My essay is intended as an archaeological dialogue on Romanization in the sense that it tries ‘to creatively discuss what we mean when we say “Rome”, across boundaries set by disciplines or scholarly traditions, fuelled by new developments in other fields, and especially in terms of material culture’ (p. 6). Therefore it is looking ahead on purpose, with the historiography of the field (only) serving as a means to an end and not so much as a point of discussion in its own right. Resuscitating, repairing or refitting the concept of Romanization itself is not at stake: on purpose I talked about ‘reinvigorating the Romanization debate’ while not discussing the feasibility of the concept in the present situation (‘When I use the term “reinvigorating” I explicitly do not mean to indicate that we should continue the debate as if Romanization were either good or bad, or to question whether or not we should use the term at all’) (p. 6). The essay, therefore, is not so much about Romanization 2.0 per se, but rather about triggering a debate to get us going again or, to paraphrase Woolf's conclusion, to get us out of the Romanization/‘Romanization’ comfort zone. The Romanization debate indeed seems to have brought us a new canon of reference works and key concerns, as Woolf rightly remarks. I argue that we should urgently try to refresh the canon by looking critically at what has become received wisdom and by simultaneously exploring what could become new key questions for our field. Canonization should be a dynamic and ongoing process.

It was very stimulating, therefore, to see all kinds of useful and intelligent applications and rejections of, and suggestions about, my propositions throughout the comments. If there ever was a discussion article that should be read along with its responses, it certainly is this one! To be clear: I have therefore no intention of creating yet another great narrative or overarching theory – how untimely that would be in our fragmented times. But I simply believe that Roman studies, and Roman archaeology in particular, can still do (even) better by engaging more ‘with the humanities and social sciences at large’ (as Van Dommelen characterizes it, p. 45). From my agenda it has become clear what, in my view, our creative discussion on ‘Rome’ should focus on at this particular moment, namely investigating Rome in the context of world history (globalization) and investigating objects we call Roman in terms of their material agency in the first place (material-culture studies). Such an approach very much decentres Rome and is therefore a useful (and much-needed) antidote to the ever-continuing ‘fetishizing’ of the Romans, a point eloquently made by Van Dommelen.

The five responses clearly show, I think, that such an agenda can count on broad assent (at least among my commenters) and, moreover, is timely. That is comforting indeed – although the important task of making it work in daily scholarly practice is still largely before us. Saying that the Roman world is a global one (and that we therefore need globalization studies to understand it better) is interesting, perhaps, but remains descriptive. In order to make the concept of globalization explanatory for the Roman world, we have to try it out and make it work. The same goes, of course, for my suggestion that we are in need of a non-anthropocentric approach to the genesis and functioning of the Roman Empire.

There are, however, differences of opinion about how all of this (which I characterized as ‘an uncomfortable and unfruitful ending’, p. 4) came to be, and about how to move forward. That we should engage with connectivity through globalization studies is not so much contested, I think; that we should (radically) follow the ‘material turn’ is. From the comments it is clear, moreover, that an important issue here is the question what the ‘material turn’ implies for the relation between historical sources and approaches in relation (or opposition?) to archaeological ones. In my reply I will first deal with what has been said about (my view of) the historiography. All five responses make important points that add to the analysis and bring it in sharper focus. Woolf, Hingley and Van Dommelen do so by discussing (parts of) the article and its suggestions proper; Hodos and Stek by zooming out and contextualizing the debate. They do this by looking at their own fields of, respectively, the archaeology of the Greek and the (pre-Roman) Italic world – thus usefully drawing in and discussing the notion of Hellenization. I will next turn to how my respondents think about globalization and material-culture studies as directions for the near future of Roman archaeology (or their own fields) and I will argue, in the spirit of dialogue, that they underestimate both the radical nature and the possibilities of these concepts, especially where it concerns ‘the material turn’.

The response by Van Dommelen is largely concerned with (my reading of) the historiography of the debate and my understanding of (what Hingley calls) ‘post-colonial Roman archaeologies’ (PCRAs). He adds a crucial element to the debate, namely the ‘nativist turn’. Indeed, alternative histories concerning people without history were already developed from the 1960s and 1970s onwards. Van Dommelen calls these approaches rather one-sided – what I characterized as anti-colonial. These ‘nativist’ perspectives would have paved the way for PCRAs; in their turn, these archaeologies would have tried to develop more nuanced approaches. Indeed, intellectual genealogies hold the key for understanding current intellectual trajectories and therefore Van Dommelen is quite right when he says that, in fact, we need more historiography. But I think that we agree on the fact that in most cases concerning PCRAs we could never really speak of post-colonialism. Much survey archaeology (but certainly not all), for instance, has until fairly recently withdrawn itself into the nativist/anti-colonial corner without really engaging in depth with what post-colonial theory has to offer. Hence, if PCRAs genuinely, and all across the field, aimed to be about investigating the location of culture from a variety of (alternative) perspectives, this archaeological dialogue would be unnecessary. However, although I certainly see innovative research emanating from PCRAs, with both Van Dommelen and Hingley, I also see many nativist/anti-colonial reflexes, as I have tried to illustrate in my text. And clearly, as also Hingley acknowledges, I am not the only one. Therefore, in my view, the ‘dynamic and transformative agenda’ can better start with what post-colonial theories are about in somewhat more general terms: contact situations and therefore connectivity, or, to use the concepts now widely used in the humanities and social sciences to study these processes, network theory and globalization. Would it not be illuminating to study Roman colonialism and imperialism from this much wider perspective? However, to be clear, this is not the main issue. Often we tend to expect too much of theory. Whether through post-colonial theory or globalization theory: what matters is the ‘dynamic and transformative agenda’, to use Hingley's formulation (p. 20). And I am more than happy (although a bit more sceptic) to share his optimism that, in that respect, PCRAs are now in a process of reinventing themselves at last.

If so, I hope that the two main points of the agenda presented in my essay will also be part of that reinvention; a recent article like Van Oyen (Reference Van Oyen2013), at least, suggests that they can be. First, I will briefly elaborate on the feasibility of globalization theory. Talking about a global Roman world is, of course, dialectically continuous with our present-day concerns, as Hingley reminds us. But I believe it is more than that. This is why I drew on the 2009 essay Pour une anthropologie de la mobilité by the French anthropologist Marc Augé and underlined that

Or, as Appadurai (Reference Appadurai2001, 1) remarked (and shows throughout his work), ‘globalization may not be a member of the familiar archive of large-scale historical shifts’. Be that as it may, I think the comments show understanding of, or even sympathy for, the idea that talking about connectivity necessitates the involvement of globalization theory. In discussing their respective fields, Hodos and Stek even use it in a rather matter-of-fact way and consider it useful. But, of course, all this remains to be seen. Woolf is certainly right in saying that talking about ‘global Romans’ is descriptive rather than explanatory. It is for that reason that we should make it part of our agenda and explore the concept by means of concrete, applicable and focused research questions. Globalizing the Romans is certainly not easy, and it is quite an intellectual responsibility, but we should try to do it – not least because it effectively decentres Rome.

Globalization thus might be a useful concept in order to understand differently the genesis of the Roman world (that is, in the context of world history) and the different scales within that particular network. Material-culture studies are useful, I argue, to develop a different conception of the relation between artefact and culture. I think there was less sympathy for or understanding of this. Woolf here really deferred his subscription; Hodos remarked that ‘the object itself still does not act’ and stated that the perception ‘that objects are agents on behalf of social actors . . . is more workable in a practical interpretational way’ (p. 28); while Stek associated it with an ‘“ahistorical” approach’ (p. 31). Unfortunately, this reply is not the place to go into this issue in depth, which I believe to be of crucial importance for Roman archaeology in particular. Let me first repeat that ‘I certainly do not want to suggest a dichotomy between “historical” and “archaeological” interpretation. However, in the current situation . . . material culture in particular seems able to nuance and challenge existing (historical) narratives’ (pp. 18–19). For this, however, we have to take the ‘material turn’ and entanglement theory seriously. Taking material culture seriously, however, is not just to draw in more archaeology and archaeological sources. This is indeed, as Woolf rightly concluded, not the way I meant it. It is to study and understand the impact of things on human thought, society and evolution (cf. Boivin Reference Boivin2008), just as we study and understand the impact of humans. Hence the point is not only that we need more material culture in a debate that is dominated by historical questions and frameworks. As both Hodos and Stek show, such an ‘archaeological turn’ has already been taking place in their fields for around two decades already, and has certainly been fecund. Still, this ‘archaeological turn’ is not about understanding material culture as constitutive of its historical context as human actors. Hodos rightly remarks, ‘Ultimately, it is the expression of social identities through the use of material culture that underpins human connectivity’ (p. 30). But I think that theorists of material agency would even see this as more radical, and would argue that it is only material culture that makes human connectivity and articulation of social identities possible. The word ‘expression’ is at the heart of the matter here, because material culture is as expressive as it is constitutive of the social realities of its time. Therefore, we should not only go ‘beyond representation’ as far as material culture is concerned; we should also take material agency seriously in order to write history. This is why I concluded my essay with a reference to Böhme (Reference Böhme2006): this is very much history writing but from the perspective of objects. And as it turns out, history looks very different from that perspective. This is also why I believe that we need to try and write a non-anthropocentric approach towards the genesis and functioning of the Roman Empire. Roman archaeologists, however, have been virtually absent from debates concerning ‘the material turn’ so far, and apparently have often not fully realized the immense potential of their own archaeological sources as history makers.

How to achieve all this and materialize such an ambitious agenda? As both Woolf and Stek usefully argue and underline, we need explicit and targeted research questions that are not dependent on grand theories or higher abstractions. I agree. At the same time, however, it remains the question whether or not we can really evade the great narrative so conveniently. I was reminded here of an observation by the social anthropologist Gerd Baumann, who urges us to dare to think in terms of structures and grand narratives (again), against what he calls ‘the fashionable but helpless reduction of all social facts to contextual contingency’ (Baumann and Gingrich Reference Baumann and Gingrich2004, xiv). This should certainly be the subject of another archaeological dialogue in the future.

Has Romanization still any role to play, then? I have tried to evade this question in my article, but the comments make it clear that I should have been more specific. Romanization as an explanatory model evidently will not suffice. It is too unspecific, too grand a theory, and it comes with much unhelpful baggage: all comments have important things to say about its faults. However, if we want to talk about change and transformation taking place in the context of an empire, and if we take into account the fact that what the Romanization debate has produced as alternatives is new words for old ones (pace Woolf), I believe that Romanization cannot but stay with us as a descriptive term. And similar to other descriptive terms (creolization, Mediterreanization, globalization), it only really means something when we try to make it explanatory. ‘Rome’ is the framework, the common denominator of what we see happening in large parts of the Mediterranean network. And unlike with Hellenization, as Hodos explains, power and politics that we call Roman do play an important role in the functioning of the network. Therefore, as long as we do not regard Rome in terms of the 19th-century nation state, Romanization is perhaps less at fault as a descriptive term than is often thought. And, in fact, is the history of the Roman Empire itself not a history of decentring Rome? Be that as it may – and perhaps most importantly – let me underline once more that Romanization 2.0 for me is a debate – canon-work – and not an explanatory grand theory.

We will achieve this, of course, only through a constant dialogue; not just with one another, but also with those in the humanities and social sciences whose debate is ours. I am confident that if we work on writing a historical anthropology of the Roman world in terms of connectivity and with a focus on material agency, as I have proposed in my essay, Greg Woolf will be applying for a subscription for Romanization 2.0 very soon. But even though this would certainly be a landmark, it would not be the ultimate goal of the exercise. That goal is a continuous intellectually creative discussion on how to understand Rome ‘across boundaries set by disciplines or scholarly traditions, fuelled by new developments in other fields’ (p. 6), leading to Romanization 3.0, 4.0, etc. Therefore: out of our comfort zones and au travail!