Dancing through time: A methodological exploration of embodied memories

This article responds to an absence of memory studies research methodologies for exploring embodied memories, including its form and content, the lived practices it involves, and its embeddedness in wider socio-political discourses. While conceptualisations of embodiment are central in the field of memory studies, its methodological consequences remain under-developed. We are proposing dance-based methods as having significant potential to address alternative ways of knowing and relating to the past. Drawing on empirical work on both professional and social dance among British Bangladeshi women in London, conducted as part of the 5-year research project Migrant Memory and the Postcolonial Imagination, we found embodied remembering to be a social form of doing that can serve to create, preserve and negotiate shared pasts.


Introduction
The multidisciplinary field of memory studies has long been concerned with the role of the body in producing, carrying, preserving and communicating memories.While the field has become increasingly interested in the role of textual forms and media and communication technologies in the transmission of temporal meaning across time and space, questions relating to embodiment, performance and affect in remembering processes persist.The everyday role of the body in the creation and transmission of memory has been conceptualised in a range of ways, from ritual communication (Connerton, 1989) to heteropathic encounters (Silverman, 1996), but memory studies research methodologies that have the body at their centre remain under-developed, particularly in contrast to the plethora of narrative and textual methodologies that have been widely adopted.As a result, empirical research on embodied memory, including its form and content, practices of remembering and its embeddedness in wider socio-political discourses, constitute a lacuna in the field.In response to the lack of methodological attention paid to the body in memory studies, this article proposes dance-based methodologies as having significant potential to address embodied ways of knowing and communicating the past.
In what follows we advocate for the development of embodied methods as a complement (rather than as an alternative) to traditional ethnographic methods such as interviewing and participant observation to develop accounts of social and cultural remembering which better account for the role of the body and its relationship with other technologies of transmission.In doing so, the body is not simply the object of enquiry but is also a methodological tool for exploring the communication of temporal meaning.We begin by assessing the existing scholarship on embodied memories, both to conceptualise the body in processes of remembering and to indicate the methodological limits which undercut the development of more refined understandings of embodied memory.We then set out how dancing, as an expression of everyday practices that communicate and negotiate vernacular memories and belonging, can be mobilised and applied in research on embodied memory by drawing on our own fieldwork which investigates the ways in which the end of empire in South Asia, postcolonial experiences of migration, and subsequent political ruptures are remembered by women from South Asian communities in the United Kingdom.While memories of the end of colonialism and subsequent political rupture and migration have often been studied with the help of approaches of ethnography and oral history, participatory methods that have the body at its centre mean that important nuances in the contemporary meaning and significance of the colonial past for diasporic communities in the United Kingdom, particularly in relation to women's experiences and remembering practices, are routinely overlooked.
This research draws on empirical work conducted as part of a 5-year research project, Migrant Memory and the Postcolonial Imagination (MMPI), 1 which explores cultural memories of the 1947 Partition of British India and subsequent decolonial processes among the South Asian Diaspora in the United Kingdom.As part of this project women's memories of the 1947 Partition of British India, the 1971 Liberation War of Bangladesh and subsequent migration to Britain among the female Bangladeshi diaspora in Tower Hamlets, London, were investigated using a range of qualitative methods.This article engages with key empirical elements of this work including 2 years of observation of both professional and social dance in women-only celebrations, interviews with 18 British Bangladeshi women, and most specifically a 2-month-long dance workshop with 13 participants.Based on this empirical work this article proposes collaborative and participatory dance-based methods as valuable to complement narrative-based approaches to studying embodied memories.

Mind the gaps: body, memory and methodology
In the field of memory studies, embodiment has become an increasingly popular concept.This builds in part on the work of postcolonial and feminist scholars, such as Frantz Fanon (1970), Luce Irigaray (1985), bell hooks (1990) and Judith Butler (1990), who have drawn attention to the body as a site of knowledge and contestation and its role in the production of meaning.This turn towards the body in social and cultural research is connected to critiques of Western scholarly traditions in which verbal modes of communication are often regarded to be of greater epistemic power.This can lead to uneven opportunities to contribute to processes of meaning-making, particularly in the public domain (Lyotard, 1979: 22-23), and insufficient recognition of the epistemic power of other modes of collective remembering (Carruthers, 2008: 18).The embodied turn has challenged such assumptions.Today in the field of memory studies, the concept of embodiment is widely if often loosely and nebulously applied to describe a range of mnemonic attributes and effects that are held within or emanate from a variety of abstract bodies such as places (Trigg, 2012), memorials (Young, 1994), buildings andarchitecture (Huyssen, 2003) and music (Van Dijck, 2006).This article is concerned with the more literal meaning of embodied memories as they are articulated through the human body.
Memory studies have always acknowledged the body as central in the preservation and transmission of pasts.Pierre Nora (1989: 13) locates 'true memory' in the body.This 'true memory' finds expression in gestures, skills and traditions, which he juxtaposes in elegiac fashion with modern archival modes memory characterised by material traces.For Nora, the move from embodied to archival memory marks the loss of mnemonic authenticity and the subsuming of memory by history.While his focus on the body as central to processes of remembering is productive, Nora's romanticisation of the body as the only vehicle of 'true memory' is problematic.It implies that the body is a repository for pure, unmediated, unfettered memory, rather than recognising the body and bodily articulations of memory as socially and culturally constructed, and the body itself as relationally embedded in wider cultural ecologies of remembering.Indeed, in contemporary memory scholarship it is rare to find such essentialised views of the body as a vehicle for memory and it is rather more commonplace to acknowledge the social and cultural contingencies of the body in the creation, preservation and transmission of memory.
Although working contemporaneously with Nora, Paul Connerton (1989: 104) locates the body as central to mnemonic processes, not only as a symbol of meaning, but as 'socially constituted in the sense that it is culturally shaped in its actual practices and behaviours'.He situates the body as one, which traverses the individual and social dimensions of memory, from bodily automatisms, to bodily habits, to the embodied performances of commemorative ceremonies (Connerton 1989: 4).Connerton particularly attends to the ways in which bodily performances of memory are embedded in, and are central constituents of, social and cultural rituals and forms of practice.For Connerton (1989: 72), bodies 'keep the past [also] in an entirely effective form in their continuing ability to perform certain skilled actions'.It is through these 'incorporating practices', that the past is transmitted via the co-presence of bodies acting in space, through ceremonies, bodily properties or bodily techniques (Connerton, 1989: 79).While he explores various ways in which the body is able to 'convey and sustain memory' (Connerton, 1989: 104), his analytical emphasis is on the stable reproduction of historical meanings, consciously or unconsciously, through the body.The body is not considered as a site of mnemonic struggle or as engaged in negotiations between competing pasts, which are performed from intersectional positions in the present.While his account certainly allows for the evolution of habitual bodily performances of the past in the present, the radical ruptures experienced by, for example, postcolonial bodies and the liminal spaces and places that they might inhabit remain beyond the scope of his analysis.
Attending to the body in remembering processes is key in redressing a tendency in cultural memory scholarship to overemphasise the stability, accessibility and intelligibility of written and oral memories (Taylor, 2003: 36-37); however, it is equally crucial that the body is not overdetermined as a privileged or static mnemonic site.Diana Taylor's approach is instructive in developing an understanding of the body as engaged in the complex, uneven and sometimes contested production of mnemonic meaning while resisting the romantic binary positioning of the body in contrast to what Nora (1989: 13) calls 'archival' modes of remembering.Taylor (2003: 20-21) positions the 'archive' and its relatively stable composition of enduring cultural materials, and 'repertoire', as the enactment of embodied memory, comprising 'performances, gestures, orality, movement, dancing, singing' and 'all those acts usually thought of as ephemeral, non-reproducible knowledge', as in constant and contemporaneous interaction as systems of mnemonic transmission.In doing so, she recoups embodied memory from Nora's binary and sequential characterisation of the relationship between embodied memory and archival forms of memory.Unlike the conceptualisation of embodied memory as ritual provided by Connerton, Taylor's conceptualisation addresses the interconnectedness of different modes of remembering and the potentially unstable character of embodied memories, which allows for an analysis of tradition and transmission, without losing sight of the potential for change, agency, negotiation and unequal power structures to be in play in embodied remembering processes.
In Taylor's work on embodied memory, and in much of the scholarship on cultural and transcultural memory more broadly, the focus of mnemonic analysis is predominantly (and appropriately) centred on the memory of those individuals and groups who are socially, politically and economically marginalised.Embodied memories have overwhelmingly been presented as of significance for the transmission of painful, interrupted, fractured, traumatising pasts in 'moments where speech fails and where the distance between past and present seem to collapse' (Hirsch and Spitzer, 2009: 158).However, this tendency has embedded a conceptualisation of embodied memory as compensatory, rather than as a modality of remembering which is embedded in a wider communicative ecology.Understanding embodied memory as part of vernacular remembering and embedded in wider cultural processes requires an analysis of embodied memory, not only where 'speech fails', but routinely and systematically as a vehicle for the transmission of the past, which is always entangled in the broad spectrum of remembering practices and processes.The rupture, dispersal and fragmentation of community can complicate the preservation and transmission of pasts (Hall, 1990: 223-224), and this can leave the body, as a continuous presence in time of rupture, as a key site in the making and transmission of memories when other mnemonic carriers such as archives, monuments and objects fail.Yet, bodies just like archives can be unreliable as they are embedded in social relationships, power structures and political processes, and are of course ultimately subject to the limits of human finitude.Understanding the precise ways in which bodies are implicated in these various socio-cultural processes is essential, as in creating, preserving and communicating memories, remembering bodies are at the locus of individual agency, social groups and cultural norms and conventions, as well as themselves being subject to the passage of time, social change and experiential ruptures, all of which significantly shape its ability to remember.In this sense, bodies mediate relations with close and distant others, and are therefore not simply generative of, or contain memory, but are active in its relational production.
As the preceding discussion suggests, accounting for the role of the body in remembering is not new.However, few scholars provide clear methodological strategies for approaching memory empirically.Despite the multidisciplinary character of memory studies, there is a somewhat narrow range of applied methods used for studying embodied memories, which is limiting the potential of the field to study memories beyond the use of oral or written language (Keightley and Pickering, 2013: 2).Qualitative interviewing is by far the most common strategy used to access memories (Mihelj, 2013: 62), and while the method has considerable strengths, it is not always the most suitable, or indeed the most ethical approach in gaining an understanding of the contemporary meanings of pasts and the mnemonic practices involved in articulating these.
Scholars working in the area of difficult pasts, such as Marianne Hirsch and Leo Spitzer (2009: 158) in the context of the Holocaust as well as Ananya Kabir (2005: 190) in the context of the 1947 Partition of India have argued for empirical memory studies to reach beyond the oral or written language in their research material.They identified that some painful memories are rarely verbalised, and propose the body as a site of remembering to offer participants alternative ways to articulate their pasts.Both propose analysing arts and performances to include embodied forms of remembrance.Besides applying approaches of oral history, Ananya Kabir (2013: 30) studies primary materials, including visual art, cinema, photography and music in the course of exploring her family's intergenerational memories of Partition.She describes her method as a critical reading of their content, and as such her empirical focus supports a largely textual mode of analysis.Similarly, Hirsch and Spitzer (2009) in their study of Holocaust memories of both victims and perpetrators in the film Shoah, in video material of the Eichmann trial, and in academic and literary texts, focus on the role of re-enactment, affect, and silence in testimonies, in bounded cultural texts.While the insights into embodied memory generated by this mode of textual analysis can be illuminating, the potential for understanding the social, experiential, embodied practices of remembering are limited in these approaches, and the sensory, affective, and relational dimensions of embodied memory still need to be accounted for.This requires attending to the role of the audience or the spectator in the production of embodied mnemonic meaning.For Hirsch (2012: 98), cultural representations of painful pasts should position audiences in ways that they are able 'to imagine the disaster "in one's own body", yet [to] evade the transposition that erases distance, creating too available, too direct an access to this particular past'.She argues that the body of the spectator should be brought into relation with the past of another, which enables affective engagement without collapsing the critical distance required to avoid sliding into a relationship of appropriation.
Performances of embodied memories can then evoke or engage bodily memories in their recipients.This is acknowledged in Kaja Silverman's (1996: 185) concept of 'heteropathic memory and identification-at-a-distance'.While Silverman (1996: 185) understands heteropathic memory as the ability of the onlooker to recognise the 'desires, struggles, and sufferings of the other' without interiorising or appropriating them, this ability can be extended beyond the 'look' into embodied participation in the performance of memories.Understanding the extent to which the bodies of audiences of embodied memory performances are engaged in the relational process of meaningmaking requires research methodologies which can gather and analyse data on social practices and contexts of doing embodied memory.In this regard, the methodological tools required for analysing embodied memory as a cultural form are well established, but less so in relation to embodied memory as a social practice.
In what follows, we therefore propose a participatory and performance-based methodology in which researcher and participants enter an ongoing social relationship that allows them to encounter memories beyond the spoken and written word and accounts for the cultural content of embodied memory performances, but is also able to capture the socially embedded meaning-making processes in which it is located.

Dance practice as a methodological approach to studying collective memory
We conceive the socially embedded body as shaped by past experiences and active in the creation, articulation and reproduction of mnemonic meaning.But bodies are also in motion, moving through changing contexts and encounters in which they engage in imaginative transactions and negotiations between past and present.Focusing on the methodological implications of such a conceptualisation, we suggest participatory dance practice as a valuable technique for observing and analysing not only the products (i.e. the specific versions of the past mobilised in the present) but also the relational processes of embodied remembering.In doing so we draw on the work of scholars in the related fields of dance and performance studies (Buckland, 2001(Buckland, , 2013;;Desmond, 1997;Foster, 2004;Mitra, 2009;Roach, 1996;Taylor, 2003) who have proposed ritualised, theatrical and participatory dance as carriers of communal identity and cultural memory.Rather than defining cultural memory as a storage system, they understand dance as an 'activity of remembering' that is generated through movements and relations (Ingold, 2002: 142).Dance is not an inevitable practice but an activity contingent on individual agency, social structures, participation and the gaze of others.It synthesises multiple modes of remembering interconnecting the repertoire with the archive (Taylor, 2003: 20-21).Embedded in these complex relations, ritualised, choreographed and participatory dance can have a stabilising effect on memory in ongoing participation and transmission of choreography and movements, but the mnemonic charge of dance practice can also make it a site of contestation (Parfitt, 2021: 3).As Lefebvre (1991: 216) suggests, the body that moves physically can also move politically, culturally, socially, affectively, imaginatively and kinaesthetically, which can imply ephemerality, disruption and negotiation of memory.
When dance is discussed as part of research methodologies, researchers often describe their observations of communication and negotiations taking place between performer and audience members (Mitra, 2009;Taylor, 2003).We are, however, interested in the participation in social dance practice.Many researchers have immersed themselves in established social dance practices such as Angela McRobbie (1984), who joined women dancing in the 1980s in London.More recently, and specifically relating to the field of memory studies and diaspora studies, Ann R. David (2015) carried out ethnographic work on Punjabi folk dance in London and Ananya Jahanara Kabir (2020: 137) immersed herself in an 'embodied investigation' of postcolonial memorialisation, creolisation of cultural practices and commercialisation in the festival economy.Kabir describes her reading of both 'archive' and 'repertoire' through her participation in dance classes, parties, concerts and festivals, which were accompanied by data generated from interviews.
The combination of interviewing and co-performance, a common practice in the anthropology of dance, has proven fruitful in getting to know an unfamiliar field (Thomas, 2002: 512) and to foster relationships with research participants (Ness, 2004: 139).Such a combination of methods also provides avenues for research participants to engage with the gaze of the researcher (Taylor, 2003: 64), can facilitate access to tacit embodied knowledge (Gore et al., 2013: 128) and recognises the interconnectedness of different mnemonic technologies (Skinner, 2010: 121).Most crucially, however, we can learn from anthropologists that the moving body in dance practice provides an entry point in forming a dynamic view of culture (Farnell, 1999: 344) as well as the physically enacted production and transmission of historically situated knowledge, identities and memories (Turner, 1985: 187).Considerations of memories and remembering are central to such anthropological thinking as dance is a mnemonic practice learnt by osmosis or training often attempting to recreate and maintain, while never being fully reproducible due to human error or the passage of time and space (Drewal, 1992: 4;Roach, 1996: 2, 27).In this regard, dance has not only been found to conserve pasts but also to be able to stimulate reterritorialisations of memories (Skinner, 2010: 119).
Those complex qualities of dance have been made use of in memory research.As in Kabir's (2020) approach, social dance offers an ethnographic opportunity for participatory or observational research; however, it is rarely considered as a guided and designed research technique to collect data on embodied memories in the field.Just as qualitative interviews use techniques of 'naturally occurring' conversations to elicit the experiences and meaning-making processes of participants, guided dance practice can be used in deliberate ways to generate data about relational and embodied mnemonic processes.What we propose is a reflexive relationship with what Taylor (2003: 20) defines as the 'repertoire'.Rather than only observing and participating in 'naturally occurring' performances of the repertoire, research can be designed to create times and spaces in which researchers and research participants can come together to co-investigate embodied memories through participatory dance practice.Movements, borrowed from the everyday as well as from diverse folk, popular and classical dance forms, can be used to introduce ideas to participants and evoke embodied responses alongside reflective conversations.Working with different dance styles and movements, those that participants consider central to their intersectional identities and those styles and movements that are positioned as other and as objects of desire, can mobilise different pasts and modes of remembering, enters them into conversation and gives insight into processes of mnemonic negotiation.The combination of repetition and ephemerality of dances allows for reenactment but also embodied re-imagination, revision, clarification and dialogues of pasts.Through joint performance, watching and copying movements of others, as well as the development of choreography, both products and processes of remembering intertwine and shared understandings of the past can be elaborated between researcher and researched.Dance as a guided research activity allows the researcher access to the transmission of non-narrative mnemonic meaning by enabling participation in the memories of others.This is achieved without immobilising participants' memories and without eliding how bodies and our interpretation of bodies are shaped by the past as well as by archival modes of remembering and present relationships.

Dances with memory: a case study
After a year of observing professional dance, participating in social dance practice during womenonly community celebrations among British Bangladeshis, and a series of interviews with professional diasporic dancers and curators, dance had been identified as a crucial practice of vernacular remembering among British Bangladeshi women in the context of cultural and gendered celebrations organised by women for women in local community centres.These celebrations take place during the day in an interstitial space between public performance and private homes, as well as sitting across and between different national, religious and cultural contexts.During these womenonly celebrations, dance enables a negotiation of different modes of femininity deriving from transactions between past and present as well as aspirational constructions of the future.We observed movements borrowed from Dhamail, a Bengali folk dance, Bhangra, a Punjabi folk dance, as well as dances to contemporary popular British and US-American music.In subsequent conversations and interviews, many women shared a desire to learn 'Indian Classical Dance', which they were familiar with from film, television and theatre.
Based on these initial observations and interviews, we recruited 13 Muslim British Bangladeshi women, aged 35-50 years and living in Tower Hamlets, to take part in eight 90-minute workshops, in which dance was used as a method to explore inherited and firsthand memories of political rupture and migration, and their relationship to processes of identification.The workshops were organised at Mulberry School for Girls in Tower Hamlets in collaboration with parent liaison officer Sabina Khan and professional dancer and choreographer Kesha Raithatha.Both advised the research design and process in different ways.Sabina Khan supported the research as a gatekeeper having close relationships with the participants.She assisted us in the recruitment process and advised the facilitation of the workshops.Kesha Raithatha, a British Gujarati dancer and choreographer, taught Kathak, a classical South Asian dance form, to both researchers and research participants providing us with a specific embodied repertoire to engage in relational communication and negotiations of the past.Kathak, a classical dance form that the research participants desired to learn, derives from a long history in South Asia often dominated by Indian Hindu communities that can serve to exclude the non-Hindu and non-Indian dancer both in the subcontinent and in the diaspora (Chakravorty, 2006: 123;Spear and Meduri, 2004: 441;Thobani, 2017: 170).
Each of the eight workshop sessions started with a focus group discussion allowing for conversations about the research questions and reflections on the last sessions.This was followed by calm or silent exercises to relax the participants.After that, we began the actual dance practice, which lasted around an hour.In the initial sessions, Kesha Raithatha taught the participants and the researchers movements borrowed from Kathak with a strong focus on the symbolism found in mythological stories, facial expressions and the hand gestures of mudras 2 (Spear and Meduri, 2004: 438).Growing up watching dancers in local festivals but also those of Bollywood cinema, these movements were not completely unknown to the participants, and evoked various memories among them.In their bodily responsiveness, the participants started to openly interrogate their relationship to such dances and movements, brought them into conversation with the intersectional positions they occupy in the past and in the present, and highlighted their dialogic relationship with the researchers and dance practitioner.After the dance practice, we concluded each workshop with another brief focus group discussion allowing for iterated reflections on the research questions, newly acquired movements and developed choreographies.These discussions covered questions about dance and body as well as about memories of home, ruptures, re-location and dislocation, and were later extended in individual in-depth interviews with the participants.Applying this combination of methods addresses the methodological challenges in using dance, a somewhat ephemeral and non-reproducible practice, as a research method and, therefore, the fear of a lack of tangible data.While it is important to acknowledge the ephemerality of dance, neither the memory of the archive nor the repertoire is fully safe from change, corruption and manipulation (Taylor, 2003: 36-37).As Danielle Robinson (2021: 231) notes, dance might be ephemeral, but the participants taking part in such dances are not, which is why an application of a combination of ethnographic methods assisted us in collecting and interpreting the mnemonic charge of their performances.In tandem with their dance, focus groups, smartphone photography and filming, notes and drawings, allowed us to bring archive and repertoire into dialogue.We were able to contextualise and reflect on the meaning produced in dance practice by including the participants' perspectives at various stages to make sense of the tacit and ephemeral knowledge inherent to the dancing body.
The dance workshops allowed for observation and analysis of embodied memory as a relational social practice and gave insight into the complex mnemonic terrains and processes the participants navigate.While each of the eight dance workshops as well as subsequent discussions, drawings, notes, photos, films and interviews provided us with answers about the relationship between memory and the body among British Bangladeshi women in Tower Hamlets, London, this found its most direct form in the last two sessions in which the participants together with researchers and dance teacher applied movements borrowed from Kathak dance to create their own choreographies.Based on the thematic focus of the research on cultural memories of decolonisation in the diaspora, the suggested topics for the choreographies included different political ruptures, spatialities and identities, such as 'Partition', 'Liberation War', 'Back Home', 'Motherhood', 'Womanhood' and 'Tower Hamlets'.Focusing on these choreographies and participants' reflections, we will illustrate the potential of dance to elicit the complexities of embodied memory.
Dance allowed the participants to engage with products and processes of remembering.They danced because they remembered, and they danced to remember.Dancing together with the research participants, we were able to observe and participate in these mnemonic processes.Monira explained, It just brings back memories.'Oh, remember this song?'I don't even realize tucking away my skirt, moving my feet, I would realize people looking at me and asking myself: 'Oh, what am I doing?' (Monira) Acknowledging the mnemonic dimension of dance practice, Monira finds that dancing comes naturally to her when she hears music.She describes her body as an 'unruly' mnemonic device and reverses the Cartesian dualism in which the mind should rule the body (Bordo, 2004: 145).By arguing that her body rules her mind when she is dancing, Monira points at the immediacy and uncontrolled dimension of embodied remembering.The sensory experience of music, which makes Monira's 'unruly body' dance, is evoked by and evocative of memories of the past.In actively providing moments in which social dancing can occur, we created moments in which such complex cross-temporal processes take place and become observable.Still, her bodily remembering is not as pure and unmediated as Pierre Nora would have argued.Monira recognises that dance makes discourses about the body visible, that bodies are always subject to interpretation, and in this way, that embodied remembering is a relational practice.
We interpreted dance not only as communicative of memories but also generating discourses, relations and mnemonic community.As Paul Connerton (1989), we find bodies to be socially constituted in processes of remembering.Monira's acknowledgement of the proximity of the bodies and gazes of others discloses communicative relationships and negotiations through dance.In thinking: 'What am I doing?', she responds to and with her 'unruly body' and engages with a hybridity of interpretations of her body and dance practice.She describes dance here as a communicative practice of embodied negotiation that functions as an intermediary to the verbal.This communication takes place through facial expression, bodily movement and choreography using the symbolism of the repertoire provided by Kathak.Dancing as evoked by memories functions to cement relations between the self and the social world through the ways in which these are collectively configured in processes of recollection.Many participants stressed the sense of community they felt during the workshops, as Ayesha shared: Whatever we were doing it was connected to everything else, to the dancing, the flowing and it was great to connect everything in your life to it and you could connect it easily to the questions that you asked.As if you are bonding as one and having people as a big group doing that passionately and I could see their passion through their dance.It was great to just let loose and get bonding.You don't get to see long lost friends and then you see them through dancing.I feel more happy during it and I feel more connected to myself and everything and people surrounding me.(Ayesha) Shared pasts and their communication heavily rely on a sense of belonging together, which many British Bangladeshi women communicated they would lack in today's fast-paced and individualistic life in London.Ayesha and others afforded dance with a quality to intervene in such isolation.This is not specific to this community, as dance has been found to provide women with a sense of community and recollection of past selves by other scholars such as Angela McRobbie (1984) and Leslie Gotfrit (1988).As Angela McRobbie (1984: 144) has found in her own work, dance represents a pleasurable practice that intervenes in the everyday, in which women's mnemonic agency can be limited due to a lack of community.Memory research must acknowledge the importance of community in processes of eliciting pasts, and dance as a participatory and often joyful practice is able to create such condition.Providing spaces for the exchange of memories and the generation of network and community are pleasurable for participants but also an important intervention, as Sabina Khan articulates: I want to see them blossom and grow.Obviously, they know who they are but throughout the years, starting their own families they sort of lose themselves a little too and coming back to these dance classes, they can be who they are.They don't have to pretend to be in a certain way or hold back.It is a safe space where they can actually express their feelings.Their emotions.Their excitement.Whatever it is that they want to do.(Sabina Khan) Sabina finds that dance allowed the participants a form of doing that otherwise needs to be held back.To further illustrate such potential of dance practice, which allowed us to observe and analyse embodied memories as a social practice involving mnemonic decision-making processes, negotiation, community, we will illustrate the choreographies developed as part of the workshops in more detail.In six pairs, participants agreed to create choreographies about the following topics: 'Back Home', 'Motherhood', 'Womanhood' and 'Tower Hamlets; and left other suggested topics relating to the 1947 Partition and the 1971 Liberation War aside.These choices provided us with insight into which pasts the participants wanted to engage through dance.As argued before dance is not an inevitable embodied practice but contingent on wider societal structures and individual agency.Reflecting on their choices, the participants explained: Julia: Remember yesterday when I suggested the topics [Partition and Liberation War], nobody chose them.Ayesha: No, no, they want something positive.
The participants did not want to engage with pasts they perceived to be negative.Nevertheless, this initial mnemonic boundary was transgressed in every single one of the choreographies.While none of the participants wanted to tell pasts relating to rupture, these experiences found expression in the themes of 'motherhood', 'womanhood', 'Tower Hamlets' and 'Back Home'.The two participants who chose the theme 'Back Home' portrayed images of nature and rural activities that they entangled with regional and national symbols.Their bodies were swinging to represent waves of the waters of East Bengal, and with their hands they portrayed swimming fish, which then turned into lotuses, a symbol of Bangladesh.In a sudden movement, their arms were raising towards the front of the room, indicating a stop.In defence, their arms and hands formed to shooting rifles.Showing their fists and enacting imaginary weapons, they imitated fighting scenes.After they were done, they explained, We did the lotus and we represented the ocean.It represents Bangladesh.That is the national flower and then did you see this part?We were like 'don't fight us, otherwise we are going to . ..'And then it was about to stop the war.We would fight back.(Sabina Khan) None of the women chose the topic of the Liberation War of Bangladesh; however, a significant part of the choreography depicted scenes of the war.The unruliness of memories allowed complicated pasts to find their way into more pleasant mnemonic terrains of folksy and romanticised rural life in Bengal.In analysing the mnemonic function of such juxtaposition, we find that dance allowed the participants to engage in creative mnemonic processes that are part of a 'doing of little things', which enunciates a repeated temporality that is not 'passive submission but of active engagement' with past, present and future (Das, 2007: 68).Inserting themselves as fighters protecting a peaceful life 'back home', dance became a process that allowed us to observe complex processes of transaction in which layered and at times divisive pasts are synthesised into meaningful, but, more importantly, also more desirable embodied mnemonic narratives.Rather than an enactment of pasts, the participants imaginatively engaged with their memories and produced newly recontextualised memories.Dance is contingent on individual agency and in its ephemerality it has the potential to provide participants with a less daunting tool to communicate and negotiate pasts.Choreography can imagine different pasts outside the authoritative and stagnant qualities often associated with the spoken and written word.This points at the accommodating qualities of dance, which is not pre-representational but still able to ease contested pasts.Dance functioned as a legitimate activity to playfully reach out to the past and the past of others, which can serve to foster and expand mnemonic community.
Dance did not only reveal mnemonic relations between remembered and remembering selves, and the shifting experiences and characterisations of community over time, but also differences among the participants as well as discontinuation of mnemonic practices.The fact that the British Bangladeshi women that took part in the workshop perceived Kathak as a distant, but nevertheless familiar and desirable embodied repertoire to engage in, discloses and negotiates preassigned lines of shared pasts among (diasporic) South Asians.Constructions of Britishness, Asianness and British Asianness were articulated and negotiated through an embodied engagement with past and present repertoires among the researchers, dance practitioner Kesha Raithatha and the participants.On one hand, the dance practice produced continuities and stability, but, on the other hand, it also allowed the participants to communicate mnemonic negotiation, differentiation, struggle, and to confront embodied unreliability.Asma, who moved a teenager from Bangladesh to London, shared, This was the kind of music and dance we had when we were growing up.It took me back to my childhood, the life I had, the life I knew.The dance would be part of our culture.That is the way I look at it because I miss it.(Asma) Asma reflects on dance as a practice that takes her back to a life among women in rural Bangladesh, where she was born in 1970, but also recognises that this kind of dance does not fully belong to her present reality.Kathak allowed Asma to mobilise memories of a past life in Bangladesh and bring those into conversation with memories of rupture and change relating to her migrating body, changing discourses that decide about shared repertoires as well as her changing body.Before the workshop, Asma had not danced in a long time, which is connected to social structures which produce tensions between her positionality as a middle-aged, married, Muslim British Bangladeshi woman and religious and nationalist discourses that define Kathak today as a dance form predominantly performed by Indian Hindus.The dance practice evoked many discussions among the participants about the meaning of Kathak as a practice relating to shared pasts in South Asia: We were one country before, it is like the restaurants in this country.They are owned by 90 per cent Bengalis, yet, it is called Indian restaurant.It is a bit like that with the Kathak.It is an Indian dance but actually it is shared.(Sabina Khan) Sabina's observation recognises how embodied practices can be subject to reinterpretation.Groups and individuals can be excluded from shared rituals and practices due to changing political climate and borders, which can lead to the abandonment of practices and the memories the practices are evocative of.However, it is not only structures and discourses that decide about the possible discontinuation of practices of embodied remembering.Moving, changing and ageing bodies can also cause abandonment of embodied mnemonic practices.Just like archives can be corrupted or destroyed, the body can be subject to processes that confuse or discontinue modes or remembering.Sadia told us, I stopped dancing after marriage.After my child was born it was difficult for me because I had a C-section, so I was scared to dance or any heavy thing.After this, after a long time, 14 years after that, I started here.(Sadia) While Sadia's statement hints at wider discourses around the gendered body, it is also revealing the ways bodies can fail as mnemonic devices, not due to trauma and rupture but due to their material and biological qualities.Due to these limitations of embodied remembering, applying it in a compensatory manner is not necessarily an expedient methodological solution in 'moments where speech fails' (Hirsch and Spitzer, 2009: 158).Many participants recognised the embodied limitations they were facing during the dance practice as they could not remember movements or their bodies were unable to reproduce movements.These moments allowed for reflection and gave insight into processes of forgetting, which some experienced as unpleasant, however, all of the participants agreed that the application of dance in research made them more comfortable and was more pleasurable than research based on verbal methods.Many of the women had previously taken part in other research projects and shared that they preferred those applying creative and arts-based methods.Yasmin shared, It was really good, it was not only teaching you how to dance but how to mentally prepare you as well and then you have to go with the flow and we have to go with the music in our heads, bringing out.Like group therapy, what you did, just follow it and listen to yourself rather than just listening to music.It was really good.Therapeutic.(Yasmin) Yasmin acknowledges the realm of pleasure in the research process and it is important to stress that dance has the potential, for many people, to be an enjoyable practice (Chakravorty, 2004: 14).This benefitted the study substantially, in terms of developing positive and trusting relationships with participants over a long period of ethnographic research, and in terms of facilitating reflective modes of engagement with mnemonic practices and meanings.

Conclusion
Drawing on empirical work on women's memories among the South Asian Diaspora in the United Kingdom, this article attempts to extend research on embodied memory by developing a methodological approach which both mobilises and explores the role of the body in vernacular remembering practices.Applying social dance as an elicitation method, we were able to study embodied memories as a relational practice that is able to construct, articulate and transmit notions of the past among social groups.We observed both the stable communication of shared pasts through the body but also processes of merging different pasts and the re-articulation of uncertain or undesirable pasts in bodily movements and reflections on them.Social dancing not only makes such processes observable for the researcher, but also allows the researcher to participate in reflexive transmission processes.Observing how participants engaged with dance repertoires, and their present-day aspirations to engage in such repertoires, allows for an exploration of shared, co-existing and competing embodied memories and notions of belonging.While dance provided a vocabulary outside the discursive rhythms and expectations of language, choreography and movement should not be mistaken as an arena free of power structures often associated with the mnemonic processes of the archive.Rather, dance and bodily movements have to be understood in conjunction with the ways in which they are narrated.Through this use of dance embedded in a wider suite of ethnographic techniques, we gained a thorough understanding of the role embodied remembering plays in the communication of complex and contested pasts by British Bangladeshi women in the United Kingdom.We conclude by stressing that conceptualisations of embodied memories and embodied methodologies can help to further understand interrupted or traumatic pasts and their reverberations in the present, but should not be limited to these.Instead, empirical research on embodied remembering can be understood as an entry point into the intricate network of different modalities of remembering across the full range of lived experiences.