Copyright Sociological Research Online, 1996

Evaluating an Empowering Research Strategy: Reflections on Action-Research with South Asian Women

Abstract

Anti-racist sociology, feminist sociology and action-research share a concern with empowerment of 'the researched'. A review and critique of the concept of empowerment in anti racist, feminist and action research is used to argue for the use of Strauss's 'paradigm' for study of the negotiation of power in the research process. Power negotiations are discussed in relation to a reflexive case study of an action- research project which worked alongside South Asian women in Glasgow. The case focuses on the project set up, which was aimed at intrinsic empowerment, then on the community action which tried to respond to local concerns, and finally on the experiences of two researchers in the project. In conclusion, it is argued that empowerment, rather than an unquestioned, universally desirable goal, remains an issue for continuing debate.

Keywords:

Action-Research; Anti-Racism; Asian Women's Action Group (AWAG); Community Work; 'Crossroads' Project; Empowerment; Feminist Research; Glasgow

Introduction

...it is hard to speak against empowerment which, along with the proverbial motherhood and apple pie, is above reproach. (Stevenson and Parsloe, 1993: p. 55)

1.1

Empowerment may be another fashionable idea, or a new term for a process in which various researchers and professionals have been involved for some time. In this paper, I argue that there is a need to problematize empowerment in three interrelated (though not generally communicating) research fields; anti-racist sociology, feminist sociology, and action research. The first part of the paper identifies issues which have not, I will argue, been fully explored. Critiques of 'white sociology' which attempt to establish anti-racist research have debated issues of power arising from a racist society, and have suggested that empowerment of 'the researched' may be a way forward. Feminist researchers have argued that their work must empower the researched, and action-research has insisted on empowerment as its raison d'être. But all three are problematic, and critics have questioned the notion of empowerment, right or wrong, which they seem to espouse. Having raised a series of issues for debate, I go on in the second part of the paper to explore them through discussion of some field material from a research project conducted and developed over several years with women of South Asian heritage in the city of Glasgow, Scotland. The project set out to be anti-racist and feminist, and to empower local people to participate in community action and research.

Problematizing Empowerment

2.1

'White sociology' has been sharply criticised (e.g. by Lawrence, 1982, and many times since) for stereotyping the experiences of black people in Britain, and therefore preventing adequate representations of black people's views and concerns. In doing so, it is said to have reflected and reproduced, rather than challenged, the processes of radicalization and the structure of racism. Miles, writing of the study of 'race relations', argues that many sociologists of the 1970s and 1980s:

....employed uncritically the common-sense notion of 'race', reified it and then attributed it with the status of a scientific concept. (1989: p. 72)

2.2

In his view, they were thus perpetuating the particular form of race-thinking basic to racialization in the society around them. Like Lawrence (1982), he sees such work as perpetuating stereotyping and racism.

2.3

More recently in the development of anti- racist sociology, the issue of whether white researchers should be involved in research with black people has been debated. Douglas (1992: p. 39) for example has argued that white women's accounts of black women's health experiences are flawed because of 'the assumption that the shared experiences uniting women outweigh the differences in relation to "race" and class'. In her view (p. 41) 'racism is paramount' and, since white women do not experience racism, they cannot understand black women's experiences sufficiently to research them. In criticism, Bhavnani (1993: p. 42) states that in such arguments, 'experience is used as a truth which silences and ends the right to argue with it'. Kazi (1986) wants to drop experience as 'the claim to truth' (Bhavnani 1993: p. 42) and argues that, provided the stereotypes are dropped and racism challenged, black and white women can usefully work together in researching women's lives and contextualizing them in the wider structures of 'race', gender and class. Ramazanoglu (1989a, 1990) concurs with this view, and Rhodes (1994) also appears to support it.

2.4

Bhavnani's (1988) paper is particularly important here, because she links anti-racist sociology with a critical notion of empowerment. She argues that a simplistic view of empowerment would use qualitative research tools to allow the expression of views by groups in society who are not normally given this opportunity to contribute to research findings. Of itself, however, raising voices, she feels, does not constitute empowerment, unless the analysis then produced takes full account of the power context in which the views have been expressed: the fetishism of 'experience' can obstruct this process.

2.5

Feminist researchers (including Kazi and Ramazanoglu) have debated issues of empowerment of the 'researched' at some length, beginning with Oakley's (1981) and Finch's (1984) observations on interviewing women, which emphasize the use of qualitative research tools as means of researching women's views. But Ramazanoglu (1989b) has noted that whilst women's voices have increasingly been heard in defining research questions, in the process of research itself, and in the interpretation of results, rather little attention has been paid to distinctions of 'race' and class within the category 'women'. 'Women's experience' has tended to be regarded as homogeneous. Ramazanoglu's (1989b) re-examination of her own work demonstrates a revealing exclusion of black women from a study on the grounds that they would distort the research results.

2.6

In black feminist writing, criticism of white feminism, including its sociological versions, is strong. Lorde (1994) for example, argues that by emphasising one social division, between men and women, above all others, feminism denies racism, and denies the recognition of black women's difference. Whether or not it attempts to be empowering, therefore, it excludes black women. Aziz (1994) also argues that differences between women must be recognised by feminism, and, further, that black women must not be seen simply as victims of racism, but also as agents in their own social lives. If such criticisms can be addressed however, feminist work appears to offer considerable possibilities for challenging stereotypes, and for examining the inter- relationships between gender, 'race' and class.

2.7

Empowerment of 'the researched' has also been an important factor in debates about action-research. Kingsley's (1985) useful review notes considerable variation between projects labelled 'action-research' in terms of clients, funders, groups or categories researched, researchers and their roles, the definition of research questions, dealing with conflict, and the aims of the projects, i.e. what type of change they aim to produce. The work she reviews shares two basic characteristics, firstly feedback, that is, some sort of explicit interaction between the researcher(s) and the field, and secondly, change, that is, projects aim to alter existing professional practice, or organisational structure, or community activity. More recently, Whyte has defined participatory action-research as research in which:

...Some of the people in the organisation or community under study participate actively with the professional researcher throughout the research process from the initial design to the final presentation of results and discussion of their action implications. (1991: p. 20)

2.8

His volume concentrates on participatory action-research in industrial or agricultural enterprises, whereas Kingsley (1985) concentrates particularly on examples from community development fields. In all the cases they describe, researchers work, to some extent, with, rather than 'on', the researched, and their interpretation of the situation is, in one way or another, challenged.

2.9

Whyte's (1991) version of action-research, and those of his co-contributors, though radical in relation perhaps to a conventional attempt at objective social science, do not match the challenging work of recent feminist action-research (Reinharz 1992). Reviewing a largely American literature, Reinharz (1992: p. 181) identifies five types of 'change oriented' feminist research, of which the work to be discussed in this paper combines 'action-research' which promotes and evaluates change, and 'participatory research', in which those being studied take part in formulating research questions and design, and in data collection and analysis. Reinharz' third type (1992: pp. 186-189) is 'prevalence and needs assessment', conducted using methods which challenge existing assumptions, and aimed at discovering the numerical extent of experiences and need for services. The fourth type (1992: pp. 189-191) is 'evaluation research' which examines (p189) 'the effectiveness of different types of actions in meeting needs or solving problems', and also evaluates evaluation itself. Finally (1992: pp. 191-4) Reinharz identifies 'demystification', a research process based on the notion that the acquisition of knowledge in under-researched areas can be a basis for change, as it can challenge the status quo.

2.10

Feminist action-research stresses participation of 'the researched' to a very marked degree, and often, therefore, creates a very close relationship between action and research. Lather (1988) for example writes of the use of an action-research approach to encourage students taking a women's studies course to engage with the emancipatory perspective she saw as the aim of the course. She reviews the work of Mies, whose action-research project included street demonstrations to awaken local consciousness to the need for a women's refuge:

The purpose was to empower the oppressed to come to understand and change their own oppressive realities. (Lather, 1988: p. 572)

2.11

Cancian's work (quoted in Reinharz, 1992: pp. 182-3), involving herself with a group of colleagues who were having difficulties coping with dual roles, made the researcher one of 'the researched'. For Lather (1988) and Mies (Lather, 1988; Cook and Fonow, 1986), their work was aimed at emancipation, whereas Cancian was seemingly concerned with empowering the women with whom she worked, and herself, to tackle a difficult situation. Cook and Fonow (1986: pp. 20-1) review a series of examples in which feminist researchers involved themselves very closely in their work, in which the personal became the researchable. In these very close relationships, empowerment occurs for different people, who may include the researcher, for different purposes.

2.12

Other feminist researchers, however, have been critical of notions of empowering research. Scott (1992) for example, who was involved in evaluative research (one of Reinharz', 1992, types of feminist action-research) on AIDS/HIV education, describes how she came to the view that an uncritical insistence on empowerment right or wrong, which, in this case, entailed a considerable degree of participation by all parties to the project in the research, reduced the prospect of effective evaluation of the work, and defeated the aim of the project. She concludes that a detached researcher is needed for effective evaluation, and that a non-participatory approach would have achieved the aims of the project. Opie (1992) is sharply critical of Lather's (1988) work on her women's studies course, and accuses her of having a particular end in view, i.e. the emancipatory perspective. She argues that this is empowerment to do what the researcher wants, not empowerment of the researched to express their own views, or take their own actions. These criticisms, and the varying strategies identified before, draw attention to important questions about the nature of empowerment, and the extent to which it is essential to feminist research.

2.13

Writing of Sociology more generally, and commenting on various 'versions of the Sociologist's role', Silverman argues that the partisan Sociologist (one who takes sides; Becker, 1967):

...makes claims to know how things really are, while all too often ignoring what people are actually saying and doing. (1985: p. 188)

2.14

He suggests that committed researchers can come to their work with a predefined agenda, expecting particular outcomes. He accuses them of elitism, for which they, in turn, have castigated the 'objective', 'disinterested' scholars of more conventional types. These comments can develop Opie's (1992) views on the potentially impositional nature of empowerment, and raise issues about listening to those who, through empowering research approaches, raise views which are unexpected, or 'unrespectable' (Berger, 1963: pp. 56-61) from feminist points of view.

2.15

In other fields, the notion of empowerment is also coming under scrutiny. In community care practice, for example, as Stevenson and Parsloe (1993: p. 55) note, the concept of empowerment has been canonized to the point that it has become difficult to criticise (cf. Solomon, 1976, for an early discussion of social work practice). Elsewhere, the universal good of empowerment has been questioned: Maseide (1991) for example argues that the exercise of power is essential to effective medical practice, and Long and Long whose concern is Third World development, have warned researchers:

Although the word [empowerment] has become wedded to a discourse that stresses the need to 'listen to the people' and to understand the 'reasoning behind local knowledge' in order to arrive at appropriate alternatives 'from below', it is difficult to deny the connotation it carries of an 'injection of power' from outside aimed at changing the balance of forces..... No matter how firm the commitment to good intentions, the notion of 'powerful outsiders' helping 'powerless insiders' slips constantly in. (1992: p. 275)

2.16

These writers are, again, questioning empowerment as a universal good, and are complicating the issue by looking at issues of power distribution and contextualization. For example, Stevenson and Parsloe (1993) note that the stress in discussions of community care has been on empowerment of the client, or consumer, and argue that if other participants in community care, such as staff or carers, are not also empowered, it is unlikely that consumer empowerment will be effective. Thus they draw attention to the need to pay attention to power throughout a structure, and to avoid narrowly focusing on consumers or 'the researched'. Such breadth of concerns can usefully be maintained by using Strauss's (1978: p. 238) 'paradigm' for approaches to understanding the negotiation of power. This entails study of the interactive, negotiated distribution and use of power, which is then placed in a wider social structural context. For anti-racist, feminist sociology, part of this wider context must be patterns of racism and anti-racism, feminism and anti- feminism, and gender relationships. Where an action-research strategy is pursued, contextualization of this kind is important to counter the tendency for such work to remain locally based and rather inward looking (and cf. Ellis, 1989, on community projects' similar tendency).

2.17

A further issue arises from Long and Long's (1992: p. 275) observation on the 'injection of power from outside'. Such a view sets up, as they point out, a dichotomy of 'powerful outsiders' and 'powerless insiders', and presents a very restricted view of power as an either/or (in or out) phenomenon. In general, such a view would ignore the probability of negotiating power: in particular, it would ignore the agency and potential agency of insiders. In research it would, for example, serve to promote the determinism of racism and sexism, to the detriment of considering resistance to these. It would lead researchers to ignore the activity of 'the researched' in, for example, presenting a best face (Laslett and Rapoport, 1975) or a public account (Cornwell, 1984) or simply lying, to mislead the researcher (Sutherland, 1975), or refusing to participate in the research. To put it another way, this can be a way of increasing the arrogance of research, as researchers may believe they have the power to emancipate, by giving power.

2.18

In the case study of an action-research project which follows, I am concerned with the relationship between the theory and practice of feminism and anti-racism in research, and with evaluating a research strategy which was believed to be empowering for 'the researched'. I review issues of empowerment at several phases of the project, and in different aspects of action research, with reference to the general points raised in this section. I therefore focus on the questions of the potential benefits and costs of an empowering research strategy, on responses to those empowered, on negotiation and contextualization of power, and on the challenges to researchers presented by empowerment. I examine how the workers in the project tried to keep faith with its feminist and anti-racist principles, and follow through some of the effects and implications of their strategies.

2.19

Following an outline of the background to the community in which the study took place, the account is presented in four sections. The first ('Designing Empowering Research') deals with setting up the project, an attempt to 'design-in' empowerment, which was based on assumptions about the need for and desirability of empowerment of the researched. In this section, the emphasis is on action-research, and the debates about power it generates: feminism and anti-racism were rather peripheral at this stage of the project. The second section ('Empowerment in Community Action') examines the community action dimension of the work, particularly the challenges presented by local people to community workers. The theme of anti-racism comes to the fore in this section. Power struggles in the local community are discussed, and limitations on empowerment considered. The analysis suggests that value free empowerment is unlikely, and that claims to commitment to the researched always require careful scrutiny. The third and fourth sections ('Empowerment in Research' and 'Disempowerment of Research') concentrate on the research experience, looking at two researchers whose power negotiations and positions differed markedly, and who illustrate potential variation in the experience of empowering research. The conclusion argues for empowerment to be seen, not as a prescriptively universal good, but as a series of issues about power, relevant to practice and research, part of a debate to which anti-racists, feminists and action-researchers can all contribute.

2.20

The data used in the case study comes from the project archives, which consist of my own field notes, kept as a record of the project process, and of more formal minutes of meetings and other documents produced for discussion by the project team. I also draw on the published work of three members of the team, Harvie, Murray and Wardhaugh. All have read this account, and, whilst they do not necessarily agree with the commentary or conclusions, which are mine, do agree that I have not misrepresented their work.

The Context of Action-Research

3.1

Glasgow is Scotland's second city, located in the West of the country, on the River Clyde. An important centre of industrial development in the late 19th and early 20th century, the city has suffered greatly from the decline in heavy industry in Britain in the latter part of the 20th century (Checkland, 1981). Glasgow has fought an external image of poverty, slum housing, violence and drunkenness in recent years, promoting alternative images of a proud working-class and a European cultural centre (Checkland, 1981; Charsley, 1986). The population of 676,300 (Glasgow District Council, 1995: p. 9) occupies centrally located 19th century tenement housing of varying quality, larger houses, which once belonged to the industrial elite, and a series of outlying areas, ranging from the smart, leafy suburbs of Bearsden and Giffnock, to deprived areas of social housing, such as Castlemilk and Drumchapel.

3.2

During the 20th century, migrant workers from the Empire and Commonwealth were attracted to settle in the city. Predominantly from the Indian subcontinent, and to a lesser degree, Hong Kong, they came to work as traders, then as transport workers, and as relatively low paid service workers, especially in catering and the distributive trades (Maan, 1992). Initially they settled in the centre of the city, in the old tenements, notably in the Gorbals area, where Irish and Jewish immigrants had settled before them. Through the sixties and seventies, the people of South Asian heritage became more widely distributed through the city, though the centre of the community remained in the city centre. In the Census of 1991, there were 14,319 people who identified themselves as Pakistani or Indian (about 2% of the city population).

3.3

In the Gorbals, and the neighbouring area Govanhill, a local community group called the Gorbals Group was set up in 1957 with the aim of working to help local people 'in facing the problems of the area' (Bryant and Bryant, 1982: p. 27). The group was involved in campaigning to improve community facilities, especially for young people, and local housing conditions. Later (1967), the group, renamed Crossroads Youth and Community Association, managed to secure local authority funding, to employ workers, to run an advice shop where people could obtain advice about housing and other benefits, to work with other local campaigning groups, and to offer training placements for students taking social and community work courses. Though receiving funding from the statutory sector, Crossroads remained, at root, a voluntary sector organisation, and valued its independence.

3.4

In 1981, when the action research project began, Crossroads was operating from shop front premises in Govanhill, in the centre of the South Asian community, but was not generally successful in providing its services to local South Asian people. People of Indian and Pakistani heritage walked past the shop all day. The street contained many sari shops and small traders catering for the South Asian community, the Central Mosque and several Sikh temples were nearby, but Crossroads had little contact with this section of the community around it.

Designing Empowering Research

4.1

In 1981, I approached Crossroads with a colleague, Sue Nowikowski, as part of a process of trying to set up 'some research' on 'race relations', as we then called it, in Scotland. Crossroads was keen to develop its community work with South Asians, but was nervous of doing so because of lack of knowledge of the circumstances and experiences of South Asians in Scotland: at that time, very little information was widely available, and, throughout Britain, social and community workers were having similar problems (cf. Ballard, 1979; Cheetham et al., 1981; Young and Connelly, 1981; Ellis, 1989). Crossroads workers and Management Committee (local white people committed to community development through self-help; see Bryant and Bryant, 1982) felt that their work needed research support: they were looking for information to support action work.

4.2

We wrote up our experience and interpretation of the early contacts with Crossroads and the evolution of the project as a reflexive process, exploring our changing orientation towards the work as the relationship with Crossroads developed (Nowikowski and Bowes, 1983). That we felt the need for reflection expresses the degree of challenge to our usual ways of working that the experience presented.

4.3

We described the initial contacts cautiously, as involving 'very different orientations' on our part and on the part of the two community workers with whom we met. Very soon, as the process of trying to plan an action-research project began, a later meeting was described as 'frankly catastrophic', and, 'close to a breakdown of communication'. Our own brisk attempts to define research issues contrasted with the lengthy, and to us, often seemingly rather pointless and repetitive, meetings used by community workers to advance the process. It was clear that two very different approaches to the work existed, and our power to define the situation was coming into question: the community workers, perceiving us as seeking power, were consciously preventing us from controlling the development of the project, and this was an unfamiliar experience for us at the time. Ultimately, they believed that the stimulus for community action and therefore action-research must come from the grass-roots (cf. Murray, 1991; Ellis, 1989: pp. 151-3) and that they, as professional community workers, and we, as professional researchers, had to be made to respond to that grassroots demand.

4.4

These very difficult experiences led us into reading about taking sides in research, looking at Becker's (1967) work on taking sides, and at the accounts of previous action-research projects such as the Community Development Projects (CDPs) of the 1960s (Lees and Smith, 1975; C.D.P., 1977), and to sociological writing about research committed to 'the researched' (e.g. Stavenhagen, 1971; Schensul, 1974) and action-research generally (e.g. Town, 1973; Touraine, 1981). We were not, at that stage looking closely at feminism or anti-racism. 'Taking sides', from this literature, seemed to involve the participation of 'the researched' in the whole process of research, from deciding topics of concern, through data collection and interpretation, to the dissemination of results. In this tradition, research was linked to action, and aimed to stimulate or support it. For us, this was a new orientation, more 'unrespectable' (Berger, 1963: pp. 56-61) than the sociology and social anthropology to which we were accustomed.

4.5

The substantive focus of the research work was largely stimulated by Crossroads workers' view of the need for community action in the South Asian community in Glasgow. They saw evidence of multiple deprivation, and little social service help, and statutory social services were reluctant to work with South Asians because of lack of knowledge of their situation. From the research literature, it was clear that similar issues had arisen elsewhere in Britain (e.g. Young and Connelly, 1981; Cheetham et al., 1981). Since there was so little information available, because so little research or action had been done, we kept the initial focus as general as we could, whilst taking an early decision to focus on women, whose needs appeared most pressing.

4.6

Trying to address some of the problems of tension between action and research revealed in the literature, we attempted to institutionalize community accountability and control in a management structure for the project. The project was overseen by a committee consisting of Crossroads management, community workers and myself, (Nowikowski left the project at an early stage): from the start, efforts were made to recruit South Asian women to this committee. Each of the researchers, Wardhaugh and Harvie (see below), were accountable to the committee. Three local Pakistani Muslim women were recruited by the community workers at an early stage in the project, after visiting the advice shop. All three were somewhat marginal to the South Asian community at that time, and their isolation had brought them into the shop, but later they proved crucial first contacts, providing the way into community work with the target population.

4.7

From the beginning, therefore, there was an attempt to ensure community input into the project (cf. Reinharz', 1992, account of participatory research): though we did not use a discourse of empowerment at the time, this was intended to be an empowering strategy. One component of it was some disempowerment of the researcher: for example, Bowes and Nowikowski's notes record a discussion about when a researcher should commence work, during which our proposal for a period of research at the University, away from the field, was vigorously opposed by community workers, who argued that a researcher could thereby lose touch with the community and its concerns, developing her or his own interests. In retrospect, there was certainly a danger in this disempowerment, or 'deskilling' of the researcher, as in Scott's (1992) evaluation work: the danger lay in the potential rejection of outsider comment, and the fetishism of experience discussed earlier. A counterbalance to such a tendency was eventually provided by my own relative distance from the research field. It was helpful not to be directly involved in the day to day process of the project, and to take on the role of academic supervisor, which included a strong element of support for researchers in their endeavour, and a need to act as advocate for them during the process of negotiating research work. As I will discuss later, the researcher roles that were set up in the project were not always easy to negotiate by the researchers themselves.

4.8

In its formal constitution, the project was certainly in sympathy with much more recent discussion in feminist research, especially in terms of input from 'the researched', who were to participate in the research process. The relationship between researcher and researched was intended to involve far more mutuality than was conventionally accepted in research projects, that is, to be collaborative and participatory. Though I had been involved in the women's movement in Britain since my late teens, and had studied and written about women, which at the time was seen as constituting feminist work (cf. Cook and Fonow, 1986), I had failed to make connections between feminism and ways of doing research. These were the very early days of explicitly feminist methodological writing (Roberts, 1981, was just published), and the debates of the 1980s had barely begun.

4.9

Looking back, though we considered ourselves to be anti-racist, there was very little explicit reflection on issues of race in these early stages of the work, beyond rather sketchy references to racism and the resulting particular vulnerability of black women. We were beginning to recognize differences between women, who we saw as experiencing sexism (white women) or sexism and racism (black women), but did not reflect on this at any length. It is not at all surprising that white women researchers have come under attack from black feminists for their failure to understand difference (see the discussion of Aziz and Lorde, above), and we were guilty of this. There was no real examination of what anti- racist work might mean, especially for white professionals: we were taking our whiteness as unproblematic. Whilst we undoubtedly saw our work as anti-racist, we were not very clear about how it was so, beyond assuming that community empowerment would ensure anti-racist work. As the action-research proceeded, it became increasingly clear that issues of 'race' were more complicated than this, and had to be separately addressed, as I will demonstrate.

Empowerment in Community Action

5.1

Issues of racism came to the fore early on in the life of the project, and were encountered within Crossroads, and in some of the community groups with which it worked. The idea for the project had come from the employees, who had to convince the management committee that it should devote resources to the work. There was heated discussion. Murray writes:

Not all members at that time accepted that racism exists, or that it was an issue which deserved the Association's proposing redirection of resources. (1991: p. 141)

5.2

We faced arguments about colour- blindness, that Crossroads worked with all deprived groups, and did not single out any group for special treatment: problems of poverty and bad housing were shared, and the same for all those who suffered them. There were also problems in the local community: some local white people complained about resources being taken from them for the new work.

5.3

Crossroads began the action work by trying to develop ethnically sensitive practice, based on anti-racist principles. The process involved a series of responses to local racism, and a constant process of re- evaluation of Crossroads' work, or constant negotiation. For example, anti-racist training for the Crossroads executive and staff was run by outside specialists in order to raise awareness of racism, and how to challenge it. Early efforts to do so included work towards equal opportunities in employment practices, challenging racism expressed by clients in the advice shop, and in community groups. For the community workers, this was not easy: Murray (1991) describes considerable opposition to the anti-racist approach. For example, during community development work against racist harassment of South Asian families, a local, all white tenants association argued that because it represented all tenants, there was no need for separate work with South Asians. Some members of Crossroads argued that forming a black group would weaken tenant action in the area, which needed to show a united front. Eventually, after considerable debate, the work with the black group against harassment went ahead. The debate was, in many respects, about power.

5.4

The work with women also involved challenge, disagreement and re-evaluation of strategies. The women who founded the community group linked with the action-research, the Asian Women's Action Group (AWAG), late in 1984, had made contact with one another through contacting Crossroads at the advice shop. From the start, this women's group set its own agenda, becoming the centre of a local support network both for its members and for other women who had indirect contact with it. The group successfully raised funds for sewing classes, language classes, office management classes, welfare rights training and, later, assertiveness training. The membership was not constant over the years, as new recruits replaced members who moved into full-time training or employment. Crossroads supported the group, despite disagreements, including debate about the involvement of a white researcher (Wardhaugh, 1989).

5.5

The work with the women's group exhibited the anti-racist strategy in a number of respects. Firstly, it involved community workers accepting challenges to prevalent stereotypes of South Asian women as passive victims of oppression, or 'abjectly submissive' (Brah, 1992: p. 75). Secondly, it was linked to decisions to divert resources towards work with specifically black groups. Thirdly, it involved community workers self-consciously examining their own roles in the project: most were white and there was much discussion of issues relating to white women working with black women (further discussed below). Fourthly, the provision of resources in the form of support and facilities for the groups ran counter to the more general restriction of access to resources experienced by South Asian women.

5.6

In terms of empowerment, Long and Long's (1992: p. 275) comments (quoted previously) are particularly pertinent for drawing attention to the perpetuation of external power. The role of community workers was to facilitate AWAG's access to Association resources, including accommodation, funding for publicity (especially in the early stages), access to e.g. welfare rights advice and training, and to the fund of experience of applying for financial aid for community projects. These resources were provided at the behest of the Association, and could have been withdrawn. There remained therefore a power relationship between the women's group and the Association, even though this was defused as far as possible, and even though, quite soon, three South Asian women became involved in Crossroads executive. One resource available to the group was the research component of the action-research project: it is important to note that although the project organisation was aimed at ensuring community input and responses to it, and although the Association was attempting to ensure anti-racist working, researchers retained considerable influence over the running of the project, as I will show.

5.7

Crossroads' power was thus exhibited in its relationship with AWAG, in AWAG's access to resources. It was also illustrated in Crossroads' relationship with local people, where it was used to challenge the views of the white tenants' association (see above), and in the advice shop, where people making racist comments were, as a matter of policy, requested to withdraw them or leave. Thus it was power to disempower (racism cf. Bhavnani, 1988) as well as to empower. Further aspects of Crossroads' power are revealed by examination of the community action. In principle, this was organizing 'among the unorganized' (Murray, 1991: p. 140), and the importance of listening to the grassroots was emphasized. But this was not a straightforward commitment to the unorganized, right or wrong, or to South Asian women right or wrong. As the community development work proceeded, there was constant judgement about appropriate activity. For example, as I have noted, there were decisions to be made about the allocation of resources, and about how support for AWAG should run. At an early stage, AWAG raised the issue of domestic violence and proposed action on it: community workers judged this issue to be unsuitable for a group in its early stages and urged a focus on welfare rights (cf. Ellis', 1989, reports from Middlesborough and Manchester). The group soon became involved in making a video in Punjabi about access to welfare payments, which was shown widely to local South Asian women (Wardhaugh, 1991), and the domestic violence issue was left aside until a later date.

5.8

In this case therefore, empowering the local community by facilitating certain kinds of community action rather than others contained very strong elements of Long and Long's (1992) 'injection of power' from outside, and assessment by those 'injecting' power about the circumstances in which this should or should not be done. Outsider control in the negotiation of power remained considerable. Furthermore, there was an element of Opie's (1992) 'appropriation' here, in the judgements about which issue was suitable for action. It would be quite wrong to interpret this particular approach as empowering in the general sense, as the research was linked to efforts to promote community action of very particular kinds: there were certainly people in the local community who would have liked to restrict South Asian women's access to resources, but there was no way that Crossroads was going to support such wishes. In rejecting domestic violence as an issue for action at this stage, the community workers were looking at the possible outcomes of empowerment, judging this issue as too sensitive for a group inexperienced at campaigning, as AWAG was, and too sensitive for community workers who were still relatively ignorant about the South Asian community. The workers, in keeping with Crossroads' ethos, believed help with access to welfare benefits would be a positive development: they did not take the possible alternative view that benefits might induce dependency. Empowering strategies seem inevitably to involve 'appropriation': it is hard to see how value free empowerment might be achieved, and hard to see when it would be desirable. Taking sides must entail appropriation.

Empowerment in Research

6.1

Wardhaugh (1989, 1991) was recruited into the action- research project in 1984 after the structure had been set up. She was registered as a PhD student at the University of Stirling, and funded by a state studentship which had been awarded following a joint application by Bowes and Crossroads for a student project generally focused on the topic 'Self Help and Community Group Work: Action- Research with Asian Women in Glasgow'. To a considerable extent then, Wardhaugh's PhD project was predefined for her: she did not have the autonomy to choose her topic that many research students have. She therefore became involved in a new process of negotiation, within the pre-existing action-research project structure, about her role in the project, the topics she would research, the way she would research them, and the eventual focus of her thesis.

6.2

A general remit for the topics requiring research attention had emerged from the action-research project management committee and the work with AWAG. The researcher was to collect relevant local and national data, monitor the action work, and report regularly to, and participate in, the committee. Wardhaugh began her work with this broad remit, and spent her first three months establishing a role for herself within the Association and the community groups, learning about the local situation, and, especially, becoming involved with AWAG. Locating this experience in relation to the research literature, Wardhaugh (1989) refers to Hammersley and Atkinson's (1983) conceptualisation of the 'funnel structure' of ethnographic research, which starts broadly and gradually develops a narrower focus.

6.3

Early in 1985, discussion focused the project onto issues of Asian housing, particularly Asian women and housing, which were appropriate as a research focus in the sense that they could be directed into a PhD thesis and were of interest to Wardhaugh, appropriate for action work (in the sense that the community workers had the necessary expertise and experience of work on housing issues), and appropriate for AWAG, in whose discussions housing issues had been raised and assessed as being of key importance. Despite the general interest of the topic, Wardhaugh (1989) herself notes that, in practice, she had much more say over the focus of the research than did the other participants in the project, who she describes as keen to discuss but unwilling to decide, preferring Wardhaugh herself to select the focus. Wardhaugh describes the character of discussion:

The process was 'democratic', but mainly in the sense that everyone had an equal say in the final decision making, rather than full participation in the whole decision-making process. (1989: p. 24)

6.4

The design of the action-research project foresaw the researcher being a participant observer, quite closely involved in the community action work and with the community groups. Wardhaugh (1989) explains how she gained acceptance into a local role, and how the role itself was affected by her gender, ethnic identity and class, the negotiation of multiple identities. As a woman working with women, she felt this gave an immediate basis, in shared experience, for a good relationship. Being female also meant it was comparatively easy for her to be with Asian women: a man would not, she argues, have been welcomed as she was, because of culturally based prohibitions on women being in contact with unrelated men. Wardhaugh was also able to work with Asian men who treated her as an 'honorary man', focusing on her educated status: such work would, she argues, have compromised the honour of an Asian woman. She therefore describes the gender identity she had in the work as 'marginal' and argues that it allowed her greater freedom of movement because of its lack of definition. Working-class origins were a further asset, as she found she shared an informal approach to social interaction with the Asian women, most of whom were also working-class. Again, however, she was marginalized, because although of working-class origins, she was not living a working-class life.

6.5

Wardhaugh, a white woman, did not share ethnic identity with the AWAG women or members of the other community groups with whom she worked. She discusses (1989) the issues of white privilege and the need to challenge racism in her work. Her ethnic identity and its consequences for her work were explicitly discussed with AWAG and the Association workers following an AWAG meeting addressed by a local black woman, representing another community group, who challenged Wardhaugh's presence, as a white women, at the meeting, and her employment as researcher on the project. One member of AWAG expressed agreement with these views, whilst agreeing to continue working with Wardhaugh. The other members did not agree, arguing that they were happy to work with anti-racist white people like her. Wardhaugh's own view, which she maintained throughout her work was:

...a recognition of the realities of racism, but nevertheless a conviction that a guilty abstention from any involvement with black people was less useful than an attempt to work together in addressing issues of racism, poverty and poor housing.(1989: p. 29)

6.6

As the project work proceeded, Wardhaugh explains that she became part of the local scenery, 'Julia', someone who was always around, who worked with community groups and whose job was to do research. Whilst on this level, her similarities and differences with local people appeared rather unimportant once she had gained acceptance; in other ways, they were kept to the forefront. She was involved in the anti-racist activities in Crossroads (discussed above) and racism was therefore an issue of constant concern. Furthermore, the research work in which she was involved also entailed awareness of issues of racism and a process of critical contextualization in relation to other sociological work, empirical and theoretical. In this respect, the academic dimension of the research was important, necessitating as it did, awareness of broader questions, including those relating to the contextual analysis of power. Bhavnani's (1988) prioritizing of context over experience would predict that Wardhaugh could negotiate a place in the project in this way.

6.7

When collecting and interpreting research data, Wardhaugh worked in a consciously anti-racist and feminist way. For example, she decided to use qualitative methods of data collection which would prevent the researcher from predefining questions, and moulding responses into preconceived forms: here, she was participating in the feminist critiques of quantitative, positivist work, important at this stage of the development of feminist methodology (Cook and Fonow, 1986), whilst criticizing early feminist methodologists like Oakley (1981) for their lack of attention to issues of 'race'. The women and men with whom she worked speak movingly through her research reports, especially of the racist harassment many of them experienced:

I have to send the children to the shops for me, but often they will come back crying, saying [people] have stolen the money from them or something. I just don't know what to do, can you do anything to help? (Wardhaugh, 1991: p. 162)

A man spoke of his family's frightening ordeal:

The worst ... was when they set light to the door, they fastened a sack or something to the door, placed petrol on it and set it alight. Luckily we were all still awake, well, I hardly dare go to sleep at night, I'm always waiting for something to happen. So no-one was hurt, but we've had enough now, they'll have to move us, we just can't stay here. (Wardhaugh, 1991: p. 163)

Such accounts reveal the impact of racism in the area on many local families.

6.8

Importantly, Wardhaugh's approach did not operate with the preconceived type of analysis so sharply criticized by Opie (1992). Material like these interview statements would fit neatly into an argument stressing the determinism of racism in every aspect of the lives of black people in Britain. Wardhaugh's faithful response to 'the researched' however reveals an important counter to this view, particularly in the account of AWAG's development and activity. As I outlined above, AWAG was to some extent 'empowered' by community work support in the form of access to resources. But this was not a simple case of power being injected from the outside and 'given' to the 'powerless' (even though it seemed in some respects to be so). As Wardhaugh (1991) points out, AWAG was built on a tradition of organizing resistance originating in the Indian subcontinent (also stressed in Parmar's (1982) important critical essay, and cf. Liddle and Joshi, 1986) and on local women's own ways of negotiating space and action for themselves in what might, from an outsider's stereotypical point of view, have seemed to be a strictly circumscribed existence. One woman revealed her way of exercising her choice at an AWAG meeting:

He thinks I'm at the shops now, you know [laughs] ... I'm not really supposed to be here. ... I can't stay too long, though, as I can't explain being out for more than a couple of hours. (Wardhaugh, 1991: p. 154)

6.9

Wardhaugh's accounts of the group's development also reveal the network of mutual support of which AWAG became the centre, and the growing confidence of members of the group to work towards their own aims. She emphasizes that the strength of the group was largely based on its own pre- existing resource, a 'natural constituency' of 'women whose needs the group aimed to serve, and to whom they held themselves accountable' (1991: p. 155).

6.10

Thus the project was initially seen by the committee which managed it in Long and Long's (1992) terms as injecting power where none existed. The process, as it turned out, was not entirely one way, despite the Association's power. Firstly, the members of AWAG were neither totally unorganized, nor completely powerless, and part of the group's strength derived from its own resources, and secondly, the project set up invited challenges to researcher and community workers. These involved, as described, the predicted challenges about racism, but also some more unexpected ones, including the debate about Wardhaugh's role and identity. In setting up the work, we had not thought through issues of agency, adopting Crossroads' view that its task was 'organizing the unorganized' (Murray, 1991: p. 140), and falling in with stereotypical notions of passivity. This was, undoubtedly, naive, but probably a stage in the development of anti-racist feminism (Parmar, 1982, also sees the recognition and study of agency as a stage in the development of anti- racism).

Disempowerment of Research

7.1

From its early days, AWAG discussed the issue of domestic violence and the particular problems of Asian women trying to leave violent husbands. As already explained, this issue was not the focus of the first research project, despite being a central concern of the group. When the time came (1986) to apply for funding for the second research project, a discussion, dominated by AWAG members, argued that it should focus on domestic violence. AWAG was supporting several local women who had left violent homes, and more and more women were requesting support. Group members were critical of existing refuge provision in Glasgow, and wanted to campaign for better support, on the basis of reliable research, thus enlisting a resource whose voice they felt would be listened to by policy makers. The size of the problem was such that AWAG would soon, the group felt, be overwhelmed, and therefore unable to continue supporting abused women effectively. Harvie, the second researcher, began work in September 1986 with a clear remit (much clearer than Wardhaugh's had been) to provide research material to support community action by documenting violence in Asian families in Glasgow, developing analysis of violence using this data and comparisons with other projects, and examining the compounding factors of cultural background, language problems and isolation (this outline paraphrases the original funding proposal). Since rather little of Harvie's work is in the public domain, discussion of the experience of her project here must remain limited.

7.2

It is important to note that AWAG's stress on the importance of domestic violence issues for Asian women departed radically from other views. As Harvie's work demonstrated, many in the local Asian community, including 'community leaders' (those who represented 'the community' to the outside world, not necessarily with its full consent), and women who had experienced violence, were reluctant to discuss it openly. For community leaders, acknowledging domestic violence as a problem would have been bad public relations. Women who experienced violence were fearful of a negative response from friends and relatives which would leave them with no support, or believed that they somehow deserved violence. Whilst there were several refuges locally (including one specifically for Asian women), women did not know about them, or had heard bad reports about them, or did not want to cut themselves off from family, friends and possibly children by leaving their violent husbands. AWAG's concerns were, at that time, not widely reflected in the literature on domestic violence, especially the British literature (Mama, 1989, had yet to appear): researchers had been slow to examine the specific problems of minority ethnic women experiencing domestic violence, and to develop analyses which took account of racism. The Women's Aid movement and other refuge providers had also not fully addressed questions about black women, especially Asian women. In response to Harvie's (1991) nation wide survey of refuge provision for Asian women, many refuge groups were awakened to their lack of thought and their uncritical 'colour-blind' approach to refuge provision. One group wrote:

...this survey has highlighted our inexperience and lack of expertise and knowledge in this area. (Harvie, 1991: p. 187)

Another stated:

We think that we are non-racist, but filling in this form has made us realize that we are not positive enough towards anti-racism. We feel that many other refuges are like us... (Harvie, 1991: p. 188)

7.3

AWAG, in raising the issue and persuading community workers and researchers to pursue it, had far-reaching effects. For Harvie, it meant that her work was very closely linked with grass-roots concerns, and that her research and writing, including the refuge survey (Harvie, 1991) and a booklet on legal issues relating to abused women in Scotland (written for local discussion), responded very closely to these concerns. Compared with the researcher who sets her own agenda for research and writes for the academic community and possibly policy makers, Harvie was radically disempowered. For the women in AWAG, there was much more open discussion of issues of abuse, and greater awareness of its existence. For some refuges which took part in the nation wide survey (Harvie, 1991), the issue of provision for abused South Asian women was raised for the first time, whilst others found themselves re-examining their existing practice. Thus AWAG's ability to raise and pursue the issue of abuse extended into an ability to exercise wide influence. The action-research work, in this respect, seems to have had an empowering effect.

7.4

Harvie's own position was, however, extremely difficult, and became more so as the work proceeded. She wrote of her early lack of awareness of the full impact of racism on black women's lives, and described how her awareness grew with her work. She wrote:

...had I the political awareness of racism which I now have, I may not have undertaken this research project. It seems more appropriate for work with Black women to be carried out by Black women. (1990)

7.5

Thus she was faced with increasing doubt, as the project proceeded, based on her experiences of talking, often very deeply, with women who faced violence at home, and racism outside. Harvie became deeply involved with and committed to the 'action' dimension of the work. It was difficult to respond to the academic demands of the 'research' dimension, with this kind of commitment. This was another dimension of her disempowerment, in respect of her academic work. But she was not powerless in this situation: the commitment she developed was her own, and she was able actively to participate in negotiating her role in the project, as Wardhaugh had been. The disempowerment here was not so much of a person, as of a perspective, which Harvie eventually found to be incompatible with her own.

Conclusions: Power in the Research Process

8.1

In recent years, sociologists have become increasingly concerned with research as a social process, and therefore with reflecting on the role of researchers themselves in the process of research, and on issues regarding the purpose of sociology. Feminist, anti-racist and action-oriented researchers are engaging with the debate in somewhat different ways - action- researchers are more inclined to trace a direct link with Becker, whereas feminists and anti-racists identify with women and racialized social groups - but all are seeking perspectives which attempt to alter the previously existing power of 'establishment' researchers and research perspectives. The effectiveness of such strategies, particularly the extent to which they can create new kinds of power relationships which have new kinds of detrimental effects, has been the subject of extended debate. For example, Silverman (1985) is critical of many of the, as he sees it, extravagant claims to moral rectitude made by some proponents of action-research perspectives, and Opie (1992), as previously noted, questions the tendency of some feminist writers to impose their own view of the world on their respondents, silencing those who depart from such a view. Rhodes (1994) suggests that racial matching of interviewer and interviewee, often seen as an anti-racist strategy, may serve to perpetuate the marginalization of black researchers.

8.2

There can clearly be no detailed prescription for a research process which will empower and disempower in all the right places to the right degree, and the notion of 'correctness' here is certainly mistaken. The experience of action-research I have reviewed illustrates above all the complexities of power relationships in such a project, and exposes, unsurprisingly, the need for these to be consciously examined by participants in the research process. The power to define is particularly important, since this lies at the root of decisions about what to research, how to research it, and how to interpret and communicate results. It is probably easier for researchers to retain this power for themselves, but, as most of those working in the tradition reviewed here would agree, questionable whether the findings of work defined in this way reward anyone other than researchers themselves. The power to define operates at the micro level, whereas other power dimensions are much more wide-ranging (or higher level): these include society-wide divisions of gender, class, 'race' and ethnicity. Here Strauss's (1978) argument about contextualization is important.

8.3

Consciousness and analysis of power in the research process brings into focus questions concerning the extent to which the society-wide divisions determine people's experiences and life-chances. Again, this is not a new issue (of course, for many, it has been the sociological issue), but the perspective suggested here offers ways of tracing some pathways between structural determination and individual agency. Our project began with the probably naive notion that research questions generated from the 'grassroots' would be different from those generated from a review of academic literature. One of the main bases of this notion was that the oppression of South Asian women in Britain, through racism, sexism and deprivation, determined their silence and inaction. Even without going into the detailed findings of the project, it should be clear that the women with whom we worked, were not structurally oppressed into silence, but were completely competent and capable of speaking, presenting their views of the world and acting upon them: they did not need an action-research project to do this, though the project did become an effectively mobilized resource which helped strategies, which might otherwise have been difficult, to be pursued. As researchers, who represented white, middle-class power, we were challenged, disrupted (Opie, 1992) and silenced (Bhavnani, 1988), and got somewhere near a partnership with the women with whom we worked.

8.4

It may be that the power of racism and racist structures in modern Britain is so strong that the only way that research with people from racialized groups can (should?) proceed is with a set up in which 'the researched' are empowered to influence and question the process of research, and which pays attention to their agency and views. Thus might a potentially racist researcher be challenged. It is not that there is a finite fund of power to be shared out, such that empowering the researched would automatically disempower a researcher: the case study has shown that this was far from the case, and I have emphasized, throughout, the need to be conscious of, and examine the process of, power negotiation. This is especially important where researchers believe that the researched have been empowered, and may be thinking of power as in their gift. For Lawrence (1982) and many of the feminist writers, including Opie (1992), the adoption of an anti-racist and/or feminist viewpoint involves challenging stereotyping assumptions, including anti-racist and feminist ones, and maintaining a critical awareness throughout a project of ways in which they may influence the work.

8.5

Along with a loss of power might go a loss of responsibility. But this would defeat the point of an empowering approach, which is at least partly intended to allow the voices of 'the researched' to speak more clearly. Ribbens argues that:

Ultimately we have to take responsibility for the decisions we make, rather than trying to deny the power that we do have as researchers. (1989: p. 590)

8.6

It is also necessary to maintain and engage with a sociological discourse, which can analyze and explain the social relations observed. As I argued early in the paper, experience cannot be seen as equivalent to sociological analysis. However 'the researched' are empowered within the context of a particular project, researchers inevitably retain their ability to address wider audiences, of other researchers, other professionals, policy makers, and, probably in a more limited way, a wider public. Communication to these audiences of work which has implications wider than for the locality of a project is vital, and researchers retain this responsibility.

8.7

Anti-racist sociology, feminist sociology and action-research, though they rarely engage with each other and are, indeed, capable of antagonism, seem to be developing variations on the theme of power. There are particular areas which each perspective needs to address more fully. Feminist research needs to develop a clearer recognition of racism, and a less stereotypical view of black women. Action-research over-emphasizes experience, and requires to discuss the nature of its research element more reflexively and in relation to the maintenance of a sociological analysis. And the debates about anti-racist research have still to continue: the issues in this area were by no means resolved in the project discussed here, and black women remain a footnote in much sociological work about women (cf. Foster, 1995). All three perspectives are capable of challenging existing power structures by opening research to voices previously unheard. This is not however simply a matter of the 'injection of power from outside' (Long and Long, 1992: p. 175), as these perspectives can also force researchers to be aware of the agency of 'the researched', power from below, which a deterministic perspective assumes cannot exist.

Notes

1. The use of terminology, 'black', 'Asian' and 'South Asian' in this paper reflects terms which were used at different stages in the project, and our debates about them.
2. Thanks to Sue Scott and the anonymous Sociological Research Online reviewers for comments on the paper.

References

AZIZ, R. (1994) 'Feminism and the challenge of racism: deviance or difference' in H. Crowley and S. Himmelweit (editors) Knowing Women: Feminism and Knowledge. Cambridge: Polity Press.

BALLARD, R. (1979) 'Ethnic minorities and the social services' in V. S. Khan (editor) Minority Families in Britain: Support and Stress. London: Macmillan.

BECKER, H. S. (1967) 'Whose side are we on?', Social Problems, vol. 14, pp. 239-47

BERGER, P. L. (1963) Invitation to Sociology: a Humanistic Perspective. Harmondsworth: Penguin.

BHAVNANI, K. (1988) 'What's power got to do with it? Empowerment and social research', Text vol. 8, pp. 41-50

BHAVNANI, K. (1993) 'Talking racism and the editing of women's studies' in D. Richardson and V. Robinson (editors) Introducing Women's Studies: Feminist Theory and Practice. London: Macmillan.

BRAH, A. (1992) 'Women of South Asian origin in Britain: issues and concerns' in P. Braham, A. Ratansi and R. Skellington (editors) Racism and Anti-racism: Inequalities, Opportunities and Policies. London: Sage.

BRYANT, B. and BRYANT, R. (1982) Change and Conflict: a Study of Community Work in Glasgow. Aberdeen: Aberdeen University Press.

CHARSLEY, S. R. (1986) '"Glasgow's miles better": the symbolism of community and identity in the city' in A. P. Cohen (editor) Symbolising Boundaries: Identity and Diversity in British Cultures. Manchester: Manchester University Press.

CHECKLAND, S. G. (1981) The Upas Tree: Glasgow 1875-1975.....and After 1975-1980. Glasgow: University of Glasgow Press.

CHEETHAM, J., JAMES, W., LONEY, M., MAYOR, B. and PRESCOTT, W. (1981) (editors) Social and Community Work in a Multiracial Society. London: Harper and Row.

COMMUNITY DEVELOPMENT PROJECT (1977) Gilding the Ghetto: the State and the Poverty Experiments London: C.D.P. Interproject Editorial Team.

COOK, J. A. and FONOW, M. M. (1986) 'Knowledge and women's interests: issues of epistemology and methodology in feminist sociological research' Sociological Inquiry, vol. 56, no. 1, pp. 2-29.

CORNWELL, J. (1984) Hard-earned Lives: Accounts of Health and Illness from East London. London: Tavistock.

DOUGLAS, J. (1992) 'Black women's health matters: putting black women on the research agenda' in H. Roberts (editor) Women's Health Matters. London: Routledge.

ELLIS, J. (1989) Breaking New Ground: Community Development with Asian Communities. London: Bedford Square Press (National Council for Voluntary Organisations).

FINCH, J. (1984) '"It's great to have someone to talk to": the ethics and politics of interviewing women' in C. Bell, and H. Roberts (editors) Social Researching: Politics, Problems, Practice. London: Routledge and Kegan Paul.

FOSTER, P. (1995) Women and the Health Care Industry: an Unhealthy Relationship. Buckingham: Open University Press.

GLASGOW DISTRICT COUNCIL (1995) Glasgow's Housing Plan for the 90s: a Sharper Focus. Glasgow: City Housing.

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

HARVIE, L. (1990) "Fighting the Whole World?" Action-research on Sexual Violence Against Asian Women in Glasgow, (unpublished ms).

HARVIE, L. (1991) 'Sexual violence and the voluntary sector: Asian women and wife abuse' in A. M. Bowes and D. Sim (editors) Demands and Constraints: Ethnic Minorities and Social Services in Scotland. Edinburgh: SCVO.

KAZI, H. (1986) 'The beginning of a debate long due: some observations on "Ethnocentrism and socialist-feminist theory"', Feminist Review, vol. 22, pp. 87-91.

KINGSLEY, S. (1985) Action-Research: Method or Ideology? Wivenhoe: Association of Researchers in Voluntary Action and Community Involvement.

LASLETT, B. and RAPOPORT, R. (1975) 'Collaborative interviewing and interactive research', Journal of Marriage and the Family, vol. 37, no. 4, pp. 968-77.

LATHER P. (1988) 'Feminist perspectives on empowering research methodologies' Women's Studies International Forum, vol. 56, no. 3, pp. 257-277.

LAWRENCE, E. (1982) '"In the abundance of water, the fool is thirsty": sociology and black "pathology"' in Centre for Contemporary Cultural Studies (editors) The Empire Strikes Back: Race and Racism in 70s Britain. London: Hutchinson.

LEES, R. and SMITH, G. (editors) (1975) Action- research in Community Development. London: Routledge and Kegan Paul.

LIDDLE, J. and JOSHI, R. (1986) Daughters of Independence: Gender, Caste and Class in India. London: Zed.

LONG, N. and LONG, A. (1992) (editors) Battlefields of Knowledge: The Interlocking of Theory and Practice in Social Research and Development. London: Routledge.

LORDE, A. (1994) 'Age, race, class and sex: women redefining difference' in H. Crowley and S. Himmelweit (editors) Knowing Women: Feminism and Knowledge. Cambridge: Polity Press.

MAAN, B. (1992) The New Scots: the Story of Asians in Scotland. Edinburgh: John Donald.

MASEIDE, P. (1991) 'Possibly abusive, often benign, and always necessary. On power and domination in medical practice', Sociology of Health and Illness, vol. 13, no. 4, pp. 545-61

MAMA, A. (1989) The Hidden Struggle. London: London Race and Housing Research Unit.

MILES, R. (1989) Racism. London: Routledge.

MURRAY, C. (1991) 'Ethnic minorities and community work: the experience of Crossroads Youth and Community Association' in A. M. Bowes and D. Sim (editors) Demands and Constraints: Ethnic Minorities and Social Services in Scotland. Edinburgh: SCVO.

NOWIKOWSKI, S. E. and BOWES, A. M. (1983) 'Action-research in multi-racial Glasgow', unpublished ms (available from author).

OAKLEY, A. (1981) 'Interviewing women: a contradiction in terms' in H. Roberts (editor) Doing Feminist Research. London: Routledge and Kegan Paul.

OPIE, A. (1992) 'Qualitative research, appropriation of the "other" and empowerment', Feminist Review, vol. 40, pp. 52-69

PARMAR, P. (1982) 'Gender, race and class: Asian women in resistance' in Centre for Contemporary Cultural Studies (editors) The Empire Strikes Back: Race and Racism in 70s Britain. London: Hutchinson.

RAMAZANOGLU, C. (1989a) Feminism and the Contradictions of Oppression. London: Routledge.

RAMAZANOGLU, C. (1989b) 'Improving on sociology: the problems of taking a feminist viewpoint', Sociology, vol. 23, no. 3, pp. 427-442.

RAMAZANOGLU, C. (1990) Methods of Working as a Research Team (Women Risk and AIDS Project Paper 3). London: Tufnell Press.

REINHARZ, S. (1992) Feminist Methods in Social Research. Oxford: Oxford University Press.

RHODES, P. J. (1994) 'Race-of-interviewer effects: a brief comment', Sociology, vol. 28, no. 2, pp. 547-58.

RIBBENS, J. (1989) 'Interviewing - an "unnatural situation"?', Womens Studies International Forum, vol. 12, no. 6, pp. 579- 92.

ROBERTS, H. (1981) (editor) Doing Feminist Research. London: Routledge.

SCHENSUL, S. L. (1974) 'Skills needed in action anthropology: lessons from El Centro de la Causa', Human Organisation, vol. 33, no. 4, pp. 203-9.

SCOTT, S. (1992) 'Evaluation may change your life, but it won't solve all your problems' in P. Aggleton, A. Young, D. Moody, M. Kapula and M. Rye (editors) Does it Work? Perspectives on the Evaluation of HIV/AIDS Health Promotion. London: Health Education Authority.

SILVERMAN, D. (1985) Qualitative Methodology and Sociology. Aldershot: Gower.

SOLOMON, B. B. (1976) Black Empowerment - Social Work in Oppressed Communities. New York: Columbia University Press.

STAVENHAGEN, R. (1971) 'Decolonizing applied social sciences', Human Organisation, vol. 30, no. 4, pp. 333-43.

STEVENSON, O. and PARSLOE, P. (1993) Community Care and Empowerment. York: Joseph Rowntree Foundation with Community Care.

STRAUSS, A. (1978) Negotiations: Varieties, Contexts, Processes and Social Order. San Francisco: Jossey Bass.

SUTHERLAND, A. (1975) Gypsies: The Hidden Americans. London: Tavistock.

TOURAINE, A. (1981) The Voice and the Eye. Cambridge: Cambridge University Press.

TOWN, S. (1973) 'Action-research and social policy: some recent British experiences', Sociological Review, vol. 21, no. 4, pp. 573- 96.

WARDHAUGH, J. (1989) Asian Women and Housing: the Potential for Community Action, unpublished PhD thesis, University of Stirling.

WARDHAUGH, J. (1991) 'Asian women: campaigning for self-help' in A. M. Bowes and D. Sim (editors) Demands and Constraints: Ethnic Minorities and Social Services in Scotland. Edinburgh: SCVO.

WHYTE, W. F. (1991) Participatory Action- research. London: Sage.

YOUNG, K. and CONNELLY, N. (1981) Policy and Practice in the Multi-racial City. London: Policy Studies Institute.