¹ To the bibliography there cited (read Pomtow, H., Klio xvii, 1921, 171Google Scholar) add P. Foucart, JS 1906, 569 (brief mention in connection with discussion of campaigns against pirates of 74–71); A. Wilhelm, JÖAI 1914, 98 (brief mention in connection with discussion of demands for military assistance by Rome); M. Cary, CR 1924, 60; J. Dobiás, Listy Fil. 1924, 13 and 94; 1925, 65 (see n. 25 below); Carcopino, J., ‘Sur la loi romaine du monument de Paul Emile’, Mélanges Glotz (1932) i, 117Google Scholar; F. T. Hinrichs, ‘Die lateinische Tafel von Bantia und die “Lex de Piratis”’, Hermes 1970, 471.

² Column IV = pl. 15, fig. 6.

³ The size of the blocks compares with those of the blocks forming the seaward wall of the magazine that flanks the trireme harbour on the North side and backs onto the Agora. There is a chance that the blocks were robbed from the upper course of this wall.

⁴ Ptolemy X Alexander I left Cyprus for Alexandria in 107. Soon after, his brother Ptolemy IX Soter II passed through Cyprus on his way to Syria, but at least one Cypriot mint continued to strike for Ptolemy X down to 105/4, and no Cypriot mint started to strike for Ptolemy IX until 100/99. It seems to follow that Ptolemy IX, although he returned to Cyprus from Syria soon after 107, did not control the island until well after his Syrian adventure of 103–102 was over (see Will, E., Histoire politique ii, 370–1; 376–7)Google Scholar; anyone mentioning the king of Cyprus between 104 and 100 might well have doubts about his identity and express them by the formula found in the Cnidos text. (We are grateful for the numismatic information to Dr. O. Mørkholm.)

⁵ The Delphi law occupies three blocks of approximately equal size on the front face of the monument of L. Aemilius Paullus. It is in the highest degree improbable that it ever occupied more than this. The relevant layers of the monument, together with the corresponding blocks of the inscription, are as follows:

Of these layers, V, II and I are almost completely preserved, with the bulk of the front face and the two side faces extant. The degree of taper of the monument appears to make it certain that no more than two layers ever intervened between Layer V and Layer II. Layer III = Block B was seen by Colin and others from its content to precede Layer II = Block C, and this is confirmed by the text from Cnidos. Layer IV = Block A presumably lay immediately above Layer III = Block B.

It is moreover impossible that any of the Delphi law ever lay below Layer II = Block C. The righthand face of Layer II bears an inscription which continues down onto Layer I, the front face of which is occupied by two copies of a proxeny decree. (It is also relevant that there are two uninscribed lacunae towards the bottom of Block C and a tendency towards crowding of the letters.) The left-hand face of Layer II is occupied by an inscription and both sides of Layer III are blank. The only place where any more of the Delphi law may be postulated is on the left-hand or right-hand face of Layer IV = Block A. But the date of the inscription on the front face of Layer I is later than the date of the law; if one is to invoke this inscription on Layer I to explain a placing of some of the law on the side of Layer IV, one has to suppose that the law was inscribed some time after it was passed, perhaps not a very plausible hypothesis.

⁶ R. K. Sherk, Roman documents from the Greek East (hereafter Sherk), pp. 18–19.

⁷ Sherk 12 (an inadequate guide to the texts of the two copies).

⁸ See T. Drew-Bear, Historia 1972, 85–6.

⁹ See also U. Laffi, Studi classici ed orientali 1967, 5.

¹⁰ R. K. Sheik, pp. 207–8.

¹¹ The part omitted relates to details of procedure for trials; Delphi had much experience in inscribing Roman documents and doubtless saw little point in inscribing all this. We find it hard to think of any other principle of selection; local relevance cannot be invoked to explain what appears at Delphi or most of what appears at Cnidos (see above).

¹² The Delphi text is hereafter referred to by the notation A, 1 (= Block A, line 1) and so on, the Cnidos text by II, 1 (= Column II, line 1) and so on.

¹³ For sc(iens) taken as s(in)e see SEG iii, p. 82, n.; the observation on the translation of sciens and s(in)e is due to Professor H. B. Mattingly.

¹⁴ The importance of the comparison is emphasized by G. Colin, BCH 1924, 91; Carcopino, J., Mélanges Glotz i, 123Google Scholar; also F. T. Hinrichs, Hermes 1970, 471; all hold that the law from Delphi was designed to provide for a great command for Marius (see n. 28).

For the text of the Bantina, Lex see CIL i ², 582Google Scholar; FIRA i, 6. For a reflection on the coinage of the climate of opinion that exalted the iusiurandum in legem see Crawford, M. H., Roman Republican Coinage ii, 605Google Scholar.

¹⁵ Presumably the correct translation of ἔπαρχοι, in view of the consistent use of ἐπαρχεία to translate provincia (so G. Colin, BCH 1924, 91; H. Stuart Jones, JRS 1926, 171–2, misunderstood by Carcopino, J., Mélanges Glotz i, 123Google Scholar), compare IGRR iii, 714. Latin possesses no word for ‘provincial governor’ in the Republican period, compare the de Termessibus, Lex Antonia (FIRA i, 11Google Scholar, II, line 6), and we assume that in the Latin text there stood here something like (magistratus) qui provinciis praesunt, rather than (with Colin) the normal sequence of titles (see on II, 13), praetor prove praetore prove consule, of which the first element is ambiguous.

¹⁶ The first exclusion in this paragraph perhaps reflects the fact that all tribunes of the year were in its favour (so H. Stuart Jones, JRS 1926, 172–3); contrast lines 11–13 and 26–28. If this is right, two points are worth making: one can calculate that the beginning of Delphi A is likely to correspond to a point on Cnidos II somewhat above the point at which our text begins; if the whole preamble of the law, including the names of ten tribunes, was inscribed at Cnidos (the preamble was not inscribed at Delphi), that would account for much of the space remaining to be filled at the top of the lost portion of Cnidos II (see p. 200). Secondly, the year 100 appears to be ruled out for the law; for Saturninus was elected only when one successful candidate had been immediately murdered and certainly ended the year in total isolation from his tribunician colleagues.

The second exclusion is doubtless to be explained by the supposition that it was not thought worth while getting an oath from current governors not immediately involved with the subject of the law.

¹⁷ Doubtless covered already as ex-magistrates (for the terminology see n. 15).

¹⁸ We do not understand how F. T. Hinrichs, Hermes 1970, 475 can equate this clause with C, 8–10.

¹⁹ These very specific provisions appear to correspond to the rather vague provision at Delphi that anyone who disobeys the law or fails to swear is not to be unpunished.

²⁰ Reading iouraverit in line 20 for the ioudicaverit of the bronze.

²¹ We are more puzzled by this paragraph than earlier commentators; in particular, we are not happy with the idea of taking ὄστις and ὅσοις to refer to the same category (with Colin, and Naber, (in his translation in SEG iii, pp. 84–5))Google Scholar. Apart from the change from singular to plural, there is also a change from indicative to imperative; we also suspect that the difference of voice between κρίνεσθαι and κρινάτωσαν is significant, and find it very difficult to see how a law can have started by identifying a possible prosecutor as ‘anyone who wishes who is free’, but have gone on to exclude people by talking of entitlement according to the law; in any case, if we are right in supposing that Cnidos V follows on from Delphi C., it is not clear where the provision concerning such an entitlement can have stood. We prefer to suppose that όσοις refers to those who may be prosecuted, and that the whole clause represents something like quibus multam irrogare ( = ἐναιτήσειν, with G. Colin, BCH 1924, 93, n. 2) quos in iudicio sisti liceat; influenced by the plural, the translator has then placed ἀγέτωσαν and the other remaining verbs in the plural, although the subject is strictly speaking the same όστις. The Latin is perhaps here hic agat sistat nomen deferat and so on. The existence of judicial proceedings in our law before nominis delatio presupposes the existence of the procedure of divinatio, on which see A. H. M. Jones, Criminal Courts 54 and 63–4.

With the right of any citizen to prosecute compare FIRA i, 18, line 35; 21, ch. 61.

²² Arguments summarized by F. T. Hinrichs, Hermes 1970, 473–86; erratic bibliography on p. 475, n. 4. J. Carcopino, Autour des Gracques 211, objects that the appearance of tribunes in the law shows them not yet to have had the automatic right to a place in the Senate which they had acquired by 102; but since in the only passage of the law which is not restored they appear in the company of every other magistrate from the dictator downwards, we do not see the force of this objection.

²³ The Fragmentum Tarentinum (Epigraphica 1947. 3) appears to contain provisions corresponding to lines 13–15, 19–20 and 20–22 of the Lex Bantina, followed by provisions for promulgation of the law (which perhaps correspond to lines 31–32 of the Lex Bantina) and then by a prohibition on fraudulent non-enforcement of the law (which recalls part of B, 15–19 at Delphi) and the formula s.s.s.e.q.n.i.s.r.e.h.l.n.r. (on which see R. Bartoccini, Epigraphica 1947, 13, n. 4).

We are not convinced by the suggestion that the Fragmentum Tarentinum bears part of the Lex repetundarum of the Tabula Bembina (H. B. Mattingly, ‘The extortion law of the Tabula Bembina’, JRS 1970, 154; half-accepted by A. N. Sherwin-White, ‘The date of the Lex repetundarum and its consequences’, JRS 1972, 83, esp. 91–2 and 99).

²⁴ 41 lines of Cnidos IV correspond to 12 lines at Delphi (bottom of B and top of C), bearing in mind that there must be one blank line at the bottom of Delphi B; the 27 remaining lines of Delphi C are then the equivalent of 92 lines of the same length as those of Cnidos IV, but 69 lines of the length of those of Cnidos V (which are either side of a third as long again); of these 69 lines some 60 may be ascribed to the missing top part of the column (see p. 200).

²⁵ G. Colin, BCH 1924, 63; FD iii, 4, pp. 47–50; E. Cuq, CRAI 1923, 130–1, argues that a law on piracy is inappropriate to the period during or after the campaign of M. Antonius, and that the mention of C. Marius and L. Valerius does not necessarily provide the date of the law, since it is not in the preamble (see also n. 16 above); but it seems to us impossible to see B, 20–2 except as a command to the governor of Asia during the consulship of C. Marius and L. Valerius. Cuq is followed in his negative arguments by J. Dobiás, Listy Fil. 1924, 13 and 94; 1925, 65, who goes on to argue for a date of 99–96 (we are indebted to Dr. M. Teich for help with these articles).

²⁶ Hermes 1970, 488, n. 1.

²⁷ For conventions governing the reception of foreign embassies in the Senate, attested by this law and by Sherk 18, line 66, attested as assigning the month of February by Cicero, , II Verr. i, 90Google Scholar; cf. ii, 76; Pseudo-Asconius 244 St., see G. Colin, BCH 1924, 72; FD iii, 4, pp. 48–9; H. Stuart Jones, JRS 1926, 169; Carcopino, J., Mél. Glotz i, 120Google Scholar; given the mention of February in 70, it seems to us preferable to attribute the whole set of rules to A. Gabinius, Tr. Pl. 139, rather than to allow A. Gabinius, Tr. Pl. 67 a hand (contra, M. Griffin, JRS 1973, 210, n. 134; note Sherk 10, B, a decree of the Senate passed on 9th February, 135 in response to envoys from Samos; Josephus, , AJ xiii, 260Google Scholar, 6th February, 126 (?), Broughton, T. R. S., MRRP i, 509, n. 2)Google Scholar.

Against Hinrichs (490, n. 3), it also seems clear to us that B, 28 is quite general in its application and does not refer to a particular year.

²⁸ G. Colin, BCH 1924, 76; FD iii, 4, 50–2; Carcopino, J., Mél. Glotz i, 117 and 130–2Google Scholar; W. Schur, ‘Das sechste Consulat des Marius’, Klio 1938, 313 (of little interest); F. T. Hinrichs, Hermes 1970, 471, noting on p. 496 the later hostility of Marius to Antonius, an d arguing a wish to outdo him now; contra, A. Passerini, Athenaeum 1934, 134–7 = Studi su Gaio Mario 87; Athenaeum 1939, 62–4 = Studi 207–9 (Passerini's own interpretation of th e law as hostile to T. Didius and designed to help the Equites does not seem to us to have any more support from the text).

²⁹ That part of the Cnidos text which corresponds to B, 27–8 now shows that Macedonia is not being removed from the normal senatorial allotment of provinces, contra Carcopino, J., Mél. Glotz i, 125–6Google Scholar. Note that our law, like the Lex Sempronia, while ‘popularis’ in tone, preserves a role for the Senate, IV, 6–7 and 27–8; C, 6 (?).

T. J. Luce, ‘Marius and the Mithridatic command,’ Historia 1970, 167–8, while taking the same view of the law of 101 as ourselves, has nonetheless persuaded himself that Marius’ journey to the East was designed to secure a command in the East (how?), and that the measures taken by Rome with regard to the Eastern provinces in the 90's, of which he mentions a few, were designed to thwart Marius.

³⁰ Appian, , BC i, 45Google Scholar.

³¹ Justin, xxxvii, 4.

³² Cicero. pro Rab. perd. 26 with T. R. S. Broughton, TAPA 1946, 35.

³³ H. B. Mattingly, ‘The date of the SC de agro Pergameno,’ AJP 1972, 412; we remain certain that Asia was taxed from 129 and that Appian, BC, v, 4 is worthless as evidence to the contrary.

³⁴ Strabo xiv, 642; the ambassador Artemidorus floruit 104–100, Geogr. Gr. Min. i, 566, 31.

³⁵ I. Priene III and 117; OGIS 440 = ILS 8870 = IGRR iv, 194; note also that 91 is a point of reference in the Lex Antonia de Termessibus with regard to land held by the city.

³⁶ Plutarch, Marius 31; Cicero, , ad Brut, i, 5, 3Google Scholar.

³⁷ Justin xxxviii, 1; Strabo xii, 540; M. Aemllius Scaurus perhaps went on the embassy involved, E. Badian, Athenaeum 1956, 120. Rome later instructed Sulla to place Ariobarzanes on the throne of Cappadocia, see E. Badian, Studies 157.

³⁸ OGIS 437, line 5 with commentary; compare the government of Sicily by (?) L. Asellio, Diodorus xxxvii, 8, and the activity of C. Claudius Pulcher, pr. 95.

³⁹ Michael Crawford hopes to return to the problem of cults of Roman magistrates during the Republic.

⁴⁰ Dio xxviii, fr. 97, 3, incidentally, provides evidence only for marginal involvement of Marius in the conviction, and is perhaps to be discounted altogether, in view of the silence of the other sources, notably Cicero (so, in part, E. S. Gruen, Historia 1966, 54–5).

⁴¹ Asconius 57C.

⁴² G. Colin, BCH 1924, 75–6.