Copyright Sociological Research Online, 1998


Bella Dicks and Bruce Mason (1998) 'Hypermedia and Ethnography: Reflections on the Construction of a Research Approach'
Sociological Research Online, vol. 3, no. 3, <>

To cite articles published in Sociological Research Online, please reference the above information and include paragraph numbers if necessary

Received: 4/6/98      Accepted: 29/9/98      Published: 30/9/98


Current interest in ethnography within social research has focused on its potential to offer insights into the complexity of the social world. There have increasingly been calls for ethnography to reflect this complexity more adequately. Two aspects of ethnographic enquiry have been particularly singled out as areas in need of redefinition: the delineation of ethnography's object of study and its mode of presentation. Both of these areas are implicated in the recent attention to the possibilities of hypermedia authoring for ethnography. The paper offers a discussion of this potential in the light of an ongoing research project with which the authors are engaged. The project is designed to enable this potential to be assessed, and to provide for the construction of what the authors call an ethnographic hypermedia environment (EHE). We believe that the promise of hypermedia lies not only in its facility for non-sequential data organisation, but also in its ability to integrate data in different media. The synthesis of the visual, aural, verbal and pictorial planes of meaning holds considerable promise for the expansion and deepening of ethnographic knowledge. Consequently, we suggest that hypermedia has implications for all stages of the research process, and argue against the current tendency to see it as merely a tool either for analysis or for presentation. These arguments are illustrated by means of a commentary on some work in progress.

CAQDAS; Computers; Ethnography; Hypermedia; Methodology; New Media; Qualitative Research; Visual Ethnography

Introduction: Ethnography and Eclecticism

Ethnography has recently moved to centre stage in sociological research. This is undoubtedly connected to the parallel movement made by the category of 'culture' in many disciplines over recent years; ethnography focuses on culture and seems to hold the promise of access to the cultural in all its complexity and depth (Van Maanen, 1995). There is a strengthening conviction that culture requires a form of 'thick description' which does not reduce it to purely functional roles. As a result, strategies for researching culture have been required to address this complexity and to enable its full specification. In spite of, or perhaps because of, the longstanding debates over its definition and raison d'Étre, ethnography promises to be the kind of research practice that can rescue the many layers of interpretation involved in researching culture (Denzin, 1997). Thus, recent writing on ethnography has focused on making it more attuned to the recovery of complexity - variously expressed as contingency, multi-vocality, intertextuality, hybridity, and so on.

It is in the spirit of recognising and representing this complexity that the beginnings of a new attention to hypertext and hypermedia are materialising on the horizon of ethnographic innovation[1] (Howard, 1988; Seaman and Williams, 1992; Hagaman, 1995; Coffey et al, 1996). This paper is an attempt to investigate these new beginnings, and, in the process, to offer an account of some work in progress here in Cardiff that is dedicated to exploring the potential of hypermedia technology for ethnographic authoring.[2] This project is in its early stages, and our remarks and observations reflect the contingent nature of our ongoing investigations. We hope, nevertheless, to offer some thoughts on the implications of hypermedia for ethnographers that will help to invigorate debate within this emerging field.[3]

Our analysis starts with the article on qualitative data analysis published in Sociological Research Online by Amanda Coffey, Beverley Holbrook and Paul Atkinson (co-members of our current research team) and the response by Ray Lee and Nigel Fielding which it provoked (Coffey et al, 1996; Lee and Fielding, 1996). We will concentrate on those aspects of the debate which are concerned particularly with ethnographic research. What concerns Coffey et al is that the increasing diversity of current perspectives on ethnographic research is being met by an inappropriate emergent orthodoxy within the field of CAQDAS (computer-assisted data analysis), centred on a preoccupation with 'code- and-retrieve' procedures. It is their contention that the potential of hypertext lies in its ability to escape the constraints of this tendency and to offer a more flexible and reflexive research tool. In their response, Lee and Fielding share the excitement about the potential of hypertext, but take issue with Coffey et al's assertion of the extent to which a code-and-retrieve orthodoxy has taken hold, and are also sceptical of their arguments about CAQDAS' affinity with grounded theory approaches.

We do not want to reopen the debate about CAQDAS and grounded theory here. However, we note that both commentaries converge in the sense that they take it as axiomatic that 'doing ethnography' requires eclectic and multiple perspectives. The disagreement focuses only on how adequately, or otherwise, current CAQDAS applications respond to this requirement. If there is, indeed, any 'orthodoxy' within ethnography at present, both articles offer support for the view that this consists, primarily, in insisting on the complexity of the ethnographic endeavour. It is these demands for ethnographic approaches that can more fully realise this complexity that have helped to ignite current interest in hypermedia. Below, we briefly explore the forms taken by some of these demands, in order to frame our discussion of how hypermedia might be able to address them.

Post-Paradigm Ethnography: Mode of Writing/Object of Study

Two decades of dispute over the legitimacy of traditional paradigms in ethnography has meant that few of its cardinal practices have survived unscathed. Particularly in the decade since the publication of Clifford and Marcus's Writing Culture, uncertainties over its subject matter, its method, its medium and its intention have all been publicly aired (James et al, 1997). Two of these areas of contention, however, have attracted the most intense critical attention. These are, respectively, its medium and its subject-matter - reflected in the large number of critiques directed at experimentation in ethnography's mode of writing, and a concern to redefine its object of study. We shall briefly examine these, in order to reflect on the current state of ethnographic methodology and on the lessons for any putative 'hypermedia turn' in ethnographic authoring.

The Retreat of the Author

Ethnography has been characterised of late by a form of radical doubt, which has made some writers wary of producing strongly 'authored' narratives. This anxiety has produced a new attention to the poetics of representation, particularly in anthropological but also in sociological applications of ethnography (Atkinson and Coffey, 1995). Strong authorial narratives seem increasingly to be a risky venture, open to charges of ethnocentrism through their claims to an authority that appears spurious and ill won.[4] This authority is understood to derive instead from a literary style - classical realism - which is inherently monologic and in which the authorial voice subdues the voice of the other (Atkinson, 1990). This has resulted in a swathe of new writing that shuns a strong authorial tone in favour of writing styles more favourable to the new mood of contingency. Claiming the title of 'author' suggests an egoistic desire to cover the ethnographic 'subject' with one's own fingerprints, thus obscuring or repressing its diversity and complexity.

The new 'hands off' approach to writing ethnography adopts various textual strategies to let the diversity and complexity of the ethnographic subject shine through (Marcus and Fischer, 1986). Writers/ethnographers may 'put themselves in the picture' in order to lessen their distance from their 'subjects' and to rupture the fiction of authorial invisibility.[5] Or there may be an attempt to foreground the heterogeneous perspectives of ethnographic participants, through an attention to 'showing' rather than 'telling'. Both reflexivity and voice have thereby become central terrains of anxiety in the production of ethnographic texts (Hertz, 1997). Whilst the traditional stance of authorial control aspires to the possibility of objective interpretation, experimental writing self-consciously embraces a more relativist subjectivity. It foregrounds the inescapability of the self/other relation as a de-authenticating mechanism.

However, this experimentation with different forms of writing inevitably runs up against an obstacle - that writing does ultimately mean authoring. James Clifford, a leading exponent of 'post-paradigm' ethnography, acknowledges that representing different voices ends up 'confirming the final virtuoso orchestration by a single author of all the discourses in his or her text', and even the 'utopia of plural authorship that accords to collaborators ... the status of writers' still fails to dislodge the ethnographer/subject relationship (Clifford, 1988: pp. 50 - 51). As Sangren (1988) notes in his (mostly overstated) critique of the 'rhetorical turn' in ethnography, there can be detected in these experiments a fantasy of direct, transparent representation, in which all power relations have been overcome through an innovation which is merely textual. We need to recognise that ethnography is a form of mediation, which inevitably translates one form of knowledge (that of the research 'subjects') into another (that of the author/readers). In this sense, 'there is no textual format that pictures the social world as a perfect simulacrum' (Atkinson, 1992b: p. 7).

The Breaching of 'The Field'

The other major focus of 'post- paradigm ethnography' (Marcus and Fischer, 1986) has been the questioning of the category of 'the field', traditionally the term that covers the object of ethnographic study. Many writers have pointed out that this term implies a bounded entity that can be entered and exited, and which exists as an objective 'place' with its own time and social/cultural character (eg. Fabian, 1983). Sociologists have long been suspicious of the notion of the spatially-defined 'community' as an object of study, and anthropologists, too, are now supplanting it in favour of 'multi-site analysis' which is more attuned to the mobility of social forces rather than to relations of fixity and 'dwelling' (Clifford, 1997). Writings on globalisation have been particularly influential here, showing that local culture is fully implicated in wider processes of political and economic change. Ethnographers, therefore, need 'to place their subjects firmly in the flow of historic events' (Marcus and Fischer, 1986: p. 44).

These critiques have helped to redefine the object of ethnographic study by questioning the project of representation that lies at its heart. Ethnographic subjects do not 'speak' their identities directly, but are caught up in much wider cultural and economic systems of distinction. Concepts such as intertextuality and intersubjectivity (Barthes, 1977; Pecheux, 1982) have suggested that ideas, images and discourse overspill the confines of any one set of texts, so that the meanings voiced by research subjects can not be understood as being 'theirs' - a uniquely characteristic local trait defining 'the people of place x'. In this perspective, the 'field' actually splits into different sites of enquiry. Clifford (1997), for example, cites Susan Harding's (1994) research into Christian Fundamentalism, which involved analysing newspaper and televisual discourse as well as the talk of research participants. Thus, 'the field' no longer equates to a temporal, spatial and social unity.

A More Holistic Ethnographic Practice

These are, then, two related areas of concern within 'post-paradigm ethnography': to rethink how ethnography's subject-matter is defined, and to radicalise how is written. Both require attending to the complexity of, firstly, what we study and, secondly, how we represent it. To some extent, these two tendencies may work against each other. For example Van Maanen (1995) distinguishes between different kinds of ethnography: 'critical ethnography', 'confessional ethnography', 'dramatic ethnography' and 'auto-ethnography'. Critical ethnographies seek to locate the practices under study within their wider social, historical and symbolic context, whilst the others stay much closer to the participants' own sphere of action. Whilst reflexive ethnographic writing may seek to foreground the personal intersubjectivities of participants and researcher, critical ethnography rather demands a less personalised, more 'authorial' mode of representation. Critical ethnographies deploy a meta- commentary that interweaves the voices of ethnographic participants with the wider context of the study, but this is not the case with ethnographies preoccupied with 'giving voice' to participants within a particular research setting. Some of the writings specifically aimed at recovering and revaluing the voices of participants can appear excessively personalised, psychologistic and de-moored from their wider contexts.[6] There is, however, no reason why critical commentary and explication can not be combined with the presentation of voices and dialogues in ways that do not fall back on traditional legitimising claims of 'objectivity' and 'subjectivity'. Reflexivity is not about pretending there is no difference between the voices of participants and that of authors; instead, it involves acknowledging and foregrounding the inevitable processes of selection and interpretation that make 'authoring' what it is. Hypermedia, in our opinion, can only aid such a project.

There is currently a danger that ethnographic writing splinters into a plethora of different genres, and thereby abandons its most valuable attribute - the ability to knit together a layered account of the many and varied practices that make up the social world. If we remain committed to making ethnographic accounts more embedded in the social, ethnographic writing should aim to reunite these different foci in a more textured form. This may involve a post-disciplinary reconceptualisation of the boundaries of ethnographic work and a (continued) decentring of the method of participant observation. For example, producing the kind of deterritorialised, messy and complex ethnographic text that is called for in much current methodological writing (Marcus, 1994b) requires rethinking the traditional observational constraint on ethnographic practice. Put another way, ethnography may now seek to document social processes and forces that are not immediately visible to the ethnographic eye within the ethnographic field.[7] This means attending to sites of meaning which are not confined to what people say or how they can be observed to act.

At the same time, ethnography is searching for a style of writing that can do justice to these diverse forms of data without reducing them to a 'thin' narrative that squeezes social practices into one plane of meaning. As the ethnographic endeavour expands and becomes more flexible, so traditional forms of writing and the academic monograph struggle to articulate it. Since ethnographic writing is an academic sense- making practice which is not content with merely assembling a data-base of empirical materials, the dilemma arises of how to maintain a structure of argumentation whilst accommodating a more multi-layered object of study. The interpretation of data and the dissemination of this interpretation are essential to the logic of doing academic research. Thus, there is an ongoing and unsolvable tension between the urge to complicate research findings and explore more dialogic forms of representation, and the countering urge to grant intellectual coherence to these findings by assimilating them within a line of argumentation.

Hypermedia: Hypertextual and Mixed Media Authoring

In the light of these difficulties, it is not surprising that hypermedia has begun to be recognised as a potentially liberating form of ethnographic representation. We are, however, alert to the fact that these intellectual expectations which have attached themselves to hypermedia may too easily declare it as a long-awaited solution to the dilemmas of representation alluded to above. Hypermedia resources should be deployed as a means of exploring these ongoing tensions, rather than by attempting to side- step them. This means, for instance, that the EHE author will still confront important questions of how to reconcile authorial interpretation and voices 'from the field'. However, the task is made easier by avoiding some of the overblown rhetoric that hypermedia has attracted. Some of these hyped-up expectations derive from a rather vague and ill-defined notion of what 'hypermedia' consists of. Our next step, therefore, is to discuss, briefly, some of the difficulties with the terms 'hypertext' and 'hypermedia'.

The simplest definition of hypertext is 'Hypertexts are electronic documents, read on the screen of the computer' (Bolter, 1993: p. 21). This notion is expanded by Jeff Todd Titon, who claims that:

Hypertexts are non-linear. Several writing spaces can appear on the screen simultaneously.... In a hypertext, the reader is always offered multiple pathways through the information, and the reading will be different depending on which pathways are chosen and what is read and not read. (Titon, 1995: p. 441)

Our working definition of hypertext conceptualises it as a form of text that is computer-mediated and contains authored 'links' that create associations between different elements in the hypertext. The term 'hypermedia' is conventionally used to describe hypertexts which incorporate other media such as video, photographic images, sound, graphics, and so on. Jakob Nielsen (1997) argues for keeping the traditional term 'hypertext' for all systems, regardless of the media they use, but writers such as Landow and Delany (1991) argue for the use of the term 'hypermedia', in the recognition that mixed media can qualitatively transform the nature of the text itself, although more recently George Landow has declared that the terms should be used 'interchangeably' (Landow, 1997: p.3). For the purposes of this article we shall use hypermedia to refer to a particular type of hypertext, namely one that includes a wide variety of media other than plain text.

Hypermedia potentially favours an expanded and more complex object of study, as well as inviting an experimental mode of authoring. These potentialities are enshrined in two principal advantages that hypermedia can offer the ethnographer. Firstly, there is the possibility of creating all kinds of multiple links between both the data assembled and the interpretative texts which comment upon these data (Howard, 1988). This facility allows the object of study to breach the boundaries of the research setting itself, since connections can be made with all kinds of intertextual resonances in mind. Different types of interpretation can be accommodated, so that both the voices of participants and the author's commentary can be more creatively integrated. For example, most hypertexts allow the creation of 'paths' through the hypertext with appropriate labelling, so that the linkages and ruptures between interpretation, the data presented and the potential 'intertexts' of the ethnography itself can be more explicitly foregrounded. Whilst these pathways are designed to guide the reader in the direction of authorial argumentation and/or suggestion, the very accessibility and 'proximity' of the data texts may open up channels for innovative interpretation and reinterpretation - both in the analytic phase and in the presentational phase.

Secondly, there is the provision for readers to trace their own paths through these chains of links. As soon as one introduces multiple links into a hypertextual document (rather than merely having a linear sequential link from one 'page' to the next), the author can no longer control how a reader will progress through the environment created, and which directions s/he will choose to pursue. Associations and lines of enquiry can thereby emerge in the act of reading that may not be predicted in advance by the author. Although there is nothing inherent to the provision of multiple pathways or trails in EHEs that will push the reader into constructing pathways of their own, the presentation of interlinking avenues of enquiry and the facility for switching among them aims to encourage readers to approach the ethnographic environment as a shifting matrix of connections rather than a fixed grid of self-contained narratives. However, the actual usage that readers make of such potential remains a matter for empirical investigation, and we hope to make use of technical facilities for mapping and recording the directions that actual readers take.

Thus, hypermedia, potentially, enables both the complexity of the object of study and the mode of its representation to be more fully and flexibly articulated. Of course, a writer can never control how a reader will interact with a traditional printed book either, so we are not suggesting here that a radically new form of communication is enabled by hypermedia environments. Any text is capable of being read in a non-linear mode. In fact, one can argue that a computer-based hypertext is more limiting than a written text. With the latter, one can physically 'link' from any word in the text to any word in any other text whereas a reader of a computer-based hypertext can only follow the links created by the author, rendering the reader less free to create their own interpretations (see Aarseth, 1997: pp. 77 - 78). What is innovative about ethnographic hypermedia environments (EHEs), however, is that the potential for cross-referencing and for multiple linkages is integral to the medium itself, and can inform all phases of the research process. We shall return to the implications of this below.

However, there is another innovation afforded by hypermedia, which has not yet been fully thought through. One of the most exciting promises of hypermedia resides not just in its capability for accommodating non-sequentiality, polyphony and multi-perspectivalism, but in its mixed-media features. The potential for integrating the visual and the textual within the same medium is considerable. It allows us to make the step from thinking of the visual merely as illustrative of argumentation spelled out through the printed word, and to see it as itself constitutive of meaning. Thus we need to consider seriously what hypermedia can do that a well-illustrated and produced book can not. There are potential gains to be derived from exploring how interpretation is simultaneously a verbal and a pictorial, a visual and an aural activity. Ethnographic knowledge can be thought of anew in terms that enrich those traditionally afforded by written narratives.

This potential for integrating the visual and the verbal links in well with recent calls to break down the divide between visual and 'conventional' ethnography (Marcus, 1994a). Traditionally, anthropologists have seen film merely as data to be slotted into a pre-existing classificatory framework, and have failed, according to Marcus, to see film as a different form of mediation. Film allows, importantly, for the representation of simultaneity, so that the 'transcultural space' of the new ethnographies can be shown as different cultural practices participating in physically separate but simultaneously unfolding worlds. Similarly, in sociology, there have been calls for a greater recognition of the role of visual representation in the social world, and warnings of the limitations of focusing exclusively on talk and texts (eg. Chaplin, 1994). In an EHE, academic interpretation can take advantage of different media. Moving images can not only complement the printed word, but can also communicate in a different way. For example, by editing together sequences shot with a digital video camera in 'the field', particular meanings can be encoded which contribute to the intellectual coherence of the environment as a whole.

Each medium - whether still or moving images, printed or spoken words, graphic representations and simulations - brings its own logic of communication. A well-constructed EHE would seek to take advantage of these, instead of seeing them as add-on 'special effects'. For example, the specific modalities of film mean that certain relations of complexity (such as simultaneity) can be opened up, but others are closed down through the 'fixing' quality of the image as opposed to the word (Hastrup, 1992). Film and photography can be nuanced and suggestive, but are conventionally read (at least in the documentary genres) as media that confer a particular experiential authority - the quality of having-been-there (Barthes, 1977). Within an EHE, it is possible to destabilise this assertion of authority of the image by creatively juxtaposing particular video clips with other narratives in the form of printed text, spoken text and other images.

Experimentation has also been occurring in the field of visual ethnography - in the films and commentaries of Trinh -T -Minha, for example. Debates among film-makers about strategies of representational practice have centred on similar areas of concern to those among ethnographers using the printed word: the possibilities of disrupting the conventional camera/object relation, for example, or dislodging the primacy of the ethnographer's narration (Taylor, 1994). Whilst Marcus (1994a) is correct to suggest that film offers new possibilities for exploring wider and more fluid conceptions of meaning (through critical juxtaposition of images, through the visualisation of 'parallel worlds', etc.), film itself is no more inherently dialogic than print. The potential for a more sophisticated representational practice comes rather from the possibilities of combining different representational forms within the same medium. It is through a careful and principled montage of different kinds of 'data' that the potential for new meanings can emerge.

Research and Developments in Hypermedia

There are a number of implications for the conduct of research when the aim is to create an ethnographic hypermedia environment. First of all, it is worth pointing out that there are several co- existing academic and commercial hypermedia systems, carrying their own biases, conventions and dedicated software packages. In general, hypertext has developed separately from the field of CAQDAS, and this has certain implications for its application to qualitative methodology.

Two different research trajectories in hypertext can be discerned. The earliest scholarship focused on using hypertext as a method of indexing huge volumes of data. This research was inspired by Vannevar Bush's article published in Atlantic Monthly in 1945. Bush (1945) believed that the contemporary library systems would prove inadequate for the needs of researchers. He proposed instead a machine he termed a 'memex' that would allow its user to store documents and then create associative trails between the documents. The user then could gather together these trails and store them as documents in their own right.[8] Subsequent researchers have appropriated this concept, which was never built, and claimed that it was a prototypical hypertext system. Most notable of these is Theodore Nelson who, inspired by Bush's article, coined the term hypertext in 1965.[9] This trajectory crucially perceives hypertextual indexing of documents as a way of augmenting a researcher's ability to find and comprehend new information. Specifically, it claims that associative indexing - rather than hierarchical categorisation - is a more natural method of thought, hence Bush's name for his article: As We May Think.

It was not until the 1980s that a new trajectory developed with the emergence of several small-volume hypertext systems. These systems could be used to create small, bounded hypertexts or to link together various different documents on the same computer in a hypertextual format. The most widely accessible was Apple's 'HyperCard' program, which was given away free to any one who purchased an Apple Macintosh computer. Although not specifically marketed by Apple as a hypertext system, HyperCard allowed users to enter information on different 'cards' and then link different cards together in an associative manner (Nielsen, 1997). It appears that some anthropologists used HyperCard as a way to store and manage fieldwork data and journals (eg. Fischer, 1994) but there does not seem to have been any sustained research effort generated that centred on the use of HyperCard.

Two non-commercial systems were developed at academic institutions during the 1970s: 'Guide' at the University of Canterbury, Kent, and 'Intermedia' at Brown University in the USA. Guide was distinguished by its use of a form of hierarchical hypertext; buttons allowed users to 'unfold' texts or display 'notes' about a specific piece of text (Weaver and Atkinson, 1994). Intermedia took a very different approach and can be seen as the first attempt at creating a truly collaborative hypermedia environment (Yankelovic et al, 1988).[10] Intermedia consisted of a series of programs linked together that allowed a user to create and edit text, graphics and video data in different files and then build links between those files. Both these systems were theoretically driven in that they embodied notions of textuality and hypertextuality. Unfortunately Intermedia has been discontinued because it was grounded in a computer operating systems that is rarely used outside of academic and other specialist institutions, while Guide maintains a schizophrenic existence as a "UNIX" based teaching tool at Canterbury University while the Windows version is marketed by a private company as a tool for creating training manuals.

In the last decade there have been two main developments in the field of hypertext. Probably the most obvious has been the emergence of the World Wide Web. Anyone with access to the web can create small volume hypertexts that link to outside documents, although no one can alter documents that belong to other users. At the same time, the increasing availability of powerful personal computers has made hypertext packages such as 'AuthorWare,' 'Director,' 'Toolbook,' and 'Atlas-Ti' commercially viable. Hypertext and hypermedia activity in the UK currently reflects a particular discipline-bound division of labour. In contrast to the widespread belief that hypertext is an inherently liberating form of postmodern authoring, due to its seemingly unstructured nature (eg. Kaplan, 1997; Landow, 1997; Moulthrop, 1993), many of its current applications outside the academy suggest quite the opposite. Hypermedia can equally be used for surveillance purposes, for assessment purposes, and for other aims that are at odds with this image of creative, non-linear thinking.[11]

In education and training, both for companies and for schools, hypermedia systems are aimed at transmitting and testing information in a fairly structured way.[12] Many use Macromedia's 'Authorware' and Assymetrix' 'Toolbook'. Both of these products are focussed on computer-based training (CBT) and computer-based learning (CBL) and thus require strictly constrained environments. The most creative usages of the technology tend be found within graphic design companies, art and design departments in universities and within creative writing courses or writers' networks.[13] Consequently, different kinds of commercial software have been developed to respond to these different demands.[14]

Doing Ethnography the EHE Way

As can be seen above, applications of hypermedia tend to divide into those which see it primarily as a tool for the presentation of knowledge which is already - to some extent - codified and defined (educational packages, graphic design presentations[15]), and those applications that see it as an aid for the accumulation of knowledge about a subject that exists only as 'data' (CAQDAS approaches). Interestingly, this divide is reflected in recent academic writing on hypertextual ethnography: whilst anthropological commentaries have focused on its promise for integrating film and field notes in the presentation of ethnographies (Seaman and Williams, 1992; Howard, 1988), sociologists have concentrated on its potential for enabling more sophisticated approaches to data analysis (Weaver and Atkinson, 1994; 1995; Coffey et al, 1996). However, writings in both areas are still extremely scarce, and we do not know of any sociological or anthropological research project that has been written as an EHE.[16] There will undoubtedly, however, be increasing interest in this area and an awareness of the potential for innovative ethnographic work. Our current research project is designed to explore this potential by creating an EHE of our own.

In pursuing this project, we have tried to integrate the two perceptions just mentioned. In particular, we wish to demonstrate the ways in which constructing an EHE has potential implications for all stages of the research process. We do not want to make extravagant claims for EHEs, and indeed our project is designed to interrogate their usefulness and feasibility rather than to promote them as radical innovations. However, it seems uncontrovertible that there are potential consequences of doing hypermedia ethnography that are not limited to the period of data analysis or presentation. There are two points to be made in this regard, the first of which concerns the relation between analysis and representation and the second to our earlier comments about the object of ethnographic study. The first observation is provided by Weaver and Atkinson in their pioneering work with hypermedia for CAQDAS purposes:

[One] of the most important features of Hypertext applications results in the blurring of the boundaries between 'data' and 'analysis' on the one hand, and between both of those and 'representation' on the other. There is no need in a fully developed hypertext environment, for instance, to force the 'analysis' into the straightjacket of a single, monologic and linear textual format. ...The use of Hypertext software to 'author' materials as well as to undertake analytic tasks, means that a highly flexible set of relationships can be facilitated: between the ethnographer and the data (and other materials) on the one hand, and between the 'reader' and the ethnography on the other. (Weaver and Atkinson, 1995: p. 165).

In other words, a radical possibility offered by EHEs is that they can be constructed in such a way as to enable both analysis and presentation in the same medium.[17] In order to make this point clearer, let us first clarify what are the procedures of analysis facilitated by hypertextual approaches.

Data analysis and Hypermedia

As Coffey et al (1996) and Kelle (1997) point out, hypertext links facilitate a cross-referencing approach to data analysis, whereby segments of data (in different forms) can be linked to each other in order to posit thematic and associative connections. This cross-referencing approach can be distinguished from the strategy of indexing - the basis of the code- and-retrieve CAQDAS software packages (c.f. Bush, 1945). Coffey et al suggest that indexing/coding, while a useful strategy for assimilating and organising one's data, should not become the only accepted approach for data analysis, and they fear that many CAQDAS programmes work to create such a consensus. Kelle (1997), on the other hand, points out that both cross-referencing and coding have often been used simultaneously and in a complementary way for the understanding and interpretation of data. Both techniques offer particular advantages to the analyst.

Weaver and Atkinson are ready to acknowledge that coding data results in a 'comprehensive and detailed reflection on the structure and content of the data' and encourages the analyst 'to be alert as to possible relationships between the categories ... devised' (Weaver and Atkinson, 1995: p. 149). They are concerned however that coding may encourage a 'culture of fragmentation' (Atkinson, 1992a), in which the decontextualisation and subsequent recontextualisation of data risks obscuring how text segments are framed within the wider meaning-structures of the text as a whole. Using hyperlinks rather than cut-and-paste techniques can certainly help to avoid losing sight of the 'wider picture', since hypertext is not predicated on the segmentation of data. Instead, links are set up between different sections of, parts of, or whole documents (or videos, sound clips, etc.) without needing to remove them to a new location. Thus, each time the analyst considers the data afresh, s/he will be attending not to segments of decontextualised or recontextualised data, but to the original sources. This potentially allows for a more embedded and holistic analytic approach.

It is true, of course, that the very process of constructing links requires the analyst to have a prior familiarity with the data. Kelle (1997) suggests this is difficult to achieve unless some coding has been carried out. However, in both coding/indexing and hyperlinks/cross-referencing, the analyst's knowledge of the data is successively built up: his/her familiarity with the data is extended and refined through the traditional techniques of intensive reading and manual note- taking in conjunction with the identification of data clusters. In standard code-and-retrieve procedures, the analyst scans the text marking significant text-segments with a tag. The choice of significant categories and which segments are assigned to which is the result of an ongoing process of interpretation by the analyst as s/he moves through the data sets. This interpretative process also pertains to the construction of hyperlinks. In other words, hyperlinks are built up in the same way as codes are - by a combination of computer-aided functions, and the analyst's intensive immersion in the data and application of relevant meta-knowledge. The advantage offered by tracing hyperlinks rather than coding lies in the fact that the data remain contextualised throughout, and more complex and multiple links can thereby emerge.

In both types of analytic approach, analysts are progressively building up their knowledge of the data. This interpretative process produces a set of intellectual categories and connections that are progressively refined as the data yield deeper levels of insight. This is the period during which discoveries are being made about the data, and significant connections are being noticed. Each emergent category can - during the same period - be written as a short interpretative text, or as an 'analytic memorandum' that serves to define it. Most state-of-the-art CAQDAS software allows these memos to be attached to the segment cluster in question. In hypertextual software, however, these interpretative texts can become quite complex early on, and can be linked in to any other text within the system. Data familiarisation and interpretation can thus proceed apace, without the need for data to be segmented into different locations with attached memos.

In non-hypertextual approaches, when the analyst is satisfied that the categories are as complete and appropriate as possible, s/he moves on to the fourth stage of the research process: constructing a narrative that weaves these categories together. It is at this point that some of the problems of having to rely on linear writing styles may well begin to impress themselves. Most analysts will have had the sensation of understating the complexity of the inter-relationships of these categories by translating them into a linear printed narrative. In hypertextual approaches, on the other hand, the process of analysis has resulted, ideally, in the construction of a linked web that is accessible in that same medium to the reader. In practice, however, we are strongly sceptical that, in the hypertextual approach, there is no need for an intervening period of 'authoring' between analysis and presentation. In reality, much work will go into the construction of links that enable a coherent line of argumentation to be pursued by the reader. However, these pathways or trails consisting of linked data/interpretative texts will be multiple and themselves interlinked. In this sense, rather than abandoning linearity, one is offering the reader an environment in which to pursue multi-linear lines of enquiry.

Hypermedia and the Object of Study

The second observation is connected to our earlier remarks concerning the redefinition and expansion of ethnography's object of study. In the staged research process, which conventionally ends with the production of a printed narrative, the nature of the final product sets the parameters of the preceding stages of research. In other words, analysis and fieldwork are conducted in the knowledge that the end product will consist of a monograph with chapters and a sequential structure. Equally, research conducted with a different end product in mind - namely a fully-fledged EHE - will have implications for how each stage of the research process is conceived. In hypermedia approaches, one knows in advance that:

All these assumptions will have their implications for how the object of study is conceived. In particular, we can postulate that a more deterritorialised and multi-layered field of meaning can emerge as the object of study.

Therefore we suggest that hypermedia should not be seen as either a tool for analysis, or a tool for presentation. Rather, it should be seen as offering an approach to research practice that is capable of accommodating some of the calls to redefine ethnography that we discussed earlier. Unfortunately, it appears that currently available software is not yet able to provide both the rigorous and disciplined aides for data analysis offered by Atlas/Ti or NUDIST, along with the impressive tools for presenting and structuring linked pathways offered by Authorware or Director, within one integrated package. The division of labour between presentation and analysis appears to continue within commercial applications. However it seems likely that new hypermedia authoring packages will be developed that allow for a more creative and flexible integration of analytic and presentational aspects.

Case Study: Preparing a Research Project in Standard Form and EHE

In this section we will offer some initial thoughts from our current project in Cardiff to indicate how some of the above issues might be tackled. In preparing to construct an ethnographic hypermedia environment, we have been aware of the need to design a project which brings with it all the usual complexities, ambiguities, multiple layers and theoretical depth of a well-conceived and original piece of research. Only by doing this could we claim to have explored fully the challenges and demands of 'doing hypermedia ethnography'. We did not want to conduct a hastily executed project that would simply serve the ends of our methodological aims, namely to reflect on the process of constructing an EHE. However, funds have been allocated, understandably, for fulfilling this methodological aim, and not for the conduct of a substantial new piece of ethnographic research.

Our solution was to work with two sets of data. The first has already been collected, interpreted and written up as a conventional research monograph.[18] The second set of data comprises an extension of that first project, and takes that work in a new direction. Both sets of data will be combined within the EHE that we are preparing to construct. This will allow us to fulfil two objectives:

  1. To consider how representing existing data/interpretation in hypertextual form transforms that interpretation and encourages different readings of it. This objective illuminates the relation between presentation and reading within a conventional and a hypermedia ethnography.
  2. To consider how conducting a self-contained piece of ethnographic research with the aim of constructing an EHE transforms all stages of the research process.

We are currently preparing the 'old' data for integration into an EHE using a mixture of Macromedia's 'Authorware' software package and Eastgate's 'StorySpace.' This has meant rethinking the ways in which those data and interpretation are structured. During analysis of the different types of data in that earlier project, it became clear that clusters of these data actually revealed several layers of knowledge. Thus, as usually happens, the theoretical model originally employed had to be expanded and complicated in order to accommodate these extra layers. In the process, the relation between different types of data was shown to be a complex and nuanced one, which required the insertion of new levels of interpretation - redefining the theoretical model on which the research was based.

It is not necessary here to describe the earlier project in detail. However, executing the project brought home how much more complicated the object of study becomes through the process of research. In Dicks' research, the original theoretical model that the project was designed to explore still held considerable value after analysis was completed but the range of different social practices and meanings that needed to be brought into the picture had widened considerably. In other words, the various strands of analysis pursued made the representation of these theoretical principles more complicated. Indeed, it meant that a rather messy and overcrowded graphic diagram was devised, which struggled to contain the various aspects of analysis required of it. When constrained to the flat two- dimensionality of the printed page, the attempt to represent graphically the theoretical relation between aspects of analysis encounters a dilemma: how to do justice to the complexity of theoretical knowledge whilst communicating the essential relations of that knowledge in such a way as to enable readers to grasp its principles. In opting to convey clarity, the complexity of theoretical relations can be sacrificed. The very flatness and two-dimensionality of the printed page can - in our experience - encourage the analyst to conceive of the research project in flat, two-dimensional terms.

In an EHE, on the other hand, various relations of three-dimensional 'depth' can be readily configured through the logic of cross-referencing using multiple hyperlinks. Interpretation and data can be interlinked in a way that suggests multiple relations of knowledge, rather than a narrative sequentiality relying on a 'if that, then this' ordering. In addition, the inclusion of data in different media allows some of these relations to be demonstrated through imagery and sound as well as through the printed word or diagram. What we are suggesting is that there may be consequences for how theoretical models are conceived once their graphical representation is no longer confined to a single-medium. We are not suggesting that there are epistemological consequences per se, since to do so would be to posit a strong form of determinism whereby thought is defined by and confined to the conventions of particular representations. In fact, we argue the opposite: that since the human mind is clearly capable of making multiple linkages and connections, and since the social world itself requires flexible and multi- faceted analysis, then the creative integration of different media may offer the reader and analyst a more adequate approximation of the richness of ethnographic knowledge.

Concluding Reflections on Constructing an EHE

The above discussion is intended only as a starting point for considering the implications of constructing EHEs for ethnographic knowledge. What we want to try and achieve with the current project is an open-minded assessment of the nature of these implications. In particular, we want to try and chart a path through the nascent field of interest in hypermedia that tries to avoid some of the polarised positions which have already emerged - for example the debate over whether hypertext impoverishes the reading experience (Birkerts, 1994). We are aiming for a creative integration of technical innovation and ethnographic authoring. For example, whilst we are attuned to the potential of non-sequentiality enshrined in hypermedia applications, we also remain committed to the production of a form of authoring which maintains intellectual coherence. In other words, the EHE must be capable of guiding the reader along pathways that make explicit certain extrapolations and bring in the benefits of the authors' meta-commentary.

In this sense, we do not discard linearity as such, but seek to offer multi-linear pathways that make up to a web of sense-making trails. This web is more than the sum of its parts. The reader can see how the same data are implicated and included in different interpretations, and can thereby retain a sense of the contingency and complexity of analysis. The data are no longer hidden from view, archived in the researcher's filing cabinets, spatially and temporally removed from the analysis and the reader.[19] Instead, the data and the analysis are embedded within the same medium and are thereby equally accessible to the reader. However, they are not simply offered to the reader as a labyrinthine web, the threads of which the reader is left to trace blindly, with no map of where s/he is going or what choices are available. Instead, we propose that hypermedia authoring involves the construction of a set of carefully mapped trails through the data/analysis, which are made explicit to the reader. The reader can see how these different interpretative trails relate to each other, and can cross from one to the other with relative ease. Ultimately the reader should be able to create her/his own trails through the EHE and have these saved by the program so that the EHE evolves with time as the reader becomes something akin to a co-author.[20]

The major challenges which we foresee in ethnographic hypermedia authoring concern how to render these trails explicit, and how to ensure that the reader can remain on course, or retrace his/her steps, without losing direction and becoming 'lost in hyperspace'. In this regard, the consistency and quality of the navigational tools created for each particular EHE are of crucial importance. Particular navigational aides can be designed to ensure that the various trails are clearly signposted, that pathways followed can be retrieved, that departures embarked upon can be logged, and that the intersections of different trails are made explicit. In addition, the question of how far the authors attempt to map the entire EHE remains to be explored. The benefits of having the entire web of interconnecting texts made graphically explicit on the screen - say in a three-dimensional diagram - may be outweighed by the risks of attempting too complete and complex a representation.[21] A sense of the whole environment should rather be able to emerge from the reader's explorations of it. Nevertheless, some form of mapping will almost certainly be required, and the form that this takes will be a matter for careful thought.

Creating an EHE involves undertaking a form of authoring which is collage-like in nature, and thus has aspects in common with other collage-type forms of communication: television, radio and newspapers/magazines, for example. These have long co-existed alongside predominantly single-author media such as books and films. Hypermedia can not aspire to the same effects as the single-authored narrative, and should not attempt to do so. Instead, we can expect both types of communicative medium to continue to co-exist and to offer their respective advantages. However, we can also expect them to evolve and change in relation to each other as new technologies become increasingly widespread. What hypermedia potentially offers is a more hybrid form of representation which allows both the collage principle and the more authored kinds of meta-narrative to be integrated within the same medium (or rather within the same multi-media form). As word-processing powers expand, the single-authored text will inevitably start to become more 'leaky' at the edges. For example, Microsoft's Word 97 already allows:

The potential for making texts more complex and less contained is clearly already with us.

What we see as the potential for hypermedia authoring is its capacity to allow a deeper and more rigorous exploration of ethnographic knowledge and a more faithful representation and communication of that depth to readers. If hypermedia were to be used merely for the stringing together of different voices and perspectives, this would not be offering much more than a collage-effect postmodern video could achieve. Instead, what hypermedia allows is the ability to integrate different types of data, interpretation and lines of enquiry, so that the interweaving of ethnographic voices and authorial commentary can be explored and opened up for scrutiny by the reader. EHEs can offer a means of further interrogating the ongoing contours of debate over reflexivity and representation. Instead of relying on the hierarchical organisation of printed narrative, where authorial commentary appears to 'enclose' the voices it cites, hypermedia opens up the possibility of innovative re-organisations of these relations both by the author and by the reader. As a result, hypermedia may allow the ethnographic object of study to be seen as more fully embedded in its social (cultural, economic, political) context, and the analysis to proceed in a more flexible manner through a thicker level of description. The objective is to allow the reader to move more easily between the general picture and the particular local situation, to adopt both the 'jeweller's eye' as well as the 'bird's eye' view of the social world (Marcus and Fischer, 1986). The actual mechanics of achieving this are mired, of course, in the interface between ethnographers and currently available technology. As authors, we recognise that the task is a potentially daunting one, fraught with difficulties of both a technical and interpretative nature. We'll let you know how we get on.


1 These terms are often used interchangeably. See the discussion in ¶3.1.

2 The project is funded by the ESRC. For a brief introduction to the project see our homepage on

3 There is a deep irony involved in writing about hypertext in this medium that is exemplified by the nature of an 'electronic' journal such as this. Academic status is print-based and grounded in the convention of linear argumentation; consequently this journal is formatted as a virtual paper journal with each article as a separate file with scholastic endnotes and a reference list. In addition each article, ours included, is composed as a linear exposition. Essentially, this article is a recreation of standard text in a hypertext environment that deals with the advantages of hypertext. We recognise that this is a far from ideal position; however, rather than trying to disguise this we have decided to remain within a conventional scholastic format in the presentation of our ideas. Experiments with scholarly articles as hypertexts will be essayed at a different time.

4 Although, particularly in anthropology, these new critiques have not gone uncontested (see Roth, 1988; Sangren, 1988).

5 For an example of such an 'auto-ethnographic' approach, see the rather unusual, and - in our opinion not very successful - article by the unnamed authors Communication Studies 298, 'Fragments of Self at the Postmodern Bar', (Communication Studies 298, 1997).

6 We tend to work within a paradigm of ethnographic investigation that can be identified as critical ethnography and our exploration of hypertext is, naturally, biased to that end. We do not think, however, that reflexivity and an exploration of multivocality are excluded from such a project.

7 We do not mean to suggest that the ethnographic viewpoint has been confined to the manifest visible/audible ephemera of social life. In sociology, participant observation is not seen as a cardinal tenet of ethnography in the same way that it has been in anthropology (Hammersley and Atkinson, 1983). Anthropology, too, has grappled with the problems of the boundaries of 'the field'. Interpretative anthropology, in particular, has stressed how observed behaviour and discourse 'in' the field have to be related quite systematically to other contextual aspects of the culture under study in order to produce 'thick description' (Geertz, 1975). However, this approach still sees observed behaviour as an index of 'wider culture', in that the observed field 'contains' the local manifestation of these wider rules and systems. It does not, thereby, invite the kind of eclectic attention to non-visible practices and discourses that have recently been called for, which may not be coterminous with the categories of observed behaviour.

8 Interested readers can select the following link from which they can attempt to download an animated example of a memex for Windows or Macintosh systems: Be aware that this site has bandwidth problems and we have never successfully downloaded the demonstration.

9 Nelson's system, called Xanadu is proposed as a method for storing every text ever published along with every alteration, comment or annotation to that text in a hypertextual format. This is intended to create a 'docuverse'.

10 George Landow's home page has extensive information about the Intermedia project. Follow this link for more detailed information.

11 For an interesting counter-argument to claims by Landow et al that hypertexts constitute a form of post-structuralist authoring see the online article Hypertext and Reading Cognition by Alec McHoul and Phil Roe

12 For example, in one case study we investigated, an outside consultant in environmental risk management was asked to generate an Environmental Management System (EMS) for an aluminium producer. European Union policy requires that the company must possess a quantifiable training system for workers at the plant about the EMS. The consultant decided to produce a hypermedia tutorial - with associated tests built in - on the grounds that, unlike a book which a worker could simply ignore, a hypertext requires the worker to sit down at a computer. In addition, by continually asking the user to press keys to continue as well as answering regular questions the user had to interact with the computer in order to get to the end. Because the hypertext used timed speech files the user was unable to skip past sections and was forced to plough through each section. Even the 'back' button option, which allows a user to return to a previous section, was removed. Supervisors were always on hand to ensure that the user was proceeding through the package in an orderly and attentive manner.

13 Not all pedagogical uses of hypermedia are, however, constraining. There appears to be a movement in literary pedagogics that encourages the creative use of collaborative hypermedia environments, inspired by the likes of George Landow and Michael Joyce.

14 The most popular of the currently available packages are:

Macromedia 'Authorware' and Assymetrix 'Toolbook': Used primarily for computer-based training (CBT) and computer-based learning (CBL), these programs tend to define user-computer interaction within an interactive question and answer format. and for the respective products.

Macromedia 'Director': This has been very popular in the field of corporate presentations and in graphic design. It focuses on synchronising different media to enable slick demonstrations.

Scolari 'Atlas-Ti': This program acts more as a theory developer in a similar vein to CAQDAS products such as NUDIST and the Ethnograph. It focuses on enabling hypertext mark-ups to code data in various media.

Eastgate 'StorySpace': This program has emerged from literary research into hypertext; one of its lead designers is Michael Joyce the noted author of avant-garde hypertexts, and its focus is on enabling writers to create literary hypertexts.

15 Creative writing applications are a rather special case, since they are not concerned with communicating knowledge as such but with the manipulation of language for creative purposes.

16 An exception is Saskia Kersenboom, whose 1995 text Word, Sound, Image: the Life of the Tamil Text has an accompanying CD-ROM. However, this CD-ROM is not claimed to be fully hypertextual, and is rather conceived as being illustrative of the various conventions of Tamil dance and music with which the book is concerned.

17 Although, as noted above, this is still a largely unrealised potential.

18 This project comprises Dicks' PhD thesis (Dicks, 1997). This was the study of a coalmining heritage park in South Wales and offers a three-pronged account of the site's significance: an investigation of how it came to be set up and its relation to local definitions of historical knowledge (encoding), an analysis of its displays and narrative structure (text), and a small scale interview/focus-group based investigation of readings made by its audiences/visitors (decoding). This was conceived as a multi-layered study requiring an eclectic range of methods and strategies of enquiry.

19 That said there are huge issues surrounding this point. For example our fieldwork has generated about twenty hours of video footage to date which would require about twenty CD-ROMs to store in digital video format. Our current estimate is that if we produce an EHE capable of fitting on a single CD then we will be able to include a maximum of 30-35 minutes of video footage. Even with the emergence of new media storage devices (eg. DVD) it is unlikely that in the near future it will be possible to make all the data of an ethnographic project available in an easily accessible form for the reader. Nor is it necessarily desirable to make all data available due to issues of confidentiality and so on.

20 Again this is more of a potential than a realised ability. For example all commercial hypertext authoring programs permit the distribution of hypertexts with a 'reader's' version of the authoring program which allows the reading of the hypertext but not its authoring. Although programs such as StorySpace do allow readers to save annotated 'readings' of the hypertext the author can never make new links, add or delete material unless in possession of a full version of the program, StorySpace.

21 For example, one of the debates surrounding StorySpace is in the appropriateness and usefulness of the 'document web' overview that each hypertext generates.


AARSETH, Espen (1997) Cybertext: Perspectives on Ergodic Literature. Baltimore, MD: John Hopkins University Press.

ATKINSON, P. (1990) The Ethnographic Imagination. London: Routledge.

ATKINSON, P. (1992a) 'Reading, Writing and Rhetoric: The Ethnography of a Medical Setting', Qualitative Health Research, vol. 2: pp. 451 - 74.

ATKINSON, P. (1992b) Understanding Ethnographic Texts. London: Sage.

ATKINSON, P. and COFFEY, A. (1995) 'Realism and its Discontents: On the Crisis of Cultural Representation in Ethnographic Texts' in B. Adam and S. Allan (editors) Theorising Culture. London: UCLA Press.

BARTHES, R. (1977) Image-Music-Text. New York: Hill and Wang.

BIRKERTS, S. (1994) The Gutenberg Elegies: The Fate of Reading in an Electronic Age. London: Faber and Faber.

BOLTER, Jay David (1993) Writing Space: The Computer, Hypertext, and the History of Writing. Hillsdale, NJ: Lawrence Erlbaum.

BUSH, Vannevar (July 1945) Atlantic Monthly, no. 176, pp. 101 - 8. Reprinted in Steve Lambert, and Suzanne Ropiequet (editors) (1986) CD ROM: The New Papyrus. Seattle: Microsoft Press.

CHAPLIN, E. (1994) Sociology and Visual Representation. London and New York: Routledge.

CLIFFORD, J. (1988) The Predicament of Culture: 20 Century Ethnography, Art and Literature. Cambridge MA: Harvard University Press.

CLIFFORD, J. (1997) Routes: Travel and Translation in the Late Twentieth Century. Cambridge MA, London: Harvard University Press.

COFFEY, A., ATKINSON, P. and HOLBROOK, B. (1996) 'Qualitative Data Analysis: Technologies and Representations', Sociological Research Online, vol. 1, no. 1, <>.

COMMUNICATION STUDIES 298 (1997) 'Fragments of Self at the Postmodern Bar', Journal of Contemporary Ethnography, vol. 26: pp. 251 - 92.

DENZIN, N. (1997) Interpretive Ethnography: Ethnographic Practices for the 21st Century. Thousand Oaks, London, New Delhi: Sage.

DICKS, B. (1997): 'The View of Our Town from the Hill: An Enquiry into the Representation of Community at the Rhondda Heritage Park'. Ph.D. Dissertation, University of Wales, Cardiff.

FABIAN, J. (1983) Time and the Other: How Anthropology Makes its Object. New York: Columbia University Press.

FISCHER, Michael D. (1994) Applications in Computing for Social Anthropologists. London: Routledge.

GEERTZ, C. (1975) The Interpretation of Cultures. London: Hutchinson.

HAGAMAN, D.D. (1995) 'Connecting Cultures: Balinese Character and the Computer' in S. Leigh Star (editor) The Cultures of Computing. Oxford: Blackwell Publishers/Sociological Review.

HAMMERSLEY, M. and ATKINSON, P. (1983) Ethnography: Principles in Practice. London and New York: Routledge.

HARDING, S. (1994) 'Further Reflections', Cultural Anthropology, vol. 9, no. 3, pp. 276 - 78.

HASTRUP, K. (1992) 'Anthropological Visions: Some Notes on Visual and Textual Authority' in P.I. Crawford and D. Turton (editors) Films as Ethnography. Manchester: Manchester University Press.

HERTZ, R. (editor) (1997) Reflexivity and Voice. London: Sage.

HOWARD, A. (1988) 'Hypermedia and the Future of Ethnography', Cultural Anthropology, vol.3, no. 3, pp. 304 - 15.

JAMES, A., HOCKEY, J. and DAWSON, A. (editors) (1997) After Writing Culture: Epistemology and Praxis in Contemporary Anthropology. London and New York: Routledge.

JOYCE, Michael (1995) Of Two Minds: Hypertext Pedagogy and Politics. Ann Arbor: University of Michigan Press.

KAPLAN, Nancy (1997) 'Politexts, Hypertexts, and Other Cultural Formations in the Late Age of Print' online at

KELLE, Udo (1997) 'Theory Building in Qualitative Research and Computer Programs for the Management of Textual Data', Sociological Research Online, vol. 2, no. 2, <>.

LANDOW, George P. (1997) Hypertext 2.0: The Convergence of Contemporary Critical Theory and Technology. Revised edition. Baltimore, MD: John Hopkins University Press.

LANDOW, George P. and DELANY, Paul (1991) Hypermedia and Literary Studies. Cambridge, Mass: MIT Press.

LEE, R.M. and FIELDING, N. (1996) 'Qualitative Data Analysis: Representations of a Technology: A Comment on Coffey, Holbrook and Atkinson', Sociological Research Online, vol. 1, no.4, <>.

MARCUS, G.E. (1994a) 'The Modernist Sensibility in Recent Ethnographic Writing and the Cinematic Metaphor of Montage' in L. Taylor (editor) Visualising Theory. London and New York: Routledge.

MARCUS, G.E. (1994b) 'What comes (just) after "Post"? The case of Ethnography' in N. Denzin and Y.S. Lincoln (editors) The Handbook of Qualitative Research. London: Sage.

MARCUS, G.E. and FISCHER, M. (1986) Anthropology as Cultural Critique: An Experimental Moment in the Human Sciences. Chicago: University of Chicago Press.

MOULTHROP, Stuart (1993) 'You Say You Want a Revolution: Hypertext and the Laws of Media' in Eyal Amiran and John Unsworth (editors) Essays in Postmodern Culture. Oxford: O.U.P.

NELSON, Theodor (1987) Computer Lib.: Dream Machines. Seattle: Microsoft Press.

NIELSEN, Jakob (1997) Multimedia and Hypertext: The Internet and Beyond. Boston, Mass.: Academic Press, 1995.

PECHEUX, M. (1982) Language, Semantics and Ideology. London: Macmillan.

ROTH, P.A. (1988) 'Ethnography without Tears', Current Anthropology, vol. 30, no. 5, pp. 555 - 69.

SANGREN, P.S. (1988) 'Rhetoric and the Authority of Ethnography', Current Anthropology, vol. 29, no. 3, pp. 405 - 24.

SEAMAN, G. and WILLIAMS, H. (1992) 'Hypermedia in Ethnography', in P.I. Crawford and D. Turton (editors) Film as Ethnography. Manchester: University of Manchester press in association with the Granada Centre for Visual Anthropology.

TAYLOR, L. (editor) (1994) Visualising Theory. London: Routledge.

TITON, Jeff Todd (1995) 'Text', Journal of American Folklore, vol. 108, pp. 432 - 448.

Van MAANEN, J. (1995) 'An End to Innocence: The Ethnography of Ethnography' in J. Van Maanen (editor) Representation in Ethnography. London: Sage.

WEAVER, A. and ATKINSON, P. (1994) Microcomputing and Qualitative Data Analysis. Aldershot: Avebury.

WEAVER, A. and ATKINSON, A. (1995) 'From Coding to Hypertext: Strategies for Microcomputing and Qualitative Data Analysis' in R.C. Burgess (editor) Studies in Qualitative Methodology. London: Jai Press.

YANKELOVICH, Nicole, HAAN, B., MEYROWITZ, N., and DRUCKER, S. (1988) 'Intermedia: The Concept and the Construction of a Seamless Information Environment', IEEE Computer 21: 81 - 96.

Copyright Sociological Research Online, 1998