Bryn Mawr Classical Review 95.02.21

Alan H. Sommerstein (ed., trans.), Thesmophoriazusae. The Comedies of Aristophanes, Vol. 8. Warminster: Aris & Phillips, 1994. Pp. xii + 242. $49.95 (hb). $24.95 (pb). ISBN 0-85668-558-5 (hb). ISBN 0-85668-559-3 (pb).

Reviewed by S. Douglas Olson, Center for Hellenic Studies.

Thesmophoriazusae is one of Aristophanes' funniest comedies, but also one of his most neglected. Indeed, no full-length commentary has been published since Rogers' edition of 1911, and Alan H. Sommerstein's new text and translation of the play is accordingly even more welcome than the previous seven volumes in his widely praised ongoing series. Libraries everywhere will need to purchase this book, and many individual classicists will want to own a copy as well.

As always, Sommerstein offers a Greek text on the left, a modestly idiomatic English translation on the right, and a commentary on the translation (keyed, however, to the numbers in the Greek text, and with constant reference to it) in the back. It is this commentary which will be of most interest to the average reader, and Sommerstein is here up to his usual high standard. His notes are crisp, clear and immensely well-informed, the bibliography is up-to-date, including some work which has appeared only in the last year, and in matters of staging in particular he seems to have thought the play through to a degree that no-one has before. Simply put, this is fine, solid work, and will be of immense value to readers of every sort. The text is slightly less helpful, for reasons, however, which I suspect are largely beyond the control of the editor himself.

Thesmophoriazusae is preserved entire in only two manuscripts, the great 10th c. Ravennas 429 (R) and a 15th c. copy of it, Monacensis gr. 492 (Mu2 = Rogers' H). In addition, portions of approximately 110 lines are preserved in four papyri, and there are numerous quotations in the Suda and similar sources. (Some mention of the latter should perhaps have been included in the "Notes on the Text" on p. 14.) Sommerstein has himself collated R from van Leeuwen's facsimile edition, and in addition reports occasional readings from Mu2, although without indicating his source for them. His text and apparatus have also benefitted from the work of Colin Austin, who has in recent years rediscovered and reattributed numerous early conjectures (Dodone 16 [1987] 61-97) and discussed various problems and cruces (Dodone 19 [1990] 9-29). Unfortunately, the critical apparatus containing all this labor and learning has been removed from the foot of the page, where it belongs and where it has been in the previous volumes in this series, and placed where it is unlikely to be consulted often or even discovered, between pp. 139 45, at the end of the text. The citation apparatus, meanwhile, has been eliminated altogether, while the standard convention of enclosing words which have been added or deleted in angle or square brackets, respectively, has been almost completely abandoned (e.g. 283, 301, 304, 605). The result of all these changes is that this actually quite troubled text appears far smoother than it is, and this is, despite appearances, no service to the reader. Equally unhappy is the way in which the Greek has been set up and printed. Line-numbers are so far off to the right as often to be swallowed up by the binding; breathing marks are difficult to make out, and when under circumflexes, largely illegible; damaged characters are printed (e.g. 276, 500, 953), and the type-face is such that other letters (particularly alpha's) are crushed together and deformed. In addition, there are a number of typographical errors, including the following: 40 SUGKLH/SAS, read SUGKLH/|SAS; 400 EA/N, read E)A/N; 595 A\GGELLW=N, read A)GGELLW=N; 602 E)/XEIS, read E)/XH|S (Goodwin, Moods and Tenses § 195); 729 H)= 'GW\, read H)= 'GW/ (?); 736 MHXANWME/NAI, read MHXANW/MENAI; 741 FH/S, read FH/|S; 813 AU)QH/MERON, read AU)QHMERO\N (e.g. Ach. 522); 908 A)LLA, read A)LLA\; 1040-1 *A(IDA, read A)I/DA; 1043 A)PE/CU/RHSE, read A)PECU/RHSE; 1055 E)/PI\, read either E)PI\ or E)/PI; 1118 EI(/LHFEN, read EI)/LHFEN; 1160 EI/, read EI). These are all relatively minor flaws, and few if any of them can be traced to Sommerstein himself. All the same, it is unfortunate that they have been allowed to escape the attention of the proofreaders and production editors, given the importance of this edition and the likelihood that it will be used and consulted for many years.

In sum, this is an important and helpful edition of a play which should now, as a direct result of Sommerstein's efforts, be much more easily accessible to a wide audience of students and scholars. His Frogs is eagerly awaited. I append a few short notes, having to do chiefly with textual matters.


  • 117-8 Somm. prints KLH/|ZOUSA SEMNA\N / GO/NON O)LBI/AN in place of what is elsewhere reported as R's KLH/|ZOUSA SEMNO\N / GO/NON O)LBI/ZOUSA, presumably in order to eliminate the second participle. (Dindorf [1835] claims SEMNA\N in 117 as his own emendation, but Somm.'s apparatus makes no mention of the word, implying that this is in fact R's reading; clarification of the point in the notes would have been helpful.) This is a substantial change, and one might argue just as easily that, if the language in Agathon's hymn is a bit odd, this may be part of the point of the parody.
  • 140 Somm., comparing Threatte, The Grammar of Attic Inscriptions, p. 477, suggests KATRO/PTOU for R's KATO/PTROU. (Dover [1993] already has this reading in his note ad Ra. 47, perhaps having seen Somm.'s text in advance.) As there is no difficulty with the verse as it stands and as Threatte's examples are all from the 4th c. (only one of them from before 350), this seems unnecessary at best.
  • 164 Somm. proposes TOU/TOU GA\R OU)=N A)KH/KOAS ('for you must actually have heard him sing') in place of R's TOU=TON etc. Once again, however, R's text makes good sense and thus requires no emendation ('for you must have heard about him,' i.e., 'although you are apparently so uncultured that you may not have heard about Ibycus, Anacreon and Alcaeus, and so will miss my meaning'), and the accusative is in any case better with the perfect, since whether In-law has heard of Phrynichus is directly relevant to Agathon's point, whereas whether he ever heard him sing is not, so that with Somm.'s TOU/TOU an aorist would be more appropriate.
  • 288 The text Somm. prints (QU/EIN E)/XOUSAN, EI) DE\ MH/, A)LLA\ NU=N LAQEI=N) appears to be unmetrical, unless we are to take MH/, A)LL- as a single long syllable (synizesis), in which case the comma seems impossible (whence Dindorf's MA)LLA/ for ME A)LLA/). In any case, an explanatory note would have been helpful.
  • 314 Somm. prints Halbertsma's XARE/NTAS E)PIFANH=SAI for R's FANE/NTAS E)PIXARH=NAI. E)PIFAI/NOMAI (unlike E)PIXAI/ROMAI) appears nowhere else in Ar., however, and participial forms of FAI/NOMAI are routine in descriptions of divine epiphanies and parodies thereof (e.g. Pax 631; Ach. 567). Not only is emendation unnecessary, therefore, but Halbertsma's proposal seems on several additional counts less likely to represent the poet's words than does the reading in R.
  • 420 Coulon, Cantarella and Dindorf all print TAU=T' (agreeing with A(\ in 418) without comment, and it is thus again unclear (cf. 117) whether Somm.'s TOU=T' represents an emendation, a typographical error, or a more accurate reading of R. In either the first or the third case, a note would have been helpful.
  • 476 Here again (cf. 288) the text Somm. prints (E)GW\ GA\R AU)TH\ PRW=TON, I(/NA MH\ A)/LLHN LE/GW) is metrical only if one takes MH\ A)/LL- as a single long syllable, and we need either a note or MA)/LLHN, just as we are given MA)LLA/ (for MH\ A)LLA/) at 646.
  • 500 Somm. prints U(PAUGA/S' (< U(PAUGA/ZW, 'to shine beneath' vel sim.), but translates 'against the light,' i.e., U(P' AU)GA/S, which is the reading of most previous editors and makes better sense. Cf. E. Hec. 1154 U(P' AU)GA\S TOU/SDE LEU/SSOUSAI PE/PLOUS.
  • 532 Fritzsche's GUNH/ TIS, adopted by Somm., is logically tighter than R's GUNAI=KES, but this does not justify removing a difficult but perfectly intelligible reading from the text in favor of an easier conjecture.
  • 557 Somm. explains his decision to print SI=TON (R and Suda) in place of Koster's generally adopted OI)=NON by noting that "no-one has ever explained convincingly what use a strigil would be in siphoning off wine." While this is true, Arist. Top. 145a23-5 suggests the trick could be done somehow (TH=| DE\ STLEGGI/DI KA)\N A)RU/SAITO/ TIS), and the verb SIFWNI/ZW ought by all rights to refer to the transfer of something liquid.
  • 581 van Leeuwen reports A)FA/RKTOIS (the text printed by Somm.) as his own emendation, and Cantarella reports A)FRA/KTOIS as the reading in R. If Somm.'s apparatus is not simply in error, therefore, a clarifying note would again have been helpful.
  • 794 Somm. prints R's KATALAMBA/NETE in place of Brunck's universally accepted KATELAMBA/NETE, but offers no explanation.
  • 809 Maas' clever *A)/NYTOS for R's difficult AU)TO/S would seem to deserve at least a mention.
  • 1013 This line is so easily emended (e.g. by the insertion of Dobree's <TOU=T'>) that it seems a shame to print it with cruces, particularly in a text otherwise seemingly intended for a general audience.
  • 1114 Somm. suggests emending R's corrupt SKU=TO to SU=KO, understood as a mispronunciation of SUKH= ('fig-tree,' i.e. 'penis'), rather than to Scaliger's widely accepted KU/STO ('cunt'). This makes better sense of the passage, but it should be noted that the evidence for SU=KON ('fresh fig') as a term for the female genitals, which Somm. in his commentary argues is part of the humor here and thus further support for the emendation, is limited to Pax 1353. Pl. 970, Strattis fr. 3 K-A, and Pl. Com. fr. 286 K-A are all irrelevant, despite Henderson, Maculate Muse § 127.
  • 1197 and 1215 Somm. (against LSJ) takes the first syllable of the rare word SUBH/NH as long, and thus omits Bothe's <SU/> in 1215. This may be right, although the location of the caesura in 1197 would then be unusual (cf. White, VGC § 121.iii; Lys. 768 is, however, a parallel). In any case, a brief note to the reader would once again have been helpful.
  • 1203 There is no reason why Euripides cannot pick up his harp after he unties In-law, and run off carrying it. As a result, there is no reason to emend R's perfectly acceptable TOUTI/ to Dobree's TAUTI/, regardless of how the Schol. of R interprets the action.
  • Commentary and Translation

  • 231 At Eq. 10 (its only other appearance in Ar.), MU\ MU/ is the sound of wailing, and since this sense is at least as appropriate as Somm.'s muffled screaming ('Mmmmmm!') here, it is probably to be preferred.
  • ad 270 Thratta (p. 175) Read "280 Thratta."
  • ad 446 Whatever the merits of the garland-maker's complaints about the effect of Euripides' impiety on her ability to make sufficient money for her family after her husband's death in Cyprus, it is important to note that the Athenian state undertook to provide support for the children of men killed in action (Crat. fr. 183 K-A; Th. 2.46.1, with Rusten ad loc.; Lys. Against Theozotides 1-46 [P. Hibeh 14] [c. 404 BCE]). Indeed, Thucydides' Pericles calls this specifically an W)FE/LIMON STE/FANON (cf. STEFANHPLOKOU=S', 448).
  • ad 485 The claim at Eub. fr. 52 K-A that every house in Thebes has it own KOPRW/N ('outhouse') should probably not be treated as serious historical evidence for sanitary arrangements there.
  • ad 986 For "Peace 383," read "Peace 381."
  • ad 1009-14 For the crane in the Theater of Dionysus, see the comprehensive discussion of Mastronarde, CA 9 (1990) 247-94.