¹ e.g. Braun, J. W. J., Die Kapitole (1849)Google Scholar; Castan, A., Le Capitole de Vesontio et les capitoles provinciaux du monde romain (1869)Google Scholar; Les Capitoles provinciaux du monde romain (1886)Google Scholar; Kuhfeldt, O., De Capitoliis Imperii Romani (1883)Google Scholar; Toutain, J., ‘Étude sur les capitoles provinciaux de l'empire romain’, École pratique des hautes études (1899)Google Scholar; Les Cultes païens dans l'empire romain vol. 1 (1907), 181–93Google Scholar; Cagnat, R. and Chapot, V., Manuel d'archéologie romaine I (1916), 157–60Google Scholar; Paris, P. et al. , Fouilles de Belo (Bolonia, Province de Cadiz) (1917–1921) (1923), 65–70Google Scholar; Bianchi, U., ‘Disegno storico del culto capitolino nell'Italia romana e nelle provincie dell'impero’, Memorie della Classe di Scienze morali e storiche dell' Accademia dei Lincei 82 (1950), 349–415Google Scholar; Barton, I. M., ‘Capitoline temples in Italy and the provinces, especially Africa’, ANRW II.12.1 (1982), 259–342Google Scholar; Todd, M., ‘Forum and Capitolium in the early Empire’, in Grew, F. O. and Hobley, B. (eds), Roman Urban Topography in Britain and the Western Empire (1985), 56–66Google Scholar; Zanker, P., ‘The city as symbol: Rome and the creation of an urban image’, in Fentress, E. W. B. (ed.), Romanization and the City: Creation, Transformations, and Failures (2000), 25–41Google Scholar; MacMullen, R., Romanization in the Time of Augustus (2000)Google Scholar, 60; Brill's New Pauly, Encyclopedia of the Ancient World (2003) s.v. Capitolium 1071–73Google Scholar.

² Aulus Gellius, Noctes Atticae 16.13.9.

³ An error noted also by Kuhfeldt, op. cit. (n. 1), 13 n. 18; de Azvedo, M. Cagiano, ‘I “Capitolia” dell'impero Romano’, in Atti della Pontifica Accademia Romania di Archaeologia. Ser. III, Memorie 5 (1941), 63 n. 426Google Scholar; Bianchi, op. cit. (n. 1), 350 and Barton, op. cit. (n. 1), 278 n. 46; although many of these works still assume that Gellius implies a physical resemblance between Rome and its colonies.

⁴ Braun, op. cit. (n. 1), 3; Daremberg-Saglio Dict. II.906 s.v. Capitolium; Janin, R., Constantinople Byzantine. Développement urbain et répertoire topographique (1964), 174–5Google Scholar.

⁵ S. Maffei, Verona Illustrata (repr. 1825), 210.

⁶ Du Cange, C. D. F., Glossarium mediae at infimae Latinitatis (1737)Google Scholar, s.v. Capitolium (sense 2); cf. Castan, op. cit. (n. 1, 1869), 15–16; Daremberg-Saglio Dict. I.2.906 s.v. Capitolium. Similarly, the scholia on Statius' Thebaid seem to call the Parthenon the ‘Capitolium Minervae’: ad 4.136: ‘Ilium Palladia] Hippomedontem dicit, qui ab Athenis, id est, a Capitolio Minervae ad bella descendit’ (‘Ilium Palladia] he means Hippomedon, who came down from Athens, that is, from the Capitolium of Minerva, to the wars’). We thank JRS's anonymous reader for this reference and suggestion.

⁷ ad 14.22: ‘arx autum, id est, Capitolium illius Urbis, est turris quae aedificata post diluvium, in altitudine quatuor millia dicitur tenere passuum … describunt ibi templa marmorea, aureas statuas, plateas lapidibus auroque fulgentes, et multa alia quae pene videantur incredibilia.’

⁸ Braun, op. cit. (n. 1); as Kuhfeldt (op. cit. (n. 1), 13 and n. 18) pointed out, of its 32 pages, only pp. 12–22 are relevant and the rest are ‘ballast’.

⁹ Castan, op. cit. (n. 1, 1869), with summary of identified Capitolia on p. 28. The identification at Vesontio is made on the basis of a medieval toponym.

¹⁰ Kuhfeldt, op. cit. (n. 1), 14 n. 20.

¹¹ Castan, op. cit. (n. 1, 1869), 30: ‘Les Capitoles provinciaux paraissent résulter de concessions gracieuses du gouvernement impérial, et cette nature de faveurs semblerait avoir exclusivement le lot des colonies, c'est-à-dire des villes admises à jouir de la plénitude des institutions romaines. En effet, sur les vingt-quatre Capitoles que nous avons reconnus, vingt-trois appartiennet à des localités soumises au droit colonique; et quant à celui de Marruvium Marsorum, qui pourrait faire exception, rien ne prouve que l'ancien chef-lieu des Marses n'a pas été, à un moment de son existence, repeuplé par quelque corps de vétérans des armées romaines.’

¹² Castan, op. cit. (n. 1, 1869), 29.

¹³ Kuhfeldt, op. cit. (n. 1), 7.

¹⁴ Kuhfeldt, op. cit. (n. 1), 8–9; Maffei, op. cit. (n. 5), 210; cf. Daremberg-Saglio, Dict. I.2.906 s.v. Capitolium; Grenier, A., ‘Les capitoles romains en Gaule et le Capitole de Narbonne’, CRAI (1956), 316–17Google Scholar; Gayraud, M., Narbonne antique des origins à la fin de IIIe siècle (1981), 270Google Scholar.

¹⁵ Cyprian, Ep. 55: ‘Quid superest, quam ut Ecclesia Capitolio cedat, et recedentibus sacerdotibus ac Domini altare removentibus, in cleri nostri sacrum venerandumque consessum, simulacra atque idola cum aris suis transeant?’ (‘What remains, but that the Church should yield to the Capitol, and that with the priests withdrawing and removing the altar of the Lord, the images and idols with their altars should pass into the sacred and venerable assembly of our clergy?’)

¹⁶ Coxe, A. Cleveland, The Ante-Nicene Fathers, Translations of the Writings of the Fathers down to A.D. 325, vol. 3 (1885–1897)Google Scholar.

¹⁷ Useful discussions along these lines in Maffei, op. cit. (n. 5), 209–10; Castan, op. cit. (n. 1, 1869), 23–8; Toutain, op. cit. (n. 1, 1899), 3–4; op. cit. (n. 1, 1907), 183–4; Grenier, op. cit. (n. 14), 316–17; Gayraud, op. cit. (n. 14), 270. For a partial rehabilitation of the evidence of the Acta in response to Castan and Kuhfeldt, see Allard, P., ‘Les Capitoles provinciaux et les Actes des Martyrs’, La Science Catholique, Revue des questions religieuses 1 (1887), 358–76Google Scholar.

¹⁸ Kuhfeldt, op. cit. (n. 1), 10.

¹⁹ Kuhfeldt, op. cit. (n. 1), 41: ‘Neque fallemur cum imperatorum Rom. certe temporibus vix ullam totius Italiae coloniam municipiumve fuisse contendemus, quod non Iovis Iunonis Minervae templo uteretur.’

²⁰ Kuhfeldt, op. cit. (n. 1), 56–7; cf. the review by Seeck, O. in Wochenschrift für klassische Philologie 2 (1884), 37–8Google Scholar.

²¹ Kuhfeldt, op. cit. (n. 1), 78 and n. 297.

²² Kuhfeldt, op. cit. (n. 1), 79.

²³ Castan, op. cit. (n. 1, 1886).

²⁴ Aust, E., ‘Iuppiter’, in Roscher, W. H. (ed.), Ausführliches Lexicon der griechischen und römischen Mythologie II (1890–1897), cols 739–43Google Scholar.

²⁵ de Ruggiero, E., ‘Capitolium’, in Dizionario epigrafico di antichità romane, vol. II.1 (1892), 93–5Google Scholar.

²⁶ Daremberg-Saglio, Dict. II.906 s.v. Capitolium.

²⁷ Cagnat, R. and Gauckler, P., Les monuments historiques de la Tunisie (1899), 1–18Google Scholar.

²⁸ Toutain, op. cit. (n. 1, 1899), I.181–93.

²⁹ Toutain, op. cit. (n. 1, 1899), 20–9; op. cit. (n. 1, 1907), I.187–8.

³⁰ Toutain, op. cit. (n. 1, 1899), 19.

³¹ Toutain, op. cit. (n. 1, 1907), 186–8; quote from pp. 186–7.

³² Toutain, op. cit. (n. 1, 1907), 190–3.

³³ Cagnat and Chapot, op. cit. (n. 1), I.157–60 (quote from pp. 157–8).

³⁴ Bianchi, op. cit. (n. 1), 399, emphasis in original; this classic passage is also cited at Barton, op. cit. (n. 1), 260.

³⁵ An early example: MacKendrick, P., ‘Roman town planning’, Archaeology 9.2 (1956), 126Google Scholar.

³⁶ Brown, F., ‘Cosa I: history and topography’, MAAR 20 (1951), 63Google Scholar.

³⁷ e.g. Brown, F., Cosa: the Making of a Roman Town (1980), 53–6Google Scholar.

³⁸ Grimal, P., Les Villes romaines (1954)Google Scholar, 60 (and repeated in the 1983 English translation by Woloch, G. M., Roman Cities (1983)Google Scholar, 50; cf. pp. 50–1.)

³⁹ Todd, op. cit. (n. 1); Gros, P., ‘Sanctuaires traditionnels, capitoles et temples dynastiques: rupture et continuité dans le fonctionnement et l'aménagement des centres religieux urbains’, in Los Asentamientos ibéricos ante la Romanización (1987), 111–21Google Scholar; É. Blutstein-Latrémolière, ‘Les places capitolines d'Espagne’, Mélanges de la Casa de Velázquez 27.1 (1991), 43–64CrossRefGoogle Scholar. Likewise, the review of comparanda for the architectural form of Capitolia by Manasse, G. Cavalieri, ‘La tipologia architettonica’, in L'area del Capitolium di Verona: recherché storiche e archeologiche (2008), 309–15Google Scholar, rests on identifications which are in many cases uncertain.

⁴⁰ Gargola, D. J., Land, Laws and Gods. Magistrates and Ceremony in the Regulation of Public Lands in Republican Rome (1995), 83Google Scholar; Mouritsen, H., Italian Unification: a Study in Ancient and Modern Historiography (1998), 76Google Scholar is more cautious.

⁴¹ OCD ³, s.v. Capitol, Capitolium. For the temple itself see LTUR s.v. Iuppiter Optimus Maximus Capitolinus, aedes, templum.

⁴² Todd, op. cit. (n. 1), 57.

⁴³ Brill's New Pauly, 1073: ‘It is probable that capitolia were originally erected in those Roman colonies laid out on the pattern of Rome (coloniae B; cf. Suet. Tib. 40,1: Capua; Vitr. De arch. 3,2: Pompeii), then in cities that wished or were obliged in particular to emphasise their adherence to the empire (e.g. foundation of the Colonia Aelia Capitolina by Hadrianus in Jerusalem, with a temple to Capitoline Jupiter on the Jewish Temple Mount). … The capitolium was as a rule situated on the forum’. Stirling, L. M., ‘Art, architecture and archaeology in the Roman Empire’, in Potter, D. S. (ed.), A Companion to the Roman Empire (2006), 80Google Scholar: ‘The advent of Roman power often meant the addition of a capitolium, a temple to the tutelary deities of the city of Rome: Jupiter, Juno and Minerva, as worshipped on the Capitoline hill. Provincial capitolia were modeled on the one in Rome, using an imposing podium to elevate the temple and emphasize its frontal aspect over all the others. Three cult chambers or niches housed the three divinities. In the forum of Gorsium in Pannonia, created under Trajan, a head of Jupiter was found in the central chamber (Fitz, J. and Fedak, J., ‘From Roman Gorsium to late-antique Herculia: a summary of recent work at Tác (NE Pannonia)’, JRA 6 (1993), 261–73Google Scholar)'.

⁴⁴ Kuhfeldt, op. cit. (n. 1); Toutain, op. cit. (n. 1, 1907).

⁴⁵ For a recent evocation of the passage in relation to the reproduction of the plan of the Roman temple in ‘Capitolia’ elsewhere, Cavalieri Manasse, op. cit. (n. 39), 308; for the same point in relation to Cologne see Hellenkemper, H., ‘Architektur als Beitrag zur Geschichte der Colonia Ara Agrippinensium’, in ANRW II.4 (1975), 808Google Scholar; and for explicit reliance on Gellius for the connection between Capitolia and colonies see, for example, Brown, F. et al. , Cosa II: The Temples of the Arx (1960), 103–6Google Scholar (cf. n. 92 below). Against this interpretation of the passage, see Zanker, op. cit. (n. 1), 41, and more generally against this passage, Bispham, E., ‘Mimic? A case study in early Roman colonisation’, in Herring, E. and Lomas, K. (eds), The Emergence of State Identities in Italy in the First Millennium BC (2000), 157Google Scholar: ‘Gellius’ account of, and commentary on, Hadrian's words have had a pernicious influence on the study of local government and autonomy in the Roman Empire.' More recently, Ando too has criticized the tendency to ‘rely heavily on the heuristic value of a single imperial text, by which to interpret the abundant but tessellated evidence from the middle and late republic’ (Ando, C., ‘Exporting Roman religion’, in Rüpke, J. (ed.), A Companion to Roman Religion (2007), 431Google Scholar), and drawn attention to the mismatch between Gellius' account of colonies as having ‘all the laws and institutions of the Roman people’ and Republican evidence such as Cicero's criticism of the colonists at Capua for trying to ape Roman institutions (de leg. agr. 2.93).

⁴⁶ In addition to the examples already cited, see for instance Pohl's claim that in colonial contexts a Capitolium is ‘an essential building to indicate the might and power of Rome’ (Pohl, I., ‘Was early Ostia a colony or a fort?’, La Parola del Passato: rivista di studi antichi 38 (1983), 124Google Scholar); Macdonald's assertion that ‘Western cities required capitolia … though they varied greatly in size and detail, it seems they were always one-ended, with porch and steps facing a plaza, usually the forum’ (Macdonald, W. L., The Architecture of the Roman Empire, Volume II: an Urban Appraisal (1986), 119Google Scholar); and Zanker's explanation that the ‘capitolia (religious centers)’ of Roman citizen colonies ‘were laid out on the axis of each forum so that overland traffic had to pass in front of them, a principle that made them very visible as symbols of the colonies’ association with Rome and submission to Roman sovereignty' (Zanker, P., Pompeii. Public and Private Life (1998), 7Google Scholar).

⁴⁷ Zanker, op. cit. (n. 1), 27–8.

⁴⁸ Zanker, op. cit. (n. 1), 28.

⁴⁹ Fentress, E. W. B., ‘Frank Brown, Cosa and the idea of a Roman city’, in Fentress, E. W. B. (ed.), Romanization and the City: Creation, Transformations and Failures (2000), 11–24Google Scholar; Bispham, op. cit. (n. 45); see also, more succinctly, Torelli, M., Tota Italia: Essays in the Cultural Formation of Roman Italy (1999), 15–16Google Scholar.

⁵⁰ Bispham, E., ‘Coloniam deducere: how Roman was Roman colonization during the Middle Republic?’ in Bradley, G. and Wilson, J.-P. (eds), Greek and Roman Colonization. Origins, Ideologies and Interactions (2006), 75Google Scholar, building on Crawford, M. H., ‘La storia della colonizzazione romana secondo i romani’, in Marino, A. Storchi (ed.), L'Incidenza dell'antico. Studi in memoria di Ettore Lepore, vol. I (1995), 187–92Google Scholar, where the definition of a colony during the Republic is seen as unfixed. See in addition Woolf, G. D., ‘Catastrophe and aftermath’, in Sweetman, R. (ed.), Roman Colonies in the First Century of their Foundation (2011), 150–9Google Scholar, also questioning the category of ‘colonization’.

⁵¹ Fentress, op. cit. (n. 49), 14.

⁵² Bispham, op. cit. (n. 45), 112; cf. Bispham, op. cit. (n. 50), 158.

⁵³ Bispham, op. cit. (n. 50), 100, 112.

⁵⁴ Beard, M., North, J. and Price, S., Religions of Rome, vol. I (1998), 333–4Google Scholar. Walbank makes a similar point in relation to pre-Augustan colonies outside Italy, that ‘we do not know whether the building of a Capitolium accompanied the original foundation of a colony’ (Walbank, M., ‘Pausanias, Octavia and Temple E at Corinth’, Annual of the British School at Athens 84 (1989), 381CrossRefGoogle Scholar).

⁵⁵ Bispham, op. cit. (n. 50), 118.

⁵⁶ Bispham, op. cit. (n. 50), 122.

⁵⁷ Beard, North and Price, op. cit. (n. 54), 335.

⁵⁸ Van Andringa, W., ‘Religions and the integration of cities in the Empire in the second century AD: the creation of a common religious language’, in Rüpke, J. (ed.), A Companion to Roman Religion (2007), 89Google Scholar. This chronology is impossible; the chronology of the temple on the forum at Sabratha cannot be so closely linked to the award of civic status, and in fact it is clear that the Capitolium at Cuicul is substantially later (see below, n. 60).

⁵⁹ Toutain, op. cit. (n. 1, 1899); op. cit. (n. 1, 1907).

⁶⁰ It clearly post-dates the macellum, one of whose entrances it blocks, an observation for which we thank Roger Wilson.

⁶¹ A structure adopted for ease of comparison with Barton's work and indeed with earlier studies which employ a similar approach.

⁶² ‘Kapitolium’ is also found: the spelling with K is simply the result of a widespread ancient view that K should be used before A, Q before U, and C before I, E and O — a means of regularizing the orthographical treatment of three letters with essentially the same phonetic value (cf. Quintilian 1.7.10).

⁶³ Varro LL 5.158: ‘cliuus proximus a Flora susus versus Capitolium Vetus, quod ibi sacellum Iovis Iunonis Minervae, et id antiquius quam aedis quae in Capitolio facta.’ See also Martial 5.22.4, 7.73.4, and Coarelli, F., ‘Capitolium Vetus’, in Steinby, E. M. (ed.), Lexicon Topographicum Urbis Romae, vol. I (1993), 234Google Scholar.

⁶⁴ For the Capitolium in Rome see Tagliamonte, G., ‘Capitolium (fino alla prima età repubblicana)’, in Steinby, E. M. (ed.), Lexicon Topographicum Urbis Romae, vol. I (1993), 226–31Google Scholar. For recent discussions of the building itself, with earlier bibliography, see Sommella, A. Mura, ‘Il tempio di Giove Capitolino: una nuova proposta di lettura’, Annali della Fondazione per il museo ‘Claudio Faina’ (2009), 333–72Google Scholar, and Arata, F. P., ‘Nuove considerazioni a proposito del Tempio di Giove Capitolino’, Mélanges de l'École française de Rome, Antiquité 122.2 (2010), 585–624CrossRefGoogle Scholar (who underscores the variety of reconstructions that have been proposed (608–22), rejects them all, and proposes, but does not illustrate, a new one of his own, arguing for a smaller temple than other scholars have assumed).

⁶⁵ We thank one of JRS's anonymous readers for this suggestion.

⁶⁶ Plutarch, Publicola, 15.1–4; Tacitus, Histories 3.72; 4.53.

⁶⁷ On the textual and iconographic sources, see Arata, op. cit. (n. 64), 586–92.

⁶⁸ Dion. Hal. 4.61.

⁶⁹ Barton, op. cit. (n. 1), with earlier bibliography.

⁷⁰ For this arrangement, see Livy 7.3.5.

⁷¹ Barton, op. cit. (n. 1), 260–1.

⁷² Aventine: Livy 5.21–3, 31, 52 (396 b.c.); ad Circum Flaminium: Livy 39.2.11 (187 b.c.).

⁷³ CIL III.1426 (Sarmizegetusa) and CIL III.640 (Philippi); see TLL 2.1393. See also Patterson, J. R., ‘A dedication to Minerva Augusta from Butrint’, in Hansen, I. L. and Hodges, R. (eds), Roman Butrint: An Assessment (2007), 40–3Google Scholar for a more recent example from Roman Butrint.

⁷⁴ Hörig, M. and Schwertheim, E., Corpus Cultus Iovis Dolicheni (1987)CrossRefGoogle Scholar.

⁷⁵ Millar, F., ‘The Roman coloniae of the Near East: a study of cultural relations’, in Cotton, H. M. and Rogers, G. M. (eds), Rome, the Greek World, and the East, vol. 3 (2006), 176Google Scholar, conveniently collects the evidence for both cults.

⁷⁶ Sampaolo, V., ‘Osservazioni sul sistema viario a nord di Capua’, Bollettino di Archeologia 39–40 (1996 [2001]), 1–6Google Scholar; De Caro, S. et al. , ‘Nuovi dati per il santuario Capuano di Giove Tifatino’, Rendiconti della Accademia di archeologia, lettere e belle arti di Napoli 67 (1997–98), 15–29Google Scholar. We thank Nicholas Purcell for pointing us towards Monte Tifata.

⁷⁷ CIL VIII.23062 (+ CIL VIII.11167 = 906); Barton, op. cit. (n. 1), 303–4. Coins: e.g. RIC 6 (Sutherland, C. H. V., The Roman Imperial Coinage, vol. 6 (1967)Google Scholar), reverse legends listed in the index at pp. 701–2 under Iovi Conservatori and variants.

⁷⁸ Kuhfeldt, op. cit. (n. 1), 6; Barton, op. cit. (n. 1), 268.

⁷⁹ Barton, op. cit. (n. 1), 260. Cf. Todd, op. cit. (n. 1), who expresses some caution on the details (p. 57).

⁸⁰ Grimal and Woloch, op. cit. (n. 38), 50; Todd, op. cit. (n. 1), 57.

⁸¹ Dion. Hal. 4.61.4, and see Mura Sommella, op. cit. (n. 64) and Arata, op. cit. (n. 64) for the various possibilities for reconstruction.

⁸² Barton, op. cit. (n. 1), 260.

⁸³ e.g. Barton, op. cit. (n. 1), 265 (Minturnae, on the forum, perhaps the aedes Iovis of Livy 36.37); 266 (Terracina, where three chambers under the cella ‘imply … a corresponding division above’); 267 (Narbo); 269 (Virunum); 270 (Aenona, where a statue of Juno was found nearby).

⁸⁴ Vitruvius 4.7.2.

⁸⁵ Bispham, op. cit. (n. 50), 100 for more Italian examples.

⁸⁶ Barton, op. cit. (n. 1), 273.

⁸⁷ A point noted already by Toutain, op. cit. (n. 1, 1899), 19.

⁸⁸ Gros, P., L'Architecture romaine (1996), 194Google Scholar points out that ‘on a trop souvent déduit du compartimentage technique des substructions de ces temples une organisation à trois cellae qui ne se vérifie que rarement lorsque les niveaux d'occupation sont suffisament conserves pour qu'il soit possible d'en juger’. To his counter-example of the Capitolium at Thuburbo Maius could be added the supposed Capitolia at Gighthis and Cuicul, and, probably, the temple of Roma and Augustus at Lepcis (Livadiotti, M. and Rocco, G., ‘Il tempio di Roma e Augusto’, in Di Vita, A. and Livadiotti, M. (eds), I Tre Templi del lato nord-ovest del Foro Vecchio a Leptis Magna (2005), 217–18Google Scholar).

⁸⁹ Vitruvius 1.7.1. There is ambiguity in the Latin over whether Vitruvius means that the community's tutelary deities are Jupiter, Juno and Minerva, or whether these three and the tutelary deities are all candidates for the highest location in the city, and it is not even certain that Vitruvius is discussing a single ‘Capitoline’ cult entity here, rather than individual shrines to Jupiter, Juno and Minerva, any one of whom may be an example of a tutelary deity. Todd (op. cit. (n. 1), 57), however, takes this as an uncomplicated claim that the ‘most prominent site in the city must be reserved for the Capitoline Triad’.

⁹⁰ cf. McEwen, I., Vitruvius: Writing the Body of Architecture (2004)Google Scholar, Wallace-Hadrill, A., Rome's Cultural Revolution (2008), 144–210Google Scholar.

⁹¹ cf. Vergil's contemporary literary construction of Carthage, where the Phoenician colony is at least in part a reconstruction of the Roman one (Aen. 1.419–40, with the brief comment at Gros, P., ‘Carthage: faillite d'un Empire et résurrection d'une capital’, in Actes du Colloque Les Mégapoles méditerranéenes (2000), 535Google Scholar).

⁹² Brown et al., op. cit. (n. 45), 103–6: ‘The name “Capitolium” has been applied to the great temple of three cellae on the Arx of Cosa in default of explicit evidence of its identity, because the circumstantial evidence admits of no other … Its presence in the colony is warranted by Gellius’ definition of the effigies parvas simulacraque of the metropolis.' See Fentress, op. cit. (n. 49) on Brown's general model; Bispham, op. cit. (n. 50), 99–103 for the detailed case against an identification of the temple at Cosa as a Capitolium. Terracotta acroterial sculptures found on the Arx depicting the abduction of Ganymede may suggest that this area had a temple to Jupiter at some point (Torelli, op. cit. (n. 49), 39), but no more than that.

⁹³ Suet., Gramm. 9.6. It is interesting to note that the Budé edition (1993) translates Suetonius' ‘statua eius (i.e. Orbilius) Beneuenti ostenditur in Capitolio ad sinistrum latus marmorea’ as ‘On peut voir à Bénévent, sur le côté gauche du forum, une statue en marbre’.

⁹⁴ Suet., Tib. 40, Gaius 57 (Tacitus, Ann. 4.57, telling the same story refers simply to templa Iovi); Silius Italicus 11.265. Cf. Kuhfeldt, op. cit. (n. 1), 14–19.

⁹⁵ CIL IX.3688 = ILS 5364. This seems likely on the basis of nomenclature and the letter-forms to go back to the late Republic or early Empire: Letta, C. and D'Amato, S., Epigrafia della regione dei Marsi (1975), 64Google Scholar (no. 49).

⁹⁶ CIL IX.5438 = ILS 5368: a reference to paving a street as far as an arch adjoining an existing Capitolium in a.d. 119.

⁹⁷ CIL IX.2842 = ILS 5362. The inscription is likely to be second-century a.d. or later, as it contains the expression v(ir) c(larissimus).

⁹⁸ CIL V.3332 = ILS 5363 = AE 1989, 325; Buonopane, A., ‘Il materiale epigrafico’, in Manasse, G. Cavalieri (ed.), L'Area del Capitolium di Verona: richerche storieche e archeologiche (2008), 287–8Google Scholar. This inscribed statue base refers to the transfer of a statue from the Capitolium at Verona during the reign of Gratian, Valentinian and Theodosius, dating the inscription to a.d. 379–83.

⁹⁹ AE 1927, 124 = AE 2005, 324.

¹⁰⁰ CIL XI.1545 = ILS 3084.

¹⁰¹ Camodeca, G., ‘Evergeti ad Ercolano. Le inscrizioni di dedica del tempio di Venere’, Rendiconti della Pontifica Accademia Romana di Archeologia 81 (2008–2009), 51–7Google Scholar; see 51–2 for the text (with 56–7 for the relevant restoration): ‘[Vibi]dia virginis l(iberta) Saturnị[na] ẹt A. Fu[rius Saturnin]us / [o]b honores sibi et suis decret[os a]ẹdem Ven[eris vetustate corr]uptam / [imp]ensa sua refectam adornaverunt pronaio a solo fa[ct]o; id[em HS – –3 c.– in Capit]oli refec /[tio]ne contulerunt et amplius HS LIIII reip. dederunt ob flamoni[u]m et dec[urionalia ornamenta? m]aximạ.’ (‘Vibidia Saturninia, freedwoman of the [Vestal] virgin, and A. Furius Saturninus because of the honours decreed for themselves and their descendants, decorated the temple of Venus which had been ruined by age and was repaired at their expense, with the porch built from the ground upwards; they also contributed HS […] for the repair of the Capitolium and moreover gave HS 54,000 to the town on account of the flaminate and the highest insignia [?] of a decurion.’) We thank Nicholas Purcell for discussion of Faesulae and Herculaneum.

¹⁰² Althiburos is an African example, as is Saia Maior.

¹⁰³ Cavaliere Manasse, op. cit. (n. 39); for the identification and earlier studies see e.g. Frothingham, A. L., ‘Discovery of the Capitolium and forum of Verona’, AJA 18.2 (1914), 129–45CrossRefGoogle Scholar; Manasse, G. Cavaliere, ‘Il foro di Verona: recenti indagini’, in La Città nell'Italia Settentrionale in età romana (1990), 579–616Google Scholar. The identification at Verona, made already by Maffei in 1732 (reprint, op. cit. (n. 5), vol. I, 209–11) ‘è uno dei pochissimi attestati inequivocabilmente dalla convergenza dei requsiti necessari, epigrafici, urbanistici e planimetrici – i suoi resti, anche se conservati solo in fondazione, non si prestano a dubbi interpretative circa lo sviluppo dell'alzato’ (Cavaliere Manasse, op. cit. (n. 39), 312).

¹⁰⁴ Cavaliere Manasse, op. cit. (n. 39); p. 83 for discussion of the cella substructures and Tav. 5.

¹⁰⁵ Although the inscription is late fourth-century, and a question may therefore arise over what sense the word ‘Capitolium’ has here, the context is not one of Christian differentiation from pagan religion, and the meaning may well be the original one, of a temple to the Capitoline Triad.

¹⁰⁶ Buonopane, op. cit. (n. 98), 272 no. 1.

¹⁰⁷ Gasparri, C. et al. , ‘Cuma (Napoli). Il foro: campagne di scavo 1994, 196–7’, Bolletino di Archeologia 39–40 (1996 [2001]), 44–8Google Scholar; Gasparri, C., ‘Nuove indagini sul Foro di Cuma’, in Gigli, S. Quilici (ed.), La Forma della città e del territorio: esperienze metodologiche e risultati a confronto (1999), 131–7Google Scholar; Nava, M. L., ‘L'attivatà archeologica a Napoli e Caserta nel 2005’, in Atti del Quarantacinquesimo Convegno di Studi sulla Magna Grecia (2006), 650–2Google Scholar.

¹⁰⁸ Muscettola, S. Adamo, ‘La Triade del capitolium di Cuma’, in I Culti della Campania antica (1998), 219–30Google Scholar.

¹⁰⁹ de Azevedo, Cagiano, op. cit. (n. 3), 16–17 and idem., Aquinum (Aquino) (1949)Google Scholar, 40 n. 22, based on a letter in the archive of the Soprintendenza. Coarelli, F., Lazio (1982), 214–15Google Scholar presents doubts about the identification, based only in part on the location of the temple ‘in una zona troppo marginale della città’ (214).

¹¹⁰ Maetzke, G., Florentia (Firenze) (1941), 49–56Google Scholar; Barton, op. cit. (n. 1), 265; Pagni, M., ‘Dalla città augustea alla Fiorentia imperiale’, in Pagni, M. (ed.), Atlante Archeologico di Firenze (2010), 144–7Google Scholar.

¹¹¹ Maison Carrée: Fiches, J.-L. and Veyrac, A., Nimes. Carte archéologique de la Gaule 30/1 (1996), 282–3Google Scholar.

¹¹² CIL XIV.32.

¹¹³ Meiggs, R., Roman Ostia (1973), 380Google Scholar; cf. Kuhfeldt, op. cit. (n. 1), 26–7.

¹¹⁴ Barton, op. cit. (n. 1), 264; Meiggs notes that the ‘direct evidence that this Ostian temple was a Capitolium is not strong’ (op. cit. (n. 113), 380). For the dating and construction of this temple, see Albo, C., ‘Il Capitolium di Ostia. Alcune considerazioni sulla tecnica edilizia ed ipotesi ricostruttiva’, Mélanges de l'École française de Rome, Antiquité 114 (2002), 363–90CrossRefGoogle Scholar and DeLaine, J., ‘Building activity at Ostia in the second century AD’, in Bruun, C. and Zevi, A. Gallina (eds), Ostia e Portus nelle loro relazioni con Roma (2002), 64–71Google Scholar.

¹¹⁵ Meiggs, op. cit. (n. 113), 380; Pavolini, C., Ostia (1983), 102Google Scholar.

¹¹⁶ Barton's arguments for the identification of one of these earlier temples as a Capitolium (op. cit. (n. 1), 263) are not compelling.

¹¹⁷ Pompeii: Pitture e Mosaici, vol. VII (1997), 305–11Google Scholar.

¹¹⁸ Richardson is unusual in arguing for a Sullan date for the original temple, on the basis that the ‘original architecture showed strongly classicizing tendencies before any restructuring took place … And the masonry faced with opus incertum of broken lava most closely resembles that of the Theatrum Tectum’ (Richardson, L., Pompeii: an Architectural History (1988), 138Google Scholar).

¹¹⁹ Maiuri, A., ‘Pompei – Saggi nel edifici del Foro’, Notizie degli scavi di antichità (1942), 213–320Google Scholar, and subsequent authors (e.g. de Vos, A. and de Vos, M., Pompeii, Ercolano, Stabia (1982), 46Google Scholar and Coarelli, F., Pompeii (2002), 69Google Scholar).

¹²⁰ e.g. de Vos and de Vos, op. cit. (n. 119), 46; Zanker, op. cit. (n. 46), 64; Coarelli, op. cit. (n. 119), 69.

¹²¹ Barton, op. cit. (n. 1), 261. The identification goes back to Castan, op. cit. (n. 1, 1869), 19–20 (but see n. 125 below); cf. also Kuhfeldt, op. cit. (n. 1), 20–3.

¹²² NM Nap 6266; for Mau, A., Pompeii, its Life and Art (1899), 63Google Scholar, the relief of two figures on the back of the torso suggested that the temple was in the post-earthquake period a marble workshop, but Cooley argues that the relief is instead likely to have been recut to make the torso (Cooley, A. E., Pompeii (2003), 32Google Scholar).

¹²³ CIL X.796. Another dedication to Jupiter Optimus Maximus (CIL X.928) was found elsewhere in the city during the construction of the Sarno aqueduct at the end of the sixteenth century; its original context is unknown.

¹²⁴ A marble head of a woman was found in the temple, as was the base for a statue dedicated by one Spurius Turranius (CIL X.797), which may account for it: Cooley, op. cit. (n. 122), 32. Coarelli, op cit. (n. 119), 69 assumes that the female head is a head of Juno.

¹²⁵ A misconception originating with Castan, op. cit. (n. 1, 1869), 20, followed by e.g. Daremberg-Saglio, Dict. II.905; and repeated as recently as 2003 in Brill's New Pauly 2.1073. Vitruvius, De Arch. 3.2: ‘ornanturque signis fictilibus aut aereis inauratis earum fastigia tuscanico more, uti est ad Circum Maximum Cereris et Herculis Pompeiani, item Capitoli’, is correctly translated as: ‘The pediments are ornamented with statues of terracotta or gilt bronze in the Etruscan fashion, as are the Temple of Ceres and Pompey's Temple of Hercules at the Circus Maximus, and the Capitoline Temple.’ Vitruvius gives examples in this chapter entirely from Rome; see Platner-Ashby, s.v. Aedes Herculis Pompeiani (A Topographical Dictionary of Ancient Rome (1929))Google Scholar for Pompey's Temple of Hercules at the Circus Maximus. Kuhfeldt (op. cit. (n. 1), 20–1) pointed out the error, and Castan accepted the correction in his 1886 publication (op. cit. (n. 1, 1886), 25–6).

¹²⁶ e.g. Cagiano de Azevedo, op. cit. (n. 3), 34; Stella, C. in Comune di Brescia (1979), 48–9Google Scholar; Frova, A., ‘Il Capitolium di Brescia’, in La città nell'Italia Settentrionale in età romana (1990), 341–63Google Scholar; Stella, C., Brixia. Scoperta e risoperte (2003)Google Scholar; accepted by Barton, op. cit. (n. 1), 264. The identification is doubted, however, by Gregori, G. L., Brescia romana. Ricerche di prosopografia e storia sociale, vol. 2 (1999), 271Google Scholar; and Degrassi suggested that it was instead an imperial cult temple (Degrassi, N., ‘I fasti imperiali romani nel Capitolium di Brescia’, in Atti del Convegno internazionale per il XIX centenario della dedicazione del Capitolium e per il 150 anniversario della sua scoperta (1975), 197–204Google Scholar).

¹²⁷ Comune di Brescia, op. cit. (n. 126), 82, 95; Stella, op. cit. (n. 126, 2003), 51–2.

¹²⁸ CIL V.4312.

¹²⁹ For the bronze statues, see Comune di Brescia, op. cit. (n. 126), 69–75; Stella, op. cit. (n. 126, 2003), 56–77.

¹³⁰ Rossignani, M.-P., Luni. Guida Archeologica (1985), 55–7Google Scholar, 65 fig. 105 (a reconstructed, schematic plan). The identification is clearly made on the basis of what we have called above ‘the standard view’: ‘Il tempio fu construito pochi anni dopo la fondazione della colonia: l'erezione di tali edifici templari (dedicati alla triada di Giove, Giunone, Minerva venerata nel Capitolium di Roma), quasi sempre collocati su uno dei lati brevi del Foro, sancisce nelle colonie il legame politico e religioso con la città-madre’ (p. 56). The identification is assumed in Rossignani, M.-P., ‘Gli edifice pubblici nell'area del Foro di Luni’, Quaderni del Centro Studi Lunensi 10/12 (1985–87), 123–48Google Scholar.

¹³¹ cf. Ward-Perkins, J. B., ‘From Republic to Empire: reflections on the early provincial architecture of the Roman West’, JRS 60 (1970), 6–13Google Scholar for discussion of several examples (Augusta Raurica, Augusta Bagenniorum, St Bertrand de Comminges, Zadar, Conimbriga, Virunum), in which none of the temples is in fact identified definitively as a Capitolium. Blutstein-Latrémolière, op. cit. (n. 39), 61–3, shows that there are in fact temples of imperial cult in the fora she identifies as ‘places capitolines’ (Barcino, Italica) — see below for discussion.

¹³² Up-to-date bibliography and cautious discussion of most of these possibilities can be conveniently found at Cavalieri Manasse, op. cit. (n. 39), 310–14; for her there are ‘poco più di venti casi’ in Italy (314). For Grumentum, Mastrocinque has argued not only that the mid-first-century a.d. ‘Temple D’ on the Forum was a Capitolium, but also that ‘il più antico monument del Foro è il [Augustan] tempio C, il cosidetto Cesareo, che dunque non nacque come tale, ma come Capitolio, data la sua posizione central nella città e la sua dimensioni: in una colonia romana non poteva mancare il Capitolio e un enorme tempio sul Foro non poteva essere che il Capitolio’ (Mastrocinque, A., ‘Grumentum: nuove ricerche’, in Grumentum Romana (2009), 253Google Scholar).

¹³³ A connection would be equally hard to demonstrate at Aquinum, Fiorentia or Ostia.

¹³⁴ Gasparri et al., op. cit. (n. 107).

¹³⁵ CIL X.3703–4.

¹³⁶ Note though Adamo Muscettola, op. cit. (n. 108), 228–30, on the dating of the cult statues.

¹³⁷ Our knowledge of the topography of the Roman coloniae in Britain such as London, Colchester, York and Lincoln is much poorer than that of other cities such as Silchester and Caistor-by-Norwich for which we have complete plans, and we cannot say for certain that coloniae in Britain lacked Capitolia. We thank Roger Wilson for this point.

¹³⁸ CIL II.1194. References to the Capitoline gods in the lex coloniae from Urso concern games to be celebrated for them, and do not mention any associated building: ILS 6087, LXX–LXXI. Rüpke discusses the ‘symbolic link to Rome’ these provisions created (Rüpke, J., ‘Religion in the lex Ursoniensis’, in Ando, C. and Rüpke, J. (eds), Religion and Law in Classical and Christian Rome (2006), 41Google Scholar).

¹³⁹ RIT 922. Cf. also RIT 34 from Tarraco, a dedication to Jupiter Optimus Maximus, Juno, Minerva, the Genius praetorii and the Dii Penates; findspot unknown.

¹⁴⁰ Mar, R. et al. ‘Arqueologia Urbana en el foro de Tarraco’, in Arqueologia, Patrimonio y desarrollo urbano. Problemática y soluciones (2010), 61–3Google Scholar for the structure near the lower forum; Alföldy, G., Die römischen Inschriften von Tarraco (1975), 403Google Scholarapud RIT 922, for the hypothesis that the Capitolium was on the citadel.

¹⁴¹ e.g. Paris et al., op. cit. (n. 1), 68–75; Keay, S., Roman Spain (1988), 136Google Scholar; Keay, S., ‘The development of towns in early Roman Baetica’, in Keay, S. (ed.), The Archaeology of Early Roman Baetica (1998), 73Google Scholar; Sillières, P., Baelo Claudio, une cité romaine de Bétique (1995), 85–96Google Scholar; MacMullen, op. cit. (n. 1), 60; Bonneville, J.-N. et al. , Belo VII. Le capitole (2000)Google Scholar; cf. Barresi, P., ‘I Capitolia di Sufetula e di Baelo Claudia: analisi dei progetti’, in Camporeale, S., Dessales, H. and Pizzo, A. (eds), Arqueología de la construcción I. Los procesos constructivos en el mundo romana: Italia y provincias occidentales (2008)Google Scholar for similarities in the design conceptions and proportions of the two complexes.

¹⁴² Sillières, op. cit. (n. 141), 91; Bonneville et al., op. cit. (n. 141), 179–95.

¹⁴³ Galán, M. Bendala, ‘Baelo Claudio y su personalidad ciudadana y urbana: diálogo desde el estudio y la amistad’, Pallas: revue d'études antiques 82 (2010), 473–7Google Scholar. He speculatively suggests that the central temple may be to Melkart/Hercules, the eastern temple to Tanit, and the western one to another Punic deity, possibly Eshmun.

¹⁴⁴ Blutstein-Latrémolière, op. cit. (n. 39).

¹⁴⁵ cf. also Gascó, C. Aranegui et al. , ‘El Foro di Saguntum: la planta arquitectónica’, in Los foros Romanos de las provincias occidentales (1987), 74–7Google Scholar; the argument for this being a Capitolium is based on the tripartite substructures and the location on the forum. Gascó, C. Aranegui, ‘Un templo republicano en el centro cívico saguntino’, in Templos Romanos de Hispania (1991), 67–82Google Scholar, is extremely cautious about the possibility of this identification.

¹⁴⁶ Gros, op. cit. (n. 39), 113; op. cit. (n. 88), 151. Likewise, he accepts (op. cit. (n. 39), 114; op. cit. (n. 88), 152) the identification of the so-called Capitolium at Ampurias purely on the basis of its situation on the forum within a U-shaped portico (porticus triplex), although he had earlier pointed out (op. cit. (n. 39), 112) that such an arrangement within a porticus triplex was also characteristic of imperial cult temples.

¹⁴⁷ Eumenius, Pan. Lat. 9 (4), 9f: ‘quod praecipuo est positum quasi inter ipsos oculos civitatis, inter Apollinis templum atque Capitolium … ibi adulescentes optimi discant … maximorum principum facta celebrare … ubi ante aras quoammodo suas Iouios Herculiosque audiant praedicari Iuppiter pater et Minerua socia et Iuno placata.’

¹⁴⁸ Sid. Apoll., Ep. 9.16 (carm. 65–72): ‘e quibus primum mihi psallat hymnus, qui Tolosatem tenuit cathedram [i.e. St Saturninus], de gradu summo Capitoliorum praecipitatum.’ The late antique martyr narrative of Saint Saturninus has a rather different story, but still involving a Capitolium: the bishop used to pass by the Capitolium on his way to his church, and the priests in the temple blamed the silence of their oracles on this fact, seized him and attempted to force him to sacrifice to their idols; when he would not, they had him tied to a bull who dragged him around the town to his death: Cabau, P., ‘Opusculum de passion ac translatione sancti Saturnini, episcopi Tolosanae ciuitatis et martyris’, in Mémoires de la Société archéologique du Midi de la France 61 (2001)Google Scholar for a provisional edition of this Opusculum de passione ac translatione Sancti Saturnini, episcopi Tolosanae civitas et martyris, where the text is dated to the early fifth century (59); we thank one of JRS's anonymous referees for this reference. Gregory of Tours, Hist. Franc. 1.28; see Kuhfeldt, op. cit. (n. 1), 63 n. 237.

¹⁴⁹ Boudartchouk, J.-L. et al. , ‘Le Capitolium de Toulouse, l’Église Saint-Pierre-et-Saint-Géraudet le martyre de l'évêque Saturnin: nouvelles donneés’, Mémoires de la Société archéologique du Midi de la France 65 (2005), 15–50Google Scholar.

¹⁵⁰ Sid. Apoll., Carm. 23, 39–45: ‘Salve, Narbo potens salubritate, urbe et rure simul bonus videri, muris, civibus, ambitu, tabernis, portis, porticibus, foro, theatro, delubriis, capitoliis, monetis’: the plural is not explicable simply by the demands of scansion (contra Castan, op. cit. (n. 1, 1869), 22), since the final syllable of the singular capitolio could also be long, but Sidonius also talks here of monetae. Cf. Ausonius 11.19.14–17, comparing a temple at Narbo to the Capitolium at Rome: ‘quodque tibi Pario quondam de marmore templum tantae molis erat, quantam non sperneret olim Tarquinius Catulusque iterum, postremus et ille aurea qui statuit Capitoli culmina Caesar?’ (11.19, 14–17) ‘Or of that temple of Parian marble, once yours, so vast in bulk that Tarquin once would not scorn it, nor again Catulus, nor finally he who last raised the golden roofs of the Capitolium, Caesar himself [i.e. Domitian]?’ We do not see a positive claim in this passage that the temple under discussion at Narbo was itself a Capitolium, though many have (e.g. Castan, op. cit. (n. 1, 1869), 21–2; Barton, op. cit. (n. 1), 267).

¹⁵¹ Castan, op. cit. (n. 1, 1869), 22–3 accepts the identification, and Grenier, op. cit. (n. 14) also argues for it; Gros, op. cit. (n. 39), 112 and n. 2 regards it as an unresolved question. See Kuhfeldt, op. cit. (n. 1), 61–2 and n. 232 for the medieval documents which mention this toponym; in the extracts which he cites the Capitolium is mentioned in connection with forts, towers and other strong points of the city's defences, which raises the suspicion that we are here dealing with ‘Capitolium’ in the medieval sense of ‘citadel’. For Gayraud (op. cit. (n. 14), 260), the arrangement of the foundations, in which walls divide the podium of the temple into three longtitudinally and two latitudinally, ‘invite à concevoir une cella tripartite’, but no more than that. On the size, see Barton, op. cit. (n. 1), 334 fig. 1.

¹⁵² CIL XII.4318

¹⁵³ CIL XII.4437, 4439, Grenier, A., ‘Appendice épigraphique’, in Carte archéologique de la Gaule Romaine 12 (1959)Google Scholar, 207 no. 6, and for a possible [ma]gister la[rum augustalium] CIL XII.4386. On the evidence for the temple's identification as a Capitolium, see Grenier, op. cit. (n. 14) and Gayraud, op. cit. (n. 14), 269–71; for the suggestion that the temple combined this with an imperial cult function, 271–2. Cf. also Fishwick, D., ‘Un don de statues d'argent à Narbo Martius’, CRAI 136.2 (1992), 381–401CrossRefGoogle Scholar.

¹⁵⁴ Praschniker, C., ‘Die kapitolinische Trias von Ödenburg-Sopron’, Jahreshefte des Österreichischen archäologischen Instituts in Wien 30 (1937), 111–34Google Scholar, especially 120–2 for the context; cited also by Cagiano de Azevedo, op. cit. (n. 3), 42.

¹⁵⁵ Praschniker, op. cit. (n. 154), 127–9.

¹⁵⁶ CIL III.4363 = 11079, accepted as a Capitolium by Toutain, op. cit. (n. 1, 1899), 6.

¹⁵⁷ Fitz, J., ‘Forschungen in Gorsium im Jahre 1979’, Alba Regia 19 (1982), 204–6Google Scholar and 209; Abb. 8 and Tav. XX; the building described in Fitz and Fedak, op. cit. (n. 43), 265 as a Capitolium with three cellae does not look especially temple-like on the plan in Fitz, op. cit. (n. 157), Abb. 8; the initial identification of the half-life size head was that it might be Jupiter (Fitz, op. cit. (n. 157), 209 no. 1; Tav. XX). Fragments of a statue of Minerva were found in the vicinity (Fitz and Fedak, op. cit. (n. 43), 265).

¹⁵⁸ AE 1989, 581.

¹⁵⁹ CIL III.6167; Barton, op. cit. (n. 1), 269: ‘since it is in seven relatively short lines it is unlikely to come from the entabulature of a temple.’

¹⁶⁰ Barton, op. cit. (n. 1), 270; Todd, op. cit. (n. 1), 63.

¹⁶¹ And indeed the name S. Maria in Capitolio is not attested before the thirteenth century; before that it was called Maria alta or Maria in altis, S. Maria in Malzbuchel, S. Maria super Malzbuchel. The church is on a low hill and we may again be dealing with the medieval sense of the term ‘Capitolium’, as citadel, with no evidential value for the Roman period. See Kuhfeldt, op. cit. (n. 1), 74–5 for the sources.

¹⁶² Hellenkemper, op. cit. (n. 45), 804–9; Todd, op. cit. (n. 1), 65; Follmann-Schulz, A. B., ‘Die römischen Tempelanlagen in der Provinz Germania inferior’, ANRW II.19.1 (1986), 735–8Google Scholar.

¹⁶³ Schalles, H. J., ‘Forum und zentraler Tempel im 2. Jahrhundert n. Chr.’, in Die römischen Stadt im 2. Jahrhundert n. Chr. (1992), 201Google Scholar.

¹⁶⁴ Dio 69.12.1. The variant version at Eusebius, HE 4.6, which says Aelia Capitolina arose after the Bar Kochba revolt, is ruled out by the evidence of hoards containing both Aelia and Bar Kochba coinage: Meshorer, Y., The Coinage of Aelia Capitolina (1989), 19Google Scholar; Goodman, M., ‘Judaea’, in The Cambridge Ancient History XI (2000), 673Google Scholar (with thanks to JRS's anonymous reader for the latter reference). Cf. Paulinus of Nola, Ep. 31: ‘nam Hadrianus … in loco passionis simulacrum Iovis consecravit’ – Patrologia Latina 61, p. 326 D (Migne); Sulpicius Severus, Chronica 2.31.3.

¹⁶⁵ Kuhfeldt, op. cit. (n. 1), 60; Castan, op. cit. (n. 1, 1869), 186–8; Meshorer, op. cit. (n. 164), 22 and 70–1 no. 1. Coins of Antoninus Pius, however, show Jupiter seated alone in a tetrastyle temple (Meshorer, op. cit. (n. 164), 27–8 and 72–3 nos 18–19).

¹⁶⁶ Kuhfeldt, op. cit. (n. 1), 54–5.

¹⁶⁷ Hesychius of Miletus 1.41bis (= Preger, T., Scriptores originum Constantinopolitanarum, vol. 1 (1901), 18Google Scholar): Ὁ δὲ βασιλεὺς οὐ μόνον οἴκους ἐν Κωνσταντινουπόλει περιϕανεῖς κατὰ μίμησιν Ῥώμης καὶ τὸ Καπετώλιον ἔκτισεν, ἀλλὰ καὶ θείους τε καὶ ἱεροὺς ναοὺς πολυτελῶς ἀνεδείματο, τόν τε τῆς ἁγίας Εἰρήνης ναὸν καὶ τῶν σεβασμίων καὶ κορυϕαίων Χριστοῦ μαθητῶν … [followed by a list of other churches]

¹⁶⁸ Janin, R., ‘Du Forum Bovis au Forum Tauri. Étude de topographie’, Revue des études byzantines 13 (1955), 85–108CrossRefGoogle Scholar; op. cit. (n. 4), 174–6.

¹⁶⁹ Dindorf, L., Chronicon Paschale I (1832), 570Google Scholar.

¹⁷⁰ C. Th. 14.9.3 (law de studiis liberalibus); cf. C. Th. 6.21.1 (law of 15 March 425, de professoribus).

¹⁷¹ Cledonius in Keil, H.. Grammatici Latini, vol. 5 (1868), 14Google Scholar.

¹⁷² Ghislanzoni, E., ‘Statua di Giove ed iscrizione onoraria agli imperatori Adriano e Antonino Pio rinvenute a Cirene’, Notizario Archeologico 39–40 (1916), 193–216Google Scholar.

¹⁷³ Mariani, L., ‘Zeus aigiochos’, Notizario Archeologico 3 (1922), 7–10Google Scholar; Bagnani, G., ‘Hellenistic sculpture from Cyrene’, JHS 41 (1921), 238–41CrossRefGoogle Scholar; Cagiano de Azevedo, op. cit. (n. 3), 46; Bonacasa, N. and Ensoli, S., Cirene (2000), 76–7Google Scholar.

¹⁷⁴ Chamoux, F., ‘Un sculpteur de Cyrène: Zénion, fils de Zénion’, Bulletin de correspondance hellénique 70 (1946), 67–77CrossRefGoogle Scholar. For the Zeus statue, see Paribeni, E., Catalogo delle sculture di Cirene, statue e rilievi di carattere religioso (1959), 78–9Google Scholar (no. 185) and Tav. 106; for the others, Smith, A. H., A Catalogue of Sculpture in the Department of Greek and Roman Antiquities, British Museum, vol. 2 (1900)Google Scholar, 255 nos 1478 and 1479.

¹⁷⁵ Livy 41.20.9: ‘… et Antiochiae Iovis Capitolini magnificum templum, non laqueatum auro tantum, sed parietibus totis lammina inauratum, et alia multa in aliis locis pollicitus, quia perbreve tempus regni eius fuit, non perfecit’ (‘and at Antioch a magnificent temple of Jupiter Capitolinus, not merely gilded, but decorated with gold sheets on all its walls, and he promised many other buildings in other places, which he did not finish because of the very short time of his reign’).

¹⁷⁶ John Malalas, Hist. Chron. 10.10 (Thurn, J., Ioannis Malalae Chronographia (2000), 178 ll. 44–5Google Scholar).

¹⁷⁷ Acta SS. Juliani, Basilissae, et Sociorum apud Acta Sanctorum, Januarii vol. 1, pp. 580 and 585 (Bolland.).

¹⁷⁸ Castan, op. cit. (n. 1, 1869), 181; RPC Online http://rpc.ashmus.ox.ac.uk/coins/9463/ (Accessed 11 December 2011).

¹⁷⁹ CIG 3153.

¹⁸⁰ CIG 2943.

¹⁸¹ CIG 3074 = IGR IV.1556, an altar of Zeus Ktesios, Zeus Kapetolios, Roma and Agathos Daimon; cf. Mellor, R., ‘The Goddess Roma’, in ANRW II.17.2 (1981), 960Google Scholar, who explains: ‘Here Zeus appears both as protector of the home (Ktesios) and as protector of treaties (Jupiter Capitolinus).’

¹⁸² Paus. 2.4.5. Walbank takes ὑπὲρ δὲ to mean ‘beyond’ and admits that ‘this sanctuary … must be well away from the city centre if it is beyond the theatre’ (op. cit. (n. 54), 367), though she makes an unconvincing case that Temple E on the Forum is a separate temple to the Capitoline Triad.

¹⁸³ e.g. P. Oxy. I.43 verso iv.3 (a.d. 295) — a house behind the Kapitolion.

¹⁸⁴ P. Oxy. XVII.2128.4.

¹⁸⁵ P. Oxy. XVII.2109.8–9.

¹⁸⁶ P. Oxy. LXVI.4541.

¹⁸⁷ P. Oxy. LIV.3757.3; 3758.78 and 156.

¹⁸⁸ P.NYU 2 40.

¹⁸⁹ P.Mil. Vogl. 4 233.

¹⁹⁰ BGU II 362; see Johnson, A. C., An Economic Survey of Ancient Rome vol. 2: Egypt (1936), 662–8Google Scholar; Lewis, N., Life in Egypt under Roman Rule (1983), 88–9Google Scholar.

¹⁹¹ Barton, op. cit. (n. 1), 266–70.

¹⁹² Both discussed with bibliography at Cavalieri Manasse, op. cit. (n. 39), 315.

¹⁹³ Barton, op. cit. (n. 1), 279–326. In each of our categories we list the examples in chronological order in so far as that can be ascertained.

¹⁹⁴ CIL VIII.929 = CIL VIII.11206, a building inscription giving a Hadrianic or later date; and CIL VIII.928 = CIL VIII.11205, an inscription of a.d. 388/392 referring to cellis Capi[toli].

¹⁹⁵ The Capitolium is approximately dated by an inscription found reused in the paving of its precinct, which originally came from the epistyle of the colonnade and records its dedication by M. Plotius Faustus (‘Sertius’) and Cornelia Valentina Tucciana (‘Sertia’): Doisy, H. Pavis d'Escurac, ‘Flaminat et société dans la colonie de Timgad’, Antiquités Africaines 15 (1980), 190, 198–9Google Scholar. This couple also built the nearby Market of Sertius, thought to belong to the Severan period because of several characteristics of the inscriptions on the statue bases to Sertius and his wife (CIL VIII.2393–9, 17904–5): the use of signa or nicknames (‘Sertius’, ‘Sertia’), the absence of the voting tribe, the formula a militiis, and the absence of imperial epithets in the names of the military units mentioned, all combine to suggest an early third-century date: Boeswillwald, E., Cagnat, R. and Ballu, A., Timgad, une cité africaine sous l'empire romain (1905), 188–9Google Scholar. Ward-Perkins, J. B., Roman Imperial Architecture (1981)Google Scholar, 394, considered on stylistic grounds that it dated ‘probably from the latter part of the second century’, but this now seems slightly too early.

¹⁹⁶ Inventory of municipal treasures held ‘in Kapitolio’ (CIL VIII.6981 = CIL VIII.6982 = ILAlg 2.1.483 = ILS 4921; CIL VIII.6983 = ILAlg 2.1.538; CIL VIII.6984 = ILAlg 2.1.539); dedication of statue possibly iuxta C[apitolium] (CIL VIII.7014 = ILAlg 2.1.591 = ILS 758).

¹⁹⁷ Temple with a single cella, with dedications to Juno Regina (IL Alg 1.1230 = AE 1909, 239) and to Minerva Augusta (ILAlg. I.1231 = AE 1906, 4 and 5 = AE 1909, 238), both on marble plaques probably from the base of cult statues. In addition, a colossal statue of Jupiter, a colossal head of Minerva and fragments which could belong to a head of Juno were found, apparently remains of the cult statues themselves.

¹⁹⁸ Single-cella temple with three niches, building dedication to the Capitoline Triad (CIL VIII.1471 = CIL VIII.15513 = CIL VIII.15514 = Dougga 31 = Dougga 32 = ILTun 1379), and a head of Jupiter. See also Cagnat and Gauckler, op. cit. (n. 27), 1–3.

¹⁹⁹ Probably single-cella temple with building inscription dedicated to the Capitoline Triad (AE 1950, 136), and a head and upper torso of Jupiter.

²⁰⁰ Double-cella temple of peculiar design, in colonnaded court, with building inscription and dedications to all of the Capitoline Triad plus the Genius of the Colony (CIL VIII.2611 = AE 1951, 121 = AE 1992, 1862; CIL VIII.2612; CIL VIII.18226; Barton, op. cit. (n. 1), 289–91).

²⁰¹ Building inscription mentioning Capitolium, dedicated to the Capitoline Triad (ILTun 699 = AE 1914, 55 = AE 1923, 106 = AE 1942/43, 111 = ILAfr 244 = ILPBardo I, 339). In addition, a head of Jupiter was found. Description of the temple (before its excavation and identification): Cagnat and Gauckler, op. cit. (n. 27), 120–1.

²⁰² Tetrastyle Corinthian temple on the highest part of town (but not the forum as built on private land), with building inscription mentioning Capitolium, dedicated to the Capitoline Triad (CIL VIII.26161 = AE 1892, 145). See also Cagnat and Gauckler, op. cit. (n. 27), 6–8.

²⁰³ Temple with a cella and two ‘transepts’, dominating a paved court opposite the forum; in the cella was found a head of Juno. The building inscription is dedicated to the Capitoline Triad and refers to the temple as a Ka[pi]tolium (CIL VIII.27769 = CIL VIII.1824 = CIL VIII.1826 = CIL VIII.1831 = CIL VIII.16470 = AE 1913, 45). Cagnat and Gauckler, op. cit. (n. 27), 8–10.

²⁰⁴ Temple in its own precinct off the forum, with a building inscription referring to K[apitoliu]m and dedicated to Capitoline Triad (IAM II.2, 355 = ILM 45 = AE 1925, 30 = AE 1926, 26).

²⁰⁵ Temple structure with a curious plan, located on one side of a paved area assumed to be the forum; the building inscription calls the building a Capitolium and is dedicated to the variant Capitoline Triad of Jupiter Conservator, Juno Regina and Miner[va Augusta] — see above, Section 1 (CIL VIII.906 = CIL VIII.11167 = CIL VIII.23062 = AE 1905, 127 = AE 1995, 1645).

²⁰⁶ Fragments of a monumental building inscription mentioning a Capitolium, reused in the paving of a late workshop: IAM-S, 861 = AE 1991, 1750.

²⁰⁷ Building inscription dedicated to the Capitoline Triad and referring to aedem Capitoli (CIL VIII.25500), apparently placed within a colonnaded enclosure entered by a monumental arch, and decorated with twelve statues.

²⁰⁸ Inscription recording, amongst other testamentary benefactions, 170 lb of silver and 14 lb of gold given ad Kapitol. (CIL. VIII.1858 = CIL VIII,16504d = ILAlg. 1.3040 = BCTH 1943/45, 112 = AE 1942/43, +56 = AE 1945, 58 = AE 1988, 1120).

²⁰⁹ Building inscription mentioning Kapitoli[um]: CIL VIII.10767 = CIL VIII.16849 = ILAlg 1.1097.

²¹⁰ Inscription mentioning Capitolium: AE 1914, 35.

²¹¹ A list of votive offerings or inventory of precious artefacts in the Capitolium: CIL VIII.1013 = 8.12464; and a fragment of marble plaque apparently mentioning the Capitolium: AE 1999, 1841 = AE 2007, 1729. The Capitolium at Carthage is also mentioned in literary sources: Cyprian, de lapsis 26 (a.d. 251); C.Th. 11.1.34 (25 February 429).

²¹² Statue base honouring a benefactor who rebuilt the forum cum aedibus e[t Capi]tolio — suggesting that the Capitolium was on the forum: CIL VIII.24095 = ILPBardo I.418 = ILS 5361 = AE 1894, 115.

²¹³ Inscription mentioning sacerdotes Kapitoli: ILAlg 1.2146 = AE 1907, 2.

²¹⁴ Dedication to Jupiter Optimus Maximus, also mentioning a statue promised in Capitolium: CIL VIII.6339 (p. 965, 1841) = ILS 3669 = ILAlg II.3, 8795.

²¹⁵ Twelve fragments of a marble plaque attesting repair of a Capitolium: AE 2003, 2004; Peyras, J., ‘Inscriptions latines du basin de Bagrada’, in Bost, J.-P. et al. (eds), Itinéraire de Saintes à Dougga (2003), 277–9Google Scholar.

²¹⁶ Building dedication to Capitoline Triad (fragmentary, but not in doubt): ILAfr 13 = AE 1909, 240.

²¹⁷ This inscription (CIL VIII.4195) has been published as follows: ‘[In honorem domus A]ug(ustae) Iovi Opt[imo Max]imo Iun[oni Reginae, Minervae Augustae: / Imp(eratori) Caes(ari) M(arco) Aurelio Antonino Au]g(usto) Armeniaco Parthico [maximo Medico] pont(ifici) max(imo) trib(unicia) pote[st(ate) XX imp(eratori) IIII co(n)s(uli) III / et Imp(eratori) Caes(ari) L(ucio) Aurelio Vero Aug(usto) A]rmeniaco Parthic[o maximo Medic]o trib(unicia) potes(tate) VI im[p(eratori) IIII co(n)s(uli) II] / […]tus Venustus leg(atus) [Aug(usti) pr(o) pr(aetore) co(n)s(ul) desi]gnatus dedica[vit].’ But this seems unsatisfactory; it would be very odd to have the phrase In honorem domus Augustae beginning the inscription and carved in larger letters than the subsequent lines, on a par with the letter sizes for the deities. Moreover, if one restores Iunoni Reginae as would be expected with IOM there is not room on the right-hand side for Minerva. But since Minerva's epithet, Augusta, is present in the inscription, it is easiest to restore her at the start; the order of divinities is unusual, but the order of names on the temple's lintel would exactly parallel the spatial arrangement of the cellae or statues of the divinities within the temple. We would restore the first line as ‘[Minervae A]ug(ustae) Iovi Opt[imo Max]imo Iun[oni Reginae]’.

²¹⁸ Building inscription dedicated to the Capitoline Triad: CIL VIII.14369 = ILAfr 435 = AE 1934, +34 = ILT 1206.

²¹⁹ Building inscription (eight fragments of entablature) dedicated to Capitoline Triad: CIL VIII.2194. Corinthian capitals and column shafts were found at the site, but how closely associated with the inscription is unclear (Moll, A., ‘Inscriptions romaines découvertes à Tébessa et dans les environs pendant les années 1858 et 1859’, Recueil des notices et mémoires de la Société archéologique de Constantine 4 (1858–1859), 178–80Google Scholar, inscriptions nos 9–13).

²²⁰ CIL VIII.27827. This dedication to the Capitoline Triad has a slightly unusual format for a Capitolium, but is apparently a building inscription: the surviving fragment is 1.2 m wide, 0.6 m high, with letters 0.24 m high.

²²¹ Given that this is true on the literary and epigraphic evidence alone, our argument about the relative popularity of the cult in Africa is not affected by Barton's point that one needs to take into account the better preservation of monuments in Africa where comparatively few sites have been built over — an observation which is in any case also true of Spain: Barton, op. cit. (n. 1), 260, where he also notes that whereas ‘in Gaul and Spain municipal life tended to be concentrated on the larger cities, each with an extended territorium, in Africa it was dispersed among numerous small civitates, more and more of which as time went on acquired municipal institutions of their own’, and he relates this to the late second-/early third-century dating of ‘the majority of the African capitolia’. He still thinks that there are a disproportionate number of Capitolia in Africa.

²²² Barton, op. cit. (n. 1), 286.

²²³ CIL VIII.15663; Cagnat and Gauckler, op. cit. (n. 27), 13.

²²⁴ ILAlg. 1.1026 = CIL VIII.28064 = AE 1909, 7.

²²⁵ AE 1978, 855.

²²⁶ CIL VIII.17511 = ILAlg. 1.470.

²²⁷ CIL VIII.24328; Barton, op. cit. (n. 1), 295.

²²⁸ CIL VIII.11198 = 921; Barton, op. cit. (n. 1), 324.

²²⁹ CIL VIII.16439; Barton, op. cit. (n. 1), 323–4.

²³⁰ CIL VIII.23876 (= 12286); Barton, op. cit. (n. 1), 283.

²³¹ AE 1949, 109; Barton, op. cit. (n. 1), 294–5.

²³² CIL VIII.25935.

²³³ CIL VIII.12014 = ILS 5412; Barton, op. cit. (n. 1), 321–2.

²³⁴ IRT 4 and 9; cf. Barton, op. cit. (n. 1), 300–2.

²³⁵ Ward, P., Sabratha: a Guide for Visitors (1970), 35Google Scholar.

²³⁶ cf. Kenrick, P. M., Excavations at Sabratha 1948–1951 (1986), 114Google Scholar. Giudi (unpublished; communication cited by Cagiano de Azvedo, op. cit. (n. 3), 48) noted Egyptianizing decoration, and considered that the temple might have been dedicated to Jupiter Ammon (Giudi, G., ‘I monumenti della Tripolitania romana’, Africa Romana (1935), 247Google Scholar), but it appears that this refers to the Egyptianizing marble reliefs found in the vaults of the temple, which Kenrick argues derive from the nearby temple of Serapis (Kenrick, op. cit. (n. 236), 115 and pl. 28).

²³⁷ Constans, L. A., Gigthis. Études d'histoire et d'archéologie sur un emporium de la Petite Syrte (1916), 29–32Google Scholar; cf. Barton, op. cit. (n. 1), 286–7.

²³⁸ Cagnat and Gauckler, op. cit. (n. 27), 14–18; Cagnat and Chapot, op. cit. (n. 1), 159–60; Duval, N. and Baratte, F., Les ruines de Sbeïtla (1973), 23–30Google Scholar; Duval, N., ‘L'urbanisme de Sufetula = Sbeitla en Tunisie’, ANRW II.10.2 (1982), 606–7Google Scholar. Cf. Barresi, op. cit. (n. 141) on similarities in the proportions of the designs between Baelo and Sufetula.

²³⁹ e.g. Bendala Galán, op. cit. (n. 143), 473–4; cf. n. 143 above on Baelo Claudia.

²⁴⁰ Quinn, J. C., ‘The reinvention of Lepcis’, in Jiménez, A. (ed.), Colonising a Colonized Territory: Settlements with Punic Roots in Roman Times, in M. Riva, Dalla and Di Giuseppe, H. (eds), Meetings between Cultures in the Ancient Mediterranean. Proceedings of the 17th International Congress of Classical Archaeology. Bolletino di Archeologia online 1 (2010), 56Google Scholar.

²⁴¹ Masturzo, N., ‘Il tempio occidentale – tempio di “Liber Pater”’, in Di Vita, A. and Liviadiotti, M. (eds), I Tre Templi del lato nord-ovest del Foro Vecchio a Leptis Magna (2005)Google Scholar, especially 57 and 129–31, building on Musso, L., ‘Nuovi ritrovamenti di scultura a Leptis Magna: Athena tipo Medici’, in Bacchielli, L. and Aravantinos, M. B. (eds), Studi Miscellanei 29, vol. 2 (1996), 115–39Google Scholar; the inscription is IRT 290.

²⁴² cf. Masturzo, op. cit. (n. 241), 57.

²⁴³ Di Vita, A., ‘Liber Pater o Capitolium? Una nota’, in Di Vita, A. and Liviadiotti, M. (eds), I Tre Templi del lato nord-ovest del Foro Vecchio a Leptis Magna (2005), 16Google Scholar.

²⁴⁴ IPT 31.

²⁴⁵ Quinn, op. cit. (n. 240), 58.

²⁴⁶ Statues found at Rapidum have been claimed to represent Jupiter (seated, with a thunderbolt) and Minerva (Chabassière, ‘Notice sur Sour Djouab et ses environs’, Revue africaine 13 (1869), 456Google Scholar and pl. IV; Gsell, S., Les Monuments antiques de l'Algerie (1901), vol. 1, 153Google Scholar; Ballu, A., ‘Rapport sur les fouilles exécutées en 1910 par le Service des monuments historiques de l'Algerie’, Bulletin archéologique du Comité des travaux historiques et scientifiques (1911), 93–4Google Scholar; Barton, op. cit. (n. 1), 299), but J.-P. Laporte has shown that the seated statue is probably Aesculapius (Laporte, J.-P., Rapidum. Le camp de la cohorte des Sardes en Maurétanie Césarienne (1989), 121–3, 170–5Google Scholar) and in any case the female statue is of a different scale, making it hard to see them as part of a cult triad. Laporte's photo (172 pl. 18) shows the statue of Jupiter to be approximately twice life-size, while the female deity (previously claimed on no real evidence to be Minerva or Juno) is said to be life-size (174).

²⁴⁷ Barton, I. M., ‘Encore un Capitole africain? Le temple de Cillium’, Antiquités africaines 25 (1989), 227–34CrossRefGoogle Scholar.

²⁴⁸ cf. Di Vita, op. cit. (n. 243), 19, who argues that the cult of Jupiter Optimus Maximus in the provinces only became popular from the reign of Domitian, when the emperor was first more or less directly assimilated to Jupiter. But the bulk of dated Capitolia are much later, and even if Di Vita's argument were true, it would not explain the opposite cases of Britain, with no Capitolia, and Africa with so many.

²⁴⁹ This point was made by Barton (op. cit. (n. 1), 278; and cf. 266–8), although he appeared reluctant to abandon the idea of a connection altogether (cf. his discussion of Sustri, pp. 306–7, which assumes that only coloniae could have Capitolia).

²⁵⁰ CIL VIII.26539; Gascou, J., La Politique municipale de l'empire romain en Afrique Proconsulaire de Trajan à Septime-Sévère (1972), 179Google Scholar. Beschaouch, A. ‘Thugga, une cité de droit latin sous Marc Aurèle: Civitas Aurelia Thugga’, in Khanoussi, M. and Maurin, L. (eds), Dougga (Thugga). Études épigraphiques (1997), 61–73Google Scholar discusses the possibility that the city's earlier title Civitas Aurelia Thugga (used in the second half of the second century) indicates some particular favour from Marcus Aurelius (r. a.d. 161–180), and suggests that this was the award of the Latin right.

²⁵¹ Barton, op. cit. (n. 1), 308.

²⁵² CIL VIII.2388 = ILS 5554.

²⁵³ See n. 195 above.

²⁵⁴ Cagnat, R. and Ballu, A., ‘Communication, Séance de la Commission de l'Afrique du Nord’, Bulletin archéologique du Comité des travaux historiques et scientifiques (1898), clvi–clviiiGoogle Scholar; Gsell, op. cit. (n. 246), 140; Eingartner, J., ‘Fora, Capitolia und Heiligtümer im westlichen Nordafrika’, in Schalles, H.-J. et al. (eds), Die römische Stadt im 2. Jahrhundert n. Chr. (1992), 234–6Google Scholar.

²⁵⁵ Carthage: see n. 211 above. Cirta: see n. 196 above. Theveste: see n. 208 above.

²⁵⁶ C.Th. 11.1.34 (law of 25 February 429).

²⁵⁷ cf. Gros, op. cit. (n. 88), 193–6.

²⁵⁸ Barton, op. cit. (n. 1), 279–81; Gsell, op. cit. (n. 246), 143–5 (Lambaesis).

²⁵⁹ Gsell, op. cit. (n. 246), 148 (Cuicul); Gsell, S. and Joly, C. A., Khamissa, Mdaourouch, Announa. Fouilles exécutées par le Service des Monuments Historiques de l'Algerie (1918), 70–3 (Thibilis)Google Scholar.

²⁶⁰ Gsell, op. cit. (n. 246), 137–9; Ward-Perkins, op. cit. (n. 253), 394; Barton, op. cit. (n. 1), 313.

²⁶¹ Althiburos (but in its own enclosure), Mopht…. (Mons), Pupput, Segermes, Thuburbo Maius, Thubursicu Numidarum, Thugga (added to the forum later).

²⁶² Abthungi, Lambaesis, Numluli, Maraci, Thamugadi, Ucubi, Volubilis.

²⁶³ ILT 699.

²⁶⁴ Barton, op. cit. (n. 1), 313: the head was 1.35 m high, and Barton calculates that the seated figure would have been some 7 m high, or five times life-size.

²⁶⁵ CIL VIII.26524 = ILAf 521; Poinssot, C., Les Ruines de Dougga (1983), 38–9Google Scholar.

²⁶⁶ Schalles, op. cit. (n. 163), 208–9; n. 60 above.

²⁶⁷ Contra van Andringa, quoted above, n. 58.

²⁶⁸ Oea: IRT 230. Sabratha: IRT 6. Timgad: AE 1968, 647. Bulla Regia: CIL VIII.25512. Aubuzza: CIL VIII.16367 = ILS 6783 = ILTun 1628 = AE 2004, 67.

²⁶⁹ Quinn, op. cit. (n. 240), 62.

²⁷⁰ Brent Shaw notes that most African Capitolia date from the Antonine period or later and are unrelated to civic status, and that their spread is correlated with the growing power of local élites, ‘more a measure of the final acceptance of Roman identity by the urban elites of the African provinces than it was any pre-emptive strike by either side to incite a new sense of belonging’. (Shaw, B., ‘Cult and belief in Punic and Roman Africa’, in Salzman, M. R. and Sweeney, M. A. (eds), The Cambridge History of Religions in the Ancient World (2013)Google Scholar, vol. 2, 246).

²⁷¹ CIL VIII.2194.

²⁷² Abthungi, Althiburos, Avedda, Lambaesis, Mopth…, Segermes, Thuburbo Maius, Thubursicu Numidarum, Tinfadi/Henchir Medkis, Volubilis, and Henchir el Gonaï (cf. Bianchi, op. cit. (n. 1), 401). The fact that at Thuburbo Maius the temple was dedicated by the proconsul of Africa (‘dedicante L. Octavio Cornelio Salvio Iuliano Aemilia[no pro]cos.’) is absolutely standard in cases of municipally-funded construction and does not imply that the proconsul paid for it; the city usually had to seek imperial permission for the expenditure of large sums on public building, and this permission was negotiated by the provincial governor, who was then frequently involved in the opening ceremony.

²⁷³ Four of the certain Capitolia (Thugga, Numluli, Sala and Timgad) and three of the possible ones (Belalis Maior, Thagura and Aïn Nechma). Cf. Barton, op. cit. (n. 1), 277; see n. 195 above.

²⁷⁴ CIL VIII.26121.

²⁷⁵ On whether gymnasium in such contexts means ‘a gymnastic display’ or a ‘distribution of olive oil’, see Fagan, G. G., ‘Gifts of gymnasia: a test case for reading quasi-technical jargon in Latin inscriptions’, ZPE 124 (1999), 263–75Google Scholar.

²⁷⁶ e.g. a statue base honouring the donor of an aqueduct at Sabratha (IRT 117 = AE 1925, 103); cf. Sandys, J. E., Latin Epigraphy. An Introduction to the Study of Latin Inscriptions (1927), 109–10Google Scholar for examples of donors offering to pay for the statues voted to them (although he misinterprets the formulae honore accepto impensam remisit, ‘having accepted the honour he repaid the expense’ and honore contentus sua pecunia posuit, ‘content with the honour alone he erected it with his own money’: the honorand is not declining the compliment of a statue, but accepting it and also paying for it).

²⁷⁷ Schalles, op. cit. (n. 163), 210.

²⁷⁸ Contra Gros, op. cit. (n. 88), 227. For other examples, see e.g. CIL VIII.15383, a dedication of a fountain also at Numluli; AE 1949, 27, dedication of the Antonine Baths at Carthage (a.d. 161/162); CIL VIII.12274, construction of public baths at Avitta Bibba (a.d. 204); CIL VIII.2706 cf. p. 1739, restoration of bath-house at Lambaesis (Severan); ILS 5776 = CIL VIII.23991, construction of fountain and tap at Giufi (dedicated also to Mercury Augustus, a.d. 233).

²⁷⁹ CIL VIII.15513.

²⁸⁰ CIL VIII.26606: ‘P(ublius) Marcius Q(uinti) f(ilius) Arn(ensi tribu) Quadratus flamen divi Aug(usti) pont(ifex) C(oloniae) I(uliae) K(arthaginis) in quinque decurias [adlectus ab imp(eratore) Antonin]o Aug(usto) Pio ob honorem flaminatus sui perpetui patriae suae theatrum cum basilicis et porticu et xystis et scaena cum siparis et ornamentis omnibus a [solo ext]ructum sua pec(unia) fec(it) idemq(ue) ludis scaenicis editis et sportulis datis et epulo et gymnasio ded(icavit).’ ‘Publius Marcius Quadratus, son of Quintus, of the Arnensian tribe, priest of the deified Augustus, pontifex in the Colony of Julia Karthago, adlected into the five jury courts by the emperor Antoninus Pius, because of the honour of his perpetual flaminate in his home town built the theatre with halls and porticus and covered walks and the stage building with the curtain mechanism and with all the decoration, with his own money from the ground upwards, and he dedicated the same with the staging of theatrical shows and by giving a handout and a feast and a gymnasium.’

²⁸¹ Lassus, J., ‘Une opération immobilière à Timgad’, in Chevallier, R. (ed.), Mélanges d'archéologie et d'histoire offerts à André Piganiol (1966), 1221–31.Google Scholar

²⁸² Fentress, E. W. B., ‘Frontier culture and politics at Timgad’, BCTH 17B (1981), 405–7.Google Scholar

²⁸³ See for instance Suet., Vesp. 8 and Section III above.

²⁸⁴ Wilson, A. I., ‘Urban development in the Severan Empire’, in Swain, S. C. R. et al. (eds), Severan Culture (2007), 307–10Google Scholar.

²⁸⁵ Timgad, apparently Severan (n. 195, above); Maraci (AE 1949, 109) and Mopth… (AE 1950, 136), both a.d. 198/211; Avedda, a.d. 212/217 (ILT 1206); T… [Henchir Medkis], a.d. 214 (CIL VIII.2194); Volubilis, a.d. 217 (ILM 45); …rda, a.d. 222/235 (ILAlg 1.1097).

²⁸⁶ Lepcis: Ward-Perkins, J. B., The Severan Buildings of Lepcis Magna (1993), 31, 52–4Google Scholar. The suggestion sometimes made, that the temple in the Severan forum at Lepcis was dedicated to Bacchus and Hercules (e.g. Thomas, E., ‘Metaphor and reality in Severan architecture: the Septizodium at Rome between “reality” and “fantasy”’, in Swain, S. C. R. et al. (eds), Severan Culture (2007), 334Google ScholarPubMed, with references to earlier literature) has no direct support, and relies on transposing Dio's mention of Severus constructing a temple to these deities from Rome to Lepcis (Dio 77.16.3); cf. Wilson, op. cit. (n. 282), 299.

²⁸⁷ Wilson, op. cit. (n. 284).

²⁸⁸ cf. Gros, op. cit. (n. 88), 196, but his suggestion that in the late second century Capitolia gave way to temples of the imperial cult is over-simplistic, being based on the temples to the Severan family at Lepcis Magna and Cuicul, and ignoring the third-century date of several of the African Capitolia (see above).

²⁸⁹ Déroche, L., ‘Les fouilles de Ksar Toual Zammel et le question de Zama’, Mélanges de l'École française de Rome, Antiquité 60 (1948), 72–9Google Scholar; Minerva may have been mentioned in the lost right-hand part of the inscription, or possibly Mars or another male deity was substituted who could have been assimilated to Caracalla.

²⁹⁰ AE 1949, 109; RIC 640, 840.

²⁹¹ Todd, op. cit. (n. 1); Blutstein-Latrémolière, op. cit. (n. 39); Eingartner, op. cit. (n. 254); Schalles, op. cit. (n. 163); Gros, op. cit. (n. 88), 220–3; although he does give consideration also to the imperial cult (and especially at pp. 229–31).

²⁹² Schalles (op. cit. (n. 163), passim, especially 211) sees an initial tight link between Forum and Capitolium in the early Imperial period breaking down by the second century, with the creation of more sanctuaries in their own enclosures; but this too is based on the assumption that many temples which are sited on the forum are Capitolia, for which the evidence is insufficient.