428

Eupolis

- Hdn. Grammatici Graeci II1.1 p. 506.19 ~ Hsch. κ 3292 κόκκυ· τό έλάχιστον
(“kokku: the least bit”)
- Suda κ 1916 κόκκυ· Άττικώς άντί του ταχύ (“kokku: Attic in place of
‘quickly’”)
- EM p. 524.52 κόκκυ· Αττικοί τό ταχύ (“kokku: Attic authors in the sense
‘quickly’”)
- Σ Ar. Av. 505 (κόκκυ) ή άντί τού ολίγον (“(kokku) or in place of ‘a
little’”).
Text Photius’ κόκκην is otherwise unattested and metrically impossible, if the
line is iambic tetrameter catalectic (as it otherwise seems to be). Tsantsanoglou
proposed retaining the reading and emending to <“) έν γάρ ταΐσι μάχαισιν
/ άποθνήσκουσιν κόκκην πρώτοι (anapaests).254 Kassel-Austin note instead
that all the other verses in fr. 268 quoted in the vicinity of this one in POxy.
2740 are iambic, and deal with the fact that the upsilon in κόκκυ (drawn from
the commentary rather than the lemma, which is lacunose at this point) is
similarly guaranteed long at Ar. Av. 505; Ra. 1384,255 by not unreasonably sug-
gesting that some vowels in such quasi-words may have had no fixed length.
The upsilon is at any rate guaranteed short in the aorist form of the cognate
verb κοκκύζω at Ar. Ra. 1380 καί μή μεθήσθον, πριν άν εγώ σφών κοκκύσω
and in the perfect form at Ar. Ec. 31 ήμών προσιόντων δεύτερον κεκόκκυκεν.256
Why Kassel-Austin chose to present this fragment separately, rather than
together with the other verses probably drawn from the same scene culled
from POxy. 2740, is unclear. I retain their arrangement for simplicity’s sake.
Interpretation Probably a bomolochic response to something another char-
acter has just said (e. g. “Brave men are rarely wounded in battle”), with γάρ
in the sense “(Yes/No), for ...” (Denniston 1950. 73-4); έν ταΐσι... μάχαισιν
taken over from the other speaker; and κόκκυ πρώτοι reserved for the end as
a punch-line. Cf. Ephipp. fr. 2 “(Heracles) Aren’t you aware, by the gods, that
Pm a Tirynthian Argive? They always fight all their battles drunk. (B.) Which

254 Tsantsanoglou argues that a decisive point in favor of κόκκην is “the double attes-
tation” in Photius, by which he means only that the form of the word in the lemma
is the same as that in the quotation, which is unsurprising and does not actually
amount to two separate data points.
255 As likely also in S. fr. 791 κοκκυβόας όρνις.
256 So too in Aesop. Prov. 119 <ε>ΐ μή αλέκτωρ κοκκύση, τάς ώρας άγνοοϋμεν, and cf.
Hdn. Grammatici Graeci III. 1 p. 525.10 τό δε κόκκυγος συστέλλουσι καί έκτείνουσιν
(“they both contract and extend the form kokkugos”). A κοκκύμηλον is a stone-
fruit” ({ κόκκος), i. e. a plum or the like; the source of the upsilon (short) is unclear.