Copyright Sociological Research Online, 1997

Romm, N. (1997) 'Becoming More Accountable: A Comment on Hammersley and Gomm'
Sociological Research Online, vol. 2, no. 3, <http://www.socresonline.org.uk/2/3/2.html>

Abstract

This article provides a response to Hammersley and Gomm's article entitled Bias in Social Research (1997). Hammersley and Gomm's proposed conception of bias is rooted in a particular view of the pursuit of scientific knowledge - a view which they call nonfoundationalist. The way in which Hammersley and Gomm account for their nonfoundationalist view and the way in which they level critiques against relativism and standpoint epistemology, are challenged in this article. The discussion is focused around my concern that their account excludes (as outside the range of relevant argument in research communities) a serious consideration of alternative epistemological orientations.

Keywords:

Accountability; Bias; Reciprocity; Research Process; Responsibility

Introduction

1.1

In their article on Bias in Social Research (1997), Hammersley and Gomm develop a proposal for conceiving bias in what they call a nonfoundationalist view of the pursuit of scientific knowledge. As part of their argument, they suggest that however researchers proceed, they have a special responsibility to work under the constraint of trying not to produce 'an account that is at odds with the evidence available about the relevant phenomena' (¶4.2). They have a responsibility to 'maximize the chances of discovering the truth about the matter concerned' (¶4.14). Bias is that which mitigates against the fulfillment of this responsibility. When researchers proceed in ways that have 'an unacceptable chance of being erroneous' and when this behaviour could have been 'recognised and minimised', they may be accused of bias (¶4.11 - ¶4.13).

1.2

The commentary that I provide in this article is based on my locating/exploring a view of the process of inquiry that (I contend) has not been sufficiently accounted for in Hammersley and Gomm's conception of bias. The suggestions of those who propose that the process of knowledge-construction is to be assessed in terms other than its maximising chances of truth-discovery as Hammersley and Gomm define it, are unaccounted for in the discussion. The proposal that knowledge-construction activities should be linked to cultivating forms of relationship which do not unfairly authorise particular ways of accounting at the expense of others, are glossed over in their article.

1.3

My attempts to engage with their argument begin by my considering their so-called nonfoundationalist response to foundationalism before I move on to consider (and criticise) the arguments that they elucidate against relativism and standpoint epistemology.

Foundationalism

2.1

Hammersley and Gomm define foundationalism (in its extreme form) as linked to the belief that 'research, when it is properly executed ... produces conclusions whose validity follows automatically from the 'givenness' of the data on which they are based' (¶2.3). According to this view, when bias tarnishes the research process, the link between conclusions and evidence becomes severed - resulting in erroneous conclusions.

2.2

Hammersley and Gomm point out that foundationalism for a long time penetrated much of the thinking behind quantitative and qualitative methodological canons of research (¶2.7 and ¶2.8). The idea that quantitative methods could be used for 'transforming data into valid conclusions' and the idea that qualitative inquiry could be used to 'depict reality in its own terms', both imply the suggestion that conclusions could be mapped onto the relevant evidence. Provided that researchers' attentiveness to 'the evidence' is not hampered through bias, it was supposed that true conclusions about the phenomena could be attained.

2.3

Such foundationalist thinking according to Hammersley and Gomm does not take account of the continuing uncertainties that beset the process of aligning conclusions to appropriate 'evidence' (¶2.6).

Hammersley and Gomm's Proposed Nonfoundationalist Conception of Bias

3.1

Hammersley and Gomm suggest that one of the features of what they call a nonfoundationalist view of the practice of scientific inquiry, is that it takes into account that the validity of statements cannot be read off from 'the evidence' in any straightforward way. Because of this, the practice of creating conclusions will always be marked by uncertainty. Therefore, it is important, they argue, to follow a more 'moderate line' (than foundationalism), admitting the uncertainty of all attempts to arrive at conclusions (¶2.6). However, they suggest that this admission of uncertainty does not and should not commit communities of inquirers to take the position that there is no point in trying to avoid error. On the contrary, it is still crucial (indeed, more so than ever) to develop a concern to attempt to 'avoid the danger of accepting as true what is in fact false' (¶4.2). Researchers, they claim, have a special responsibility to 'do their utmost to find and keep to the path which leads towards knowledge rather than error' (¶4.3).

3.2

This responsibility of researchers, Hammersley and Gomm suggest, requires an awareness of the importance of collegial accountability. Colleagues' arguments about the way research methods are being used and the way evidence is being appropriated need to be taken into account, so that accounts produced are less likely to be at odds with the evidence available about the relevant phenomena (¶4.2). This is a more difficult process than is admitted within a foundationalist view of knowledge. But as long as researchers are committed to developing accounts with maximum chance of avoiding errors that could have been avoided, there is some hope, according to Hammersley and Gomm, of offering better accounts (than if such commitment is not evidenced).

3.3

According to Hammersley and Gomm an essential quality of researchers is that they are (or need to be) specifically committed to trying to avoid 'systematic and culpable error' (¶4.13). This implies avoiding the temptation to tailor research to serve goals other than the search for knowledge and/or the temptation to tailor the research process (collection, analysis or interpretation of evidence) to bolster predetermined conclusions. Errors caused by bias are errors that researchers 'should have been able to recognise and minimise, as judged either by the researcher him or herself (in retrospect) or by others' (¶4.13). Accountability in research communities implies that researchers are on the look-out for such errors.

3.4

Hammersley and Gomm's article can be read as a plea to try and foster this kind of community of inquirers. However, their definition of bias already presupposes a certain view of what kinds of considerations are relevant to discussions about knowledge-creation in society. Some alternative considerations are explored below.

Comment on Hammersley and Gomm's Argument

4.1

Hammersley and Gomm contend that the goal of scientific inquiries is to strive to produce accounts of independently existing phenomena (those relevant to the inquiry at hand). Researchers, they claim, should be oriented in this direction, in order to operate their responsibilities as researchers. A problem with Hammersley and Gomm's view, however, is that it fails to provide for definitions of inquiry which emphasise the inclusion of moral considerations at the moment of knowledge-creation. In terms of such definitions, inquiry is to be conceptualised in terms other than its supposed ability to forward insight about independently existing phenomena. For instance, it might be conceptualised in terms of its possible involvement in invoking what Lather (1991: p. 60) calls 'maximal reciprocity' in the inquiry process itself. The notion of maximal reciprocity may be seen as a guiding commitment by those wishing to experiment with its meaning(s) in practice - meanings which in Lather's terms (1991: p. xvi) incorporate polyvocal complexity, paradox, and ambiguity. Dilemmas arise, for instance, as soon as we recognise that our own input in creating constructions will make a difference to others' conceptions of possibilities for seeing- and-acting; but that we cannot avoid the responsibility of creating such constructions. Our 'accountability' then becomes linked to our being aware of this potential problematic in the knowing process - so that it can be at the forefront of our consciousness, penetrating our own sense of possibilities for relating in 'the world' (Romm, 1996a: pp. 25 - 27).

4.2

It could be argued - from a moral point of view - that when researchers operate with a conception of knowledge such as that endorsed by Hammersley and Gomm, they already might be contributing to a process of sustaining unnecessarily certain forms of authoritative relationship in society (both in so-called research communities and in relation to other participants in society). (See, for instance, Brown, 1977, 1989; Stanley and Wise, 1983, 1993; Lather, 1991, 1993, 1995; Gergen, 1994; McKay and Romm, 1992; Romm, 1994, 1995; Weil, 1996; Humphries, 1997.) This is because inquiry practices operated in terms of the supposed following of 'the path which leads towards knowledge' become regarded as being necessarily of higher status than inquiry practices deemed as falling outside of this authorised epistemological ambit. Hammersley and Gomm do not appear to see the point of this complaint that has been leveled by those proposing alternative epistemological orientations. They argue that researchers must recognise the responsibility to keep on the path that leads towards knowledge (as they have defined it). Research operated by those accepting this responsibility rightfully, they suggest, has a higher acclaim than research which does not try to operate with this intention.

4.3

Thus, for example, Lather's suggestion that researchers might consider themselves as being accountable (in some way) to 'people's struggles for self-representation and self-determination' (Lather, 1995: p. 42) would be regarded in terms of Hammersley and Gomm's epistemology as a possible detraction from the proper research aim of getting closer to 'the truth about the matter concerned'. But to be accountable, as Lather proposes, to people's struggles for self-determination and self- representation means that inquirers' responsibilities become seen as linked to searching for ways of relating such that their desired research agendas and their ways of constructing/defining realities in research processes, provide maximum opportunities for others too to develop constructions (and attendant choices). The way in which research constructions might come to impact in the social world (as they impact on others' understandings) and the way in which they are authorised, are, in terms of this argument, not issues that can be left out of discussions about proper responsibilities. The complaint of those who raise such issues is that Hammersley and Gomm's proposed epistemological stance detracts from a concern with exploring these responsibilities.

Comment on Hammersley and Gomm's Conception of Relativism

5.1

Hammersley and Gomm dispense with what they see as relativist epistemological positions. They aver that the relativist argument that we live in a world of 'multiple realities' is itself presented as a universal claim to validity (¶3.3). They see as unmanageable the contradiction between proposing that all statements are relative and claiming to know that there 'are' multiple realities.

5.2

But this 'contradiction' can be re- examined from a different angle, which provides a way of shifting the terms of the contradiction. It is possible to offer a different account of the character of the claim that there 'are' multiple realities. The claim need not be referring to some feature of existence operating independently of the way that people in social life orient themselves. As inquirers we can relate to 'reality' as Hammersley and Gomm do - by proclaiming the need to orient ourselves to the (posited independently existing) phenomena that can best inform our judgements (even accepting that these may be somewhat fallible). But we can also decide to engage with 'reality' with a different intention. We can suggest that it is important to cultivate a sense of our presence in constructing (and ongoingly reconstructing) our world, so as to make room for others to cultivate their (perhaps different) relationship with 'reality'. The claim that 'there are multiple realities' can be read in this light. The statement may be seen as referring to the speaker's sense that s/he has a responsibility to allow for people to have different ways of presencing their worlds. To suggest that phenomena 'have no existence' independently of accounts (a suggestion which Hammersley and Gomm reject - ¶4.2) is to offer an alternative way of participating in reality-construction to that offered by Hammersley and Gomm. It is to offer a different way of being-in-the-world - a way which strives to make provision for people to develop their choices of vision and of action by recognising, working with, and bearing in mind, others' possible choices and concerns. This, in any case, is one way of treating the 'relativist' claim that there 'are' multiple realities.

5.3

Hammersley and Gomm assert that relativism often has the additional weakness of allowing for a proliferation of realities. This is related to their assertion that it might display an 'undiscriminating tolerance' (¶3.3). It might be noted here that if the claim that there 'are' multiple realities is read as incorporating in some way a moral stance (as was suggested above), then this claim can be accompanied by suggested, albeit tentative, criteria to prevent tolerance from becoming 'undiscriminating'. Criteria can be proposed to challenge positions that are deemed as likely to have the outcome of excluding others' rights to expression. Lather, for instance, does not shy from proffering certain notions of validity which she sees as serving to endorse the drawing out of different forms of expression. This supplies a counterweight to the definition of validity proffered by Hammersley and Gomm (¶4.6) - which defines it simply in terms of the power of statements to represent relevant phenomena. Using Lather's alternative notions of validity, inquirers can challenge claims which seem to embody 'controlling codes' or which can be argued to deny 'multiple openings' or which fail (somehow) to 'bring ethics and epistemology together' (Lather, 1995: pp. 54 - 55).

5.4

Statements which are seen as likely to create, say, unduly controlling, settled, fixed, or morally threatening outcomes (in terms of judgements that admittedly are themselves fallible) would be challenged with the hope to curtail their proliferation - by trying to locate ways in which they might be becoming exclusionary.

Comment on Hammersley and Gomm's Account of Standpoint Epistemology

6.1

Hammersley and Gomm suggest that because of the internal difficulties which beset relativism, many relativists supplement their stance with a 'standpoint epistemology' (¶3.4). In terms of this epistemology, as Hammersley and Gomm define it, some particular adopted standpoint is seen as providing access to 'the truth'. They argue that certain Marxist positions, as well as feminist ones, can be labeled under this category (¶3.5). Standpoint epistemology privileges the views of those considered capable of offering 'insight into the nature of the world' by virtue of their social location (¶3.5). They point out that there are versions of standpoint epistemology which appreciate that not all those belonging to the 'right' social location possess the appropriate insights. This is supposedly because they are misled by the dominant ideology, which mitigates against their attaining potential insights which they otherwise might have. But Hammersley and Gomm argue that once it is admitted that people in the 'right' locations may be misled, standpoint epistemology faces a problem (¶3.8). It can no longer appeal to social location as a located 'source' of truth-discovery to ground proposed insights.

6.2

Hammersley and Gomm note that sometimes (though not often these days) standpoint epistemologists refer to some form of scientific practice to find a grounding for insight (¶3.8). Hammersley and Gomm trace this solution back to the Hegelian view (which they see Marx as modeling) that the distinction between knower and known can eventually be overcome so that 'true knowledge' can be realised (¶3.5). Once this Hegelian position is supported in some guise, it is suggested (according to Hammersley and Gomm's characterisation) that although certain knowers might not be inclined to practice logic or science in the correct fashion due to biases arising from their social location, these are still in principle sources of knowledge. This argument, as Hammersley and Gomm seem to admit (¶4.2) bears some similarity to their own realist position, in that 'science' is now seen as the source of advancement towards the chance of truth-discovery. The proviso within Hammersley and Gomm's position is that it is recognised that seeking the path of knowledge is a matter of maximising chances of truth-discovery - rather than a matter of ever being able to claim with certainty that 'true knowledge' is being attained.

6.3

But more importantly for this article, it must be remembered that there is much dispute in both Marxist and feminist discourse regarding the way in which social inquiry should be directed. Gouldner (1980) points out that within what he calls 'Critical Marxism' (as ideal-typically distinct from 'Scientific Marxism'), proffered arguments can be read as invitations to evoke certain discussions in society - rather than supposedly operating to advance 'insight into the nature of the world' (to use Hammersley and Gomm's terminology). Expanding upon Gouldner's explorations, Romm (1991) offers a way of envisioning the character of Marxist statements, and attendant processes of inquiry, when these are seen as linked to a nonrealist Marxist stance. In similar vein, Pleasants (1996) offers an exploration of implications of pursuing a reflexive Marxism which shies away from representational speech. The idea here is that Marxism as a discourse is not to be understood as pursuing ways of generating concepts which can reflect mechanisms operating in some posited external reality. Rather, it is to be understood as posing insights which can become the basis for discussing the possibility of generating new forms of social relationship in society. The status of the 'insights' is not to be considered as connected with whether or not they supposedy 'represent' posited mechanisms operating in social reality.

6.4

As far as feminism is concerned, Williams (1993) indicates that many feminists too have not posed their arguments in terms of claims to be able to generate insight about 'the nature of the world'. The goal of enriching mutual 'understanding' (on the part of so-called professional researchers and others in society) within constructivist feminism is seen as linked up with a commitment to (try to) set up more collaborative knowledge- construction processes. Collaboration here does not imply that some consensus about how to view experiences/phenomena is sought as part of the inquiry process. What is sought is a way of creating 'openings' for various contending positions to be brought into consciousness, as a basis for the involved participants to move beyond the confines of their initial perspectives; so that they can in turn envisage fresh ways of experiencing 'the world'. In this sense their understanding can be said to be enriched through the process of encounter with alternatives.

6.5

Feminist researchers' ways of proceeding in the inquiry process - when underpinned by such a view of discursive encounter - are aimed at setting up arenas of inquiry where new 'openings' can be explored. The goal is not to discover something 'about' social reality as if this ever can be appreciated independently of self-engagement, but to ensure that the process of allowing for mutual encounter in the experience of being-in-the-world operates in a fair manner. (See, for instance, Griffiths, 1990; Stanley and Wise, 1993; Aphane et al, 1994; Lather, 1995; Lincoln, 1995; and Weil, 1996, for accounts which can be interpreted in this vein.) This concern is what marks off these feminist positions from those adopting a more realist epistemological stance. The concern with mutuality of encounter implies, in Lather's terms, that the connection between ethics and epistemology is central to the way of proceeding in the inquiry process, as well as to judgements of validity (which can always be revisited). In each situation, difficult decisions on the part of inquirers have to be made as they themselves (and bearing in mind their encounter with others) judge the fairness of their way of exercising choices. Continuing discursive entries into conversations about what fairness in different situations might amount to, are, of course, invited within such an epistemological orientation.

6.6

Hammersley and Gomm gloss over these kind of 'radical' arguments for the development of an alternative epistemology (to the one that they proffer). They prefer to discuss both 'relativism' and 'standpoint epistemology' in terms of pretended claims to offer 'insight into the nature of the world'. And they concentrate in their article on showing that these claims are suspect.

Cutting across Hammersley and Gomm's (Epistemological) Standpoint

7.1

Space in this commentary on Hammersley and Gomm's account does not permit a detailed exploration of 'other' paths that research can follow - that is, other than provided for in their account of 'the path which leads towards knowledge'. As I have attempted to show in this article, it is possible to cut across their proposed conception of the path of knowledge. This implies cutting across their proffered epistemological standpoint. Smaling (1995: p. 22) points out that it is possible to draw out some of the orientations that might be required on the part of researchers wishing to operate outside of the ambit of a realist epistemological stance. One quality that he suggests as important is that of 'letting-be and letting-go' as we encounter one another's experiences (1995: p. 24). With such an orientation our attention is not directed towards trying to generate descriptions and explanations of some 'object of investigation' (using others' views simply as some kind of reality- check). Rather, our attention is directed to allowing ourselves, and others, to 'let' one another 'be' (that is, find ways of operating that are meaningful) and in this process to 'let go' (of ideas which might stifle our own or others' capacity to 'let be').

7.2

This way of 'being' has implications for the way that research methods are operated and their utility judged. Instead of seeing them as tools to aid us in our appreciation of (posited) independently existing phenomena and instead of judging them in terms of whether their use is likely to help fulfill this purpose, we can see them in a different light. Let us consider some examples.

7.3

We can start with the example of some experiments undertaken by Schratz and Walker (1995). People were asked to sit side by side and touch an object and then engage in discussions about it. Each was given a different part of 'the object' - an apple - to touch (1995: p. 25). From their experiment (undertaken a number of times with different participants) Schratz and Walker indicate that they found people reluctant to alter their opinion regarding the object through listening to the experiences of others (1995: p. 34). This reluctance is interpreted by Schratz and Walker as linked to the participants' desire to 'search for agreement' (1995: p. 31). They infer from this that if people were less inclined to search for agreement between their different experiences, they would be better able to reach an accommodating perspective (1995: p. 34).

7.4

Schratz and Walker indicate that they do not wish their own sense-making of these experiments to be treated as an attempt to supply an account of 'the evidence' regarding participants' responses. This is not the point of their inquiry. They indicate in their book that they are using a narrative style which is conversational - so as to allow the reader (and all subjects in society) to 'ask questions, make suggestions, take paths that run off at tangents ... ' (1995: p. 14). They hope that others will participate in deciding how to treat the 'information' offered. Schratz and Walker clearly do not wish their experiments to be considered as tools for generating accounts of regularities written into social life. Instead they are presented as devices for generating visions of experiences for people to (re-) engage with, as they work with, passed, and around, the tendencies discerned/located by Schratz and Walker.

7.5

Survey research too can be used with alternative intentions to those suggested within Hammersley and Gomm's account of the purpose of scientific inquiry. Survey questions can be asked about issues that have been raised of concern in society, and this information can be compared with other information - while at the same time recognising that 'the findings' reflect ways of drawing out and creating 'information'. Most survey research findings are not treated in this fashion. The surveys are normally 'validated' by stating the 'degree of confidence' with which stated descriptions and correlations (comparison of information generated by asking different questions) are likely to represent the relevant relationships 'in reality'. But there is no reason why survey research methods and statements springing from them need be validated in this way. They could rather be seen as opportunities for allowing people to 'let-be' and 'let-go' - by expecting different people (including the initiating researchers, participants and audiences) to engage in different ways with 'the information' arising from the inquiries. (Romm, 1996b: pp. 188 - 192, offers some examples of how such engagements might be effected.)

7.6

Golding (1996), for his part, offers an indication of the way in which ethnographic research can be undertaken without committing to a realist epistemological orientation. An example is his work exploring working relationships in a manufacturing company. He elucidates ways in which (in his experience) chains of command could be said to be perpetuated in the texture of relations of a company which he calls the 'Wellblown manufacturing company' (1996: p. 79). He indicates how in the course of his involvement therein he discerned 'violence to the spirit' through 'intimidatory schemes', and he refers to other 'rituals of oppression' (1996: p. 89). However, in his conclusions he acknowledges (to the reader) that he has imposed a particular 'observer derived framework' for shaping his analysis. He also indicates that he is aware that 'literal descriptions and analyses can ... produce ossification' (1996: p. 93). He indicates that it is possible (and desirable) for others in society to reconsider the relevance of his study of ritual for some setting in which they are engaged and at the same time to reshape/re-appropriate it. In this way Golding contributes to a form of knowledge relationship wherein 'the evidence' is treated as a resource for raising matters for contention, rather than an indication of 'something' to be (more or less) correctly appropriated.

7.7

Furthermore, many forms of in-depth interviewing and also focus group interviewing (whether or not they are seen as part of ethnographic inquiry) have been used with the purpose of creating opportunities for inquirers to explore with others new ways of being-in-the-world (see, for example, Smaling, 1992: pp. 172 - 175; Aphane et al., 1994: pp. 12 - 14; Etter-Lewis, 1996: pp. 125 - 127, for some accounts.) The idea here is to utilise these methods as a way of practising inquiry processes which allow for the opening of new paths for experiencing and revisiting our worlds. The research is accounted for in terms of its possible (argued for) contribution to this practice. It is accounted for by considering it in terms of its attempts to be an instance of this practice.

7.8

The above serve as some indication of what it might mean to pursue alternative paths of knowledge to that argued for by Hammersley and Gomm. It must be emphasised that this is not to say that attempts to operate alternative forms of knowledge-relationship in the research process are to be regarded as being dilemma-free. (See, for instance, also in this respect, the work of Gouldner, 1975; Morgan, 1983; Billig et al, 1988; Stanley and Wise, 1993; Brown, 1994; Lather, 1995; Flood and Romm, 1996; Weil, 1996.) However, it is argued here that Hammersley and Gomm's account of 'the path which leads towards knowledge' does little to aid an appreciation of the difficulties facing those who might wish to make some research contribution without unduly impinging their input into the terms of social discussion. The difficulties remain. For an allegiance to Hammersley and Gomm's conception of the pursuit of scientific knowledge through the attempted removal of avoidable 'error' already might be seen as prejudging the way in which reality-construction is approached. This is the danger of Hammersley and Gomm's account of the knowledge seeking process.

7.9

There may be scope for the symbol 'bias' to be used in less exclusivist way than is provided for within Hammersley and Gomm's argument. It is possible that it can be used in a way which does not exclude the relevance of alternative epistemological stances. For example, it could be used to evoke some appreciation of our capacities to extend our own and others' engagements with our lived world(s), as part of the process of developing an orientation of discursive accountability (Romm, 1996a). Bias can then be defined as that which is seen to mitigate against the practice of such an orientation in inquiry processes. What might be involved in this practice is of course a contentious issue. But at least when the symbol of bias is used in this way it does not presuppose that we have to accept some espoused view of the pursuit of scientific knowledge in order to raise discussion about our responsibilities as inquiring persons.

Conclusion

8.1

This article was focused on locating epistemological arguments which I tried to show were not sufficiently accounted for in Hammersley and Gomm's conception of bias. I suggested that their specific view of research communities might operate to exclude (as outside the ambit of relevant argument) the concerns of those who challenge their proposed conception of the practice of research. I suggested that it is possible to create less exclusivist spaces for discourse around the notion of 'bias', including discussions around researcher responsibilities. The creation and vitalisation of this space is part of the process of extending 'researcher accountability'. This accountability in turn implies a recognition that the proper validating communities for assessing processes of social inquiry need not be confined to those subscribing to Hammersley and Gomm's proposed epistemological stance.

References

APHANE, M.-J. et al (1994) Inheritance in Swaziland. Swaziland: Women and Law in Southern Africa Trust (WLSA).

BILLIG, M. et al (1988) Ideological Dilemmas. London: Sage Publications.

BROWN, R. H. (1977) A Poetic for Sociology. Cambridge: Cambridge University Press.

BROWN, R.H. (1989) Social Science as Civic Discourse. Chicago: The University of Chicago Press.

BROWN, R. H. (1994) 'Reconstructing Social Theory after the Postmodern Critique' in H.W. Simons and M. Billig (editors) After Postmodernism: Reconstructing Ideology Critique. London: Sage.

ETTER-LEWIS, G. (1996) 'Telling from Behind her Hand: African American Women and the Process of Documenting Concealed Lives' in R. Josselson (editor) Ethics and Process in the Narrative Study of Lives. London: Sage.

FLOOD, R. L. and ROMM, N. R. A. (1996) Diversity Management: Triple Loop Learning. Chichester: Wiley.

GERGEN, K.J. (1994) 'The Limits of Pure Critique' in H.W. Simons and M. Billig (editors) After Postmodernism: Reconstructing Ideology Critique. London: Sage.

GRIFFITHS, V. (1990) 'Using Drama to Get at Gender' in L. Stanley (editor) Feminist Praxis: Research, Theory and Epistemology in Feminist Sociology. London: Routledge.

GOLDING, D. (1996) 'Management Rituals: Maintaining Simplicity in the Chain of Command' in S. Linstead, R.G. Small and P. Jeffcutt (editors) Understanding Management, London: Sage.

GOULDNER, A.W. (1975) The Dark Side of the Dialectic. Amsterdam: Sociology Institute, University of Amsterdam.

GOULDNER, A.W. (1980) The Two Marxisms: Contradictions and Anomalies in the Development of Theory. London: Macmillan.

HAMMERSLEY, M. and GOMM, R. (1997) 'Bias in Social Research', Sociological Research Online, vol. 2, no. 1, <http://www.socre sonline.org.uk/socresonline/2/1/2.html>.

HUMPHRIES, B. (1997) 'From Critical Thought to Emancipatory Action: Contradictory Research Goals?', Sociological Research Online, vol. 2, no. 1, <http://www.socre sonline.org.uk/socresonline/2/1/3.html>.

LATHER, P. (1991) Getting Smart: Feminist Research and Pedagogy with/in the Postmodern. London: Routledge.

LATHER, P. (1993) 'Fertile Obsession: Validity after Poststructuralism', Sociological Quaterly, vol. 34, no. 4, pp. 673 - 693.

LATHER, P. (1995) 'The Validity of Angels: Interpretive and Textual Stategies in Researching the Lives of Women with HIV/Aids', Qualitative Inquiry, vol. 1, no. 1, pp. 42 - 68.

LINCOLN, Y. S. (1995) 'Emerging Criteria for Quality in Qualitative and Interpretive Research' , Qualitative Inquiry, vol. 1, no. 3, pp. 275 - 289.

McKAY, V. I. and ROMM, N. R. A. (1992) People's Education in Theoretical Perspective. Cape Town: Longman.

MORGAN, G. (1983) 'Exploring Choice: Reframing the Process of Evaluation' in G. Morgan (editor) Beyond Method: Strategies for Social Research. London: Sage Publications.

PLEASANTS, N. (1996) 'A Wittgensteinian Social Theory?: Introducing Reflexivity to Marxism', Philosophy of the Social Sciences, vol. 26, pp. 397 - 416.

ROMM, N. R. A. (1991) The Methodologies of Positivism and Marxism: A Sociological Debate. London: Macmillan.

ROMM, N. R. A. (1994) 'Symbolic Theory' in N. R. A. Romm and M. Sarakinsky (editors) Social Theory. Johannesburg: Heinemann.

ROMM, N. R. A. (1995) 'Knowing as Intervention', Systems Practice, vol. 8, no. 2, pp. 137 - 167.

ROMM, N. R. A. (1996a) 'Inquiry-and-Intervention in Systems Planning: Probing Methodological Rationalities', World Futures, vol. 47, pp. 25 - 36.

ROMM, N. R. A. (1996b) 'A Dialogical Intervention Strategy for Development' in J. K. Coetzee and J. Graaff (editors) Reconstruction, Development and People. Johannesburg: International Thomson Publishing.

SCHRATZ, M., and WALKER, R. (1995) Research as Social Change: New Opportunities for Qualitative Research. London: Routledge.

SMALING, A. (1992) 'Varieties of Methodological Intersubjectivity - The Relationships with Qualitative and Quantitative Research, and with Objectivity', Quantity and Quality, vol. 26, pp. 169 - 180.

SMALING, A. (1995) 'Open-mindedness, Open- heartedness and Dialogical Openness: The Dialectics of Openings and Closures' in I. Maso et al (editors) Openness in Research: The Tension between Self and Other. Assen, The Netherlands: Van Gorcum.

STANLEY, L. and WISE, S. (1983) Breaking Out: Feminist Consciousness and Feminist Research. London: Routledge & Kegan Paul.

STANLEY, L. and WISE, S. (1993) Breaking Out Again: Feminist Ontology and Epistemology. London: Routledge.

WEIL, S. (1996) 'From the Other Side of Silence: New Possibilities for Dialogue in Academic Writing', Changes, vol. 14, no. 3, pp. 223 - 231.

WILLIAMS, A. (1993) 'Diversity and Agreement in Feminist Ethnography', Sociology, vol. 27, no. 4, pp. 575 - 589.

Copyright Sociological Research Online, 1997