Divided Britain and the meanings of the past in the era of Brexit

The connections between identities and narratives of the past in the United Kingdom conform in several ways to more general dynamics that, in recent years, have seen a resurgence of populist nationalism (cf. e.g. Niklasson & Hølleland Reference Niklasson and Hølleland2018). However, like any individual case-study there are particular trends at work in the UK, and a particular manifestation of these trends in the EU membership referendum of 2016 and its aftermath. These events have had a profound, and entirely negative, direct impact on the practice of archaeology in the UK, affecting research funding and collaborations in universities, the staffing of commercial field units, and more indirectly the vitality of the museums and heritage sector (Gardner & Harrison Reference Gardner and Harrison2017). In this paper, though, I want to concentrate on the way particular identity dynamics have been manifested in the debate around the place of Britain in the world, and the relationships between the constituent nations of the UK, which are key themes of the Brexit years and which draw strongly upon symbols of the past in the making of diverse political arguments. Scholarship on these themes has been building since the referendum (e.g. Bonacchi et al. Reference Bonacchi, Altaweel and Krzyzanska2018; Gardner Reference Gardner2017), but of course much is still unfolding. It is clear already, though, that different phases of the British past, from at least the Iron Age through to the twentieth century, enjoy considerable currency in these arguments—often on both sides of the Brexit debate. In the first part of this section I want to look at the particular theme of empire in British identity, which brings with it the complexity of Britain’s status as both a colony and a colonizer, before examining the intra-national politics of the UK and the relationships between English and other identities in the British Isles and Ireland.

The events of the EU referendum and its aftermath are well known and can be succinctly summarized. While a small but vocal anti-EU movement had been evident in British politics for some time (the UK Independence Party, UKIP, was founded in 1991, adopting that name in 1993), and previous Prime Ministers had certainly had difficulties with aspects of European policy before, Prime Minister David Cameron included a commitment to a referendum on EU membership in the Conservative party manifesto of 2015, having announced this objective in 2013. This was fairly clearly an attempt to shore up Tory support in the wake of the unusual period of coalition government since 2010, and also perhaps to distract from the consequences of the austerity economics that had been imposed on the UK in that period. Neither of these factors boded well for the outcome of the referendum, in fact, though Cameron’s confidence in winning the argument for Remain may have been bolstered by the results of the Scottish independence referendum—on which more below—in 2014, which the pro-independence movement lost (Bale Reference Bale2022; Dorey Reference Dorey2021). In any event, the referendum on EU membership took place on 23 June 2016, and produced a narrow majority to leave the European Union (52% Leave, 48% Remain). The political consequences of this result have dominated British politics ever since, most obviously during the premiership of Theresa May, from 2016–19, but even after Boris Johnson won an election in 2019 on a promise to ‘get Brexit done’, the Withdrawal Agreement that led to Britain formally leaving the EU on 31 January 2020 can hardly be said to be the end of the matter. One of the fallacies of Brexit—and of course I write as a Remain supporter, like most British academics—is that it will ever be ‘done’, as each new government will be obliged to renegotiate aspects of the relationship, to deal with the continual list of problems surrounding trade, security, the border of Northern Ireland, and academic research funding, among many other issues (Baines et al. Reference Baines, Brewer and Kay2020; O’Toole Reference O’Toole2021). The reasons for the outcome of the vote will also continue to have consequences, insofar as they reveal significant divisions of opinion, of lifestyle and of ideology among sectors of the UK population. A significant component of these divisions is bound up with the symbolism of the past and its relationship to identity.

Before delving into these to explore something of the semiotics of identity in modern Britain, it is important to note that there are, of course, many reasons why people voted a certain way in the referendum—or, indeed, did not vote at all (turnout was 72.2%). Numerous surveys and studies have taken place in the aftermath of the result which shed light on various of the demographic patterns, particularly those relating to age, education and prosperity, but these intersect in some complex ways. It is broadly true, for example, that older people without university degrees were more likely to vote Leave, but these are related variables, as access to degree education has significantly widened since the 1990s (Dorey Reference Dorey2021; Dorling & Tomlinson Reference Dorling and Tomlinson2019, 21–31). Economic factors are complex, too, as while austerity is certainly a relevant part of the context of the referendum result, and a lot of media attention focused on northern English, working-class voters who did indeed also support Boris Johnson in the 2019 election, many working-class voters in London voted for Remain, and many relatively well-off middle-class voters in southern England voted Leave (Dorling & Tomlinson Reference Dorling and Tomlinson2019, 31–7). This compels us to bring in more cultural factors, both to do with everyday experience (more multicultural or monocultural communities, for example) and to do with perceptions of continuity and change in status and identity. This, in turn, is where the past comes in. While more ancient periods of British history were invoked as symbols of both sides of the Brexit divide, as will be discussed more below, the most dominant and encompassing theme, especially in the immediate aftermath of the referendum, was the British Empire. Crucially, the actual history of the Empire, from its very first stages in the English conquest of the British Isles and Ireland, through its global expansion, connections with slavery, relationship to emigration and immigration, up to the role of the Empire in the Second World War and its rapid collapse thereafter, all played a subtle but significant part in shaping the structures of British life and identity in 2016. At the same time, misconceptions and selective appropriations of this history were more overtly involved in the referendum and its aftermath, both as deliberate strategies and as part of the received attitudes of many voters. There is, therefore, in this situation a complex interplay of structures and agents, and symbols and practice. Theoretical tools which help us get to grips more firmly with it will be discussed in the next section, but first we need to exemplify the kind of dynamics at play more specifically, and also examine their relationship to intra-national identities.

The referendum result can be seen as the final stage in the collapse of the British Empire, or perhaps a hangover after that process, where an intoxicating cocktail of nostalgia for an imaginary past and amnesia about the realities of empire, and of the position that its collapse left Britain in during the 1960s and early 1970s, combined to motivate people to believe in the idea that Britain could detach itself from its nearest neighbours. There was a good deal of explicit reference to finding new fortunes by re-establishing trade with the Commonwealth countries, and other former colonies, in the immediate aftermath of the referendum (Olusoga Reference Olusoga2017). A memorable and revealing remark by Andrea Leadsom, one-time Conservative leadership contender, in a speech in October 2016, should be seen in this context. Leadsom, speaking in Paris, suggested that ‘tea, jam and biscuits’ should be central to Britain’s post-Brexit trade. The comments were ridiculed at the time (Smith Reference Smith2016), but the confusion they reveal about the sources of certain goods is actually a good example of the complex semiotics of British identity and its relation to empire. Tea is, of course, not a native crop in Great Britain, but a beverage imported from India and China, becoming popular in Britain from the eighteenth century (Alibhai-Brown Reference Alibhai-Brown2015, 143–4). The fact that it has become a powerful symbol of British identity, both internally and externally, is an example of the way that the process of Empire had a profound, transformative effect on a label and a grouping that, before the eleventh century, applied mainly to people living in what is now Wales. ‘British’ identity is, indeed, highly flexible over time, and was appropriated and expanded first by Anglo-Norman rulers of the English empire which conquered Ireland and Wales, and sought to do the same to Scotland, and then served as a tool of global empire. In this role, it was open to hybridization from the colonies, and at the same time enabled people from all parts of the Empire to identify with it, at least on some level (Colley Reference Colley2014; Davies Reference Davies2003). Indeed, the process of masking the central role of England in dominating the Empire, in terms of economic and political power, was so successful that not only did it create semiotic confusions like the symbolism of tea, but it left England with an identity crisis when the Empire fell away (Nairn Reference Nairn2021; Niven Reference Niven2019). Compared to Scotland, Wales and Ireland, England has few strong symbols of identity, authentic or otherwise. This explains two seemingly contradictory elements of the Leave campaign. On the one hand, nostalgia for the British Empire was clearly a part of the ideology of that campaign. On the other hand, many of the people who voted Leave identify now more as English than British (Dorling & Tomlinson Reference Dorling and Tomlinson2019, 51–7; Gardner Reference Gardner2017). The contradictions of England without an empire, and a lack of collective engagement with the realities of this situation, how it came about and what it means for the future, are a major element in the unhappy sentiment that drove Brexit.

This point brings in the further complexities of the semiotics of national identity in modern Britain, most of which are also connected to the history of empire. The symbolic capital of the non-English regions of the UK is in many ways more distinctive than that of England itself because the identities of those regions both had to be constructed to resist Anglicization and because part of the process of Empire-building is often the reifying and fixing of particular ‘ethnic’ characteristics of conquered peoples—or even, as we have already seen, their appropriation to a more pan-imperial level (Colley Reference Colley2009; Kidd Reference Kidd1999). This dual process was happening in Wales, Scotland and Ireland through the Middle Ages, as the first English empire was being created. From the early seventeenth century, as the project of empire became gradually globalized, some of these internal tensions were erased, or at least were purported to be. People from all of the ‘home nations’ took part in the global empire, whether at a level that might be said to still involve colonial exploitation, serving in British armies perhaps, or conversely benefitting from empire as industrialists, traders or civil servants. Yet due to the earlier history of the formation of ‘Britain’, England remained dominant, and the prejudice shown by English members of imperial society towards Irish, Welsh and Scots Britons continued alongside new forms of prejudice against people from other parts of the world (Dorling & Tomlinson Reference Dorling and Tomlinson2019, 38–74; Hechter Reference Hechter1975; Kidd Reference Kidd1999). The collapse of the British Empire after the Second World War, therefore, readily allowed those tensions to resurface. Growing pressure for a change in the status quo governing the Union was already clear in the 1990s, leading to the devolutionary moves of the Blair administration to attempt to stave off a greater crisis. The Scottish Independence referendum of 2014, arising out of a period of electoral success for the Scottish National Party, demonstrates that devolution only postponed the question. Scotland’s majority vote for EU membership in the 2016 referendum was a further manifestation of that direction of travel, as was the similarly pro-Remain vote in Northern Ireland and the more Welsh-speaking parts of Wales (Dorling & Tomlinson Reference Dorling and Tomlinson2019, 51–7; Kearney Reference Kearney1991; Nairn Reference Nairn2021). The other most pro-Remain areas in England were the major cities and university towns. In the case of the former, this aligns with the parts of England where ‘British’ identity is still more keenly felt than ‘English’ identity, because metropolitan populations are much more mixed and home to many second- or third-generation descendants of migrants from the former Empire. Given these complex and contradictory dynamics of identity in early twenty-first-century Britain, absolutely shot through with the legacy of empire, it is no surprise that symbolic use of the past has also become confused.

To bring the discussion back to a mooring in archaeology, two examples illustrate how the deeper history of Britain is played into these identity dynamics in ways which highlight the malleability of the past. The meanings of particular events, episodes or periods are slippery, and are only pinned down in particular interactions. Moreover, these selective interpretations of the past often compound the contradictions of post-imperial British identities. The Roman occupation of Britain has long been a challenging period to integrate into the story of the British nation (see e.g. Hingley Reference Hingley2000). Like several other European nations, but perhaps more acutely than most of these, British self-identity has struggled with the tension between resistance and accommodation to Rome and how these responses resonate with later imperial history, seen from the position of the colonizer, not the colonized. In a detailed programme of research on the use of the past in Brexit debates on Facebook, Bonacchi et al. (Reference Bonacchi, Altaweel and Krzyzanska2018) have revealed how the idea of the Roman Empire served both sides of the debate, though in somewhat counter-intuitive ways. With Rome typically equated with the EU, pro-Remain arguments often highlighted the (outdated) narrative of the Roman Empire as a civilized and multi-cultural entity, while pro-Leave supporters asserted an image of Roman occupation as violent and oppressive and of the Empire as doomed to fail because of uncontrolled immigration (Bonacchi et al. Reference Bonacchi, Altaweel and Krzyzanska2018: 182–5). This is interesting, not just because of the multivalent signification of a particular period and culture, but also because these meanings run counter to the attitudes to the British Empire typically held by pro-Remain and pro-Leave groups, as discussed above (Dorling & Tomlinson Reference Dorling and Tomlinson2019, 41–6). A similar set of confusing contradictions applies to a later event, also a well-known invasion of Britain, which appeared as a symbol on both sides of the political divide. The Norman invasion of 1066, and the Battle of Hastings, is perhaps the most famous event in British history (cf. Sellar & Yeatman Reference Sellar and Yeatman1930). The incorporation of Britain into a continental cultural network was again celebrated on the Remain side, with a local East Sussex group attending several of the many pro-EU protest marches after the referendum wielding a banner claiming ‘Hastings ♥ Europe since 1066’ (Fig. 1). Meanwhile, among the rising far-right activity that accompanied the referendum period, a subsequently proscribed group known as National Action were photographed with a banner referring to the ‘Spirit of 1066’ (Stewart Reference Stewart2016). This evokes ideas of Anglo-Saxon indigeneity and, interestingly, a ‘lost cause’ mentality that is very common among contemporary far-right groups in Europe and the US, and which bears some similarity to the ‘heroic failure’ motif which is more pervasive in British culture (Belew Reference Belew2019; Janney Reference Janney2020; Niklasson & Hølleland Reference Niklasson, Hølleland and Niklasson2023; O’Toole Reference O’Toole2019). These two examples highlight both the resonance that the past has in the present for the construction of identity and also the plasticity of the meaning of symbols of that past. Clearly, context shapes these meanings in particular directions, but in order to understand how that occurs, and how we might even begin to comprehend some of the processes of identity construction that happened in the deeper past, we need to investigate how symbols are invested with meaning in action. That is where a theoretical turn that archaeologists have hitherto neglected comes into play.

Figure 1.

Hastings ♥ Europe banner, pro-Remain march, central London, 2019. (Photograph: author.)

The semiotics of identity: a practice approach

The meanings of identity symbols may be fluid, but they are not inconsequential, as the previous section demonstrates. The way in which people draw upon the past in creating identities in the present is a fundamental aspect of the relevance of archaeology to the future. It will not be easy to escape this association with identity politics, despite the wishes of some archaeologists (e.g. Shennan Reference Shennan2002, 9–14; cf. Kohl and Fawcett Reference Kohl, Fawcett, Kohl and Fawcett1995), not least because the connection has been built up over the last century or more, and of course many of the most pernicious abuses of the past in the present are simply based on old archaeological interpretations (cf. Hofmann et al. Reference Hofmann, Hanscam and Furholt2021). Rather, we might learn from the complexity of the contemporary semiotics of identity and try to understand the social dynamics of identification more fully, in order to be more sensitive to the workings of these in the past. That, in turn, should inform our future interpretations to be less open to misappropriation. Clearly, simple theories of how past identities were built from normative representational characteristics have long proved inadequate, while recent evolutionary and ‘new materialist’ approaches reduce identification, in different ways, to an irrelevance—but in doing so, introduce a whole range of other problems (e.g. McGuire Reference McGuire2021; Van Dyke Reference Van Dyke2021). In finding a new way to approach identity, we might do worse than to excavate more fully the tradition which began to be explored in the 1990s, in the heyday of interpretative archaeology, before it was largely abandoned as the theoretical bandwagon moved on to new pastures of post-humanism. Siân Jones, in her book The Archaeology of Ethnicity (Reference Jones1997), was a key advocate of a practice approach to situational, multi-dimensional identities, alongside a few other publications around the same time (e.g. Fewster Reference Fewster2007; Lightfoot et al. Reference Lightfoot, Martinez and Schiff1998; Meskell Reference Meskell1999), but the social theory of Pierre Bourdieu and Anthony Giddens which underpinned much of this work was soon passed over. Archaeologists, collectively, barely even scratched the surface of this theoretical tradition, whose roots lie much deeper. Parts of the practice programme have persisted in overtly Marxist archaeology (e.g. McGuire Reference McGuire2008; Saitta Reference Saitta2007), but the foundational work of the American Pragmatist philosophers and their descendents in the school of Symbolic Interactionism (SI) has, with a handful of exceptions (e.g. Crossland Reference Crossland2014; Gardner Reference Gardner, Amundsen-Meyer, Engel and Pickering2011; Reference Gardner2012; McDavid Reference McDavid, Daniel Mouer, Holtorf and Karlsson2000; Preucel Reference Preucel2006; cf. Carreira da Silva & Baert Reference Carreira da Silva, Baert, Gardner, Lake and Sommer2014; Saitta Reference Saitta and Smith2014), been ignored. This is unfortunate, as there are few bodies of theory which address so directly the problem of the mobilisation of symbols of identity in action.

The Pragmatist philosophers formed only quite a loose school from the beginning, and subsequent generations even more so, but the core figures usually recognized are Charles Sanders Peirce, William James, George Herbert Mead and John Dewey, along with Jane Addams and Charles Horton Cooley. A diverse spectrum of topics is covered in the writings of these figures, ranging from the semiotics of Peirce—which has been advocated as an approach in archaeology by Preucel (Reference Preucel2006) and Crossland (Reference Crossland2014)—through the social psychology of James and Mead, to the socio-political and ethical writings of Dewey and Addams. What they share is a concern with overcoming the traditional dualism of materialism and idealism, finding a solution to that relationship in the primacy of action and practice, and advocating applications of philosophy to problems of progressive politics and democracy, which are just as vital now as in the heyday of these philosophers in the late nineteenth and early twentieth centuries (de Waal Reference de Waal2005; Joas & Huebner Reference Joas and Huebner2016; Musolf Reference Musolf2003). Various strands of influence flow out from these thinkers, including in neo-Pragmatist philosophy (e.g. Hilary Putnam and Richard Rorty), and more pertinently here in the symbolic interactionist tradition of social theory, developed initially by Herbert Blumer, a student of Mead’s, and later prominently featuring the hugely influential theorist of interaction, Erving Goffman (esp. Reference Goffman1959; Scott Reference Scott2015, 15–17, 82–111). In the later twentieth and into the twenty-first century, this classic body of social theory has influenced numerous works on identity theory, often alongside the slightly distinct lineage constituted by Bourdieu and Giddens, whose thinking was informed more by engagements with structuralism, Marxism and phenomenology, but with which pragmatism shares many common interests (e.g. Goff Reference Goff1980; Rosenthal & Bourgeois Reference Rosenthal and Bourgeois1998). The range of work developing the Pragmatist and SI traditions is considerable, but includes major theoretical syntheses on identity by Richard Jenkins (Reference Jenkins2014), Susie Scott (Reference Scott2015) and Kath Woodward (Reference Woodward2002), among a huge literature in various disciplines (e.g. Azmeary Ferdoush Reference Azmeary Ferdoush2018; Iossifova Reference Iossifova and Scott2020; Nicolini Reference Nicolini2012; Reckwitz Reference Reckwitz2002; Simpson Reference Simpson2009; Wenger Reference Wenger1998). In many ways this represents the most cogent theory of identity on offer today.

So what defines such an approach, and what might it offer to archaeology? Of most relevance to this paper is that it offers a processual, symbolic model of the acting self. Leaving aside the formal semiotics of Peirce, with its more elaborate and nuanced framework for meaning than classical Saussurean structuralism, the pragmatist notion of the self is predicated on humans being symbol-using beings for whom the self is a symbol. Practice mediates between actors and structures in the perpetual, ongoing unfolding of life, and within this flow each individual is also engaged in a constant dialogue between the self-as-subject and the self-as-object, the ‘I’ and the ‘Me’ in Mead’s terms (Jenkins Reference Jenkins2014, 58–61; Mead Reference Mead1934, 173–226; Wiley Reference Wiley1994, 18–73). This accommodates both the driving force of agency in the cluster of intentions and instincts of the ‘I’ and the socially shaped self that relates to other individuals and social groupings, largely through their symbolic representation, of the ‘Me’. This is, therefore, a relational approach but one which does not flatten the distinctive character of human agency. Rather, it situates it in particular social contexts wherein human agency takes particular forms in particular societies, with particular symbolic repertoires of personhood and identity (Shalin Reference Shalin and Ritzer2000, 339–41; cf. Barrett Reference Barrett2022a,b; Emirbayer & Mische Reference Emirbayer and Mische1998). These symbols are not abstractions, though, but constantly engaged in practice, which is the primary substrate of social life, and which determines their meaning in each context of action and interaction. What people do, and the way this mingles their own intentions and ambitions, with socially agreed conventions of appropriate or inappropriate conduct, gives shape to social life, to individual lives and to organizations and institutions. Practices thus become patterned, and insofar as these patterns enfold various forms of material culture, from spaces and buildings to portable objects, these patterns give structure to the form and distribution of that material culture (Musolf Reference Musolf2003, 117–98; Schatzki Reference Schatzki2010). The archaeological ‘record’ is thus primarily a record of past practices which were once the medium of people’s lived negotiations of identity, simultaneously of self and group, and mobilized as such because of the symbolic power of those practices within a world of knowledgeable actors (Barrett Reference Barrett2022b; Giddens Reference Giddens1984, 355–68; Scott Reference Scott2015, 4–11). The meaning of symbols is inherently polysemic and slippery because they are engaged in so many diverse practices over even a short span of time, and their meaning is constantly and recursively being created and negotiated. And yet, by virtue of the perpetual co-creation of these meanings in countless interactions within a social community—a community of practice—shared meanings take shape which have a degree of fixity to balance against this fluidity, and these are perhaps easier for an archaeologist to detect. Any period of the past contains a tension between these tendencies to fixity and fluidity, and both of these have consequences in the lived experience of individuals and in the shapes of structures and organizations. All of the foregoing discussion of identity in Brexit Britain bears this out, in the manifold and paradoxical symbolism attached to seemingly deep-rooted and ancient identities, which are simultaneously constantly reconfigured, such as Britishness itself. An example from later Roman Britain further shows how the enfolding of tradition and transformation in practices, in the context of imperial fragmentation, pre-empts some of the trends we see in more recent British politics.

Practices and cultural change in late Roman Britain

Late Roman Britain, between the fourth and fifth centuries ce, spans a similar period of imperial disentanglement, but of course in this case Britain had been at the edge of the empire concerned, rather than its centre. The semiotics of identity in this context are distinctive and particular, but at the same time the effects of an imperial system on the dynamics of idenification play out in some similar ways to trends we can see in the British Empire much later on. For example, just as in later colonies, local and regional identities which are selectively reified by the imperial administration from a rather more fluid pre-conquest situation come to be more and more important to people through the course of Roman occupation, which in Britain lasts for at least 350 years. This is evident both at the level of the civitates, the administrative districts into which the Roman province(s) of Britain are divided, and at the level of ‘British’ identity itself, which scarcely existed for people in the British Isles prior to the Roman period but did thereafter (Geary Reference Geary2002, 63–92; James Reference James1999, 90–109; Matthews Reference Matthews and Leslie1999; Mattingly Reference Mattingly2007, 520–28; Moore Reference Moore2011; Woolf Reference Woolf, Kaminski-Jones and Kaminski-Jones2020). ‘Roman’ identity was also subject to some of the same processes as ‘British’ identity was during the formation of the British Empire, shifting and transforming to become ever more hybridized as the wider empire, and particularly its frontier regions, shaped more and more of the symbols of being ‘Roman’. Contrary to the traditional ‘Romanization’ narrative of the Empire homogenizing the culture of its inhabitants, by the fourth century ce to be a Roman was to be many different things, but prominently included practices such as participating in Christian worship, from the eastern edge of empire, or wearing clothing derived from the traditions of the northern frontiers (Bispham Reference Bispham and Bispham2008, 225–33; James Reference James, Brody and Hoffman2014; Miller Reference Miller, Mathisen and Swan1996). Like later ‘British’ identity, this flexibility was perhaps a necessary feature of a long-lasting imperial culture. It was also one that had limitations, though, in how much local diversity could be accommodated (Kearney Reference Kearney1991; cf. Given Reference Given2004, 52–68; Woolf Reference Woolf2012, 233–53). Just as in more nakedly political terms, the constellation of identities within an imperial system is vulnerable to fragmentation, and in late Roman Britain we see some of the dynamics which drive that process playing out.

A good example which illustrates both the simultaneous fluidity and fixity of the meanings of identities, their negotiation in practice, and also the generative role of boundary interactions (cf. Gardner Reference Gardner2022a) is the development of a new writing technology in the western frontier zone of Roman Britain, very late in the period of Roman occupation. The modern country of Wales constitutes an expansive frontier region in the Roman period, facing the Irish Sea and routes of trade and other interaction with the polities of Iron Age Ireland. Roman military campaigns in Wales were protracted and difficult, lasting from the late 40s ce (just four years after the Claudian invasion began) to the early phases of the governorship of Agricola in the 80s ce. A relatively dense network of forts and roads established in this period lasted until the Hadrianic period, the 120s ce, but from then on the garrison network was reduced as the majority of Roman forces moved to the north. However, through the rest of the Roman period—nearly another 300 years—some military presence remained in forts like Caernarfon in the northwest, and at the two legionary fortresses of Caerleon and Chester, in the southeast and northeast of the region respectively. In the later third and fourth centuries, indeed, this network was supplemented with a new fort at Cardiff and a number of small coastal bases. Apart from a couple of towns at Caerwent and Carmarthen, and the emergence of some villa farms in the south, the settlement pattern across much of Wales remained relatively unchanged by Roman occupation. It thus bears similarities to the hinterland of Hadrian’s Wall in the north, a zone of continuing military control and economic under-development (Gardner Reference Gardner2022b, 165–7). This does not mean, though, that nothing changed in the lives of the inhabitants after the imposition of military infrastructure. To the contrary, across Wales and the Irish Sea zone, new cultural signifiers and new opportunities to use them came into play as time went on. The third and fourth centuries ce, in particular, were a time of considerable interaction across the Irish Sea, and this produced changes visible in the archaeological record on both of its shores. In Ireland, there is more enclosed settlement in this later period and more items of Romano-British material culture, particularly metalwork, making its way into a variety of contexts on Irish sites (Cahill Wilson Reference Cahill Wilson, Cahill Wilson, Dowling, Bevivino and Barry2014; Dowling Reference Dowling, Cahill Wilson, Dowling, Bevivino and Barry2014; Waddell Reference Waddell2010, 395–401). The contacts via which objects moved might have included trade, raiding, migration and military service, or probably a mixture of all of these (and more)—part of the typical pattern of multi-valent frontier interactions which had impacts on both the imperial power and those on its edge (Gardner Reference Gardner2022a; cf. Ferguson & Whitehead Reference Ferguson, Whitehead, Ferguson and Whitehead2000). Innovations in practices, and their semiotic function, are to be expected in such situations.

In the Irish Sea zone, one such innovation in particular stands out. This region is known for its megalithic stone monuments in prehistory, while in Roman Wales, there was relatively limited participation in the habit of erecting inscriptions on stone, outside small numbers of public and funerary commemorations in the major settlements. Yet, in the late fourth century and increasingly from the fifth to the seventh century, inscribed stone monuments using a new script system appear on both sides of the Irish Sea. There are around 150 of these in Wales, concentrated in the southwestern and the northwestern peninsulars; many are in Latin, some in the new Irish script, Ogham, and some in both (Fig. 2). In Ireland there are about 360 inscribed stones, using Ogham, and mainly found in the southeast (Edwards Reference Edwards2001, 15–18; Handley Reference Handley, Hill and Swan1998, 339–41; Mytum Reference Mytum1991, 30–35; O’Sullivan & Downey Reference O’Sullivan and Downey2014). The Ogham script, with 20 letters comprising various horizontal and diagonal lines around a central stem, is well suited for the practice of carving along the edge of an upright stone. If the Welsh examples are taken as representative, the use of this system is part of the changing religious milieux of the late Roman west, with Christianity and associated burial practices becoming visible by this means. Many of the inscriptions are personal names and are treated as memorials, though functions to do with land-claims and boundaries might also be served (Edwards Reference Edwards2001; Johnston Reference Johnston2013, 9–16; O’Sullivan & Downey Reference O’Sullivan and Downey2014; Stevenson Reference Stevenson1989). The significance of these stones in the semiotics of identity of this region at this time is considerable, and they illustrate the tendencies towards both fixity and fluidity already discussed—the inherent malleability of identities, on the one hand, and conversely the desire to make them more permanent. An innovative writing-system produced through interaction across the sea between the Irish-speaking world and the Latin-inscribing world is made manifest on the most substantial, ancient-seeming of media, large upright stones embedded in the landscape (cf. Williams et al. Reference Williams, Kirton, Gondek, Williams, Kirton and Gondek2015). They are used to give weight to and fix the place of individuals in a zone of population mobility and cultural change, with the collapse of the Roman provincial structure in Britain at the beginning of the fifth century and the emergence of new power structures in what became Wales: new British identities being formed in the wake of Rome, but still drawing upon its symbolism. The fusion of traditional and new practices is thus illustrative of the mixture of both in the identities of the people living in this time and place, and shows the simultaneous embracing and defiance of change that we see in the continual forging and re-forging of British identities down to the present day. The benefit of a specifically Pragmatist/SI approach to these processes is that it attunes us to the significance of these practices, these choices and these symbols in the shaping of both agents and structures in this context. The frontier region of the Irish Sea, like others in Britain and across the Empire, was always dynamic and characterized by a shifting mosaic of opportunities and constraints for soldiers, the wider military community, and other inhabitants on both sides of the boundary (cf. Gardner Reference Gardner, Tanner and Gardner2024). As imperial authority waned, the structures connecting the region to the rest of the western Empire receded, but the local groupings that the Empire had fostered and which had been reified by administrative practice over time gained importance, providing one form of anchorage to the past—as evinced by the use of the name of the civitas of the Ordovices on the fifth-/sixth-century Latin-inscribed Penbryn stone (Edwards Reference Edwards2007, 184–8). The actors engaged in erecting such monuments, and their intended audiences, sought such continuities at the same time as they were literally engaging in new forms of expression and new power relationships. This interplay of creative enactments of continuity is manifest too in other frontier regions in Roman Britain, such as in the repeated communal feasting activity evident at some of the northern forts such as Binchester, south of Hadrian’s Wall, in the same period, intended to cement local relationships within established places of power at a time of transformation (Petts Reference Petts2013). In both cases, actors—themselves situated in different, and changing, structural positions—used resources from the past in innovative ways to try to shape the future. And in both cases, the critical step for the archaeologist is to think in terms of practices—open-ended and dynamic activities—rather than static objects or their distributions. The latter are only the means to the end of understanding what people did to influence their own present moment when the future could always be otherwise (cf. Giddens Reference Giddens1984, 9, 14–15), through the medium of practices.

Figure 2.

Latin/Ogham ‘Vortipor’ stone, Carmarthenshire, ?sixth century (Macalister Reference Macalister1945, 342).

Conclusion: meaningful identities

The two examples that have been briefly sketched out in this paper show how the semiotics of identity are unstable and malleable, with symbols having manifold and contradictory meanings. They also show, though, how those meanings gain at least temporary stability and power, as they are mobilized in practice. Indeed, identity labels and their symbolism can be extremely long-lasting, as with many of the cultural groupings across the British Isles, but they are always being re-contextualized and re-invented as part of the ever-changing constellation of practices that makes up the social fabric of people’s everyday lives. The diverse ideas of ‘Britain’ being negotiated through the Brexit process bear little similarity to those of the Roman period, except insofar as those were, also, inherently diverse. The strongest connecting thread between the two periods illustrated in this paper’s case-studies is actually the way that empire plays a major role in the structuring of identities in each time-frame. ‘British’ identity is a creation of one empire, an artefact of Roman imperial control at its edges and already transforming right as that control slipped away in the late fourth and early fifth centuries, open to innovation on the frontier. It was then transformed again, many times, through the long period of English and then British imperialism, becoming itself an imperial identity, as ‘Roman’ had been before, and now also susceptible to transformative pressures from the edges of that empire, ultimately leaving it adrift of symbolic connections to England as the British empire, in turn, collapsed. Through all of these developments, it is the semiotics of practices, from inscribing names on stone to drinking tea to campaigning for or against Brexit, that gives shape and meaning to these identities. As actions in the material world, these practices necessarily involve material culture and, moreover, create patterning in that material culture. Understood as a record of practices, archaeological evidence therefore gives us much to work with in trying to reconstruct the identitiscapes of the past.

Indeed, such an endeavour is more necessary than ever. As the pendulum of archaeological interests has swung once more towards the scientific, generalizing end of the spectrum with the ‘third science revolution’ (Kristiansen Reference Kristiansen2014) and innovations in ancient DNA and isotopic studies of human ancestry, relatedness and migration, so the risk of falling back into normative stereotypes of identity has grown significantly (Wolinsky Reference Wolinsky2019). Some would see post-humanist approaches as offering a way out of this dangerous territory (Crellin & Harris Reference Crellin and Harris2020), but the attribution of objects with human-like properties was also a hallmark of the culture-historical era and so, like others, I would prefer to focus on the potential of practice approaches to break down simplistic relationships between things and identity without abandoning the embeddedness of these processes in very human worlds of power and inequality (Gardner Reference Gardner2021; cf. e.g. McGuire Reference McGuire2021). Symbols of past identities, or at least objects, images and places which are treated as such, are increasingly mobilized in the political practices of many groups in the present, but particularly those on the far right with violent, exclusionary and authoritarian visions of the future (e.g. Niklasson & Hølleland Reference Niklasson, Hølleland and Niklasson2023). If archaeology wants to retain both its capacity to critique abuses of the interpretation of the past and also to engage in progressive political action (cf. e.g. McGuire Reference McGuire2008), then challenging these mobilizations of identity in present practice means effectively understanding their dynamism in past practices. The identities that people cleave to in the present were never as permanent, unchanging, pure or homogenous as they are often imagined to be. They might have been fixed in meaning for a time and in a place, but always remained open to change as their use in practice changed. To be able to demonstrate the flexibility of human social processes over the long term is one valuable contribution that archaeology can make to an uncertain future.