How can we grasp the effects of memory politics on the quality of peace? How can we instil peace with meaning so that it can be analysed empirically? This chapter lays out our comparative methodology for analysing memory politics, elaborates on the idea of mnemonic formations, and introduces an analytical framework for investigating mnemonic formations through four conceptual entry points: sites, agents, narratives and events.

With the unfolding of the so-called memory turn in the social sciences over the last couple of decades, a rich body of literature on individual case studies in and of conflict-affected societies has emerged. So far, however, there has been little research that attempts to draw out some generalisable observations across different cases and across time and space concerning the relationship between memory politics and peace. The analytical framework elaborated in this chapter allows us to take this step and compare across cases, look for patterns, similarities and differences, and to systematise and assemble emerging knowledge.

As outlined in the Introduction, mnemonic formations are thematic clusters of particular salience around which sites, agents, narratives and events gather and thicken. Mnemonic formations provide a multidimensional link between memory politics and peace, functioning as diagnostic sites; it is through interrogating these that we discern patterns of memory politics that allow us to understand better the relationship between memory politics and the quality of peace. While we search for patterns, we are cognisant of the fact that any comparative endeavour inherently and necessarily poses ‘a threat to replace thickness with universal concepts or standards’ (Niewöhner and Scheffer, 2010a: 533). Inspired by ethnography’s key strengths, that is, detailed, nuanced and thick description à la Geertz (1973), we use soft comparison, allowing the cases to speak to each other by combining within-case analysis with cross-case comparison (George and Bennett, 2005). There is thus a balance between the depth of interrogation of a case and the comparability between cases, allowing for insights to be generated within cases, as well as teasing out what this can mean conceptually, through comparison. There is of course a trade-off between, on the one hand, the depth of contextualisation necessary to describe a phenomenon adequately in order for its meaning to be rendered understandable, and on the other the generality necessary for the concept to be applicable in other contexts. Our point of departure resembles comparative ethnography (Simmons and Smith, 2019: 341–2). We are therefore interested in the specificities of each of the five cases of mnemonic formations not only in and of themselves, but also with regard to how the insights from each individual case stand in relation to the other cases.

In line with the idea of thick comparison, the object of comparison is ‘produced through thickening contextualisations, including analytical, cross-contextual framings that are meant to facilitate comparison’ (Niewöhner and Scheffer, 2010b: 3, emphasis in original). Yet with what degree of precision should this thick comparison be drawn (Prus, 2010: 502)? The framework developed in this chapter enables a nuanced, thick description of individual cases as well as subsequent cross-contextual comparison by reading the mnemonic formations through the four conceptual lenses of sites, agents, narratives and events. While the optimal level of abstraction (see Sartori, 1970) is possibly unclear in general in ethnographic comparison (Prus, 2010: 503), we believe the framework we propose strikes the right balance for the study of mnemonic formations. It allows us to draw wider conclusions about how the quality of peace is intrinsically connected to the ways in which memory is configured and entangled, moving beyond the particularities of each case and enabling concepts and insights to be extended across cases and to generate theory.

Five different cases of mnemonic formations are explored, derived from Cyprus, Bosnia and Herzegovina, South Africa, Cambodia and Rwanda. The conflict-affected societies focused on are indeed diverse, and each displays its own unique historical trajectory and conflict dynamics. Nevertheless, they all share the characteristic that a violent past continues to be a divisive issue, with memorialisation emerging as a potential tool for building peace and/or perpetuating conflict. There are contestations around the ways in which the past could and should be remembered which translate past violence into present conflict. Importantly, the selected mnemonic formations are situated within a wider societal context of a number of topics that are commemorated – as well as some that are not. It is our close familiarity with these cases that allows us to select significant mnemonic formations within this broader context.

Reading mnemonic formations: sites, agents, narratives and events

The framework for the five case studies brings into play the theoretical traction of the conceptual lenses of sites, agents, narratives and events. We posit that the dynamic and frictional characteristics of peace processes are not fully captured by separate studies of these key elements and that we need to study their mutual constitution.

Through the emphasis on sites of memory, such as memorials, monuments or museums, we are able to capture that ‘matter matters’ for memory politics. Since memory politics is a process with, and a result of, the work produced by individuals and organised groups, it is crucial to acknowledge the role of agents in our understanding of mnemonic formations. Given the power of language and discourse and its substantial effects on societies, we bring to the fore narratives of memory. A focus on memory events, lastly, recognises the importance of performativity in effecting social relations. All these conceptual entry points are combined in the SANE framework, which can be systematically applied to either individual or comparative cases.

In what follows, the four interrelated elements are unpacked to bring out their relevance for a systematic analysis of mnemonic formations and how they relate to the quality of peace and the potential for a just peace. For the sake of heuristics, they are here analytically separated, but we understand them to be in constant productive interaction with each other.

Sites: pinning memory politics to place

In the most general sense, sites are places where something is commemorated. They may take the form of cemeteries, memorials, museums or artefacts. They fix meaning to physical locations and material structures, and pin memory politics to place (see Buckley-Zistel, 2020; McDowell and Braniff, 2014). Sites are, in our understanding, material representations at particular locations (Björkdahl and Buckley-Zistel, 2016; Kappler, 2017). Sites become invested with a particular meaning through social practice and by being tied into narratives about past events, present conditions and future aspirations. They are thus important nodes for memory politics as they enable a group to congeal around them (Halbwachs, 1992: 204). To make the link to the past tangible in the present and to give it a sense of permanence, groups tend to produce topographies of memory so that sites not only frame memory but ‘situate and spatially constitute group remembrance’ (Till, 2003: 291). Hence, sites where memory takes place are not per se places of memory; rather, it is through the social practices of place-making that they become meaningful for memory, politics and peace.

At the same time, geographic locations of atrocities are often turned into sites of memory. Keeping alive the ostensibly authentic aura of an original site of atrocity – often despite decay and degeneration – is part of memorial aesthetics and has a strong emotional impact on visitors. Some sites are presented as museums or memorial museums (often at ‘authentic’ places such the Tuol Sleng Genocide Museum in Cambodia), with household objects, texts, videos and/or other displays to educate visitors about the past. These locations may appear as physical scars on the landscape, in the form of mass graves, buildings or areas previously used for confinement, torture or execution, or ruined religious buildings. They may include traces left by explosions, or remnants of divisive walls and crossings. To mark these locations and to invest them with meaning, museums and memorials are built, commemorative events held or burials performed, providing an important reference point for narratives and for the construction of collective identities. The materiality of objects at memorial sites, moreover, shapes collective identities, communicates particular norms and values, and transmits certain narratives about the past (Beckstead et al., 2011: 194). Here, too, social, material and spatial elements mutually constitute each other, in that people may construct memorial objects but at the same time their identity is (re)constructed as they engage with these objects. Memorial sites contain information and produce meaning in a metaphorical way, but also in a highly material way (Buckley-Zistel, 2021). They may serve as archives of facts and artefacts, or as storage spaces for documents and images and sometimes also for victims’ clothing and bones, as at certain sites in Rwanda and Cambodia. Here, ostensible authenticity is linked to truth and material evidence, conveying a powerful narrative about the horror of the violence. Other sites consist of objects that are highly symbolic and culturally situated, such as statues, crosses, Buddhist stupas, abstract figures or walls filled with names, which may all be very meaningful to visitors.

The focus on sites of memory allows us to explore the social construction of sites and how they are linked to the efforts or needs of particular groups to construct a collective identity. A spatial analysis enables us to explore what effect sites have on particular groups and how different groups may compete over the interpretation of a site and its narratives. Sites, to us, are thus not fixed topographic entities but constructed locations and the result of place-making, where place-making refers to ‘the set of social, political and material processes by which people iteratively create and recreate the experienced geographies in which they live’ (Pierce et al., 2011: 54).

Beyond what the eye can see, memorial sites are affective by making us ‘feel’ the past (Buckley-Zistel, 2021; Tolia-Kelly et al., 2018: 3). Their display and design might trigger emotions such as sadness, a sense of loss, despair and despondency. Even though this is always personal and may differ widely from visitor to visitor, it is by no means random; it is situated in the wider commemorative practices of a society and the way a society remembers. These emotions are often beyond representation and cannot be captured by words. They expand the effect of these sites from what is visible to what is invisible, or from what they mean and what they do (Waterton, 2014).

Some sites are not formally marked but still form part of the cultural heritage of conflict. They may entail empty yet meaningful spaces, voids where individuals, families or communities once lived; places of pain where terrible things happened yet the crimes committed are collectively denied by new inhabitants. Rape camps may not display any acknowledgement of the past crimes. Physical gaps may remain in the townscape after the erasure of religious buildings. These voids are not meaningless, yet they may disrupt hegemonic remembering and forgetting. One particularly interesting example that we analyse is the so-called Sarajevo Roses, a citizens’ monument in the shape of red colourings on the city’s pavements and streets. These mark the craters left by mortar shells that killed residents during the siege of the city and are present in the cityscape for people to engage with or not in their everyday comings and goings (Björkdahl and Kappler, 2017). In that sense, people’s movement across sites of commemoration, their reluctance to cross bridges or their determination to cross into areas of the other can be important markers of how they engage with the landscape of memory. Such a memoryscape necessarily carries the past into the present, where it can be re-engaged with in various ways.

Where and how memory is manifested spatially has a strong influence on how it is politicised and who can access it (Buckley-Zistel and Schäfer, 2014a). For instance, memorial sites may be erected to obtain some recognition, to draw attention to past atrocities and to serve as symbolic reparation. For victims, they become moments of assertion of their rights – rights they were deprived of in the past – giving them back at least some form of agency and dignity. For many, having their voices heard may be of great personal and collective significance. At the same time, these sites may be used for personal, quiet mourning (Viebach, 2014). Their material, style and form may provoke sentiments such as sadness, intimidation, despair or claustrophobia, turning them into places where heritage ‘hurts’ (Sather-Wagstaff, 2017; see also Mannergren Selimovic, 2020).

Agents: exercising memory politics

Memory politics after war and violence is driven by multiple agents with varying capacity to impact on the memoryscape. Agency is essentially about a person’s or a collective’s capacity to act, and is thus always about power relations. To exercise agency is to bring about an effect of some sort on the world through one’s actions or, in the words of Anthony Giddens (1984: 9), agency is present if an ‘individual could, at any phase in a given sequence of conduct, have acted differently. Whatever happened would not have happened if that individual had not intervened’. However, this agency is not exercised in a vacuum but in a spatial and temporal context that can enable and disable agency in various ways (Björkdahl and Mannergren Selimovic, 2016a; Giddens, 1984). Agency is further constructed in a social world that shapes the opportunities and resources available, in a constant interplay of practices and discourses (Ahearn, 2001: 112). We thus locate agency in the intersubjective relations between people and groups, rather than as a possession contained within individuals. To exercise agency is thus to engage in activities that form fields of relationality (Ahearn, 2001: 130; see also Arendt, 1985), involving varying degrees of friction.

Agency is essential for the making of the world that we inhabit. Its construction is mediated by social relations that provide ‘a shifting set of possibilities that ha[ve] as much to do with objective realities as subjective and intersubjective understandings of changing conditions and pressures’ (Fujii, 2009: 18). These shifting constraints or enabling opportunities are key to understanding agency at different points of time because they condition if and how memory agents can influence the memoryscape. When agents are constrained in their mnemonic agency, their narratives can become silenced and they may be absent from sites deemed meaningful by them. At the same time, though, constraints on agency may shift in time and absences and silences can ultimately be overcome by agents, affecting the structure of the memoryscape accordingly.

Even if not all memory agents’ voices are equally heard or perceived, their practices not equally visible or their memories not afforded the same degree of legitimacy, in any memoryscape we can identify a whole host of different agents. They may have considerably diverging levels of power and understandings of the past, but they share an interest in shaping how that past is remembered. The array of different agents participating in memory politics may include politicians, victim associations, museum curators, international peacebuilders, media and tourist entrepreneurs, representatives of the international community and ordinary citizens. These agents may be formal or informal; they may be organised in a coherent, professional manner or they may collaborate in a less official way. Furthermore, while agency is located within social relations, the agents of interest here can be both collective and individual. While individuals can sometimes speak on the behalf of groups and even be endowed with a high level of legitimacy, it often happens that collective agents are able to position themselves more prominently in the memoryscape. We can find both elite or grassroots agents, with the former strongly embedded in state and civil society structures, while the latter can challenge this – for instance by developing a certain agency of resistance through practices of broader social mobilisation.

Agents can be located at a local, national or international level yet transgress these scalar levels (Björkdahl and Kappler, 2019). In any given mnemonic formation global agents – various UN agencies such as UNESCO, as well as peace-building agencies – are frequently present and influential, as memory work becomes increasingly global and transnational. The international is also represented in a more transient sense by tourists. While tourism entrepreneurs are important in shaping the memoryscape in a way that is economically viable for their businesses, tourists themselves as agents are meaningful above all through their presence or absence at specific sites (Buckley-Zistel and Williams, 2022). The importance afforded to a certain site can legitimise certain narratives over others or expand the agency of particular individuals or groups.

Besides this multitude of memory agents, there are ideas on how agency can have a mnemonic effect beyond human agents, even though any agency attributed to non-humans remains controversial. Most frequently this is discussed with regard to the agency of objects (see, for example, Lindstrøm, 2015, 2017; Ribeiro, 2016; Sørensen, 2016, 2018). Proponents of this idea suggest that materiality can also possess agency, in the sense that objects participate in processes and affect social outcomes and thus one could argue that they are imbued with agency (Feldman, 2014). This is particularly pertinent as objects – like people – have different agentic capacities in differing (cultural) contexts, structuring the relationship between the individual and the object in ways that allow agency to unfold. Others have suggested that the most useful way of thinking about object agency is with respect to their emplacement in biographies (Hoskins, 2006). This is a particularly useful suggestion in the context of this book, as events of remembered violence can be tied to specific parts of an individual’s life and associated with specific material objects. The affective personal objects on display in the War Childhood Museum in Sarajevo are a powerful example. However, any rendering of objects possessing their own intentionality is controversial, as it weakens the conceptual clarity of the term agency and its differentiation from concepts such as effect (see Lindstrøm, 2015). This is in line with our approach to the relational nature of agency, as discussed earlier in the chapter. We adhere to the idea that objects are imbued with agency primarily through the mnemonic significance they assume for certain individuals or the meaning they have in social relations.

Agency is by no means fixed over time, but we can see a constant struggle for, and reconstitution of, agency in social relations. Agents participate in an on-going making and remaking of the memoryscape, engaging in memory work both in formal settings and in the everyday. While agency exerted in the construction of a site may be seen as fixed, the everyday work of upholding the meaning of the site, as well as the maintenance of the site, requires agents’ on-going commitment (Viebach, 2014). Moreover, the organising of events requires agency, and in the case of recurring events this requires on-going agency so as to allow the memory work to be performed and repeated.

It is in the upholding of such structures and compliance with them, or resistance to them, that the limitations of agency become visible. We recognise the fact that agents are to a lesser or greater degree embedded in relations of power and interdependence (MacLeod, 1992: 533–4). The analyses of agency must therefore also look at its constrained and restricted nature in order to understand when there are limitations on a full exercise of agency; for example, gendered notions of acceptable practice or gendered access to spaces and voice might be inhibitive for women (Björkdahl and Mannergren Selimovic, 2015).

And yet we find that it is important to recognise not only reactive but also proactive agency, which exercises power of initiative. In other words, the ‘ability to act in an unexpected fashion or to institute new and unanticipated modes of behaviour’ (McNay, 2000: 22) is crucial in understanding possibilities of change. A search for agency and agents in unexpected, ignored or hidden spaces discloses that transformation can take place even if it does not assume overt, organised forms. The creative dimension of agency hints at the shortcomings of the rationalist and determinist notions of agency; it alludes to the expressions of agency that do not reify or reproduce but which instead challenge structures and make something new.

Narratives: making meaning and coherence

Narratives are texts or speech acts that create particular meanings. In addition to spoken or written language, narratives can be communicated through exhibitions, symbols, and other spatial/emplaced and performative expressions. They can be produced and recounted through material objects and through sites, or expressed through commemorative events. They may be visually, sonically and viscerally experienced through performances and artefacts functioning as manifestations that the narrative can be pinned on, such as a bench, a museum photograph, a video recording or the displayed bones referred to earlier in the chapter.

Historical events as well as experiences of everyday life are turned into stories that can be told and retold, yet also changed. Narratives thus ‘impose a structure, a compelling reality on what we experience’ (Bruner, 2002: 89). Telling a particular narrative in a particular way is constitutive of identities, both individual and collective. These narratives connect the past with the present in that they refer to events and experiences of the past but are told and retold in the present. We thus understand narratives not as a reflection of reality ‘out there’, but as a meaning-making instrument. To explore memory narratives in the context of the framework we have adopted is to ask what is distinctive about a particular narrative, why it is narrated in this way and not another, and how it helps people make sense of their world. At the same time, we recognise that dominant narratives especially are rarely presented in a pure form. They are mediated, edited, translated and curated (Fernandes, 2017). They must, therefore, not be viewed in isolation from the power structures that produce and promote them (whether those cement a hegemonic or a transformative narrative). Some narratives will gain more traction than others, depending on the power of the narrative agents. Some become more or less ritualised and stabilised in society, while others may be prevalent but tend to be expressed and circulated in the everyday and through informal practices. The state has an unmatched capacity to shape narratives about the past by employing the various tools and resources at its disposal (Wertsch, 2008: 128). It has the power to institutionalise collective narratives through textbooks, statues, museums and other inscriptions in public space. The divided memoryscape of Cyprus is a case in point, demonstrating the power of the state in shaping mnemonic narratives. In Rwanda, moreover, the state has almost exclusive authority over the way the country’s past is narrated and the events around the genocide interpreted. Nevertheless, as we also see in the deeply divided societies that we study, other collectives can create their own narrative realms below or parallel to that of the state. In this process, some stories and experiences are deemed irrelevant, shameful or even dangerous, so that they become marginalised and silenced. Denial and revisionism are the most blatant expressions of silencing in post-war societies, but it can also take more subtle expressions. When it comes to remembering victims (with ensuing acknowledgement and reparations), some crimes are given more importance than others. Moreover, even though many victims experience multiple harms, memory work tends to construct narratives that concentrate on select, easily defined crimes while silences grow around other pains. Indeed, sometimes the narrative power of silences can tell us at least as much as public vocalisations (Eastmond and Mannergren Selimovic, 2012; Mannergren Selimovic, 2020). Conflict-related sexual violence, in particular, is a crime often shrouded in silence, for multiple reasons. By being cognisant of narrative silences in memory work we can trace how such forms of marginalisation have consequences for the quality of peace. The degree to which such largely silent narratives are entangled with the wider mnemonic formation is telling for their particular role in the construction of peace.

With respect to the politics of memory, the power of narrative to shape our understanding of the past and of ourselves as a group is well documented (Bell, 2006: 5; Buckley-Zistel, 2014; Hammack and Pilecki, 2012; Wertsch, 2008: 122). Collective mnemonic narratives often aim to uphold boundaries between a collective us and a collective them, and attempt to weave plural and heterogeneous experiences and expectations within a particular group into a coherent story. Individual memories can thus merge into larger cultural and social frameworks that follow some generalised patterns (Wertsch, 2008). In Hannah Arendt’s thinking, storytelling is what connects the individual experience to the public sphere at the centre of politics. Individual experiences are ‘transformed, deprivatized and deindividualized, as it were, into a shape to fit them for public appearance’ (Arendt, 1985: 50). Narratives thus produce certain generalised templates for remembering that bring meaning and coherence as they powerfully narrativise ‘the experience of the social category to which the individual belongs’ (Richardson cited in Elliott, 2005: 13). Some narratives bring forth commonly used tropes such as the role of victimhood in constructions of guilt/innocence that centre on ‘chosen traumas’ (Volkan, 1991) – a concept that suggests a group remembers a traumatic event in order to reproduce its collective identity. Another key trope is the idea of ‘never again’ which, originally coined by Holocaust remembrance, attempts to create narrative moral coherence by drawing a line between an atrocious past and a more peaceful present and future.

Crucially, while collective narratives may provide meaning for some, they are often juxtaposed with competing narratives that have a very different understanding of how to make the past, present and future morally coherent. Negotiations between competing stories are always on-going, as individuals attempt to make sense of and narrate their own experiences of trauma, upheaval and fragmentation. Individuals or collectives without the power to hegemonise the discourse may embrace, resist or transform collective narratives, sometimes (partly) producing and using them, sometimes (partly) rejecting them (Whitebrook, 2001: 10). These other voices may be difficult to hear, and we may have to listen more carefully and be prepared to listen to silences, as some agents who have little power over the collective narratives that dominate may choose silence as a way of navigating through polarised environments. To understand more fully the narrative work of memory politics, it is therefore important not to listen only to the ‘privileged story-tellers … to whom narrative authority … is granted’ (Campbell cited in Milliken, 1999: 236).

Given that narratives bring the past, present and future into a meaningful order, we are interested in detecting how memory work is inherently non-static and constantly evolving. Our case studies therefore look at narrative developments over time. For example, we explore how the narrative of South Africa’s collective past has moved from a near-total exclusion of colonialism to a tentative inclusion, uncovering the connections between the country’s colonial past and its apartheid era. This temporal dimension of narrative is also evident in the frictions between memories of Sarajevo pertaining to the last war and the present-day city, and how negotiations between these different identities affect the possibilities for a just peace.

Events: performative mnemonic practices

We understand events as meaning-making performative mnemonic practices that forge collective identity. They are not permanent but temporary, even though they might be planned, regular and repeated at particular intervals. They may be rooted in one place – often in relation to a memorial – or may lay claim to wider spaces, as in the case of peace walks or parades. They may be solemn and quiet, as with the reburial of excavated bones, or violent and loud, like a riot. Importantly, these mnemonic practices are assigned significance that is recognised by relevant agent groups. In this book, we are interested in events of remembering and commemorating past atrocities. Importantly, these atrocities can themselves be referred to as events: the event of the Rwandan genocide, the event of the siege of Sarajevo, and so on. We are not examining the remembered event, however; instead, we are interested in remembering as an event. This analysis recognises that events are part of a wider memory structure which they constantly reproduce (Wagner-Pacifici, 2015) yet potentially also challenge. Mnemonic practices only become relevant when they are infused with meaning which they (re-)produce through performance. They may involve some form of embodied practice such as gestures, movements or articulations enacted by an individual or several people, or an installation or art intervention. Events can take the form of parades, ceremonies, public protests and burials, special media broadcasts of a public ceremony, tourist tours, peace marches or national days of commemoration. They may be highly ritualistic or ephemeral and fleeting as people come together for political action. They are visual and may involve the display of flags, coffins and other material objects. Recurring events become rituals – as happens with anniversaries or holidays – that help maintain mnemonic continuity.

The event of remembering is intrinsically linked to a remembered event, as is the case with memory more generally. The event as a performative act therefore ‘rehearses and recharges the emotion which gave the initial memory or story embedded in it its sticking power’ (Winter, 2010: 12). Atrocities are commemorated in the present to mark their significance and to reinforce particular narratives about this past. In this sense, ‘[e]‌ach commemoration focuses on an event and the events are linked into a story whose meaning lies in the whole, rather than in any commemoration (or the historical event commemorated) by itself’ (Papadakis, 2003: 253–4). For instance, in Cyprus, Greek Cypriots celebrate Greece’s War of Independence from the Ottoman Empire by observing Greek Independence Day on 25 March. Here, a war that began in 1821 is a past event transported into the present by a national holiday.

In memory politics, events are frequently repetitive, as with annual commemorations, lighting candles or laying wreaths, and are embedded in, or borrow from, a long symbolic tradition (which often has its origin in religion). They often have the character of a ritual, including activities that have symbolic meaning and follow certain rules; they circulate around objects of thought and feeling that are of great value to participants (Lukes, 1975: 291). Drawing on shared objects or performances helps to form and sustain deep emotional bonds among those taking part or watching (Etzioni, 2000: 45). These kinds of rituals are therefore a crucial mechanism for the recreation of a collective identity for victim groups, citizens of a nation and the international community. As mnemonic practices, events can be understood as predictable occurrences that build on foreknowledge and previous experience, with the aim of reinforcing certain narratives of collective memory and re-establishing individuals’ bond to the group as well as to the narrative (Drozdzewski et al., 2019: 261).

Events also require recognition and participation to be visible in the public sphere. A family member praying quietly at a memorial site receives little attention, yet a head of state visiting the same site and laying a wreath gains a large audience. A lone perpetrator in the Rwandan genocide mumbling softly ‘I’m sorry’ goes unnoticed, while the pope’s visit and public apology for the role of the Catholic Church in the massacres generates headlines. Events as mnemonic practices require an audience for which they are performed, with the audience being part of the meaning-making process.

Many mnemonic events offer the opportunity for participation, mass enrolment and emotional purchase. They are often transmitted by television, radio and social media to reach a wider audience. This participation is important because it ‘connect[s]‌ participants together in commemorative moments, and provide[s] a sense of connection to people they imagine have performed or will perform the same ritual in the past or future’ (Drozdzewski et al., 2019: 263). There is a strong physical aspect to participating in events that moves beyond their cognitive effect. The intensity of an event experienced through many people coming together, getting excited, feeling powerful, singing songs and chanting declarations strongly affects participants beyond words (Jerne, 2020), again contributing to forging a sense of community based around a particular memory.

Events may serve to maintain the existing memory narrative they reflect, but they may of course also be transformative in how they attempt to challenge dominant mnemonic structures. While memorial events may serve to (re)create a community they might also be used for the opposite purpose: to be divisive and to disrupt the community fabric, to question remembrance in and of itself, or to provide an alternative performance. Similarly to memorials, they are open to (conflicting) interpretations and can be subject to contestation. This is apparent in the case of South Africa, where, after apartheid, the ANC government decided to do away with the old national holidays linked to colonialism or apartheid and instead introduced new public holidays such as Youth Day and Reconciliation Day, with Mandela Day marked as a global observance day.

Events are often organised by elite memory entrepreneurs and are therefore potentially highly effective for the production of hegemonic collective memories. As identity-shaping mnemonic practices, however, they need to be re-enacted time and again, which requires resources. Forms of power, such as access to the media or control over national holidays, are necessary to mobilise larger groups to participate in an event. Nevertheless, other events can take place more spontaneously, as when a piece of art is created or a dance performed, or as with a light spectacle, graffiti or a flash mob. They are then not placed in the longue durée of repetitive performances but instead challenge those by breaking out of ritualised structures. This might take the form of competing events that confront the hegemonic role of more recurrent practices.

Events hence have a high potential for mobilising large numbers of people and can therefore become highly political. Such events may trigger conflict as much as they foster expressions of solidarity and belonging. They are often what holds a memory landscape together in terms of serving as a constant reminder of the importance of honouring the dead, keeping memory alive and establishing a sense of unity. At the same time, they can also hinder transformation as they keep performing the past and projecting its assumed significance onto the present.

Sites, agents, narratives, events: productive interactions

In this chapter, we have separately discussed sites, agency, narratives and events as conceptual entry points through which we can investigate mnemonic formations. Again, this can only be done heuristically for analytical purposes, as all four points relate to and constitute each other. The next step is then to demonstrate the dynamic interplay between these four elements.

We view sites as material representations that fix meaning to physical locations and material structures, make the link to the past tangible in the present, pin memory politics to place and underpin a mnemonic formation (see Björkdahl and Buckley-Zistel, 2016; Kappler, 2017; McDowell and Braniff, 2014). Memorials, museums or marked structures are invested with particular meaning by being tied into narratives about past events, present conditions and future aspirations. Such sites are important to memory-making as groups may congeal around them (Halbwachs, 1992: 204), although these groups are never static and are always in flux. Hence, we argue, through social practices of place-making such as commemorative events memory agents make particular sites meaningful for memory, politics and peace.

We also suggest that in any given mnemonic formation an array of memory agents that may be formal, informal, local, national, international, transnational, collective or individual participate in memory politics. Such agency is relational and reconstituted in social interactions. It can be exercised through formal, public actions with political objectives or through fleeting action in the margins of the mundane to uphold or challenge existing power relations. Agents participate in an on-going making and remaking of mnemonic landscapes.

The narratives we focus on are specific articulations of the past that are evoked in memory politics and underpin the mnemonic formations. Such narratives connect the past with the present in that they refer to events and experiences of the past but are told and retold in the present. As such they are constitutive of individual and collective identities, so that they are a meaning-making instrument. There are hierarchies of memory, as narrating memory involves deciding what to remember and what to forget. Some narratives will through this process gain more traction than others.

Finally, we understand events as the performance of memory in a set of enactments using speech, movements and gestures to make reference to past events. They are meaning-making performative mnemonic practices that are part of a wider mnemonic formation. Such events may be ritualistic, or organic and fleeting as people come together for political action, and they are attributed significance recognised by the community. Events may serve to maintain existing memories, but they also have transformative powers.

It is precisely the relational constitution of these four conceptual elements that we are interested in. It is only in an analysis of their interplay that the full complexity of the mnemonic formation becomes visible. The synthesis through the SANE framework is dynamic and allows these elements to be examined for their mutual constitution of each other, as well as for how they may change each other. Ultimately, the application of this framework to the mnemonic formations means that empirical insights from each case can be compared and contrasted, as we do in the final chapter. Based on the analysis of the interplay between sites, agents, narratives and events in each mnemonic formation, and informed by an entangled understanding of memory, the final chapter is where we present our conceptual findings on the impact that politics of memory has on the quality of peace.