Copyright Sociological Research Online, 2001

 

John Roberts (2001) 'Dialogue, Positionality and the Legal Framing of Ethnographic Research'
Sociological Research Online, vol. 5, no. 4, <http://www.socresonline.org.uk/5/4/roberts.html>

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

Received: 4/7/2000      Accepted: 19/1/2001      Published: 28/2/2001

Abstract

The question of positioning between the research Self and the research Other is a much discussed issue within qualitative research, especially within ethnographic approaches. Yet what is distinctive about many of these accounts is that they begin their respective analyses from a concrete level. In other words, many who champion placing their Self in the research process do so by focusing upon face-to-face encounters between their Self and the Other. This often entails a rejection of a positivist and objectivist informed accounts of social research in favour of a more humanist approach. This latter standpoint, humanism, is certainly interested in themes such as bias, power, regulation and domination constructed during the research process. But the structured, layered and ideological nature of the research context itself, namely its non-humanist properties, is often neglected in humanist explanations in favour of the more concrete interpretive moment. What this amounts to is a lack of sensitivity towards the positioning of Self and Other by the unobservable and ideological structures of a specific research context. As a result, discussion about the necessity of dialogue between all participants involved is one-sided. This closes down considerably the impact we, as researchers, have on a research context. In addition such a standpoint closes down the positioning effects of a research context upon our own research Self. By drawing upon the work of radical ethnographers and the discourse theorist Mikhail Bakhtin, this paper seeks to outline some of the problems that arise in humanist ethnographic accounts as regards positionality and dialogue. This opens the way for some observations about how ethnography might take into account 'non-humanist' structures such as the state, law and governance in capitalist societies in respect to the issues of positionality and dialogue. I flesh out these theoretical observations through a brief discussion of my own ethnographic experience researching Speakers' Corner in London.

Keywords:
Bakhtin; Dialogue; Ethnography; Ideology; Positionality; Social Form; Speakers' Corner

Introduction

1.1
The question of positioning within a research context is a much discussed issue. Within qualitative research the discussion often revolves around the problem of whether the researcher operates as either an 'insider' or 'outsider' within a delimited research context. A frequent assumption made about this dualism is that the most advantageous position to adopt is an 'insider' position. It is often said that to gain empathy and understanding (verstehen) of a research context it is best to immerse one's Self within the research context in question. In practice this implies acquiring knowledge of the processes, histories, events, language and biographies of the research context. Only when such a position is adopted can meaningful dialogue between researcher and respondents follow (cf. Winch 1963).

1.2
It is also true to say that many of these 'humanist' accounts underline the fact that the categories 'insider' and 'outsider' are not fixed and static but are mobile, unstable and subject to change during the research process. As Herod (1999) notes:

(T)he researcher's identity is not necessarily fixed in some absolute sense...but it may translocate through categories and identities, such that at some times and places the researcher may emphasize certain positionalities and identities and not others (Herod 1999: 321).

1.3
And of course, lest we forget, as researchers we are frequently positioned by respondents. Just as we might wish to be perceived as an 'insider', for instance, respondents might in fact position us as an 'outsider'.

1.4
In stressing the need for reflexivity about positionality in the research process, however, many humanist accounts tend to throw the causal baby out with the objective bath water. By focusing predominately upon the concrete interpretive moment, humanist ethnographies often fail to fully acknowledge the more abstract social characteristics of a research context. This is not to deny that humanist ethnographies cannot address issues of power, bias, regulation and domination in the research process. But the structured, layered and ideological nature of the research context itself, namely its non-human properties, is frequently neglected in humanist explanations in favour of the more concrete interpretive moment. As a result humanist ethnographies often betray a lack of sensitivity towards the positioning of Self and Other by the often unobservable and ideological structures of a specific research context (Burawoy 1991a). If this is the case then discussion about the necessity of dialogue between all participants involved becomes one-sided because a substantial slice of social reality is bracketed from the research context, thus closing down considerably the impact we, as researchers, can have on the world.

1.5
In this article I want to explore in more detail the ramifications for ethnography when a more structured, non-humanist account is adopted. I will try to show that dialogue and positionality within the research context operates at many different social levels. Specifically, I want to argue that where a research context is situated within specific relations of exploitation and power, then the research context in question is best view as a refracted moment, or 'social form', of the totality of those social relations of exploitation and power. By making this claim I am not suggesting for one moment that research contexts merely reflect in a rather mechanistic way particular social relations. What I wish to argue instead is that each research context reflects a specific set of social relations in its own unique way. Hence the reason why each social form refracts the commonalities of those wider social relations. But once this much is taken on board, how we view dialogue and positionality within the research process must change accordingly. For it will be a central part of my argument to also insist that where a research context is situated within a wider set of dominant power relations, then it is crucial to explore how a research context is subject to particular modes of governance and regulation. These modes are important positioning devices insofar that they position subjects within the ideological limits of specific power relations. This being the case, we can also say that particular modes of governance and regulation substantially influence the dialogic relationship between Self and Other. By taking this theoretical detour, therefore, it will be possible to explore the positioning strategies of the Other upon our Self through deeply embedded ideological dialogue. Thus the paper proceeds as follows.

1.6
In the next section I briefly discuss how a common, humanist strand within ethnography views both dialogue and positionality. I suggest that for all of their stress upon the need to be reflexive and open about the influence of Self upon the Other and vice versa, these accounts never the less espouse a dualism which unintentionally separates the Self from the Other. Following this I suggest that (i) a more structured, layered and ideological account of dialogue is required which (ii) embeds the Self integrally within the activity and social context of the Other. I draw upon a non-humanist (materialist) reading of Mikhail Bakhtin to develop this explanation. Having defended the idea that dialogue operates at many different levels of abstraction within the research process, I then focus upon one such level in more detail within a particular set of social relations, namely capitalist social relations. Here I argue that law acts the main ideological form for the positioning and regulation of individuals within capitalist relations of exploitation and power. It follows that the legal framing of a research context can be seen as being the most ideologically robust positioning mechanism of both Self and Other within capitalism. I embed these somewhat abstract theoretical remarks within a brief discussion of my own ethnographic experience of researching the legal framing of free speech at Speakers' Corner in Hyde Park, London. I then make some concluding observations.

Being Reflexive about Dialogue and Positionality

2.1
For any ethnography it is obviously important to try to gain an understanding of the concrete, lived experience of subjects within a social context (Hammersley and Atkinson 1983). Underpinning this assumption is the idea that such an understanding rests upon the added requirement to have actually have 'been there' and observed, perhaps participated in, the social activity which constitutes the context in question. Once returned from the field you must think about your research material by 'being here', back in your own lived context. "Being an ethnographer is to be in two places at the same time" (Pearson 1993: ix). Part of the reason for 'being there' in the first place is, in many respects, part of the process of "trying to think oneself into the situations of the people one is interested in" (Armstrong 1993: 5). An ethnographer must therefore learn to take seriously the 'natural settings' through which social behaviour is daily acted out. Related to this moment is the belief that the ethnographer must seek to explore the 'symbolic world' of respondents.

By 'symbolic world' we refer to the meanings people apply to their own experiences, meanings developed through patterns of behaviour which are some way distinctive by comparison to the outside world (Fielding 1993: 157).

2.2
This particular ethnography rests upon the assertion that although there are 'facts' out there, these are only true within the social context of their discovery and only within their textual representation (Van Maanen 1995: 22-23).

2.3
Another integral moment of the ethnographic process is reflexivity. According to Steiner (1991), reflexivity in research practice denotes the turning back of one's experience upon oneself. Caught up in this act is the recognition of the socially constructed nature of both research Self and research experience.

We are talking about a circular process, in which reflexivity is the guiding relationship allowing for the circularity. This looping back may...unfold as a spiralling, if we allow for multiple perspectives, and acknowledge that 'the same self' may be different as a result of its own self- pointing (Steiner 1991: 21).

2.4
Reflexivity refers to the ability of the researcher to tell a story about their Self so that they become aware of their own research activities. Escaping from the subject-object divide implies recognising the negotiated nature of the researcher, research context and the researched. Only then will meaningful dialogue between all three develop.

2.5
Often the term used to capture the relationship between these symbolic practices and the reproduction of a social context is that of 'the field'. This rather lush and tranquil metaphor for what is frequently a chaotic, disturbing and thoroughly frustrating research experience is described by Coffey (1999) in the following way.

(The field is) a heterogeneous group of locations and contexts. Everyday life as an arena of social enquiry makes the boundaries of observation and analysis almost limitless. While generalizations about the field are difficult, and often unhelpful, all fieldwork sites will have at least one common factor. The field is a site peopled by social actors and, implicitly, by the social researcher. The primary task of the fieldworker is to analyse and understand a peopled field. This task is achieved through social interaction and shared experiences. It follows, therefore, that fieldwork is dependent upon and guided by the relationships that are built and established over time (Coffey 1999: 39; my italics).

2.6
On such an account, interaction between the research Self and the research Other is accomplished primarily at a 'concrete level'. On the question of positionality, Coffey seems to admit that the research Self can affect both the research context and the research Other. The very act of immersion within the research context therefore implies that the researcher automatically adopts the position of being both an insider and an outsider (Coffey 1999: 22).

2.7
However, Coffey's modest idea that ethnographers should not assume the heroic role as a researcher who embarks on a quest from being a strange outsider to informed insider belies, I believe, a different form of dualism in research practice. As the italicised sentence of the above quote from Coffey makes clear, meaningful contact is said to be constructed through social interaction and shared experiences. But such a position can easily slide into a one- sided ethnographic analysis. It can do so because 'the field' is primarily defined here through social relationships and social interaction between individuals. What this account therefore neglects are those social processes which cannot simply be reduced to the concrete level of social interaction. For example, and it is a point which will be explored in greater depth below, this standpoint begs the important question of how we are to account for the ideological structuring effects of regulation upon a social context through the state, law and governance. Thus, in one important sense, a crucial structuring moment of the research context is denied. Put simply, each social context has a unique ideological identity which cannot be reduced merely to the level of 'the concrete' symbolic activities of the individuals who inhabit the context in question.

2.8
There are many negative results for ethnographic research as a result of this one-sided analysis. Perhaps most ironically this type of ethnography often comes to be filtered primarily through the interpretive framework of the ethnographer. I say that this is somewhat ironic because most ethnographers pride themselves on exactly not imposing their own bias upon the research context. The work of Clifford Geertz is instructive in this respect.

2.9
Geertz (1993a) famously insists that ethnography should strive for 'thick description'. What Geertz means here is that ethnography should seek to discover 'a stratified hierarchy of meaningful structures' by which certain social acts are important to the people carrying out those acts (Geertz, 1993a: 5). This is to isolate the 'flow of social discourse' (Geertz, 1993a: 20) between participants within a social context so that the meaning of what is said is rescued. To achieve this task the researcher must pay close attention to the micro processes of social interaction. Once thick description is accomplished the anthropologist can move 'from local truths to general visions' (Geertz, 1993a: 21) in order to gain a sense of the wider social picture of which the initial moment in which the anthropologist was interested. But such a move must always generalise within thick description and so must always work with the meaning collected.

2.10
From the comments sketched out above, however, Geertz's account must be found wanting in at least one important respect. For it would seem to be the case that thick description is premised upon separating the practical moment of data collection from the theoretical, interpretive moment which is conducted away from the research context. Accordingly Geertz acts as Other to the encounter he witnesses. As Dwyer (1982) suggests:

Geertz, then, restricts the active role of the anthropologist to the moment of writing, denying an active role to the anthropologist in the direct encounter with the Other. In Geertz's approach, the anthropologist has again become a passive observor or recorder, and the interdependence of Self and Other is supplanted by the interdependence of the anthropologist and the text he or she constructs, a text in which the Other's constructions are treated in isolation and as having been expressed spontaneously (Dwyer 1982: 263; see also Lemaire 1991; Scholte 1986).

2.11
Geertz's rightly acclaimed study of Balinese cockfighting provides a good example of Dwyer's observation.

2.12
According to Geertz (1993b), the practice of cockfighting amongst Balinese men is not simply about gambling. Indeed, betting on a cockfight is not judged primarily on whether one will increase their monetary input. Rather, betting is concerned first and foremost, "about fights against such-and-such a cock of so-and-so which cock which your cock demolished, not on how much you won..." (Geertz 1993b: 440). Cockfighting is therefore a symbolic set of practices made up of a whole host of emotions and sentiments. This leads Geertz to assert that cockfighting is really about public reflection by Balinese men on their cultural ethos and private sensibility. A cockfight acts as collective text for Balinese men to read and learn about the themes of "animal savagery, male narcissism, opponent gambling, status rivalry, mass excitement, blood sacrifice" (Geertz 1993b: 449). Read as a collective text, therefore, cockfighting is transformed into an external manifestation of the private emotions of Balinese men.

2.13
What is noticeable about the 'thick description' of Balinese cockfighting is how little Geertz himself figures in his own narrative. We never learn, for example, how the very presence of Geertz structured the particular encounter he was witness to. Indeed, Geertz's use of the 'text' metaphor places him in the role of a passive and neutral observer of Balinese society. His job is merely to note the symbolic and textual representations of Balinese society and not to explore the politics of power associated with the production of the social context he observes. As Scholte (1986) remarks, Geertz is more interested in the question, 'what is considered meaningful?' at the expense of the more probing questions: 'How is meaning produced? How is meaning maintained, controlled and distributed?' (Scholte 1986: 10; see also Burawoy 1991b). Insofar that Geertz brackets out these latter questions in his methodology so does he bracket out his Self in his research practice. Inasmuch as this view prevails, then both the historical and social situatedness of researcher and researched are denied, along with the historical impact of the social sciences and humanities upon research contexts. We are thus left with an ahistorical notion of Self who gazes objectively into the cultural landscape of the Other.

2.14
One of the main points to therefore emerge from the discussion so far is that ethnographic research should take seriously the use of historical and theoretical methods. In this respect Burawoy's methodological advice about the application of what he terms as the 'extended case method' for ethnography is useful.

The extended case method looks for specific macro determination in the micro world...It seeks generalization through reconstructing existing generalizations, that is, the reconstruction of existing theory (Burawoy 1991b: 279; see also the discussion of Evans-Pritchard by Mattick, Jr. 1986).

2.15
The use of the term 'extended case method' is intended to show how each 'micro' social context is determined by a constellation of 'macro' institutions located in time and space and at different levels of abstraction. These institutions shape domination and resistance within a specific context and imbue that context with a certain ideological, or 'social', form. To abstract away from this social form is therefore to abstract away from a historically and ideologically shaped context which, in turn, is a moment in a wider set of historical and ideological social relations. Before I say more about the concept 'social form' it is worth briefly pausing to consider the term 'abstraction' in a little more depth.

2.16
The method of abstraction is informed by a rich and deep ontology that seeks to discover the structured, layered and necessary properties and relationships of a social context. We can say that a world exists both independently of our knowledge whilst recognising also that access to that world is achieved through knowledge. Properties, relationships, causal powers, or whatever, may never be observed directly, though we can gain adequate knowledge of them through our (theoretical) practice (Bhaskar 1978; see also Benton 1981; Collier 1994; Sayer 1992). An interesting illustration of these theoretical points is provided by Porter (1993) in his ethnographic study of how racism affects occupational relationships between doctors and nurses in an intensive care unit of a large metropolitan hospital in the north of Ireland. According to Porter, racism has 'ontological depth' to the extent that it predates and positions individuals in various and enduring ways. At the same time, the structure of racism operates in 'open systems' where it will interact with many other structures. What Porter found in his study was that racist attitudes were clearly expressed by nurses towards doctors from ethnic minorities. However, this was often expressed through the veneer of the structure of professionalism. This latter structure is characterised in the main by universalistic-achievement values such that individuals are seen to have gained their occupational status through their own ability. Attached to this 'pattern variable' is the idea that the judgment of an individual should be made on the basis of scientific rationality. This is bound up with the belief that individuals should also be judged on their merit and skills as a particular professional rather than on factors contingent to the occupation at hand. Through observation and interviews, Porter insists that 'naked racism' within this particular hospital context was deemed 'irrational' because the structure of professionalism was also present. Thus racist attitudes were seen as justified if the doctor to whom the racist remark was directed was deemed to be 'professionally incompetent'. Porter argues that both structures were often evident in the conversations he witnessed between nurses.

2.17
Porter's analysis demonstrates well the importance of abstracting social structures for ethnographic analysis. But the method of abstraction is also important because it transforms our understanding of what it means to be reflexive within a research setting. As Cain (1990) suggests, a reflexive ontology attuned to the causal efficacy of social structures implies that we have to take very seriously the use of social theory as a qualitative methodological research tool.

In presenting our field work we have to be personally reflexive, that is, present a description of our changing relationship with the researched population...But in order to decide which site to speak from or to understand a pre-given site we have to engage in theoretical reflexivity, based on the new kind of theories we are constructing. Theoretical reflexivity is a key concept and a key practice in the identification of standpoints...Theoretical reflexivity means thinking about oneself in terms of a theory and understanding theoretically the site one finds oneself in (Cain 1990: 132-133; see also Cain 1986).

2.18
Burawoy's use of the extended case method implicitly adds to Cain's suggestion the idea that personal and theoretical reflexivity must also be accompanied by a type of historical reflexivity. This is so in the simple sense that each research context is structured over time by a set of specific (macro) historical social relations such as capitalist social relations (see also Roberts 1999a; b; 2001 for further comments on the notions of abstraction, historical reflexivity and social form). In order to come to grips with these wider set of historical social relations the most appropriate theoretical move would seem to be one that first established the determining ideological moment of those historical set of social relations. Advice from Karl Marx can be of some assistance here. He encourages us in Volume 3 of Capital to seek out how,

the specific economic form, in which unpaid surplus-labour is pumped out of direct producers, determines the relationship of rulers to ruled, as it grows out of production itself and, in turn, reacts upon it as a determining element. Upon this, however, is founded the entire formation of the economic community which grows out of production relations themselves, thereby simultaneously its specific political form (Marx 1966: 791).

2.19
What we might say, then, is that each research context is a distinct moment in this contradictory relationship between rulers and ruled. That is to say, each research context assumes a specific ideological and contradictory form in the reproduction of those social relations at different levels of abstraction. This standpoint also has the advantage of asking how meaning is constituted, maintained, distributed and controlled. We simultaneously concretise the production of our own research techniques by saying something about the historical and political context through which we, as ethnographers, gain knowledge (Scholte 1986: 10-11). In respect to the example of racism discussed above we would therefore try to isolate not only the structure of racism but also the capitalist social form of racism (see Virdee 2000).

2.20
In the next two sections I suggest that it is imperative for ethnographers to take seriously the complexly and historically determined nature of the research context. In particular I argue that ethnographers must re-focus their ideas around the terms 'dialogue' and 'positionality'. By drawing upon the work of Mikhail Bakhtin and recent developments in critical theories of law and governance, I try to map out some of the characteristics of a more 'non-humanist ethnography' that opens itself up to structures of ideology, conflict and struggle existing at a number of levels of abstraction. I begin this discussion by once again noting some of the problems produced by humanist accounts when addressing these issues.

The Need for Dialogue

3.1
Qualitative researchers frequently stress the need for good rapport and dialogue to be encouraged during the research process. Only then will we, as researchers, really begin to understand how and why we embody a multitude of identities within the research context through positioning strategies by respondents (Jorgensen 1991; Oakley 1981). This was certainly the conclusion discovered by Back and Solomos (1993) during their research experience of studying the local politics of race in Birmingham. Back and Solomos hoped to gain knowledge about the ideologies and practices of those individuals involved in carrying out policy programmes in Birmingham City Council and to what extent these programmes reproduced or ameliorated racial inequalities. In particular they were concerned to assess the impact of black political participation was having on the political landscape. Whilst they consciously sought to avoid taking sides during the research process, they nevertheless discovered that they were positioned with various identities by respondents. This was demonstrated most dramatically when Back and Solomos found that ballot rigging had occurred during the voting procedure for the next prospective parliamentary Labour Party candidate in a Birmingham constituency. As a result of ballot rigging, a white candidate had been elected as the prospective candidate instead of two prominent Asian councillors who were also standing. Back and Solomos wrote a paper about the events leading up to the selection of the candidate. However, the outgoing Labour Party MP for the constituency accused Back and Solomos of bias against the chosen white candidate. The MP also indicated that he believed that their research was invalid because of this perceived bias. Thus Back and Solomos found themselves being positioned as politically biased and incompetent researchers. They concluded from this incident:

The point that follows from this is the simple fact that as researchers we give accounts of our work that are dependent upon the context in which our speaking positions are being evaluated. We are not arguing here that anything goes. We are simply saying that as researchers involved in political struggle we may find it necessary to take on strategic academic identities (Back and Solomos 1993: 194).

3.2
Back and Solomos's account makes some very perceptive observations about power relations and their relationship to positionality within a research context. However, it is also true to say that their account can be situated within a humanist perspective. As a result it displays several weaknesses . To begin with, Back and Solomos believe that it is important to first adopt a position of neutrality within a research context in order to grant due weight to different dialogic encounters with different respondents. Accordingly they suggest that positioning occurs only during and after the research event. However, from the discussion in previous sections this would seem to be a somewhat spurious epistemology to adopt. This is because Back and Solomos explore positioning as a concrete phenomenon. Subsequently they miss not only the positioning they share with the people of Birmingham through their subjection to the same ideological social relations, they also miss the positioning practices operating at a high level of abstraction within any social context as regards forms of regulation and governance. In one important respect, therefore, Back and Solomos are subject to an evaluative positioning as soon as they enter their research context. But if this is the case then we need an explanation of dialogue which can help to explain this more non-humanist notion of positioning. For clearly, if the process of positioning is enacted at a high level of abstraction then we must have an account of dialogue that can take note of the deeply embedded ideological nature of dialogical interaction.

3.3
I believe that the work of Mikhail Bakhtin can begin to present us with such an account of dialogue. This is because dialogue for Bakhtin represents communication between individuals within a socially, culturally and historically embedded context. This is so, according to Bakhtin, for a number of reasons. First, Bakhtin (1987) relates an individual to her historical context. Whatever a speaker speaks about has already been articulated and subject to much discussion in different spatial and temporal localities. An utterance is already pregnant with many viewpoints before it comes to be used in practical activity. "The speaker is not the biblical Adam, dealing only with virgin and still unnamed objects, giving them names for the first time" (Bakhtin 1987: 93).

3.4
Second, Bakhtin stresses that speech is achieved through interaction. Dialogue can only emerge as a response to another speaker. Thus an utterance not only responds to another speaker but also anticipates a counter-response from the speaker or from another speaker or from a whole host of speakers.

From the very beginning, the speaker expects a response from them, an active responsive understanding. The entire utterance is constructed, as it were, in anticipation of encountering this response (Bakhtin 1987: 94).

3.5
A chain of communication is actively constructed which looks to the past and looks to the future.

3.6
Third, dialogue for Bakhtin is relational. Every individual is a social being, each bringing a collection of social baggage into the dialogic encounter. As such utterances, as distinct from mere words and sentences, carry with themselves the quality of addressivity. "Both the composition and, particularly, the style of the utterance depend on those to whom the utterance is addressed, how the speaker (or writer) senses and imagines his addressees, and the force of their effect on the utterance" (Bakhtin, 1987: 95). The utterance is concerned not only with an object, but also anticipates another speaker's evaluation of the object.

3.7
Fourth, Bakhtin suggests that dialogue always works through 'typical forms', namely speech genres (Bakhtin 1987). A person's style of utterance is mediated through genres so that pure, individual stylistic endeavours are impossible. Styles often correspond to specific genres and genres often correspond to specific conditions of speech communication. Genres elicit not only styles of utterances, they also enable 'themes' to emerge which in turn give rise to compositional unities: "to particular types of construction of the whole, types of its completion, and types of relations between the speaker and other participants in speech communication..." (Bakhtin, 1987: 64). Thus genres structure, and are structured by, specific ideological social forms of life (Bakhtin and Medvedev: 1928/1978).

3.8
For Bakhtin, therefore, words are self-contained ideological units which have a number of often contradictory 'accents' associated with them based upon their cultural, historical and social embeddedness. These words gain a new level of meaning when situated within a particular social form and thus can go on to structure generic types of activity within the social form in question. This is why words are an arena for social and class struggle. Often, a dominant ideology will attempt to impose a universal, monoglossic form of dialogue over the heteroglossic, subversive utterances of subordinate groups (Voloshinov 1930/1973).

3.9
Bakhtin's insights are thus important for ethnography for at least four reasons. First, they alert us to the idea that each utterance not only expresses the intention and position of the speaker, but also reflects the individual to which the utterance is addressed. In this sense each utterance seeks to influence the dialogic position occupied by another subject. For example, an utterance by a respondent addressed to a researcher might endeavour to position the researcher in a certain manner (cf. Oinas 1999). Second, dialogic strategies operate through specific genres, and such genres are ideologically embedded within distinct ideological forms of existence. Genres, therefore, are complexly structured and layered and can be examined beyond an individual responding to a question. Positioning by a respondent of a researcher carries with it deeply embedded ideological connotations. Third, and following on from the previous point, positioning through dialogue is also structured by social conflict, contradiction and struggle. To stabilise a dialogic social context and to avert a crisis of legitimacy, therefore, modes of 'dialogic governance and regulation' must be introduced. Finally, these forms of governance and regulation are positioning devices, but they are positioning devices that can be methodologically isolated irrespective of the actions of the respondent in any concrete situation. For example, it is possible to isolate the regulatory devices of particular legal signs that act as ideological positioning devices (see the next section). Albeit, these are all positioning devices which impact upon the research process at a concrete level whether or not we are conscious of their effects. This further implies that the 'non-humanist' ideology articulated by all of these processes also impacts upon the respondent's actions in positioning the researcher, again whether or not respondents or researchers are actually conscious of this process.

3.10
In the next section I say a little more about the structuring of dialogue and positionality within specific social forms. I suggest that a crucial mechanism for the non-humanist positioning of dialogue within capitalist social relations is that accomplished through the legal framing of a research context. Legal framing, encompassing as it does the three regulation mechanisms of the state, law and governance, ideologically positions both researcher and respondents. At the same time, therefore, legal framing acts as a device which regulates utterances, dialogue, and genres between researcher and respondent. This means that legal framing is integral to the process of ethnographic research within capitalist societies and thus must be accounted for. In other words, once we bring into the research process the idea of positioning through legal framing then the need for the notion of dialogue defended in this article becomes ever more urgent. I make these observations by drawing upon my ethnographic experiences of researching the legal framing of free speech at Speakers' Corner.

Legal Framing, Positionality and Speakers' Corner

4.1
If a research context, or social form, does refract the contradictions, etc., of wider social relations, each research context not only assumes a unique ideological form which presupposes those wider social relations, it also exhibits a sense of a instability. This is so for at least two reasons. First, as a distinct ideological entity, each social form can prove problematic for the 'functioning whole' of wider social relations (cf. Jessop 1990). For example, even though the capitalist state emerged internally from capitalist socio-economic relations, the capitalist state is at the same time a distinct ideological entity which must answer to a whole range of social forces beyond the remit of socio-economic relations. Thus the state can, and often does, prove problematic for socio-economic relations. Second, a research context internalises contradictions, etc., in its own unique way. As such it must contain those unique contradictions through both coercion and consent (cf. Gramsci 1986). Yet the containment of contradictions also open up gaps in the governance of a social form which allow meanings of opposition to break forth (Jessop 2000).

4.2
Subsequently, if a social form is unstable and requires regulating then we must, where possible, explore the historical nature of such regulation. But what forms do these modes of regulation assume under capitalist social relations? According to Woodiwiss (1985; 1990a; 1990b), law appears under capitalism as a set of state enunciated discourses that seek to interpellate individuals as 'law abiding'. In order to achieve this mammoth task in the face of resistance, Woodiwiss further argues, law must maintain a 'disciplinary network' by producing a 'background ideology-effect' and act as a 'second-order' discourse for a diverse range of discourses that can be invoked when these primary discourses are challenged. Law is thus a reactive mechanism to discursive word-signs, utterances, dialogue and genres that have already been articulated within a social context. But so that these rules are not seen to be arbitrarily composed, law depends upon a system of norms which are not seen to contradict one another. This implies that legal discourse must be consistent, coherent and enable a type of closure over its pronouncements. Consistency confers a level of autonomy upon capitalist law which is absent in other discourses. But consistency can never appear as itself but must always be articulated with a substantive sign such as 'liberty', 'equality', 'citizenship' or 'free speech' (Woodiwiss 1985, 1990a, 1990b; cf. Norrie 1993). In effect, legal signs maintain the mobility of the regulatory functions of the 'fixed' state-form. For these reasons law remains one of the most powerful mechanisms for the articulation of ideology within capitalist societies.

4.3
Legal discourse also provides a unifying force for governance mechanisms in the face of resistance. Developed primarily as means to analyse the complex, subtle and dispersed forms of power in modern societies, governance for Foucault (1975/1991) refers to a shift in modern societies from a form of discipline centred around sovereign law and based upon explicit and elaborate spectacles of punishment to a more localised form of discipline based upon a plural and local form of governmentality which propagates an ethical outlook. Modern types of power rest primarily upon a political 'rationality' and 'technology' of power - the moral and ethical construction, maintenance and regulation of objects to govern (Foucault 1975/1991).

4.4
If successfully constructed within a particular place, and if forms of political technology and rationality have been rendered unambiguous by those subject to them, then governance enables law to rule at a distance (cf. Rose 1999). More importantly, the substantive signs articulated by legal discourse provide a constituting link between governance and the state. If the constituting link between all three regulating entities is successful, we can say that the legal framing of a context, or social form, has been established. Principally, legal framing produces a relatively stable configuration of genres within a particular social context by ideologically investing utterances and dialogue with themes congruent to a specific state project. That is to say, legal framing involves 'subsidising' dialogue by putting in place the twin processes of government-as-regulator and government-as- speaker within a social context (Post 1996; Tushnet 1999) thus imposing monoglossia upon a place. As researchers, therefore, we are immediately positioned ourselves by the framing mechanisms of the state, law and governance as soon as we enter a research context.

4.5
In order to illustrate how the research Self can be said to be legally framed within a research context I will now draw upon my own ethnographic research experience at Speakers' Corner. However, the following observations about my research experience should in no way be taken as a substantive account. Their purpose is merely to flag up some illustrative material with which to demonstrate the main points of my argument.

4.6
I spent nine months researching free speech at Speakers' Corner in Hyde Park, London back in 1996. As you may or may not know, Speakers' Corner is a legally designated site in Hyde Park, London, where every Sunday about 15 to 20 speakers get up on their makeshift platforms and talk about different subjects such as current affairs to an audience that reaches into its hundreds. (Obviously speakers and other participants are subject to rules and regulations about public speaking in Hyde Park.) There are a vast array of different speakers and they range from Muslims, green speakers, gay and lesbian speakers, speakers talking about Black politics, Marxist speakers, and speakers discussing meditation and philosophy. There are even some speakers who only speak if they are spoken to. As well as speaking from a platform, many regulars attend in order to ask questions from 'the floor' and/or to heckle a particular speaker. Many who attend are non-regulars and these individuals often engage in debate and discussion or just listen to an ongoing discussion.

4.7
Understandably, therefore, it is possible to say that an important association exists between the place name, Speakers' Corner, and the legal sign, 'free speech'. This association is certainly made by many of the regulars at Speakers' Corner. On one occasion I was chatting with a Marxist speaker about public protest in the former Eastern European communist countries. He was telling me about the relationship between public speaking by dissent groups in these former communist states and their relationship to the eventual collapse of those states. His point was that the right of free speech in particular locations was an integral moment in radical public protest. On his estimation, "Speakers' Corner is just a geographical location, if you like, to epitomise that concept of free speech and democracy as free speech" (Interview 21st May 1996). Moreover, he said, Speakers' Corner represented a crucial 'intermediate' place for free speech during moments of relative decline in radical political movements because it was a permanent place for the articulation of inclusive citizenship. For this particular speaker, therefore, Speakers' Corner personified a special locality for radical utterances at a time when radical movements were in decline.

4.8
But many regulars at Speakers' Corner would also often emphasise the association between Speakers' Corner and free speech if they felt that a particular discussion was getting too heated. 'Take it easy. This is a place for free speech' was frequently uttered. The 'calming effect' behind words such as these seemed to be that the one which respected, in some way or another, the dialogic nature discussion. Thus the ideological word-sign, 'free speech', opened up a type of dialogue between participants. That is to say, 'free speech' did not, it seemed to me, close down dialogue between participants but actually ensured that it continued. As the Marxist speaker told me:

Well that's the beauty of it. It's a geographical place where there is this culture of speaking. And because there's a culture of speaking there's a culture of tolerance. Even though it may be expressed in intolerance, sometimes, you know, by homophobic ideas, by racist ideas, or sexist ideas, or whatever, there is a tolerance there. And that tolerance is a great thing in my opinion because it enables you to push the boundaries about things and to raise issues about things so that even if somebody comes out with a homophobic remark you can remark on it. If somebody comes out with a homophobic remark in a pub and you say you're gay, you're gonna get your face punched in. Whereas at Speakers' Corner I've never been attacked...which I consider either luck or it's part of this cultural acceptance (Interview 21st May 1996).

4.9
This 'cultural acceptance' of a generic form of speaking that came to be encapsulated within the word-sign 'free speech' ensured that Speakers' Corner as a place carried with it an ideology of 'inclusive visibility'. This was so to the extent that those utterances frequently excluded from 'conventional' political mechanisms could be publicly expressed in this space at Hyde Park. Thus 'free speech' was a dialogic word-sign in the Bakhtinian sense of that term because it maintained a certain heteroglossic inversion. More to the point, 'free speech' produced particular generic forms of speaking. As another speaker told me, Speakers' Corner had a 'festive atmosphere' which allowed debate and discussion to proceed in an enjoyable and interactive manner. I was told that such an atmosphere was based, in part, upon the ability of speakers to 'invert' reality, to highlight its more absurd facets. Yet this inversion was deeply serious for most regulars because it enabled them to make visible 'hidden' power relations and thus provided an opportunity to articulate their own identity and their set of right claims. One notable example was Abdul, a speaker who talked about black politics. In the words of a white speaker:

Abdul is a very good speaker, no doubt about it...You look at his rapport with the audience, a mostly black audience, but his rapport with the audience is fucking brilliant. His jokes are inversions of white racism. And I don't find them hostile because I know what he's going on about. He's doing the opposite (of white racism) and it's brilliant the way he does it (Interview 9th June 1996).

4.10
But the association between Speakers' Corner and free speech is not merely confined to the geographical borders of Speakers' Corner itself. Recently, for example, the Home Secretary, Jack Straw, announced plans for the establishment of speakers' corners across 300 cities in the UK. The rationale behind the proposal was summed up by Straw in the following manner.

This government is committed to the rejuvenation of towns and cities. This is a small but constructive step in the best British tradition of free speech and public debate (The Independent, 16 November 1999).

4.11
Thus the more I reflected upon the link between Speakers' Corner and free speech the more I realised that it was important to ascertain the ideological nature of free speech in order to see what type of dialogue was being produced.

4.12
The obvious place to begin this critical enquiry was to go back in history in order to explore the moment when the state legally framed Speakers' Corner. This occurred in 1872 when Gladstone's Liberal government passed the 1872 Royal Parks and Gardens Regulation Act. The 1872 Act, itself a reaction both to a number of radical campaigns in London of the time campaigning for the right of free speech in royal parks and to a specific campaign for free speech at Hyde Park, does not mention 'free speech'. The legal sign most frequently cited in the 1872 Act is 'public address'. This is an interesting legal sign to the extent that it sanctioned dialogue of a particular nature. To begin with, the First Schedule of the 1872 Act contained no less than twenty specific regulations. Each regulation constructed a moral profile of what the state considered to be an ideal-typical deviant using the royal parks. Regulation 13, for example, stated:

No person shall commit any act in violation of public decency, or use profane, indecent, or obscene language to the annoyance of other people using the parks

4.13
The most important Regulation, Regulation 8, stipulated:

No person shall deliver, or invite any person to deliver, any public address in a park, except in accordance with the rules of the park

4.14
This Regulation is a crucial legal framing mechanism for at least two reasons. First, Regulation 8 associated 'public address' with the other regulations in the 1872 Act. Thus Regulation 8 effectively positioned those exercising 'free speech' within the moral boundaries of the ideal-typical deviant user of the royal parks. As such the state had a wide remit through which to intervene and limit public dialogue within a royal park if it considered that particular utterances were in some way or another considered deviant. At the same time state intervention was premised upon the distinct ideological meaning of the discursive terms used. Regulation 13, for example, includes the sub- sign 'public decency'. In Victorian Britain the term 'public decency' was often associated with 'fecklessness' and 'immorality'. Those possessing these latter characteristics were those unfortunate enough to find themselves in grinding poverty (Johnson 1993). Yet many radicals campaigning for 'free speech' during this period sought to represent the interests of these individuals in their wider political battle for an extension of citizenship rights. Counterposed by the state to these negative characteristics was the sign 'responsibility'. To be 'responsible' in Victorian Britain was also to be 'moral' decent' and 'educated'. Regulation 13 therefore articulated this particular Victorian ideology within the confines of the royal parks. By associating 'public address' with these sub-signs the 1872 Act limited the form of dialogue at royal parks.

4.15
Second, the 1872 Act was an important positioning device because it defined 'public address' 'in accordance with the rules of the park' in question. This meant that each royal park could enact a specific set of governance mechanisms that responded to its unique spatial boundaries. At the same time these governance mechanisms were legally enforced by an Act of parliament. In respect to Hyde Park a legally designated area for 'public address' was mapped out that stretched from Marble Arch to Hyde Park Corner. With this legally designated area a set of rules could be developed that responded to the unique place of Hyde Park. However, these rules were drawn up by Parliament rather than being left to the discretion of park rangers, thus ensuring the overriding power of the state. Overall, therefore, the 1872 Act represents a contradiction. At the moment that the state recognises 'public address' at Hyde Park so does it forget 'free speech' at Hyde Park (I discuss the 1872 Act and the events leading up to its final ratification in much greater empirical depth in Roberts 2000a, b). In this respect I would want to argue that the legal sign, 'public address', was non- dialogic in a Bakhtinian sense because it encapsulated a certain monoglossic tendency.

4.16
The basic ideological form of the 1872 Act was still in place in 1996 and in fact exists up until today.[1] As a result of this historical investigation I came to realise that the word-sign, 'free speech', had a specific contradictory and ideological identity within the confines of Speakers' Corner itself. Yet this ideological identity was complexly structured through a historical narrative unique to Hyde Park itself and through wider social relations such as the state, law and governance. As a result 'free speech' at Speakers' Corner was an ideological word-sign that existed beyond the concrete interaction of those participants I observed at any moment in time during 1996. Moreover, it was an ideological word-sign that carried deep ideological significance for those involved in that concrete interaction whether they were conscious of this or not. And part of this ideology was related to the implicit and explicit dialogic confrontation between 'free speech' and 'public address'.

4.17
By exploring the 1872 Act, therefore, I was able to begin to reflect upon how the framing mechanisms of the state, law and governance positioned my Self at Speakers' Corner. Two general research questions seemed especially pertinent in this respect. First, to what extent does legal framing imply that we, as researchers, also share a basis of commonality with those whom we research simply because we are positioned by the same ideological mechanisms of legal framing? Second, to what extent does legal framing and legal positioning imply that respondents, through their own ideological utterances, position the researcher in order to circumvent, in whatever measure, modes of governance and regulation through legal framing. From my own experience it was certainly the case that respondents often used my research Self to resist the legal framing of Speakers' Corner. For example, I remember standing listening to a Muslim speaker talking about the representation of Islam in western societies. I had interviewed the speaker a few weeks previously. He saw me in the crowd and before I knew what was happening he engaged me in dialogue about Christianity in western countries and its negative effects upon Islamic culture. This rupture in the sequence meant that I was momentarily positioned as a western Christian. I tried to argue back by saying something to the effect that all religions, in my opinion, were a set of ideas under which lay a political agenda. A number of other Muslims intervened in the crowd and also started to debate with me about my views. In one important respect, therefore, it was through this initial positioning of my Self by the speaker, and by his refusal to accept wholly my status as a researcher, that he opened up a space to resist the legal positioning of his utterances through the legal sign 'public address'. He also opened up a gap for other Muslims to do the same. Again, this was an interesting dialogic encounter and it opens up a number of questions. How satisfying does the Islamic speaker view this rupture in the research process? How will we interpret our next encounter with each other? To what extent does the positioning of my Self enable the Islamic speaker and other Muslims to circumvent the legal framing of his Self at Speakers' Corner? Can we detect an implicit notion of free speech at work in the generic style of the Islamic speakers' utterances? Is this alternative 'subaltern' expression of free speech in contradiction with the legal sign 'public address'? Unless the researcher is aware of the legal framing and legal positioning engendered by the research context as this occurs through time then it is likely that they, as researchers, will remain unaware of the ideological ramifications of the positioning strategies of respondents (cf. Dwyer 1982).

Conclusion

5.1
In this paper I have tried to suggest that positionality and dialogue work at many different levels of abstraction within the research process. Often, however, ethnographers particularly and qualitative researchers generally fail to take account of these different levels. Such a neglect tends to reimpose a dualism between Self and Other, thereby closing the gap regarding positions of both difference and commonality between both. At the same time we are in danger of losing sight of ideological mechanisms of regulation that, again, have important ramifications for the relationship between Self and Other.

5.2
By setting out some albeit brief notes for why we should take a non-humanist ethnography seriously, I have tried to show that (abstract) theory construction is crucial for ethnographic research. Indeed, the fact that we all use some kind of theory when addressing empirical issues does suggest that the more we are conscious of this then the more that we might become aware of the ideological pitfalls of using ideas in a non-reflexive manner. In addition, and perhaps more importantly, we can also be attuned to the practical use of research in trying to change the world.

5.3
But I hope that the main points of my argument are not taken to mean that I support a 'death of the research subject' thesis. I emphatically do not. Actually, I believe that my argument reinforces the necessity of taking the subject very seriously indeed. For I have tried to show that qualitative research can begin to enter into a complex and serious level of dialogue with other people only when it takes seriously the complex and contradictory processes of determination and regulation upon the research context. To that extent the argument above can be seen as a contribution, in a small measure, to the establishment of a radical ethnography.

Notes

1Even though I do not have the space here to discuss how the regulation of Speakers' Corner has changed over the years, it is interesting to note that the current Act which regulates Speakers' Corner, the 1997 Royal Parks and Other Spaces Regulations Act, still fails to mention 'free speech'. The general tenor of the 1997 Act is similar to that of the 1872 Act in the sense that 'public address' is defined in relation to the ideal-typical user of the royal parks. It should also be noted that Jack Straw's speakers' corner scheme is to be implemented through a non-governmental organisation. This raises important questions about the democratic mandate of the scheme.

References

ARMSTRONG, G. (1993) Like that Desmond Morris? In Hobbs, D. and May, T. (eds.) Interpreting the Field: Accounts of Ethnography. Oxford: Clarendon Press.

BACK, L. and SOLOMOS, J. (1993) Doing Research, Writing Politics: The Dilemmas of Political Intervention in Research on Racism. Economy and Society, 22(2): 178-199.

BAKHTIN, M. and MEDVEDEV, P. (1928/1978) The Formal Method in Literary Scholarship, trans. by A. Wehrle. Baltimore: John Hopkins University Press.

BAKHTIN, M (1981) The Dialogic Imagination, trans. by C. Emerson and M. Holquist, ed. M. Holquist. Austin, Texas: University of Texas Press. BAKHTIN. M. (1987) Speech Genres and Other Late Essays, trans. V.W. McGree, ed. C. Emerson and M. Holquist. Austin, Texas: University of Texas Press.

BENTON, T. (1981) 'Objective' Interests and the Sociology of Power. Sociology 15 (2): 161-184.

BHASKAR, R. (1978) A Realist Theory of Science 2nd edition. London: Verso.

BURAWOY, M. (1991a) Reconstructuring Social Theories. In Burawoy, M et al. (eds.), Ethnography Unbound: Power and Resistance in the Modern Metropolis. Berkley and Los Angeles, California: University of California Press.

BURAWOY, M. (1991b) The Extended Case Method. In Burawoy, M et al. (eds.), Ethnography Unbound: Power and Resistance in the Modern Metropolis. Berkley and Los Angeles, California: University of California Press.

CAIN, M. (1986) Realism, Feminism, Methodology and Law. International Journal of the Sociology of Law, 14: 255-267.

CAIN, M. (1990) Realist Philosophy and Standpoint Epistemologies or Feminist Criminology as a Successor Science. In Gelsthorpe, L. and Morris, A. (eds.) Feminist Perspectives in Criminology. Milton Keynes: Open University Press.

COLLIER, A. (1994) Critical Realism. Verso, London.

COFFEY, A. (1999) The Ethnographic Self: Fieldwork and the Representation of Identity. London: Sage.

DWYER, K. (1982) Moroccan Dialogues: Anthropology in Question. London: J. Hopkins Press Ltd.

FIELDING, N. (1993) Ethnography. In Gilbert, N. (ed.), Researching Social Life. London: Sage.

FOUCAULT, M (1975/1991) Discipline and Punish: The Birth of the Prison, trans. by Alan Sheridan. London: Penguin.

GEERTZ, C. (1993a) Thick Description: Toward an Interpretive Theory of Culture. In Geertz, C. The Interpretation of Cultures. London: Fontana Press.

GEERTZ, C. (1993b) Deep Play: Notes on the Balinese Cockfight. In Geertz, C. The Interpretation of Cultures. London: Fontana Press.

GRAMSCI, A. (1986) Selections from Prison Notebooks. London: Lawrence and Wishart.

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

HEROD, A. (1999) Reflections on Interviewing Foreign Elites: Praxis, Positionality, Validity, and the Cult of the Insider. Geoforum 30: 313-327.

JESSOP, B. (1990) State Theory: Putting Capitalist States in their Place. Cambridge: Polity Press.

JESSOP, B. (2000) Governance Failure. In G. Stoker, G. (ed.), The New Politics of British Local Governance. London: Macmillan.

JOHNSON, P (1993) Class Law in Victorian England. Past and Present 141: 147-69.

JORGENSEN, J. (1991) Co-Constructing the Interviewer/Co-Constructing 'Family'. In Steiner, F. (ed.), Research and Reflexivity. London: Sage.

LEMAIRE, T. (1991) Anthropological Doubt. In Nencel, L. and Pels, P (eds.), Constructing Knowledge: Authority and Critique in Social Science. London: Sage.

MARX, K. (1966) Capital vol. III. Moscow: Progress Publishers.

MATTICK, Jr., P. (1986) Social Knowledge: An Essay on the Nature and Limits of Social Science. London: Hutchinson.

NORRIE, A. (1993) Crime, Reason and History: A Critical Introduction. London: Weidenfeld and Nicolson.

OAKLEY, A. (1981) Interviewing Women: A Contradiction in Terms. In Roberts, H. (ed.), Doing Feminist Research. London: Routledge and Kegan Paul.

OINAS, P. (1999) Voices and Silences: The Problem of Access to Embeddedness. Geoforum 30: 351-361.

PEARSON, G. (1993) Talking a Good Fight: Authenticity and Distance in the Ethnographer's Craft. In Hobbs, D. and May, T. (eds.), Interpreting the Field: Accounts of Ethnography. Oxford: Clarendon Press.

PORTER, S. (1993) Critical Realist Ethnography: The Case of Racism and Professionalism in a Medical Setting. Sociology 27(4): 591-609.

POST, R.C. (1995) Constitutional Domains: Democracy, Community, Management. Cambridge, Mass.: Harvard University Press. POST, R.C. (1996) Subsidized Speech. The Yale Law Journal 106(1): 151-195.

ROBERTS, J.M. (1999a) Investigating a 'Criminal' Public Sphere. In Brookman, F., Noaks, L. and Wincup, E (eds.), Qualitative Research in Criminology. Aldershot: Ashgate.

ROBERTS, J.M. (1999b) Marxism and Critical Realism: The Same, Similar or Just Plain Different? Capital and Class 68(Summer): 21-50.

ROBERTS, J.M. (2000a) Speakers' Corner: The Conceptualisation and Regulation of a Public Sphere. Ph.D. Dissertation, University of Cardiff.

ROBERTS, J.M. (2000b) The Enigma of Free Speech: Speakers' Corner, the Geography of Governance and a Crisis of Rationality. Social and Legal Studies 9(2): 271-292.

ROBERTS, J.M (2001) Realistic Spatial Abstraction? Marxist Observations of a Claim Within Critical Realist Geography. Progress in Human Geography, forthcoming.

ROSE, N. (1999) Powers of Freedom: Reframing Political Thought. Cambridge: Cambridge University Press.

ROYAL PARKS AND GARDENS REGULATION ACT (1872), The Public General Statutes, vol. VII, chapter 15: 139-43.

ROYAL PARKS AND OTHER OPEN SPACES REGULATIONS (1997). <http://www.hmso.gov.uk/si/si1997/97163901.htm>

SAYER, A. (1992) Method in Social Science: A Realist Approach 2nd edition, London: Routledge.

SCHOLTE, B. (1986) The Charmed Circle of Geertz's Hermeneutics: A neo-Marxist Critique. Critique of Anthropology 6(1): 5-15.

STEINER, F. (1991) Introduction: Research as Self- Reflexivity, Self-Reflexivity as Social Process. In Steiner, F. (ed.), Research and Reflexivity. London: Sage.

TUSHNET, M. (1999) The Possibilities of Comparative Constitutional Law. The Yale Law Journal 108(6): 1225-1309.

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

VIRDEE, S. (2000) A Marxist Critique of Black Radical Theories of Trade-union Racism. Sociology, 34(3): 545-565.

VOLOSHINOV, V.N. (1930/1973) Marxism and the Philosophy of Language, trans. by L. Matejka and I.R. Titunik. London: Seminar Series.

WINCH, P. (1963) The Idea of a Social Science and its Relation to Philosophy. London: Routledge and Kegan Paul.

WOODIWISS, A. (1985) A Theoretical Prologue to a Socialist Historiography of Labour Law: Law, Discourse and Transpositioning. International Journal of the Sociology of Law, 13: 61-78.

WOODIWISS, A. (1990a) Social Theory After Postmodernism: Rethinking Production, Law and Class. London: Pluto Press.

WOODIWISS, A. (1990b) Rights vs. Conspiracy: A Sociological Essay on the History of American Labour Law. New York and Oxfrod: Berg.

Copyright Sociological Research Online, 2001