Introduction: Beyond ‘Romanisation’
Cultural expressions were changing on three continents under Roman domination. Some might wonder why a homogenous ‘imperial’ culture was not developing after so many generations of Roman occupation. Though we may identify certain imperial discourses about the benefits of the empire, like pax, prosperity, and humanitas (Haeussler 2013a: 29–30, 58–63), the available evidence often indicates an increasing diversification of the various regions in this vast, multi-cultural and multi-ethnic empire, especially if we do not focus on artificial units, like provinces, but on local circumstances, often on a microcosmic scale, as each context provides a unique environment for societal, cultural, and religious changes to take place. Having jettisoned the outdated and inadequate concept of ‘Romanisation’ years ago (see discussion in Woolf 1997; Millett 2004, Mattingly 2011, Haeussler 2013a, and Hingley’s 2014 response to the endorsement to re-instate the ‘Romanisation’ paradigm by Versluys’ 2014), we have also freed ourselves from this simplistic (and colonialist) concept that takes for granted the diffusion and spread of Roman culture from the ‘centre’ to the ‘periphery’. Over the past 25 years, the rejection of ‘Romanisation’, notably in Anglophone scholarship (for France, cf. Le Roux 2004), has opened up new ways to re-think our methodologies, deconstruct ‘Romanisation’ into its constituent parts, and explore people’s diverging experiences of Roman imperialism in the various provinces, whilst emphasising the development of local cultures.
Since Martin Millett’s publication of Romanization of Britain in 1990, we have seen a staggering number of publications on sociocultural change in the Roman world, many of which continued to employ the Romanisation terminology while experimenting with innovative and useful methodologies to advance the subject (see review in Haeussler 2013a: 27–73). Post-colonial perspectives, for example, have provided food for thought since Marcel Benabou’s ‘La resistance africaine à la romanisation’ (1976). At the 1995 Roman Archaeology Conference, John Barrett was an outspoken critic of the term ‘Romanisation’ and advocated to research instead ‘the mechanisms of agency’ (1997: 63). Edward Said’s post-colonial term ‘discrepant experience’ (1993: 35–50), though more fixated on the fundamental, almost irreconcilable experiences between colonisers and colonised in modern times, was already adopted in 1995 by David Mattingly to supplant the despised Romanisation paradigm (see e.g., Mattingly 2011). Over the years, we saw an astounding number of works on ‘identities’ (military, diaspora, etc.) in the Roman world (see e.g., Haeussler 2008; Eckardt 2010; Gardner 2010). Our understanding of sociocultural change in Roman times was further advanced by embracing globalisation and glocalisation studies (Haeussler 1998: 17; 2013a: 60–61; see also Versluys 2014; van Alten 2017; Cobb and Haeussler forthcoming), network analysis (e.g., Woolf 2016) as well as creolisation (Webster 2001) and bricolage (Roth 2003)—in addition to koineisation and cosmopolitanism. Other models have gladly fallen out of favour during the 1990s, like acculturation, assimilation, and core-periphery models, though they still flare up occasionally (see review in Haeussler 2013a: 27–38, 40–41, 54–59).
We are certainly closer to understanding some of the underlying developments of socio-cultural change across the Roman world as well as the actors and their motivations. But many methodologies also have shortcomings that need to be addressed. Many of these ‘post-colonial Roman archaeologies’ (to use Hingley’s term [2014: 20]), though providing valuable methodologies, are unfortunately based on limited source material, often focussing on only one province, most frequently on Britannia. Studies that follow in the footsteps of Benabou and Said offer an anti-colonial discourse of ‘cultural resistance’ that is more akin to 20th-century colonialism, but hardly suitable for Roman times (see already critique in Haeussler 1998: 14). And apart from concentrating on more high-status protagonists grouped in artificial categories, like ‘military’, ‘urban’ and ‘rural’, many studies still fail to address what John Barrett was looking for 23 years ago (1997), namely an understanding of the mechanisms of agency?
Based on the existing scholarship, we want to focus on human agency and human interaction in the Roman world. Each human being has agency, but the outcome of someone’s agency depends on multiple factors: in a small, intimate community, agency can be more significant in bringing about change compared to large, notably urban communities (see Fuchs 2001: 39–40; see also Jürgen Habermas’ seminal 1981 work on ‘communicative action’; on agency and archaeology, see e.g., Gardner 2010; Robb 2010; Rüpke 2015). For us, the individual, the choices (s)he took and the motivations that triggered these choices are at the centre of our research. In contrast to, for example, Caspar Meyer (2013: 306–307), for whom ‘the focus on agency and cultural perceptions’ is controversial ‘in archaeological studies of cultural interaction’, we consider agency essential for understanding change in any society. Moreover, the Roman world also provides us with an abundance of evidence for individuals and their actions, choices, and aspirations: a million inscriptions, papyri, writing tablets and graffiti from Roman times, written not only in Latin and Greek, but also in Aramaic, Celtic, Egyptian, Etruscan, Palmyran and many other languages. Many of these also attest the individual person’s bilingualism and code switching (for which, see Adams 2003). Many texts were handwritten by sub-elite people without the help of a scribe or stonemason. The archaeological evidence is even more overwhelming and ranges from grave goods to dress items, artistic representations, architectural choices, and many more, as we shall see below.
We are dealing with a myriad of individual acts of agency that together produce a creative process that shaped people’s lives in the Roman world. And we are not just focussing on the rich and wealthy in the empire, on these seemingly ubiquitous—but ambiguously defined—‘elites’ who are seen as the main protagonists and architects of change in many studies. In many ways, we are more interested in the sub-elites, whatever their gender, age or status, like the merchants, craftsmen, tenant farmers, those in the logistics business, the small holders, soldiers, veterans, day labourers, midwives, dancers, gladiators, prostitutes, slaves, ex-slaves, and children to mention but a few. Each of them had the chance to make their own choice, like their choice of dress and jewellery, their choice of diet, the choice of visual language, techniques, artistic repertoire, naming practices, and the choice to engage, ignore or transform cultural and mythical connotations. These choices may not always be spontaneous and completely free – the level of spontaneity depends on time and place: in the first centuries BC and AD, new material evidence reveals major new developments which resulted from countless actions by individuals who had a more profound effect compared to other periods because existing sociocultural understandings and behavioural patterns were losing their validity during periods of uncertainty.
Most of his/her lifetime, the individual was facing many expectations and group pressures (and we do not just mean pressure from the ‘local elite’ as they in turn were facing pressures from ‘below’ and their peers). Even though we are dealing with a huge empire, the majority of inhabitants lived in small-scale communities, from the family units, to farms, villas, villages, and vici, to the medium-sized provincial towns of 10,000 or 15,000 inhabitants where many people would have known each other and attended the same events. Apart from experience, performance and human interactions that were framing people’s activities, an individual’s actions and decisions can also be expected to have repercussions on a local or regional level: it is, therefore, important to aim for an understanding of group dynamics (v. infra). Roman imperialism instigated many developments, some of them inadvertently. One of the most fascinating ‘side effects’ is the level of individualisation that we can identify in the archaeological and epigraphic record across many Roman provinces, notably during the first to third century AD. Important factors that made this individualisation possible include migration and urbanism in conjunction with monetisation, the commodification of social transactions, social mobility, and civitas Romana to name but a few (see Haeussler 2013a: 66–67; 2013b: 185–187; Haeussler and King 2017). Liberated from pre-existing social constraints, we can identify processes that were empowering people of various social statuses. Having acquired status and wealth, many sub-elite people, like traders, craftsmen, and ex-slaves (and some slaves), could make choices, such as what dress to wear, what hairstyle, what food to eat, what table ware to use, what language and writing style to use or what ‘guilds’ (collegia) to join; do you invest in your business, aspire for public office, or perhaps move to another city or province?
Our focus is therefore on understanding the cultural dynamics in the Roman world as a bottom-up development which leads us to re-think existing theories and methodologies, notably those relating to the subaltern, like creolization and bricolage, and build up a new concept that we name ‘creolage’.
Are We All ‘Creoles’ and ‘Bricoleurs’?
‘Nous sommes tous les creoles’1 according to the famous French philosopher from the Carribean, Édouard Glissant (1998). For him, multiculturalism is merely a convenient way to get over cultural problems where one might take ‘a bit of everything and mix it’ (‘on prend un peu de tout, on mélange tout ça’). By contrast, creolisation for him goes beyond this: in any given point of the world, cultural contacts did not only produce, in Glissant’s words, ‘un simple métissage, mais une résultante imprévisible’2 (1998). The unpredictable outcome is, in a way, a further stage in cultural evolution, creating a new culture, just like the Caribbean cultures Glissant was experiencing, analysing, and studying (1998). This may mirror our understanding of many contemporary societies. We often talk about multiculturalism while we often may merely tolerate each other’s cultures. Within our global world, are we currently in a process of creolising, of developing new local cultures? Or is there too much ‘resistance’ since people are afraid of losing their identity? Does not ‘postmodern “conditio humana” resemble … the “conditio Caribbeana”’ (Palmié 2006: 434) and are we not all ‘Carribeans now in our urban archipelagoes’, in the words of Clifford (1988: 173).
How similar, one may wonder, was the Roman empire in this respect (if only superficially). It was a multi-ethnic, multi-cultural, and multi-lingual world and we see an unprecedented scale of cultural developments, social interactions, and spatial mobility in all Roman provinces as well as interactions with the lands beyond its frontiers, from Ireland and the Baltic Sea to India and China (see e.g., Cobb and Haeussler, forthcoming). In this world, the concept of creolisation can provide a useful paradigm to develop a bottom-up approach to cultural changes in the Roman provinces, notably in response to colonial/imperial power, as already suggested by Jane Webster (2001). Though increasingly common in the humanities—the ‘universal creolisation’ in the words of Christine Chivallon (2013; see also Palmié 2006)—it is rarely applied in Roman archaeological studies. Having been developed for modern colonial situations in the Americas, creolisation points us away from local societies merely ‘imitating’ or ‘mimicking’ Roman culture, and focus instead on the creation of a new culture, different from both the ‘Roman’ and ‘native’ ones. Indeed, if we look at certain Gallo-Roman or Britanno-Roman artefacts, like the sculptures of deities, we may often wonder whether a pre-Roman inhabitant might have been able to recognise that, for example, the horned deity surrounded by Apollo and Mercury on the famous relief from Reims is meant to depict a ‘Celtic’ god—supposedly ‘Cernunnos’—or that a goddess sitting on a throne and surrounded by horses was meant to represent Epona (see Haeussler 2012 for discussion of such interpretationes with further bibliography).
The research on many Roman provinces has shown that there never developed any coherent ‘Gallo-Roman’ or ‘Britanno-Roman’ culture. Do we just need to create smaller catchment areas for these new cultures, like a civitas territory? More importantly, we need to consider that different cultural discourses were going on simultaneously depending on people’s status and gender, their profession, age, and wealth. Do we only see this ‘creole’ pan-Mediterranean culture in the late Roman Empire when Christianity was imposed on all Roman citizens? Again, the answer is no since material culture in the fourth to sixth centuries had once again become very localised and only certain elite networks—notably in relation to the church—appear more ‘uniform’. Creolisation is a long-term process; it is the end of a complex process of interactions, adoptions, and adaptations which may have resulted in the creation of a ‘new culture’, but these cultures are usually very localised.
And while new cultural understandings and representations were being developed that were neither ‘native’ nor ‘Roman’, we must also be aware of the shortcomings of ‘creolisation’: it not only puts too much focus on two ‘cultures’ interacting with each other, but we also have to acknowledge that ‘creolization’ in the Americas developed in particular historic conditions that include deracinated and deported (or ‘displanted’) individuals combined with a racialised concept of slavery and extreme social violence, a situation in which, one may argue, one sees the emergence of a ‘counter-culture’ or even a ‘substitute’ to the dominant colonial culture, virtually creating a kind of parallel culture in the Caribbean in areas—like Afro-Christian religions, Voodoo, and carnivals—where it cannot threaten the colonial and/or dominant power discourses (see Chivallon 2013: 51–52, 67). Despite the term’s usefulness, the central aspects that created ‘creolisation’ processes in the Americas are therefore quite different from the Roman world.
Let us look at the concept of bricolage that has gained increasing popularity in Roman cultural studies in the past twenty years (Roth 2003; Haeussler 2013a: 58–61). For instance, Ian Haynes (2013: 239–241) wrote about the ‘armoury of the bricoleur’ when studying the ‘disparate origins’ of military equipment used by auxiliary units in the Roman world. The bricoleur appears as the do-it-yourself improviser who puts pre-existing features (material and non-material), which are easily at hand, together in new ways, for example by using them for a purpose that they were not intended for. This use of the term bricolage is individualistic and uniquely creative, and therefore ideally suited to study the mechanisms of social agency, as we discuss below.
As Véronique Altglas (2014b) has demonstrated, all possible disciplines nowadays describe everybody who is innovating as ‘bricoleur’. Though one likes to quote Claude Levi-Strauss’ The Savage Mind (1962) for the origin of this concept, the term has often been abused and re-defined. When we go back to Levi-Strauss’ seminal work, we not only have to insert it in the wider anthropological/sociological scholarship of his time and the focus on the creation of myths, which in turn is based on his 1955 work on ‘The Structural Study of Myth’ (see discussion by Johnson 2012: 357–358; Altglas 2014a; 2014b): ‘The characteristic feature of mythical thought is that it expresses itself by means of a heterogeneous repertoire which, even if extensive, is nevertheless limited. It has to use this repertoire, however, whatever the task in hand because it has nothing else at its disposal. Mythical thought is therefore a kind of intellectual “bricolage”—which explains the relation which can be perceived between the two’ (Lévi-Strauss 1962: 16–17).
In the realms of structuralism, Lévi-Strauss’ concept is less interested in the ‘bricoleur’, the artisan or author, but in processes of production of myths and cultures (Soar and Tremlett 2017: 428). And while Lévi-Strauss focuses on the ‘intellectual “bricolage”’ in creating myths, insisting on the ‘incidental nature of myth-making’ (Altglast 2014b: 476), it was Derrida (1978: 285) who expanded the concept when suggesting that ‘every discourse is bricoleur’ or even more strongly in the words of Deleuze and Guattari (1984: 1) for whom ‘we are all bricoleurs’. The use of the term has thus moved more and more away from structure and patterns to ‘fragmentation, arbitrariness, and eclecticism’, and from ‘myth-making’ to ‘individual practise’ (Altglas 2014b: 482): the term is equally used today to understand cultural resistance of the subaltern strata of society as well as the subversion of the dominant culture, for example in a colonial context as well as in modern-day youth cultures. Nowadays, in a postmodern fragmented world, highly individualistic ‘bricoleurs’, freed from collective norms and social values, may be free to mix and match whatever they like, thus having rendered, in the words of Altglas, ‘bricolage utterly ephemeral and eclectic to the point that the survival of styles themselves could be questioned’ (2014b: 480). But is this ever the case? Even in modern societies, ‘bricolage is neither random nor boundless’ since there are ‘pre-constraints and limits in contemporary bricolage’, whereby the ‘relations of power and culture … remain relevant to conceptualise bricolage’ (ibid.: 489–490). Bricolage is a useful concept to understand diversification of cultural repertoires, inter alia because it allows us to understand the actions of the ‘bricoleurs’ and how they ‘negotiate (with difficulty) with the foreignness and particularisms of what they appropriate’ (ibid.: 489).
Consequently, we need to discuss motivational forces that shape the individual social agent’s decisions. Being able to make your own cultural choice implies that people were—at least to some extent—freed from collective norms and values. For Thomas Luckmann, religious ‘bricolage’ in modern societies was made possible when religious institutions lost control over society: ‘The privatization of individual existence is linked to the privatization of religion’ (1979: 136). Individualisation and bricolage are quite apparent in the religious sphere during the Principate when we can identify a multitude of cults and cult places—some of a more makeshift character—that reveal the lack of overall control by the official religious institutions, like the local ordo and its municipal priests (see Haeussler 2011): one reason is of course that the demographic composition of the various social groups within a civitas have become so heterogeneous and changeable that local myths and the local pantheon are no longer important factors for the identity of people that engage within a much wider, entangled and global world.
How is this do-it-yourself approach to cultural expressions, to art, religion, and identity, possible in the Roman world? One may suggest that our ‘bricoleurs’, our social agents of whatever status or gender, ‘choisiraient et assortiraient librement des ressources symboliques dans des combinaisons personnelles, fondées sur la subjectivité individuelle’3 (Altglas 2014a: 326). Unlike Lévi-Strauss, we wish to focus on the social agents making choices, both elite and sub-elite. This does not imply that bricolage was used in Antiquity to subvert the dominant culture of the colonisers, as in some modern examples, though we need to keep this possibility in mind. In the changing and entangled world of the Roman Empire, bricolage was used by people from all ways of life and we need to examine their decisions whatever their status, gender, or occupation.
Is not every (free) person a bricoleur in the Roman world? Each individual can just pick and mix artefacts and ideas that are readily available in her/his ‘environment’, and then ‘cobble together’ their own cultural repertoire: dress, jewellery, hairstyle, personal names, myths, deities, architectural design, art, sculpture, food, religious preferences, and so on. But this bricolage is not random or arbitrary. Despite the astonishing diversity of cultural expressions in the Roman world, we are not dealing with an endless number of individualised and unique results of bricolage. In other words, the majority of personal decisions must have been influenced by other factors that need to be accounted for. We must insert people’s decision-making process within wider social, economic, and cultural discourses, taking into account people’s individual experiences as well as group and peer pressures. When we talk about the subaltern strata of society, we need to ask, for example, whether they had the means to make their own decisions; they can cobble together a new identity, but there might be clear limitations in the resources and tools available to them (see also Toner 2015). This ‘bricolage’ or ‘tinkering’ seems to have had some profound impacts. Among others, many of the late Iron Age cultures and artistic styles largely disappeared in the Roman West, like La Tène culture. And this also opened up the possibilities for people to develop new artistic and cultural representations in the first centuries BC and AD which has often been interpreted as a resurgence of ‘native’ styles (Webster forthcoming).
Bottom-Up Approach: Individual vs. Group Pressure
Many of these problems outlined in the previous section can be addressed by employing some of the useful aspects from cultural theories, like creolisation and bricolage. But we must also go beyond them if we wish to focus on the individual and on collective thinking regarding the developments in Italy or any of the provinces. In other words, the individual’s choices and actions were taking place within certain collective expectations and cognitive understandings, but at the same time, human actions re-produce these cognitive understandings and societal actions—hence, there is always scope for change (Haeussler 2013a: 64–65). This will allow us to understand the singularity of self-display and some exceptional choices, styles, and creations in art, religion, onomastics, dress, etc., taking into account multi-lateral interactions on a local, regional, and global scale (see also Meyer 2013; Pitts 2019). We need to advocate a bottom-up approach and be aware of the multiple experiences, expectations, norms, and social pressures that shaped the individual social agents in diverse geographical settings in the Roman Empire.
If we look at art, we see many new developments across the Roman world taking place, new visual languages and iconographies being developed. Was this the result of merely ‘arbitrary’ bricolage or do we witness the development of a new creole culture? Neither! And perhaps a bit of both. There are so many factors that could have contributed to the creation of a new artistic representation or a new artistic style. Among them there are often very localised developments, unique to a community’s understanding of art; social pressure on the social agent; diverse financial resources; educational background; the individual’s experience beyond the local community; and the zeitgeist, i.e. the cognitive trends, to mention just a few aspects. Within a few miles, quite different types of artefacts were produced, while at the same time we often see the use of imported products that were commonplace across several provinces. People were developing new unique—often local or regional—forms of jewellery (e.g., dragonesque brooches, hairpins, glass beads), pottery styles (face jugs, local Samian ware imitations), glass styles; they are also adapting traditional forms to new materials and techniques, for instance for textiles (e.g. use of looms, dyes, materials [silk], etc.).
Let us investigate some examples that combine several important features, like visual language, writing, craftsmanship, socio-economic status, and social expectations among sub-elite people. In the north-west Italian city of Turin (Augusta Taurinorum) we find high-quality tombstones of traders and craftsmen that depict the occupation of the deceased (see Figure 1). They contrast heavily with the rather unusual and irregular tombstones discovered in various villages in the Canavese, just a few kilometres to the north of Turin (see Figure 2). Though so different from each other, both reflect ‘new’ developments during the first and second centuries AD. On the one hand, there is the successful, wealthy craftsman or trader, rarely visible in pre-Roman times, who was emulating (or mimicking?) elite-style tombstones, but adapted them to proudly represent his own profession depicting the symbols of his trade on the stele, be it a comb maker, cloth merchant, wheel maker, and so on. This also includes women, such as Cornelia Venusta who was a clavarius, ‘keymaker’, and Antistia Delphidis, a tessarius lignarii, ‘wood carver’ (Guillamón 2003: 170–171; Haeussler 2013a: 282–285).
Here we see a phenomenon that happened across the Roman world: social groups—both existing and emerging ones—were acquiring sufficient wealth to compete with elite forms of representation, from the dress and lifestyle choices, mingling in selected social groups, like collegia or initiation cults, to the ostentatious burial, and also the wish to advertise one’s achievements for generations to come by leaving a memorial.
These high-quality tombstones of traders and crafts(wo)men stand in stark contrast to those we find across the villages of the Canavese. These are no isolated peripheral communities, as they are part of a heavily centuriated countryside between two Roman urban foundations, Eporedia (modern-day Ivrea, founded as colonia civium Ronanorum in 100 BC) and Iulia Augusta Taurinorum, founded in triumviral times. And like the rural population in other parts of the empire, the people in the Canavese must have travelled to marketplaces, like Eporedia, Augusta Taurinorum, or Bodincomagus, to sell their products and buy supplies. And yet, we can recognise how different forms of new cultural expression developed virtually side-by-side and almost simultaneously. Are we dealing with economic differences among the deceased or limited access to a professional stone mason? Not necessarily since those people probably owned their own land and since Caesar’s grant of citizenship in 49 BC, they should have been Roman citizens just like their contemporaries in Turin and Ivrea.
Here we see what some people might call ‘bricolage’ or ‘creolage’. Our various ‘bricoleurs’—male or female—cobbled together a new form of representation by combining traditional and Roman-style features, among them the use of local stelae of irregular shape and rather primitive writing with some unusual formulae. This is no ‘cultural resistance’ or ‘vulgar Romanisation’ (see Alcock 2001), but the result of conscious decisions embedded in people’s cognitive understandings. For instance, these irregularly shaped stelae are hardly suitable for writing; their usage reflects late pre-Roman Iron Age traditions and we see a similar development further south in Piedmont around the modern town of Cuneo. Although the area had been Roman since 222 BC (and reconquered after the Second Punic War, 196 BC), the material culture in the Canavese was very much embedded in La Tène culture, down to the early Principate. Because of the hardness of the stone, the inscription had to be short. The people (or the stone cutters) had knowledge of writing and used some Roman abbreviations they found useful, like VA for vixit annos. We see another form of reasoning in this decision. Age, especially high age, like 70, 75 or 80, is more important in these rural communities than the Roman concepts of career and success. In other words, these inscriptions reveal rather different cognitive understandings. For the importance of age in the Canavese inscriptions, we may suggest external pressure within the more competitive economic world of the early Principate: age can also be employed as a marker of ancestry and long-standing land ownership.
Like in other parts of the Roman world, individuals were thinking ‘outside the box’ in the Canavese. This means that their unconventional and novel thinking did not just result in copying or emulating a certain ‘Roman’ epigraphic culture, unlike their urban contemporaries in Turin. Instead, they created their own style. It is likely that an individual setting the first of these stelae created a kind of domino effect, though other individuals added extra features. The idea was taken up by people in other communities in this region, part of the local network. In a way, bricolage by an individual was setting in motion a process that led to a microcosmic form of ‘creolisation’, developing a unique cultural and artistic form of status display for this rural area. Are we dealing here with an attempt to subvert the dominant culture, the colonial culture in Turin and Ivrea? This seems unlikely, though it is probably part of a coping mechanism in times of change.
More than ever before, the individualisation of societies across the Roman world, especially during the first to third centuries AD, allowed people to make personal decisions. It is important to take into account societal developments across the Roman world. By the first century AD, we can identify a myriad of individuals in our sources that attempted to improve their status, their wealth, and their lifestyle. How many people left their home community and their family in order to start a new life in another region, in another province; how many men joined the Roman legions and auxiliaries, ending up in far-flung provinces after their retirement where they married and had children (see e.g., Eckardt 2010; Haynes 2013)? How many people tried to find a lucrative niche market to make a profit? In this respect, Maaike Groot’s (2008) analysis of a sub-elite farm near Nijmegen effectively exemplifies how subsistence farmers were adapting to this ‘global’ word by specialising in a profitable niche market: specialising in horse breeding meant having to buy food and imported products on the local market, thus being forced to ‘bricolage’ and ‘glocalise’ their cultural identities. Roman society put so much emphasis on an individual’s success: it is much less about the ‘collective’, but about individual survival strategies, about wealth, status, and one’s political and military career. Whether farmer, craftsman, or trader, wealth became synonymous for status and can there be a greater honour than the council setting up a statue in your honour as a benefactor to your community (whether your hometown or your chosen home, and independent of your status). One can already visualise the multitude of motivational forces that were pressurising the individual.
Can we possibly imagine the social and cultural complexity that an individual was facing when living in one of the hundreds of cosmopolitan and multi-ethnic urban centres that were thriving in Roman times? There were the farmers who commuted to the city to sell their produce or moved there to start a new life; there were immigrants of various cultural backgrounds who brought their ‘traditions’ and aimed to leave a mark, like the famous case of Barates from Palmyra who set up a Palmyran-Latin bilingual tombstone, discovered south-west of Southshields fort in Britain (RIB 1065; CSIR-GB I.1, 247); we can identify ex-slaves who tried to survive, some of them managing to climb to the top of their local community, as seviri and benefactors; there are the day labourers who made a living in a monetised economy; there are the ex-soldiers and their experience across the empire; there also was a need for lawyers knowledgeable in Roman laws as well as patrons and flamines that provided a connection to the centre of power, Rome, and domus divinae. Local landowners were trying to hold onto their power and status, for example through their ostentatious status display, but they also had ambitions far beyond their community: how many held offices in several cities, how many became rich equestrians and even senators? This means that members of the local elites may have become less interested in local affairs (discussion in Haeussler 2013a: 43–46).
In these new urban centres—many of them Roman creations—who is providing an example of a behavioural model? The elites are hardly setting an example for everyone: for instance, we cannot expect sub-elite people to walk around in a toga or indulge in luxuries. No-one? If you could, you might try to become a member of one of those many trade and transport collegia (‘guilds’), like the centonarii for traders and manufacturers of textiles (Liu 2009). This was important for your social status in society, but also an important economic benefit (see e.g. Verboven 2011). But this also provided peer pressure: whenever you join a ‘club’, you need to abide by its rules how to dress up, how to behave, which deities to worship within the collegium, and so on. And a collegium also needed a patronus or patrona—an influential member of the local elite—who could represent their interests, promote business, and provide financial endowments (Haeussler 2013a: 44, 171; Liu 2009). And despite the need for an imperial ‘authorisation’ as utilitas publico, it is important to emphasise the regional diversity among these collegia (Liu 2009).
Creolage: Individualisation vs. Collective ‘Identities’
This of course leads us to the question of how collective, local, or regional ‘identities’ are developing? Some of them may have been based on pre-Roman cultural understandings or boundaries. But we are mainly dealing with a new development—our ‘creolage’—where individuals were responsible for any form of innovation, some of which were taken up and became part of local or regional cultural expressions. We do not only have to think about the constraints and pressures that might have restricted the individual’s choices—or room for manoeuvre—but also on the impact one decision can have on a collective, be it a collegium, vicus, or civitas. How people interact within a group or a network is important if we aim for a bottom-up approach to cultural change and social cohesion.
To understand the interaction-based bottom-up approach to group identities, we can make use of sociological research that analyses how collective identities are developing, especially out of diverse groups of people. Most of our Roman communities and collectives must have had a rather heterogeneous membership. In this respect, research on modern social movements can provide food for thought. Timothy Gongaware, for example, discusses the impact variations in membership can have on a ‘group’, for example by adding or eliminating networks and strategies; there is also the interaction with other groups that ‘may influence … collective action frames’ (Gongaware 2012: 7). Apart from the impact that the environment and globalisation may have on collective identities, we need to consider inner-group interactions, opportunities and constraints provided within a group (Melucci 1985), as well as emotions that may impact/constrain people’s decision making; collective memory—or perceived collective memory—can be an important factor in sustaining the status quo or in stimulating change (Gongaware 2012). Across the Roman world, collective memory can be an important factor in motivating people’s actions: do the conquered peoples forget their own collective memory (Woolf 1996)? Are certain places becoming ‘memory sites’, like abandoned oppida and rural cult places (see e.g., Haeussler 2020; on cultural memory in the Roman world more generally, see also papers in Galinsky and Lapatin 2016)? As always, memory is not static and will be developed within the various social groups.
Why are these modern studies important for understanding the Roman world? They help to understand the complexity of individuals working together in a group. For instance, one often talks about the local elite. But one seemingly ignores that the members of the local ordo were a highly heterogeneous group of people with disparate views on culture, politics, religion, and history, partly due to their different economic interests, their different origins, and statuses. The same is true for any other group in the Roman world, like the countless collegia. Hence it is important to keep in mind how collectives are operating. Creating a collective identity is important (and the Romans did even impose a ‘dress code’ in their colonies, at least for some groups, like the decurions and priests, as in the lex de colonia Genetivae). If one joins an existing group, it means following certain rules, adopting certain symbols, engaging in collective activities. This is important for our study since any form of collective—or collective identity—is still based on each individual being able to make his/her decision, responding to group/peer pressure.
It is more difficult to understand how emerging groups were developing their distinctive behavioural identity. This leads us to the concept of ‘identity work’ that explores how ‘multiple individuals can come to share the feeling and cognitions that make up collective identity’ (Gongaware 2012: 8; see Schwalbe and Mason-Schrock 1996; Snow and McAdam 2000). By interacting with each other, people create a dynamic process that affects both the collective identity as well as their personal identities. As a dynamic progress, modern studies suggest that collective groups can also develop sub-groups, like factionalism (Snow and McAdam 2000). Therefore, we must be careful with traditional top-down models for Roman societies: though a highly hierarchical society, the various institutions—ordo, collegium, priestly colleges, and so on—all have their own internal dynamic resulting from individual’s actions and diverging views; how can one maintain a collective identity over several generations? How to avoid dissonance or factionalism? Whether individual, collegium, vicus, or civitas, their association with a powerful patronus/patrona will change how people make decisions: for instance, if your group profits from close links with Rome, then many individuals may take up new career choices, e.g. by joining the praetorian guards (Haeussler 2013a: 219–220).
We do not want to expand any further on the theme of the ‘identity dilemma’ (in the words of McGarry and Jasper 2015), but some of the above-cited sociological studies provide food for thought for understanding the context in which people act and interact. One may argue that modern society is of course so much more complex. Yet, the problems we want to address with ‘creolage’ in Roman times do not lack complexity. Not only does any individual have multiple identities in any one community, depending on his/her age, gender, wealth, status, etc., but more importantly, our sample groups are not restricted to one well-defined collective; we are dealing with any individual interacting with many different groups and other individuals, and the interactions go beyond the local community or civitas. Based on the available epigraphic evidence, let us take the example of a carpenter or stone mason who works for the rich and wealthy in his community, for the sub-elites who can afford his services, and sometimes for public buildings (e.g., temples or theatres, or as required by the Roman army); he/she employs people from all strata of society, has probably some slaves, buys products on the local market and has contacts with suppliers elsewhere in the region and beyond, that supply wood or stone; he/she participates in collective events (sacrifices, theatre, public baths) and may endeavour to join a local collegium to enhance his/her social status and economic network; his/her family might have come from another province, perhaps his/her father received Roman citizenship after serving in the auxiliaries, or perhaps his/her parents were liberti? And what future did he/she imagine for his/her own children? What educational decisions did parents take, regarding language, schooling, or even just names for their children: children often have Roman personal names while the parents had indigenous ones: were Latin names just a fashion or did parents consciously intend to make their children’s lives easier, to avoid discrimination?
Individual, Iconography, and Deities
The religious sphere provides a wealth of evidence to study individual actions, diverse choices, and group dynamics. It is beyond the scope of this paper to discuss this in detail. Let it suffice to say that polytheistic, non-dogmatic cults/religions, often tied to a particular social group as well as to a specific place, were facing challenges in the increasingly ‘global’ world. Our focus is on the individual’s capacity to innovate and trigger change. The interconnected Roman world can be experienced by an individual, for instance, by engaging with Graeco-Roman myths, reading philosophical works, attending theatrical performances, participating in the ‘imperial cult’, meeting people from other ‘faiths’ in your community or group (including military units) as well as travelling to other regions and provinces. People decided to go to a renowned healing sanctuary, attend famous festivals and processions, ask an oracle, become initiated in a mystery cult, ascent famous mountains, like Mount Kasios, or go to one of the famous academies in Athens or Alexandria. The level of communication and mobility which the individual was engaging with resulted in an emerging ‘global’ narrative and ‘global’ sacred landscape which in turn causes individuals—acting alone or within certain groups—to consciously or unconsciously alter local religious and mythical narratives as well as cult practices (see Chiai et al. 2012; Haeussler and King 2017).
We must jettison the idea of a homogenous ‘local elite’ or priestly ‘caste’ who makes these decisions: all these groups were extremely heterogeneous with individuals increasingly pursuing their personal interests. At many sacred sites, we can identify ‘visitors’ of diverse status, origin, and experience who were instigating acts that may have a variety of outcomes, starting with the introduction of new media of religious communication, votive offerings, sacrifices, writing, and sculptural representations, to the organisation of performative events, like festivals and processions. Whether the first person to set up an inscribed altar to a local deity is a local inhabitant, a newcomer, or passer-by, (s)he had to make choices based on personal religious and cultural understandings, education, and local pressure: is one allowed to set up an altar; does one keep the indigenous theonym or use a Graeco-Roman ‘equivalent’ on the dedication; which formulae and iconography to use (see Haeussler 2012)? We frequently see examples of code-switching (see also Wallace-Hadrill 2008: 23): in a text or in art, one switches from Latin to indigenous words, theonyms or iconography, and back again. While votive altars seem so commonplace in the Roman period, there are countless sites where writing never caught on or remained extremely basic (see e.g. sites in Roman Provence: Haeussler 2011).
The same is true when setting up an anthropomorphic stone statue for a local cult: does one buy a Classical statue off the shelf or create something unique? Let us look at a more high-status pick-and-mix ‘bricolage’. A stone mason, his client, and probably others in the local community in the civitas of the Vocontii decided to create a monumental, larger-than-life (206 cm) stone statue of their god of thunder, but different from the highly evolved sculptures from Iron Age southern Gaul (Figure 3; Esp. I 303; Carré 1978). The god’s common symbol across Southern Gaul—the wheel—figures prominently on the new statue (75 cm). The human male, who holds it in his right hand, may be identified as Jupiter by the eagle to his left. The combination between Jupiter and the local wheel god is so well attested in the region that his ‘native’ name has not been preserved (apart from the name ταρανοου ‘thunderer’ on a Gallo-Greek cippus from Orgon – RIG G-27). But the representation of Jupiter is rather uncommon as his military dress resembles more Mars Ultor or a Roman emperor. In addition, there is a serpent emerging from the trunk of an oak tree to his left that is not part of Jupiter’s common iconographic repertoire. Nearby at Vaison-la-Romaine, the sculpture of a human-serpent hybrid, the ‘monstre anguipède’ (Esp. XII 7998; CAG 84/1, 169*: 177), may provide a link to the Jupiter-Giant depictions of Eastern Gaul: if we are dealing with a similar concept, the social actors involved in creating the Séguret statue made different decisions to represent their god of thunder, but also decisions that did not become popular (was the site too remote and not well connected?).
Avoiding problematic terms like ‘appropriation’, ‘cultural resistance’, ‘hybridisation’, or ‘vulgar Romanisation’, we see a creative process by which individuals conceived a unique sculptural representation reflecting the zeitgeist (v. infra). Among the many altars combining Jupiter and wheel in southern Gaul, the one from Psalmodi (Saint-Laurent-d’Aigouze) is unique: wheel and thunderbolt were combined with the attributes of the (chthonic) mallet god: hammer and olla. The dedicant, Gaius Octavius Pedo, decided to employ the Latin theonyms, Jupiter and Silvanus, although the Celtic theonym Sucellos is attested at nearby Nîmes. But what is quite exceptional is the unprecedented, ‘fluid’ arrangement of the attributes on the altar (Figure 4; Esp. IX 6849; CAG 30/3, no 276, 2*: 635).
Introducing a new sculptural representation—be it by a wealthy donor, by decision of the local ordo, or by sub-elite persons—does not mean it will be accepted by the majority of the local devotees: but at sites with a higher frequentation, group dynamics may set in motion developments that—perhaps after some generations—will lead to people internalising new cult understandings. The natural site of a spring was the religious focus of the Iron Age ‘oppidum’ Nîmes/Nemausus before the emperor Augustus ‘appropriated’ it and transformed it into, as some people might say, an ‘Augusteum’. While it provides a venue for the rich and wealthy, like the decurions, seviri, and flamines to set up honorific statues, it can have hardly been envisaged to turn it into a hub where individuals worship countless deities, many of which are ‘non-Roman’ (Nemausus, Matres, Proxumae, Sucellos, etc. – Haeussler 2011: 404–411). Wealthy individuals sponsored certain cults: interesting is the anonymous philanthrope who consolidated the worship of Isis and Serapis by giving each of Nîmes’s decurions five sesterces to ‘dine in public’, while also backing Diana, Mars, Vesta and Somnus (CIL XII 3058).
Let us look at the agency of a sub-elite individual, the primus pilus Gaius Iulius Tiberinus: Roman citizen, legionary, centurion, perfectly Latin name, and a native of Beirut (domo Beryto). He set up an impressive marble altar to the local god Nemausus, closely associated with Nîmes’s sacred spring, and the only known epigraphic dedication in Narbonensis to the omnipotent Near Eastern Jupiter Heliopolitanus (i.e., Ba’al) (Figure 5; CIL XII 3072; Turcan 1972: 109).
What motivated Tiberinus to make such a unique dedication?
- Being a native of Beirut, a major cult centre for Baal and close to Baalbek, must have motivated his dedication to Jupiter Heliopolitanus, a deity that was becoming increasingly popular in the Roman West (Hajjar 1977; Bellelli 1997). For Kropp (2010: 230), this may reflect an ‘advertisements for local cults driven by pride, patriotism and local identities’.
- It seems likely that he advised the stone mason on the correct iconography: Jupiter Heliopolitanus wears the typical ependytes (imitating a military cuirass: see Kropp 2010), holding a whip in his right hand and a bunch of herbs in his left; there are traces of two busts of Helios and Selene on his chest; the statue is surrounded by two bulls (Hajjar 1977: 355–356). These features can also be found in Baalbek and Beirut, though Jupiter Heliopolitanus’ depictions were not canonised and the Nîmes relief is different from, for example, Marseille (Hajjar 1977: 349–352; Kropp 2010: Figure 4).
- Even if Tiberinus was motivated by ‘pride’ or ‘patriotism’, his choice of setting up the altar at Nîmes’s sacred spring needs an explanation: was he inspired by the spring’s sacredness and/or its role for elite self-display? Other sub-elite people, too, used this site to advertise their social status, like Titus Flavius Hermes (probably an ex-slave), exactor operis basilicae marmorarii et lapidarii (‘foreman of the basilica works, marble and stonecutter’), who made the only other dedication to Jupiter and Nemausus (CIL XII 3070).
- Why did Tiberinus combine Jupiter Heliopolitanus and Nemausus? Was it a public statement of integration by combining his ancestral home deity with the god of his chosen home, Nemausus? Perhaps he saw common ground between them: the ancestral god of Nîmes’s sacred spring and his ancestral god Ba’al, both related to springs and fertility. Also, Heliopolitanus’ sacred animals, the bulls, are emblem animals for Nîmes’s region, renowned for cattle breeding.
- Tiberinus’ most interesting decision is the design of a unique iconography for Nemausus, consisting, not of an anthropomorphic deity, but of two military items: carnyx and a Celtic oval shield. The ritual function of carnyxes can be seen in Tintignac in Aquitania where they were ritually deposited at the end of the Iron Age (Maniquet et al. 2008) and on the famous Gundestrup Cauldron as part of a ritual procession that involved a large cauldron into which humans were inserted by a male deity. But any connection with the cult of Nemausus seems speculative; the survival of Iron Age knowledge into the Principate in one of Gaul’s main Roman veteran colonies seems doubtful. Perhaps the stonecutter copied the iconography from the Arch of Orange. The iconographic choice may have intended to show Nemausus’ ‘ancestry’ which would insinuate that some people still identified with ‘Gallic’ ancestors.
Tiberinus’ case also demonstrates the importance of individuals in establishing personal links—here between Nîmes and Beirut and their cults—that by-pass Rome and Roman-style hierarchies. The presence of Jupiter Heliopolitanus in a public space inspires the Nîmois to engage with ‘foreign’ deities from ‘exotic’ places, making them aware of the ‘global’ world they were living in. The ‘spread’ of many cults in the Roman world was primarily due to individual social agents, not institutions.
Elite and Sub-Elite Discourses, and Collective Cult Understandings
Let us explore another scenario where we can analyse how the individual’s decision and individuals interacting with each other affected religious understandings. At some sites, people were engaging in different discourses, like the rural population that practised age-old rituals in contrast to people who were engaging in a more ‘elite discourse’, informed by the knowledge of Greco-Roman mythology and cult practises in a Roman colonia. It is therefore often no surprise to find seemingly incompatible finds in one site.
There are lots of possible case studies. Due to the wealth of material and textual evidence from the second century BC to the late fourth century AD, the sanctuary of Uley is interesting for a study in the longue durée (Woodward and Leach 1993). Judging from the archaeological evidence, there was no rupture after the Roman conquest. Among others, the nature of the existing animal sacrifices continued throughout the centuries, with only minor changes, as shown by Anthony King (2005: 233–234): among c. 230,000 animal bones, sheep and goat bones remained relatively stable for centuries and continued to be predominantly killed in autumn, insinuating major ritual activities at this time; cattle bones were gradually decreasing, like in other sanctuaries, either due to economic reasons or a movement away from collective sacrifices to more individual offerings.
Despite demographic changes, most people in the sanctuary’s catchment area continued to worship this deity; any newcomers seem to have followed established practises. This is surprising since the Colonia Nervia Glevensium (Gloucester), just 15 miles away, was founded by emperor Nerva c. AD 96–98 on the site of an earlier Roman fort. In this new context, one can envisage contradictory motivational factors: continue with the established rituals and practises, or adapt them to suit the new zeitgeist of the Principate? A dilemma for the local social agent?
Again, it takes us back to human interaction and their engagement with different discourses. Many of Gloucester’s original colonists—veteran soldiers—may have participated in and advocated a more imperial (‘Roman’) discourse. But cultural change does not operate top-down, but in multiple directions (see also Christensen 2019 on ‘great and little tradition’). In other words, our new settlers—and the people who move to the new urban hub—also engage with the local people and their cultural, religious and linguistic understandings, thereby creating—in the long run—new ‘traditions’ that are typical for Roman Gloucester and its territorium. To put it simply: local discourses can influence regional ones and subaltern discourses elite ones. We can see this throughout the Roman world: the Celtic goddess Epona, for example, can be found across the Roman West, including Rome; this was no evocatio or senatorial decision, as with Juno and Magna Mater, but we can envisage a bottom-up movement by which individuals, for example traders and horsemen, adopted Epona and spread their perception of her cult across wide distances. And by worshipping ‘indigenous’ deities, like emperor Diocletian worshipping the ‘Celtic’ Belenos at Aquileia, these deities become part of a global narrative. Since there was no ‘right belief’ or ‘faith’ in ancient religions, it is natural to have awe and respect for deities and divine forces of a topographical feature, and hence individuals engage in a ‘dialogue’ on the nature of the divine, which can result in misunderstandings and re-interpretations (see Haeussler 2012).
Despite the continuity of cult practices at Uley, there were also changes. Soon after Glevum’s foundation, the sanctuary was rebuilt in stone (Woodward and Leach 1993), though not as a Roman podium temple, but as a Romano-Celtic temple, which is worth noticing, among others, since the construction of these rather standardised Umgangstempel across western provinces reflect networks and an exchange of ideas that again completely by-pass Rome and Italy. At Uley, the deity’s name and sculptural representation acquired Graeco-Roman characteristics, notably a Classical Mercury/Hermes sculpture. Considering that most of our veteran soldiers at that time probably originated from the Germanic, Gaulish, and Iberian provinces, it is possible that Mercury was more a ‘Celtic’ interpretatio than a Roman one. Were Glevum’s founding fathers consciously re-shaping local cults, based on their own experience? Did certain decuriones, local landowners, and other visitors to Uley, who participated in a wider ‘Roman’ discourse, perhaps with a ‘liberal education’ in Greek and Latin, intend to make their imprint on the local sanctuary, either out of devotion or to stamp their personal religious authority and patronage over the rural population? Is it all down to one individual setting up a statue to Mercury and promoting the cult, or was the local ordo or a priestly college consulted? Were change and adjustment a question of survival of the sanctuary?
Whatever may have been the motivation, individual acts triggered new forms of interaction and engagement in the cult place. This does not mean that sub-elite worshippers instantly accepted this. An anthropomorphic sculpture may have alienated them. Again, it is a question of experience and interaction: by participating in the ancestral cult, but in a physically different environment, in the long run, as the textual evidence shows, local people started to evoke only the god Mercury (unlike the nearby Nodens sanctuary at Lydney or Sulis in Bath).
We also seem to witness changes in religious practices. Among others, worshippers no longer deposited their offerings in ditches and pits as they had done prior to the construction of the Romano-Celtic temple. Moreover, chicken became important for animal sacrifice after AD 100 (King 2005: 333). The suggestion that the ram and cock were Mercury’s companion animals (equally depicted on bronze figurines and on a limestone altar from Uley – CSIR-GB I.7) does not mean that one was sacrificing the god’s companions: no-one ever sacrificed eagles to Jupiter or owls to Minerva. At least one object from Uley represents a seemingly ‘indigenous’ type of deity, a ‘horned god’ with a masque-like head (dare I say, Cernunnos); it is a trapezoid fitting, perhaps from a bucket or cauldron (Martin Henig, in Woodward and Leach 1993: 98–99, Figure 85.2; King 2005: 334); a complementary fitting represents a rather ‘primitive’ depiction of Sol (Henig, ibid. Figure 85.1). It would be fantastic if this ‘bucket’ were an object used in regular ritual activities at Uley, but it may just have been an individual’s donation to the god and the connection between the horned god and Mercury is therefore speculative.
As we have tried to show, one was theoretically free to make any decision, but one’s choices were constrained. If you were participating at a large collective religious event at Uley—or another sanctuary—you were unlikely to break existing traditions as this might upset the validity of a collective sacrifice or offering. Although social pressure may have forced you to attend and participate at major events in the local religious calendar, you were of course free in your interpretation of the cult: there was no dogma. But how would you behave if you came to Uley on your own on a quiet day to make a vow to the deity? Would you just follow the practices you experienced in other religious sites? Or was there anything specific you needed to observe? Did you need to check whether you are allowed inside the temple? You were probably told whether you had to cover your head (like the Romans do) or uncover it (like Greeks and Egyptians). Perhaps there was a priest or temple attendant who told you to purify yourself before you approached the shrine; a local farmer might sell his chicken and lamb if you wish to do a sacrifice.
People from various backgrounds and statuses would visit Uley and each had their own religious understanding, their own experience of religious practises, whether it was a soldier or officer, a decurion, travelling merchant, local farmer, landowner, or craftsman, not to forget any Roman senators and equestrians visiting the province. Individually they might have made their personal decisions and dedications. But all these individuals were interacting with each other, either through their actions and their material remains and face-to-face at important festivals. This leads us to the psychology of group dynamics (or more condensed in Wiggins’  psychological oppositions of agency/passivity and communion/dissociation). At Uley, we deal with a kind of ad hoc group consisting of individuals and social groups coming together at religious festivals, like in autumn when sheep were sacrificed in larger numbers. Despite individuals potentially having discrepant ‘worldviews’—in the widest possible sense—the setting of a religious gathering can provide a sense of ‘communion’: a shared experience and sharing the sacrificial meat may help to strengthen human bonds while consolidating and re-producing social hierarchies. This form of multilateral interactions shapes and reshapes people’s religious understanding and behaviour.
It is in this context that we must place the introduction of cursing at Uley during the second century AD. How did this concept of curse tablets develop? Does it go back to pre-Roman understandings that are not preserved or did one copy the practices at Aquae Sulis, just 25 miles south? Was it a single act that inspired other people? And who wrote the first defixio? Perhaps it was an outsider from another province who visited the sanctuary or conceivably a priest who saw it as a new form of revenue. The curses’ subject matter predominantly indicates sub-elite, rural individuals, like Cenacus complaining to Mercury about his stolen draught animal.
Though the concept of a defixio came from mainland Europe (Brodersen and Kropp 2004), including Gaulish first-century AD defixiones from Larzac and Chamalières (Lejeune 1984), the British texts are different in content, layout, and formulae. Although each was written by a different hand (Tomlin 2008), they follow patterns, like using a quasi-legal language that echoes people’s participation in a Rome-dominated legal discourse. Unlike monumental inscriptions, the curses reflect personal communication with a deity and therefore more frequently attest local vernacular language and code-switching, including the use of Celtic words, like gabatas for ‘plates’, and Celtic personal names, like Atavacum, Biccus, Cenacus, or Senebellena (Tomlin 1993: Uley nos. 1 and 4; Hassall and Tomlin 1995: Uley no. 33; Douglas 2015).
Overall, we can see how the decisions of individuals, though framed by existing ‘structures’, can change existing behavioural patterns. As in the case of Tiberinus in Nîmes, we find wealthy donors and benefactors acting in sacred places across the Roman world. In Asia Minor, wealthy men and women used the imperial cult as a means of self-promotion, among other by introducing gladiatorial games—an unusual form of entertainment for the Greek East, one would think (see van Bremen 1996). This does not mean that local worshippers accepted that form of imposition or even understood its meaning. We must remember that across the Roman world, many cult places have not yielded any inscription or even an altar. This raises the question whether people made the conscious decision not to use writing, not to employ Roman-style altars, and instead continued with what they perceived to be ancestral traditions, like ritual depositions in ditches and shafts, as at the Folly Lane sanctuary (Verulamium, St Albans), first–third centuries AD (Niblett 1999). Therefore, seemingly out-of-box thinking may sometimes be so much more in line with an individual’s understandings and that of his/her social group. We can think of the sometimes seemingly ‘primitive’ artistic attempts—like representations of the deceased on the above-mentioned Canavese stelae or the representations of ‘warrior gods’ from Roman Britain, as in a ‘graffito’ discovered in a third-century pit at Tre Owen (Powys) including a basic Latin dedication (Figure 6; CSIR-GB I.5, no. 15; Haeussler 2012: 167–168, fig. 10). And coming back to Mercury, he also appears in unusual forms, like a face pot from Lincoln (Figure 7; RIB II.6: 2499.1).
Epilogue: Creolage, Interactivity, and the Complexities of Life
To talk about ‘Romanisation’, ‘discrepant experiences’, or ‘acculturation’ is not only a stark underestimation regarding the complexity of simultaneous, often contradictory developments that were taking place in the ancient world, but it also ignores the complexities of life itself. Every individual had its very own personal experiences. Since birth, the individual is accumulating numerous experiences and group memberships that shape his/her behavioural ‘repertoire’, starting with one’s family, household, gens, community, and ethnicity (village, city, tribe, pagus, civitas, etc.), one’s education, gender-specific groups, and religious group(s), to one’s work colleagues (e.g., amici subaediani – CIL X 6699), co-members in a collegium or societas, or contubernales and commilitones in the army, not to mention the membership of an ordo, priestly college, or even equestrian or senatorial order.
Multiple memberships obviously imply multiple, often permeable identities (see e.g. Rebillard 2015: 428–430). But for our study, it is important that each individual needs to ‘socialise’ in each group, including learning particular behavioural patterns, whilst expanding his/her personal network at the same time (Haeussler 2013a: 316–320). Our view of a network in this paper goes far beyond elite or patronage/matronage networks, but it is a complex, entangled almost infinite matrix of a person’s human contacts—both past and present: even lost contacts from the past can still influence a human’s actions in the present.
We also must not forget that many individuals were the victims of warfare, deportation, or slavery, others suffered discrimination or poverty, while only a small number of individuals were the profiteers in Roman societies, accumulating more and more wealth to the detriment of others. As a result, any social group would have been rather heterogeneous in character, developing its own group dynamics and theoretically unpredictable decisions, though some group members can be expected to be more passive than others.
As a result, an individual faces an extremely complex and often unpredictable matrix of connections, interactions, and dependencies, especially in an empire of that scale, size, and diversity. Often unknowingly, individuals were partaking in multilateral ‘discourses’ taking place at various levels in society and in their networks, but ‘local’ vs. ‘global ‘(or ‘little vs. big tradition’) is a stark oversimplification. The individual needs to find his/her place in society—and albeit motivated, pressurised, and hemmed by family and local traditions, economic needs, personal aspirations, or the imperial discourses about the empire’s opportunities and benefits, it is the social agent who makes the decisive decisions. Our sources demonstrate thousands of individual fates in the Roman world. And we must not ignore the various forms of group pressure the individual was facing, for example when buying or selling products at the local market, or participating in long-distance trade activities, or when joining the Roman legions, auxiliary units, praetorian guards, or the navy, when participating in collective events, from theatre performances to religious festivals, or when becoming a member of particular social and religious groups, like Greco-Oriental mystery cults or collegia. Your local community might have been quite small, perhaps only 10,000 or 15,000 inhabitants for many provincial cities; but this means that everybody knew each other, and everybody needed each other’s services, and the pressure to follow certain norms and values was extremely high.
And yet, there was change. People across the Roman world decided to break with ‘traditions’. They did not just ‘cobble together’ new art styles, deities or identities. Instead of ‘bricolage’ or ‘creolisation’, we see a highly individualised process that was embedded within local social, cultural and economic understandings. How can any woman, man or child survive in the complex and heterogeneous societies of the first to third centuries AD? Among others, code switching and cultural navigation must have been the norm. The same person might make different decisions depending on context: private or public? How to dress, which language to use in conversation or in writing, how to name your children, which deities to worship, and choice of material culture depending on context: at which occasion was someone going to use the fine (Samian) table ware and silver spoons? The list is endless. Even a wealthy villa owner might want to impress visitors with the latest ‘fashions’, like elaborate mosaics depicting Greco-Roman myths. At the same time, the same person might deal with more ‘local’ customs in public, like financing a temple or shrine in local style or worshipping deities with non-Roman epithets or theonyms. We must not fall into the same trap and focus too much on the ‘elites’ as the major protagonists. Elites are happy to preserve the status quo. They are responding to pressures from below, for example when their status and symbols of authority are challenged by people who acquired wealth and started emulating local elite culture, by locals who served under Roman command in other provinces, or by newcomers with different cultural concepts. People were moving across the Roman world, like the above-mentioned Tiberinus from Beirut who ended up in Nîmes (see also papers in Lo Cascio and Tacoma 2016 and PIR 1933–2015 for people’s mobility across the empire in the first to third centuries AD). As a result, we see people from all ways of life happy to switch between different cultural languages, often within the same medium, like in epigraphy and sculpture.
Creolage aims to account for regional diversity and diversity within one community. It focuses on individualisation in society and consequently the role of the social agents to shape their identities and expressions within certain ‘structures’. In the entangled world of the Roman empire, this puts less focus on Rome and more on the wider developments that shaped people’s understanding. Key words include zeitgeist and cognitive trends. With hindsight, we might be able to recognise the zeitgeist—the ‘spirit of the time’—in a particular Roman period and region. But what created the zeitgeist in the first place and how did it affect people’s lives? Monika Krause attempts to develop ‘zeitgeist as an analytical specified concept (…) for social and cultural analysis’ (2019: 1). One of her case studies is the spread of baroque art and architecture across geographic and ideological boundaries in Europe during the 17th and 18th centuries. This is not too dissimilar to the spread of Roman-style architecture, the sumptuous villa and domus with its mosaics, wall paintings and facilities. And in both cases, we are not just dealing with ‘high art’ and an architecture of power, but also an influence in popular culture. But as in our approach to creolage—focussing on the local context and the individual social agent’s decision making process—‘zeitgeist’ is also a ‘puzzle’ for modern sociology (Krause 2019, based on Garfinkel 1988’s approach that social order is ‘always and only locally achieved’). Whether baroque or Roman, one may use Max Weber’s concept of the Trägergruppe (‘carrier group’), but with Krause’s addition: not just people, but also ‘objects, media, and technologies’ are carriers of culture. These non-human carriers also ‘last longer than situations’ and can be ‘copied, used, and reinterpreted’: practises and carriers ‘add to macro-cultural phenomena’ (Krause 2019: 8).
This helps us to understand developments in Roman times, both on a macro-regional and a micro-cosmic level. We must face the challenge: cultures, objects, artefacts, and practices do not have meaning per se. They are contested and re-interpreted. Concepts, like ‘zeitgeist’, ‘tradition’, or ‘fashion’ all need to be individually explained out of individual social agents’ actions and decision. They create—and maintain—the ‘zeitgeist’ at any period of time. And this ‘zeitgeist’ is clearly changing and developing, becoming more global, less collective, and less localised. This, in turn, gives room to manoeuvre for sub-elite people freed from certain social and economic constraints. Artefacts and architecture in Roman times are—as Krause suggests—long-lasting Trägergruppen that influence people long after their creators are gone, like the Roman tombstones, villas and temples that shape people’s decisions, while technological changes, like in pottery production, engineering, or glass blowing, were shaping people’s lives for generations to come. We also see the use—and adaptation and alienation—of certain visual languages and certain techniques that were employed to create localised works of art, visual expressions of identities and myths, and many more. As we have discussed above, the multilateral dialogues that take place, more or less simultaneously, at various levels are essential for our understanding of (societal, cultural, artistic, linguistic or religious) changes. Among others, there are the ‘dialogues’ (or disagreements) between local and supra-regional discourses, between elites and sub-elite groups, between local histories, myths, and traditions versus colonial or imperial discourses. All of these inform the actions of the individual social agent, whatever his status, gender, profession, or age. And because there is a sense of friction, this leads to a creative process, to new and unprecedented innovations. It is futile to categorise artwork or objects as ‘Roman’ or ‘native/pre-Roman’: we are dealing with a constant flux where every generation has, in a way, to re-invent its cultural repertoire. In this view, it is not ‘discrepancy’ that needs to be explained, but the replication of local cultural expressions from one generation to the next. Explaining sociocultural developments present us with a complex problem and there is no simple answer when it comes to understanding these mechanisms, and no answer that can ever fit all encounters in the Roman world. A bottom-up approach, acknowledging the multiple experiences, expectations, norms, and social pressures that shaped the individual social agents in diverse geographical settings in the Roman empire, helps to understand the ‘mechanisms of agency’ across the Roman world.