¹ Date: Timaeus in Ps-Skymnus, 11. 211–14; foundation legend: Justin, XLIII. 5; Athenaeus XIII. 576. Subsequent excavations have confirmed the archaeological dating established by Villard, F., La céramique grecque de Marseille (VI^e–IV^e siecles), essai d'histoire économique (1960), 76–81Google Scholar.

² Strabo IV. 1. 5: its chora could support olives and vines, but not grain.

³ Jullian, C. in ‘Aries grecque et romaine’, Journal des savants n.s. 20 (1922), 97–113,Google Scholar at 100.

⁴ Strabo IV. 1.4. Post-War excavations suggest that the coastline in antiquity was on average some 100 m back from the modern north quay, but that the process of encroachment into the Vieux-Port was already well under way in antiquity: see e.g. the reports of Rolland, H. and Benoit, F. in Gallia 5 (1947), 155–7Google Scholar; (1948), 208; 8 (1950), 116; Benoit, F., ‘L'évolution topographique de Marseille: le port et l'enceinte à la lumière des fouilles’, Latomus 31 (1972), 54–7Google Scholar. Cf. Gantès, L.-Fr. and Moliner, M., Marseille, itinéraire d'une mémoire, cinq années d'archéologie municipale (1990), 41–2Google Scholar, for more recent data.

⁵ Avienus ll. 704–12; Caesar, BC II. 1. 3. Avienus' choice of language is at the mercy of his metre. Caesar is concerned to emphasize the difficulty of besieging Marseille. One of these springs was almost certainly the Lacydon, the name of a local water-divinity (Jullian, C., ‘Le port du Lacydon et le ruisseau sacré des Marseillais’, Provincia 1 (1921), 1–6)Google Scholar, by whose name the whole Vieux-Port was known in antiquity (Mela II. 5. 77).

⁶ e.g. Pirenne, H., Mahomet et Charlemagne ³ (1937)Google Scholar, passim, ranging from triumph (p. 77: ‘Marseille nous donne tout à fait l'impression d'un grand port’) to disaster (p. 168: ‘Marseille est mort à cette époque’).

⁷ Marseille is absent from the index in both Hodges, R., Dark Age Economics: the Origins of Towns and Trade, A.D. 600–1000 (1982)Google Scholar and Randsborg, K., The First Millennium A.D. in Europe and the Mediterranean, an Archaeological Essay (1991)Google Scholar. It figures but once in Hodges, R. and Whitehouse, D., Mohammed, Charlemagne and the Origins of Europe: Archaeólogy and the Pirenne Thesis (1983), in an inaccurate reference (p. 23)Google Scholar.

⁸ There is as yet no definitive publication of the Bourse, the key site and pioneering excavation (1968–84) of French urban archaeology, but the late antique material has been discussed in a series of excellent surveys: Bonifay, M., ‘Eléments dévolution des céramiques de l'antiquité tardive à Marseille d'après les fouilles de la Bourse (1980–1981)’, RAN 16 (1983), 285–346CrossRefGoogle Scholar; Foy, D. and Bonifay, M., ‘Eléments d'évolution des verreries de l'antiquité tardive à Marseille d'après les fouilles de la Bourse (1980)’, RAN 17 (1984), 289–308CrossRefGoogle Scholar; Bonifay, M., ‘Observations sur les amphores tardives à Marseille d'après les fouilles de la Bourse (1980–1984)’, RAN 19 (1986), 269–305CrossRefGoogle Scholar. The results of the various excavations within the city in the late 1980s are now conveniently assembled in L.-Fr. Gantès and M. Moliner, op. cit. (n. 4). Post-war finds in general are summarized in Gallia 5 (1947). 155–60; 6 (1948). 207–9; 8 (1950). 116–17; 11 (1953), 100–6; 12 (1954), 426–9; 18 (1960), 286–90; 20 (1962), 687; 22 (1964), 580–5; 25 (1967), 404–5; 27 (1969), 423–30; 30 (1972), 520–4; 32 (1974), 512–18; 35 (1977), 520–5; 44 (1986), 413–26; Gallia Informations (1987–1988) ii, 244–9; (1990) i–ii, 168–74. For pre-War discoveries, Clerc, M., Massalia: histoire de Marseille dans l'antiquité, des origines à la fin de l'empire romain d'occident, 2 vols (1927–1929)Google Scholar, a comprehensive historical survey which although inevitably dated remains fundamental, and the catalogue in Benoit, F., Forma Orbis Romani: carte archéologique de la Gaule romaine, V, Bouches-du-Rhône (1936), 17–41Google Scholar and map (hereafter FOR). For other major recent excavations within the city, see also d'Archimbaud, G. Demians, ‘Les fouilles de Saint-Victor de Marseille’, CRAI (1971), 87–117Google Scholar; d'Archimbaud, G. Demians, Allais, J.-M., and Fixot, M., ‘Saint-Victor de Marseille: fouilles récentes et nouvelles interprétations architecturales’, CRAI (1974), 313–46Google Scholar; and for early Christian topography in general (and an excellent bibliography), Guyon, J., ‘Marseille’, in Picard, N. Gauthierand J.-Ch. (eds), Topographie chrétienne des cités de la Gaule iii (1986), 121–33Google Scholar. Both Les dossiers d'archéologie 154 (1990) and the local journal Marseille 160 (1991) are devoted to ancient and medieval Marseille and contain much valuable (and similar) discussion within their popular formats: in particular, both Bonifay and Guyon stress the positive aspects of the city's history in late antiquity. In a wider context, Rivet, A. L. F., Gallia Narbonensis (1988)Google Scholar, provides a convenient recent consideration of the Roman province, but its discussions of the topography of individual cities require significant modification in light of the weight of archaeological material recently excavated and published in the Midi. Despite this, however, the rural archaeology of the Marseille basin remains virtually non-existent: for purposes of comparison, see e.g. the recent summary of rural settlement in the Roman period around the nearby Étang de Berre: Leveau, P., ‘Villas and Roman settlement in Basse-Provence’, in Barker, G. and Lloyd, J. (eds), Roman Landscapes: Archaeological Survey in the Mediterranean Region (1991), 169–75Google Scholar. Finally, Février, P.-A. et al. , La Provence des origines à l'an mil (1989)Google Scholar, though general, is full of up-to-date analysis and insights.

⁹ The baptistery survived in an increasingly mutilated form until modern times: Grosson, J. B. B., Recueil des antiquités et monuments marseillois qui peuvent intéresser l'histoire et les arts (1773), 50, 168–9Google Scholar, pl. 23. It was excavated in a haphazard fashion between 1850 and 1854 during the demolitions required for the building of a new cathedral: Roustan, F., La Major et le premier baptistère de Marseille (1905)Google Scholar; but see also Barral, X.Altet, I. and Drocourt, D., ‘Le baptistère paléochrêtien de Marseille’, Archéologia 73 (1974), 6–19,Google Scholar for a critical evaluation of these excavations and of Roustan's work.

¹⁰ The largest of these is the baptistery at Aix (approximately 14 m a side): the others are at Fréjus, Riez, and Cimiez: Guyon, J., ‘Baptistères et groupes épiscopaux de Provence: élaboration, diffusion et devenir d'un type architectural’, Actes du XT^e congrès international d'archeologie chrétienne (1989), vol. 2, 1427–49Google Scholar.

¹¹ Roberti, M. Mirabella and Paredi, A., Il battistero ambrosiano di San Giovanni alle fonti (1974)Google Scholar.

¹² Roustan, op. cit. (n. 9). The mosaics outside the baptistery must belong to the contemporary episcopal group, otherwise unknown. Besides the mosaics, four massive white marble column bases (of the sixteen dictated by the plan), and traces of wall and floor decoration in polychrome marble were among the finds.

¹³ Dating: Stern, H., ‘Mosaiques de pavement préromanes et romanes en France’, Cahiers de civilisation médiévale 5 (1962), 14–15,CrossRefGoogle Scholar and figs 1–3, facing p. 16. Not all the excavated mosaics belong to the late antique series: ibid., 14.

¹⁴ Guyon, op. cit. (n. 10), 1443–6, for the difficulty of accurately dating the Provencal group. Barral I Altet and Drocourt, op. cit. (n. 9), 18–19, for cautious attribution to Proculus. But the dangers of attributing buildings to the famous are obvious, and there are now some archaeological grounds for dating the whole episcopal group at Aix, for example, including by implication the baptistery, as late as c. 500: Guild, R., Guyon, J. and Rivet, L., ‘Les origines du baptistère de la cathédrale Saint-Sauveur: étude de topographie aixoise’, RAN 16 (1983), 182–5, 213–14CrossRefGoogle Scholar.

¹⁵ Pan.Lat. VI. 18–19.

¹⁶ Olympiodorus, fr. 22. Gregory of Tours, Hist. VI. 11.

¹⁷ The date derives from pottery recovered from its foundation trenches: M. Euzennat, ‘Fouilles de la Bourse à Marseille’, CRAI (1976), 543.

¹⁸ Ancient observations: FOR, op. cit. (n. 8), 18, nos 61–2. Bourse excavations: Euzennat, M. and Salviat, F., ‘Les fouilles de Marseille (mars–avril 1968)’, CRAI (1968), 144–59Google Scholar. See also Euzennat, op. cit. (n. 17), 529–52; Euzennat, et al. , ‘Les fouilles de la Bourse à Marseille (campagnes 1975–6)’, RAN 10 (1977), 235–46Google Scholar. Butte des Carmes: Bertucchi, G. and Gantès, L.-Fr., ‘Les fortifications de Marseille et les couches archaïques sur la Butte des Carmes’, AMM 3 (1981), 61–72;Google ScholarGallia 44 (1986), 419–21;Google Scholar Gantès and Moliner, op. cit. (n. 4), 43–4.

¹⁹ The omission of the low-lying vallon du Lazaret is implied for practical defensive reasons, and confirmed by the existence in this area of a poorly-known Roman cemeterial zone: Clerc, op. cit. (n.8), 276–8; FOR, op. cit. (n.8), 24, 26, 29, 32, no. 59. The Pistoles and Phocéens sites were however occupied from the fifth century B.C. onwards (Gantès and Moliner, op. cit. (n. 4), 25–6, 31–5) and seem likely to have been included within the Hellenistic rampart.

²⁰ Guyon, op. cit. (n. 8), 126.

²¹ Siege: Caesar, BC I. 34–6, 56–8; II. 1–22. Fore-wall: Bonifay (1983), op. cit. (n.8), 287–9; Gallia 44 (1986), 418Google Scholar. Documentary, structural, and cemeterial evidence all point to the general survival of the earlier circuit: cf. Bonifay, M., ‘Fouilles récentes sur le chantier de la Bourse: niveaux de l'antiquité tardive et du haut moyen-âge’, Archéologie du Midi Médiéval 3 (1981), 42;Google ScholarFévrier, P.-A., ‘Aux origines de quelques villes médiévales du Midi de la Gaule’, RSL 49 (1983), 324–8Google Scholar. Evidence from the Bourse suggests that the siege of 49 B.C. caused only superficial damage: Euzennat et al., op. cit. (n. 18), 245. The sum spent on the walls by Crinas in the first century A.D. (Pliny, NH XXXIX. 5. 9–10) probably went on repairs rather than a wholesale rebuild.

²² See e.g. Butler, R. M., ‘Late Roman town-walls in Gaul’, Arch.Journ. 116 (1959), 25–60;Google ScholarFévrier, P.-A. in Duby, G. (ed.), Histoire de la France urbaine. 1. La ville antique (1980), 399–421Google Scholar.

²³ Gantès and Moliner, op. cit. (n.4), passim, and summary, 93–6. Other late antique occupation levels within the city: Gallia 44 (1986), 423–5 and (less certainly), 6 (1947), 156–9; 18 (1960), 267–8, 22 (1964), 583; Benoit, F., ‘Topographie antique de Marseille: le théâtre et le mur antique de Crinas’, Gallia 24 (1966), 1–12CrossRefGoogle Scholar.

²⁴ Gantès and Moliner, op. cit. (n. 4), 41–2.

²⁵ Gallia 42 (1984), 423–6; Gallia Informations (1987–88) ii, 321–7.

²⁶ Février, P.-A., ‘Approche de villes médiévales de Provence: réflexions à partir de deux fouilles faites à Fréjus et Aix’, Rendiconti di atti della pontifida accademia rumana di archeologia 53–4 (1980–1982), 369–82,Google Scholar esp. 371–2, for the absence of grey stamped ware at the Clos de la Tour site, implying abandonment before the late fourth century.

²⁷ Gallia 28 (1970), 448–51; Barruol, G., ‘Riez: un centre administratif et religieux des Alpes du Sud’, Archeologia 21 (1968), 20–6Google Scholar.

²⁸ Abandonments, e.g. the Jardin du Grassi, École des Beaux Arts, Enclos Laugier, Aire du Chapître sites: Gallia 16 (1958), 419; 18 (1960), 300; 35 (1977), 512; 44 (1986), 386; Boiron, R., Landure, C., and Nin, N., Les fouilles de l'Aire du Chapître (1986)Google Scholar. Disused wall-circuit and burials: Benoit, F., ‘Recherches archéologiques dans la région d'Aix-en-Provence’, Gallia 12 (1954), 294–300CrossRefGoogle Scholar. Christian topography summary and full bibliography: Guyon, J., ‘Aix-en-Provence’, in Topographie chrétienne, op. cit. (n.8) ii (1986), 17–28Google Scholar; medieval topography: Fevrier, op. cit. (n. 26), 377–82.

²⁹ For summaries of late antique occupation on the Bourse site, see Bonifay (1981), op. cit. (n. 21), 37–48; Bonifay, M. and Guéry, R., ‘L'antiquité tardive sur le chantier de la Bourse à Marseille’, in Archéologie médiévale en Provence-Alpes-Côte d'Azur 1970–1982 (1983), III–14Google Scholar.

³⁰ For the Puget III site: Gantès and Moliner, op. cit. (n. 4), 53–8.

³¹ The adjacent area of the port, the Corne, can only have been in limited use in this period (cf. Section 11), and the suburb extends well away from it to the north. The road into the city, or the springs with their potential for industrial activity, may have been more significant in its development.

³² There is no evidence whatsoever for the existence of a church in this area. A vital extramural focus of Christian funerary activity did exist, but opposite the city on the south bank of the port around an elaborate church complex at Saint-Victor. Occupation may conceivably have extended into this traditionally cemeterial zone (Gantès and Moliner, op. cit. (n. 4), 61–2), but it would be very premature to claim this as part of a general phenomenon of suburban settlement around the port.

³³ Some of the interventions at the Bourse may belong to this period (Foy and Bonifay, op. cit. (n. 8), 290), but the resumption of activity here is first clearly marked by the appearance of tenth-century forms of vetrina pesante: Bonifay, M., Paroli, L., and Picon, J., ‘Ceramiche a vetrina pesante scoperte a Roma et a Marsiglia’, Archeologia medievale 13 (1986)Google Scholar, esp. 85–6. For the general problem, cf. M. Fixot, ‘Archéologie médiévale en Provence’, Prov.Hist. 40 (1990), 455–64.

³⁴ Paulinus of Pella retired to a domus urbana at Marseille, (Eucharisticos 1. 527)Google Scholar. Marseille still seems to have contained large houses in the late sixth century: Gregory of Tours, Hist. VI. 11 (Dynamius' house); IX. 22. But further north, Venantius Fortunatus, who regularly corresponded with the Marseille élite, could bring himself expansively to praise wooden houses (Carm. IX. 15), although contemporary evidence suggests that stone was still much the preferred building medium (e.g. for churches, Gregory of Tours, GC 71).

³⁵ Clothing: Ep. Arelatenses (MGH, Ep. III) 49, 53; Gregory I, Reg. VI. 10. Slaves: Reg. VI. 10; Vita Eligii 10; Vita Boniti 3. Only the latter source refers explicitly to Marseille rather than Provence in general. Timber was exported from the papal patrimony to Alexandria under Gregory the Great, but nothing proves that the relevant estates were those in Provence.

³⁶ Recent summaries for DSPs: Association CATHMA, ‘La céramique du haut moyen-âge en France méridionale: éléments comparatifs et essai d'interprétation’, in La ceramica medievale nel mediterraneo occidentale (1986), 27–54, esp. 40–2; Rigoir, Y. and J., ‘Les derivées des sigillées dans la moitié sud de la France’, SFECAG, Actes du Congrès de Reims (1985), 49–56Google Scholar; Rivet, L. in Février, P.-A. and Leyge, F., Les premiers temps Chrétiens en Gaule méridionale (1986), 176Google Scholar.

³⁷ Foy and Bonifay, op. cit. (n. 8), 308.

³⁸ In advance of his forthcoming book on the subject, see Bertucchi, G., ‘Le vin de Marseille’, Dossiers d'archéologie 154 (1990), 44–9Google Scholar.

³⁹ Carthage: e.g. Fulford, M. G., ‘Carthage: overseas trade and the political economy, c.A.D. 400–700’, Reading Medieval Studies 6 (1980), 68–80Google Scholar.

⁴⁰ Euzennat, op. cit. (n. 17), 545–50; Bonifay (1983), op. cit. (n. 8), 289–90. The Corne had become a graveyard for shipping by the third century. For a graphic summary of the fifth-century environmental conditions, Jourdan, L., La faune du site gallo-romain et paléochrétien de la Bourse (1976), 302–4Google Scholar: numerous flies and various species of vulture hovered over the accumulated debris.

⁴¹ For the phasing, Bonifay (1983), op. cit. (n. 8), 290–7; ibid., 303–26 for dating, with important revisions in Bonifay (1986), op. cit. (n. 8), 270–1. Keyhole excavations on the other side of the Corne corroborate much of this sequence: Cavaillès-Llopis, M.-T., ‘Céramiques de l'antiquité tardive à Marseille’, Documents d'Archéologie Méridionale 9 (1986), 167–95CrossRefGoogle Scholar.

⁴² See n. 8 for full references and Hitchner, R. B., ‘Meridional Gaul, trade and the Mediterranean economy in late antiquity’, in Drinkwater, J. F. and Elton, H. W. (eds), Fifth-Century Gaul: a Crisis of Identity? (1992), 122–31Google Scholar, for an attempt to put this information in its regional economic context.

⁴³ Only twenty or so sherds of the eastern Late Roman C were found in all periods in the main Come excavation: Bonifay (1983), op. cit. (n. 8); cf. Cavaillès-Llopis, op. cit. (n. 41), 171, fig. 6.

⁴⁴ Bonifay (1983), op. cit. (n. 8), 304; cf. Cavaillès-Llopis, op. cit. (n.41), 171, fig. 8.

⁴⁵ Association CATHMA, op. cit. (n. 36), 34–5. Even the contemporary phase at Saint-Victor, just across the Vieux-Port, has produced far lower proportions of ARS, perhaps indicative of the particular redistributive role of the port area.

⁴⁶ Full publication of these and other common ware imports to southern Gaul by Bonifay et al. is forthcoming in Actes du IV^e congrès international sur la céramique mediévale en Méditerranée occidentale, held at Lisbon in 1987. But Bonifay (1986), op. cit. (n. 8), 299 implies that the quantities involved are small. Cf. Fulford, op. cit. (n. 39), 72, for imported coarse wares at Carthage.

⁴⁷ Bonifay, M. and Villedieu, F., ‘Importations d'amphores orientales en Gaule (V^e-VII^e), in Déroche, V. and Speiser, J.-M. (eds), Recherches sur la céramique byzantine (1989), 17–46Google Scholar; Bonifay, M., Congès, G. and Leguilloux, M., ‘Amphores tardives (V^e-VII^e siècle) à Aries et à Marseille’, in Amphores romaines et histoire économique: dix ans de recherche (1989), 660–3Google Scholar.

⁴⁸ e.g. Keay, S. J., Late Roman Amphorae in the Western Mediterranean, a Typology and Economic Study: the Catalan Evidence, 2 vols (1984), 428–30Google Scholar; Panella, C., ‘Le merci: produzioni, itinerari e destini’, in Giardina, A. (ed.), Società romana e impero tardoantico, iii, le merci, gli insediamenti (1986), 451–4Google Scholar.

⁴⁹ Keay, op. cit. (n. 48), 430–1; Fulford, op. cit. (n. 39), 71.

⁵⁰ Arthur, P., ‘Naples: notes on the economy of a Dark Age city’, in Malone, C. and Stoddart, S. (eds), Papers in Italian Archaeology (1985) iv, 247–59,Google Scholar at 256.

⁵¹ Keay, op. cit. (n.48), 427–8; Arthur, op. cit. (n. 50), 255–6; Fulford, op. cit. (n. 39), 75–6; Hitchner, op. cit. (n.42), 161–2.

⁵² Bonifay (1986), op. cit. (n. 8), 300.

⁵³ Bonifay and Villedieu, op. cit. (n. 47), 37–9.

⁵⁴ On this basis it was suggested from the imported pottery that there might have been an increase in commercial activity between the early fifth and late sixth centuries: Bonifay (1983), op. cit. (n. 8), 345. But if anything the amphora data imply the opposite: Bonifay (1986), op. cit. (n. 8), 298. I am unconvinced that any meaningful comparisons between periods can be made for absolute volumes of trade.

⁵⁵ Bonifay (1986), op. cit. (n.8), 300–1.

⁵⁶ Papyrus: Hist. v. 5. The same passage (quoted n. 94) highlights oil among the other (unnamed) commodities imported to Marseille. Oil and liquamen: Hist. IV. 43: ‘igitur advenientibus ad cataplum Massiliensim navibus transmarinis, Vigilii archdiaconi homines septuaginta vasa quas vulgo orcas vocant olei liquaminisque furati sunt’.

⁵⁷ Marculf, , Supplementum 1 (MGH Formulae, ed. , Zeumer, p. 107)Google Scholar.

⁵⁸ Gesta Dagoberti 18; Levillain, L., Examen critique des chartes de Corbie (1902), no. 15, 235Google Scholar. Cf. Section III.

⁵⁹ Bonifay (1986), op. cit. (n. 8), 301, offers a number of alternative explanations, which are not convincing in the face of documentary evidence for the continuing importance of oil in Francia and of its storage at Marseille.

⁶⁰ Gaza wine at Tours: Hist. VII. 29; at Lyon: GC 64. Carthage LRA 4 at the Bourse: Bonifay and Villedieu, op. cit. (n.47), 20–1; in Gaul: ibid., 27–9, fig. 9, p. 30.

⁶¹ Hist. IX. 22.

⁶² Hides from Cordoba are mentioned in a royal grant to Corbie in 716 on imports to Fos: see Levillain, op. cit. (n. 58), and n. 72 below. For increasing use of perishable containers in late antiquity, see e.g. the remarks in C. Panella, ‘Le anfore tardoantiche: centri di produzione e mercati preferenziali’, in Giardina (ed.), op. cit. (n.48), 251–72.

⁶³ Bonifay, Congès, and Leguilloux, op. cit. (n.47), 662, fig. 1.

⁶⁴ Bonifay and Villedieu, op. cit. (n.47), 39, graphs 7, 8.

⁶⁵ Cassiodorus, Variae in. 41; Gesta Dagoberti 18.

⁶⁶ Austrasian corridor: Duprat, E.-H., ‘Le couloir austrasien du VI^e siècle’, Mémoires de l'Institut Historique de Provence 20 (1943–1944), 36–65Google Scholar (with caution); on the partitions in general, Ewig, E., ‘Die fränkischen Teilungen und Teilreiche (511–613)’, and ‘Die fränkischen Teilreiche im 7.Jahrhundert (613–714)’, collected in his Spätantikes und Fränkisches Gallien (1976) i, 114–230Google Scholar.

⁶⁷ Conflicts connected with Marseille and its magnates: Hist. IV. 43; VI. 11; VI. 24; vi. 31; VI. 33; VIII. 5; VIII. 12; VIII. 20; IX. 22. I cannot enter into the multifaceted dynamics of these disputes or the various partitions (n. 66) in this paper, although I believe that they have often been misinterpreted: I hope to return to these subjects elsewhere.

⁶⁸ Provençal patricii: Buchner, R., Die Provence in merowingischer Zeit (1933), 92–6Google Scholar.

⁶⁹ Hist. IV. 43.

⁷⁰ Gesta Dagoberti 18.

⁷¹ MGH Dipl. no. 61 (Clovis III, 691); no. 67 (Childebert III, 695); n. 82 (Chilperic II, 716). Childebert's diploma records the abbey's concession of the right in return for land in Berry, but the diploma of 716 repeats the terms of the tractoria of Clovis III. It is impossible to know what underlies this exchange and its apparent repudiation: cf. Ganz, D. and Goffart, W., ‘Charters earlier than 800 from French collections’, Speculum 65 (1990), 906–32,CrossRefGoogle Scholar at 914.

⁷² Levillain, op. cit. (n. 58). Chilperic II here confirms grants of his predecessors, now lost. See the translation and discussion of this document by Pirenne, op. cit. (n. 6), 71–3.

⁷³ Pirenne, H., ‘Le cellarium fisci: une institution économique des temps mérovingiens’, reprinted in his Histoire économique de l'occident médiéval (1951), 104–12Google Scholar; Ganshof, F.-L., ‘Les bureaux de tonlieu de Marseille et de Fos. Contribution à l'histoire des institutions financières de la monarchie franque’, Études historiques à la mémoire de N. Didier (1960), 125–33Google Scholar.

⁷⁴ Lafaurie, J., ‘Monnaies de bronze marseillaises du VI^e siècle’, BSFN (1973), 480–2Google Scholar; Brenot, C., ‘Rapport préliminaire sur les monnaies des fouilles de Marseille’, Actes du VIII^e congrès international de numismatique (1976), 217–26Google Scholar; Brenot, C., ‘Monnaies en cuivre du VI^e siècle frappées à Marseille’, in Mélanges de numismatique, d'archéologie et d'histoire offerts à Jean Lafaurie (1980), 181–8Google Scholar; Lafaurie, J., ‘Les monnaies de Marseille du VI^e au VIII^e siècle’, BSFN 36 (1981), 68–73;Google ScholarLafaurie, J., ‘Deux monnaies en argent du VI^e siècle trouvées à Saint-Blaise’, BSFN 38 (1983), 411–13Google Scholar. The copper coinage has now turned up on sites in Provence besides Marseille, but the city remains by far the most likely mint site.

⁷⁵ Brenot (1980), op. cit. (n. 74), 185. See also her summary in Février and Leyge, op. cit. (n. 36), 197–8.

⁷⁶ Grierson, P. and Blackburn, M., Medieval European Coinage. I. The Early Middle Ages (1986), 115–16Google Scholar.

⁷⁷ Brenot (1960), op. cit. (n. 74), 185–6.

⁷⁸ Grierson and Blackburn, op. cit. (n. 76), m–14, 117–28.

⁷⁹ ibid., 128–31; S. E. Rigold, ‘An imperial coinage in southern Gaul in the sixth and seventh centuries’, Num.Chron.6 14 (1954), 93–133, including the fullest catalogue; Grierson, P., ‘The patrimonium Petri in illis partibus and the pseudo-imperial coinage in Frankish Gaul’, Revue belge de numismatique 105 (1959), 95–111;Google Scholar J. Lafaurie (1981), op. cit. (n. 74), 70–3.

⁸⁰ Hence the coinage is now generally called quasi-rather than pseudo-imperial, because it was not intended to pass as a genuine imperial issue.

⁸¹ Isolated coins are known from other southern centres, but only on a very irregular or imitative basis: Rigold, op. cit. (n. 79), 103. All four main mints struck their first quasi-imperials in the name of Justin II: Grierson, op. cit. (n. 79), 97.

⁸² Grierson, op. cit. (n. 79), 95. This article provides the most closely argued review of the problem, but its contentions are retracted in Grierson and Blackburn, op. cit. (n. 76), 130, which broadly follows Lafaurie (1981), op. cit. (n. 74), in attributing the introduction of the coinage to some sort of royal numismatic compromise. If this is so, it would emphasize further the importance of Marseille in an integrated Merovingian economy, but the theory begs several questions, which I cannot develop here.

⁸³ For the emergence in the fifth century of episcopal sees at these hitherto secondary centres, see e.g. S. T. Loseby, ‘Bishops and cathedrals, order and diversity in the fifth-century urban landscape of southern Gaul’, in Drinkwater and Elton, op. cit. (n. 42), 144–55. Viviers overlooks the Rhône and controls the Défile de Donzère. Uzès lies on the overland route to Aquitaine, and close to the border with Visigothic Septimania.

⁸⁴ e.g. Rigold, op. cit. (n. 79), 103–22.

⁸⁵ ibid., 119.

⁸⁶ The activity of the royal master-minter Eligius at Marseille and Arles (and the superior fineness he temporarily maintained in his issues there) nevertheless implies the enduring significance of this coinage: see J. Lafaurie, ‘Eligius monetarius’, Rev.Num.6 19 (1977), 111–51.

⁸⁷ Grierson and Blackburn, op. cit. (n. 76), 130–1, for a summary account of the royal coinage.

⁸⁸ ibid., 117.

⁸⁹ Explicit in Gregory I, Reg. VI. 10, requesting payments in kind, not coin, from the director of the papal estates in the Marseille-Arles region because Gallic solidi were not legal tender in Italy. The Pope had however accepted them previously: Reg. III. 33.

⁹⁰ Distribution map of Marseille issues and discussion in Lafaurie (1981), op. cit. (n. 74), 70–3: it is significant that the quasi-imperials appear to have influenced contemporary Frisian issues.

⁹¹ cf. James, E., The Merovingian Archaeology of South-West Gaul (1977), 230–4Google Scholar. But there remains insufficient information to hypothesize sensibly about the workings of this system, especially given the dangers of extrapolating the dynamics of trade from coin finds: Grierson, P., ‘Commerce in the Dark Ages: a critique of the evidence’, TRHS ⁵ 9 (1959), 123–40Google Scholar.

⁹² Reg. XI. 34, 38, 40–2. Note that Gap and Angers are not on any direct route north and indeed the letter to Gap seems to have been dictated by other imperatives, while the identification of Licinius as bishop of Angers is not entirely certain. For routes, cf. the letters Gregory wrote for Augustine in 596 to Lérins, Marseille, Aix, Arles, Vienne, Lyon, Autun, Tours: Reg. VI. 49–54, 56–7.

⁹³ Marculf, op. cit. (n. 57), Suppl. 1: Gesta Dagoberti 18.

⁹⁴ Hist. v. 5: ‘O si te [Felix] habuisset Massilia sacer — dotem! Numquam naves oleum aut reliquas species detulissent, nisi cartam tantum, quo maiorem opportunitatem scribendi ad bonos infamandos haberes.’

⁹⁵ Hist. VI. 2: ‘Nam cum Marsiliensim portum propter regum discordias adire ausi non essent, Agathae urbem … advenerunt.’

⁹⁶ The Jews who preferred to leave Clermont rather than accept baptism returned to Marseille, suggesting the city already housed a significant Jewish community: Hist v. 11: see Goffart, W., ‘The conversion of Avitus of Clermont and similar passages in Gregory of Tours’, in Neusner, J. and Freriches, E. R. (eds), To See Ourselves as Others See Us: Christians, Jews, and ‘Others’ in Late Antiquity (1985), 473–97Google Scholar. They faced similar attempts at forcible conversion there: Reg. 1.47, a letter in which Pope Gregory refers to Jews travelling in the region of Marseille pro diversis negotiis. See also Gregory of Tours, Hist. VI. 17; GC95.

⁹⁷ Hist. VIII. 1; vi. 32.

⁹⁸ Hist. VI. 11: ‘cum signis et laudibus diversisque honorum vexillis’.

⁹⁹ For pre-Roman Marseille, Clavel-Lévêque, M., Marseille grecque (1977)Google Scholar, supplemented by recent excavation data (summary in Gantèsand Moliner, op. cit. (n. 4), 67–84) and the forthcoming proceedings of two recent conferences devoted to ‘Marseille grecque’ and ‘Marseille et la Gaule’.

¹⁰⁰ Caesar, BC I. 34–6: Velleius Paterculus II. 50. 3 and Dio XLI. 19 for less jaundiced accounts of the city's actions.

¹⁰¹ BC 1. 36.

¹⁰² Strabo IV. I. 6. Archaeology has begun to confirm this: e.g. Arcelin, P., ‘Arles protohistorique’, and ‘Les fouilles du Jardin d'Hiver’, in Sintès, C. (ed.), Du nouveau sur l'Arles antique (Revue d'Arles 1) (1987), 17–31Google Scholar.

¹⁰³ Maps of Gallic civitates in Histoire de la France urbaine I, op. cit. (n. 22), 9,97. Summary of territory left to Marseille: Rivet, op. cit. (n. 8), 222–4.

¹⁰⁴ Especially Strabo IV. 1.5.

¹⁰⁵ Seneca, de Clem I. 15. Clerc, op. cit. (n. 8), ii, 313–36 for an account of the cultural role of the early imperial city.

¹⁰⁶ Tacitus, Agricola 4.

¹⁰⁷ The documents do give mixed reviews of its wine: see e.g. Bertucchi, op. cit. (n. 38). Marseille is singularly deficient in surviving inscriptions in all periods, but the lack of inscriptions in other cities referring to its inhabitants or associates also tends to emphasize its isolation within the regional economy, compared for example to Aries, Nîmes, or even Aix.

¹⁰⁸ Grenier, A., ‘La Gaule romaine’, in Frank, T., An Economic Survey of Ancient Rome iii (1959), 473–9Google Scholar, assembles the economic data for Arles in summary form. For the navicularii in particular, see CIL XII. 672, 692, 704, 718, 853, 982 and esp. CIL III. 14165⁸, and the articles by Christol, M., ‘Remarques sur les naviculaires d'Arles’, Latomus 30 (1971), 643–63Google Scholar and ‘Les naviculaires d'Arles et les structures du grand commerce maritime sous l'empire romain’, Prov.Hist. 32 (1982), 5–14.

¹⁰⁹ e.g. Expositio totius mundi 58; Ausonius, Ordo urbium nobilium 10.

¹¹⁰ Ep. Arel., op. cit. (n. 35), 8.

¹¹¹ For recent excavations in Arles, see summaries in Gallia 44 (1986), 388–402; Gallia Informations (1987–1988) ii, 229–39; (1990) i–ii, 229–39; C. Sintè et al. (1987), op. cit. (n. 102), and Sintès, C. et al. , Carnets de fouilles d'une presqu'île (Revue d'Arles 2) (1990)Google Scholar. Preliminary pottery analysis: Piton, J., ‘Étude comparative entre les importations africaines et les productions de la Vallée du Rhône, fin III^e–début IV^e siècle’, SFECAG, Actes du congres d'Orange (1988), 81–90Google Scholar.

¹¹² e.g. the mid-sixth-century abandonments of the circus and the late antique occupation around it, and the seemingly related sequence nearer the city on the Hôpital Van Gogh site, with structures and the road leading to the circus going out of use around this time: Sintès et al. (1990), op. cit. (n. 111), 59–62; (1987), op. cit. (n. 102), 44–8. Occupation of the Esplanade site in the southern suburbs had also ceased by this time (ibid., 37). The luxury houses in the Trinquetaille suburb across the Rhône had already been deserted after some fifth-century reoccupations (ibid., 80–8).

¹¹³ Conc. Gall, i (CCSL 148), 14.

¹¹⁴ ibid., 54–5. cf. Griffe, E., La Gaule chrétienne à l'époque romaine (1964–1966), i, 336–40Google Scholar. For Proculus’ aspirations see also n. 123 below.

¹¹⁵ Jerome, Ep. 125.

¹¹⁶ Gennadius, , De viris illustribus 62Google Scholar. Marrou, H.-I., ‘Le fondateur de Saint-Victor de Marseille, Jean Cassien’, Prov.Hist. 16 (1966), 297–308Google Scholar; Chadwick, H., John Cassian (1967)Google Scholar. The sites of these monasteries remain unknown, despite the force of local historiographic tradition.

¹¹⁷ Refugees: Prinz, F., Frühes Mönchtum im Frankenreich (1965), 47–58Google Scholar.

¹¹⁸ Brown, P. R. L., Augustine of Hippo (1967), 356Google Scholar, where the phrase is applied to early fifth-century Palestine.

¹¹⁹ Euch. ll. 520–1. For some of these holy men, Gennadius, De vir.ill. 68, 80, 81, 100, 101.

¹²⁰ Marseille appears as Massilia Graecorum on the Peutinger Table (Miller, K., Die Peutingersche Tafel (1962)Google Scholar, Segm. 11, 1) and in the Notitia Dignitatum (Not.Dig.Occ. XLII, 16). The city's Greek heritage is reflected under the Empire in the names of its buildings, officials, and individuals (e.g. CIL XII. 410; v. 7914). Greek was still used in the early third century to record the career of a high-ranking imperial functionary from Marseille (a rare bird) (CIG III. 6771). Like Naples, Marseille was never ‘typically Roman’: Arthur, op. cit. (n. 50), 247.

¹²¹ Marseille remained a vital interface between late sixth-century Gaul and the eastern Mediterranean world on numerous levels. I will not enter into the political manifestations of this here, but as far as mentalités are concerned, Bishop Serenus of Marseille's precocious reaction against icons illustrates how his city continued to hold intellectual tensions in common with the East, but alien to the western Church: see Gregory I, Reg. IX. 208; XI. 10; Markus, R., ‘The cult of icons in sixth-century Gaul’, JTS 29 (1978), 151–7CrossRefGoogle Scholar.

¹²² Notably in the clothing and the elaborate manner of the burial: see Boyer, R., Vie et mort a Marseille a la fin de I'antiquite (1987), 45–93Google Scholar. The body was laid in one of the finest sarcophagi produced by a local sculptural atelier, the late fourth- and fifth-century activity of which is another important index of local Christian wealth: Drocourt-Dubreuil, G., Saint-Victor de Marseille: art funeraire et priere des morts aux temps paleochretiens (IV-V siecles) (1989), esp. 80–7Google Scholar.

¹²³ For the efforts of the fifth-century bishops of Marseille to rival Aries (and Aix) in the evolving ecclesiastical hierarchy, their success in maintaining some sort of anomalous status within it well after Proculus’ death, and the city's importance as a nexus of power within the Church of southern Gaul at this time, see the recent detailed analysis by Mathisen, R. W., Ecclesiastical Factionalism and Religious Controversy in Fifth-Century Gaul (1989)Google Scholar, passim, esp. 22–5, 28–30, 51–60, 117–22, 187–8, 213–14, 219–20, 224–8. In this highly competitive context, the remarkable scale of the baptistery at Marseille makes perfect sense (cf. Section 1, above).

¹²⁴ Amantius of Clermont: Sidonius, Ep. VI. 8; VII. 2; VII. 7; VII. 10; IX. 4.

¹²⁵ e.g. Grenier, A., Manuel d'archéologie galloromaine ii (1934), 499–509Google Scholar. See also Russell, R. J., ‘Geomorphology of the Rhône delta’, Annals of the Association of American Geographers 32 (1942), 149–254Google Scholar.

¹²⁶ For the Fossae, Pliny, NH III. 4. 34; Mela II. 5. 78 and esp. Strabo IV. 1. 8, recording the profit to Marseille of tolls on canal shipping and the handover to Arlxses, but also the resumption of silting and the general difficulties of entry. The Fossae are not recorded thereafter. For theories as to their course, see e.g. Grenier, op. cit. (n. 125), and Constans, L.-A., Arles antique (1921), 195–205Google Scholar.

¹²⁷ Peutinger Table: Miller, op. cit. (n. 120), Fos–Segm. I, 5; Ostia–Segm. v, 1. Levi, A. and M., Itineraria Picta: contributo allo studio delta Tabula Peutingeriana (1967), 124–30Google Scholar; Bosio, L., La Tabula Peutingeriana: una descrizione pittorica del mondo antico (1983), 149–62Google Scholar.

¹²⁸ For the settlement and the geomorphological problem of its disappearance, see e.g. R. Beaucaire, ‘Les fouilles sousmarines de Fos’, Prov.Hist. 14 (1964), 16–25; Monguilan, L., ‘Un port romain dans le golfe de Fos’, Caesarodunum 12 (1977), ii, 359–70Google Scholar; Liou, B., ‘Les découvertes archéologiques du Golfe de Fos et le tracé du littoral antique’, in Déplacements des lignes de rivage en Méditerranee, C.N.R.S. (1987), 59–65Google Scholar; Oldham, R. D., ‘Historic changes of level in the delta of the Rhône’, Quarterly Journ. Geological Soc. 86 (1930), 64–92CrossRefGoogle Scholar. The underwater remains have sadly been much disturbed in the modern redevelopment of the Fos gulf by the petrochemical industry. See also Gallia Informations (1990), i–ii, 292–3, for recent survey data showing considerable late antique and early medieval rural settlement in the Fos area, another pointer to regional vitality in this period.

¹²⁹ This is not to deny that Arles functioned as a seaport in the Roman period (although significant archaeological traces of its port are still lacking), simply that this operation can never have been straightforward.

¹³⁰ e.g. Portus: Meiggs, R., Roman Ostia ² (1973), 170–1Google Scholar; Luni: Smith, C. Delano, Gadd, D., Mills, N., and Ward-Perkins, B., ‘Luni and the Ager Lunensis’, PBSR 54 (1986)Google Scholar, esp. 123–41; Ephesus: Foss, C., Ephesus after Antiquity: a Late Antique, Byzantine and Turkish City (1979), 185–7Google Scholar.

¹³¹ Bede, HE IV. 1, refers to Theodore and Hadrian, en route to England in 668, taking ship to Marseille and then travelling by land to Arles.

¹³² Procopius, Bella V. xiii. 14–29.

¹³³ Despite recurrent internecine conflicts, it seems that in normal circumstances passage between Frankish kingdoms was uninhibited on due payment of tolls: Gregory of Tours, Hist. IX. 32, for an exception which proves the rule.

¹³⁴ See e.g. the machinations of Bishop Theodore of Marseille, arraigned by Guntram not just for his part in the Gundovald affair, but as one of those chiefly responsible for arranging the assassination of his brother Chilperic. Gregory of Tours, Hist. VIII. 5. The works of Venantius Fortunatus also highlight on a cultural level the integration of the southern élite into the social milieu of the royal courts in the North: George, J. W., Venantius Fortunatus, a Poet in Merovingian Gaul (1992), 141–50Google Scholar.

¹³⁵ Werner, J., ‘Fernhandel und Naturalwirtschaft im östlichen Merowingerreich nach archäologischen und numismatischen Zeugnissen’, Bericht der römischgermanisch Kommission 42 (1961), 307–46Google Scholar.

¹³⁶ See Buchner, op. cit. (n. 68), 37–9, n. 24.

¹³⁷ For the condition of Italy in this period, see e.g. T. S. Brown, Gentlemen and Officers: Imperial Administration and Aristocratic Power in Byzantine Italy, A.D. 554–800 (1984); Christie, N. J., ‘The archaeology of Byzantine Italy: a synthesis of recent research’, Journ. Mediterranean Arch. 2/2 (1989), 249n93,Google Scholar esp. 259–63, which interestingly highlights the relative abundance of imports in western Liguria, the adjacent region to Provence.

¹³⁸ cf. introduction and n. 6. His interpretation of the demise of Marseille is another matter.

¹³⁹ Agathias, Hist. I. 2. 1–3.