New materialism: towards a new understanding of the Roman world?
In a 2014 discussion article published in Archaeological Dialogues, Miguel John Versluys proposed a new understanding of the much-debated concept of ‘Romanisation’ as one of ‘objects in motion’. Alongside a strong focus on ‘globalisation’ theory, he urged Roman archaeologists to recognise the heuristic value of what he calls “diasporas of material culture” (Versluys Reference Versluys2014: 15), and to follow the ‘object turn’ or ‘material-cultural turn’, making “material culture, with its stylistic and material properties (and thus agency […]), central to our understanding of the Roman world” (Versluys Reference Versluys2014: 16). Although some aspects of his manifesto have stimulated extensive discussion, his argument for object agency has remained largely unchallenged. Yet, alongside advocacy of ‘globalisation’ theory (Pitts & Versluys Reference Pitts and Versluys2015), such object agency represents the most contentious aspect of his revision of ‘Romanisation’. Based on the ‘material turn’, Versluys (Reference Versluys, van Oyen and Pitts2017: 192) invites us “to rewrite history as a particular relationship between objects and people with things as the agents provocateurs of (historical) change”.
Our issue with this object-agency approach is that it represents a soft-culturalist perspective that offers an unbalanced view of the workings of imperialism, marginalising hard power, violence and extreme social hierarchies. Despite some recent efforts to promote this approach further (van Oyen Reference van Oyen, Lichtenberger and Raja2017), the implications of adopting this type of ‘new materialism’ in the study of Roman expansionism is best assessed by examining other disciplines, such as art history, anthropology and visual culture studies, which have opened themselves to this particular strand of post-humanism over the past 20 years (Green Reference Green2012; Gamble et al. Reference Gamble, Hanan and Nail2019).
It is not the purpose of this debate piece to present a comprehensive overview of the diverse, and sometimes partly entwined, approaches encompassed under the ‘umbrella’ term of post-humanism (Ferrando Reference Ferrando2013). Within archaeology, they include, but are not limited to, new materialisms, symmetrical approaches and the ontological turn (cf. Fernández-Götz et al. Reference Fernández-Götz, Gardner and Díaz de Liañoin press). Some of the critiques offered by these trends have rightly identified shortcomings in traditional approaches, while also offering new directions for reflection. Nevertheless, the ways in which some scholars have applied these novel perspectives is problematic. First, the concept of ‘flat ontologies’, as advanced by proponents of symmetrical archaeology (e.g. Witmore Reference Witmore2014) under the inspiration of actor-network theory (Latour Reference Latour2005), often fails to identify inequality, power differences and causal relationships effectively (Preucel Reference Preucel2012; Hodder Reference Hodder2014). This can limit archaeology's capacity for social critique, a weakness that is particularly evident when trying to understand strongly hierarchical state formations, such as the Roman Empire.
Second, the emphasis of some post-humanist views—particularly those advocating object-oriented ontologies—on ‘things-in-themselves’ (Olsen Reference Olsen2010) has been criticised on philosophical grounds (Nielsen Reference Nielsen2019) and for leading to a new form of object fetishism or antiquarianism (Barrett Reference Barrett2016). From our perspective, archaeology's ultimate aim should not be the study of things, but of people through things.
Finally, the expansion of the concept of agency to objects without establishing a differentiation between human and non-human forms of agency constitutes a further problem. Following Robb (Reference Robb2015), we could consider that things have a type of agency in the sense that they ‘act back’ on people, but this would be different from human agency. Although things undoubtedly ‘shape the mind’ (Malafouris Reference Malafouris2013) and actively construct people's identities and their ‘being in the world’, this recognition of the importance of objects is not the same as granting them that power without the human component (Van Dyke Reference Van Dyke, Acuto and Franco2015; Ribeiro Reference Ribeiro2019). As an alternative, we could consider Stockhammer's (Reference Stockhammer, Abu-Er-Rub, Brosius, Meurer, Panagiotopoulos and Richter2019) concept of the ‘effectancy of things’, as this acknowledges the effects that objects have on people (rightly recognised by Gosden in his Reference Gosden2005 study of Britain's incorporation into the Roman Empire), but serves as a counter-notion to human agency while simultaneously avoiding the risk of anthropomorphising objects.
Where are the ethics?
Our main concern with some new materialist approaches resides in how they ascribe agency to objects, transforming them into historical players in their own right to the detriment of the study of human agency and social structure. In the context of Roman archaeology, the new materialist approach is exemplified by, among others, Woolf (Reference Woolf, van Oyen and Pitts2017: 216):
As we move beyond representation and instrumentality we take a great weight of responsibility off the shoulders of Roman actors. Humans are no longer the sole drivers of Roman success and Roman failure […] Neither expansion nor collapse is to be explained primarily in terms of human values, ideologies, beliefs or motivations […] Taking things seriously allows us to put people in perspective.
This and other similar perspectives risk detracting from the darkest aspects of social life in Rome and other imperial powers, such as enslavement, mass violence and sharp inequalities, thus emphasising what González-Ruibal (Reference González-Ruibal2019) labels the ‘soft politics’ or ‘political agnosticism’ of many new materialist approaches (see also Gardner Reference Gardner, Millett, Revell and Moore2016). By shifting attention from people to objects, views based on the agency of Roman things might contribute to the production of a ‘sanitised past’, which has implications for the present (Rekret Reference Rekret2016). Blaming objects for the sins and bad intentions of the people who create and use them could, at one level, serve as a witty thought-experiment about the daggers used in the assassination of Julius Caesar, but it becomes a serious issue if we adopt the same ‘agency of things’ perspective when considering the weapons used to kill millions of people during the Roman conquests. While some might argue that making these considerations about events that took place 2000 years ago constitutes a harmless intellectual exercise, the implications of some new materialist approaches become more evident when the same logic is applied to more recent events. Thus, the same perspective could be used in defence of international stockbrokers during the financial crash of 2008, arguing that they were simply overwhelmed by complex algorithms rather than by the rapacious mindset of neoliberal capitalism. Or, in its more extreme form, it could be used to take responsibility away from states and individuals in cases of modern wars and crimes against humanity, such as the Yugoslav Wars, the Rwanda genocide or the Iraq invasion. Recognising the active role of material culture in social life should not lead us to underestimate either human responsibility or suffering.
When analysing Roman expansionism, the above reflections are about much more than the material culture of a bygone empire. Focusing on the ‘agency of things’ instead of on human actions and social structures results in an opaque narrative, which emphasises the metahistorical qualities of objects over their value as historical sources. De-historicising means de-humanising, and, more crucially, it means moving away from the possibility of developing empathy through our understanding of the past (Snyder Reference Snyder2017). Thus, both Versluys's (Reference Versluys2014: 19) statement that “Romanization is about understanding objects in motion” and van Oyen's (Reference van Oyen, Lichtenberger and Raja2017: 287) proposal that “the historical issue at the roots of the Romanization debate revolves around a particular patterning of material culture” risk obscuring or forgetting the human stories behind the process of Roman expansion, including the suffering caused by military actions. These approaches can lead to the aggressive acts of conquest that caused the death or loss of liberty of millions of people being understated or even ignored. Appian's (Roman History 4. The Celtic book 1.2; McGing Reference McGing2019) claim that Caesar killed one million Gauls and enslaved another million, out of a total population of four million, illustrates the point. Although these figures might be exaggerated, if we assume only a half or one-third of these totals, the impact in terms of the percentage of overall population would make such actions comparable to some of the cruellest episodes in human history, such as the First and Second World Wars.
A predatory political economy
As a (partly complementary) alternative to previous models that aim to analyse Roman expansionism, we introduce the concept of the ‘predatory regime’ to define the political economy of Late Republican and Early Imperial Rome (second and first centuries BC and first century AD). Our use of the term is based on the application by González-Ruibal (Reference González-Ruibal, Leone and Knauf2015: 424) who, following Mbembe (Reference Mbembe2001), describes predatory regimes as being “characterised by the militarization of power and trade, pillage as an economic strategy, the pursuit of private interest under public command and the conversion of brute violence into legitimate authority”. Rome's foreign policy was largely driven by the personal interest of elite factions and individuals who used warfare and the extraction of external resources (both human, i.e. slaves, and non-human, e.g. minerals, grain, textiles) as a means of increasing their wealth and prestige. The period of the Late Republic was marked by social and political violence, both internalised in the form of civil war and externalised by conquering new territories (Barrandon Reference Barrandon2018; Maschek Reference Maschek2018; Lange & Vervaet Reference Lange and Vervaet2019). Rather than pursuing a state-driven grand strategy, the process of conquest can often be characterised as the pillage of foreign lands for personal gain. In this model, state gain was frequently just a secondary outcome of individual and familial agendas. While Caesar's Gallic War provides a clear example, there are also cases of longer campaigns of military engagement and exploitation, such as on the Iberian Peninsula.
The characteristics of a predatory regime can indeed be observed in second- to first-century BC Hispania, where Roman actions encompassed the range of activities listed by González-Ruibal (Reference González-Ruibal, Leone and Knauf2015: 425):
Predatory activities can be destructive (pillage, slave raids, total war) or productive (trade, mining, plantations). In the latter case, however, they always imply a systematic exploitation of resources beyond the threshold of social or natural sustainability.
It should be noted, however, that the predatory model did not have the same intensity in all the territories occupied by Rome, nor over the entire duration of the Roman State. The model of violent military conquest and large-scale plundering applies to some territories, such as large parts of Gaul, Iberia and Dacia, but not to others, such as Noricum, Cyrenaica or Cappadocia. Moreover, while neither ‘Romans’ nor ‘natives’ were homogeneous groups, applying binary terms like conqueror/conquered, Roman/native or domination/resistance can be useful for some instances of military conquest and for the decades afterwards, although such application loses its significance in later times.
By conceptualising the expanding Roman State as a predatory system, we aim to contribute to decolonising Roman studies by paying more attention to the ‘dark sides’ of imperialism. We argue that instead of moving the focus from humans to things, we should concentrate more on the human component, and place subaltern groups at the forefront. This involves making visible those ‘left behind’ by the process of conquest, as well as drawing attention to the mechanisms of imperial domination. Our theoretical perspective resonates with the work of Latin American scholars such as Mignolo (Reference Mignolo2011), who applies decolonial thought to highlight the dark sides of modernity. Materiality plays a key role in predation, from military equipment and infrastructure to landscape transformation and the circulation of goods through raids or trade (González-Ruibal Reference González-Ruibal, Leone and Knauf2015). Changes in the material world make it possible to carry out warfare and exploitation in new ways, but not in the sense of an extreme version of the ‘agency of things’ that takes agency and responsibility away from humans; rather, we see such agency as part of a bidirectional process in which people create material culture while simultaneously being constructed by it.
The dark side of the Empire: two case studies
Two examples of themes in which archaeology is providing new insights into the ‘dark sides’ of Roman state power serve to illustrate our stance.
Beyond limes archaeology: new perspectives on the Roman conquest
Archaeological research of the past few decades has revolutionised our knowledge of the Roman conquest of Western and Central Europe, with new investigations sometimes confirming information from written sources, and others challenging official accounts or uncovering conflict where it was previously unknown (Fernández-Götz & Roymans Reference Fernández-Götz and Roymans2018; Fitzpatrick & Haselgrove Reference Fitzpatrick and Haselgrove2019). In addition to the ever-expanding corpus of data, there is also a qualitative difference in the way a growing number of scholars approach the Roman military. Traditional studies focused predominantly on so-called limes (frontier) archaeology, that is, the Roman military infrastructure in the provinces during the post-conquest period. While this approach remains valuable (Breeze Reference Breeze2018), there is increasing interest in studying the moment of conquest itself, as well as its immediate aftermath. Within the dominant paradigm of limes archaeology, the Roman army is viewed as the defender of provincial peace and civilian life against external enemies. By contrast, our complementary approach analyses the role of the Roman military as aggressor during the expansion, including as perpetrators of episodes of mass violence, enslavement and sometimes genocide (Roymans & Fernández-Götz Reference Roymans and Fernández-Götz2019).
Archaeology can make a fundamental contribution by investigating the spatial dimension and the direct social impact of the conquests on the societies that they affected. The increasing use of techniques such as lidar, geophysics and systematic metal-detector surveys is identifying numerous Roman marching camps and evidence for sieges of indigenous hillforts and battles, while the detailed study of settlement patterns allows for assessment of the demographic impact of the conquest in test regions.
Northern Gaul and northern Iberia are two regions where archaeological studies have substantially changed previous narratives (Camino et al. Reference Camino, Peralta and Torres-Martínez2015; Peralta et al. Reference Peralta, Camino and Torres-Martínez2019; Roymans Reference Roymans2019). Some two decades ago, Caesar's actions on the northern frontier (58–51 BC) and Augustus’ Cantabrian Wars (29–19 BC) were almost untraceable in the archaeological record. This led many scholars, both archaeologists and ancient historians, to conclude that the conquest of those regions had limited societal impact, despite Classical sources referring to the cruelty of the campaigns and the fierce native resistance. Our understanding has now changed completely, with archaeology providing ample evidence of the massive scale of Rome's military engagement and the dramatic consequences for many local communities. The fortification of Thuin (Hainault) in northern Gaul, for example, has been identified as the oppidum of the Aduatuci, who were conquered in 57 BC. Following this event, the entire population of over 50 000 individuals were sold as slaves. The site of Kessel/Lith (Brabant) has been reinterpreted as the scene of the Roman massacre of tens of thousands of Tencteri and Usipetes in 55 BC, including women and children—an act classifiable as genocidal mass killing (Figure 1). On a larger geographic scale, analysis of settlement evidence suggests a significant population decrease in the territory of the Eburones that can be plausibly linked with Caesar's campaigns (Roymans Reference Roymans2019). In northern Iberia, archaeological work has provided spectacular evidence for the Roman attack and destruction of indigenous hillforts—such as La Loma and Monte Bernorio (both in the province of Palencia)—identified over 60 Roman camps and traced the routes followed by the legions in their advance into the Cantabrian Mountains (Camino et al. Reference Camino, Peralta and Torres-Martínez2015; Peralta et al. Reference Peralta, Camino and Torres-Martínez2019).
Human remains from a battle-related find complex dredged from the River Meuse at Kessel-Lith (the Netherlands), probably linked to Caesar's massacre of the Germanic Tencteri and Usipetes in 55 BC (after Roymans Reference Roymans2004: 128; photograph by M. Ydo).
The archaeology of rural communities as part of a hierarchical imperial system
Until recently, rural archaeology in the north-western provinces was dominated by the ‘Romanisation’ narrative, with its focus on the ‘Pax Romana’ and themes including the emergence of villa landscapes and the development of markets. This narrative reflects an ideal of ‘civilisation’, in which the Roman army is portrayed as the guardian of peace. The influence of post-colonialist thought and critical heritage, however, is creating more space for alternative narratives of dramatic rural transformations that underline the key role played by imperial agency and extreme social hierarchies. This is illustrated by three major transformations that had a profound impact on rural populations in Germania inferior (Roymans et al. Reference Roymans, Derks and Heeren2020).
First, a phase of the extremely violent Caesarian conquest of the Lower Rhine resulted in dramatic demographic losses for the indigenous population. Here, Caesar left behind landscapes of trauma and terror, and archaeology can study the impact of his campaigns through palaeodemographic research and by identifying major conflict sites (Roymans Reference Roymans2019).
Second, a post-conquest phase of fundamental reorganisation of the Lower Rhine is characterised by the large-scale immigration of ‘Germanic’ groups from the east bank of the Rhine. This process was largely initiated and managed by the imperial administration, and resulted in the ethnogenesis of new tribal formations, which often comprised a mixture of immigrant groups and surviving autochthonous populations (Roymans Reference Roymans2004). Archaeology has the potential to assess the historical model of migration and ethnogenesis through conventional material culture studies combined with isotope analysis. Rural populations in Germania inferior were intensively exploited as a breeding ground for auxiliary soldiers, following the imperial policy implemented in the Lower Rhine region, especially in the first century AD. The constant drain on recruits for the Roman army had profound social effects, including the rapid diffusion of elements of Roman military culture into rural communities.
Finally, the second half of the third century AD witnessed a further dramatic demographic decline among rural groups. The archaeological record indicates the almost complete depopulation of rural areas in the southern Netherlands and northern Belgium. While explanations for this collapse are still debated, the evidence suggests that imperial agency played a significant role. In this context, we can refer to written sources about the forced deportation of Lower Rhine groups to interior Gaul (Heeren Reference Heeren, Roymans, Derks and Hiddink2015).
Conclusions
Modern scholars should be careful not to sanitise the past, particularly when ancient sources and archaeology provide ample evidence for the darkest sides of imperialism (Figure 2). While processes of collaboration, integration and hybridisation undoubtedly existed, we cannot ignore the most aggressive side of Roman expansion when we attempt to produce more holistic histories. It is not our intention to disregard new materialism and the concept of ‘objectscapes’ in their entirety, as some applications certainly demonstrate valuable insights (e.g. Pitts Reference Pitts2019; some contributions in Selsvold & Webb Reference Selsvold and Webb2020). The predatory model is therefore not necessarily incompatible with new materialist perspectives, but it clearly places the focus on the exploitative aspects of imperial power. The centring of discussions on Roman expansionism around ‘objects in motion’ and the ‘agency of things’, on the other hand, risks missing or minimising crucial human aspects, such as aggression, resistance and suffering.
The destruction of a Germanic village by Roman troops during the Marcomannic Wars (AD 166–180); scene from the Column of Marcus Aurelius in Rome (image copyright of Alamy).
Imperial agency and asymmetrical power relations are largely omitted by many new materialist approaches, which therefore do not offer a balanced perspective on the functioning of empires, both ancient and modern. Some of the perspectives resemble neoliberal narratives that portray an idealised view of modern-day globalisation focused on the movement and consumption of goods and ideas, leaving little or no mention of the more negative sides associated with the exploitation of people and resources.
All empires have ‘bright’ and ‘dark’ sides that are fundamentally entwined. The bright aspects are often reflected in monumental public buildings, sumptuous elite residences and developed infrastructure. But the dark sides equally need to be taken into account. For the Roman Empire, the dominant narratives have traditionally emphasised the bright aspects, from the spread of literacy to villa landscapes, public architecture and high-quality tablewares. In order to develop a more inclusive and balanced account, however, we also need to acknowledge the more brutal sides, from mass enslavement and destruction to huge wealth disparities. Our application of a predatory model is therefore not aimed at demonising Roman expansionism, but to provide a new framework that contributes to a better understanding and allows comparisons with similar cases of expanding imperial powers throughout history. If we want to reclaim the ‘people without history’ (Wolf Reference Wolf1982) and give voice to the voiceless, their stories of suffering and oppression need to be made visible: the killed, the enslaved, the marginalised, the displaced, the oppressed. A holistic history should include winners and losers, and all those who cannot be easily assigned to one of these two poles and simply tried to adapt as best as they could to the changing world in which they lived.
Acknowledgements
This article has been produced with the support of the Philip Leverhulme Prize.
