEAcPtooe €K Y7]e oWfLa 7TEAa~Et ••. ), und erst danach beginnt die 
erschiitternde Klage des Peleus. Die Szene der 'Phoenissen' wird aller­
dings anders fortgefuhrt: die Totenklage ist den nachsten Angehorigen 
des untergehenden Geschlechtes vorbehalten. Kreons Beteiligung 
wiirde die Wirkung der leidenschaftlichen Szene nur abschwachen 
und sein eigenes Verhalten ins Zwielicht rUcken." As Erbse has been 
kind enough to demolish his own case, Di Benedetto's arguments for 
removing Creon from 1307-55 (Fraenkel pp.71-74) need not be re­
peated. 

One difficulty in Creon's speech, however, does deserve mention. 

1310 JI I ~, " , , '" '\ 
O£fLOL, T£ opaew; 7TOTEP EfLaVTOV 7] 7TO/\£V 

, ~ , " 'c:. " ''/'' eTEVW oaKpveae, 7]V 7TEPt<; EXEt VE'I'0e 

~ .. ~"A' ., TOCOVTOV weTE ot XEpOVToe LEVat; 

Why does a cloud of doom surround the city after Menoeceus' death? 
Since the moment when Tiresias declared that his death would save 
the city (834-959, especially 918, 947-52), no-one has doubted his 
words-not Creon (915-90), not Menoeceus himself (977-1012), not 
the chorus (1058-59), not the messenger (1092), not Iocaste (1206). 
According to Erbse (p.19), however, Creon Hkennt doch am Beginn 
dieses Epeisodions den Ausgang des Kampfes noch nicht, weiss vor 



462 INTERPOLATION IN GREEK TRAGEDY, IT 

aHem nichts vom thebanischen Sieg! Er kann also auch den Tod des 
Sohnes noch nicht, wie Teiresias verheissen hatte, als Vorbedingung 
der Rettung ansehen ... Erst am Ende des folgenden Botenberichtes 
wird ihm bestatigt, dass die Leiche des Menoikeus nicht Symbol des 
aHgemeinen Unterganges, sondern ein Zeichen der Erlosung sei. "15 

Besides dreaming up this idea, Erbse deals with TE in 1313 (p.18) in a 
way that looks responsible enough but can only astound anyone who 
comes to him from Fraenkel (pp.78-79). 

To return to 1584. 

1585 Twv8E 8', Ol8t7TOV, A6ywv 
" [ , \ ~ ~ ~,,~ , 

CXKOVCOV· cxpxac TTJCOE yr}C EOWKE fLOL 
'E \ , ~, , .I. \ ~ ~ \ 

TEOKI\E"f}C 7TaLC coc, yafLwv -pEpvac O£Oovc 

A ft , \ , 'A ' '() LfLOVL KOpTJC TE I\EKTpOV vnyoV1Jc CE EV. 
, .,.. ", ,~ _ 'A" 

OVK OVV C Eacw T"f}VOE yr}V OLKE LV En· 

.I.~ \].,. ", , " 
ca-pwc yap Et7TE.l ELpECLac OV fLTJ 7TOTE 

~ ,~ ~ ,~ l' -1.1: '\ 
COV T"f}VOE yr}V OLKOVVTOC EV 7Tpu..:, ELV 7TOI\LV. 

«If the dowry was given to Haemon, how is it that Creon becomes 
sovereign 1" (Pearson, cited by Fraenkel p.87 on behalf of the deletion 
indicated above). Erbse replies (p.20) that «Kreon Vormund (KVpLOC) 

Antigones wird" and «der zuktinftige Gemahl das gesamte Vermogen 
der Erbtochter (J7TtKATJPOC) , etwa gar das Konigsamt, schon zu Leb­
zeiten Kreons erhalt. Die Herrscherwiirde geht damit also auf den 
gesetzlichen Vormund (Kreon) tiber ... Bei Billigung dieser Dber­
legungen versteht man, weshalb Oidipus und Antigone Kreon als 
neuen Herrscher widerspruchslos anerkennen." On the contrary, 
Creon is taking a liberty, because Eteoc1es merely confirmed -rTjv 86cLV 

-rTjv 7Tp6c(}E, which naturally did not include the kingdom: 

757 , ~, ,~ \.1..:; 'A' ~, ~ 
yafLovc 0 aOEI\-P'IC VTLYOV1JC 7TaLOOC TE cov 

A ft , , ~, , \ .I. \ ~ 
LfLOVOC, Eav TL T"f}C TVX'YJC EyW C-pCXI\W, 

\ \ '\ () ' ''' , "'", COL XPTJ fLEI\EC aL· T"f}V OOCLV 0 EXEYYVOV 
\ , () ~ ~ "'1: ,~ , ~ 

T"f}V 7TpOC E 7TOLW VVV E7T E5000LC EfLCXLC. 

Since Creon appeals only to Eteoc1es' instructions, Attic law is irrele­
vant except in so far as they comply with it. If it were relevant beyond 

10 The passage quoted is actually part of his answer to a point of less consequence raised 
by the deletion of 1312 (which he rightly opposes), but it would obviously have been his 
answer to the important point if he had answered it. 



MICHAEL D. REEVE 463 

that, the first thing that would have to be explained is why Oedipus is 
not Antigone's KVpWC. The less said about Creon's authority, therefore, 
the better. For the audience it is surely enough that he xcfJpac A'A€t7TTa~ 

ILoVVOC c/>vA.a~ (OT 1418). 

Fraenkel maintains that the language of 1595-1614 is unworthy 
of Euripides, and so Erbse does his best to demonstrate that it is 
not. 

1597 \ \ , .J..~ \, ~ ,~ 

Ka~ 7TpLV €C .,...WC p:YJTPOC €K YOVTJC P.Ofl€LV 
" 'A'\ \ A" "0 ' ayovov 7TOflflWV au.t> P. E €C7TLCEV 
.J..' , 0 ' l' " " .,...ov€a Y€l'€C aL 7TaTpoc W Taflac €yw. 
, \~" I "to t I \ €7TEL 0 €Y€Vop.7Jl', av LC 0 C7TEtpaC 7Ta77Jp 

, '" .J..' KTELV€~ P.€ vop.LCac 7TOIl€P.WV 7T€.,...VK€VaL. 

«Man wird aoOLc wohl als ein ungeschicktes Fiillsel ansehen miissen" 
(Fraenkel p.90). "Der Text deutet die Moglichkeit eines Anschlages 
des Sohnes auf den Vater an. Der aber setzt sich zur Wehr (1600-1). 
Das umstrittene aO()Lc ist also recht sinnvoll (,seinerseits'). Man ver­
gleiche Ion 312 ... , Hik. 743, Soph. O.T. 1403 und andere Stellen" 
(Erbse p.21). In all these passages aoOLc means not 'for his part' but 'in 
turn'. 

1602 7T'P.7TEL 8' P.E 
p.acTov 7ToOovvTa 07Jpdv aOAwv {3opav-
00 ccp~6p.€c(ja - TapTapov yap wc!>€A.€V 
"(j ~ T7 (j \ '''{3 , €Il €LV fl.L aLpwv HC a vcca Xacp.aTa, 
., , , ~ " . "'\ ~ ,~ , oc p. OV OLWIl€C aflfla OOVfl€VCaL T€ P.OL 

Salp.wv g8WK€ lloAv{3ol' &p.c/>' 8€C7T077Jl'. 

"00 is awkward, as the antecedent ... has to be inferred from 7T'P.7TEt 
••. {3opav" (Pearson). Erbse (p.22) translates 'he sent me to the place 
where-I was rescued': 'dorthin, wo-ich gerettet wurde'. The dash 
is informative. 

"Ob ... die Verse 1606-7 wirklich korrupt sind, darf man bezwei­
feln. Es gibt zwar keine exakten Parallelen, aber doch sehr ahnliche 
Konstruktionen, vgl. besonders E 267f. (eyc1 8' K€ TOL Xap{,.wl' p.{al' 
(mAOT€paWl' I ScfJcw ()7Tvd.p.€l'aL)" (Erbse p.22). Iliad 14.267 would be a 
parallel for 8alp.wv p.0£ gDWK€ llOAvf30V D€C7TOT7JV &P.c/>LDOVA€VCaL, if that 
were what the interpolator had written. 

"No mention is elsewhere made of Oedipus having ever been the 
slave of Poly bus" (Paley). "Euripides gebraucht DOVA€VW in der weiten 



464 INTERPOLATION IN GREEK TRAGEDY, II 

Bedeutung <abhangig sein' [here follows a string of references, every 
one to cSov.\€v€£V in the usual sense <be a slave to' or <serve']. Gewiss ist 
Oidipus kein Sklave im gewohnlichen Sinn, aber er war als Kind und 
Jiingling in seiner Korinther Zeit von seinem Pflegevater abhangig­
die Verse 43-4 (Ka~ .\afJwv dX~f£aTa I IIo.\vfJ<p TpOCP€1. cSlcSWCLV) bringen 
zum Ausdruck, wie er seine Stellung angesehen hat-und empfindet, 
seit sich herausgestellt hat, dass Polybos nicht sein leiblicher Vater 
war, jenes Verhaltnis als cSov.\€la" (Erbse p.23). Children dependent on 
their parents would be surprised to hear that they are the cSo17.\o£ of 
their parents and their Tpocp€l.a are a token of their cSov.\€la ;16 and as 
for the piece of psychology with which Erbse winds up his argument, 
it rests on a sense of the word cSov.\€la that he has just gone out of his 
way to dispose of. 

The daggers are Erbse's. "FRAENKEL (a.O. 91) hat im Anschluss an 
Paley ... mit Recht an die bekannte Regel erinnert, dass ~f£&. (im Gegen­
satz zu Teif.L&.) nieht elidiert werden darf. Der Wortlaut des Verses 1613 
ist also verdorben, indessen (wenn ich recht sehe) nicht eindeutig 
heilbar. Das darf aber nieht zu der Folgerung verfiihren, der Satz 
1612-4 konne nicht im 5. Jh. geschrieben worden sein (vgl. FRAENKEL 
a.O. 92)" (Erbse p.23 n.4). Whether the rule about Ef.L&.17 was bekannt 
to Erbse or anyone else before Fraenkel drew attention to it is some­
thing that may fairly be doubted. What of Erbse's daggers? WeT€ is 
necessary as a link between OU and 1614, and Ele is necessary alongside 
EC T' Ef.LWV 'TTaWWV fJlov. Is it the embarrassing Ef£', then, that is "nicht 
eindeutig heilbar" ? Erbse might have taken the bull by the horns and 
proposed T&'cS'. 

Back to dramatic technique. "Ware diese Rede [1595-1614] nicht, 
dann wiirde Oidipus, sobald er die Ausweisung erfahren hat, FRAEN­
KELS Wunsch gemass nur fragen: 'Was solI ich nun tun ... , wer wird 
mich fiihren ... , wovon solI ich leben?' Der Zuschauer aber konnte 
kaum ein Mitleiden mit solchen Besorgnissen des Greises haben" 

18 Let Erbse make what he will of Supp. 361-62 Tote T£KOVCt yap I 8t$CT1]VOC ClCTtC p.~ &v7t-

8oVA£V£t T'KVWV. 

17 It applies equally to ca on its own, but ca after Ta can be elided even if other words 
intervene (Soph. El. 1499, Men. Epitr. 43). 



MICHAEL D. REEVE 465 

(Erbse p.24). Antigone might have had something to say about 
that.18 

Fraenkel then tries to show that the question of Polynices' burial, 
first raised by Eteocles in 774-77 and argued out by Creon and An­
tigone in 1627-82, has been superimposed on the original under the in­
fluence of Antigone. Erbse is at pains to play down both the similarity 
of the situation and the significance of the verbal echoes. 

1627 II€KPWII 8E TWII8€ TOll fLEII Ec 80fLoVC Xp€WII 
"I:' If I I:' 1:"" I 1\ 

7J07J KOfLLo,,€W, TOIlO€ 0 , OC 1T€PCWII 1TO/UIl 

1T(xTp£8a CVII &'\'\OLC ~AO€, IIoAvlI€tKovC IIlKVJI 

• f3 '\ '''0 ~ 1:" ., "i: 0 I €K al\€T a a1TTOIl T1}CO 0PWII E~W X OIlOC. 

K7Jpvg£TaL 8e 1Ta.Ct Ka8fLE{OLC T&8e 
"" \ I 1:" " ',,1,.' \ ~ oc all II€KPOIl TOllo 7J KaTaCT€'f'wlI al\ctJ 
'" ~ \1 0 ' '\\Ii: 7J "In KaI\V1TTWII allaTOIl allTal\l\a~ €TaL • 

1634 • ~ 1:" " \ ",,1,.' ~ f3 ' €av 0 aKl\aVTOV, aTa'f'0v, OLWVOLC opav. 

cv 8' €KAL1Tovca TPL1TTVXOVC 0P'r}vovc II€KPWV 

18 Diller, op.cit. (n.6) 645-46, is of the opinion that if 1595-1614 are removed the resultant 
use of (iltv in 1615 is unidiomatic: HAuch wenn das Wort eine Rede einieitet, hat der Spre­
chende vorher schon Wiinsche oder Einwande zu der Lage vorgebracht, in der er sich be­
findet und zu der er nun, von einem neuen Standpunkt aus, sich wieder aussert"; "Immer 
wird in diesen mit eUv beginnenden Reden etwas registriert und beiseite gelegt, das, so 
lange es nicht gesagt war, den Sprechenden von dem nun folgenden Schritt im Handeln 
oder Reden zuriickhielt." 

As regards the first statement, would it really have been unidiomatic for Iphigenia to 
begin her speech at IT 342 with eUv if Euripides had passed straight from 239 to 260 and 
omitted the intervening dialogue between Iphigenia and the messenger? To rephrase 
slightly Diller's second statement, the function of eltv in the passages he examines is to 
indicate acceptance, often resigned acceptance, of new information or a new situation, and 
at the same time to introduce a reflexion or decision occasioned by that new information 
or situation. It is just this function that it fulfils at Phoen. 1615 if 1595-1614 are removed. For 
a similar use in stichomythia cf El. 618. 

Supp. 1094, which Diller considers his strongest card, differs significantly from Phoen. 
1615. The loss of his son moves Iphis to lament the irremediable mistake he made in 
begetting him. 'If I had known what it is like for a father to lose his children, my present 
plight would never have befallen me', Gcne "'VTEvcae Ka~ l'favlav TfKWV I /lP'CTOI', Eha Toi/8e 
vVI' eT"plcKop.a, (1092-93). Having summed up his plight in these last three words, he 
proceeds "ltv· Tl 8~ xp~ TOil Tw..al7TWpOv P.f 8pav; (1094), where 8~ in effect means ToiJ8e 
eTEp,cKOP.fiVOV. At Phoen. 1615, on the other hand, the one thing that Oedipus' recital of 
his woes does not include is a summary of his present plight, and yet he proceeds £IiI" 
T' 8pa.cw 8ijO' 0 8vc8alp.wII 'yw; What then is the function of afjTa? 

Even if there were any substance in Diller's objection, it would not follow that Fraenkel 
was wrong to ascribe 1595-1614 to an interpolator. The interpolator might have sub­
stituted lines of his own for lines written by Euripides (or eUv for alai, as Mr Barrett points 
out). 



466 INTERPOLATION IN GREEK TRAGEDY, II 

, Y "A ' ~, " KO/U~E ca1J77]v, v-nyoV7J, OOP.WV ECW, 
\ .Q "''''' f'1 Kat 7TapUEVEVOV rYJV E7TtOVCav TJp.Epav 

, ".$; \ , A" , P.EVOVC • EV :I CE I\.EK7'POV tP.OVOC P.EVEt. 

HDas Verbot gilt ab sofort. Es untersagt also symbolische Bestattun­
gen, bevor der Tote (nach Abzug der Feinde) fortgetragen werden 
kann, es untersagt aber ebenso, dass er in die Heimat zuruckgeholt 
und dann erst bestattet werde.19 Wie man sieht, weichen die sach­
lichen Voraussetzungen von denen der <Antigone' erheblich ab" 
(Erbse p.26). The only significant difference is the one intended by 
Euripides and ignored by Erbse, that the body has been brought into 
the dty (1480-1581). Erbse also ignores Fraenkel's observation (pp. 
101-02) that VEKPWV 8€ 7'WV8E 7'bV P.€V ••• , 7'bv8E 8' ... (1627-28) is in­
compatible with the presence on the stage of three bodies (1482, 1502, 
1635, 1693-1702). 

Ka7'aC7'I.cpwv (1632), which Fraenkel suspects of being borrowed in­
opportunely from Ant. 431 7'OV vI.KVV C7'I.CPEt, Hheisst <nach ritueller 
Vorschrift schmiicken', wie aus Herakl. 124 und I.A. 905 besonders 
deutlich hervorgeht (vgl. auch PEARSON z.St.)" (Erbse p.26 n.2). Once 
more Fraenkel has disposed of this interpretation in advance (pp.l02-
103), but it would not have occurred to him that Held. 124 and IA 905 
are relevant: in the former passage the object of the verb is an altar, in 
the latter Iphigenia. So long as Erbse understands the word, the pro­
priety of the operation it denotes is apparently of no interest to him. 

HIn 1637 sind die Formen €7Ttovcav und Elnovcav, palaographisch 
gleichwertig, also beide iiberliefert. Dass €7TtOVcav vermutlich nicht 
<unmetrisch' ... ist, hat C. PRATO ••. gezeigt" (Erbse p.26 n.2). 
Where metre is concerned some Italians can prove anything, but it is 
unusual to see their results acclaimed in Germany. 

19 Erbse's reason (p.26) for suspecting 1634, where the syntax alone breaks down, is 
quite baffling: "Es mag zwar ublich sein, einer positiven Formulierung die negative folgen 
zu lassen; hier aber wiirde diese Aussage nur fUr die Zeit gelten, in welcher sich der Leich­
nam noch auf thebanischem Hoheitsgebiet befindet." The argument seems to run as fol­
lows: the body is "noch auf thebanischem Hoheitsgebiet" (even the interpolator knows 
that it is on the stage); it must be neither symbolically buried where it is nor brought into 
the city for burial (a distinction not drawn in the text); only where it is can it be eaten by 
birds; therefore Creon cannot have added 'it must be left where it is to be eaten by birds'. 
The falSity of the first premise, the fanciful nature of the second, and the sheer irrationality 
of the conclusion, leave the argument rather too weak to prove the line spurious. 

In case anyone is bewildered by the passage, here is a paraphrase of it: the body of Poly­
nices (which is now on the stage) is to be thrown outside Theban territory and neither 
crowned nor covered with earth but left unburied for the birds to eat. 



MICHAEL D. REEVE 467 

In 1644, where Antigone asks a question that Creon does not answer, 
Erbse (p.26) commends" die leichte und wohl evidente Anderung" of 
,,.{ to el, which "befreit den Text von der ersten (scheinbaren) Frage 
(1644): Tl 7'()v8' iJ{3pl'etc 7TCl.Tlp' cbocTlAAwv x8ovoc"; but he does not ex­
plain the force of this el, and it does not emerge from the translation 
he quotes: 'Jagst freve1nd du den Vater aus dem Land, warum I 
Schaffst du, zum Hohn dem Toten, uns ein neu Gesetz?' 

Fraenkel (pp.106-07) regards it as the most surprising feature of the 
passage that nothing comes of Antigone's avowal eyw C¢€ Oar/J(n, lCav 

a7T€wl7Ty/ 7ToAtC (1657), from which she gradually retreats until she is 
forced to change the subject (1673). Erbse replies (p.27): "Antigone 
r~agiert eindeutig" to 1656, "wie es einer Antigone20 zukommt: €yw 
C¢€ 8aljJw (1657). Aber sie muss . .. Schritt fur Schritt !{uruckweichen: sie 
darf den Toten nicht waschen, darf seine Wunden nicht verbinden, 
ihm nicht einmal einen Abschiedskuss reichen. Der Dichter ... lasst 
sie jedoch die Aussichtslosigkeit ihres Vorhabens nicht eingestehen: 
nirgends nimmt die das Versprechen des Verses 1657 (eyw C¢€ 8aljJw, KaV 
a7T€wl7TTJ 7TOAtC) !{uruck." How do the two italicized assertions square 
with each other ?21 

How is Antigone to bury Polynices (1657) if she accompanies her 
father into exile (1679)? The difficulty was noticed by a scholiast:22 
ou T'YJP€'i TO CVfI-¢WVOV, aAAa KCl.Ta TO Op&fl-CI. U7TOTl8€TCl.t· 7TWC yap OaljJ€t 
nOAvvelK7JV 'AvTLYOV7J cvfl-¢evyovcCI. TCp 7TCl.Tpl; According to Erbse (p.29), 
these sentences "sind ... kein Produkt interpretatorischer Klugheit; 
denn sie set!{en ganz naiv voraus, dass ein dramaturgisches Genie wie 
Euripides die Teile seines Werkes nicht habe aufeinander abstimmen 
konnen." They setzen nothing whatever voraus: they merely state in 
an unexceptionably sober way that the text contains an inconsistency. 
It is Erbse himself who setzt something gan!{ naiv voraus, namely the 
third proposition in the following syllogism: Euripides was a genius; 
a genius is never guilty of an inconsistency; Euripides wrote this pas­
sage; therefore there is no inconsistency in this passage. Unlike some 
of Erbse's others, this syllogism is at least valid; but a more alarming 
method of arriving at the conclusion can scarcely be imagined. 

20 The phrase 'einer Antigone' is a concise illustration of the preconceptions with which 
Erbse approaches this part of the play. 

21 The object of Erbse's factitious distinction is presumably to deflect suspicion from 
1743-46, where Antigone is still proposing to bury Polynices. 

22 Whether or not the author of such a comment was Byzantine is of purely historical 
interest. 



468 

1703 

INTERPOLATION IN GREEK TRAGEDY, IT 

OED. 
ANT. 

OED. 
ANT. 

OED. 

A ,.,. A A 1:.' , 
VVV XfY'lCf.LOC, W '1T(xt, OS'OV '1T£PCXW£TCXt. 

co .... '\ \'.,., ....,.... I 
o '1TO tOC; CX/\/\ 7] '1TpOC KCXKO tC £p£tC KCXKCX; 

, ~ 'Af)' f) A " \ , £V TCXLC 7]VCXLC KCXT CXV£LV f.L CXI\Wf.L£VOV. 

A' , 'A f)!<;i: '" 'I:. '1TOV; TLC C£ '1TVpyOC T woc '1TpOCO£<:, £TCXL; 

• \ V \ ' "', f)" , f) A L£pOC nOI\WVOC, OWf.LCX L'1T'1TtOV £OV. 

In his defence of these lines, an addition suggested by ~C, Erbse makes 
light of 1687 and 1734-36, which are inconsistent with them: 

1687 
1734 

OED. '1T£CWV cmov f.LOL f.Lo'ipcx K£lcof.LcxL '1TEScp, 
A ,'" ", '\ 'f) NT. TCXO£ C £'1T£f.L£v£ f.L£I\£CX '1TCX £CX 

.1..'''' ,"'" , .,.,vycxocx '1TCXTPLOOC CX'1TO y£v0f.L£vov, 

1" f) A 

W '1TCXT£P, CXV£ LV '1TOV. 

In his translation of the latter passage he misses out the very word that 
introduces the inconsistency, '1TOV. At no point does he acknowledge 
that a reception anywhere (1706 '1TPOCSEg£TCXL) , let alone in Athens, 
must be regarded as a mitigation of Oedipus' misery (1687 '1T£CWV, 1690 
TWV CWV KCXKWV, 1710 cPvy«v 'TaAcxLvcx.v, 1723 SVc'TUx£c'Ta'Tcxc cPV')'ac, 1734 
f.L'A£cx 7Taf)ECX). 23 

Whether or not Erbse thinks 1737~6 are interpolated is not clear 
(see p.30 n.2), but he treats them (p.32) in the same way as many 
passages that he defends: H Halten WiT uns nUT an den uberlieferten 
Wortlaut, dann ergibt sich: Antigone versichert dem Vater in den 
Versen 1737-9 zum letztenmal, dass sie mit ihm in die Verbannung 
ziehen werde. Wenig spater gibt sie die Absicht kund, den Bruder 
trotz Kreons Verbot zu bestatten. Wenn Oidipus sie daraufhin auf­
fordert, bei den Freundinnen, bei den Heiligtumern oder im Gebirge 
Zuflucht zu suchen, kann er doch wohl nur an die Zeit nach Erfiillung 

18 Diller, op.cit. (n.6) 646, is unhappy about the function of aM' Eta if 1703-07 are deleted: 
"Diese Interjektion steht gewohnlich innerhalb einer Rede; sie besagt, dass der Sprechende 
einen Gedanken nunmehr in die Tat umsetzen will ... Wenn wir Ph. 990 der von F. 
gebilligten Personenverteilung der Oberlieferung folgen, steht dort aM' Era zwar bei Per­
sonenwechsel, aber innerhalb der bisherigen Handlungslinie. Dasselbe gilt rur Hel. 1429. 
Wenn Oedipus 1708 Antigone mit aM' Ela von der Leiche des Polyneikes wegrufen wollte, 
so ware das nach dem Sprachgebrauch wohl einfach ungehorig." Here too, however, the 
speaker wishes to translate an idea into action (Antigone's idea of accompanying him into 
exile, which he has implicitly accepted at 1693 'ITpoeayay' ~ JLE, JLTfTpoe we y,avew cl8EV). 

and since the idea was discussed only 15 lines previously and the intervening passage is a 
preliminary to carrying it out, his aM' Eta most definitely stands "innerhalb der bisherigen 
Handlungslinie." 



MICHAEL D. REEVE 469 

des ersten Auftrages, also an die Zeit nach seinem Tode, denken. 
Diese uberaus naheliegende Deutung . .. " 

Two years before he turned to Phoenissae, Erbse took issue24 with 
an earlier article by Fraenkel on Septem.25 Fortunately most of the 
interpolations he discusses are Single lines of no significance outside 
their context, and his misinterpretation of the play26 neither stands 
nor falls with them; but it may be as well to show that he has not 
undermined Fraenkel's arguments at any point. 

470 \ - "'..I.. \, \..1..' Kat TepOE ..,.,WTL 7TEfL7TE TOV ..,.,EPEYYVOV 
'\ " A ~ ~'\ r ' 7TOI\EWC a7TEtpYELV 7"7'}COE OOVI\LOV ,=,vyOV. 

E [ ' ,,,"<:.' , ~ \ , <:.' , ] 
T. 7TE/L7TOL/L av 1)01) TOVOE, cvv TVX'D oe Tep 

\~, I " .... " Kat 01) 7Te7Te/L7TTaL KOfL7TOV ev XePOLV exwv 

Meyapevc 

Del. C.G.,27 Fraenkel. "Auch in der Tragodie find en sich Beispiele fur 
Verwendung der Junktur Kat. O~ im Satzinneren (vgl. Soph. O.c. 173; 
DENNISTON, Gr. Part. 2251" (Erbse p.9). In the section that Erbse refers 
to, Denniston presents examples of Kat 07j 'in response to a definite 
command, often with a word of the command echoed', e.g. Ale. 
1117-18 T6A/La 7TpoTE'ivaL xe'ipa .•• - Kat O~ 7TpoTEtVW, Aves 175 {lM~ov 
KaTw. - Kai o.ry (lM7TW, and three of the five cited by Fraenkel from 
Aeschylus himself, PV74-75 cKlA1) o€ KtPKWCOV {ltc!. - Kat O~ 7Tl7TpaKTaL 

TO-3pyOV, Supp. 437-38 TaOe eppacaL ••• - Kat 8~ 7TleppaC/LaL, 506-07 
KAaoovc iJ-€V aVTov Ae'i7Te ••• - Kat. O~ cepe Ae{7TW. What is in dispute is 
not whether Kat S7j can come in the middle of a sentence28 but 
whether it is a coincidence that the deletion of 472 brings the passage 
into line with all these others. A coincidence it might be if the trans­
mitted text gave better sense than 'I will now send this man, and 
rather fortunately he has already been sent'.29 

2' Hermes 92 (1964) 1-22. 
26 SBMunchen 1957, Heft 3, 3-61= Kleine Beitriige I (Rome 1964) 273-328. 
26 See infra n.29. 
21 See R. D. Dawe, Repertory of Conjectures on Aeschylus (Leyden 1965). 
280C 173 is the worst example Erbse could have chosen to prove that it can, because 

!/Javw "a~ 8~ behaves in all other respects as though it were "a~ 1)~ !/Javw. The examples he 
wants are on p.252 of Denniston in section (iv), e.g. Hdt. 4.102.1 "TWV 8f "a~ 8~ oE /Ja"M€c 
CVV€ABOV"Tc;C ifJov)""OOV"To. It is in this section that Denniston ought to have put Sept. 472 if he 
trusted the text. 

29 That the seven commanders have already been sent to their several gates, 'already' 
meaning while the chorus were singing 287-368, is a theory to which Erbse devotes the 
first nine pages of his article. The history of the theory is reviewed by von Fritz, Antike und 



470 

451 

457 

INTERPOLATION IN GREEK TRAGEDY, IT 

\ I , "\ \ "\ \ , 1\ '\ I /\t:y a/\/\ov a/\/\a£c €V 7TV/\a£c €L/\TJXOTa. 

E [ \ \ \ , ~() \ I \ 1\ ] T. KaL p.TJV TOV €V'T€V €V /\aXOVTa 7TpOC 7TV/\a£C 
\ 'e I \'E I \ 
/\€s W' 7pLT<[J yap T€OK/\<[J... 

Del. H. Wolf. Fraenkel objects amongst other things to the vagueness 
of 7TpbC 7TvAaLc, but according to Erbse (p.ll) it is "nicht unbestimmter 

I · V 376" " 1\" ~\ '\" S' 'h a s 1m ers : wc T €V 7Tv/\aLC €KaCTOC €£/\TJX€V 7Ta/\ov. Ince ow at 
the gates' quite plainly means 'at which gate', the resemblance be­
tween the two uses is purely superficial. 

510 

514 

e I '" \ \ I ", I", 
SVVOLC€TOV O€ 7TO/\€P.LOVC €7T aC7TWWV 

() 1« \ \ I "'.J...~'" 
€OVC 0 p.€V yap 7TVP7TVOOV .1 V'f'WV €X€L, 

«y: Q I '" \ Z \ \' " '''-7T€pfJL<fJ O€ €VC 7TaTTJp €7T aC7TLOOC 

cTaoaioc ~cTaL, OL~ X€PbC fJEAoc r/>Mywv. 

[ " 1'..- z~ , , ] KOV7TW TLC € LO€ "Iva 7TOV VLKWP.€VOV. 

TO£cfo€ P./VTO£ 7Tpocr/>lA€£a oaLp.6vwv· 
''''' , ., I C'~' t , 

7TpOC TWV KpaTOVV'TWV €CP.€V, 01. 0 TJCCWp.€VWV. 

, \ ..- \ .!.e "'" "'''-'' , €LKOC O€ 7TPUAi;€LV avopac wo aV'T£CTaTac, 

'z' T:J..~ , , €t €VC y€ V'f'W KapT€pWT€POC p.axrr 

• y 7T€pfJl<[J T€ 7TPbC A6yov TOV c~p.aToc 
, , ,on. Z \ "'''!~ , 

CWTTJP Y€VO£T av €VC €7T aC7TWOC 'TUXWV, 

Erbse accepts this order from Fraenkel but refuses to part with 514 
(del. Hermann). His elaborate paraphrase of the passage, which leads 
to the predictable conclusion "dass der Gedanke des Verses 514 in 
dieser Deduktion nicht fehlen darf" (p.13), never comes into contact 
with Fraenkel's argument: "ein solcher Gedanke konnte an sich sehr 
wohl den Abschluss des Vorhergehenden bilden, er darf sich aber 

. nach den Regeln aeschyleischer Gedankenfiihrung nicht zwischen 
511-3 (6 p.~v y~p ••• Tvr/>wv' €X€£, 'Y7T€pfJl<[J o~ Z€VC ••• ~cTaL ••• r/>Mywv) 

und 515 (7"oLcfo€ P.EV7"OL KTA.) eindrangen."3o It may be added that the 
invincibility of Zeus is less to the point than his victory over Typhon. 

moderne TraglJdie (Berlin 1962) 193-226, whose verdict Erbse is appealing against. Erbse's 
only contribution is the argument that the future tenses in 408 and 621 do not refer beyond 
the speech (better, do not refer beyond themselves: cf. Eur. Andr. 323, Or. 560, Pind. O/. 
9.25,10.79, and Bundy, Studia Pinaarica I [U.Cal. Publ. in CI.Phil. 18.1,1962] 21-22), which is 
possible but leaves von Fritz's case untouched (cf. Erbse p.5: "natiirlich konnen diese Verb­
formen nicht eine der Vergangenheit angehorende Handlung bezeichnen"). 

80 Fraenkel "muss allerdings zugeben, dass die tadellosen Worte nicht von einem 
Interpolator verfertigt sein konnen" (Erbse p.12). Fraenkel merely says that the line 
"macht ... gar nicht den Eindruck. als sei er das Flickwerk eines Bearbeiters." The notion 
that interpolators cannot write presentable lines is an assumption of Erbse's own similar to 



553 

559 

MICHAEL D. REEVE 

" ~, , ~~'" \' "A ,<:; ECTtV OE Kat TCfJO OV I\EyHC TOV pKaoa 
'" " \ ~, ~ "" \ ~ I 

aVYJp aKofLTTOC, Xetp 0 0PCf 'TO opaCtfLOV, 

"AK'TWP, aOeAcP0C 'TOU 7T£lpoc AeAeYfLl.vov 
t\ '" \_ , I " 
OC OVK EaCH Yl\wccav epYfLa'TWV a'Tep 

" \~., '\<:; I , eLCW TTVI\WV peovcav al\OaLVetV KaKa, 
,~" ~.,. il , 'il' '" I ovo Etcal-'Et'f'aL UYJPOC EXULCTOV oaKOC 

[ ' , .J..' \" ., 1<:; ] eLKW 'f'EpOVTa TTOI\EI-'Lac ETT aCTTLOOC 

;gwfJev eZcw. 'T0 cPl.pOV'Tt fLl.fLrfe'TaL 

TTVKVOU KPO'T7JCfLOU 'Tvyxavovc' tmo TT'TOAw. 

558 oaKOc V: oaKOVC cett. 

471 

Weil emended oaKOVC to oaKOC and deleted 559, but Wilamowitz, who 
adopted both changes, was the first editor to punctuate after eZcw 

rather than after 559. Erbse makes the suggestion, entirely ground­
less, that Wilamowitz would never have accepted the deletion of 559 
but for the mistaken punctuation that led Weil to it. Some other 
passage, therefore, will have to serve as a lesson about the persistence 
of error.31 

others that he makes elsewhere, e.g. "schwerlich kann diese Beziehung von einem Inter­
polator geschaffen oder auch nur kenntlich gemacht worden sein" (p.6 of the other article, 
on Phoen. 1376=756 of all things), "ich wiirde diesen Kontrast als sehr wirkungsvoll be­
zeichnen. K6nnte ihn ein Interpolator geschaffen ... oder gar durch Zusatze in den Text 
gebracht haben 7" (ibid. p.28, on Phoen. 1680). 

"Die beliebte Annahme, ein 'Leser' habe die Worte aus einem anderen Drama des 
Aischylos heriibergenommen ('an den Rand geschrieben' ... ), soil das Ratsel [of how the 
line got in] 16sen" (p.12), with the footnote: "Die Annahme ist natiirlich moglich, aber sie 
erklart so lange nichts, als nicht bewiesen wird, dass die spatere Uberlieferung notwendig 
ausschliesslich auf dieses interpolierte Exemplar zuruckgeht. Wenigstens seit einem hal ben 
Jahrhundert (seit Bekanntwerden der Papyri) weiss man, dass ein solcher, fur die Glaub­
wurdigkeit der oben referierten Annahme unabdingbarer Beweis nicht gefiihrt werden 
kann." The argument of these two sentences seems to be: (1) before a line can be ascribed 
to an interpolator, it must be proved that all the mediaeval manuscripts go backto an inter­
polated copy; (2) the papyri show that in antiquity the tradition had several branches; 
therefore (3) even if someone interpolated a line in one copy, it would not spread through 
the tradition; but (4) 514 is found in all the mediaeval manuscripts; therefore (5) 514 is not 
an interpolation. This chain of argument prompts three questions: (1) could it be proved, 
except of course by the discovery of interpolations, that all the mediaeval manuscripts go 
back to an interpolated copy 7 (2) do the papyri show that no interpolation established it­
self before the tradition branched? (3) was the tradition free from contamination right up 
to the mediaeval period? An authority on Uberlieferungsgeschichte like Erbse really ought to 
have worked out other answers to these questions than the ones implied by his argument. 

31 "Wer sich den Gang der Diskussion urn diese Verse vor Augen halt, wird sich die 
Frage, weshalb wir den ganzen Rattenschwanz von Vorurteilen und ubereilten Folge­
rungen nicht langst losgeworden sind, sicherlich selbst beantworten konnen" (p.1S). 



472 INTERPOLATION IN GREEK TRAGEDY, II 

Erbse (p.15) would like to know how Fraenkel, who follows Wila­
mowitz, Hvom neutralen Objekt DaKOC zum femininen Subjekt des 
Verbs p.ll4ETat zu gelangen gedenkt." Kuhner-Gerth § 359.3 (a) will 
tell him. 

Fraenkel's observation that the Sphinx and not the ElKciJV must be 
the subject of p.lp.!frETat makes no impression on Erbse. A further ob­
servation may be offered, that yAwccav €pyp.aTwv aTEp and OTJpoc €XOlc­
TOV DaKovc ElKW ¢lpov-ra32 are unequal yokefellows. 

617 aM' olDEV WC C¢E Xp~ TEAEV7"'ijcat p.aXJJ, 
El KafJ7Toc ECTat OEc¢aTotct Ao~tov. 
[cPtAEi DE ctyav ~ MyEtV Ta Kalpta.] 

Del. Pearson. Erbse (p.16) accuses Fraenkel of suppressing any refer­
ence to Eum. 614ff, a passage that proves nothing except what no one 
doubts, namely that Apollo's prophecies come true. Erbse himself, in 
propounding for 618 the novel interpretation 'if he will have fruit by 
means of Apollo's oracles',33 suppresses any reference to another 
passage in Eumenides that Fraenkel does cite, 713-14 KaywYE XPTJcp.ovc 

TOVC €p.ovc TE Kat. Llloc I Tap{3Eiv KEAEVW P.TJD' aKafJ7TWTOVC KTlca£. If a 
line cannot be defended except by such means, Erbse's 'unprejudiced 
reader' may be forgiven if he concludes that it is indefensible. 

649 ,... ' 't' ,\ 'i: I TOtav-r EKEtVWV Ecn Ta!:,EVPTJp.aTa· 
[ \ t" " "'t' ~ L) I I t' A ] CV 0 aVTOc 7J07J YVwut Twa 7TEP.7TE£V OOKEt. 

I' W " ~ \ .... ~ I WC OV7TOT Wopt T4JOE K7JpVKEvp.aTWV 
I ./~ \ t" " ~ Q ,~/, P.EP.'f'U· CV 0 aVTOC yvwut VaVKI\7JpEW 7TO/UV. 

Del. Paley. Erbse's defence of the line (pp.17-19) is built on the hasty 
assertion, for which he cites Tucker's note (Hin K7JpvKEvp.a lies the 
sense of conveying a message entrusted"), that K7JpVKEvp.aTwv (651) 
H nicht die vom Spaher vorgetragenen Berichte meinen konne. Man 
wiirde statt dessen aYYEAp.aTWV erwarten ... " Cf Trach. 97-99 TOVTO 
Kapvgat TOV ' AAKp.7}vac, 7TOOt P.Ot 7TOOt P.Ot vaLEt 7TOT', IT 527 VOCTOC 8' 
'AXatwv €ylvEO', WC K7JPVCcETat;, probably OT 737 TOVT' [the murder of 

32 Erbse reduces the clutter of genitives to order by punctuating round £x8lC'TOV MKOVC 
(p.13). Mr Barrett would delete 559 and read lx8,c-rov MKOC after the example of Hipp. 
646-47 acp80yya .•. MIC1J 0"7pwv (and see his note on Hipp. 333-35). 

83 "ITALIB erkannte, dass zu Kap7TOc lC'Ta, der D:u:iv des Profitierenden (attr<p) erganzt 
werden muss, wahrend 8EccpaToLc, Ao~tov ein Instrumentalis ist" (p.16), with the footnote 
"ITALIB bemerkt zur Stelle: 'ElKap7TOC lCTaL i.e. El pl>J.E' Kap7TOv cptPE'V'." ltalie, for what it is 
worth, surely means El pi>J.E' Kap7Tov cp£pnv TeX O£ccpaTa. 



MICHAEL D. REEVE 473 

Laius] JI(7JP";X07J 7TdAE£ (,nuntius venit' Ellendt-Genthe, though Erbse 
could argue that Creon had the news proclaimed by a Kfjpvg). 

Lastly, a word about 803-20, where Erbse thinks he can rescue the 
credit of the transmission by defending the position of 809 and treat­
ing 804 and 820-21 as author's variants designed to follow 810. His 
remarks about 809 are irrelevant unless he is saying that sinceo~8~ fL~v 
is not a mere negation of Kat, it is not a negation of Kat at all.34 His 
speculation about author's variants is simply irresponsible, even if 804 
and 820-21 were marked as alternatives in the Aristophanean text ;36 
for what use are variants to the author of a play written for a single 
performance? He also ignores the difficulty of fLaVTtC €lfL~ TWV KaKwv 

(808) when the messenger has already said av8p€c T€(Jv(iCLV EK X€PWV 

a~ToKTdvWV (805)36 and the even greater difficulty of {JapEa 8' OQv OfLWC 

rppacov (810) after both 805 and o~8' afLrptAEKTWC fL~V KaT€C7To8'Y)fLEvOt 

(809). If this is a way of showing that the methodological possibilities 
have not been exhausted,37 then they never will be. 

That bad arguments are often put up in defence of suspected verses 
is not a startling or important discovery, and it no more reflects on 
good ones than the extravagances of deletors reflect on their suc­
cesses; but it had to be demonstrated not only that there is nothing 
disreputable in principle about deletion but also that between 
deletors and defenders nature has not been unfair in her distribution 
of skills. The first demonstration was concluded in the first part of 

U "Dass der Vers 809 an der richtigen Stelle steht, hat man seit G. HERMANN mehrmals 
bezweifelt, vielleicht deshalb, weil man 0~1l~ /-L~V fur eine blosse Negation der Kopula hielt. 
Das trifft nicht zu, wie die beiden anderen Belege bei Aischylos dartun; man vergleiche 
Bum. 471 ... , Sept. 668 (Eteokles uber das Schildzeichen des Bruders, negierte Behauptungen 
gehen voran) ... Der Vers 809 besagt also, beide Helden seien so in den Staub geworfen 
worden, dass an ihren grasslichen Tod auch nicht der geringste Zweifel moglich ist" (p.2l). 
It is not clear what the last sentence is supposed to contribute to Erbse's argument. 

8S If Aristophanes treated 804 as an acceptable variant, he must have been half asleep, 
because {lac,","ic does not scan, {laCtAEEC is not admitted in trimeters, and {lac'AEwe O/-LOC7TOPO' 
does not mean 'the king's sons'. 

38 C" AdM· I • ." , ,_. , I ".. , :J' n r. 1070-73 ESS. w/-L0' /-Lo,' o,ac 0 TI\Tf/-LWV ayyEl\wv TfKW TvXac co, T , W YEpa'E, 
Ka, cf>D..0", 1l€C7TOTOV. I PEL. alai' 7Tp0/-LaVTtc 6v/-Loe we Tt 7TpoelloK~. I MESS. OUK €CTt co, 7TaEc 
1I'a,lloc ... The scholiast's interpretation of 808, followed by Tucker, has against it not only 
Schutz, Hermann and Weil, but also the tense of El/-Ll. 

37 "Der nachfolgende neue Anordnungsvorschlag erhebt keinerlei Anspruch auf Dber­
zeugungskraft, geschweige denn auf Evidenz. Ich trage ihn nur vor, urn darzutun, dass 
man in der bisherigen Diskussion nicht alle methodischen Maglichkeiten erschopft, son­
dem (von der Anderung einzelner Waner abgesehen) immer nur Umstellungen und 
Athetesen befiirwortet hat" (pp.19-20). 



474 INTERPOLA nON IN GREEK TRAGEDY, n 

this trilogy, and the second part has been a contribution to the second. 
The third part, which will be given over to a number of deletions not 
accepted by recent editors and commentators, will make up for the 
absence from the first part of any but the broadest critical principles. 

EXETER COLLEGE, OXFORD 

June, 1972