^{page 1 note 1} MSS. have οἱ τδε νικ⋯ν: see below, p. 2, n. 5.

^{page 1 note 2} The Greek is quoted in the following paragraph.

^{page 1 note 3} See below, p. 6.

^{page 1 note 4} Sc. a σκολι⋯ ῥ⋯τρα: above, p. 63, n. 4. Εὐθεῖα is the formal opposite of σκολι⋯: the connexion of this notion with πρ⋯σθεσις and ⋯φα⋯ρεσις (Plut, . Lyc. 6. 7Google Scholar, cf. 13. 3) is illustrated by the lines of Theognis referred to on p. 2, n. 1, τ⋯ρνου κα⋯ οτ⋯θμησ κα⋯ κα⋯ γνpώμονος … οὔτ' ⋯φελών, etc.

^{page 1 note 5} Tyrtaios fr. 4 (Diehl), πατ⋯ρων ⋯μετ⋯ρων πατ⋯ρες. At least two generations, since the words might be understood more indefinitely; cf. Iliad 20.308, κα⋯ πα⋯δων παῖδες το⋯ κεν μετ⋯πισθε γ⋯νωνται.

^{page 2 note 1} How grave a view might be taken of this we may judge from the lines of Theognis, 805–10: ‘straighter than a straight-edge the man must be who receives oracles from Delphi: if you add to them, nothing can heal you: if you subtract, your guilt abides.’ Cf. Herodotus 7.6.3: Onomakritos was evidently forgiven when they were all in trouble together. Theognis no doubt doth protest too much: but oracular business (one may say without cynicism) had to combine pedantic principles with liberal practice. The conjuror will insist that there is no deception. I need not remind my reader that I am not discussing real behaviour (I do not myself believe that the two Kings inserted a clause into an oracle): I am discussing whether Aristotle could allow himself tro believe such a story. It seems to me no worse than Cicero's story of the Augur's Wink. The real behaviour which I do posit, and which may not be ‘straighter than a straight-edge’, is Tyrtaios'. Did he really know that the Pythioi had in their archives an ancient oracle such as he cites? I do not feel censorious about it.

^{page 2 note 2} It is true that stories of oracles are often illogical. Plutarch speaks (doesn't he?) as if Lykourgos wrote the Rhetra himself (6. 4, προσηγ⋯ρευκεν and esp. ⋯ν⋯ψε: see above, p. 63, top): we could suppose that Aristotle here thinks of the Rhetra as oracle-inspired rather than as Apollo‘s own words: the inspiring oracle, then, was said by the two Kings to involve Clause III, and Tyrtaios confirms their claim. This saves Aristotle's moral niceness at the cost of his consistency:P it still goes some way towards explaining why he thought (still on Tyrtaios' evidence) that the Rhetra was an oracle; only his notion of what he meant by an oracle becomes more hazy. Plutarch's sentence introducing Tyrtaios' lines could no doubt be under-stood in other ways still, once we allow this haziness.

^{page 2 note 3} The poem is so named by Aristotle (Pol. 1307^a) and Strabo (8. 4.10). Aristotle quotes no lines, Strabo quotes fr. 2 (Diehl). Our lines (fr. 3b, Diehl) as well as Diodoros' variant (fr. 3^a) are very frequently ascribed to it by modern scholars.

^{page 2 note 4} Andrewes thinks it does, because of the plural (p. 99), and that this is why in Diodoros' version the opening couplet is rewritten. But why must Lykourgos, more than any other inquirer, go without company?

^{page 2 note 5} Οἴκαδ' s an emendation, the MSS. have οἱ τ⋯δε νικ⋯ν: SC. Οἴταδ' ἔνεικαν? ‘They brought the oracles from Delphi to Oita.’ This might suit very well with Aigimios and his sons, but will not suit at all, I think, with e.g. the tence of μ⋯λει in line 4. We must accept οἴκαδ': the Herakleidai (or whoever) did not bring their oracles with them when they came from central Greece, but fetched them thence after they had settled in Sparta.

^{page 2 note 6} Not necessarily: I mean that Aristotle was not bound to infer this, though apparently Hellanikos and Xenophon did. See pp. 4–5.

^{page 3 note 1} The marginale has been unfairly spat upon. Dindorf writes: codex inepte ⋯ Πυθ⋯α, etc.: quasi sequentia, quae sunt Tyrtaei, sint Pythiae. I question how much of the lines is Tyrtaios': certainly their relevance for Diodoros was that they contained an oracle.

^{page 3 note 2} Our corrections may be wrong, but I am protesting against e.g. Andrewes p. 98 ‘the dubious gain … is not worth the cost of a triple corruption in two lines of Diodorus …’ Andrewes's resolve to think badly of Diodoros' verses makes him unduly tender towards the Excerpta's text.

^{page 3 note 3} ‘Certainly the last couplet must be separated from lines 3–6. The hexameter … amounts to a direct denial of the doctrine’ of these lines (p. 98). He then quotes Meyer's view that the couplet is part of a polemic against Tyrtaios, ‘but the collocation is more probably a pure accident’ (ib. note 2).

^{page 3 note 4} ‘The resemblance even of Diodorus' lines to the Rhetra has been greatly exaggerated’ (Andrewes, p. 97). Captain Andrewes allows me to say he would withdraw this, and is now prepared to think that even Plutarch's lines refer to the Rhetra. This will (as I see it) involve the abandonment of his view that the Rhetra is later than Tyrtaios (his p. 96 with notes 1, 2, and 3: see my next note). That view is not fundamental to his paper, whose clear thinking has done much to elucidate Sparta's constitutional problems.

^{page 4 note 1} Aristotle Pol. 1307^a speaks of the danger of great wealth and great poverty coexisting, κα⋯ μ⋯λιστα ⋯ν τοῖτο γ⋯νεται. συν⋯βη δ⋯ κα⋯ το⋯το ⋯ν Λακεδα⋯μονι ὑπ⋯ τ⋯ν Μεσσηνιακ⋯ν π⋯λεμον. This was the occasion of Tyrtaios' poem. Andrewes argues from this (p. 96, note 1) that Tyrtaios wrot during the war; that ὑπ⋯ τ⋯ν Μ. π⋯λεμον must mean this, since it is an instance of ⋯ν τοῖς πολ⋯νοις, SC. he takes συν⋯βη το⋯το as equivalent to το⋯το γ⋯νεται. I understand Aristotle to say that these things come into being (γ⋯νεται)( during wars: this will not prevent the crisis being precipitated (συν⋯βη) after the war is over. I would put the war (and Tyrtaios) a little later than he does (cf. p. 1, n. 5 above), and the Rhetra a little earlier.

^{page 4 note 2} Plut, . Lyc. I. 2Google Scholar, confirmed by the Epitome of the Λακ. πολ. (= Arist. fr. 611, § 10). Lykourgos is contemporary with the First Olympiad. Theopompos will be two generations from Lykourgos, Tyrtaios two from Theopompos.

^{page 5 note 1} F. gr. Hist. 4 F. 116.

^{page 5 note 2} Xen. Λακ. 10. 8: quoted by Plutarch, , Lyc. i. 5–6Google Scholar. The fact that Xenophon uses the phrase κατ⋯ τοὺς 'Hρακλε⋯δας as an indication of date shows that Plutarch is right in understanding him to mean the first Heraklids to settle in Sparta.

^{page 5 note 3} The synchronism with Ol. I: see p. 4, n. 2. This rested on the diskos, which (pace Meyer and Jacoby) I am confident he did not dicover for himself but owed to Hippias.

^{page 5 note 4} King Pausanias' pamphlet containing many oracles is mentioned by Ephoros, F. gr. Hist. 70 F. 118: Ed. Meyer suggested that the oracles in Diodoros 7 are from this source, via Ephoros (Forsch. i. 215 ff.). Though Meyer misconceived the pamphlet's tendency (see next note), I believe his main thesis still stands. Diodoros 7. 12 (Vogel) gives Herodotos' oracle (1. 65) with two extra lines.

^{page 5 note 5} Pausanias' pamphlet was κατ⋯ τ⋯ν Λυκο⋯ργευ ν⋯μων, the κατ⋯ being guaranteed by the Vatican palimpsest of Strabo: Ehrenberg, Neugründer, p. 14.

^{page 5 note 6} Or does ⋯πιστ⋯μενος depend on εἰμ⋯ understood? An unusual construction; cf. Od. 4. 231. In Solon fr. 1 (Diehl) line 52, ξυλλ⋯γεται β⋯οpτον can be understood more easily. I know no satisfactory explanation of the participle φυλασσ⋯μενον in Theognis 806 (the passage referred to on p. 2, note 1 above). It is tempting to put the two couplets into recta and write ἔπεσθε for ἔπειτα:

ἄρχετε μ⋯ν βουλ⋯ες,

οἶσι μ⋯λει Σπ⋯ρτης ἱμερ⋯εσσα π⋯λις,

πρεσβ⋯τα⋯ τε γ⋯ροντης ἔπεσθε δ⋯, δημ[⋯ται ἄνδρες,

εὐθε⋯αις ⋯νταπαμειβ⋯μενοι.

^{page 5 note 7} As aginst Andrewes and Meyer: above p. 3, n. 3.

^{page 6 note 1} This doctrine is repeated in de Pyth. or. 19 (Mor. 403 E), αἱ ῥ⋯τραι δι' ὧν ⋯κ⋯σμησε τ⋯ν Λακεδαιμον⋯ων πολιτεἰαν Λυκο⋯ργς ⋯δŹθηαν αὐτῷ καταλ⋯δην. Elsewhere (e.g. Mor. 227 B, C) he treats the three Rhetrai as utterances of Lykourgos himself.

^{page 6 note 2} See Appendix I.

^{page 6 note 3} IG v, fasc. 1, no. 20, lines 2–3: no. 1498, line 12.

^{page 6 note 4} Xen. Anab. 6. 6. 28. The youthful Wilamowitz (Hom. Unt., 1884, p. 280, note 16) said it here means Vertrag, and in a Spartiate's mouth. Ed. Meyer corrected him over 50 years ago (Rh.M. xliii, 1887, p. 82, note 2)Google Scholar but Liddell and Scott still say it means covenant. The man is probably an Arkadian, though he is pleading against a Spartiate defendant and before a Spartiate judge. The ῥ⋯τρα he refers to is an enactment of the Army: it is given in 6. 6. 2, εἴ τις ×ωρ⋯ς ⋯πελθὼν λ⋯βοι τδημ⋯σιον ἔδοξεν εἶναι.

^{page 6 note 5} The Tabulae Heracleenses: IG xiv, 645, Schwyzer, Dial. gr. ex. ep. pot. 62: early Hellenistic. Lines 95–185 are headed συνθηκα Διονυσω χωρων: if the parties do not fulfil their obligations, they shall be answerable: 145–6 hυπολογοι εσσονται κατ τας ρητρας και κατ τααν συνθηκαν, in accordance with the laws and with this agreement: 151 hυπολος εσσηται κατ τας ρητρας. The usage in this Spartan colony is confirmed for Tarentum by Photius s.v. ῥ⋯τραι: συνθ⋯και λ⋯γοι ⋯μολο⋯αι (this is the Homeric use). ταραντῖνοι δ⋯ ν⋯μον κα⋯ οἶον ψηφ⋯σματα. παρ⋯ Λακεδαιμον⋯οις ῥ⋯τρα Λυκοργου ν⋯μος, ὡς ⋯κ χρησμ⋯ν τιθ⋯μενος (Plutarch's view).

^{page 6 note 6} IvO 2: Schwyzer, op. cit. 409.

^{page 7 note 1} IvO 10,11: Schwyzer 414, 415.

^{page 7 note 2} IvO 7: Schwyzer 412.

^{page 7 note 3} Schwyzer 687 A; Tod, Selection, 1.

^{page 7 note 4} IvO 9: Schwyzer 413.

^{page 7 note 5} This means ‘he drafted a motion’: in 9.1, the motion has been drafted, but not yet accepted by the Gerontes. In the Spartan document of Hellenistic date, IG v, fasc. 1, No. 1498, the same words are used of the publication on stone of the enacted law: ταν δε ρητραν ταυταν γραψαντες εν σταλαν λιθναν ανθεντω τοι βιδυιοι. Ehrenberg, Neugründer 18, calls the phrase ῥ⋯τραν γρ⋯φειν an ‘Ungenauigkeit der späteren Autoren,’ as if a Rhetra could not be written down. Why ever not? They wrote them on bronze at Olympia, from a very early date. In the archaic inscription from Mykenai, IG iv. 493, Schwyzer 98, the words κατα (i.e. κατ τα) fεfρεμενα no doubt refer to something which was in writing, as εἰρημ⋯νον in Thuc. 5. 39–3 and εἴρητο in 5. 46. 2 certainly do.

^{page 8 note 1} Cf. the variants Ἑτοιμοκλ⋯ς (Anth. Pal. 7. 720): the latter form recurs in Xen. Hell. 5.4. 22 and 6. d. 33.

^{page 8 note 2} He made a set speech (δημηγορο⋯ντος), and unless he is one of the Gerontes, his γνώμη must be an amendment (κατ⋯ τ⋯ν ⋯κε⋯νου γνώμην ψηφ⋯ξεσθαι μελλ⋯ντων).

^{page 8 note 3} He would read δαμώδων γορ⋯αν: the latter word will be part of the Illyrian (‘hylleisch’) element in Laconian Greek, and correspond to heriam in an Illyrian inscription. For δαμώδων see Hesychios δαμώδης: the word is exactly apt, and I think posibly the true original is δαμώδ(ων) ⋯ν(τα)γορ⋯αν. The asyndeton will be an advantage.