Copyright Sociological Research Online, 1997


Hammersley, M. (1997) 'A Reply to Humphries'
Sociological Research Online, vol. 2, no. 4, <>

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

Received: 27/10/97      Accepted: 12/12/97      Published: 22/12/97

A Response to Humphries

In a recent article, Beth Humphries discussed my book The Politics of Social Research, describing the view of research it represents as 'traditional (positivist-influenced)', and contrasting it with the emancipatory approach exemplified by Patti Lather's Getting Smart (Humphries, 1997; Hammersley, 1995; Lather, 1991). In the abstract to her article she declares that the debates between these two positions 'are based on stereotypical views which obscure important characteristics held in common'. The implication seems to be that the arguments on each side are to be examined with a view to identifying their commonalities, strengths and weaknesses. Yet, it is striking that from the beginning there is a sharp difference in the treatment they receive. Humphries immediately disparages my views, commenting that the arguments presented in the book are 'depressingly similar' to those offered in an article 'published some years ago' (¶2.2).[1] Moreover, while she quotes me at one point, she immediately adds a critical comment (¶2.2); and for the rest she relies on brief, superficial and sometimes inaccurate references to my arguments, these being directly linked to her own criticisms. Ironically, her main charges against me are that I 'fall into the trap of stereotyping and caricaturing alternatives' (¶2.3), and that my arguments are 'structured in simplistic, reductionist and binary terms' and amount to a 'discourse of derision' (¶2.5). By contrast, the claims of those in the 'emancipatory camp' (her militaristic terminology not mine) are quoted and presented without critical commentary: there is no consideration in this introductory section of the criticisms that have been made of emancipatory arguments (including those in The Politics of Social Research), and in the rest of the article she simply raises questions arising from the tensions between emancipatory approaches and poststructuralism.

Humphries would perhaps defend her mode of discussion as stemming from her rejection of the principle of objectivity; there is no doubt about which side she is on. But it is worth noting what the function of that ideal was intended to be. In the context of qualitative research, for example, it was designed to minimise the danger of the researcher imposing his or her own views on the people being studied, to reduce the chances of misinterpretation. And the same point extended to treatment of the work of fellow researchers: by trying to be objective, initially suspending one's own evaluations, the scope for understanding others' views was maximised. The consequences of abandoning this approach are amply illustrated in Humphries' article: there is a variety of respects in which she misrepresents my position.

She describes my book as 'a critique of Lather and other explicitly political research' (¶1.1). One of my central themes is certainly criticism of politically oriented research, but explicitness is not a matter for complaint on my part, far from it. Moreover, this theme is not the only one in the book, nor do I provide a critique of Lather's work: a much more sustained discussion would be required to do justice to that.

Humphries is correct that I do not explain why I believe the natural sciences should remain the primary model for social research, though I noted that 'it must not be copied slavishly' and that 'we can learn much from the humanities' (Hammersley, 1995: p. 18). The reason, and it is a sign of the times that it needs spelling out, is that in my view natural science still represents by far the most successful form of enquiry. However, Humphries seeks to 'explain' my commitment to this model by quoting Ramazanoglu's assertion that my work displays an 'uncritical privileging of reason in some sort of established scientific community' (¶2.3). I replied to this criticism when it was originally made, but Humphries does not cite the reply or take account of the arguments presented there (Hammersley, 1994). These days 'uncritical' is apparently used to refer to anyone who appears to believe something that the speaker or writer takes to be untrue. What is at issue, of course, is what is and is not legitimate criticism, and there is room for reasonable (as well as unreasonable) disagreement about this. But if any progress is to be made in resolving these disagreements it is essential to engage with the arguments of those who take a different view, rather than simply dismissing what they say. That is why my book discusses critical, feminist and anti-racist views about social research at such length.

Another of Humphries' criticisms is that there is a contradiction between my admiration for positivist philosophers' emphasis on clarity of expression and my acceptance that feminist critiques have revealed a hidden bias in much past social research. I do not see any tension between these two views unless one assumes that clear writing is transparent, allowing us to see the world through it. That is not a view of language that is defensible in my judgment: accounts can be clear yet false. The virtue of clarity is that it facilitates assessing the validity of what is argued.

Humphries repeats Ramazanoglu's and Williams' criticism of my work that I do not take account of the heterogeneity of feminist research, adding that I ignore some developments in feminist theory that she thinks are important. One of the points I made in response to the original criticism was that my aim had not been to capture the diverse character of feminist research but to address the arguments that some writers had put forward in claiming that there is a distinctive feminist methodology. An article examining what methodological approaches feminist researchers have adopted in practice, or discussing the range of methodological views to be found among them, would have been very different in character. Also, to suggest that I 'excluded' the work of hooks, Lorde, Birke et al is to imply that one could and should include reference to everyone whose writings are potentially relevant to the issues being discussed, a requirement that I doubt could ever be met. What Humphries does not make clear is how she thinks the work of these people counts tellingly against the arguments I put forward. Depressingly, to borrow a word, like most of the earlier critics she neglects to address those arguments in any detail.

At one point Humphries suggests that my view of research is tied to 'a metanarrative of emancipation'. This is not the case. It is true that I believe the pursuit of knowledge relevant to human concerns is of value in itself, as one of many activities that have intrinsic value; and I also hold that knowledge can contribute to the achievement of other kinds of good. However, I do not believe that knowledge has always contributed to the good, or that it will always do so in the future. Indeed, in my book I note that it can sometimes lead to evil (Hammersley, 1995: chapter 7). Furthermore, for me it does not make sense to talk about emancipation in the abstract way that Humphries and others do. One can only be emancipated from particular constraints, not from all constraints; not least because what is a constraint is a matter of evaluation not of simple description. For these reasons, I do not put forward an emancipatory metanarrative, nor am I committed to one.

Humphries also claims more generally that 'traditional' approaches involve Enlightenment assumptions. There is some truth in this, despite the superficial paradox. But much depends on how one specifies those assumptions; given that the term 'Enlightenment' does not refer to a single and coherent body of views, but rather to a complex and indeed contradictory mixture. At one level, as regards belief in the value of knowledge, it is difficult to see how any kind of research could avoid being committed to Enlightenment values. At other levels, though, for example in terms of acceptance of the idea that the true and the good are two sides of the same coin, there is much room for disagreement; and there is no reason why traditional approaches should be committed to this, as I explained in my book. So, while I share some views with many eighteenth century 'Enlightenment' thinkers, including belief in the usefulness of natural science as a model, there are other Enlightenment ideas that I reject, including the assumption that 'the truth will make us free'. And the latter is a crucial ingredient of most emancipatory narratives.[2]

Humphries comments at one point that I 'appear to see no contradiction between [my] criticism [of emancipation as a goal of research]' and my 'apparent approval of the radical and liberatory goals of early positivism' (¶2.4). There would certainly be a contradiction if I agreed with those early positivists who believed that there is a close relationship between 'scientific enlightenment' and 'progressive political action'. However, I did not express approval of this, I simply noted that it is at odds with the current stereotype of positivism. Furthermore, I spent most of Chapter 7 of my book explaining why this 'enlightenment' view is unconvincing. I find it difficult to understand how Humphries could overlook this.

More importantly, this point also shows that she is wrong to claim that the only difference between traditional and emancipatory researchers is that the latter have 'exposed' the 'capitalist and male-centred interest at the root of claims to "neutrality" in the construction of scientific knowledge' (¶3.2). And this purported revelation of the reality behind the neutral appearance of traditional research exemplifies the failings of emancipatory rhetoric. We might ask, for instance, about the grounds on which it is founded. In the book to which Humphries refers I outline the epistemological alternatives available to critical and feminist researchers, and the problems with them. But she does not tell us which of these she thinks is sound and why. It is also worth noting how essentialist and reductionist a claim this is, and what problems there would be in providing evidence for it. It is important to underline that those who adopt the principle of objectivity or value neutrality are committed to trying to be objective, they do not claim that they have been or are objective. It has long been recognised that it is quite possible for an account to be biased even though produced by someone who sought to be objective; just as those who reject objectivity as an ideal may nevertheless sometimes reach true conclusions. This means that even the facts about bias are not easy to establish (Hammersley and Gomm, 1997). However, the assertion that there is bias 'at the root' of claims to neutrality goes well beyond such factual matters. Indeed, it is difficult to see how it can be anything more than an article of faith.

There are also questions to be raised about Humphries' claim that the issue of power has not been foregrounded by traditional researchers (¶3.6). The role of power in the context of research has long been recognised (see, for example, Horowitz (1967) and Beals (1969)). And in Chapter 6 of my book I discussed the issue at length, and did not treat it as a problem 'to be solved through greater reliability of research instruments or through the application of ethical standards' (¶3.6). Indeed, one of my central arguments was the need to sustain research communities committed to the pursuit of knowledge, and that this means defending them from both the internal and the external attacks that are currently taking place. What Humphries means, of course, is that 'traditional' researchers have not taken the same view of the relationship between research and power as she does. But that is not the same as ignoring it. One of the points I made in The Politics of Social Research was that whether research is political depends on what one means by the term. I discussed a number of different interpretations, according to some of which research is necessarily political, while in terms of others it need not (and in my view should not) be. Humphries' discussion is vitiated by the fact that she does not recognise these distinctions. She interprets the traditional view of 'objectivity' as disinterestedness and as involving the claim that the knowledge produced by 'traditional' research is 'innocent' because 'untainted by political agendas' (¶2.1). Here, as elsewhere, she does not address the specific arguments I put forward; resorting, instead, to caricature.

Humphries also claims that I treat 'feminist, anti-racist, critical and emancipatory "truths"' as 'outside the norms of legitimate research' (¶4.6). For me the truth is the truth whoever discovers it: it does not belong to one camp or another. And I do not deny that feminist and anti-racist research can and has produced truths. However, I do deny the legitimacy of research that is immediately directed towards achieving some practical or political goal rather than the production of knowledge. I do this not least on ethical grounds: because such an approach involves trading illicitly on the generally held assumption that research strives to be objective. If funders and lay audiences believed that research were primarily concerned with serving researchers' (or someone else's) political goals it would not survive; and it would not deserve to do so. Of course, some of those committed to emancipatory approaches assume that they can pursue the true and the good simultaneously. This is one of the key points where emancipatory and poststructuralist approaches come into conflict. I note in my book that many critical researchers draw on poststructuralism in an expedient way, as a source of radical epistemological critique of other approaches, neglecting the extent to which that critique undermines their own position. Humphries' discussion of poststructuralism suffers from precisely this weakness. She uses the poststructuralist critique of science to criticise traditional research, but effectively puts on one side the radical implications that it has for emancipatory approaches, appealing to McNay's (1992) argument that 'feminists cannot afford to relinquish either a general theoretical perspective, or an appeal to a metanarrative of justice' (¶4.7, emphasis added). Not being able to afford to do something is different from justifying not doing it. Similarly, she does not explain how she can 'throw off regulating regimes of truth' (¶4.8), given that to make any kind of political proposal or to criticise any other approach is necessarily to participate in some regime of truth. The implication seems to be that there should be no regulatory regime, and this relates to her apparent opposition to all exclusion of 'subjugated knowledges'. Does this mean that she believes we should give voice to, for example, 'holocaust denial', or at least that we should respect 'the differend' in this case, given that this view is currently suppressed morally and sometimes legally (see Lyotard, 1983)? I doubt that, in practice, she would avoid 'excluding' some 'subjugated' views that are believed to be true by their proposers. But the question is: on what grounds would she accept some views as legitimate and reject others (see Lather, 1986; 1993; Hammersley, 1992)?

Again, the implication of the fact that those who would emancipate others are necessarily involved in power relations is not just that 'emancipation cannot be conferred on one group by another' (3.5) but that liberty as a practice frequently involves the exercise of power over others. This constitutes a fundamental challenge to any view which assumes that freedom is always good, and/or that the exercise of power is necessarily oppressive. What it points to, it seems to me, is the need to recognise that there are more and less legitimate forms of both. If 'emancipation' implies complete autonomy from others it is a mirage; or, more probably, a nightmare.[3]

Finally, I would like to emphasise that nowhere did I dismiss emancipatory approaches as 'prejudiced, ignorant and ideological' (¶4.6): I discussed the arguments in favour of these approaches at length, albeit critically. Humphries does not tell us what a discourse of derision is, but it might include presenting other people's views in a superficial and inaccurate way and dismissing them by means of what is no more than counter-assertion. If so, her own article comes close to being an example; and, as I have suggested, this may result from her rejection of the principle of objectivity. Indeed, I am left unclear as to whether Humphries' position allows the possibility of rational argument between traditional and emancipatory researchers. Does she have any means by which to stand aside from the 'stereotypes'? Despite her claim to be able to do so, the approach adopted in her article suggests that the answer may be negative.


1 As I make clear in the acknowledgements, the chapter on feminist methodology in my book is a 'slightly revised' version of the article published earlier, so it is not surprising that they are similar.

2 The origin of this phrase points to the way in which the Enlightenment and its successors, including Marxism, drew on theological predecessors: see Löwith, 1949.

3 There are, of course, various arguments to the effect that such autonomy can be achieved through collective action, in the form of 'positive freedom', but the problems with these arguments are well-known: see Berlin, 1969.


BEALS, R. L. (1969) Politics of Social Research. Chicago: Aldine.

BERLIN, I. (1969) Four Essays on Liberty. Oxford: Oxford University Press.

HAMMERSLEY, M. (1992) What's Wrong with Ethnography? Routledge: London.

HAMMERSLEY, M. (1994) 'On feminist methodology: a response', Sociology, vol. 28, no. 1, pp. 293 - 300.

HAMMERSLEY, M. (1995) The Politics of Social Research. London: Sage.

HAMMERSLEY, M. and GOMM, R. (1997) 'Bias in social research', Sociological Research Online, vol. 2, no. 1, <>.

HOROWITZ, I. L. (editor) (1967) The Rise and Fall of Project Camelot: Studies in the Relationship between Social Science and Practical Politics. Cambridge MA: Massachusetts Institute of Technology Press.

HUMPHRIES, B. (1997) 'From Critical Thought to Emancipatory Action: Contradictory Research Goals?', Sociological Research Online, vol. 2, no. 1, <>.

LATHER, P. (1986) 'Issues of Validity in Openly Ideological Research', Interchange, vol. 17, no. 4, pp. 63 - 84.

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

LATHER, P. (1993) 'Fertile Obsession: Validity after Poststructuralism', Sociological Quarterly, vol. 34, no. 4, pp. 673 - 93.

LOWITH, K. (1949) Meaning in History. Chicago: University of Chicago Press.

LYOTARD, J.-F. (1983) The Differend: Phrases in Dispute (English translation). Manchester: Manchester University Press.

MCNAY, L. (1992) Foucault and Feminism. Oxford: Blackwell.

Copyright Sociological Research Online, 1997