More visualising, more methodologies: on video, reflexivity and qualitative research Sarah Pink Abstract This article was conceived as a response to Ruth HoUiday's article published in The Sociological Review in 2001 (48 (4): 503-22) 'We've been framed: visualising methodology'. Whilst recognising that HoUiday's work makes both an important contribution to her substantive area and describes an innovative use of video in qualitative research, her critique of visual anthropology as a discipline that uses reflexivity as a muse to hide its positivist truth quest has some serious problems that need to be redressed. Here I shall draw from existing work to discuss how reflexivity has been a key theme in the development of visual anthropology since the latter part of the twentieth century. With the recent availability of high quality digital video technologies at affordable prices the use of video as a method of qualitative research and representation has become 'revived' as a topic in discussions of qualitative methodologies. Whilst these 'new' explorations of the potential of video for the social sciences refer to the context of changing visual technologies (digital video, non-professional editing, and on-line or CD Rom publications) they also appear to be part of a current trend to which the title of this article alludes; the recent deluge of publications on visual research methods (Emmison and Smith, 2000; Bauer and Gaskel, 2000; van Leeeuwen and Jewitt, 2000; Rose, 2001; Ruby, 2000; Holliday, 2000; Banks, 2001; my own Pink, 2001a). In this article I shall explore the potential of video for a reflexive approach to qualitative research in sociology and anthropology. As my title indicates, I shall not be the first to do this. However my readings of recent texts on visual methods have led me to feel that as much as presenting important challenges and opportunities for the development of innovative and beneficial methodologies, the new space that the visual is coming to occupy in the social sciences is also becoming a site of inter- (and intra) disciplinary conflict, some of which is neither necessary nor very constructive. Therefore I shall also discuss how some 'new' texts have staked their claim to the visual. Although I shall be critical, I do not intend to enter into combat with the writers whose work I discuss here. I believe each of them makes an important contribution to understanding how visual practices might be part of qualitative work. < The Editorial Board of The Sociological Review 2001. Published by Blackwell Publishers. 108 Cowley Road. Oxford OX4 lJF, UK and 350 Main Street. Maiden. MA 02148. USA. More visualising, more methodologies Whilst recent publications have some common strands, such as an emphasis on reOexivity, a departure from the realist, positivist and scientific pasts of modern anthropology and sociology, and a focus on performance theory, they also stand apart from one another in ways that imply that their writers are both less informed than they ought to be about the disciplines they critique and too keen to identify the visual as 'naturally' aligned with their own theoretical ideas and methodological practices (which, to their credit are usually innovative and well informed). For example Emmison and Smith, disregarding the rather different epistemological bases and actual practices of visual anthropology and visual sociology (see Pink, 2001a: 7-13 for a discussion of this), bundle the two together because in their view "... to a significant extent the boundary between contemporary visual inquiry in sociology and anthropology is impermeable' (2000: 24), According to them 'the use of photography by anthropologists, sociologists and ethnographers has generally led to an insular and uninspiring sub-field' because 'Visual inquiry has for the most part failed to connect with the wider currents in social theory in these disciplines' (2000: 55) which has led to 'the widespread tendency to use visual materials (photographs) in a purely illustrative, archival or documentary way rather than giving them a more analytic treatment'. The result of this, they propose, is that 'most other sociological researchers simply aren't interested in what visual sociologists have to say' (2000: ix). In fact, regarding the visual as 'data' rather than 'representation', Emmison and Smith argue that rather than picking up digital video cameras, sociologists should not be producing images at all as 'visual data should be thought of not in terms of what the camera can record but of what the eye can see. Photographs may be helpful sometimes in recording the seen dimensions of social life. Usually they are not necessary' (2000: 4), Whilst this critique could be applied to the work of some visual sociologists, others (eg, Chaplin, 1994; Schwartz, 1993; Harper, 1998) and visual anthropologists (eg. Ruby, 2000; Banks and Morphy, 1997; Edwards, 1997; MacDougall, 1997) have engaged quite extensively in theoretical, analytical and critical discussions of the relationship between visual and 'mainstream' sociology and anthropology, yet, in their sweeping dismissal of the development of these disciplines Emmison and Smith do not recognise this significant body of work. Instead, departing from what they characterise as an 'uninspiring' field of visual research, their own is an 'observational' approach that concentrates on the analysis of anything that can be seen - from objects to social interaction and body posture and avoids most other contact between researcher and subject (with the exception of recommending that interviews might sometimes be useful). They fmd covert research more convenient because it evades the 'usual problems of normative responding' (2000: 110) which would imply that they are less interested in how knowledge is produced through the intersubjectivity of encounters between the researcher and the social and material world that she or he is investigating, than in 'controlling' that research situation. However, whilst their approach may be in conflict with those visual anthropologists and sociologists who insist that to be ethical visual research and representation 1: The Editorial Board of The Sociological Review 2001 587 Sarah Pink ought to be collaborative, reflexive, and to represent the 'voices' of informants, Emmison and Smith also lay their own claim to reflexivity, suggesting that becoming 'more reflexive' and 'methodologically skilled' in using what they call visual 'data' should 'enhance the quality of our research' (2000: x). Holliday also launches a sometimes misguided attack on some key anthropological texts, decontextualised from the long term theoretical debates of which they form a part, to argue that reflexivity of the anthropological kind 'becomes a mere buzz word generated within a pseudo-positivist approach still concerned with gaining greater degrees of "truth" and objectivity' (2000: 507), whilst her own brand of queer studies' reflexivity, of course, does not. To me this implies an ironic lack of reflexivity about how these authors' own written texts are structured as competitors in the race to assert the 'truth' about the value of visual methodologies as the key to the future of a reflexive social science. Especially in Holliday's case this seem incongruent as, in her attempt to obliterate her 'opponents' from other disciplines, she constructs a masculine textual narrative that feels out of place with a queer theory that has its roots in feminist theory. However, to be fair, and before I am accused of hypocrisy, I would not claim to have avoided using similar textual strategies in my own attempts to convince readers of the value of a particular approach to 'visual ethnography' (Pink, 2001a). Of course positivist, scientific and realist approaches have formed part of the history (and present) of most social science disciplines that developed over the twentieth century. However it is important to note that: first, disciplines are not homogenous univocal masses; second, such approaches need to be historically and culturally situated; and finally that change, whether it refers to 'social and cultural change' or changing theoretical and methodological approaches is usually uneven, complex and characterised by series of appropriations, continuities and differences. In this sense, any critique of disciplinary approaches needs to recognise such multiplicity and to correspondingly see academic discourses as situated voices. Whilst Holliday characterises such approaches as fundamental to the queer studies she aligns herself with, she does not apply this principle to her own analysis of academic voices. As I shall show below by exploring how anthropologists and sociologists have actually used and theorised video methods it becomes clear that there is an existing body of work that does investigate how video images and technologies become part of the encounters through which visual and other knowledge is produced in both fieldwork and representation of qualitative research. In pursuit of reflexivity In some recent work, including my own (Pink, 2001a) video methods (and visual methods in general) have been situated as part of a reflexive methodology. Those who advocate reflexive methods tend to regard others 588 <• The Editorial Board of The Sociological Review 2001 More visualising, more methodologies as having not engaged sufficiently with reflexivity. They suggest that this has led to a rejection of video as usefully objective recording method and simultaneously a failure to appreciate the potential of video as inevitably constructed representation. For instance, reflexivity, at the centre of Ruby's argument, is defined as when: ... the producer deliberately, intentionally reveals to his or her audience the underlying epistemological assumptions that caused him or her to formulate a set of questions in a particular way, to seek answers to those questions in a particular way, and finally, to present his or herfindingsin a particular way. (2000: 156) This, Ruby argues is synonymous to doing 'proper' anthropology (2000: 167). Ruby's book Picturing Culture (2000) is a revised compilation of selected key articles written through out his career. Here, reasserting the arguments critiquing scientific and positivist empiricist approaches to anthropology he made 20 years previously he laments that he see little evidence that reflexivity has been integrated into the practices of anthropologists and attributes this to a fear of relativist and post-modem approaches. Ruby situates the beginning of a reflexive visual anthropology in the earlier twentieth century work of Flaherty and Mead and Bateson and the films of Jean Rouch. Lamenting that few anthropological film or video makers have since approached the methodological issues they raised. He suggests understandings of film as objective data recording have precluded reflexive approaches to anthropological filmmaking. Instead Ruby advocates an anthropological cinema that incorporates reflexive statements of method, and is based on new understandings of visible and pictorial aspects of culture that depart from the formalist and realist schools. For Ruby a reflexive approach to visual anthropology is also an ethical approach. Arguing that it is the ethical responsibility of the anthropologist to reveal the constructedness of her or his text. HoUiday also fails to discuss any engagements with reflexivity in existing sociological or anthropological work that she finds satisfactory. This seems curious since she insists in an endnote that she is aware that some anthropologists and sociologists are critical of 'traditional' assumptions.^ She proposes that queer theory provides a foundation from which to critique the quest for truth that she equates with the contemporary projects of visual anthropology and visual sociology. Whilst some of Holliday's points may be valid for the work of some visual sociologists (see also Emmison and Smith, 2001), she does not in fact recognise the existence of the sub-discipline of visual sociology, but instead launches into a misinformed attack on visual anthropology. There are some significant problems with Holliday's rendering of visual anthropology. First she neglects the issues and debates that have been raised during the last 25 years of development in theory and practice of visual anthropology. Her quotations from Hockings' (ed.) 1995 text actually refer to a second edition of a 1975 text that includes many of its original chapters -^ and © The Editorial Board of The Sociological Review 2001 589 Sarah Pink lead her to characterise visual anthropology as a sub-discipline that, maintaining an art/science dichotomy, strives to produce 'objective' 'scientific' anthropological fihns that avoid the artistic subjectivity of cinema. These views hardly represent the contemporary visual anthropology advocated by Ruby (see above), or by anthropological filmmaker and theorist David MacDougall who proposes that anthropologists 'look at the principles that emerge when fieldworkers actually try to rethink anthropology through use of a visual medium' (1997: 192), rather than attempting to fit visual anthropology into a scientific paradigm. This would 'involve putting in temporary suspension anthropology's dominant orientation as a discipline of words and rethinking certain categories of anthropological knowledge in the light of understandings that may be accessible only by non-verbal" means' and 'a shift from wordand-sentence-based anthropological thought to image-and-sequence-based anthropological thought' (1997: 292-3). Second, according to HoUiday, anthropology (unlike her own discipline, queer studies) is not a 'properly' reflexive discipline in that, she claims, it uses reflexivity as a muse to hide its positivist truth quest. Thus she posits: Reflexivity is the latest in a long hne of (not specifically anthropological) techniques aimed at ensuring the production of greater degrees of 'truth', and is particularly espoused by Clifford and Marcus, anthropological gurus extraordinaire. (2000: 506). Given that Clifford (in Clifford and Marcus (ed.) Writing Culture) is especially well known for suggesting that 'Ethnographic truths are ... inherently/^arf/a/ committed and incomplete' (1986: 7) (italics in original), it is worth quoting him more fully to explain why Holhday is misguided in her critique because far from seeking greater degrees of truth, Clifford's approach notes the impossibility of telling 'the truth' in ethnographic text because 'Even the best ethnographic texts - serious, true fictions - are systems or economies, of truth' (1986: 7). Clifford means that ethnographies are 'fictions' in the sense that they are 'made or fashioned' in the sense of the Latin fingere which he emphasises also 'implied a degree of falsehood' and thus the meaning of 'not merely making, but also of making up, of inventing things not actually real' (1986: 6). In fact the issues raised by Clifford and Marcus have had an important and lasting impact in anthropology. As James, Hockey and Dawson sum up ... the 'Writing Culture' debate has alerted anthropologists to the need to pay closer attention to the epistemological grounds of their representations and, furthermore, has made them consider the practical import of that process of reflection, both for the anthropological endeavour and for those who are the subjects of any anthropological enquiry (1997: 3) I would suggest that whilst there has been resistance to the visual and to reflexivity in both sociology and anthropology, it is in fact unjustified to 590 © The Editorial Board of The Sociological Review 2001 More visualising, more methodologies characterise either discipline, or their visual sub-disciplines as advocating realist, objectifying practices. Clearly there is demand for (and actual practice of) reflexivity in anthropology and sociology, reflected in Ruby's and Holiday's polemics, and the increasing engagement with the visual is echoed in both the series of recent pubhcations on visual methods and visual anthropology and sociology courses on offer in the UK and US. Video and visual research methods Below, limiting my discussion to video, I shall discuss HoUiday's case study in relation to existing approaches and methods developed in sociology and anthropology. I shall argue that by resituating her work in relation to both historical and contemporary developments in visual anthropology and sociology it makes an interesting contribution to existing debates about visual methods, but that it represents neither a 'new' nor uniquely 'queer' method. In short, both in fieldwork and in one's dealings with other disciplines a collaborative approach that recognises multivocality and seeks not to objectify the 'other' as in either 'other' people or 'other' disciphnes, is more appropriate. HoUiday's visual research method involved asking her respondents to make 'video diaries' with camcorders that would 'demonstrate visually and talk about the ways in which they managed or represented their identities in different settings in their everyday lives' (509). She sees 'The use of video as a process in the research' that allows 'a representation of the performativity of identity to show through' along with 'the narrativization of identity (through respondents' commentaries)' and reflects 'the selection, editing and refining that constitute identity and performativity as a process in all our lives' (2000: 509-510). HoUiday identifies two styles of diary: those that are 'full of performances dancing and singing, jokey telephone conversations ...' which she interprets as implying 'a high level of self-consciousness on the part of the diarists' which 'appears to be the result of performing informant on a known other' when the diarist was accompanied by a friend. However she found that when diarists were alone with the camera they tended to take a more 'confessional stance', appearing to 'lose all inhibitions, disclosing the most intimate details about themselves to the camera' (2000: 510). Whilst it is unsurprising that HoUiday's respondents would 'perform' different aspects of their identities in different social and technological contexts, there is, as HoUiday notes, something curious about their use of confessional discourse. HoUiday suggests that the use of confession is empowering because 'In terms of the video diaries, then, the power to present one's subjectivity may override the risk of having that speech appropriated by others ... Thus the fullest confession opens up the greatest space to talk and affords the greatest power at that moment' (2000: 512). Whilst this may be correct I would suggest that it is at this point that HoUiday's own reflexive practice might consider the negotiations through which the © The Editorial Board of The Sociological Review 2001 591 Sarah Pink knowledge produced through these diaries was represented. By asking her respondents to create visual diaries she was also asking them to interpret what a diary would be. In doing so it is likely that some of them adopted an existing confessional dairy narrative. In my own experience of interviewing with video I have found that my collaborators in these videos have also used known (to them) Spanish and English cultural narratives though which to conduct their roles in these interviews, selecting for example 'counselling' and 'gossip magazine' narratives as ways of talking about their identities, homes and lives in them, I would suggest, therefore, that by using the diary as a concept for her respondents to work on Holliday was also active in this process as she facilitated their empowerment through the way she constructed the research process. Moreover, the 'video diary' is now well established on British television and the 'confessional' narrative was used in the late 1990s Video Diaries series, in, for example 'My Demons' in which the protagonist takes a painful journey through her unhappy childhood. What it seems to me is important about Holliday's method is that it allows her to discuss how her respondents produced their self-representations on video, with a focus on their reflexivity. The video diaries also seem to work on different analytical levels. First as cultural products they visually represent embodied performances of identities and the ways these are objectified in materials displays and objects. Second by seeing video diaries as processes rather than products, they are performances of identities in that their production is itself a practice through which identities are negotiated and performed. Finally as Holliday notes herself, 'identities may also be expressed in the very structure of the diaries themselves, which frequently borrow textual and visual codes from queer television and film' (2000: 513-4), Holliday rightly rejects the realism that she wrongly claims is the pursuit of visual anthropology. She argues that a post-positivist sociology should not be concerned with the accuracy of data, as much as with the abihty of that data, to provide 'a perspective on the social world from a subject situated within it' (2000: 517), This, she apparently claims is where anthropology and sociology have failed, but where a 'Queer Methodology' succeeds. It is here that Holliday stakes her claim for queer studies as a critical discipline that 'disrupts the notion of a "natural world" to study, and is particular suitable for visual methods. Instead it seeks to uncover "truths" (instead of a universal truth) as they are experienced and represented, though the context-specific discourses available to participants in particular locations' (2000: 518), This however, is nothing new, nor unique to queer studies; it forms a fundamental principle of much contemporary anthropology. She also suggests that a queer methodology might be closely aligned with a visual research method, since the visual dimension has close association with queer identities and is one in which queer subjects have particular skill (2000; 518), However, I fail to see why, although they might be differently skilled, queer subjects might be more skilled at visual construction and interpretation of identities than other people living in Britain or elsewhere might be, 592 © The Editorial Board of The Sociological Review 2001 More visualising, more methodologies Holliday sees her video diary approach as 'a new form of research with consequences that have not yet been forseen' (2000: 517). The 'video diary' approach is innovative in some ways and, I would suggest, especially appropriate in a culture where self-awareness and diary keeping are part of contemporary practices and where asking informants to keep written diaries is not an unusual method. However, the practice of giving cameras to native collaborators in research has formed an important part of the last 25 years of visual anthropology that Holliday skipped in her critique. As I shall outline below in my discussion of Ruby's outline of a future for visual anthropology, the potential for empowerment, self-representation, and the implications this has for the changing role of visual anthropologists have indeed been explored in detail (for example in the work Ginsburg and Turner). In outlining a way forward for a reflexive and ethical approach to anthropological film. Ruby argues anthropologists should give the subjects of films 'some say in the construction of their image', suggesting they may do so 'as facilitator and analyst of indigenous production or as collaborator', with the proviso that in the case of the latter this will mean creating a context in which the subjects are able to participate fully as collaborators (2000: 219). As an example of appropriate video research Ruby celebrates the work of Eric Michaels, who, before his early death in 1988, developed an innovative project with Australian aboriginals, facilitating their use of video media to represent themselves. Here I summarise this, not because I simply want to use an example already developed by Ruby to make my own point, but as a way of also representing Ruby's strategy. He effectively resituates Michaels 198O's project by giving it (not undeserved) centre stage as a chapter in a book that will doubtlessly be widely read by researchers and students. In the 1980s Michaels was contracted to 'assess the impact of television on remote Aboriginal communities' (Ruby, 2000: 225). To do this he developed a project designed to not only 'study' the aboriginals, but also give them agency. Rather than simply carrying out ethnography of their television viewing he trained them in video production and facilitated their setting up of a low-power transmission facility so that they may respond to television through their own broadcasts (2000: 227). Michaels' work showed how the Aboriginals' approach to the ownership, inheritance and use of information is different from European Cultures' who strive 'to 'broadcast' as much information as possible to as many people as possible' (2000: 229) and that 'the values of Australian Aboriginal society and European Society clash as regards the rights of individuals to produce and use pictorial information' (2000: 230). Michaels' work may be seen as a video method that not only allowed him to understand differences between Aboriginal and European uses of visual representation and information but also one that empowered his informants to respond in a global media context in which they might be seen as marginal or 'invisible' (a claim that Holliday also makes for her Queer subjects). Thus Ruby rightly argues that using the inexpensive technology of video, which is 'decentralised and almost impossible to control' it is possible to resist the monopolies of © The Editorial Board of The Sociological Review 2001 593 Sarah Pink 'multi-national image-making corporations' (2000: 237) and to allow more than one visual construction of 'truths' to take the stage. In my book Doing Visual Ethnography (Pink, 2001a) I have written chapters on uses of video in ethnographic research and representation. My purpose here is not to repeat the arguments I make there, but to bring attention to two projects that I felt it was important to refer to in my book and that are particularly relevant for the discussion here. The first can be seen as a feminist sociology project, in which the researchers facilitate their informants' self-representations, serving to empower them in a way that Holiday's model does not, Barnes, Taylor Brown and Wiener's (1997) project involved facilitating the production of videotapes in which mothers with HIV recorded messages for their children to view after their deaths, that they felt would represent them appropriately to her children. By helping these women to create such documents the researchers also intended to 'study the interactive aspects of mothering and the significance of impending maternal death from a stigmatising illness' through a framework of 'the concept of "eternal mothering"' (1997: 7), Conscious of the positivist tradition that has informed their discipline, and potential critique of their method Barnes et al. acknowledge that as an 'experimental' project their work is limited in that it offers few opportunities for triangulation and that the 'reaUty' recorded would have been affected by the presence of researcher and camera. Departing from a scientific experimental stance, they follow Chaplin's (1994) feminist approach to visual sociology, to produce knowledge not about, but/or women and in doing so to situate these women 'at the centre ofthe production of knowledge' (1997: 13), Thus they argue that 'the method offers the spontaneity and vividness of an uninterrupted stream of information from the individual, as the mother is allowed to talk without researcher intrusion in the form of questions' (1997: 13), For their own research about mothering the researchers (again following Chaplin) do not expect these video representations to tell one 'truth' but that 'the study offers a range of suggestions and an opportunity to construct a constellation of meanings about mothering' (1997: 14), They note how these representations also have to be situated not simply as messages from mother to child but that also in relation to 'their attitudes about how mothers care for and protect their children, how their impending death from AIDS influenced their mothering and how stigmatisation from AIDS may be transferred from them to their children' ,,, 'their self-presentation' (1997: 21), As regards the apphed aspect of the project Barnes et al. see video as an empowering visual media, it 'offers women, minorities, HIV infected people, and other marginalized groups, an opportunity to reproduce and understand their world as opposed to the dominant representation depicted in the mass media' (1997: 27). Again here video is conceptuahsed as a means of empowerment and of 'making visible' marginalized groups or individuals. However this project adds a new dimension to the possibility of video as empowerment because, in this project the researchers facilitated the women's use of video to create permanent visual representations to be used as a means 594 © The Editorial Board of The Sociological Review 2001 More visualising, more methodologies of negotiating and performing identities in private domains and intimate relationships, rather than to make public statements. A different example of a reflexive approach to video in the production of ethnographic knowledge is demonstrated in Ferrandiz's anthropological video work with Venezuelan spirit cults (1996, 1998). Situating the role of video in his fieldwork in relation to the cult's existing relationship with, and experience of media representation, Ferrandiz pays particular attention to the way the video recording developed through the intersubjectivity between himself and his informants. In some instances the video became a catalyst that helped create the context in which it was used, as in the case of a ceremony that was organised by his informants as part of the event of videoing it. However of particular interest is that when Ferrandiz began shooting video, six months into his fieldwork, the informants with whom he was closely collaborating also took the camera to shoot footage themselves, each of them creating 'completely different visual itineraries of the same place' (1998: 27). Ferrandiz takes his analysis further than merely the question of how different people created different video narratives of the same context. He forms continuities between the video making and the ritual activities in which his informants were involved; the visual practices of video recording and the ritual practices coincided as people moved in and out of trance and in front of and behind the camera's viewfinder as the ceremony proceeded. In this research the video camera became part of the material culture of the ritual and its recording capacity an aspect of the ritual activity. Therefore Ferrandiz was able to leam about ritual practices through his use of video. To take this further we can also see how this particular use of video makes explicit some issues about 'visual truth' and different ways of seeing. In Ferrandiz's footage both people who were not themselves and spirits who were not the people whose bodies they inhabited appear as different individuals move from in front and behind the camera and in and out of trance. We need to be aware not only that visual images do not represent objective truths, but that also, reality itself is not necessarily visible. As these anthropological and sociological studies show, theoreticallyinformed reflexive approaches to the use of video in collaborative or facilitating projects that both empower informants and inform researchers' interpretations of visual knowledge and understanding have been developed and strongly argued for in these disciplines. If, as Holliday implies, these or similar methods, and theories of difference and the importance of seeing (visual and other) knowledge, subjectivities and interpretations as inevitably situated, are of interest to queer studies, then significant, constructive and even collaborative connections could be made between anthropological, sociological and queer studies approaches; they need not be pitched against one another. Academic representation Above I have discussed both informant own productions of visual knowledge through video, and the ways that researchers produce and interpret knowledge © The Editorial Board of The Sociological Review 2001 595 Sarah Pink themselves through visual research processes. One question remains, if video is to play such an important role in the way knowledge is produced and represented in research and in the ways informants represent themselves to researchers, their famihes or in a global media context, what sort of role should video play in the academic representation of visual knowledge? Ethnographic fihns, or more commonly now videos, have been the dominant medium of visual anthropological representation. However this form of anthropological film, influenced by the conventions of documentary filmmaking and sometimes television production values has recently come under criticism by Ruby who argues that 'Ethnographic film production has been dominated by the professional expectations of the film world regarding equipment, production values, and consequently, cost, not by the interests and needs of anthropology. It is time to change that' (2000: 21). Instead he proposes that film should be used by anthropologists to represent their work in ways 'parallel to, but not necessarily less significant than, the printed word' (2000: 22). He calls for not only the production of new visual anthropological texts, but also of new visual anthropologists who will take control of ethnographic film, and not cede it to eitherfilmmakers,or the (usually) misrepresented categories of woman/native/other because, he argues, 'the purpose of anthropology is to allow people to see the native through the eyes of the anthropologist'; a way of seeing that he regards different from but no more truthful than films made by other groups (2000: 32). HoUiday argues that when video clips are used in academic presentations 'respondent empowerment and emotional engagement is made available through visual/verbal representation - through their greater presence' (2000: 519). She earlier indicates the importance of academic framing because 'queer subjects tend to draw largely on essentialist logics in their self identification and that is a product of limited access to other discursive positions (such as those that circulate in the academy for instance) (2000: 516). She suggests that since she is situated differently from her respondents 'the subjects of my study may not necessarily agree with my interpretations but they do get to represent themselves with minimal interference at least' (2000: 518). She thus appears, hke Ruby, to advocate that academic representations should be made and framed by academic perspectives rather than those of 'Others'. Holliday asks how the 'presence' of her respondents may t« similarly represented in academic publications. She answers this herself to some extent by suggesting that visual texts might be integrated into academic publication by the use of video 'programmes' or video 'papers' that might be viewed on the internet, 'by anyone anywhere' (2000: 519). How then might such video texts be conceived? In suggesting a departure from existing traditions in ethnographic film Ruby offers one direction. He proposes that anthropologists produce their ownfilms,that communicate anthropological insights in a way that parallels but is different from the production of written ethnography. He proposes that If ethnographic fikmnakers were to produce films that tell the story of their field research, and the story of the people they studied, in a reflexive manner 596 © The Editorial Board of The Sociological Review 2001 More visualising, more methodologies that permitted audiences to enjoy the cinematic illusion of verisimilitude without causing them to think they were seeing reahty, then an anthropological cinema would be born, (2000: 278) To my knowledge the 'video paper' is a genre yet to be developed, although some anthropological work could be seen as a starting point for such projects. For example, Kwame Braun's Passing girl, riverside; an essay on camera work (1996), a reflexive video text that explores and critiques the ethics and negotiations of its own production raises issues that are pertinent to any academic who takes responsibility for representing 'others'. Another option, that both Ruby and Holliday touch on, although in their cases as a means of dissemination for video essays, is the use of hypermedia - CD ROM, DVD or Internet, Uses of hypermedia to develop visual and multimedia representations of ethnography have already begun to develop. As I have described elsewhere (see Pink, 2001a, b,c) a number of on-line and CD ROM projects (see for example, Biella, 1997) have begun to create reflexive texts that combine video, still photography and written words. Hypermedia as a medium that can be designed (if not experienced) multilineally provides researchers with scope to both acknowledge and represent different written, visual, academic, and informant narratives and voices, without necessarily privileging any of these. It offers academics the opportunity to engage in written theoretical debates, that use written words in a way that only words can be used, and to also use video clips and photographic stills to communicate the types of knowledge that they best communicate. In Holliday's case this might mean using video clips to give her respondents' self-representations a voice or space in the text. To sum up, in this paper it has been my intention to be critical. Not of 'bad research', because in my assessment the developments in the practice of visual methods in reflexively comprehending research and representation and in seeking ways to allow the subjectivities of informants a space in academic texts that both empowers them and acknowledges the 'fiction' (see Clifford, 1986) of any ethnographic representation, indicate that visual methods are being 'properly' developed, I hope to have demonstrated that much of Holliday's critique of anthropology and sociology was misdirected, and that in fact by referring to the ways some academics working within these disciplines have dealt with both the visual and with objectifying and realist approaches, and to some ofthe historical debates upon which these developments have been based, a context in which visual methods have been used to empower informants, achieve good academic understandings of how informants represent their identities, and to represent such work is already developing. In this sense, visual anthropology is not a narrative foil to be jousted with in 'combat', but a reflexive self-conscious discipline whose practitioners might be willing to collaborate with both their informants and with other disciplines, Loughborough University © The Editorial Board of The Sociological Review 2001 Received 2 March 2001 Accepted 22 June 2001 597 Sarah Pink Notes 1 Holliday justifies her dismissal of the majority of contemporary visual anthropology by arguing that whilst the 'traditional assumptions' (presumably meaning those assumptions that informed the development of visual anthropology in the 1960s and 70s) she critiques have been criticised by anthropologists and sociologists, she wants to 'characterise a field' because 'its principles are currently taught to students' (2000: 520, end note 3). However it is unclear which 'field' she wishes to characterise, and still less clear how this applies to visual anthropology as she does not site where and by whom such 'principles of visual anthropology' (the title of Hockings (ed.) (1997 [1975] book). In my own experience they are taught as part of the history of ideas in visual anthropology, and as part of a debate - not as a model for contemporary visual anthropological practice. References Banks, M., (2001), Visual Methods in Social Research, London: Sage. Banks, M. and Morphy, H., (eds) (1997), Rethinking Visual Anthropology, London: Yale University Press. Barnes, D.B., Taylor-Brown, S. and Weiner. L., (1997), ' " I didn't leave y'all on purpose": HIV-Infected Mothers' videotaped legacies for their children', in Gold, S.J. (ed.), 'Visual methods in Sociological Analysis', Special Issue of Qualitative Sociology, vol. 20, No 1. Bauer, M. and Gaskel, G., (2000), Qualitative Researching with Image, Text and Sound, London: Sage. Biella, P., (1997), 'Mama Kone's Possession: Scene From an Interactive Ethnography', in Visual Anthropology Review, 12 (2): 59-95. Braun, K., (1998), Passing Girl, Riverside: An Essay on Camera Work, Documentary Educational Resources, USA. Chaplin, E., (1994), Sociology and Visual Representations, London: Routledge. Clifford, J., (1986), 'Introduction: Partial Truths', in Clifford, J. and Marcus, G. (eds). Writing Culture: the Poetics and Politics of Ethnography, Berkeley: University of California Press. Clifford, J. and Marcus, G., (1986), Writing Culture: the Poetics and Politics of Ethnography, Berkeley: University of California Press. Edwards. E., (1997a), 'Beyond the Boundary: a consideration of the expressive in photography and anthropology', in Banks, M. and Morphy, H. (eds). Rethinking Visual Anthropology, London: New Haven Press. Emmison, M. and Smith, P., (2000), Researching the Visual, London: Sage. Ferrandiz, F., (1996), 'Intersubjectividad y video etnografico. Holguras y textxuras en la grabacion de ceremonias espiritistas en Venezuela', in Garcia, M, Martinez, A. Pitarch, P Ranera, P and Fores, J. (eds), Antropologia de los sentidos: La Vista, Madrid: Celeste Ediciones. Ferrandiz, F., (1998), 'A trace of fingerprints: displacements and textures in the use of ethnographic video in Venezulean spiritism', in Visual Anthropology Review, vol. 13. No. 2: 19-38. Harper, D., (1998), 'An Argument for Visual Sociology', in Prosser, J. (ed.). Image-based Research: A Sourcebook for Qualitative Researchers, London: Falmer Press. Hockings, P., (ed.), (1995 [1975]), Principles of Visual Anthropology, (second edition). The Hague: Mouton. Holliday, R., (2000), 'We've been framed: visualising methodology', in The Sociological Review, 48 (4): 503-522. James, A., J. Hockey and A. Dawson, (1997), After Writing Culture, London: Routledge. MacDougall, D., (1997), 'The visual in anthropology' in M. Banks and H. Morphy (eds). Rethinking Visual Anthropology, London: New Haven Press. Pink, S., (2001a), Doing Visual Ethnography: Images, Media and Representation in Research, London: Sage. 598 © The Editorial Board of The Sociological Review 2001 More visualising, more methodologies Pink, S., (2001b), 'Ethnography, Experience and Electronic Text: a disciission of the potential of hypermedia for teaching and representation in anthropology' in I. Edgar and D. Drackle, (eds). Teaching Social Anthropology in Europe, Oxford: Berghahn. Pink, S., (2001c), 'Visualising Ethnography: transforming the anthropological vision' Visual Evidence, International Seminar Series 2000-2001, University of Leeds. Rose, G., (2001), Visual Methodologies, London: Sage. Ruby, J., (2000), Picturing Culture: Explorations of Film and Anthropology, Chicago: University of Chicago Press. Schwartz, D., (1993), 'Superbowl XXVI: Reflections on the Manufacture of Appearance' in Visual Sociology, Vol 8(1): 23-33. van Leeeuwen, T. and C. Jewitt (2000), Handbook of Visual Analysis, London: Sage. © The Editorial Board of The Sociological Review 2001 599