We shall begin with a miracle that once took place in the city of Arles in the late 420s.Footnote ¹ It was early in the morning of the feast day of the martyr Genesius. A huge crowd of devotees, made up not just of city dwellers but also many others, who had flocked into the city from its surroundings, was travelling across the river Rhône, on the famous bridge of ships, towards the shrine on the western bank that commemorated the place of the saint’s martyrdom. It was then that disaster struck: the sheer weight of the pressing crowds caused the bridge to break apart – a frequent occurrence, our anonymous narrator claims, rather nonchalantly. The same narrator picks out certain characters and vignettes from the procession and paints a highly decorous picture: the well-to-do had come not only decked out in their finery but also accompanied by key accoutrements and symbols of their status. They were accompanied by horses in livery; virgins had come in fine jewellery with smart hairstyles; servants not only carried heavy drinking cups for their elegant mistresses but also carried these same mistresses in litters so they did not have to walk themselves. Happily, disaster was averted when thanks to the aid of Saint Genesius (very much achieved through the prayers of the presiding bishops, Honoratus),Footnote ² all the devotees were brought to safety unharmed (if wet) and were able to continue their procession to the shrine. By contrast, we will turn next to an episode in the Life of Honoratus’ successor, Hilary (bishop from 429/30 to 449), which replaces decorous urban harmony with urban conflict. This is a tale in which an angry crowd confronts the bishop and receives divine punishment in recompense:

The first story presents a very much idealized vision of the city of Arles, a neatly sanctified version of what Ausonius addressed as ‘duplex [double] Arelas’.Footnote ⁴ It depicts the annual procession across the city, and across the river, from the site of Genesius’ tomb in the large cemetery at Alyscamps to the actual site of the martyrdom, in the more industrial and commercial area of Trinquetaille (see Map 3). And yet this story about late antique Arles is incongruous in several ways. Our literary record does not generally have very much to say about the cult of martyrs in Arles – indeed, the great miracle attributed here to Honoratus does not even appear in the sermon celebrating his life, which was composed by Hilary, his successor.Footnote ⁵ Meanwhile, the archaeological evidence has not yet revealed the kind of lavish domus that would have housed the litter-travelling matronae. The archaeological record indeed suggests that a number of the areas traversed by the procession would likely have been in a state of disrepair or even abandonment at this time. The second story likewise raises a number of questions – and we shall return to it later in the chapter – but it does certainly suggest a much more conflictual picture of life in Arles than that given by the first, and one where the people of the city were far from being united behind their bishop. The satisfaction the author of Hilary’s Vita takes in the fire that destroyed a large part of the city strikes a rather odd and unpleasant note. The archaeological record does in fact bear witness to a number of instances of destruction by fire at various times that cannot be firmly linked with any known episodes in the historical record. The city of Arles, along with the other cities of southern Gaul, was undergoing substantial change in this period, and we will come across a number of puzzles and contradictions in our attempts to understand our urban landscapes as locations for popular culture.

In what follows I shall focus in large part on Arles: a city of great importance in late antiquity, as we have already seen, the home of Caesarius as well as Honoratus and Hilary, and the city for which we have the best opportunity to combine textual and archaeological evidence to build up a more comprehensive picture. I shall also look at Marseille, for which recent archaeological work provides a good sense of the economic transformations of the period. The cities of Aix, Nîmes and Narbonne in turn also provide instructive examples of changes in topography and infrastructure. However, I shall also explore urban life as it appears through the prism of the literary and ecclesiastical texts. In this way in this chapter we can build as full a picture as possible of the lived culture and environment of urban dwellers as well as of the broader developments of this period.

There are already a good number of excellent syntheses dealing with the ‘transformation’ of the cities of southern Gaul in late antiquity, many of which focus largely on topographical changes, and specifically the ‘christianization’ of the urban landscape.Footnote ⁶ I shall seek to go beyond this well-trodden approach in order to make an inquiry into the cultural and social transformations of the period that considers more than changes in architecture and even infrastructure, and looks beyond the perspective of classical urbanism. In particular, I will seek to examine the cities of south-eastern Gaul as far as possible from the perspective of the non-elite urban dweller, challenging as this is. Ultimately, what I aiming to do here, in line with my approach which understands popular culture as fundamentally embedded, is to establish the spaces in which popular culture was constructed and examine the social and economic relations in which it took its form, as well as the dominant ideological structure. As this was a period of fundamental change, I shall look to see how far these various transformations impacted the broader framework itself.

In what follows I shall first introduce the general built and urban environment, before turning to look in more detail at the inhabitants of the city in terms of occupation, status and identity. Next I shall turn to the impact of the church upon the city, social as much as topographical; urban social relations will come to the fore, forming the basis for the final discussion of performance, leisure and the transformation of urban culture more generally. In the previous chapter I highlighted some key features of popular culture, as revealed in the evidence from Pompeii and Aphrodisias, and we shall see elements of both continuity and change in what follows.

The Urban Environment

At the start of our period, at the turn of the fifth century, Arles was enjoying its greatest period of glory. The city’s importance as an imperial residence had been established under Constantine and his family, while its prestige grew further still at the turn of the century with the transfer of the Prefecture of the Gauls from Trier, as we saw in the previous chapter. The city’s ecclesiastical status likewise rose to a new pre-eminence at this time with the granting of metropolitan status in place of Vienne.Footnote ⁷ Nonetheless, this stability would not last: the city was besieged in 425, 430 and 458 by the Visigoths, who captured it again in 473, and consolidated their control in 475/6. These recurrent sieges nonetheless demonstrate the continuing importance of the city. The first half of the sixth century also saw turbulent and traumatic periods: Arles was besieged by the Franks in 507–8, eventually coming under Ostrogothic control, under which a period of calm ensured, but this situation was not to last: the city was taken by the Franks in 536/7. While these events cannot be traced as such in the archaeological record, it is unsurprising that the urban fabric underwent some substantial changes during this period. Thanks to detailed and nuanced archaeological attention – if not extensive excavation – the broad lines of its urbanistic trajectory in our period have become progressively clearer in recent years.Footnote ⁸

What was the built environment of Arles like in late antiquity (see Maps 3 and 4)? In common with other cities in the region, its urban fabric had suffered during the third century, with particular damage to suburban areas.Footnote ⁹ The fourth century, however, had been a time of investment in public building under the patronage of both visiting and resident emperors. The northern ‘Constantinian’ baths and a new gallery at the forum, but also a civic basilica and quite possibly too a triumphal arch and obelisk, all date from this time.Footnote ¹⁰ Presumably a number of new administrative, residential and ceremonial buildings were also constructed during the same period, such as the so-called Palais de la Trouille.Footnote ¹¹ Ecclesiastical building probably also began in the course of the fourth century, although there is no surviving archaeological evidence of fourth-century churches. The marvellous collection of sarcophagi on display in the archaeological museum in Arles testifies to the prosperity of the city’s elites at this time.Footnote ¹²

In the course of the fifth century, however, we can see a clear change in urbanistic focus. It was probably at this time (though conclusive archaeological evidence is still lacking) that a new reduced wall circuit was built, which was maintained until at least the end of the sixth century.Footnote ¹³ Meanwhile, archaeological evidence clearly indicates the spoliation of traditional prestige public monuments: the forum, theatre and amphitheatre. (As we shall see later, the circus remained in use.) Puzzles regarding this development or indeed degradation of public monumental areas at this time remain, notably in the case of the forum. Archaeological evidence shows that the forum paving was already beginning to be dismantled in the first half of the fifth century,Footnote ¹⁴ and that the space was being taken over in part by new types of construction, including shops. The walls of these new constructions re-used spolia from the portico of the forum, while at the same time galleries of the forum cryptoporticus were being re-used as dumps and depots, but also likely as cellars and even housing.Footnote ¹⁵ Even so, we cannot entirely rule out the possibility that the forum was still in use in some form as a public civic space. In an intriguing letter, Sidonius Apollinaris describes a visit to the forum of Arles in spring 461, where nervous individuals hid behind the statues and columns in order to avoid him.Footnote ¹⁶ This passage has long puzzled scholars and presents several different options, none conclusive. One is that this passage refers to an alternative forum, most likely constructed as part of the Constantinian building programme.Footnote ¹⁷ If not, Sidonius’ account could suggest that even if the forum and its associated buildings look ‘different’ to us, they could still be used as part of everyday civic life by the inhabitants of late antique Arles, who did not necessarily have an awareness of themselves as living in a ‘late antique’ city. There is of course another option altogether: that the whole episode is largely fictional, one of a number of classically inspired vignettes that make Sidonius a lively read but a highly problematic witness to the social and cultural history of late antique Gaul. The passage does nonetheless clearly indicate the strength of the traditional ideological associations of the ‘forum’ for aristocrats like Sidonius. This is indicative of a period where ideas and ideals of the city were in transition, alongside new patterns of use and spoliation; ecclesiastical building, meanwhile, continued apace, as we shall see.

Turning from public to private, excavation has so far failed to uncover the quantity of luxurious city housing one might expect, given the other evidence for the prosperity of the fourth-century city in particular. Over the last thirty years, however, excavations have uncovered traces of late antique housing, including the domus type but also what is often (often not very helpfully) described as ‘parasitical’ construction.Footnote ¹⁸ This latter type includes the substantial dwellings built into the structures of the circus, beginning in the first decade of the fifth century, and abandoned along with the circus itself in the middle of the sixth.Footnote ¹⁹ Marc Heijmans has suggested that some of this accommodation was in fact arranged by the authorities in order to ease pressure on housing; given the systematic nature of the construction, this seems the most likely scenario.Footnote ²⁰ Describing such constructions as ‘parasitical’, or this occupation as ‘squatting’, would therefore clearly be wrong; the persistence of such terms in scholarship is evidence of the continued top-down and indeed elitist assumptions of traditional approaches to late antique urbanism. Overall, our knowledge of the nature of housing in late antique Arles remains, sadly, hazy at best. In future, scholars could perhaps deploy existing archaeological evidence to consider living conditions more systematically, given the will. For instance, examination of the faunal remains found in domestic sites, including the build-up of seashells, illustrates how stench, vermin and small carnivores would have abounded in areas which were being cleaned less and less often, as part of a degradation of the urban environment.Footnote ²¹ We can be sure that urban living conditions in sixth-century Arles would have been different from those in the city’s Constantinian heyday, for all members of society, but we need more clarity about the specific nature of the changes rather than relying on vague notions of either ‘decline’ or, more positively, ‘transformation’.

Turning to the economic trajectory of Arles in late antiquity, its importance during the fourth and (at least the first half of the) fifth centuries is suggested by our textual sources, all of which focus on the importance of the city as a river port (with access to the sea via Fos) (see Map 2). The anonymous Expositio totius mundi et gentium (c. 360) describes ‘Arles, which takes in goods from the whole world and supplies them to [Trier]’.Footnote ²² Ausonius’ famous and flattering description of ‘duplex Arelas’ in his Ordo urbium nobilium, written in the 380s, describes the well-known bridge of boats across the Rhône, but the only other feature of the city he describes, in a similar vein to the Expositio, is its role in importing merchandise from across the Roman world and distributing it across Gaul.Footnote ²³ The continuing prominence of the port and the scale of its commercial activity into the fifth century is highlighted by Honorius’ constitution of 418Footnote ²⁴ but it is difficult to match the archaeological evidence with these accounts.Footnote ²⁵ As we shall see, we can only speculate, for instance, regarding the role of small-scale productive activity in the urban landscape of the kind that has been identified in the towns of North Africa at this time,Footnote ²⁶ and which clearly provided an important context for popular culture.

An interesting contrast is provided by the case of Marseille, a city that actually experienced economic growth in our period, enjoying the longest ‘late antiquity’ of any city in Gaul.Footnote ²⁷ Indeed, Marseille is a city whose history is much better understood for late antiquity and the early middle ages than during the earlier Roman period.Footnote ²⁸ Intramural occupation continued in the fifth and sixth centuries, while there was also development outside the walls. The substantial modern excavations of the city, notably in the port area, have shown the importance of Marseille as both a centre for manufacturing (notably of ceramics, discussed further later) and as an emporium for exchange. Marseille benefited from its location and enjoyed a privileged position in a continuing Mediterranean-wide network exchange until at least the end of the sixth century.Footnote ²⁹ Like Arles, Marseille for a long time got off relatively lightly among the political and military upheavals of the period, and became something of a hub for refugee intellectuals – though the construction of a new ‘avant-mur’ around 470 does show (unsurprising) concerns for security at this time.Footnote ³⁰ The prosperity of the city in the fifth century is further demonstrated through a substantial programme of church building, as will be discussed later.

What about elsewhere in the region (see Map 2)? The city of Narbonne had already undergone a substantial contraction of its intramural territory during the third century.Footnote ³¹ Nîmes saw ‘profound’ transformations in its urban topography during the fifth century; large areas were abandoned and public buildings destroyed. Much of its late antique urban fabric still remains obscure, however, including the early cathedral and suburban churches.Footnote ³² In the case of Aix, it is clear that some areas had suffered from decline and abandonment in the third century, while recent work is beginning to give a clearer picture of the late antique period. The theatre, as with entertainment buildings elsewhere, was reoccupied and despoiled.Footnote ³³ Around 500 a big new episcopal complex was built in the main monumental centre on the site of a large public building.Footnote ³⁴ As for smaller urban agglomerations in Provence (see Map 5), several of them are now best known for their surviving ecclesiastical buildings, such as Riez and Fréjus,Footnote ³⁵ but elsewhere we know little in terms of urbanism.

The overall picture, therefore, while not straightforwardly uniform, is one of widespread and substantial change. As part of this change we can see clear shifts in the relative status and prosperity of cities. We have already seen both the rising fame of Arles and the ‘late’ economic boom of Marseille. However, we can also observe smaller urban areas rising to prosperity and visibility in late antiquity, such as Agde in eastern Languedoc, first mentioned as a bishopric only at the start of the sixth century. A variety of factors are at work here: while political changes and currents are crucial, economic developments (albeit often closely interrelated with political currents) were also of great importance.Footnote ³⁶ These changes would of course have impacted upon the living conditions of the inhabitants of these urban centres.

All too often discussions of late antique urbanism, especially when relating to topographical change, and questions of transformation/decline can seem rather too abstract when it comes to the question of the lives and identities of the actual inhabitants of these cities, particularly in the case of the non-elite.Footnote ³⁷ I shall now therefore turn to look at the city dwellers themselves.

Who Lived in the City? Occupations and Identities

Even the elite can be hard to find when we look at the late antique city – including the relatively well-known case of Arles. We met the urban elite at the start of the chapter, laden with jewellery, accompanied by their families, slaves, retainers and horses, engaging in a conspicuous display of wealth and status.Footnote ³⁸ The city was certainly a magnet for the powerful, including both civilian and military elites, due to its prominent position across various regimes in our period, as well as the local landholding elites. We get the best glimpses of the elites of the city of Arles in our literary sources: the courtiers and bigwigs attached to the court of Majorian during his stay there,Footnote ³⁹ Hilary’s well-resourced supporters and his ‘proud’ opponents,Footnote ⁴⁰ the well-connected aristocrats Firminius and Gregoria who Caesarius called upon when he arrived in the city, and the famous teacher Pomerius.Footnote ⁴¹ The Arlesian elite of the fourth century are well represented in the material record in the form of the wonderful collection of sarcophagi in the Musée de l’Arles Antique. However, as we saw, tracking their traces in the urban fabric is trickier, especially when it comes to housing. The decline of the epigraphic habit in late antiquity, meanwhile, also takes its toll on the visibility of the well-to-do and important – let alone their social inferiors. Meanwhile, while we know that our cities retained their town councils, filled by curiales, we know very little more than this.Footnote ⁴² Nonetheless, I shall try to sketch out a picture of our urban inhabitants, urban identities and sub-cultures, giving as rich and non-homogeneous a portrait as possible, in order to build up a context for the popular culture under examination. I shall look at our evidence for the different industries and professions, turning to what little we know of the organization of labour before considering the different statuses and sub-cultures we find in our cities. The commercial cities of southern Gaul, notably Arles, Marseille and Narbonne, certainly housed diverse populations, both as visitors and permanent residents, including all kinds of traders and travellers such as river boatmen and sailors, as well as resident slaves and freed people. Finally, as we shall see, Jews made up an acknowledged part of the civic community throughout antiquity.

But firstly, what about the ‘poor’, who have been the focus of much scholarship in recent years?Footnote ⁴³ John Chrysostom once told his congregation that they should imagine their city as made up of one tenth rich, one tenth ‘the poor who have nothing at all’, with the rest as middling.Footnote ⁴⁴ Peter Brown pointed out that such a figure does not in fact sit too oddly against what we know (or estimate) of late medieval or early modern cities.Footnote ⁴⁵ Caesarius defined a pauper as one in need of food and clothing, and paired him with a mendicus. However, his teacher, Pomerius, had stressed that the church should support only the weak and infirm, leaving the able-bodied poor to earn their living through their labour.Footnote ⁴⁶ The Council of Orléans in 511 likewise specified that the bishop should only help the poor or ill, those incapable of manual labour.Footnote ⁴⁷ Our texts from Arles do not explicitly mention a matricula, a register of the poor, supported by the church, familiar from other Gallic sources.Footnote ⁴⁸ However, the physical presence of the indigent and infirm outside the church, as well as the bishop’s residence, was indeed a regular feature in late antiquity.Footnote ⁴⁹ The poor are not nearly as prominent as recipients of episcopal charity in the Life of Caesarius as are captives, and yet we still learn of their regular appearance outside the bishop’s doors.Footnote ⁵⁰ They were, however, clearly greatly outnumbered by those able to make their own living.

How then did people make a living in the cities of late antique southern Gaul and how does this work provide a context for our understanding of popular culture? Our cities, sited on major trade axes, included some of the most vibrant commercial and indeed artisanal urban centres of the ancient world. Inscriptions from earlier centuries testify to an impressive diversity of occupations: a study of funerary inscriptions from Narbonne found a total of fifty-one professions across eighty-one individuals, representing a high level of division of labour. These jobs ranged from metalwork to hairdressing, including a number of textile-related jobs, as well as culinary, medical and other professions.Footnote ⁵¹ The epigraphy of second- and third-century ce Arles includes many examples relating to marine and river trade stemming from the professional organizations of ship-builders, carpenters, boatmen and so on.Footnote ⁵² Historians of late antiquity have far fewer inscriptions to work with, although legal evidence points to the continuing importance of naviculari, nautae and their organizations.Footnote ⁵³ Moreover, a variety of literary and archaeological evidence can be used to learn more about the world of urban work in late antiquity.

Information regarding the professions and industries of late antique Arles is fleeting and incidental in the sermons of Caesarius, who we can view as sharing with his fellow bishops a general disinterest in the details of the working lives of his urban congregation.Footnote ⁵⁴ In a sermon on tithing Caesarius summarizes the trades of his congregation as comprising artisanship/manufacture, commerce and military service.Footnote ⁵⁵ In another sermon he specifies the work of goldsmiths (aurifices) and carpenters (fabri) as forms of artisanship.Footnote ⁵⁶ Textiles are mentioned on several occasions in connection with women: firstly in a repeated attack on the ‘superstition’ that supposedly prevented women from performing their weaving work on Thursdays, in honour of Jupiter,Footnote ⁵⁷ although the most detailed discussion of female textual production comes in the Regula ad virgines.Footnote ⁵⁸ There are a number of references to commerce: negotiatores, merchants or businessmen, are in fact the most frequently mentioned profession in the city, alongside several mentions of trade in general.Footnote ⁵⁹ It is surely not a coincidence that this is likely to be an occupation held by the richer, more socially elevated members of the urban congregation.

We need to look at the archaeological evidence in order to get a more detailed sense of the world of work of our late antique urban dwellers. The city of Arles itself is not very helpful. An ‘artisanal quarter’ developed across the Rhône in Trinquetaille around the turn of the first century ce, with evidence showing a range of different types of activity, including metallurgy, marquetry and pottery; however, there is no evidence of this production continuing into late antiquity.Footnote ⁶⁰ The evidence for urban and suburban artisanal activity elsewhere in the region is also rather patchy. For instance, metallurgical work was carried out in the area of the disused baths in Aix,Footnote ⁶¹ while at Nîmes there seems to have been a concentration of potters’ workshops in the south-west of the city, within the walls, at least at one time.Footnote ⁶² However, excavations at the ‘ZAC’ des Halles at Nîmes in 1999 showed that this urban area was given over to agricultural activity during the fifth century and would not be ‘re-urbanized’ until the later middle ages.Footnote ⁶³

Late antique Marseille offers a different and much fuller picture. Suburban areas such as that excavated around the Bourse provide evidence of ‘teeming urbanism’, comprising a whole range of activities, including work in ivory, bone, leather and wood.Footnote ⁶⁴ The most significant ‘industry’ by far in Marseille, however, was clearly that of ceramics; indeed, since its foundation Marseille had been an important centre of ceramic production. In late antiquity it seems that amphorae, an important category of earlier production, were no longer in production but instead large quantities of both the so-called DSP (‘dérivées des sigillées paléochrétiennes’), a red slip fine ware which imitated African red slip, and common grey ware were produced.Footnote ⁶⁵ Marseille was in fact one of the major centres of DSP production (along with Bordeaux and Narbonne) and as a centre of production is notable for both the quality and wide diffusion of its products, as well as for the large variety of forms.Footnote ⁶⁶ The production of DSP in Marseille reached its height in the middle or late fifth century (most likely due to a decline in overseas imports at this time).Footnote ⁶⁷ In the course of the sixth century there was change, with a decline in the number of forms produced, and a reduction in the amount of decoration, but production nevertheless continued, without much further alteration, though on a smaller scale, until the end of the seventh century.Footnote ⁶⁸ There is little direct evidence of workshops or production (no urban kilns have been identified to date), but it must be assumed that ceramic production took place in locations within the city, outside the walls, but within the territory of Marseille.Footnote ⁶⁹ But how was urban artisanship organized and how might it have functioned as a context for popular culture? We know nothing about the organization of the ceramics industry in Marseille, although we can look at better-understood cases elsewhere for comparison. For instance, the suburban pottery quarter of Sagalassos in Turkey was still in use up to the second half of the sixth century ce. Here a number of small individual workshops have been identified, containing around ninety kilns in total, although their shared use of the same raw material suggests some kind of integration of production, albeit loose.Footnote ⁷⁰

What we know for sure is that the urban world of work was highly stratified. Nicolas Tran has aptly commented on Arles in an earlier period: ‘[t]he harbour society of Arles was no homogeneous ensemble, and the occupations and associations that emerged from it were unequal in dignity and prestige’.Footnote ⁷¹ Likewise, while the wide range of occupational inscriptions from Narbonne testifies to the dynamic state of the local economy in the early empire, it also demonstrates economic and social competition among Narbonne artisans and traders.Footnote ⁷² As discussed in Chapter 1, recent scholarship has illuminated ‘plebeian’, ‘artisan’ or indeed ‘middle-class’ culture in the towns and cities of the Roman empire, including Miko Flohr’s case study of the fullo.Footnote ⁷³ Epigraphic evidence has proved of particular use for probing this social history, while analysis of visual images, especially in funerary contexts, has demonstrated the importance of occupational identity for the non-elite.Footnote ⁷⁴ In Pompeii we saw how inscriptions and graffiti can illustrate the political capacity of occupational groups.Footnote ⁷⁵ Meanwhile, for southern Gaul, including Arles, the organization and roles of the various type of collegia and professional organization have been well studied.Footnote ⁷⁶ However, this material does inevitably lead us to the upper echelons of the urban artisanal and mercantile community. The epigraphic evidence will never take us to the poorer, more socially marginal workers at Arles or indeed elsewhere; it is, nonetheless, suggestive of sub-elite values as well as structures of economic and social organization. One striking example is the metrical epitaph of the Arlesian cabinet maker Quintus Candidus Benignus, dating to the second or third century ce: it describes his trade in the unexpected terms of doctrina and ars.Footnote ⁷⁷ Andrew Wallace-Hadrill sees this as an example of the aping of elite culture by the sub-elite, keen to differentiate themselves from the workers: an instance of ‘reproduction’ rather than inversion.Footnote ⁷⁸ However, we can surely interpret this differently: talking about carpentry in terms of doctrina and ars, even if not subversive as such, is surely clear evidence of non-elite self-confidence and pride in non-elite work, and an assertion of its value in a competitive marketplace.Footnote ⁷⁹

Alas, epigraphic evidence of this kind simply does not exist for the late antique period because of well-known changes in the epigraphic habit. Therefore, unfortunately, we hear no more about even the most important associations, such as that of the naviculari of Arles, after the start of the third century.Footnote ⁸⁰ Modern scholarship rejects earlier depictions of late Roman collegia as the coercive instruments of a statist planned economyFootnote ⁸¹ but the fact remains that we simply know very little about the organization of professional associations and their broader social, cultural and economic role in southern Gaul. Nonetheless, examples from the Greek east in particular would suggest that such associations continued to play a key role in the cities.Footnote ⁸² The importance of Arles as a river port in at least the first part of our period surely implies the continuance of a substantial support infrastructureFootnote ⁸³ – but we remain pretty much in the dark when it comes to understanding the organization of labour in our cities in the period. In an important article on late antique Rome, Nicholas Purcell stresses the continuing significance of ‘a structure of dependence which united the vast majority of the inhabitants’,Footnote ⁸⁴ but others are beginning to present a world with more space allowed for horizontal organization.Footnote ⁸⁵ We are of course even less able to reconstruct these kinds of structures of dependency in the cases of Arles or Marseille than for Rome, but in order to provide a context for understanding popular culture, it is important to try to understand the social relations in which this culture developed.

Our lack of knowledge when it comes to the operation of wage labour extends likewise to slave labour. On one occasion Caesarius makes an unusually direct reference to wages paid to slaves for their work (pro opere suo mercedem suam);Footnote ⁸⁶ this specificity is unusual, and we wish we knew more. What is clear is that slaves were part of the urban community in late antiquity, an expected part of the general picture of daily life. For instance, we have already seen household slaves appear in procession in Arles on the feast day of St Genesius as part of the prestige apparatus of elite households.Footnote ⁸⁷ Domestic slaves are explicitly referred to (and bequeathed) in the will of Caesarius: two slaves are mentioned by name, one of whom, Agritia, is sent to serve at his sister Caesaria’s monastery, while the rest of the domestic slaves are passed on to Caesarius’ successor and co-heir, Auxanius.Footnote ⁸⁸ Slaves appear in Caesarius’ sermons as examples of negative behaviour: guilty of theft, neglect of property and drunkenness.Footnote ⁸⁹ Broader anxieties about household slaves are also expressed in the anonymous fifth-century Gallo-Roman play, Querolus, where the clever slave Pantomalus evokes a topsy-turvy nocturnal ‘counter-world’ of the slaves, based on sensual pleasure, including the use of the household baths at night.Footnote ⁹⁰ While fascinating as an elite view of the potentially subversive thoughts and activities of slaves, such ‘evidence’ inevitably does not get us any closer to accessing the subjectivity of enslaved people.

Moving on, while servile origin is often associated with the use of the Greek language in the cities of the west, this is just one aspect of its long history in southern Gaul. Marseille was originally founded by Greek settlers and enjoyed continued Greek influence, with the later arrival of Greek-speaking traders, as well as slaves and ex-slaves.Footnote ⁹¹ The epigraphy of Marseille indeed shows a number of names of Greek origin, and we should not assume them to be necessarily derived from an ex-servile population. Indeed, a number of professions seem to be associated with Greek names, even when there is no sign of servile origin.Footnote ⁹² While a sizable proportion of Marseille’s inscriptions from the early imperial period are in Greek, there are so few late antique inscriptions in toto that it is not feasible to draw any statistically valid comparisons.Footnote ⁹³ We know that Greek continued to be used in church in Arles thanks to the detail in the Vita of Caesarius that the bishop taught his congregation to sing hymns in both Greek and Latin.Footnote ⁹⁴

That our cities possessed a substantial element of ethnic diversity is evident, and this bears discussion, given our interest in subaltern cultures. Late antiquity saw a continuation of the elite rhetorical tradition that inveighed against easterners, most visibly in the frequent mention of ‘eastern’ merchants, often designated as ‘Syrian’.Footnote ⁹⁵ Salvian provides the most striking example in a passage that is clearly indebted to the satirical tradition. After considering the poor reputation of a number of barbarian groups, he comes to Syrian merchants:

Even late antique legal texts employed highly coloured and indeed xenophobic rhetorical language referring to the ‘Oriens’;Footnote ⁹⁷ an orientalizing tone is still to be found, meanwhile, in some twentieth-century scholarship.Footnote ⁹⁸ More neutral in tone, interestingly, is a canon from the Council of Narbonne in 589 which refers to a population ‘both slave and free’ made up of ‘Goths, Romans, Syrians, Greeks and Jews’ (all of whom are to refrain from work on Sundays where possible).Footnote ⁹⁹ While this is clearly intended to be deliberately comprehensive, the specifications are nonetheless of interest. Epigraphic evidence for ‘Syrians’, meanwhile, is generally very limited or indeed inconclusive, although scholars have been oddly keen to identify ‘Syrians’ from among the few late antique Greek inscriptions from late antiquity.Footnote ¹⁰⁰ Attempts to identify Syrian influence, and the presence of Syrian craftsmen in the art of southern Gaul, are not conclusive.Footnote ¹⁰¹ Overall, it seems we should be wary of ‘Syrian’ clichés as outmoded at best.

What about ‘Goths’, as we saw specified as a category in the council held in Visigothic Narbonne in 589? Or indeed Franks? There has been a huge amount of debate and discussion on the issues of the nature of barbarian settlement, ethnic identity and ethnogenesis, but most of it is simply not really relevant when it comes to thinking about our urban populations.Footnote ¹⁰² As previously discussed, it seems most likely that the new regimes relied on the local Gallo-Roman elites to run the cities. Clearly, we must imagine the presence of billeted troops, but the ‘Roman’ armies had been ‘barbarized’ for many generations. An anecdote in Caesarius’ Life refers to comites civitatis et reliqui militantes as throwing their weight around, demanding hunting rights on the estate of the suburban monastery, but this clearly regards the Gothic elite.Footnote ¹⁰³ In theory the Visigoths were Arian ‘heretics’, but we hear very little about their presence as such – and Klingshirn is surely right to suggest that Arianism ‘was confined to Gothic soldiers and officials, whose religious influence on the local population would have been minimal’.Footnote ¹⁰⁴ We do get mention of ‘Arians’, however, at times of crisis, such as when they were rhetorically paired with Jews as supposed enemies of Caesarius.Footnote ¹⁰⁵

More can be said about the Jews of southern France. It is clear that they made up a recognizable part of the urban community in our period; indeed, it is in late antiquity that a Jewish presence becomes evident again in southern Gaul after a long lacuna in the evidence.Footnote ¹⁰⁶ However, even more so than in the case of ‘Syrians’, the literary ‘sources’ prove to be ideologically and rhetorically loaded in the extreme, and must be read with great care. Jews appear in late antique ecclesiastical texts in order to serve varying ideological purposes but often, as Avril Keely has put it, as ‘agents of differentiation’.Footnote ¹⁰⁷ At the same time, these texts clearly do reflect, in the words of Paula Fredriksen and Oded Irshai, ‘a social world wherein Jews – distinctive, different, singled out – still remained integrated within the lingering urban framework’.Footnote ¹⁰⁸

Our evidence from southern Gaul does not feature the grim stories of forced conversions found elsewhere, such as in Minorca (in 418), or, closer to home, in Clermont, where the synagogue was burned in 576.Footnote ¹⁰⁹ Indeed, Gregory writes that although more than 500 Jews were baptised in Clermont, the rest of the community left for Marseille.Footnote ¹¹⁰ Nonetheless, we can clearly see the presence of contradictions and tensions in the material from Provence. At times it suited the rhetorical and ideological purposes of Christian writers to include Jews within the bounds of the imagined urban community: it was more or less a hagiographical cliché that Jews were present in the funerary processions of bishops, as indeed claimed by the Vitae of both Hilary and Caesarius.Footnote ¹¹¹ (More interesting is the specific claim that the Jews at Hilary’s funeral sang in Hebrew, to which we shall return.)Footnote ¹¹² At other times our texts seek very much to ‘other’ – and worse – Jews, and the Life of Caesarius is a useful case in point, especially the depiction of the actions of the Jews of Arles during the Burgundian siege in 507/8.Footnote ¹¹³ According to the Vita, a young clerical relative of Caesarius handed himself over to the enemy troops. Once this became known, Caesarius himself came under attack from a ‘mob’, including ‘a crowd of Jews’ (popularium seditione, certe et Iudaorum turba), accusing the bishop himself of sending his relation to betray Arles to the enemy.Footnote ¹¹⁴ The authors stress that ‘especially the Jews’ but also ‘heretics’ (a common coupling) made the accusation.Footnote ¹¹⁵ As a result of these charges (unfounded, of course, according to the Vita), the bishop was placed under armed guard.Footnote ¹¹⁶ The Life further alleges treachery by Jews themselves during the siege: a member of the Jewish troops within the city (apparently a defined group with their own section of wall to guard)Footnote ¹¹⁷ supposedly offered (by means of a letter tied to a stone) to help Jewish troops outside the walls enter the city, on the understanding that no fellow Jewish resident should suffer. This subterfuge was allegedly swiftly uncovered and punishedFootnote ¹¹⁸ but provokes comments from the authors of the Vita about the destruction of the ‘hateful’ cruelty of the Jews.Footnote ¹¹⁹

These accounts raise a number of questions but demonstrate clear social tensions that existed across the social spectrum, both clerical and lay, in Arles.Footnote ¹²⁰ The reported ‘perfidy’ of the Jews can be understood as part of broader social conflict and competition which would have existed at various levels in the urban society of Arles, and would of course be exacerbated at times of crisis. Nonetheless, at other times in Arles, just as elsewhere in the late antique world, we can see the efforts of churchmen to set boundaries to their Christian communities, at times deliberately excluding both Jews and ‘Jewish’ practices.Footnote ¹²¹ Gallic church councils attempted to stop first the clergy, and then the entirety of the Christian community, from attending Judaeorum convivia.Footnote ¹²² Should we here imagine particular sorts of meals held by/in the Jewish community, for instance in accordance with specific festivals or rituals? Or was this just another attempt to stop Christians from joining in the general convivial culture of the broader community? (In later chapters we shall see Caesarius’ attempts to restrict his congregation’s participation in wider social life.) Again, we cannot know. Overall, our evidence shows the existence of some tensions, of scapegoating and of anti-Jewish clerical discourse. On the other hand, it also shows the evident involvement of Jews in the social world of the urban community.

What our Christian and indeed legal sources cannot, of course, provide any insight into is how the Jews themselves envisioned their place in the culture and community of their cities. How culturally distinctive were the Jews of southern Gaul in the first place? In terms of material culture, there are only a very few examples of distinctively Jewish artefacts in use, and even here we would want to express caution about too simple an identification between ‘Jewish’ material culture and Jewish users.Footnote ¹²³ Linguistic evidence might seem to lead us to firmer ground. We noted the rare claim that the Jews at Hilary’s funeral sang in Hebrew. In general there is very little evidence for the use of Hebrew at this time, though Gregory of Tours claims that the Jews in Orléans praised King Guntram in their own language (and Syrians likewise!).Footnote ¹²⁴ The only two identifiably Jewish inscriptions we have from southern France in our period are in Latin and share the same linguistic formulae as their Christian counterparts.Footnote ¹²⁵ Indeed, David Noy has argued that the increasing use of Hebrew later, in the early middle ages, is a sign of reaction to the increasing hostility of the authorities towards the Jews and a growing desire to express a shared identity.Footnote ¹²⁶ In our own, late antique sources we can see attempts to define a Jewish community at least from the outside and we certainly have evidence for a diverse and at times divided wider urban community. It was in this community, in which different degrees of subalternity, as it were, existed, that popular culture was constructed and experienced. One notable aspect of this, unfortunately, was the focus on and scapegoating of minorities and other outgroups, a theme to which we will return in Chapter 6.

The Church in the City and its Impact on the Built and Social Landscape

While we have been considering the wider city in large part through ecclesiastical sources, it is now time to focus on the particular impact of the church itself as a key factor in the transformation of not just the built environment but also the social and political structures of the city. In what follows we shall look at the development of both as key themes for our understanding of late antique urbanism as a context for popular culture.

The emergence of a Christian topography in the cities of southern Gaul has been much discussed.Footnote ¹²⁷ This process was part and parcel of the development of a Christian infrastructure which accompanied the growth in power and prominence of the bishop in the city. These developments are well known, even if accounts tend to take a rather ‘top-down’ view of the process. My concerns are somewhat different and therefore I shall just briefly sketch the key features of the religious landscapes, including the ecclesiastical topography, of our cities. When it comes to non-Christian topographies, there is in fact rather little to be said, at least with our current state of knowledge. The classical urbanistic layout had included a number of temples and Narbonensis was particularly rich in Greco-Roman-style classical temples. That ‘pagan’ temples were no longer being maintained in the cities in our period is clear; their actual physical condition at this time, however, is not.Footnote ¹²⁸ As we shall see in Chapters 4 and 5, the continued existence and indeed use of various types of ‘pagan’ shrines and sites in the countryside was a source of annoyance for Caesarius, among others, but none of our ecclesiastical sources complain of a similar alternate sacred landscape in the city, which is telling in itself. There are no archaeological traces of cult buildings used by Jews in the cities of Gaul in antiquity.

A broadly understood model for the development of Christian urban topography in southern Gaul is as follows: a church was built in the fourth century in a somewhat marginal topographical position, often up against the city ramparts, with a move into the monumental centre coming sometime later (normally at some point in the fifth century) as Christianity became more established locally. Indeed, a number of cathedral churches have been identified, often as part of a full ‘episcopal group’, comprising, at minimum, a church, baptistery and episcopal residence. The evidence does not always present so clear-cut a picture, however. In Aix, for instance, a cathedral was built, probably at the very end of the fifth century, in the heart of the monumental centre. This cannot have been the first ‘episcopal’ church of Aix, as there had been a bishop of Aix for more than a century by this point and the ‘original’ church has not been definitively located.Footnote ¹²⁹ In Marseille the remains of a very large baptistery, built in the fifth century in the north-west of the ancient city, are the most visible remnants of the ‘episcopal group’, including an accompanying cathedral, itself lost due to construction works in the nineteenth century.Footnote ¹³⁰ As for Arles (see Maps 3 and 4), the chronology of church building remains unclear. Only one church is mentioned in the Vita of Honoratus (bishop from 426 to 429), written by his successor, Hilary, in c. 430.Footnote ¹³¹ The Vita of Hilary himself, however, mentions three distinct Christian cult buildings, including a cathedral church associated with St Stephen.Footnote ¹³² For a long time it was assumed that this lay under the present cathedral, in the ancient monumental centre, thus fitting the standard pattern described above. However, this assumption has been challenged by the more recent discovery of the very large church built by Caesarius in the ‘enclos Saint Césaire’, in the south-east of the city, against the rampart, showing that even in the sixth century the major church of the city was ex-centred, albeit within the walls.Footnote ¹³³ This does give a rather different slant to our understanding of the topography of the city and its ‘centre’ in our period.

Perhaps the concentration on ‘cathedral’ churches has been too dominant, however. It is clear that other types of church buildings also made up the ecclesiastical landscapes of our cities. In Marseille the extra-mural funerary church under the Abbaye de Saint Victor, which developed around a necropolis in the fifth century, has been known for a very long time;Footnote ¹³⁴ more recently, a funerary church with a substantial memoria was unearthed during the construction of a car park on the Rue Malaval.Footnote ¹³⁵ That more finds of this nature are likely is suggested by the recent discovery of a suburban funerary church dating to the fifth century in Nîmes.Footnote ¹³⁶ As we have seen, the exact nature of the Christian topography of Arles in late antiquity remains unclear, despite various mentions of different churches in our textsFootnote ¹³⁷ (see Maps 3 and 4). By the middle of the fifth century there were two sites associated with the Arlesian martyr, St Genesius: his tomb in the cemetery at Alyscamps and the supposed site of his execution on the other side of the Rhône at Trinquetaille. The medieval church of St Honorat still stands in the Alyscamps cemetery, presumably on the site of the earlier church to St Genesius – but no substantial remains of the early church or the martyr’s tomb have been found.Footnote ¹³⁸ In Trinquetaille, likewise, there are no traces of the early Christian chapel we know stood there, despite the later ‘Chapelle Saint-Geneste’.Footnote ¹³⁹ Finally, monasteries formed another, rather different, element of the urban Christian topography: as discussed in the previous chapter, Caesarius became abbot of a suburban monastery prior to his election to the bishopric. The original female monastery begun by Caesarius as bishop, destroyed in the siege of 507/8, was built outside the walls (most likely in the area of the Alyscamps);Footnote ¹⁴⁰ he built its replacement within the walls, close to his own cathedral church.Footnote ¹⁴¹ Right at the end of our period, Aurelianus (bishop from 546 to 551) built two large monasteries within the city walls.Footnote ¹⁴²

The scholarly concentration on the development of the cathedral church is linked to a broader focus on the growing institutional prominence of the church and the expansion of the role of the bishop as civic leader, major themes in the historiography of late antiquity.Footnote ¹⁴³ Public construction is indeed one arena in which we can clearly see the bishop acting very much as a traditional member of the ruling elite of the city. The development of the building complex often described as the ‘episcopal group’ makes the important civic role of the bishop visible in the fabric of the city, still tangible (to a varying degree) today at Cimiez, Riez, Fréjus, Aix and Marseille.

It is worth now moving just a little further afield, to Narbonne: while it is rare that we have detailed evidence of the chronology and personnel involved, there is a welcome exception when it comes to the case of Rusticius, bishop of Narbonne from 428 to 461. Two dedicatory inscriptions from Narbonne record not just the bishop’s own activity but also the nature and extent of the support that Rusticius received from elsewhere, thus demonstrating the interlinking of political, ecclesiastical and aristocratic power in the fifth-century city.Footnote ¹⁴⁴ We learn that Rusticius rebuilt the cathedral at the urging, and with the substantial financial backing, of Marcellus, praetorian prefect of the Gauls.Footnote ¹⁴⁵ We also learn that Rusticius’ building work was further supported by high-ranking aristocrats, as well as his ecclesiastical colleague and the faithful of the church.Footnote ¹⁴⁶ Rusticius’ ecclesiastical connections were notably impeccable: he was both the son and nephew of bishops and had cemented important relationships among the ascetic network of southern Gaul as a monk in Marseille.Footnote ¹⁴⁷ His example then is indicative of how tightly connected secular and ecclesiastical elites could be.

The bishops of Arles, a city of particular political importance in our period, were often represented as being close to secular power. Hilary presents Honoratus’ deathbed as graced by the most powerful men in Gaul, including the praetorian prefect, and his funeral as packed with those of the highest ranks.Footnote ¹⁴⁸ Hilary’s own accession was secured with the aid of the otherwise unknown ‘illustrious’ Cassius and his troops.Footnote ¹⁴⁹ However, relations with local elites and power brokers were not always smooth: Hilary admits that Honoratus’ election had been contested and that winning over opposition had been a key priority at the start of his term.Footnote ¹⁵⁰ Hilary himself publicly rebuked the praetorian prefect (seemingly Rusticius’ friend Marcellus) in the middle of mass in the Basilica Constantia and expelled him from the building.Footnote ¹⁵¹ More seriously, while having friends in high places could help secure election, bishops could also pay the price for these connections, as examples from Arles in the early fifth century clearly show in dramatic fashion, as we saw in the previous chapter. Heros was installed as bishop of Arles in 408 (alongside Lazarus at Aix) with the support of their supporter, the usurper Constantine III, but removed after his fall in 411. Heros’ successor Patroclus was himself assassinated in 426 by ‘a certain tribune’ called Barnabus at the ‘secret’ order of the magister militum Felix, according to Propter.Footnote ¹⁵²

The civic role of the bishop had several more or less political aspects. As we shall see, he was in theory at least appointed through public episcopal election, therefore in the mode of traditional civic life.Footnote ¹⁵³ In times of crisis bishops could be sent to negotiate treaties, as when the bishops of Aix, Arles, Marseilles and Riez were sent to Toulouse to negotiate with the Visigothic king Euric as the emissaries of the emperor Julius Nepos in 475.Footnote ¹⁵⁴ In this turbulent period the bishop was also not infrequently called upon to play his role in the ransom of prisoners, seen as one of the most important responsibilities of the bishop by the fifth century.Footnote ¹⁵⁵ These varying aspects coalesce into the notion of the bishop as the defender of the city, the true defensor civitatis; Sidonius lauded the generosity of Bishop Patiens of Lyon, who supplied corn at his private expense as far south as Riez and Arles after the depredations caused by the forces of Euric.Footnote ¹⁵⁶ Meanwhile, the bishop also had a judicial role, authorized by Constantine, with the establishment of the episcopalis audientia.Footnote ¹⁵⁷ The particular notion of the bishop as the protector of the poor is an important component of the ideology of the episcopal role in the late antique city, as has been influentially argued by Peter Brown.Footnote ¹⁵⁸ But can we actually see our southern Gallic bishops as consistently acting as this new type of patron, with the ‘poor’ as their clientela? The answer would have to be in the negative. While praise for the charitable activities of the bishop is a standard part of episcopal eulogy and hagiography, when it comes to the bishops of Arles the poor play a very minor role. Southern Gallic bishops tend to come much more clearly into view as allies or indeed clients of the powerful. What is very clear is that the growing power of the bishop was an important factor in both social and political relations in the city.

Finally, while scholarship has tended to focus very heavily on bishops – and on aristocratic bishops in particular – they were of course not the only clergy present in the city. Gallic councils, including the Council of Agde as presided over by Caesarius, paid a good deal of attention to regulating the behaviour of clergy, who they intended to be a separate caste, separated by their lifestyle and behaviour (including sexual), their obedience to their bishop and even their clothing.Footnote ¹⁵⁹ We can thus see the higher clergy in the cities as a distinctive group. While most would never make it that far, an ambitious deacon or presbyter would hope that his ordination would be a first step en route to the episcopacy.Footnote ¹⁶⁰ However, the status of minor or junior clergy – such as subdeacons, acolytes and exorcists – is harder to parse; Lisa Bailey aptly comments that they ‘occupied something of a grey zone’.Footnote ¹⁶¹ Unfortunately, we know little about their activities in our cities and can only speculate as to how they might have been involved in some of the contentious episodes we will consider next.

Power and Dependency in the City

As we have already seen, our ecclesiastical sources, including episcopal Vitae, offer some important hints as to social tensions within the late antique city insofar as they impacted upon the position of the bishop – especially notably in the case of such pugnacious and controversial figures as Hilary and Caesarius – but also, as we have seen, as regards their lesser-known predecessors, Heros and Patroclus.Footnote ¹⁶² Of course, the field of conflict went far beyond the individual cities and extended into broader local and regional rivalries, as well as tensions with Rome, as discussed by Ralph Mathisen.Footnote ¹⁶³ Hilary is an especially notable figure in this regard. He was accused by Pope Leo of travelling around his (disputed) territorium with an armed retinue.Footnote ¹⁶⁴ Within the city of Arles itself there are subtle hints in Hilary’s Vita regarding local opposition to the bishop from the ‘proud’,Footnote ¹⁶⁵ as well as the far less subtle anecdote about a run-in between Hilary and the praetorian prefect discussed earlier.Footnote ¹⁶⁶ These accounts – as well as the accusations made against Caesarius discussed earlier – are suggestive of inter-elite competition: what can we infer about a broader web of social relations and social tensions, taking in the subaltern classes of our late antique cities?

It is scarcely surprising that Hilary was all too happy to pick a fight with his social inferiors. Indeed, at the start of this chapter we considered an intriguing passage from the Vita in which a heckling crowd was punished by a divinely sent fire. This story is itself introduced in the narrative with an anecdote recalling how Hilary used to shout at people leaving church after the Gospel reading, threatening them with hell. The narrative then proceeds to the account of the confrontation between the crowd and the bishop: ‘A crowd of people had been vainly roused to come to him, ill-advised and misguided [inaniter excitata popularum turba et inconsulta deceptaque venisset]’,Footnote ¹⁶⁷ which was followed by the burning of ‘the greater part of the city’ (civitatis pars maxima). This certainly suggests a situation of conflict, but what exactly was going on? The Vita seems to suggest that the popularum turba – who must surely be understood as lower-status residents of the city – had themselves been stirred up by some other element or elements. Are the ‘crowd’ here being directed by members of the city elite (as the Vita perhaps imagines) or were they acting on their own initiative? We shall return to the question of agency and initiative later.

Sidonius Apollinaris’ account of his visit to the court of Majorian in Arles in 461 provides a further anecdote that could possibly be suggestive as regards the socio-political activities of the non-elite. Sidonius recounts how the attribution to himself of an anonymous, dangerously satirical text about Majorian’s court caused people he met to act strangely. We can assume that these are members of the local elite who feared being implicated – but what about people lower down the social scale? Sidonius claims that the whole(?) city was agog and angry (inter haec fremere Arelatenses).Footnote ¹⁶⁸ In his satirical description of the scenes in the city’s forum, he describes the uncomfortable Arlesians as a turba factosiorum.Footnote ¹⁶⁹ At the centre of the account is Sidonius’ accuser, one of the targets of the satire in question, the former prefect Paeonius, described as a persona popularis and as a classically flavoured demagogue, who stirs the people into a ‘sea of sedition’ and is later described as a contionator.Footnote ¹⁷⁰ To further add to the republican-era flavour, Sidonius both quotes and alludes to Lucan.Footnote ¹⁷¹ Sidonius’ literary game is all too evident: it is hard to imagine any genuine popular element at work here.Footnote ¹⁷² As common with elite texts – and perhaps especially visible in the case of Sidonius – our author’s literary concerns on the one hand and his concern with uniquely intra-elite politics and relations on the other make him of very limited use as a ‘source’ for the activities and motivations of non-elites.

Despite the persistent use of cliché, we can nevertheless glimpse in this letter a broader and telling elite concern with the public presence of the non-elite in political or indeed semi-political contexts. Latin texts make persistent reference to circuli: informal open-air gatherings of the urban plebs, associated with the kind of urban loitering of which elite writers so disapproved, not only the kind of loitering for which Pompeii provides such clear evidence but also, intrinsically, the idea of popular speech.Footnote ¹⁷³ In the case of fourth-century Rome, Ammianus Marcellinus discusses, disparagingly, ‘the many circuli gathered together in the fora, at the crossroads, in the streets and other meeting-places in which people were engaged with one another in quarrelsome strife, some (as you’d expect) defending this, some defending that’.Footnote ¹⁷⁴

In fifth-century North Africa Augustine uses the same word circulus to describe (disparagingly) the debates of the uneducated;Footnote ¹⁷⁵ the term clearly designates unauthorized speech, a theme we will return to in Chapter 4. Back in Arles, Sidonius refers to circulatores clustering around Paeonius, and hanging on his words.Footnote ¹⁷⁶ The circulator was a figure closely associated with the circulus, a public entertainer or storyteller who made his living from going around the people, a figure whose speech, again, was not sanctioned by the elite.Footnote ¹⁷⁷ In late antiquity these terms remained widely used to connote popular gatherings in public space, and help us – despite the negative spin – to populate the city with popular culture.

While the day-to-day activities of the non-elite in the city often remain obscure, urban rioting and violence are distinctive features of late antique narratives, such as Ammianus Marcellinus’ accounts of riots in RomeFootnote ¹⁷⁸ or Libanius’ lively depictions of urban violence in Antioch.Footnote ¹⁷⁹ In the case of Arles there are only a few instances of urban violence that we know of; Hilary’s confrontation with the insulting crowd should probably be counted as a very minor instance.Footnote ¹⁸⁰ The image of baying lynch mobs and religious violence remains a cliché of the period, nonetheless.Footnote ¹⁸¹ We can, however, look to more nuanced assessments, such as those provided by Julio-Cesar Magalhães de Oliveira, who has identified a significant trend of increasing competition and conflict over public spaces and indeed new forms of ‘popular’ power in the late antique city.Footnote ¹⁸² We could indeed interpret the account of the altercation between Hilary and the unhappy crowd as an instance where the urban populus make their views known to the bishop, and therefore as an episode suggestive of both urban tension and tension resolution (of a kind!).

Far more serious in its violence is the case of the attempted attack on the notary Licinianus, who had brought charges of treason against Caesarius in 504/5. Caesarius’ Vita recounts that when the bishop was released and returned to the city in 506, the populus responded not just by coming out to welcome him in a formal adventus but also by attempting to stone Licinianus, his accuser.Footnote ¹⁸³ The hagiographers clearly found this coincidence uncomfortable and tried to avoid making a connection between the two outcomes: hence the Vita narrates the stoning of Licinianus at a slightly earlier juncture in the narrative and ascribes the punishment to the orders of the Visigothic ruler, Alaric.Footnote ¹⁸⁴ It also recounts that the stoning was only halted at the last moment by the intercession of Caesarius himself, who requested pardon for his accuser.Footnote ¹⁸⁵ Modern scholars have cast serious doubts on this framing of the event, pointing out that stoning is not a Roman or Visigothic legal punishment, although it is the biblical punishment for treachery and false witness.Footnote ¹⁸⁶ So what actually happened? One possibility is that the clerical authorities themselves egged the population on, only to call them off again, as part of a basically staged ritual of reconciliation and episcopal authority.Footnote ¹⁸⁷ Another possibility gives more agency to the urban community, seeing them as spontaneously acting in this extra-judicial way in support of their bishop.Footnote ¹⁸⁸ We could thereby see them as involved in a ritual of community participation, exercising what they saw as popular justice, a legitimate power of the people.Footnote ¹⁸⁹

These episodes are suggestive of the different ways in which the urban non-elite could act as a body. Episcopal elections represent an obvious occasion at which the notion of a populus Dei was constituted, even if accounts of these tend towards hagiographical commonplaces on the one hand and historiographical stereotypes on the other.Footnote ¹⁹⁰ Late antique literary sources often praise unanimous elections: as when Leo the Great wrote in praise of the election of Ravennius to the episcopacy of Arles in 449 by ‘clergy, notables and plebs’.Footnote ¹⁹¹ At the other end of the scale, Sidonius gives typically (and typically suspiciously!) lively accounts of contested elections further north, in which a large population took an interest.Footnote ¹⁹² Peter Norton, in his study of late antique episcopal elections, stresses that ‘even in a society as undemocratic as late antiquity, the people played an important role in the choice of what was for most of them, the most important local official’.Footnote ¹⁹³ Magalhães de Oliveira, meanwhile, sees this as just one of the ways in which there was more space for urban non-elites to assert their collective favor (or, indeed, its opposite) in late antiquity.Footnote ¹⁹⁴

The urban plebs, it turns out, are visible in our literary sources, appearing as active agents in the public sphere, even if their motivations remain difficult to disentangle. It seems likely that professions and occupations still provided important means for constructing identity and organization. Public spaces, even taking into account the changes in urbanism in our period, still offered a place for the non-elite to gather and exchange views. Developments in late antiquity also offered new opportunities for the exercise of collective action, as with the rise to prominence of the bishop as a new urban patron and focus. And as we shall see in Chapter 6, the late antique city remained an important locus for festive behaviour, where social relations could be both modelled and challenged in different ways.

Performance and Leisure

With festive behaviour in mind, it is time, finally, to turn to what is the traditional place to start looking for popular culture in the ancient city: with the spectacular entertainments and leisure.

After the troubles of the third century, the traditional spectacles were all but dormant in late antique Gaul: even the most generous recent research struggles to find much sign of traditional ludi and munera after the middle of the fourth century.Footnote ¹⁹⁵ Although dating remains difficult, archaeological evidence suggests that, in most cases, entertainment buildings were abandoned or incorporated into new city fortifications from the fourth century onwards. In the fifth and sixth centuries they tended to undergo new forms of occupation, notably artisanal and domestic, as well as suffering from demolition and spoliation. This applies to southern France as much as elsewhere in late antique Gaul. For instance, the theatre at Apt seems to have been abandoned as early as the end of the third century, with re-occupation setting in around the end of the fifth century,Footnote ¹⁹⁶ which seems to have been a key period for re-occupation.Footnote ¹⁹⁷

Arles, home to a theatre, amphitheatre and circus, was exceptional. Even the very presence of a circus in Roman Gaul was unusual – the only other known examples are at Lyon, Trier and Vienne. It seems unlikely that the theatrical and amphitheatre games were still ongoing in late antiquity. As elsewhere, conclusive archaeological evidence for the definitive date of the abandonment of the theatre and amphitheatre is lacking (not helped by enthusiastic nineteenth-century restoration), though a date early in the fifth century seems most likely for both, thus fitting the broader regional pattern. For the theatre, the most recent assessment dates its incorporation into the new fortifications of the city during the fifth century.Footnote ¹⁹⁸ An episode from the Vita of Hilary has been influential in this dating. It recounts how Hilary miraculously healed the foot of a deacon who had been injured while supervising the removal of marble from the proscenium of the theatre for re-use in the construction of the bishop’s new church (it landed on his foot).Footnote ¹⁹⁹ It is of course possible that performances were still put on in the theatre after this point, but this seems unlikely. As for the amphitheatre, there is evidence of spoliation for construction material and of ‘parasitical’ constructions, as well as the presence of mid-fifth-century coins in the subterranean areas, strongly suggesting re-occupation at this point, although this cannot be conclusive.Footnote ²⁰⁰ The circus, however, presents a rather different picture: the continuity of the circus games in late antiquity was no doubt due to the prestige of Arles as an imperial capital, and, furthermore, down to the occasional presence of the emperor in person.Footnote ²⁰¹ It benefited from substantial improvements in the early fourth century (which might have included the erection of the obelisk that today stands in front of the Hôtel de Ville).Footnote ²⁰² At the start of the fifth-century, housing was built abutting the exterior of the circus in what seems to have been an official, planned project.Footnote ²⁰³ However, this did not stop the chariot races, which seem to have continued even up to the middle of the sixth century, after which they seem finally to have been abandoned.Footnote ²⁰⁴

When it comes to the place of the traditional spectacles in the life and ideology of the late antique city, there is a striking disconnect between the archaeological and literary ‘evidence’. While the archaeological evidence testifies to the cessation of traditional civic spectacle entertainment, ecclesiastical discourse manifests a continuing obsession with the ‘immoral spectacles’.Footnote ²⁰⁵ The most notorious case is Salvian (writing in the 440s), who devotes a goodly portion of his jeremiad on the state of things to decrying the immorality of the spectacles – before admitting that they had generally already been discontinued in Gaul, for financial reasons.Footnote ²⁰⁶ Salvian thereby demonstrates just how good the spectacles were to think with – or rather to rail against. By the time he wrote, rhetorical condemnation of the shows – as immoral and polluting, as tainted by their pagan origins – was a firmly established and popular (read, highly clichéd) topic in preaching. The shows, especially the theatrical and gladiatorial games, functioned as a potent symbol of the profane world against which the late antique church sought to define itself, and perhaps too as a historicized form of ‘profane’ culture.Footnote ²⁰⁷

We see even more seemingly ‘practical’ texts continue to make ideological capital out of the spectacles. The canons of church councils include strictures aimed at both performers and those attending the shows, especially around the mid-fifth century; that is, during the period when other evidence suggests the ludi were largely in abeyance. Excommunication was prescribed not just for charioteers and theatrical performers but also for anyone attending the spectacula after church.Footnote ²⁰⁸ It is not surprising then to find Caesarius repeating traditional condemnations of the ‘cruel and shameless’ spectacles on a number of occasions.Footnote ²⁰⁹ We find him, as so often, in full ‘cut-and-paste’ mode. His vocabulary describing the games is traditional (cruenta, furiosa, turpia), including the old chestnut about the shows being the pompae diaboli.Footnote ²¹⁰ His stock attacks tend to be brief and formulaic, and couple together all the traditional types of entertainment, as is also typical elsewhere.Footnote ²¹¹ The essentially symbolic nature of the critique of the spectacula developed by Caesarius is most clearly highlighted in Serm. 152, where he develops a metaphorical account of the amphitheatre games at some length. He begins his account, as is common, with Paul’s claim that humans have been made ‘a spectacle to God and the angels’,Footnote ²¹² and develops the idea of a life as an arena mundi. He then goes on to claim that within each of us is a ‘spiritual amphitheatre’ and proceeds to elaborate upon this idea, listing a wide range of animals, representatives of different sins, that we find in our internal amphitheatre; that is, in our consciences.Footnote ²¹³ The animals listed certainly go far beyond what one might expect to find in the amphitheatre at Arles, including snakes, pigs, elephants, panthers and vultures.Footnote ²¹⁴ Using Caesarius’ sermons as evidence for the continuation of theatrical and arena games would therefore be unwise.Footnote ²¹⁵

The cessation of the traditional spectacles was clearly caused by much more than the dislike of bishops and monks. It speaks to some of the fundamental changes in the late ancient city that we have been considering. In the place of classical urbanism, funded by traditional euergetism, we have seen the built environment become in many respects a more utilitarian space, in which the all-too-real threat of military invasion made massive entertainment structures the most attractive as buildings to incorporate into fortifications. As for the funding of the games, it is clear that local elites were clearly either no longer willing or able (or both) to act as traditional public benefactors. The late continuation of the circus games in imperial Arles is an interesting exception that demonstrates the role of contingency and the need for a more nuanced interpretation of the fate of public entertainment. The view of the church was not necessarily the same as that of imperial government (whether that of Rome or the successor states) or indeed that of local elites (whose views remain unfortunately obscure). The idea that civic festivals and their spectacles provided for a laetitia populi, that should be preserved by a good emperor,Footnote ²¹⁶ held a strong ideological pull that the church had considerable difficulty in countermanding.Footnote ²¹⁷ We will indeed see this kind of triangulation in detail in Chapter 6 in the discussion of celebration of the Kalends of January. In the meantime, we can begin to think about ways in which the transformation of the ancient city facilitated new forms of popular culture.

In our search for popular culture in late antique Arles, as in other cities in southern Gaul, we must look beyond the traditional Roman spectacula. The demise of the majority of these left space of various kinds – for the church to exert their claim to social and ideological control but also for others, as individuals or as groups, to develop alternative practices. As part of a study of the transformation of popular culture in late antiquity we will not least have to think in terms of a more ‘bottom-up’ popular culture. Not all types of performance required grand structures or official funding. Some of the later Gallic canons do indeed seem to have had more informal entertainments in mind, for instance, the Council of Vannes in 461 was concerned with dances at social gatherings, and their stricture against clergy attendance at such events was repeated exactly at the Council of Agde, presided over by Caesarius, in 506.Footnote ²¹⁸ Archaeological evidence unfortunately cannot help us explore these activities, just as it cannot enable us to view other sites of urban popular sociability such as taverns or indeed brothels, as can be done for Pompeii. Nonetheless, we do need to think in terms of a ‘do it yourself’ popular culture, involving such activities as drinking, gaming, singing and dancing, as well as various types of informal performance, held in the context of informal sociability and conviviality rather than sponsored by the local civic elite. I shall return to this theme in more detail in later chapters, looking more closely both at the practices involved and the ecclesiastical opprobrium they provoked.

Conclusions

Clearly, the evidence from cities like Arles and Marseille cannot provide us with as rich a picture of popular culture as can be gleaned from Pompeii or even Aphrodisias. However, considering both the urban environment on the one hand and key topics such as employment, identity and social relations on the other provides a framework for understanding popular culture as fully embedded in late antique society and culture. Southern Gallic cities were home to a culturally and socially diverse non-elite population who maintained some of the traditional modes of self-expression but also made use of new opportunities, responding to the changing circumstances of the period. We have been able to see, even through and despite the stereotyped depictions given by the elite, the presence of the urban plebs in public space and their capacity to make their voices heard.

A shift of focus from the tired debate of transformation/decline of the classical city is needed in future research in order to think about how the non-elite experienced urban life in late antiquity. There were substantial changes in terms of both public and private space in our period, changes that impacted upon living standards and opportunities for the experience of popular culture alike. Despite the changing use of public spaces, longstanding forms of non-elite communication and self-expression continued. However, changes in civic leadership and ideology – the decline in the prominence of the traditional civic elite and the rise of the figure of the bishop – also provided new opportunities for this expression and communication, which could be taken advantage of by the urban populus. The widespread desuetude of the sponsored spectacles entailed the development of more ‘do it yourself’ forms of entertainment, as well as facilitating spaces for ideological takeover by ecclesiastical discourse. Later chapters will probe more deeply into both substantive aspects of popular culture (what did people actually do?) and interpretations of this culture, but for now it is time to turn to the countryside, where looking at the material record in particular will again enable us to see the social and economic contexts of this culture more closely.