¹ e.g. Jullian, C., Histoire de la Gaule (Paris, 1920–1926) iv, 332–41Google Scholar; Duval, P.-M., La vie quotidienne en Gaule (Paris, 1952 (rev. 1976)), 208Google Scholar; Thévenot, E., Les gallo-romains,⁴ (Paris, 1972), 20Google Scholar; and Ternes, C.-M., La vie quotidienne en Rhénanie romaine (Paris, 1972), 97 f.Google Scholar

² Jullian, loc. cit.

³ This conclusion must result from Jullian's own references and from any reading of the standard works on this subject, e.g. Mommsen, T. and Marquardt, J., Manuel des antiquités romaines i, Paris, 1889, 200–68Google Scholar; Abbott, F. and Johnson, A. C., Municipal administration in the Roman empire (Princeton, 1926), 57 ff.Google Scholar; Stevenson, G. H., Roman provincial administration² (Oxford, 1948), 170 ffGoogle Scholar. (Cf. Collingwood, R. G. and Myres, J. N. L., Roman Britain and the English settlements² (Oxford, 1937), 165Google Scholar, and Frere, S. S., Britannia (London, 1967), 206 f., for the similar influence of the Mediterranean model on Romano-British studies.)Google Scholar

⁴ Full colonies: Lyon, Nyon, Augst, Cologne, Xanten; promoted civitates: Convenae, Elusates, Lingones, Mediomatrici(?), Morini, Nemetes, Segusiavi, Sequani, Treveri, Vellavi, Viducasses (see Vittinghoff, F.. ‘Römische Stadtrechtsformen der Kaiserzeit’, Zeitschrift für Savigny-Stiftung, romanistische Abteilung 68 (1951), 480–85Google Scholar; Wolff, H., ‘Kriterien für latinische und römische Städte in Gallien und Germanien…’, BJ 176 (1976), 45, 55)Google Scholar. The Helvetian town of Avenches is sometimes seen as a full settlement colony (so Wolff, loc. cit.), but this seems disputable and the Helvetii are better added to the second list (see Reynolds, J., ‘Legal and constitutional problems’, in The civitas-capitals of Roman Britain (ed. Wacher, J. S., Leicester, 1966), 70 and n.Google Scholar; also the remarks of Sherwin-White, A. N.,The Roman citizenship² (Oxford, 1973), 370)Google Scholar. The Gallic colonies are now generally regarded as ‘Latin’ colonies, the equivalent of the Latin municipia found elsewhere in the western Empire under the early principate; their colonial status was a ‘local archaism’, but this did not affect the way in which they were run (Vittinghoff, art. cit., 449–54, 476; Sherwin-White, op. cit., 369).

⁵ A possible reference to such a ‘unilateral’ adoption of Roman ways is Strabo, iv. 4. 3: Αριστōκρατικαί δ' ᾖσαν αἱ πλείōυς τῶν πōλιτειῶν… νυνἱ δέ πρōσέχōυσι τōῑς τῶν ‘Ρωμαίων πρōστάγμασι τὀ πλέōν (discussed further below, p. 91 and note 11). Vittinghoff, art. cit. (note 4), 18, Reynolds, art. cit. (note 4), 70, and Sherwin-White, op. cit. (note 4), 364, all seem to assume the working of this process; also Stevenson, op. cit. (note 3), 161.

⁶ Citizenship was derived from origo (Vittinghoff, art. cit. (note 4), 438 f.). For the assumption of such a dual form of government see ibid. 481 (among the Treveri); cf. Sherwin-White, op. cit. (note 4), 368–71.

⁷ Wolff, art. cit. (note 4), 49–54; cf. Vittinghoff, art. cit. (note 4), 484 – the evidence for the colonial status of the Sequani.

⁸ See Mann, J. C., ‘Civitas, another myth’, Antiquity xxxiv (1960), 222 f.CrossRefGoogle Scholar; idem, ‘Civitas – a further comment’, Antiquity xxxv (1961), 142 f.; idem, ‘City foundations in Gaul and Britain’, in Jarrett, M. and Dobson, B. edd., Britain and Rome (Kendal, 1965), £109–13.Google Scholar

⁹ I follow here the arguments of Frere, S. S., ‘Civitas–a. myth?’, Antiquity xxxv (1961), 29–37.CrossRefGoogle Scholar

¹⁰ e.g. I envisage that civitas – decurions who had their principal country residences far away from the new colonia would still rank as colonial decurions and would still be expected to undertake magistracies which involved the administration of the former civitas-capital as well as that of the tribal lands. In coming to this conclusion I have been especially impressed by the arguments of Wolff, art. cit. (note 4), 103–16, idem, ‘Civitas und Colonia Treverorum’, Historia 26 (1977), 213–40. However, I have not followed him in his view that a complete civitas could be officially regarded as a colonia; Roman tradition would have demanded at least the ‘cosmetic’ application of the term to a specific urban centre, even if in practice the rest of the civitas was not thereby reduced to the level of a subject territorium. (See below, p. 93, for a praefectus coloniae whose junior position must indicate his authority within a town, not a complete civitas.)

¹¹ It seems to me that the greatest weakness in Wolff's case is his interpretation of Strabo iv. 4. 3 (above note 5; Wolff, art. cit. (note 4), 51 f., 56–61). It is far from certain that Strabo was remarking on constitutional changes, as opposed to the stark realities of military conquest (cf. iv. 4. 2: νυνι μἐν ōὖν ἐν εἰρήνη πἁντες εἰσἰ δεδōυλωμένōι καἰ ζῶντες κατὰ τὰ πρōστὰγματαα τῶν ἑλόντων αὐτōὐς ‘ΡωμαІων…) Further, even assuming that Strabo was here describing the introduction of Roman-style magistracies into Gallic local government, the notion that such a reform could not have taken place without some form of imperial approval implies the occurrence of a far-reaching official promotion of most, if not all, civitates under Augustus or Tiberius. Unfortunately, however, there is no record of such an outstanding act of generosity in our other sources, and moreover epigraphic evidence reveals the continuing survival of certain indigenous titles (yergobret, princeps etc.) which could not have fitted into any strictly ‘municipal’ structure. The only way round the problem was to develop von Braunert's theory (‘Ius Latii’, Corolla memoriae E. Swoboda (Graz, 1966), 68–83), also recently adopted by B. Galsterer-Kroll, ‘Zum ius Latii in den keltischen Provinzen…’Google Scholar, Chiron 3 (1973), 277–306Google Scholar, and to assume that all that was allowed to the Gallic tribes was general permission to take on Latin rights, and hence a Latin constitution, but with no obligation so to do, and even with no obligation, having done so, to change the title of the community to colonia. This much looser grant of Latin privileges went unnoticed by the historians, and allowed a certain variation in the constitutional development of the Gauls, some progressing more speedily than others. The whole notion of personal citizenship being distinct from communal status has, however, already been severely criticized by Sherwin-White, op. cit. (note 4), 360–67.

¹² CIL xiii 1390 (Lemovices), 2899 (Senones).

¹³ e.g. CIL xiii 1910, 1911, 1920,1922, 1925, 3693, 6225, 6244, 6339, 6384, 6404, 6462, 6467, 6482, 6705, 6733, 6769, 7062, 7062(a), 7063, 7064, 7266, 7321, 7352, 7386, 8853, 11690, 11696, 11810, 12013; Finke 3, 183, 306; Nesselhauf 106.

¹⁴ P. 95. A complete collection of the available material is now to hand in G. Rupprecht, Untersuchungen zum Dekurionstand in den nordwestlichen Provinzen des römischen Reiches (= Frankfurter althistorischen Studien 8), Kallmünz, 1975.

¹⁵ Two further examples would be provided by CIL xiii 548, and 2585, if IIVIR.Ǭ could safely be read as Ilvir, q(uaestor).

¹⁶ The duality of the office of vergobret is disputed–see Wolff, art. cit. (note 4), 51 and n. 12. It seems to me, however, that whatever else can be made of Strabo iv. 4. 3 (above, notes 5 and 11), he at least implies the demise of ‘dictatorial’ magistracies under the Romans: ἑνα δἡγεμόνα ᾑρōῦντō κατ'ἐντòυ τὸ παλαιόν …νυνІδὲ… Cf. Jullian, op. cit. (note 1), 337 and n. 4.

¹⁷ The interpretation of this inscription is not helped by the fact that it is written in verse – see below, note 29.

¹⁸ Single references to quaestors and aediles are also found in CIL xiii 4291Google Scholar, 5415, 5682, 7370, 11553, Nesselhauf 77, AE 1953, 56. The pronounced eastern disposition of these inscriptions (Mediomatrici, Sequani, Lingones, Taunenses, Vangiones, Tricasses) is very striking, perhaps suggesting the influence of the highly Romanized Rhineland – cf the case of the decurions, discussed above.

¹⁹ The Mediterranean evidence reveals that the quaestorship was prone to a somewhat haphazard occurrence, but certainly not the aedileship – Mommsen and Marquardt, op. cit. (note 3), 234.

²⁰ Except, of course, in the case of the full Roman colonies – see, for example, the magistrates of Lyon. Galsterer-Kroll, art. cit. (note 11), 283–305, has argued that the patchy and incomplete local cursus of the Celtic west reflect the existence of widespread and hitherto unrecognized Latin coloniae, in that the ius Latium, by automatically making Roman citizens of all who held any sort of post in a municipality, encouraged the under-taking of short careers: peregrine aristocrats would hold a position or two, receive their citizenship, and leave. The faults in this argumentation have been pointed out by Wolff, art. cit. (note 4), 62–76. As he says, and as will be discussed below, the common phrase omnibus honoribus apud suos functus clearly shows that men followed longer careers, details of which have been subsumed in a characterless form of words. In any case, the non-hierarchical system which Galsterer-Kroll postulates would be no way to run a civitas.

²¹ op. cit. (note 1), 341 f.

²² Reflected most directly as far as local communities were concerned in the development of the office of curator – see Jones, A. H. M., The Later Roman Empire (Oxford, 1964), i, 11 ff.Google Scholar; and below, p. 97.

²³ cf. Jullian, op. cit. (note 1), 342 and n.

²⁴ ibid., 352 and n., dismisses the position of magister pagi as a ‘nomination sur place’.

²⁵ i.e. in the careers of Verus, Hannarus and the anonymous Burdigalan – above p. 92.

²⁶ Despite the similarity of title I would argue (I hope not too perversely) that C. Iulius…'s prefecture cannot be equated with the post of praefectus municipii, ‘deputy duumvir’, as laid down, for example, in the Lex municipii Salpensani (FIRA 24, xxv; cf. 21, xciiii). I base my reasoning upon the obviously junior nature of his appointment, and on analogy with the first office held by Amatius Paterninus. Cf. Galsterer-Kroll, art. cit. (note 11) 283 (for the normal high standing of praefecti pro Ilviris); also Jullian, op, cit. (note 1), 352 and note.

²⁷ On the disappearance of the title of colonia in the later period see Frere, art. cit. (note 9), 30 f.

²⁸ C. Amatius Paterninus's career has aroused a good deal of discussion in the controversy over the definition of civitas – above, p. 90 and note 8.

²⁹ The mention of the multiple magistracies of pagi to be found in at least one (CIL xiii 5) of the earlier inscriptions possibly hints at a similar practice in the first century, i.e. that an individual was not magister pagi in successive years, but in the intervals between his civitas-appointments. (I would myself be inclined to read Q as quater or some such, in line 3 of CIL xiii 412Google Scholar, were it not for the fact that this would cause difficulties with the supposed metre.)

³⁰ See below (the case of L. Campanius Priscus and T. Flavius Postuminius).

³¹ There seems to have been no set title for this office. Strictly speaking, all local imperial priests should have called themselves flamines, the title of sacerdos being reserved for the high priest of the federal Altar. It could be, however, such was the prestige attached to the latter office that its name was ‘poached’ by local dignitaries. This has led to some confusion between priests and high priests, but it can safely be said that unless the context makes their position perfectly plain, e.g. a dedication at Condate itself, the latter can be easily distinguished by their overwhelming desire to demonstrate the interprovincial nature of their office – sacerdotes ad Aram, Arenses, inter Confluentes, ad Templum etc. The vast majority of ‘plain’ sacerdotes must be local. (For a general lack of uniformity throughout the empire see D. Ladage, Städtische Priester- und Kultamter in lateinischen Westen des Imperium Romanum zur Kaiserzeit (Cologne diss., 1971), 41 ff.; (H. Wolff, art. cit. (note 10), 207, may have a point in distinguishing a provincial sacerdos Romae et Augusti from the usual flamines of the Treveri, but in view of the above I remain not wholly convinced by this). I plan to publish a more detailed discussion of this question in due course. The office of flamen brings us back to the problem of status since, as in the case of quaestors, aediles and duumvirs, it may be argued that this position could only be held in coloniae or municipia, or in communities granted colonial or municipal privileges (see above, p. 90 and note 7). In fact there is about as much firm evidence for flamines in civitates which are not known for sure to have had colonial status (Aedui: CIL xiii 2585Google Scholar; Bituriges Cubi: 1376/7; ?Meldi: 3024; Senones: ?1684, 2940; ?Sequani: 1674) as there is for them in those which are (Convenae: CIL xiii 548Google Scholar; Helvetii: 5009, 5010, 5063, 5012; Segusiavi: 1629; Treveri: 4030). Once again it could be argued that here we have evidence either for hidden grants of municipal status, wholesale or piecemeal, or for plain copying of Roman ways; but once more, whatever the explanation, the final outcome seems to have been some sort of uniformity in local institutions.

³² A possible exception to the rule is provided by CIL xiii 1376/7Google Scholar, which have been restored to show the two sons of L. Julius Equester as flamines Romae et Augusti with no indication of their having been duumviri. Either the restoration is wrong, or one must accept a certain degree of flexibility in the system – their father had twice been duumvir and so might well have expected a certain amount of leeway in the premature promotion of his sons.

³³ AE 1969/70, 405.

³⁴ For the title see above, note 31. If I am correct, then the Gallic situation would provide a remarkable contrast with the Spanish where, according to Etienne, it was extremely unusual for provincial priests to have been municipal flamines or even sacerdotes of conventus. Indeed, there was a sizeable majority of Iberian provincial priests who had held no local offices at all – ‘qui ne devaient qu'à leurs qualités personelles et à leur position sociale d'être délégués à Merida, Cordoue et Tarragone, et surtout d'être élus' (Etienne, R., Le culte impériale dans la péninsule lbérique (Paris, 1958), 158Google Scholar; I owe this reference to Professor S. S. Frere).

³⁵ Sollemnis's career has been examined in detail by Pflaum, H.-G., Le marbre de Thorigny (Paris, 1948), passim.Google Scholar

³⁶ cf. Jullian, op. cit. (note 1), 364 and n.

³⁷ See J. Deininger, Die Provinziallandtage der römischen Kaiserzeit ( = Vestigia 6) (Munich, 1965), 102 ff.

³⁸ CIL's designation of L. Besius Superior as an eques has since been refuted by Rougé, ‘Les rapports de Lyon avec l'ouest-nord-ouest Gaulois’, Revue archéologique de l'Est 25 (1974), 139; details of the career of Q. Otacilius Pollinus are based on the new reading of AE 1972, 352.

³⁹ Despite Magnus's curiously archaic filiation the latter inscription must date to the third century, when the Cenabenses were divided off from the Carnutes proper – see Moreau, J., Dictionnaire de géographie historique (Paris, 1972), ad. loc.Google Scholar

⁴⁰ For the emergence of curators see Mommsen and Marquardt, op. cit. (note 3), 225 f.; Ruggiero, E. de, Dizionario epigrafico (Spoleto, 1910)Google Scholar (s.v. curator reipublicae); Abbott and Johnson, op. cit. (note 3), 90 f.

⁴¹ The technicalities of the post are discussed in Audin, A., Guey, J., Wuilleumier, P., ‘Inscriptions latines découvertes à Lyon’, REA 56 (1954)CrossRefGoogle Scholar, which examines the inscription mentioned below.

⁴² See G. Alföldy, Die Hilfstruppen der römischen Provinz Germania Inferior ( = Epigraphische Studien 6) (Dusseldorf, 1968), 81 ff., 115 f. cf. my article, Latomus 37 (1978), 817–50.Google Scholar

⁴³ Alföldy, op. cit. (note 42), 100, 121; Birley, E., Roman Britain and the Roman Army (Kendal, 1961), 142.Google Scholar

⁴⁴ Alföldy, op. cit. (note 42), 121.

⁴⁶ For the Gallic warrior tradition see Caesar, BG vi, 15. 1: … omnes in bello versantur; the necessary qualification of omnes follows shortly after (vi. 31.5) with the observation that Catuvolcus, joint-king of the Eburones, committed suicide because his age prevented him from fighting.

⁴⁶ For a somewhat different reading of the last inscription see Wolff, art. cit. (note 10), 206 f.

⁴⁷ Alföldy, op. cit. (note 42), 100 ff., 116, 121, 129.

⁴⁸ See Krier, J. and Schwinden, L., ‘Die Merscher Inschrift’, TZ 37 (1974), 123–47Google Scholar and here especially 139 f.

⁴⁹ Krier and Schwinden, art. cit. (note 48), 146 f., isolate seven men with careers similar to that of the Treveran; five of these went on to higher equestrian posts. (Of the remaining, one became flamen of Hispania Citerior; the subsequent career of the other is unknown). An even odder career is that of C. Iulius Serenus of the Convenae – quattuorvir, sacerdos Romae et Augusti, praefectus alae VII Phrygum, c. A.D. 100 (ILTG 76–80). Serenus's command of an auxiliary cavalry regiment apparently followed a series of local appointments, but was not preceded by a posting to the prefecture of a cohort, nor (so far as we know) was it succeeded by any further equestrian office. The positively ‘Julio–Claudian’ neglect of the tres militiae, coupled with the fact that the VII Phrygum was a Syrian ala, (involvement in which would have taken Serenus far from home), make Serenus's public life very difficult to explain in a second-century context. (Sapène, B., ‘Caius Julius Serenus, personnage de Lugdunum Convenarum vers l'an 100’, Revue de Comminges 64 (1951)Google Scholar, which may well explain these anomalies, was unfortunately unavailable to me in the preparation of this paper.)

⁵⁰ E. Birley, op. cit. (note 43), 138 ff.

⁵¹ ibid. 141.

⁵² RE xii 1572–86; 1551Google Scholar.

⁶³ cf. above, p. 96. In the early (Julio-Claudian) period of the Altar's history it may well have been that such men were the main recruits as federal officials – before curatores and those with more commercial interests came on the scene. Acceptance of the existence of such tribunes as a matter of course would explain Tacitus's casual reference to tribuni in Gallia geniti (Hist. iv. 61) in the context of A.D. 70. Cf. Fabia, P., ‘Officiers gaulois dans les légions romaines’, REA 14 (1912), 285 f.CrossRefGoogle Scholar

⁵⁴ For the development of the office see RE xxii 1323–26.Google Scholar

⁵⁵ cf. RE xxii 1289.Google Scholar

⁵⁶ E. Birley, op. cit. (note 43), 138.

⁵⁷ cf. CIL vi 1641Google Scholar, with Pflaum, H.-G., ‘La monnaie de Trèves à l'époque des empereurs gallo-romains’, Congrès Internationale de numismatique (Paris, 1953), Actes (1957), 273–80 (concerning the unorthodox career of a postmaster and mintmaster, apparently under the Gallic Empire)Google Scholar.

I am grateful to the Sheffield University Research Fund for financial help towards the preparation of this paper.