Beyond Romans and Natives

In 1997 Greg Woolf published the article ‘Beyond Romans and natives’, in which he called for ‘a new view of the nature and genesis of Roman imperial culture’ (Woolf Reference Woolf1997, 339). He subsequently presented such a new approach and fleshed out the theoretical model with his 1998 landmark book Becoming Roman. Having digested the alternative perspectives on Rome exemplified by Millett (Reference Millett1990) – and thus writing history ‘from below’, while making extensive use of archaeological sources and focusing on ‘provincial civilization’ – there are, however, no ideological value judgements in his analysis of the process of transformation wherein both Romans and Natives participate. Roman and Native are, in his view, strictly relative categories to the extent that one could become Roman, something unimaginable in the context of 19th-century imperialism. Woolf views Roman power as only one out of many factors that explain cultural change, and, following earlier research by others, he therefore proposes the term ‘Roman cultural revolution’ (Woolf Reference Woolf, Keay and Terrenato2001) to properly describe the transformations taking place around the time of Octavian. This is not to say that differences between Roman and Native did not matter – they mattered a lot – but rather that we would be dealing, to a large extent, with constructed ethnic and cultural identities that were fluid and permeable. It remains indeed surprisingly difficult to answer the simple question ‘who were the Romans’ (Woolf Reference Woolf2012, 219; see below), but we are not likely to find answers in the right direction by constructing absolute dichotomies between Roman and Native. As Woolf summarized recently (Reference Woolf2012, 222), ‘Considerations like these mean we can never study “Roman societies” without including many who were not Romans. Yet if we treat all provincials as “in some sense” Romans, we obscure distinctions that mattered enormously at the time’.

David Mattingly (Reference Mattingly2006) begins his historical account of Roman Britain (Part I of the Penguin History of Britain series) as follows: ‘This book tells the story of the occupation of Britain by the Romans’. In the very first sentence the three main constituents of his narrative are immediately clear: there is Native Britain, there is Roman, and the relation between them is characterized by the word ‘occupation’; later on the same page he characterizes the period in question as ‘four centuries of foreign domination’; further on (ibid., 7) he concludes that ‘Britain in the Roman Empire was a colonized and exploited territory’. Mattingly is clear about his agenda for writing up the history of Britain in the Roman Empire as ‘an imperial possession’: ‘In this book, considerably more emphasis than usual will be placed on the negative aspects of imperial rule and their impact on the subject peoples’ (ibid., 12). His 2010 book Imperialism, power and identity. Experiencing the Roman Empire and an earlier article for the general public both summarize his reasons for pursuing that agenda: ‘Despite an increasingly critical treatment of the reputations of many modern empires (the British empire included), the consensus verdict on the Roman empire remains surprisingly favourable’ (Mattingly Reference Mattingly2007) – and Mattingly clearly wants to change what he considers to be the consensus.

Briefly summarizing and characterizing the work by two leading (Anglo-Saxon) scholars on Romanization illustrates, I hope, the difference in their approach: Woolf focuses on ‘cultural transformation’, while ‘imperialism’ is central to Mattingly's understanding of the Roman world. It is interesting to note that both are (originally) historians making extensive use of material culture while putting a historical concept at the centre of their general interpretation: for Woolf, Rome is about ‘empire’; for Mattingly it is about ‘colonialism’. Both views on the Roman world are, of course, legitimate and important, but if we want to pursue and reinvigorate the Romanization debate, I suggest that we focus on ‘(cultural) transformation taking place in the context of empire’ rather than on ‘imperialism and colonialism’. Why?

Over the past decades, a large number of studies have shown the relative, contextual character of what we used to call Roman and Native, especially where material culture is concerned. This is illustrated by the now ubiquitous use of the term ‘hybrid’ to characterize artefacts and archaeological contexts – especially in the Roman world everything is, indeed, in one way or another, a ‘ hybrid’ (for critique of the use of this concept, see below). This approach is summarized by Andrew Wallace-Hadrill, whom I consider the third leading Anglo-Saxon scholar in the Romanization debate, for whom Rome is about bilingualism and code-switching. Building on the important insight that in order to ‘be Roman’ you could ‘go Greek’, Wallace-Hadrill (Reference Wallace-Hadrill2008) compares Romanization to the drawing and pumping of blood to and from the heart, with diastolic and systolic phases. This metaphor – and Wallace-Hadrill's ‘classical’, exclusive focus on Greece and things Greek in a Roman context – is not unproblematic in itself (cf. Versluys (Reference Versluys, Pitts and Versluys2014) and below), but its importance lies in the circularity that it puts at the centre of its understanding of transformation. What I have above characterized as the ‘handicap’ of anti-colonial studies here becomes particularly apparent: its focus on (good) Natives presupposes (bad) Romans – it simply needs two different cultural containers – and as such it leaves little room for circular processes whereby Natives become Romans and Romans behave as Natives. Still, as we now increasingly find out, that is exactly what happens in large parts of the republic and the empire, with various and varying forms of power and agency involved. An exclusive focus on ‘imperialism and colonialism’ will only continue to result, probably, in rather one-dimensional interpretations; Robin Osborne (Reference Osborne2008) even characterized such a persistent focus as ‘colonial cancer’.

Does that make Mattingly's interpretation of what goes on in Roman Britain less feasible? No – in order to make such statements we would have to review the evidence, which is not what this article sets out to do. It does demonstrate, however, that extrapolating his (laudably explicit) ideological, anti-colonial interpretation of this one area from the Roman world to a more general theory on Romanization is dangerous, because both ‘power’ and ‘Roman Britain’ are very specific points of view (and of departure). Before I try to indicate, in the next section, how, then, we should arrive at views that hold wider relevance for the Roman world as a whole (and perhaps even for Roman Britain itself as part of that world), two methodological observations pertain.

The first observation is that an individual scholar's view of Romanization appears to greatly depend on the area that he/she studies, as well as on the historical and archaeological sources available for that particular region. Britain in the first centuries A.D. made Mattingly speak of colonial exploitation; not many scholars would like to use those terms for, let us say, Greece or Syria in that period. The situation of (and preconditions for) becoming Roman in Gaul are markedly different from those in Britain, and it is probably only a scholar familiar with visual culture who could propose to make bilingualism and code switching central to the functioning of Roman identity.

The second observation is the dominance of the concept of ‘power’ – and the concept of ‘imperialism and colonialism’ along with it – in much Anglo-Saxon scholarship on Romanization. Note how even a recent article that thoughtfully discusses the theoretical agenda of Roman archaeology (Gardner Reference Gardner2013) is entitled ‘Thinking about Roman imperialism’ (my emphasis). If we should do away with anything, I would argue that it should be the terms ‘imperialism’ and ‘colonialism’, rooted as they are in 19th-century nation-state discourses and their 20th-century deconstructions (cf. Cannadine Reference Cannadine2013). As we are increasingly discovering nowadays, what we call Greek and Roman imperialism and colonization functioned markedly differently from our modern understanding of these concepts: these are ‘deceptive archetypes’ (Terrenato (Reference Terrenato, Hurst and Owen2005), cf. also Osborne (Reference Osborne2008); for a different perspective see, however, Van Dommelen (Reference Van Dommelen2012)).

Why, then, are these terms in their current understandings still so important for many scholars? What are the difficulties of thinking beyond Romans and Natives? Perhaps it lies in the fact that, because of their more general approach to culture, Anglo-Saxon scholars seem to be possessed by relatively stronger ideas of ethnicity and race than are their Continental colleagues (cf. Leersen (Reference Leerssen, Beller and Leerssen2007), with examples and references). It may also be related to the ‘invasion narrative’ so important for an island culture, as Laurence (Reference Laurence2001) has suggested for Roman archaeology in particular. Perhaps there are thus indeed ‘English peculiarities’ where reflections upon culture and power are concerned, as Gibson (Reference Gibson2007, chapter 3) has argued. This is not the place to explore these questions, but for writing a historiography of post-/anti-colonial Roman studies (see above) it is certainly important to describe it as part of the ‘culture wars’ of the 1980s and 1990s. In a book from 1989, James W. Carey argued that ‘British cultural studies could be described just as easily and perhaps more accurately as ideological studies in that it assimilates, in a variety of complex ways, culture to ideology. More accurately, it makes ideology synecdochal of culture as a whole’ (Carey Reference Carey1989, 97). This seems an apt summary of what happened to large parts of the Anglo-Saxon Romanization debate, in which case it is remarkable that this perspective is still so strongly represented in Roman historical and archaeological studies today. The concept of ‘power’ is important for all historical narratives, but it is only a concept. We should therefore certainly continue to think about ‘power’ in Roman historical and archaeological studies, but only if we seriously theorize the subject – and not in terms of imperialism and colonialism alone.

Local and global: beyond provincial Roman archaeologies

In order to get ‘beyond Romans and Natives’ we will have to get rid of these static taxonomies in an absolute sense, while simultaneously providing enough room in our interpretations for the fact that the categories themselves held much significance in the Roman world. For such an exercise it is crucial to rethink what exactly ‘Roman power’ is. The intention here would explicitly not be to picture a world in which Romans and Natives are happily joining together in building a new world full of aqueducts, bathhouses and other civilized amenities – such interpretations would focus too much on elite negotiation alone and would unhelpfully bring back aspects and associations of the mission civilisatrice (see the just criticism in some of the articles in Keay and Terrenato (Reference Terrenato and Hingley2001), Mattingly (Reference Mattingly2004) and Dench (Reference Dench2005)) – but rather to better understand Romanization as a cultural process (see further below).

The fact that we need to rethink our categories of analysis is the other conclusion that follows from the two methodological observations above. Roman archaeology is hampered by dichotomies between ‘classical’, ‘Mediterranean’ and ‘provincial’, while all are talking about (material culture from) the same social context. Specialization within these dichotomies is even more extreme due to the creation of distinct traditions of ‘provincial Roman archaeology’ – with scholars working on these Roman ‘provinces’ who often, or even exclusively, especially in north-western Europe, are citizens of the 19th-century nation states that were founded with the help of such ‘ancestor cults’ (Terrenato Reference Terrenato and Hingley2001). Scholars often still think, analyse and publish in terms of ‘the archaeology of Roman Britain, the Low Countries, Gaul, Germania, Spain, the Balkans, Italy, Greece, Asia Minor, Syria, Egypt, North Africa’ as if these are proper units of analysis (cf. Woolf Reference Woolf2004). They are only to a limited extent, and what makes things worse is the fact that, as described above, a scholar's specialization in one of these areas and its archaeology strongly influences his/her ideas on what Romanization looks like. Within the Romanization debate this is most clearly visible in what is designated ‘the leaning to the west’ (or ‘the north-western inclination’) of Roman archaeology: the fact that thinking about Romanization is often largely based on studies about the north-western provinces alone. If Martin Millett (Reference Millett1990), in his essay on archaeological interpretation, had analysed Roman Syria instead of Roman Britain, the Romanization discussion in the English-speaking world would have been markedly different (as will be immediately clear from reading Butcher (Reference Butcher2003), for instance; cf. also Versluys (Reference Versluys, Merz and Tieleman2012)). Despite several attempts (like Hoff and Rotroff (Reference Hoff and Rotroff1997) and Alcock (Reference Alcock1997), or, although differently, Ball (Reference Ball2000)), the field still very much needs ‘an eastern inclination’.

To get beyond Romans and Natives we will therefore have to do away with provinces as the main structuring principle of Roman historical and archaeological studies and try to consistently analyse on a local and global level simultaneously (cf. already Witcher Reference Witcher, Herring and Lomas2000). This will likewise result in a better integration of the various research traditions within Roman studies. Scholars who actively use the approaches of both Mattingly and Wallace-Hadrill in discussing Romanization are not widespread (enough). Still, Wallace-Hadrill's arguments about code switching – although undoubtedly primarily a metaphor borrowed from literary studies – are relevant in order to understand, for instance, ‘provincial’ realities at the lower Rhine. When the Batavian chief Iulius Civilis ‘revolts’ against the Romans, he has his hairstyle changed from the Roman custom into a typically Germanic style (cf. Slofstra Reference Slofstra2002a). This is very telling of (‘Native’) power, about becoming Roman and about cultural competence, and indeed we need all three perspectives to understand what goes on – especially in the context of the Batavians, an ethnographic category and identity created in a Roman context alone.

Its ideological development, however, has made the English-speaking Romanization discussion neglect or even forget subjects like the centre(s) of empire, the cities, the monumental, literature and ‘art’. But these themes should be actively incorporated again, especially because (quite a few) scholars working in those fields have learned the post-colonial lesson and initiated similar deconstructions. Note, for instance, how Peter Stewart followed up his Roman art from 2004 with The social history of Roman art in 2008, giving the Romanization discussion an important role. There certainly is a common language now, even if not in all parts of those fields. If provincial Roman archaeologists start taking Classics seriously again, and Mediterranean archaeologists art history, it will soon become even clearer that they are all talking about similar problems in the same social contexts. Here also different (national) research traditions seem to play a role. I was (happily) surprised to see that at the first RAC/TRAC taking place on the (not Anglo-Saxon) Continent (Frankfurt 2012) there was a much wider range of approaches represented than at earlier conferences in Britain (and Amsterdam).

‘How then do you “do” Roman archaeology “beyond provinces” on local and global levels simultaneously?’ one might ask at this point. The answer is simple: by regarding the Roman world and the areas it thought of as the oikumene as one single cultural container. From that perspective we do not deal with acculturation between separated cultural groups in terms of the adopting (voluntarily or not) of cultural traits from A to B or B to A, but with cultural and social interactions within the same group (Le Roux (Reference Le Roux1995) was already moving in this direction by talking about acculturation permanente). With this interaction, all kinds of (invented) ethnicities and (imagined) identities are played out, of course. However, by regarding them as strategies of identity and alterity it becomes clear that indeed ‘the main cultural tensions in the Roman Empire were between small conservatism and global trends, between customary power and Mediterranean-wide political games, between traditional forms of surplus circulation and elements of market economy – more than between Romans and natives or colonizers and colonized’ (Terrenato (Reference Terrenato, Hurst and Owen2005, 70); cf. Versluys (Reference Versluys and de Rose Evans2013), where I work out in detail what is only briefly summarized here). Terrenato's theoretical conclusion following this statement is equally apt: ‘This means that we are moving on a completely different plane from that of most post-colonial discourse, at least in its most frequently used forms’ (see further below).

If we are looking for terminology that rules out ethnic entities and cultural-container thinking – and we therefore have to rule out (post-/anti-)colonial solutions to the problematics of how cultural processes work, like acculturation, colonization, creolization, resistance or even hybridization – then a key element must certainly be connectivity. The Mediterranean and Near East are characterized by an increasing connectivity from the middle of the second millennium B.C. onwards. An intensive circulation of goods in the Bronze Age soon results in trade revolutions and, along with those, a diaspora of Phoenicians and many other Mediterranean peoples like (what we call) Greeks all across the Mediterranean. In the Hellenistic era this ‘global’ world has already become so interconnected that it even starts actively developing its own ‘culture’ – and with an immense velocity all kinds of religious, social and cultural concepts are translated from one context to the other. The Roman oikumene is the ‘outcome’ of this accelerating process of interconnectivity. We will never know what Polybius’ agenda was, when he wrote (in his Histories 1.3, to be dated somewhere between 160 and 120 B.C.) that ‘from this point onwards [after the Second Punic War] history becomes an organic whole: the affairs of Italy and Africa are connected with those of Asia and of Greece, and all events bear a relationship and contribute to a single end’. But it clearly shows that ideas of living in one, single, global world did exist.

The fact that Mediterranean history is all about connectivity within a single framework, in economic and social terms, was strongly argued by Horden and Purcell (Reference Horden and Purcell2000) already over a decade ago, and they pictured the Roman world as a quintessentially Mediterranean society driven by communication with good reason. Although scholars have certainly been interested in these new perspectives theoretically, archaeologists in particular have done too little with them in practice, in the sense of trying to understand what this connectivity implies for (their understanding of) material culture (important exceptions include Van Dommelen and Knapp (Reference Van Dommelen and Knapp2010) and Maran and Stockhammer (Reference Maran and Stockhammer2012)). The realization that the Roman world is fundamentally about connectivity and communication has led to many interpretations about the ‘hybridity’ of material culture, but, as has been argued above, the problem lies in the fact that such interpretations remain within the (post-/anti-)colonial framework because they reason from separate ethnic and cultural containers. Metaphors of hybridity thus presuppose (if not produce) static, cultural dichotomies (cf. Flood Reference Flood2009). Saying that something is (a bit) Roman and (a bit) Native is not going beyond Roman and Native at all. For the Roman world there seems to be something much more radical at stake and it might well be possible that the continuing dominance of the anti-colonial framework within Roman studies prevents us from moving in that direction (cf. more in general Weinstein (Reference Weinstein2005), where this state of affairs is characterized as ‘the post-colonial dilemma’). Here my analysis differs from that of Gardner (Reference Gardner2013), who considers post-colonialism and globalization as the two main (theoretical) approaches taken by scholars to get ‘beyond Romanization’ so far (on this question see now also Hingley (Reference Hingley, Pitts and Versluys2014b)). I argue, on the contrary, (1) that post-colonialism was, in fact, often anti-colonialism and has left the Roman–Native dichotomy intact – something which has resulted in replacing Romanization with ‘Romanization’ and the impasse we are in at present, and (2) that the exploration of globalization theory within Roman archaeology has remained very limited and that the radical consequences (and possibilities) of the concept (cf. Appadurai Reference Appadurai2001) have not yet materialized at all.

In his 2009 essay Pour une anthropologie de la mobilité, the French anthropologist Marc Augé has analysed the difficulties that scholars encounter when describing a world that is fundamentally characterized by people and objects in motion. We are all well aware, for instance, that we live in a postmodern world (surmoderne, as Augé calls it, perhaps more aptly) but we still (and rather naively) use the conceptual toolbox designed for understanding modernity to describe and understand it. I think that the same holds true for understanding connectivity in antiquity. Taking connectivity as the defining characteristic of the Roman world has immense implications and forces us to move our intellectual concepts from acculturation to globalization, from history to mnemohistory, from traditions to the invention of traditions, from being to becoming, from communities to imagined communities and from conceptualizing in terms of cultures to thinking in terms of cultural debates (Versluys (Reference Versluys and de Rose Evans2013) and (Reference Versluys, Pitts and Versluys2014) both provide more background to this claim as well as many examples).

The paradox of the current situation is the fact that, while most scholars seem to agree on the great advantages of this new theoretical position – Horden and Purcell (Reference Horden and Purcell2000) has been hailed as a landmark publication and has even been described as a true paradigm shift – at the same time they do not seem to take it seriously enough. Studies looking for more specific understandings (and critique) of the general notion of connectivity – like Morris's concept of Mediterraneanization (2005) and Woolf's (Reference Woolf, Keay and Terrenato2001) or Wallace-Hadrill's (Reference Wallace-Hadrill2008) idea of a cultural revolution – remain limited. By staying within the (post-/anti-)colonial framework many studies, in fact, strengthen the borders that the new paradigm sets out to undermine and dissolve. In order to genuinely take connectivity in antiquity seriously – and to arrive at a historical anthropology of mobility for the Roman world – we should, I propose, focus on two things: (1) globalization theory and (2) the entanglement of things and people (material-culture studies).

The first aspect can only be briefly dealt with here; in a forthcoming book Martin Pitts and I (2014) will discuss the perspectives and opportunities that globalization theory offers for studying the Roman world critically and at length. I refer to that publication for all further details. Together with the small monograph by Richard Hingley (Reference Hingley2005), this is the first book within Roman studies to explicitly deal with globalization theory from a variety of perspectives, underlining, I think, how limited its application has been so far. Within the social sciences globalization is now simply the way to talk about connectivity; simultaneously those social scientists are asking archaeologists and historians for a deep historical perspective on the connectivity of our modern world(s) (cf. Appadurai Reference Appadurai2001). At the same time, Roman studies are (still) struggling to get beyond Romans and Natives and beyond doing Roman–provincial-area studies by means of focusing on connectivity and (as we have seen, still held back perhaps by the ideological characteristics of the anti-colonial Romanization debate) not really getting there. The only logical conclusion seems to be that we should ‘push the globalization analogy harder, applying to the ancient Mediterranean the same tough questions that scholars ask about connectedness in our own time’ (Morris Reference Morris and Malkin2005, 33). Globalization is not about (American or Roman) power destroying local and authentic cultures; quite the contrary. Globalization theories are about investigating diversity within a single cultural framework, with complex power structures between all kinds of different groups that have shifting boundaries, but also with unintentional results of connectivity and communication. And it is about the transformative capacities of intercultural encounters. ‘Glocalization’ is only a word, but the basic questions of the Romanization debate can be very usefully reconceived within this framework – as explained and illustrated, for example, by Witcher (Reference Witcher, Herring and Lomas2000), Pitts (Reference Pitts2008), Alexandridis (Reference Alexandridis, Hodos and Hales2010) or Mol (Reference Mol2012). Globalization is therefore an excellent tool to make us think local and global, and to get us beyond provincial Roman archaeologies and beyond Roman and Native at last. The second focus, on material-culture studies and the entanglement of people and material culture, will be the central point of the next and final section.

An archaeological perspective on Romanization

Images of Rome in European culture are omnipresent, strong and persistent (Hingley Reference Hingley2001). One of the foremost associations ever made with Rome is that of military conquest and empire building (cf. Terrenato Reference Terrenato and Hingley2001, who identifies lawmaking and engineering as distant seconds). As we have seen above, also for many scholars this is what the quintessential story of Rome is about; for them Rome is always, and simply out of logical consequence, about asymmetric power relations, about imperialism, about colonization and about clashing identities (see Mattingly (Reference Mattingly2010); cf. Woolf (Reference Woolf2012, vii): ‘All histories of Rome are histories of empire’). Now let us for a moment radically put aside these ideas about Rome (as if that were possible, cf. Reece (Reference Reece, Blag and Millett1990)) – and along with them let us put aside all Roman literary sources that were likewise spellbound by the need to explain empire – and let us do what archaeologists should always primarily try to do: look at material culture in its own right first. What do we see then?

We see diasporas of material culture. Material culture of all kinds of different styles, forms and materials. This repertoire is as broad and varied as it is omnipresent all across the Roman world. If it were possible to have a map displaying all artefacts from the Roman world that have been preserved from a certain period, and if an anthropologist with no real knowledge of the Roman world and its history were asked to identify different clusters of material culture on that map to try and make sense of it, he would, I imagine, have a very difficult time drawing such clusters in. Something as clearly recognizable as an amphitheatre or an aqueduct would be evident all across the map, as also, and with even more examples, would be the case for an architectural form called ‘Greek temple’ or an object form called ‘Greek statue’. Objects and architecture in an ‘Egyptian’ style clearly stand out in terms of stylistic properties: they are clustered at some places on the Nile but also at some places in Italy, and the remainder of them are to be found everywhere. Stucco wall paintings, mosaics and terra sigillata pots certainly show small-scale local clusterings, but again they are everywhere to be found on this map, which represents around five million square kilometres – as are large stones with Latin inscriptions, although these are found predominantly in the western half of the Mediterranean. Military forts are everywhere, stereotypical and thus (perhaps) clearly recognizable as ‘Roman’; however, within these forts the variety of (cultural) artefacts is bewildering and impossible to cluster. Even the frontiers of the Roman world would not be as easy to point out as one might expect, if at all. I am not concerned with these specific examples, but with the general picture that emerges from them: in material-culture terms there is no identifiable Roman culture or Roman Empire (on the fundamental question of the meaning of the concept ‘material culture’ see, in general, Hicks (Reference Hicks and Beaudry2010)). From the many circles that our anthropologist would draw on the map – probably constantly correcting himself when, on closer inspection, he notices an example of what he has just labelled as category X in a very different area – it would be impossible to distinguish centre from periphery, colonizer from colonized, and, indeed, Roman from Native. If Roman archaeology were prehistory, Roman imperialism would be quite invisible in the archaeological record. Even the city of Rome itself would be invisible, albeit as a clustering of probably the largest amount of different styles in a single context (cf. Edwards and Woolf Reference Edwards and Woolf2003). A mind map of Rome in material-culture terms is therefore not first and foremost about military conquest or about empire building in terms of imperialism and colonialism at all. In material-culture terms Rome is about the reworking and redistribution of a bewildering variety of (what we call) Celtic, Greek, Mediterranean, Near Eastern and Egyptian forms and styles of material culture. Here Wallace-Hadrill's (Reference Wallace-Hadrill2008; see above) metaphor of the drawing and pumping of blood to and from the heart springs to mind as an alternative narrative. There is, however, no single heart: the system is quintessentially polycentric. And there are no diastolic and systolic phases – there is constant circularity and certainly not between ‘Greece’ and ‘Rome’ alone (‘Egypt’, for instance, plays an important role as well; see Versluys (Reference Versluys2010)). In other words, Rome is globalized and is globalizing.

One could characterize this perspective of looking at material culture in its own right as ‘beyond representation’ (cf. Malafouris Reference Malafouris, DeMarrias, Gosden and Renfrew2004). Nicolas Thomas (Reference Thomas1999, 16) characterizes the importance of this approach very well when he writes that ‘the interpretative strategy of regarding things essentially as expressions of cultural, subcultural, religious, or political entities, depends on too static and literal an approach to their meanings’. Still, as we have seen, this is what most (post-colonial) Roman archaeology is still doing, often under the heading of ‘identity’ (cf. Pitts Reference Pitts2007; Van Oyen Reference Van Oyen2013 for critique). Thomas adds (Reference Thomas1999, 18–19) that in his case, ‘This way of seeing things perhaps also helps us move beyond the long-standing dilemma of historical anthropology in Oceania, which has lurched between emphasis on continuity and discontinuity, between affirmation of the enduring resilience of local cultures, and critique of the effects of colonial history’. I argue that the same is true for the long-standing dilemma that is the Romanization debate.

But one could even take this ‘beyond-representation’ perspective one (radical) step further and make material culture, with its stylistic and material properties (and thus agency; see below), central to our understanding of the Roman world. This would be following what is now commonly designated the ‘object turn’ or ‘the material-cultural turn’ (cf., out of a quickly growing bibliography, Knappett and Malafouris (Reference Knappett and Malafouris2008), Saurma-Jeltsch and Eisenbeiß (Reference Saurma-Jeltsch and Eisenbeiß2010), Hicks and Beaudry (Reference Hicks, Hicks and Beaudry2010) and Hodder (Reference Hodder2012), all with references to the texts by Latour, Appadurai and Gell fundamental to this approach; Hicks (Reference Hicks and Beaudry2010) is an important historiography and outlook from an anthropological and archaeological perspective). Scholars in fields ranging from political theory and literature to sociology are now moving away from an understanding of the world centered on people and texts, and instead are moving towards a reconsideration of the interrelationships between all things, including humans. This ‘material turn’ is, of course, explicitly relevant for the three disciplines that have always been centered around the object, and that now seem to rediscover its agency in cultural-historical terms: art history, anthropology and archaeology. Given the fact that, as has just been argued, in archaeological terms Rome is a world fundamentally characterized, in all aspects, by objects in motion, this seems like a promising direction for Roman archaeology. It will immediately be clear how much the basic questions of the debate are reconceived through and within this framework: a central question for Roman archaeology now becomes whether or not Romanization can be understood as a relatively dramatic punctuation of connectivity that was primarily brought about by objects in motion and, of course, by the implications of this human–thing entanglement. We should, to phrase it radically, try and understand the era we call Roman primarily in terms of materiality – and thus in some ways similar to how we commonly understand the Bronze Age, the Iron Age, etc.

Material culture is an active agent in its relationship with people, rather than simply a representation of (cultural) meaning (alone). It is important to emphasize this point because, as we have seen, material culture often has (only) been made to represent a lot of different things in the various meanings of ‘Rome’ that have so far been discussed. But if the (potential) meaning of things in Roman contexts is of such bewildering complexity and fluidity – especially when looking for meaning as the outcome of the use of stylistic and material properties in a fixed relation to identity – we should perhaps focus more on what the object in question does. And in order to focus on what it does, we should investigate its agency: the way it determines its viewer, its immediate context and, consequently, its historical context (cf. Boivin Reference Boivin2008). An object called ‘Greek statue’, for instance, has no fixed meaning as such. It has not necessarily anything to do with ethnic Greeks and often cannot be connected with a desire to acquire a ‘Greek identity’ in a particular context. Even as a cultural or social concept it is evasive; the same form of Greek statuary can simultaneously be found on a bone amulet worn by a slave and, in original Greek bronze, in an imperial collection. This indicates, again, that we should not so much focus on what things with their stylistic and material properties would represent – or to what historical narrative they testify – but on what things do in a certain context. And in order to reconstruct what we then should call the experience of the object (form) called ‘Greek statue’, we should seriously take its agency into account. Understanding (Roman) material culture is about human–thing entanglement in which the ‘thingness’ of the object has an important part to play (cf. Jones and Boivin (Reference Jones, Boivin, Hicks and Beaudry2010), who call this ‘material agency’).

It is already well known that the specific stylistic and material properties that make up the diasporas of material culture that we call the Roman world had specific associations that actively affected people. With good reason Tonio Hölscher (Reference Hölscher2004) has talked about ‘the language of images’ – note how this phrasing puts the agency with the objects – but material culture without images had also agency. When these aspects of materiality and the agency of material culture in the Roman world are brought into connection with the potential of globalization studies as discussed above, things become particularly interesting. If we can really describe material culture as actant, as playing a role in networks of (social) relationships, the Roman world is a very special case. As has been described above, from the period around 200 B.C. onwards the network had become so ‘hyper-’ and interconnected that it would be better to call it global in order to indicate the degree of connectivity at stake. More things and people came together in the Roman world than ever had done before, which implies that (potential) human–thing entanglements exploded. Gosden (Reference Gosden2004) is therefore certainly right, I think, in understanding colonialism as crucially a relationship with material culture, a particular grip that material culture has on people. But I would understand this more generally and see colonization as just one out of many more forms (and degrees) of connectivity. It is the degree of connectivity that matters when understanding ‘material agency’; this is the point of calling the Roman world ‘global’. Following Appudurai (Reference Appadurai2001) and others, it is interesting and important to think about the Roman world as a device trying to handle all these objects in motion and to make sense of all the human–thing entanglements that constantly kept coming back, because, as we have seen, Rome was globalized and globalizing. The urban landscape of Rome itself testifies to this idea of ‘handling objects in motion’ in many respects, so it seems, because all the (important) object forms that circled within the system had their place in Rome: from ‘Greek temples’ and ‘Egyptian obelisks’ to ‘Celtic armour’. Things from the Eastern Mediterranean always were most prominent within this system: might that be the reason why the Roman Empire, as a device to handle objects in motion, eventually had to move east and make Constantinople its new capital? Be that as it may, I argue that we are in need of a non-anthropocentric approach towards the genesis and functioning of the Roman Empire (cf. Knappett and Malafouris Reference Knappett and Malafouris2008).

There is, of course, much more to be said about what I wished to indicate here as a fundamental alternative perspective. The choice to take material culture seriously in its own right forces us to critically rethink Romanization: be it in terms of ‘beyond representation’ or in terms of (radically) following the ‘object turn’. My reasons for underlining the archaeological/material culture/object perspective in the context of this essay are still rather limited and applied. Studying material culture in its own right, like having agency in a human–thing entanglement, makes clear at a glance that ‘Rome’ is not about empire building and imperialism or about Romans and Natives. If thinking in terms of globalization is one way to overcome that paradigm, then focusing on material culture and its agency certainly is another. We should therefore consider redirecting part of our research agenda from a focus on territories towards networks (or, as James Clifford famously remarked, from roots to routes: this is what globalization studies can do) and from texts towards things (as material-culture studies does; cf. Flood (Reference Flood2009)). Let me stress immediately that such a redirection should, of course, be understood as an alternative perspective alone. Material culture certainly has its own problems when used for historical analysis and also I certainly do not want to suggest a dichotomy between ‘historical’ and ‘archaeological’ interpretation. However, in the current situation, which I have analysed in the first part of this essay, material culture in particular seems able to nuance and challenge existing (historical) narratives. We have seen that historical narratives made ‘Rome’ into a story of military conquest and imperialism, while a focus on (networks of) material culture and its agency might tell very different stories. Especially Roman archaeologists, therefore, should take the ‘object turn’ seriously. But that does not imply that they should forget thinking about territories and texts or about colonialism and imperialism. It also does not imply that the only things that matter in understanding ‘Rome’ are longue durée developments of these kinds regarding the power of things and the flow of cultural transformations: événements and conjonctures, institutions and politics, also hold much significance and indeed Rome is more than a cultural process alone. However, I believe that focusing on ‘Rome’ through the lens of globalization theory and material-culture studies has, at this very moment, the potential to reconceive the debate and to move us beyond some long-standing dilemmas.

Hartmut Böhme (Reference Böhme2006) has beautifully shown that one can write ‘Eine andere Theorie der Moderne’ when taking ‘die stummen Dinge im Aufbau der Kultur’ seriously. This is what we now should try to do for the Roman world. That prism will make clear that, in material-culture terms, ‘Rome’ does not so much refer to a culture or a culture style, but rather indicates a period of remarkable connectivity and its material/human consequences. Artefacts we call ‘Roman’ are therefore not in the first place expressions of Romans or of ideas about Rome. They are concrete material presences part of a spatial relation in (historical) time and (geographical) space: Romanization is about understanding objects in motion.