^{page 44 note 1} Meritt, B. D., The Athenian Calendar in the Fifth Century B.C. (Harvard, 1928), pp. 121–2Google Scholar. Curiously enough, Aly, , Philologus, Supplbd. xxi, Heft 3, p. 160Google Scholar, had arrived at the same conclusion by identifying the Aristion of VI. 35 with the archon of 421/0. The attempt of Vollgraff, W., Mnemosyne, 1948, p. 257Google Scholar, to date VI to the late summer of 411 involves emending εῐκοσι in VI. 44 and πρὼτην in 45, making a mountain out of the linguistic molehill πλεῖν ⋯ (44), misinterpreting πειμ⋯^chi;ητοσ in Ar. Av. 1404 (cf. Ar. Thesm. 319!) and—as it seems to me—misunderstanding the political conditions of 411.

^{page 44 note 2} Thuc. 8. 68; [Plut, .] Vit. Ant. 22Google Scholar. See also Ferguson, W. S. in Mélanges Glotz (Paris, 1932), i, pp. 349–66Google Scholar.

^{page 44 note 3} So 417–414, Schmid-Stāhlin, , Geschichte der griechischen Literatur, i. 3 (Munich, 1940), p. 107Google Scholar, following Blass.

^{page 44 note 4} Blass, , Die attische Beredsamkeit, i² (Leipzig, 1887), p. 178, n. 1Google Scholar.

^{page 44 note 5} Breuning, P. S. in C.Q. xxxi (1937), pp. 67 ff.Google Scholar, takes V. 78 to refer specifically to the capture of Antandrus by the exiles and so dates the speech to 424.

^{page 45 note 1} Cf. Wilamowitz, , Sb. preuss. Akad. 1900, pp. 410–11Google Scholar, and especially Solmsen, F., Antiphon-studien ( = Neue philologische Untersuckungen, No. 8, Berlin, 1931)Google Scholar.

^{page 45 note 2} Loc. cit., pp. 413 ff.

^{page 45 note 3} But cf. (ed.) Berge, H. M. ten, Antiphon's Zesde Rede (Nijmegen, 1948), pp. 25–36Google Scholar.

^{page 46 note 1} See Weise, H., Kön. Mar. Gymn. Stettin Osterprogramm, 1890, p. 13Google Scholar; Cucuel, C., Essai sur la langue et le style de Vorateur Antiphon (Paris, 1886), pp. 14–16Google Scholar; Bignone, E., Studi sul pensiero antico (Naples, 1938), p. 178Google Scholar; and (ed.) Wijnberg, S., Antiphon's Eerste Rede (Groningen, 1938), p. 116Google Scholar. The [subtle] prefix seems to have been commented on by Aristophanes as early as 427; fr. 198 (δαιταλ⋯σ), 11. 7–9:

A. τι υποτεκμαιρει κα⋯ κακ⋯σ ανδρασ λ⋯γεισ καλοκ⋯γαθιαν ⋯σκουντασ;

B. οἵμ' ⋯ ⊗ρασυμαχε, τισ τουτο τ⋯ν συνηγ⋯ρων τερατευεται;

^{page 46 note 2} P. Gen. 164–7, ed. pr. Nicole, J., L'Apologie d'Antiphon (Geneva, 1907)Google Scholar. Nicole's edition contains a facsimile of fr. 1 (cols, i–iii); I should like to express here my gratitude to Prof. Victor Martin for his kindness in procuring me a photograph of the remaining three fragments (usually, but misleadingly, called cols, iv–vii) and for answering my inquiries about readings. Many of Nicole's supplements are bizarre; for the improvements of others, see the editions of Thalheim (Teubner), Gernet (Budé), and Maidment (Loeb). I shall quote throughout from Gernet's text for cols, i–iii, from my photograph for the remainder.

^{page 46 note 3} van Cleef, F. L., Index Antiphonteus (Ithaca, N.Y., 1895)Google Scholar is invaluable, a lexicon as well as an index. Cucuel is interesting, but statistically often misleading. There are some useful notes by Richards, H. in C.R. xx (1906), pp. 148–53Google Scholar.

^{page 46 note 4} Not, as Gernet, ‘un procès privé qui fût imminent.

^{page 47 note 1} ευρ⋯πωσ A pr., ευρ⋯πωσ A corr.², ευῤ ⋯πωσ N. Schömann's ευρ⋯πωσ is uncalled for. ευροποσ exists (Anth. Pal. ix. 543), and cf. παλιρροπον γóνυ Eur. El. 492.

^{page 47 note 2} Cf. Lys. 20. 3.

^{page 47 note 3} So Nicole. On my photograph I can read only ✹ο✹✹✹[, which Prof. Martin confirms. ‘Litterae praeter O incertae’, Gemet.

^{page 48 note 1} Nicole read ] ɸερ[. The letter-space before ɸ appeared to me in the photograph to be vacant, the papyrus being broken in such a way as to make any letter but o (which is written very high and very small in this papyrus) impossible. But Prof. Martin, in reply to my inquiry, decided for ] ɸερ[.

^{page 48 note 2} Especially the gnomic tendency of 1–6 and 15, the sacrifice of precision of meaning to brevity and form in ⋯ γ⋯ρ παρουσα ⋯ν⋯γκη ⋯κ⋯στῴ ισχυρο τ⋯ρα ⋯στι τ⋯σ μελλουσησ εσεσθαι (25), π⋯ρισα like τ⋯ρα ⋯στι τ⋯σ μελλουσησ εσεσθαι (31), and the tumid rhetoric of the closing sections. See farther, ten Berge, pp. 22–4.

^{page 48 note 3} Cf. Denniston, J. D., The Greek Particles (Oxford, 1934), pp. 217, 394–5Google Scholar. There are other phenomena shared by V and the Defence, but not significant in the present connexion. For example, both use λυσιτελειν (V. 10, Def. col. iii. 21), but VI neither λυσιτελειν nor σνμɸ⋯ρειν Again, V uses a variety of nomina agentium in -τησ, and fr. 1 (Thalheim) seems to show a similar taste: κατηγορηκεν ωσ στασιωτησ, ἦ και>⋯γὠ και ⋯ π⋯πποσ … ουκ ἃν τουσ μ⋯ν τυραν νουντασ ηδυνηθηαν οι κολασαι τουσ δε. But Harpocration's interpretation is surely wrong here; στασιωτησ does not mean στασιωτησ rather στασιωτησ— cf. the Marxist use of ‘lackey’.

^{page 48 note 4} Wilamowitz, , Hermes, xxii (1887), p. 198Google Scholar.

^{page 49 note 1} Schmid-Stählin, i. 3, p. 104; (ed.) Thiel, J. H., Antiphons erste Tetralogie (Groningen, 1932), p. 22Google Scholar; Maidment, p. 12; Blass, pp. 193–4, and Aly, p. 159 treated it as undatable—with Thalwhich, to judge from his silence, Wijnberg seems to agree.

^{page 49 note 2} See Wijnberg ad loc., who decides for ἦ final and ⋯ and ληθ⋯σ = σαɸ⋯σ, and Sykutris, J., κριτικ⋯ εἰσ αντιɸ⋯ντα (Athens, 1925), pp. 4–7Google Scholar. Thalheim's ( = ⋯ξῇ) ⋯ληθὦσ had been defended at length by Kohm, J., Neue Antiphonstudien (Wien, 1899), PP. 1–3Google Scholar.

^{page 49 note 3} Non-inferential ου with an interrogative can mean ‘but…’ or ‘well then, if that is so …’, but I can find no example of its meaning ‘I mean to say,…’ (as it would have to mean here). Thalheim (followed by Wijnberg) rescued the sentence by inserting it after τòν ⋯μ⋯τερον in 6; which necessitates taking ουν as = γ⋯ρ (imposible), unless και ου is altered to καιτοι (unpleasant and unlikely). The sentence is surely a doublet, as Mätzner saw; cf. VI. 51 ~ 47–50. or Dem. 9. 7–8.

^{page 50 note 1} See Schäfer, H., De nonnullarum particularum apud Antiphontem usu (Göttingen, 1877), pp. 13 ffGoogle Scholar. Here and below I ignore τε … και.

^{page 50 note 2} See Cucuel, pp. 25–8; Bignone, p. 180, n. 1; Wijnberg, p. III. Naturally substantival peri-phrases are rarely the exact equivalents of plain verbs, but are used to avoid ambiguity or clumsiness: ‘un mezzo ingegnoso e speciale di cui nontutti saprebbero servirsi o non si servirebbero conuguale frequenza’ (Bignone).

^{page 51 note 1} V. 19 χαλεπον γε τα εκ πολλου κατεψενσμ⋯να και ⋯πιβεβουλευμενα ταυτα παραχρημα is probably not an example, though Thalheim prints a comma after επιβεβουλευμενα.

^{page 51 note 2} Thiel, , Antiphons erste Tetralogie, p. 11, n. 2Google Scholar, explains the use of -σσ- in I and V as due to the speaker's nationality; he rightly points out that the speaker of I must be illegitimate. Agreed, but we do not know that his mother was non-Athenian, and, even so, why should he use his mother's pronunciation rather than his father's? In any case, Antiphon used αττα in the speech περι του δινδíων (fr. 27), which was pre-sumably spoken by a Lindian, as the περι του was by a Samothracian (ημεισ fr. 50); but the Rhodians used -σσ- (Bechtel, Gr. Dial, ii, p. 631; cf. Schwyzer, D.G.E., No. 284, I. 12)Google Scholar. The fact that Thucydides uses αττα but elsewhere -σσ- does not show that early Attic prose could not use ἄσσα, ἄσσα in [Xen.] Resp. Ath. 2. 17 all MSS. have ἄσσα (which Kühner-Blass, i, p. 612 wants to emend). The other common orthographic criterion, ξυν|συν, is unhelpful: I, V, VI all use συν-, except for ξυν- once in V. 87 and four times in V. 93; I do not know the reason for this distribution.

^{page 52 note 1} See Foumier, H., Le Verbe ‘dire’ en grecancien (Paris, 1946), pp. 152 ffGoogle Scholar. Cucuel, p. 95, is too dogmatic about Antiphon's usage, explaining VI. 17 wrongly and VI. 21 and V. 27 unconancien vincingly.

^{page 53 note 1} Aly, pp. 158–9.

^{page 53 note 2} Cf. Thiel, , Mnemosyne, 1927, pp. 333–4Google Scholar.

^{page 53 note 3} I follow Thalheim's numbering of the fragmentary speeches (Gernet gives only a selection).

^{page 53 note 4} Aly, pp. 161–2 suggests (i) a textual corruption, ιπποκρατουστουστρατηγου ιπποκρατουστουστρατηγου by haplography, then ιπποκρατουστουστρατηγου by correction, (ii) that the case was lost by Hippocrates ⋯ξ ⋯ρ⋯μου because he had just been killed at Delium. For this he adduces the not very happy parallel of Phrynichus.

^{page 53 note 5} [Plut.] may mention the speech because Hippocrates was a famous man, or because it was a good one; the context makes it extremely difficult to decide, but I incline very slightly to the former alternative.

^{page 54 note 1} Though Schmid-Stählin, i. 3, p. 102, n. 1 does not hesitate to do so.

^{page 54 note 2} I.G. i². 65; for the latest text see Athenian Tribute Lists, ii, p. 52.

^{page 54 note 3} In the Thracian panel of the assessment of 425/4 (Athenian Tribute Lists, ii, p. 43) some amounts are preserved, and some fragmentary names, but nowhere both a name and an amount. There is no regular order of names in the panel.

^{page 54 note 4} So Maidment, p. 292.

^{page 55 note 1} But not much out of date, for Demus was active as a trierarch c. 390 (Lys. 19. 25).

^{page 55 note 2} Its identification with the πολιτικ⋯σ by Wilamowitz, , Hermes, xi (1876), pp. 295–8Google Scholar, is most plausible but not relevant to our present considerations.

^{page 56 note 1} For the early dating, see Zuntz, G., Classica et Mediaevalia, ii (1939), p. 142Google Scholar (who, however, misinterprets ῥ⋯τωρ in I.G. i². 45. 21); for the later, Mühll, P. von der, Museum Helveticum v (1948), pp. 1–5Google Scholar (who believes that the Tetralogies show conscious reminiscence of VI–I–V and the Defence, and that their author was ‘ein aus Ionien zugewanderter Schüler Antiphons’). Thiel, , ed. cit., pp. 19–22Google Scholar, treats 427 as an upper terminus because of what he regards as specifically Gorgianic influence. Blass, i², p. 154, is vague: ‘rather after 420 than before’.

^{page 57 note 1} See Thiel, , ed. cit., p. 20Google Scholar.

^{page 57 note 2} I follow the classification of Kühner-Gerth, ii, pp. 46–50 and 77.

^{page 57 note 3} The closest parallels are to be found in some Thucydidean speeches particularly the debate on Mytilene and that at Camarina.

^{page 57 note 4} About the theoretical fourth possibility, that they are by an unknown author but not in his normal forensic style, obviously nothing can be said.

^{page 58 note 1} Though von der Mühll still adopts it. Richards, loc. cit., points out some interesting differences between the three Tetralogies; it is as though the writer were making a stylistic effort fictithat could not be sustained throughout.

^{page 58 note 2} This was denied by Dittenberger, , Hermes, xxxi (1896), pp. 271 ff.Google Scholar; Lipsius, , Berichte Sächs. Ges. lvi (1904), pp. 192–204Google Scholar demolished some of his most important arguments. Dittenberger, in his reply, Hermes, xl (1906), pp. 450–70Google Scholar, gave way gracefully on several points, but still maintained (i) that III. β. 10 is inconsistent with the information of Hesychius and E Il. 9. 632 on απεν7iota;αμτισμóσ, and (ii) that if the sentence ⋯ ν⋯μοσ, ῷ πιστευων … ειργοντι μητε αδικωσ μητε δικαιωσ ⋯ποκτεινειν (III. β. 9, cf. IV. δ. 8) occurred in a fourth-century orator it would not be mere rhetoric but would mean that the law did, in so many words, forbid δικαιοσ πóνοσ. After this, commentators (e.g. Thiel, pp. 13–15) have taken refuge in calling the law of the Tetralogies fictitious or ‘universal’. But I am willing to believe the author capable of almost any rhetorical exaggeration for the sake of effect. See further, Gernet, pp. 6–16.

^{page 58 note 3} See Collitz-Hoffmann, , Samml. d.gr. Dialektinschriften, 4. iv. 2, p. 976Google Scholar.

^{page 58 note 4} Plut, . Per. 36Google Scholar.

^{page 58 note 5} Pl. Phdr. 273 B.

^{page 58 note 6} von der Mühll gives due emphasis to πολλασ, which is rather overlooked by those who suggest 428 as an upper terminus and leave it at that, e.g. Aly, p. 166, Schmid–Stählin, i. 3, p. 108, n. and Nestle, W., Vom Mythos zum Logos (Stuttgart, 1942), p. 394Google Scholar.

^{page 59 note 1} So Navarre, O., Essai sur la rhétorique grecque avant Aristote (Paris, 1900), p. 148Google Scholar, followed by Gernet, pp. 6–7.

^{page 60 note 1} Bignone, , op. cit., pp. 175–215Google Scholar; Luria, , Riv. Fil. liv (1926), pp. 218 ffGoogle Scholar.