¹ The text is as printed in M.L. West, Aeschyli tragoediae (Stuttgart 1990). I would prefer to keep νιν in line 55 and to punctuate with a full stop at the end of line 69. It appears from the critical apparatus that the use of the wedge at lines 64–65 has not given rise to serious doubts about the text. In proposing the emendation recorded on lines 69–70 (CJ 3 (1811) 77) T. Briggs merely observes that he would supplement the change already proposed for line 69 by reading όράιζ in line 70, but he gives no reason for his suggestion.

² The fundamental discussion is by Keramopoullou, A.D., Ό Άποτυμπανισμός (Athens 1923).Google Scholar who deals with PV from this angle on pp. 61–66. His views, as far as the mode of execution is concerned, are largely accepted by Gernet, L. in his discussion of capital punishment in REG xxxvii (1924) 261–93Google Scholar, and by Bonner, R.J. and Smith, G., The administration of justice from Homer to Aristotle ii (Chicago 1938) 279–87.Google Scholar

³ Frag. 193. 5–9, in TrGF iii, ed. S. Radt (Göttingen 1985) 311. The fragment is discussed by Jocelyn, H.D., YCS xxiii (1973) 90–111Google Scholar, who suggests that Cicero in his translation may have paraphrased the Greek original with Roman crucifixion in mind. However, the wedge driven through the chest seems inescapable in PV, where the phrasing of line 65 (στέρνων διαμπὰξ πασσάλευ) certainly suggests that it is being used as a fastening implement; and if the chest, then perhaps the arms and legs too may be so secured. And indeed the plurals hos … cuneos and artus seem overly emphatic for poetic plurals. On any interpretation the use of a wedge for the operation described seems odd; one might have expected a wedge to be used to tighten the bonds without piercing the body directly, but PV 65 appears to be incompatible with this view. Cicero, at least, must have thought the word cuneus applicable to a fastening implement which pierces, perhaps a sharp, tapering rivet with a broad blunt end, or he would hardly have written lines 6–8; and if so, Aeschylus may have used σφήν likewise.

⁴ Under the generalship of Xanthippus a Greek force avenged atrocities committed by a Persian governor of Sestos by pegging (διεπασσάλευσαν) the perpetrator alive to a board (Herodotus vii 33); the prefix perhaps implies nailing; cf. προσηλῶσθαι (Demosthenes xxi 105).

⁵ See the illustration in Gerhard, E., Etruskisehe Spiegel Bd. 2 (Berlin 1845)Google Scholar, Tafel 139, where Prometheus is apparently held by a rivet through the centre of a band fixed across his wrist. The mirror, probably of the late fourth century BC, is sketched and discussed by K. Bapp, ‘Prometheus’, Roscher Lex. col. 3093, fig. 5(a), and discussed also by Eckhart, L., ‘Prometheus’, RE xxiii 1 (1957)Google Scholar coll. 712–716, who notes that the mirror is known only in Gerhard's drawing, so that precise dating is impossible.

⁶ See Hesiod, , Theogony, 522.Google Scholar The wedge is so taken by Griffith, M., Aeschylus. Prometheus Bound (Cambridge 1983) 64–65.Google Scholar

⁷ Examples of such an accidental dislocation compounded by a deliberate attempt at correction by transposition are hard to come by. For an instance exhibiting a degree of complexity similar to that which 1 suppose for PV 1 refer to Euripides Heraclidae 683–91, as corrected by Zuntz, G., The political plays of Euripides (Manchester 1955) 113–14Google Scholar and printed in Diggle, J., Euripidis Fabulae 1 (Oxford 1984)Google Scholar, whose text I give:

In the manuscripts two groups of three lines (684–686 and 688–690) have changed places, the error being partly explicable on mechanical grounds (e.g. homoearchon at lines 688 and 684) leading to the omission of some lines which would be later reinstated at the wrong place. However, if the scribe's eye went straight from line 683 to line 684 he would have overlooked not just three lines but four (688–687 as printed above), which would yield a line order 683, 684, 685, 686, 688, 689, 690, 687, 691. Zuntz explains the arrival of line 687 in the place where it is actually found as a subsequent deliberate transposition to patch up the text produced by the accidental disturbance. If this is correct, the transposer moved line 687 in order to preserve the stichomythia, for otherwise there would be consecutive lines for the servant (686, 688) and for Iolaus (687, 691). The implied level of awareness of the stichomythic pattern on the part of the transposer resembles that which I ascribe to the would-be corrector in PV who introduced the second stage of corruption in an attempt to preserve the pattern of exchanges. Zuntz's emendation is accepted by Wilkins, J., Euripides Heraclidae (Oxford 1993).Google Scholar A very similar double process of corruption is posited at Euripides, IT 766–84, by Jackson, J., cf. Marginalia scaenica (Oxford 1955) 9–12Google Scholar (not accepted by Diggle). For an example of a two-stage corruption at the very simplest level in Ovid, see Willis, J., Latin textual criticism (Urbana 1972) 149.Google Scholar