¹ Canevaro, M. and Harris, E.M., ‘The documents in Andocides’ On the Mysteries', CQ 62 (2012), 98–129CrossRefGoogle Scholar.

² They accept as genuine, however, the lists of names of persons denounced for impiety in 415 b.c. found at earlier points in the text (Andoc. 1.13, 15, 35, 47) because these lists ‘contain names not provided by the orator but confirmed by the Attic stelae [which record the confiscation of the property of those denounced and convicted]’ (C&H 100).

³ The provision (Andoc. 1.87) that a law may be passed relating to a specific individual if, in a secret ballot, at least six thousand vote for it (on this see C&H 117–19).

⁴ There is also a mention of ‘the stele of Demophantus’ in Demosthenes' speech against the law of Leptines (Dem. 20.159), referring back to what was probably a fuller discussion of the decree in a preceding speech by Phormio (cf. ibid. 51), and saying that it included an oath that anyone ‘to whom anything happened when he was acting in defence of democracy’ should be given the same honours as Harmodius and Aristogiton; there is a similar clause in D, more carefully drafted to refer to anyone who met his death while killing, or attempting to kill, a person in one of the categories condemned by the decree.

⁵ This may mean merely ‘the Athenian people’, and need not be taken as evidence for a regular formal renewal of the oath.

⁶ He refers four times to betrayal (once each in §§124, 125, 126 and 127), three times to subversion of democracy (once each in §§124, 125 and 126), and once to tyranny (§125).

⁷ Such as the decree of Isotimides, under which Andocides was prosecuted (Andoc. 1.71).

⁸ The Packard Humanities Institute (PHI) database (http://epigraphy.packhum.org/inscriptions/main) also cites the verb from Ath.Agora xvi 50 (c. 365–355), a treaty with the Siphnians, but there it is a restoration.

⁹ In some cases C&H present two or more logically independent arguments in a single paragraph; I have considered these arguments individually and labelled each separately.

¹⁰ It is used in all other decrees inserted in texts of the orators (if the proposer is named at all), including those at Andoc. 1.77 and 83.

¹¹ Lit. ‘by the bean’.

¹² Similarly the fall of the Four Hundred was marked by the resumption of meetings of the Assembly in its old location, the Pnyx (Thuc. 8.97.1).

¹³ e.g. Hignett, C., A History of the Athenian Constitution to the End of the Fifth Century B.C. (Oxford, 1952), 372–3Google Scholar; Rhodes, P.J., A Commentary on the Aristotelian Athenaion Politeia (Oxford, 1981), 412Google Scholar; Kagan, D., The Fall of the Athenian Empire (Ithaca, NY, 1987), 204Google Scholar.

¹⁴ Cf. Thuc. 8.86.6, where Alcibiades tells the envoys of the Four Hundred that he has no objection to rule by the Five Thousand, but that ‘they must get rid of the Four Hundred and re-establish the Council of five hundred as before’.

¹⁵ Thuc. 8.97.1; [Arist.] Ath. Pol. 33.1. Rhodes (n. 12), 412 points out, furthermore, that under the Five Thousand, in contrast with the subsequent democratic regime, the secretary of the Council could be a member of the tribe currently in prytany ([Plut.] Lives of the Ten Orators 833e, citing a decree ultimately from Craterus [FGrH 342 F 5]: the secretary, Demonicus of Alopeke, and the ἐπιστάτης of the prytaneis, Philostratus of Pallene, were both members of the tribe Antiochis).

¹⁶ It is worth noting that among all the decrees that appear as inserted documents in manuscripts of the orators, this is the only one in which the secretary (or indeed the ἐπιστάτης) is named at all, as they regularly are in inscribed decrees of this period (e.g. IG i³ 91, 97, 101, 104; cf. also [Plut.] Lives of the Ten Orators 833e [see previous note] and Ar. Thesm. 372–4).

¹⁷ IG i³ 53.5–6, 71.56.

¹⁸ The CQ referee further points out that ‘in 410/09 the archon's year and the secretary's year were not yet coterminous’; hence in specifying the date from which a decree was to have effect, archon and secretary were not interchangeable.

¹⁹ MacDowell, D.M., Andokides: On the Mysteries (Oxford, 1962), 136Google Scholar.

²⁰ Strictly speaking, this is a slight backdating, since, as the prescript makes clear, this Council was already functioning when the decree was passed; the symbolic significance of tying the decree to the ‘date zero’ of the democratic restoration is obvious.

²¹ This is not, strictly speaking, an objection to the authenticity of D, but an attempt to rebut a counter-argument.

²² Such a Verschlimmbesserung would be facilitated by the fact that names ending in −ογένης are common and those ending in −ειγένης extremely rare.

²³ Dem. 20.48 speaks of ‘those who overthrew the Four Hundred and … those who did good service when the demos was in exile [sc. under the Thirty]’; Aeschines 2.176 of ‘the Four Hundred and the impious Thirty’.

²⁴ C&H 125 date the real decree of Demophantus ‘probably after the trial of Andocides in 400/399’ – though this seems rather too long after the overthrow of the oligarchy, considering the ferocity of the oath (attested by Lycurgus).

²⁵ Not in the late fifth century, at any rate; it was different in the time of Arthmius of Zeleia, as Dem. 9.44 explains (see below).

²⁶ Cf. also the nouns αὐτόχειρ and αὐτοχειρία, which are found altogether some sixteen times in the orators from Antiphon (5.47, 62) to Lycurgus (Leoc. 122).

²⁷ The one relevant inscription (IG i³ 131.4–5, of the 430s) speaks merely of τοῖς [Η]α[οδίο καὶ τοῖς Ἀριστογεί]τονος. In a law of the 350s (Dem. 20.18, 127, 128, 160) the descendants were called τῶν ἀφ' Ἁρμοδίου καὶ Ἀριστογείτονος, again without the use of any noun at all.

²⁸ See Shear, J.L., Polis and Revolution: Responding to Oligarchy in Classical Athens (Cambridge, 2010), 136–9Google Scholar.

²⁹ We may presume that the oath-taking was done collectively, deme by deme, the members of each of the 139 demes coming forward together to the place of swearing, one of their number (perhaps the demarch) reciting the oath, and all the others confirming by word and gesture their adhesion to it (cf. Ar. Lys. 209–37). Shear (n. 28), 140 apparently envisages a procedure in which each individual recited the oath separately; this is incompatible with her plausible assumption (ibid. 136) that the whole ceremony was completed in a single day.

³⁰ Cf. Shear (n. 28), 208 (suggesting that the same procedure was used for the reconciliation oath of 403/2).

³¹ The CQ referee is ‘rather … worried … by the suggestion that one genuine document has been inserted when the others are all forgeries’. The interpolator inserted a genuine document because he happened to know of a seemingly appropriate one that was readily available; he fabricated the others because he did not know of any appropriate, extant, genuine document that he could use.

³² This article was first drafted as an appendix to my discussion of the Demophantus decree in Sommerstein, A.H. and Bayliss, A.J., Oath and State in Ancient Greece (Berlin, 2012), 130–1CrossRefGoogle Scholar, but the appendix proved to be several times the length of the discussion. I am indebted to CQ's anonymous referee for some valuable comments.