Skip to main content Accessibility help

Civitas—A Myth?

  • Sheppard Frere

Mr J. C. MANN holds that ‘each civitas of Gaul (and of Britain) was presumably a city, and like all true cities, each will have consisted of a single physical city surrounded by its territorium, the whole under the control of its magistrates, council and assembly, the last no doubt usually of little importance.’ I, on the other hand, following Jullian and Stevens, have suggested that the civitas was the whole tribal area, its capital (caput) normally ranking merely as a vicus where the government happened to have its seat. This, Mr Mann suggests, is a myth, which he doubts whether a Roman lawyer would understand.

The evidence admittedly is not all that it might be; yet as the question is important for a correct understanding of the development of local government in Roman Britain (and elsewhere), I may venture to take it up once more, especially as Mr Mann’s note appears to me to be erroneous in one or two particulars, and to exhibit the triple weaknesses of a too narrowly legalistic outlook, an inadequate account of historical development, and a neglect of some of the relevant facts.

Hide All

1 Cf. BG, v, 12; Britanniae … incolitur … maritima pars ab eis qui … ex Belgio transierant qui omnes fere eis nominibus civitatum appellantur quibus orti ex civitatibus eo pervenerunt. Or ibid, v, 20; Trinobantes prope firmissima earum regionum civitas …

2 Pauly-Wissowa, RE, Suppl. 1, 300-17.

3 Vitruvius, 32, 1; cf. 19, 9; 23, 21, etc.

4 Seneca, Epist. Mor., XCI, 10, describing the great fire there in A.D. 65; civitas arsit opulenta. Tac, Hist. I, 19; 11, 92; IV, 2. This usage is admittedly the equivalent of πóλι, but it has little relevance to the question of the provincial civitates of the West.

5 pro Sestio, 42, 91. He is tracing the evolution of civic progress… . quodam tempore homines nondum neque naturali neque civili iure descripto fusi per agros ac dispersi vagarentur … Turn res ad communem utilitatem quas publicas appellamus, turn conventícula hominum, quae postea civitates nominatae sunt, turn domicilia coniuncta, quas urbis dicimus, invento et divino iure et humano moenibus saepserunt.

Contrast also de Repub., 26, 41; coniunctionem tectorum oppidum vel urbem appellaverunt with ibid., 25, 40, quid est autem civitas nisi multitudo hominum in quoddam vinculum redacta concordiae.

6 BG, I, 12.

7 Writing under Domitian: Lachmann, Gromatici veteres, I (1848), 35-6. The whole passage is notable at its date and deserves quotation. Prima enim condicio possidendi haec est ac per Italiam; ubi nullus ager est tributarius sed aut colonicus aut municipalis out alicuius castelli out conciliabuli out saltus privati. At si ad provincias respiciamus, habent agros colonicos eiusdem iuris, habent et colonicos qui sunt immunes, habent et colonicos stipendi aries: habent autem provinciae et municipales agros aut civitatium peregrinarum.

8 Contrast also, e.g., Hist, IV, 70; legiones in Mediomatricos sociam civitatem abscessere (the tribal area) with ibid. 1, 63, Divoduri (Mediomatricorum id oppidum est) quamquam omni comitate exceptos subitus pavor terruit, raptis repente armis ad caedem innoxiae civitatis … (the town); or Ann. in, 41, haud ferme ulla civitas intacta seminibus eius motus fuit, sed erupere primi Andecavi ac Turani (a tribal rebellion) with Hist. 1, 54, miserai civitas Lingonum vetere instituto dona legionibus (the tribal government).

9 Digest, 1, viii, 9. sacra loca ea sunt quae publice sunt dedicata, sive in civitate sunt sive in agro, where the contrast is between town and country; ibid., L, XV, 4, forma censitali cavetur ut agri sic in censum referantur. Nomen fundi cuiusque: et in qua civitate et in quo pago sit, where on the tax-return the address of each farm must be given, both its district and ‘county’. The two usages are distinct, and are closely parallel to those found in Tacitus, despite the fact that Ulpian is in all likelihood writing after the Constitutio Antoniniana.

10 See note 35.

11 de Repub., 1, 26, 41.

12 He uses the word oppidum, not civitas, in describing the cities of Gaul recovered by Julian, XX, iv, 1 (cf. XVI, iii, 3). For Britain compare XXVIII, iii, 2, in integrum restituit civitates et castra, with XXVIII, iii, 7, instaurabat urbes et praesidaria {ut diximus) castra where the civilian programmes are not explicitly identical.

13 Codex Theod., 1, 29, 6; XII, 1, 12, etc.

14 The earliest such usages appear on the milestones of the third century, e.g. C.I.L., XIII, 8953 Rennes, A Cavitate) R(edonum) L{eugae), A.D. 237; ibid. 8911 Limoges, U(emovicibus) XXXIII, A.D. 243; 9040, Rheims, C(ivitate) Rem(orum) L(eugae) IIII, AD. 265-70, etc. It is clear that Jullian was correct (Histoire de la Gaule, IV, 525 ff.) in seeing here the result of the Edict of Caracalla, whereby Roman citizenship was extended to all free inhabitants of the Empire, with consequent changes in the standing and significance of the towns.

15 IV, vi, 2. It is interesting to note the differences in this passage from its late Republican analogue in the Lex Iulia Municipalis (I.L.S. 6085, passim): municipiis, coloniis, praefecturis, foris, conciliabulis avium Romanorum. The latter is designed for communities of Roman citizens, the former adapted for the altered requirements of the Empire, namely the inclusion of the peregrine communities of the provinces.

16 Jullian, op. cit., IV, 32; Grenier, Manuel d’Archéologie Gallo-Romaine, I, 323; Hatt, Histoire de la Gaule Romaine, 91.

17 IV, i, II.

18 C.I.L., XIII, 221.

19 On Forum Segusiavorum see also Vittinghoff, Römisene Stadtrechtsformen zur Kaiserzeit, Zeitschrift der Savigny-Sliftung für Rechtsgesckichte, Romanistische Abteilung, (81) LXVIII (1951), 483.

20 On the difficulties of this see Vittinghoff, op. cit. (n. 19 above), p. 475, note 131.

21 See footnote 7 above.

22 Rivet, Town and Country in R.B., 65; J. E. Bogaers, Civitas en Stad van de Bataven en Canninefaten (Nijmegen, 1960), p. 18 ff. and note 225.

23 For the date and number of such coloniae in Gaul see Jullian, op. cit., IV, 261-2.

24 Arch., LIX, 120; Ephem. Epigr., IX, 1012; Haverfield, Romanization of Roman Britain, 59; id., Roman Occupation of Britain, 185. Cf. also the Wroxeter Forum inscription dedicated by the Civitas Cornoviorum. Atkinson, Excavations at Wroxeter 1923-7, Frontispiece and p. 177; JRS, XIV, 244; Richmond, Roman Britain (Britain in Pictures), 23. In both these cases it is the whole community, not the town alone, at work through its government. The formula of the Caerwent stone, though 3rd century in date, is parallel to that of Wroxeter under Hadrian, and this despite the fact that in Britain the tribal names as far as we know were not transferred to the towns to anything like the same extent (if at all) as in Gaul (p. 31). There would have been nothing at this date to inhibit the use of the name Venta in the formula except its non-correspondence with the political reality described.

25 Grenier long ago made this point ‘une ville qui n’est pas colonie n’est elle-meme qu’un vicus; un quartier, une rue, un faubourg de cette ville est également un vicus’, Manuel d’Archéologie, 11, ii, 695; cf. Pauly-Wissowa s.v. Vicus 2091-2.

26 Metz, C.I.L., XIII, 4301 (= I.L.S. 4818), possibly Nantes; ibid. 3106 (= I.L.S. 7051).

27 cf. the opinion of Hirschfeld, C.I.L., XIII, p. 443.

28 See Birley, Trans. E. Riding Antiquarian Soc, XXVIII, iii (reprinted as Hull Museum Pubi. 206), pp. 229-32; also more shortly in JBAA, VII (1942), 27-8; JRS, XXVIII (1938), 199.

29 Antiq. Journ., XL, 61-2; cf. JBAA, VII (1942), 8-13.

30 The early military vicus must have been elsewhere.

31 ANTIQUITY (1959), 67-8 : see also Past and Present, 16 (November, 1959), 4.

32 Of the organization of the latter a convenient account will be found in Stevenson, Roman Provincial Administration, Oxford 1939, ch. VI.

33 C.I.L., XIII, 1048, 2585 (=I.L.S. 7040, 7045).

34 Cf. Jullian, Histoire de la Gaule, IV, 336-8.

35 Cf. I.L.S. 7041, Marco Lucterio … civitas Cadurcorum … publice posuit; 7043, 7044; 7045, C. Sulpicii Galli … cui ordo … statuas publice ponendas decrevit.

36 ANTIQUITY (1947) 75; cf. also the case of Evreux, Gallia, I, ii, 192. For the as yet incompletely Romanized social structure of Gaul in the first century see Syme, Tacitus, Oxford 1958, 456-61.

37 Jullian, op. cit., IV, 71 : for the general question see also 326.

38 Dio, IX, 22

39 ibid., 23.

40 Is it possible that a tradition of this state of affairs in his sources underlies Dio’s repeated statement that two cities (in LXII, 7, 1 he calls them Roman cities) were destroyed by Boudicca? These must be the colonia of Camulodunum and the municipium of Verulamium: no mention of London, then doubtless only a vicus. This is so notably different from the account of Tacitus (Ann., XIV, 37) who emphasizes both London’s richness and her fate, that it calls for comment. Dio is restricting the word πóλις to the official foundations.

41 C. E. Stevens in J. M. Wallace-Hadrill and J. McManners, France: Government and Society (1957), pp. 22-3, where he shows that the ‘four kinship groups’ (coriae) of the pre-Roman tribe were perpetuated in the four archdeaconries, with a fifth at Périgueux itself, into which the Roman church, basing itself on the civil administration, divided the diocese.

42 Zur Geschichte der peregrinen Gemeinden in Pannonien, Historia, VI (1957), 490, 492, etc.

In the September number of A NTIQUITY (1960, 222) Mr J. C. Mann took Mr Sheppard Frere to task over the definition of the Gallic or Romano-British civitas. Here Mr Frere, Reader in the Archaeology of the Roman Provinces in the University of London, discusses this controversial issue.

Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

  • ISSN: 0003-598X
  • EISSN: 1745-1744
  • URL: /core/journals/antiquity
Please enter your name
Please enter a valid email address
Who would you like to send this to? *


Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed