In The 'Professional Stranger': the role of the fieldworker in workplace studies for CSCW, eds. Lydia Plowman, Richard Harper & Yvonne Rogers. CSRP 428, School of Cognitive & Computing Sciences, University of Sussex, 1996.
A case is presented for the active intervention of ethnographers in workplace settings they study. The author's reasons for this position are discussed, as is its links to various social theories; and some of the flaws in the position are addressed.
This is a paper about morality, not methodology. On the one hand, this statement is accurate in that my aim is not to engage in the debates about how to conduct fieldwork, or to propose a new kind of fieldwork that takes into account certain issues. On the other, it misrepresents my purpose here, which is not to set up a dualism between what people (social researchers) do and what they feel, but rather to deny the existence of such dualisms. To begin again: this is a paper about integrity, not methodology or morality.
Integrity, says the dictionary, has two meanings: "moral uprightness or honesty" and "wholeness or soundness" (Concise Oxford Dictionary, 1990 edition). My concern is with the latter, and particularly with how one can be a whole person while doing fieldwork.
Ethnographers need to take moral stances. It is not sufficient for them simply to observe situations that use or might use CSCW technology and be value-neutral about their potentially damaging effects on individuals and society. There is an urgent need to consider the emancipatory agenda within CSCW.
This is linked to the attempt in the call for this workshop to treat the ethnographer as a real, situated human being, with gender, race, cultural background and so on. It is as iniquitous to ignore these matters when considering the nature of the researcher as it is to ignore them (as is done in positivist science) when considering the subjects under study. By ignoring these questions in the case of the subjects, we treat them as machines only distinguishable by the categories we are studying, an activity that is morally dubious (in denying them their full personhood) but also leads to inaccurate research, insofar as it is extremely difficult to determine which parts of the subjects' identities are in fact irrelevant and able to be ignored.
This necessity to treat the subjects of study as real people has been well covered by the various critics of the positivist methodologies, and has permeated thinking in the CSCW community to such an extent that it could be thought of as folklore. However, less well addressed is the consideration of the researcher as an active subject (although the work on reflexivity within the sociology of scientific knowledge [19], bears some resemblance to this). Again, we have a moral and a pragmatic argument. The moral one is similar: by regarding the researcher as an uninterested observer, we deny their identity as full human beings, and treat them as if they were interchangeable. The pragmatic argument is slightly different: it is concerned with the nature of interpretation. All individuals will interpret a situation differently, according to their theoretical biases but also according to their individuality. This is true in an experimental setting, but even more so in a workplace study, where often the data that are captured are solely those in the ethnographer's notebook, making even data capture a question of interpretation. Even if recordings of speech and actions are available, they are as much open to interpretation as any text - one researcher's reading of a transcript may be radically different from another's.
I am specifically interested here in the effect of the ethnographer's morality upon their study. There seems to be a prevailing attitude in CSCW research that taking a moral stance is simply not appropriate for ethnographers - that their correct role is to observe the situation but not to intervene in it, rather allowing participants to deal with their own situation based on the rational information passed to them by the ethnographer. For ethnographers to actively intervene in situations, it is suggested, would be manipulative and would deny the participants the right to command their own destiny. (This argument was put to me by John Hughes in personal communication.)
This last argument is a powerful one. The position of power that the ethnographer is placed in makes them vulnerable to having their personal opinions on a situation interpreted as Science. It would indeed be manipulative to allow that impression to go uncorrected. But should ethnographers stand by and observe a situation they hold to be damaging to the individuals involved in it without comment or action, or should they act to try to change that situation?
I suggest that to require an ethnographer to stand by is not only to repress their identity as a moral person, with beliefs about life that are as valid as those of the participants, but may also crucially ignore the power balances involved in the situation. The circumstances when ethnographers may want to intervene tend to be those which are disempowering to the individuals in the situation, and particularly to those lower down the organisational hierarchy. Many things may disempower individuals: other individuals (e.g. managers), the organisational structure and culture, or the technology being used. In the first two cases, the ethnographer, if they are in the situation in the role of outside consultant, may well have considerable power to convince management of the need for change. This is a more positive use of their status mentioned above.
This intervention may seem an impossible prospect, and a matter of persuading people to do something they don't want to. Managers seem most unlikely to want to be changed, after all. This may not in fact be the case: there is a lot of change towards empowerment taking place within management just now, as a result of business process re-engineering and total quality management. While these are in many cases more to do with hype than actual empowerment, that the words are used so widely suggests that they may have more possibility for acceptance than might at first seem apparent to those used to decades of machine-like organisations following the cues of Taylor and Ford.
Finally, there is a mighty need to positively act to emancipate individuals because of the strong disempowering drives found in many methodologies for experimentation or for software engineering.
In the case of experimentation, the problem is that discussed above: the whole point is to isolate particular features of subjects that are interesting to researchers while ignoring them as full people, leading to "an almost Frankenstinian preparation, which consists of a brain attached to two eyes, two ears and two index fingers ... only to be found inside small gloomy cubicles" [4, p.13]. This critique has been well taken on board by CSCW researchers insofar as they have substantially given up on experiments, but whether they yet consider subjects/participants/members as full people is open to debate.
Software engineering methods, particularly those going under the name of structured analysis and/or design methods [18], are perhaps more of a problem. Most were originally developed in the 1960s and 1970s, for use in the military or in the data processing departments of massive, hierarchically structured, corporations. We know that all methods and theories, in whatever area, carry with them a background of assumptions and paradigms [3], so this history cannot be ignored. Even in the enlightened and empowered modern corporation - if we believe the consultants' rhetoric that such a thing truly exists - the assumptions behind these methods are going to lead to treating those who end up using them as cogs in the machine. So this emancipatory approach is needed to act as a counterweight.
So taking a moral stance is needed for ethnographers - sitting by and watching is not enough. I would urge all ethnographers to come down from the fence (or out the closet?), recognise their personal moralities, and take moral stances towards the emancipation of those they study.
Discussing these matters with many different people, and thinking them over myself, several issues have arisen for me.
The first issue is that the paper contains a certain sense of smugness, of moral superiority in it. It's full of 'ought' and 'should' (implicitly at least). It sounds rather as if I thought I'd invented CSCW-with-morality (or even information systems development with morality) single-handed. This is no doubt partly a side-effect of my writing style. I would emphasise here that I fully recognise that moral approaches aren't confined to any one place or method.
To mention some of the related writers on this would then be appropriate. The most clearly related academic area is that part of social science which derives from critical theory, and goes under the name of critical management, critical systems, or critical social science. Critical theory is an approach, deriving from the work of Marx and Freud, to the emancipation of society from the harmful effects of modernism and capitalism [12] by examining its prevailing ideologies. It was developed originally by a group in Frankfurt from the 1920s onwards, notably Max Horkheimer, Theodor Adorno and Herbert Marcuse [13, 14, 15]. More recently, it has been further extended by Jürgen Habermas, who wrote of the different power balances served by different kinds of research, and the importance of supporting the emancipatory interest; and also, useful here, the harmful effects of technology [10, 11].
Habermas' work in particular has been the philosophical base for many of the authors who have sought to apply critical theory to organisations. A good example is [1], which uses particularly the work of Habermas but also that of Michel Foucault on power [8] (although Foucault could not be called a critical theorist, his emancipatory intent was similar). An approach to systems theory that uses critical theory, based primarily on Habermas, can be found in [7], which derives substantially from the journal Systems Practice.
Other relevant work is the approach in education research known as critical ethnography, which seeks to use ethnographic methods to bring about social transformation [2]. A mention must also be made here of the work on participatory design of computer systems, carried out over twenty years, which has part of its aim the restructuring of power balances in organisations [6, 9, 16]. And while I haven't seen how it might be brought together, there are strong links in my comments above on integrity to issues in feminist attitudes to research (see, for example, Eevi Beck's paper in this collection); and in the liberation theology practiced by priests in Central America.
So the approach has in fact plenty of antecedents in academia. My argument is not actually based on those, however, but on an emotional statement: I hold strongly to a set of beliefs about the worth of the individual, and their right to fulfilment: my basic core belief is the nurturing and emancipation of the unique human spirit. This is supported by my personal background in liberal Christianity and especially Quakerism. This was my point about integrity with which I started: to believe in these things and yet not to practice it within organisations would be, for me, hypocrisy.
Other people, of course, will not necessarily have this moral/political stance. Dave Randall made this point with regard to Lucy Suchman's work:
one might draw attention to a certain naivety in what is essentially a 'liberationist' politics by asking why other politically derived theories, for example Marxism, cannot be invoked. After all, in the drive for predictability in organisational life it is not as if managers and owners of corporation manage those corporations with indifference towards the profit motive [17, p.47].
I have rather assumed in my Assertion that everyone is like me and generally committed to emancipation, and I thus urge ethnographers to "recognise their personal moralities", assuming that those moralities are ones which require one to be active as an emancipator. This is completely incorrect of course - ethnographers presumably have the same range of beliefs, moralities and politics as other folks. This puts me in a position rather like those American free speech advocates who found themselves defending fascism: I am arguing for morality, and yet may end up encouraging those whose position I would vehemently disagree with. It must never be forgotten that Frederick Taylor genuinely believed he was doing good for those he scientifically managed.
An unresolved issue in this approach, which I mention here not to give a solution but because so many people have commented on it, is whether it wouldn't be rather more empowering to those studied to let them change their situation, rather than let others do it for them. Put like that, this is rather persuasive. It has been put to me by well-meaning people, but has a hint to my mind of the laissez-faire politics of 19th century liberalism: and like those attitudes, crucially ignores that people may be unable to change their situation themselves alone, and that others may be required to assist them, or to educate them so that they are able to change it themselves. If one is in a situation where that change is necessary and one has the power to assist, I feel it much better to do something than to do nothing. The old slogan has it, "Give a man a fish and he'll eat for a day; teach a man to fish and he'll eat for a lifetime" - maybe the answer is to teach people how to change things.
A final point of concern: how to convince anyone to let you do this kind of work! Managers aren't typically very keen on emancipation ('empowerment' is a different matter, but see Andrew Clement's excellent article on that point [5]). I have suggested above that a way in is through the empowerment talk within re-engineering and the like. This may be overly optimistic, given that even access for ethnographers is difficult. Are companies really likely to want to listen?
To end on a more polemic note again. In some ways, this whole paper doesn't go nearly far enough for me. I get very angry at situations of disempowerment and injustice within organisations, and I feel it my duty to try and change them. This is an essentially emotional position rather than a rational one. I can't tell others that they should be doing likewise, but I do get bothered by people who have the awareness of situations of injustice, and the power to do something about them, but don't act against them. This is the stance I have tried to work against in this paper.
This paper in its developing forms has benefited greatly from discussions with and comments from John Hughes, Jon O'Brien and Fides Matzdorf, the organisers and participants of the Professional Stranger workshop, and two anonymous reviewers.
1. Alvesson, M. and Willmott, H. (eds). Critical management studies. London: Sage, 1992.
2. Anderson, G. "Critical ethnography in education: origins, current status and new directions". Review of Education Research 59(3):249-70, 1989.
3. Burrell, G. and Morgan, G. Sociological Paradigms and Organisational Analysis. London: Heinemann, 1979.
4. Claxton, G. "Cognitive Psychology: a suitable case for what sort of treatment?", in Claxton, G. (ed.) Cognitive Psychology: New directions. London: Routledge and Kegan Paul, 1980.
5. Clement, A. "Computing at Work: Empowering Action by 'Low-Level' Users". Communications of the ACM, 37(1): 52-63, 1994.
6. Ehn, P. Work-oriented design of computer artifacts: 2nd edition. Stockholm: Arbetslivscentrum, 1989.
7. Flood, R.L. and Jackson, M.C. (eds). Critical systems thinking: directed readings. London: John Wiley, 1991.
8. Foucault, M. Discipline and punish: the birth of the prison. London: Allen Lane, 1977.
9. Greenbaum, J. and Kyng, M. (eds). Design at work: cooperative design of computer systems. Hillsdale, NJ, USA: Lawrence Erlbaum, 1991.
10. Habermas, J. Knowledge and human interests. Boston: Beacon Press, 1971.
11. Habermas, J. Toward a rational society. London: Heinemann, 1971.
12. Held, D. Introduction to critical theory: Horkheimer to Habermas. Hutchinson, 1980.
13. Horkheimer, M. Critical theory: selected essays. New York: Herder and Herder, 1972.
14. Horkheimer, M. and Adorno, T. Dialectic of enlightenment. London: Allen Lane, 1973. Originally published c.1945.
15. Marcuse, H. One dimensional man: studies in the ideology of advanced industrial society. London: Routledge & Kegan Paul, 1964.
16. Mumford, E. and Weir, M. Computer systems in work design - the ETHICS method: effective technical and human implementation of computer systems. Manchester: Associated Business Press, 1979.
17. Randall, D. A Comment on Lucy Suchman's "Do Categories Have Politics". CSCW: An International Journal, 3(1):47-50, 1995.
18. Sommerville, I. Software Engineering (5th edition). Wokingham, UK: Addison-Wesley. 1996.
19. Woolgar, S. (ed). Knowledge and reflexivity: new frontiers in the sociology of knowledge. London: Sage, 1988.
Magnus Ramage 6 August 1996