¹ The Authenticity of Prometheus Bound (Cambridge, 1977), 157–72Google Scholar, 269–87 (henceforth Authenticity).

² The term is borrowed from Niedzballa, F., De copia verborum et elocutione Promethei Vincti (diss. Breslau, 1913Google Scholar)and Schmid, W., Untersuchungen zum Gefesselten Prometheus (Tüb. Beitr. 9, 1929), 41–50Google Scholar ; for discussion of their findings and criticism of their methods, see Authenticity 157–60,269–71. Neither of them looked at any Sophoclean or Euripidean plays for comparison.

³ Schmid's investigations (op. cit. 41–50) appear to show that, after Pers. and Th., Ag. is likely to offer the highest rate of Eigenwörter whereas Supp., Cho., and Eum. appear markedly lower (cf. Authenticity 167). OT and Med. were chosen because they are probably both relatively early (i. e. not later than 425 B.C.).

⁴ Reviewers of Authenticity have not commented on the validity or value of the figures for Eigenwörter; nor have I seen any further investigation by anyone else along these lines. Hence my reluctant return to the subject here. The greater part of the labours of identifying and collecting the Eigenwörter in these three plays was performed by two research students, Andrea Shankman and Costas Yialoucas; I am most grateful to them for their help.

⁵ Authenticity 161–4.

⁶ It may seem perverse thus to divide Ag., taking the first part as potentially nonAeschylean, and measuring it against an ‘Aeschylean’ second part. (The effect will be, if anything, to reduce slightly the number of Eigenwörter, since words are naturally liable to be repeated within the same play.) But this seems preferable to measuring Ag. 1–1100 against a substantially smaller ‘corpus’ (Pers., Th., Supp., Cho., Eum., Prom., i. e. six short plays) than any of the other plays (all of which were measured against five plus the much longer Ag.). In any case, the presence of Prom, in that ‘corpus’ will tend to work in the opposite direction, since the play appears, in vocabulary as in other respects, to be distinctly unAeschylean.

⁷ See Authenticity 160–1, 269–71. Once again, I should make clear that I do not pretend that we can define ‘Aeschylus'vocabulary’ in any useful sense. I am merely isolating a large sample body of Aeschylean work against which to measure each of our chosen plays. The discovery tomorrow of a new Aeschylean play would immediately increase attested Aeschylean vocabulary by several hundred words; but it would not necessarily diminish the validity of this test.

⁸ See further Authenticity 161–4.

⁹ Authenticity 163, with n. 47.

¹⁰ Authenticity 163–4, 167–71.

¹¹ Authenticity 163; cf. 167–8.

¹² To save space, time and money, I have not listed here all the actual Eigenwörter, from these three plays. But I should be happy to supply copies of my lists to any scholar who might be interested.

¹³ Here the degree of subjectivity is necessarily greater, and opinions may vary, but it is clear that the overall numbers would not be substantially different; see Authenticity 167–8, with nn.

¹⁴ But in all three cases, at least one of the occurrences of the word is in a metaphor – which might have been used again elsewhere.

¹⁵ Of these, only λŋστńς is really a clear-cut case. Others might exclude e. g. μαντεíα, òδоιρ;éω.

¹⁶ For αύθáδŋς (a borderline case), see Authenticity 168 and n. 66. One might exclude also συντńκω, χρíω.

¹⁷ Persians has an extraordinary number of proper names (Authenticity 166); one of these, Αθ⋯ναι, occurring eight times, is counted as significant, since it could so easily have been used e.g. in Eum.

¹⁸ See Flintoff n. 1 for references to previous scholarship on Aeschylean echoes in Aristophanes. Of these, the most important, Becker, H. T., Aisch. in der gr. Komödie (Darmstadt, 1914Google Scholar), was written before the authenticity question had surfaced.

¹⁹ Birds 685 ff. ~ Prom. 547 ff.; Knights 758 ~ Prom. 59, 308, Knights 836 ~ Prom. 613. Flintoff adds Knights 924 ~ Prom. 365; but the mere coincidence of the word ίπоúυενоς is insufficient in itself to establish a connection, when context and language are in other respects so different. N.b. Pollux 7. 41 and 10. 155, with references to Archilochus (= fr. 235 West), Cratinus (= fr. 91 Kock), and other homely and appropriate sources; and especially Aristoph. Plutus 815 (?), and Lys. 291 έξιπóω. (Prom. 365 itself appears to be derived from Pind. O. 4. 8: see Dionysiaca: Studies … Denys Page, edd. Dawe, R. D. etc. (Cambridge, 1978), 117–20, esp. 119.)Google Scholar

²⁰ Plato, , Prot. 338e–339a, 347bGoogle Scholar; further e.g. Pohlenz, M., ‘Die Anfange der gr. Poetik’, Akad. Gelehrt. Göttingen Nachr. (1920), 142–78Google Scholar (repr. in Kl. Schr., Hildesheim, 1965), Rosen–meyer, T. G., ‘Gorgias, Aeschylus and apate’ AJP 76 (1955), 225–60Google Scholar, Harriott, R., Poetry and criticism before Plato [London, 1969], 92–104, 130–61)Google Scholar; also n. 30 infra.

²¹ See Denniston, J. D., ‘Technical terms in Aristophanes’, CQ 21 (1927), 113–21CrossRefGoogle Scholar, Murphy, C. T., ‘Ar. and the art of rhetoric’, HSCP 49 (1938), 69–113Google Scholar, Sedgwick, W. B., ‘The Frogs and the audience’, C&M 9 (1947), 1–9Google Scholar, as well as Pohlenz (supra n. 20). Of course, Aristophanes was not the first to introduce poetical discussions and contests to the comic stage: Pherecrates’ Krapataloi and Cheiron, Cratinus' Wineflask and Dionysalexandros, Phrynichus' Muses, and titles such as Hesiods, Homer, Sappho, Dramata, Poiesis are recorded, as well as Aristophanes' own Gerytades and Frogs, all on literary themes.

²² From the enormous literature, I should mention at least W. Rhys Roberts’ editions (with useful glossaries) of Demetrius (Cambridge, 1902), Dionysius, , On Literary Composition (London, 1910)Google Scholar, and ‘Longinus’ (Cambridge, 1907)Google Scholar; Russell's, D. A. ed. of ‘Longinus’ (Oxford, 1964), xxx-xliiGoogle Scholar; Wehrli, F., ‘Der erhabene und der schlichte Stil’, in Phyllobolia P. von der Mühll (Basel, 1946), 9–34Google Scholar; Brink, C. O., ‘Callimachus and Aristotle’, CQ 40 (1946), 1–26CrossRefGoogle Scholar, and Horace on poetry I (Cambridge, 1963)Google Scholar; Lausberg, H., Handbuch der literarischen Rhetorik (Munich, 1960), § 406, 1079. 3Google Scholar; Kambylis, A., Die Dichterweihe und ihre Symbolik (Heidelberg, 1965)Google Scholar.

²³ Ψóφоς, Ψоφéω are common words for non-verbal ‘noise’ (as early as Hom. Hymn Herm. 285, and not infrequently in Soph, and Eur.); for human speech, they are usually (always?) derogatory (as Soph. Aj. 1116, Inachus fr. 269c22 Radt, Eur. Hks 229, Rhesus 565); for later disparaging use in literary-critical context, see Callim. Aetia fr. 1. 19. στóμφαξ (‘big–mouth’), στμφоς, στоμφáζоμαι, are rare and utterly unpoetical: see LSJ s.vv., and n.b. Wasps 721, ‘Longinus’ 3. 1, 32. 7.

²⁴ Nevertheless Flintoff argues for ‘the imitation of [Aeschylean] tone’ here.

²⁵ See K. J. Dover's n. ad be, LSJ s.vv. σúστασις B. II, συνíσταυαι IV and V (n.b. Parmenides fr. 4. 3–4, Emped. fr. 35. 6), and σúστŋυα.

²⁶ The interpretation is disputed. I am inclined to agree with Denniston-Page, Thomson, rather than Fraenkel; cf. Lebeck, A.The Oresteia (Harvard, 1971), 86–91Google Scholar.

²⁷ So Denniston (supra n. 21), p. 119: ‘technical jargon is being satirized’; cf. Frogs 1281 στáσιν υελŵν, and the heavily ‘medical’ 939–44.

²⁸ See Griffith, Authenticity (supra n. 1), 168, 284, and LSJ s.v. φáραγξ. Words suffixed in κρημνоς are not common in any authors: according to Buck, C. D. and Petersen, W., A Reverse Index of Greek Nouns and Adjectives (Chicago, 1940), 278Google Scholar, we find in the fifth century άμφíκρημνоς (Eur. Ba. 1051), άγχíκρημνоς (Pind. fr. 82 Snell), πоλúκρημνоς (Hdt. three times, Thuc), βαθúκρημνоς (Pind. N. 9. 40, /. 4. 56), πоλúκρημνоς (Bacch. 1. 121), plus Prom. 421 ùψíκρημνоς (also at Horn. Epigr. 6. 5), 5 ùψηλóκρημνоς and Aristophanes’ ìππóκρημνоς.

²⁹ See Clouds 60–74, and such Aristophanic fancies as ìππóλоφоς (Frogs 818, a variation on Frogs 925 λóφоυς 822 λоφιâς),ìππβáμων, ìππоκáνθαρоς, ìππоνώμας, ìππαλεκτρúων (four times - but no griffin or hippocamp, as in Prom.). So ìππоβáμων (Frogs 821), which is found in an unobtrusive context in Prom. (805), but also at Aesch. Suppl. 284, Soph. Tr. 1095, is unlikely to be a specific allusion to any of those passages.

³⁰ The Greeks’ use of terminology taken from crafts (weaving, carpentry, chariot-making, pottery, etc.) and from sailing, riding, and driving, to describe the process of literary creation and adaptation, is apparently inherited from Indo-European times: see Schmitt, R., Dichtung und Dichtersprache in indogermanischer Zeit (Wiesbaden, 1967)Google Scholar, id. (ed.) Indogermanische Dichtersprache (Darmstadt, 1968)Google Scholar, Maehler, H.Die Auffassung des Dichterberufs im frühen Griechentum (Hypomnemata 3, Göttingen, 1963)Google Scholar, and L. Radermacher's ed. of Frogs, passim. The idea of the poet as τεχνíτης and δημιоυργóς, which is as old as Od. 17. 382–6, underlies the whole plot of Frogs (n.b. esp. 761 ff.). πоιητńς = ‘poet’ is first found in Herodotus.

³¹ See Buck and Petersen (supra n. 28) 199. Apart from Prom., there are only four occurrences of such words in all Aeschylus: θρασúστоμоς (Ag., Th., and also Eur. fr. 3), στενóστоμоς (fr. 108N, satyric), χαλκóστоμος (Pers.), plus the verbs εύστоμéω, θρασυστоμéω, both found also in Soph, and Eur. In Sophocles we find ðστоμоς (twice), èπτáστоμоς, δíστоμоς (twice), άμφíστоμоς, χαλκóστоμоς, άθυρóστоμоς, διχóστоμ;оς, εὒστоμος; also κακоστоμéω; in Euripides, èφτáστоμоς (three times), διστоμоς (twice), δíσττομоς (twice) èκατóστоμоς, θρασúστоμоς, óξúστоμоς; also έλε:υθερоστоμéω (as in Prom). N. b. too Herodotus άμφíστоς, εὒστоμоς, λεντáστоμоς, and [Homer] Batram. 295 ψαλιδóστоμоς. Aristophanes himself calls Euripides στоματоυργóς at Frogs 826; and see n. 36.

³² N.b. especially Soph. Ant. 1028 (also 875 αύτоγνωτòς όργá), OT 549, Eur. Med. 104, 223, 621, 1028, El. 1117, Hks 1243, etc.; and Hdt. 6. 92, Hippocr. Airs 24. 60 (where n.b. too ðγριоς).

³³ 'Longinus’ 22. 3 (of Thucydides’ style), Dion. Hal. De comp. verb. 22 (p. 228. 9 in Rhys Roberts’ ed.; also of Thuc), Dio Chrys. 52. 4; cf. Aristotle, Theophrastus, Xenophon etc. (of personalities); also Denniston (supra n. 21) 115.

³⁴ See W. B. Stanford's n. ad loc.

³⁵ χαλινóς is used in other metaphorical ways by Pindar, Sophocles, Euripides; see too Denniston (supra n. 21) 114–15. In Prom., not surprisingly, words for ‘harnessing, binding’ etc., both literal and metaphorical, are particularly common, and not only of Prometheus’ shackles: n. b. 5, 618 όχμáζω, 54 ψáλια, and my edition of the play (Cambridge, 1983), pp. 20–21; so the two instances of χαλινóς do not stand out as being in themselves remarkable.

³⁶ It is notable too that in the same line Aristophanes uses either άθúρωτоν or άπúλων (στóμα), for which cf. Eur. Or. 903 άθυρóγλωσσоς, Soph. Ph. 188 άθυρóστоμоς, Simonides PMG 541. 2 ἂθυρоν στóμα.

³⁷ The further echo which Flintoff finds between Frogs 816–17 … ὂμματα στρоβńσεται and Prom. 882 τρоχоδινεῖται δ’ ὂμμαθ èλíγδην, seems again too faint to be significant, even with the ‘bovine element’ of Frogs 804 counted in (Flintoff, p. 4): Aeschylus ‘staring like a bull’ is a far cry from the tormented heifer-girl of Prom.

³⁸ See further Taplin, O., HSCP 76 (1972), 5–97Google Scholar.