This volume, and the meeting it derives from, represents a refreshing new direction in the engagement between Roman archaeology and archaeological theory. This process of interaction has a complex history, frequently characterised by scepticism from both directions, if not the wilful maintenance of ignorance – by practitioners in both domains – of what the other group has to offer. In this respect, Roman archaeology has traditionally sat alongside other historical sub-disciplines like Medieval archaeology, taking its interpretive cues largely from the textual record, however partial or fragmentary, and regarding insights from other disciplines dealing with human societies with suspicion, and as at best relevant only to prehistory (Frere 1988, 36; cf. Scott 1993a, 6; Gardner 2007, 35–9). It is perhaps no surprise then that ‘representational thinking’ (Van Oyen and Pitts, this volume) has been prevalent in Roman archaeology, as objects were almost literally illustrative material for text-derived narratives until very recently in the history of the field, and the idea that texts are objects too has not yet had the impact it might (Johnson 1999; Laurence 2001). Another reason for conservatism has been that Roman archaeology is an international field and therefore, like Egyptology or Assyriology, it has straddled a wide range of different national traditions (Andrén 1998). As with the availability of textual evidence, this can of course be a great strength, but as the continental European tradition of Roman archaeology has generally been less engaged with the theoretical debates in Anglophone prehistory since the 1960s, which have come to define ‘archaeological theory’ in general, this has arguably acted as a further brake on theoretical engagement in the sub-discipline. When this did begin, sporadically from the late 1970s but really taking off in the 1990s, it is no accident that it was primarily in Romano-British archaeology, closer to those prehistoric debates in both scholarly language and in having relatively little in the way of relevant written sources compared to other parts of the empire.
Over the last 25 years, there has been much catching up, and not only has Roman archaeology developed its own trajectory of theoretical innovations (cf. Gardner 2016a), but the debate has increasingly spread further within Roman studies. This is all to the good, but the aim of this chapter is to ask whether Roman archaeology can yet claim a place at the forefront of wider theoretical debate, using the contents of this volume as a springboard. Also – and before tackling that question – I wish to explore how theory-building actually happens in Roman archaeology. Part of the answer to this lies in the dynamics of the contexts in which theoretical discourse takes place, and in the forms which that discourse takes. The annual Theoretical Roman Archaeology Conference (TRAC) has been particularly important for the broadening of theoretical debate, albeit mainly within a British context (both in terms of conference location and audience, and provincial subject-matter) – though there is a long-standing engagement with the Romanist community in the Netherlands and the conference has travelled there and, with the Roman Archaeology Conference, to the USA, Germany and Italy. TRAC is just one conference, though, and while it has undoubtedly nurtured the theoretically-informed work of many of the current generation of Roman archaeologists, including most contributors to this volume, there is a need for other venues and formats. In particular, it is vital that more in-depth theoretical explorations find a home, to enable, both in the interactive space of a meeting and the published space that might result from it, a fuller engagement with what should be profoundly challenging concepts (cf. Gardner 2016a, 10–11). The ‘Rethinking Artefacts in Roman Archaology’ Laurence Seminar at the University of Cambridge provided just such a venue (with longer papers and more discussion than tends to be possible in bigger conferences) and, in this volume, just such an outcome, and this is to be applauded. Opening up more time and space for discussion is not the only crucial step, though – so too is the hard work of wrestling with concepts that might reveal markedly different goals and principles among practitioners. In the remainder of this brief contribution I try to map out some of the key points of coincidence, and discord, and relate these to the wider archaeological landscape.
The aim of this volume, as I see it, is to explore new ways of comprehending the definitive material transformation that the Roman world presents us with, in particular those inspired by the ‘material turn’ (from which I would see the ‘ontological turn’ as a progression; cf. Van Oyen and Pitts, this volume). As the editors state at the outset, the sheer materiality of the Roman period in most parts of the empire is striking, and in terms of explaining or understanding this materiality, self-fulfilling narratives like ‘Romanisation’ have proven insufficient. The influence of such traditional paradigms on the ways in which artefacts have been excavated, classified and interpreted is profound, and in some ways we need to begin again from the ground up, to rework how we deal with artefacts, both theoretically and methodologically, and in highlighting avenues forward with this project each of the papers has much to offer. How this work generates new higher-order narratives is less immediately clear. Partly this is because such a goal must be seen as a long-term project, only just beginning, if it is going to be possible at all – and how we can go about determining both the length and the viability of such a project as a research community is something I return to below. Partly, also, this may have to do with the nature of the ‘material turn’ and some of the selectivity of the associated theoretical toolkit, which needs to be evaluated carefully. What is clear is that important new themes are emerging from the work in these papers, which are certainly of theoretical importance, as well as showing that Roman archaeologists are innovating methodologically, and making good use of the high-quality data that they have access to. Earlier ‘discussion’ contributions in this volume have examined some of these themes in detail, so I have just a few points to add.
In terms of the practice of Roman artefact studies, many of the papers highlight the ways in which getting to grips with the materiality of finds (sometimes literally) opens up new insights, from the role of the capacity and portability of inkwells in writing practices (Eckardt) to the various ways in which the shape and material consistency of dice played a part in gaming (Swift). These point towards the importance of phenomenological engagement with finds as a method, albeit one that raises interesting questions of the universality of embodied experience, which is a recurrent and unresolved dilemma in archaeological theory (cf. Brück 2005; Hamilton and Whitehouse 2006; Lindstrøm 2015). Classification is a related issue which several further papers address, and from a range of different perspectives, but all highlighting aspects of the familiar emic/etic debate (cf. Scott 1993a, 6, 20), as well as the intersection between categories drawn from contextual studies of artefacts and those drawn from written sources (i.e. artefacts with writing on them). Thus, Collins re-examines some of the problems of relating categories of things to categories of people, Mol explores the Roman classification of ‘Egyptian’ material, and Van Oyen discusses not only the phenomenology of concrete use, but also its variable classification in antiquity. Through the elegant case studies presented in each of the five papers mentioned so far, it is clear that the ways in which people in the Roman period categorised and interacted with objects must underpin future narratives of that period. In a way there is something of a pendulum swing here, in that some of the earliest attempts to classify Roman objects started with things known (or believed to be known) from textual sources, and thus believed to be classifiable in ‘Roman’ taxonomic terms (e.g. ‘samian’, Tyers 1996, 6–7). The increased attention in the twentieth century to objects less well-represented in such sources (e.g. coarse pottery) and, latterly, a reaction against textual determinism beginning in the 1970s under the distant influence of the New Archaeology, led to more ‘archaeological’ classifications, albeit perhaps bypassing some of the important work on the emic/etic problem that had been a key part of the emergence of that paradigm (e.g. Ford 1954; cf. Millett, this volume). Returning to efforts to understand Roman-period categories, but on a firmer multi-evidential basis, is certainly to be welcomed.
The other four main papers in the volume shed equally important light on a further related issue, to do with the variability and standardisation of the Roman material world as, in itself, a crucial characteristic of that specifically ‘Roman’ world. Pitts, Poblome et al., Murphy, and Jiménez all address the marked standardisation, but also the potential for local variability, in material culture (particularly pottery) across a range of geographical contexts. That a focus on the particular material characteristics of the Roman empire directs us towards such themes is certainly a beneficial consequence of the ‘material turn’, and one which offers much potential also for methodological innovation in how we chart these phenomena. That each paper uses different concepts to articulate some of the broader narrative significance of their particular cases – such as globalisation, koine, provincial networks – is both appealing in highlighting the flexibility of current theoretical work in Roman archaeology, and, I believe, revealing in its implications for some of the limitations of that ‘turn’, on which I will expand in the next section.
This volume does indeed move the agenda forward in Roman archaeology with respect to the interlocking questions of the nature of (Roman) things and their articulation with (Roman) narratives. In doing so it also draws into view existing debates around these issues in Roman archaeology and far beyond which bear a little revisiting, in order to illuminate my central concern with how we go about doing theoretical Roman archaeology. The ‘material turn’ is, for all its concern with ‘things’, also an interesting case study in the history of ideas and their inter-disciplinary biographies. Partly this is because, like the ‘linguistic’, ‘practice’ and other ‘turns’, this movement is multi-stranded, drawing on different sets of disciplinary experiences and philosophical roots, with sometimes contradictory consequences; in this respect it is (like those other turns) rather intimately associated with post-processualism as an eclectic theoretical disposition (cf. Shanks 2008; Lindstrøm 2015, 209; Thomas 2015, 1287–8). Unlike those other turns, though, the origins of several major approaches within the ‘material turn’ lie quite close to home, springing from the evolving work of, in particular, Christopher Tilley, Daniel Miller, and Michael Shanks, at the intersection of archaeology and anthropology. Indeed, the former two of these have long worked in the UCL Anthropology department, which has nurtured numerous other key figures developing a ‘materiality’ approach (Buchli 2002; Miller 2005), while the latter has, with several of his students at Stanford University, been a driving force in the development of ‘symmetrical archaeology’ (Shanks 2007; Webmoor and Witmore 2008), taken up also in the work of a dispersed network of other scholars (e.g. Olsen 2010). While the ‘symmetrical’ approach draws considerably on ideas developed in Science and Technology Studies, particularly the work of Bruno Latour (1993; 2005), the ‘material turn’ has been seen as a project which archaeology can claim leadership in, enfolding also the phenomenological strand pursued by Julian Thomas and others (Jorge and Thomas 2007), a more cognitive-processual strand developing particularly in Cambridge (DeMarrais et al. 2004; Malafouris 2013), and other elements including those based upon Gell’s art-as-agency approach and Peircean semiotics (Gosden 2005; Knappett 2005).
A couple of interesting points arise out of this potted intellectual history, which necessarily oversimplifies much (cf. Mol; Murphy; Van Oyen; Van Oyen and Pitts this volume; also Gardner 2003 for earlier stages). One is that there is much to be said in favour of a set of approaches which seek to address fundamental issues of the materiality of human existence across time and space, and which have been genuinely influential in other social science fields (Dant 1999; 2005; Attfield 2000; Woodward 2007; Matthewman 2011). The other is that there are also some interesting problems posed by such approaches which generally lead archaeology towards a very anthropological scale and style (in terms of evidential demands) of research, and away from a more sociological one, even as the social structures of the discipline work against the formulation of coherent positions on some of the aforementioned fundamental issues.
By this, I mean that the increasingly fragmented nature of debate in archaeological theory, influenced by a whole host of structural characteristics of the current practice of archaeology (whether at a global scale – see e.g. Mizoguchi 2015; cf. Bintliff 2011; Thomas 2015 – or the smaller scale of the Roman archaeological community – e.g. Gardner 2016a), makes it difficult for individual scholars, or groups of scholars, to develop long-term engagements with an ever-expanding web of ideas and a rapid cycling of priorities. This leads to inconsistencies between rhetoric and practice (Johnson 2006, 125), to terminological and conceptual slippage (e.g. with regard to agency; Gardner 2011b; 2016b; Johannsen 2012; Lindstrøm 2015; Ribeiro 2016), and to curious juxtapositions between what are in many ways incompatible intellectual programmes (e.g. between Darwinian memetics and some materiality approaches, see Gosden 2005, 198; Hodder 2012, 148–57).
All of this, I would argue, further feeds the process of fragmentation. It also creates – perhaps alongside other challenges like increasingly vast databases (Bevan 2015) – new problems of crafting syntheses and narratives. This returns us directly to this volume, where the linkage between new approaches to material culture and new narratives is explicitly on the agenda. I would argue, as others have (Jones 1996; Miettinen 1999; Barrett 2014; Lindstrøm 2015), that for all their novel attention to fundamental aspects of human experience (to retain a degree of anthropocentrism), aspects of some approaches within the ‘material turn’ positively inhibit developing broader-scale narratives which deliver analytical insights. Thus, for example, the flattening of agencies which is a key part of a symmetrical archaeology diverts us from structures of power (Gardner 2011b, 72–5; 2014; Matthewman 2011, 120–2; cf. Thomas 2015, 1293), while credulity towards alternative ontologies of the ‘animated’ material world (increasingly referred to as the ‘ontological turn’), makes the consideration of the role of ideologies in sustaining social inequalities difficult (Spriggs 2008; Wilkinson 2013; Lindstrøm 2015; cf. Thomas 2015, 1290). Distributing agency can also have troubling moral consequences (Ribeiro 2016). All of these issues of power, inequality and ideology at least need to be up for debate, in part because they are critically important to the role of archaeology in the present day, and strong arguments, which remain compelling, have been made for highlighting them, in the not-too distant past (e.g. from a Marxist perspective, Leone 2005; Leone et al. 1987; McGuire 2002; 2008). Crucially for the present volume, it is precisely these kinds of questions and approaches which need to be retained from earlier phases of the development of archaeological theory if a new appreciation of materiality is to be connected also with broader Roman histories. This might require some new ways of working if we are to move forward.
In framing this volume around the problem of relating objects and historical narratives, the editors have therefore squarely targeted a key contemporary challenge in archaeological theory. That, in this respect, Roman archaeology might find itself at the cutting edge of theoretical debate is not only to do with the increasing appetite for new approaches among Roman archaeologists, but that this appetite is tempered both with the perspective that catching up on previous trends lends to some of those developments (Scott 1993b, 4), and with the inescapability of the scale of the Roman empire itself, no matter how hard we try to re-imagine it (Barrett 1997). This scale presents its own problems for synthetic narrative, referred to above, but it also furnishes a constant reminder of the need to think not only about the phenomenology of particular clusters of people and things but also major social-structural forces. The case studies provided across the chapters comprising this volume do, as already discussed, furnish some key themes with which to make that articulation work, and demonstrate some of the combinations of different approaches to material culture that the editors pre-figure in their Introduction. As the editors acknowledge, though, and as is inevitable within the constraints of multi-chapter volumes, this marks a beginning, not an end, and there is much still to do. In particular, there is scope for deeper engagement with the purely theoretical issues, to highlight further opportunities for which Roman archaeology is particularly well-suited, but also to overcome some of the contradictions and inconsistencies that we have inherited, and to avoid reinventing old truths or perpetuating old myths. I would have liked to see more of this in some of the chapters collected here, though I am well aware that there is a perpetual problem in theory-building which has to do with the balance between cases and concepts in short-form writing. While a relatively oft-cited critique of media such as TRAC papers (Laurence 1999), I think this is really another symptom of the much wider fragmentation of archaeological debate noted above. In the face of increasingly expansive data resources, and an equally increasingly variegated theoretical landscape, with ever more stuff to think through with ever more ideas, we pursue increasingly individuated and accelerated pathways, partly because that is how one gets (and keeps) a job in the humanities branches of academic archaeology. This situation is unsustainable; it does not serve those who work in archaeology in whatever role, nor does it serve the many wider publics for whom Roman archaeology is an important reference-point. To counter this trend requires collective and collaborative labour, encompassing theory and practice. There are many ways of pursuing this (see Gardner 2016a for some practical suggestions), but the Laurence Seminar in May 2015, and this resulting volume, are exemplary of precisely this kind of labour. Simply put, we need a lot more like them.
Many thanks go to the editors for their invitation to participate in the Laurence Seminar in 2015, and this volume, as well as for their forbearance with my fussiness over the terminology of ‘material agency’. Thanks also to all of the contributors for the excellent papers which stimulated much of this chapter, other aspects of which have been developed in dialogue with a wide range of colleagues at TRAC and elsewhere over the last couple of years, including Richard Hingley, Matthew Johnson, Lisa Lodwick, and Kathryn Piquette.
* UCL Institute of Archaeology.