Abstract
I welcome Mike Lynch’s response to my article and thank him for it. It is, perhaps, necessary to reiterate that the article was not primarily about ethnomethodology, or even about ethnomethodological work in Science and Technology Studies (STS), but about a particular line of argument – what can crudely be called radical constructionism – which has long been part of STS and continues to be influential in some quarters there, as Lynch acknowledges. My discussion of ethnomethodology in the article pointed to a parallel between this line of argument and some ethnomethodological sources; this probably stemming from the influence of the latter on the former. I have written about ethnomethodology itself at length elsewhere, in publications referenced in the article (for example Hammersley 2019).
Lynch claims that I ‘lump […] ethnomethodology together with latter-day constructionism in the anti-representationalist camp.’ I tried to make clear that I was referring to the work of some ethnomethodologists, rather than to ethnomethodology as a whole. He questions my interpretation of a quotation from Coopmans et al. (2014, 2) but what he presents as an alternative is his own view of representation, whereas my focus was on that of one of the other editors: Steve Woolgar. Later in the article I quote from Woolgar and also from the work of John Law. Part of the second half of Lynch’s paper is a critique of Law’s work, but, while I agree with much of what he says there, I’m not sure why any of it is relevant as a response to the points I made. I presume that Lynch is aiming to show that Law’s position is very different from that of ethnomethodology. But I never claimed otherwise. To suggest there is a parallel between radical constructionism and ethnomethodology is not to propose identity or even overlap (though I would be inclined to argue for the latter in the case of Pollner 1987).
A key question raised by Lynch’s response is whether he is correct that there is a third way between the radical constructionist argument I challenged and the rather conventional position I defend. There are, of course, diverse views in the field of epistemology, but Lynch appears to distance himself from philosophy just as much as from radical constructionism in STS: he refers to ethnomethodology as offering a ‘deflationary approach to the topics of philosophical inquiry’ (Lynch 2022, 1 emphasis in original). This involves a re-specification of the questions addressed by philosophers, so as to examine empirically how practical actors deal with epistemic issues on specific occasions. But here something like the equivocation I mentioned in my article reappears: Is he simply suggesting that there are other questions of interest besides those addressed by philosophers? Or is he also proposing that the philosophical questions are spurious, perhaps along the lines indicated by Wittgenstein? If the former is his proposal, it leaves everything as it is, in philosophy and social science. If the second question captures his position, this is itself, of course, a general epistemological, and normative, argument. There is no escape, it seems to me.
Lynch (2022, 8) declares ‘To say that “science produces representations referring to objects or processes that exist independently of it” (Hammersley 2022, 132) is to use the term “science” evaluatively and indiscriminately.’ I do regard the term ‘science’ as evaluative, and believe that there are demarcation problems that must be addressed, not simply described; just as Lynch (2016) thinks there is a demarcation problem about what counts as ethnomethodology that needs to be addressed. Aside from this, to me, the quoted phrase states what is obvious, in other words beyond reasonable doubt (at least until further notice). It does not say, of course, that all the representations produced by science are true, only that they refer to some objects and processes that exist. For example, we can question the accuracy of DNA profiling without denying that DNA, chromosomes, and cells exist, and certainly without denying that human bodies containing flesh and fluids exist; as Lynch recognises in his excellent study of DNA profiling (Lynch et al. 2008). The statement of mine he quotes could only be rejected on the basis of a broad, if not global, epistemological scepticism. This makes me wonder, once again, whether there is any ground on which Lynch’s middle way could rely.
Even if we assume that epistemic issues can be avoided, or resolved unproblematically, by ethnomethodologists, their work focusing instead on documenting how such issues are handled by others in the course of practical action, a further problem arises that I mentioned towards the end of my article: about the value of some of this work. Lynch (2022, 5) writes at one point about what is ‘a worthy topic of investigation’, but he does not tell us how he is assessing worth. Once scepticism has been abandoned, the fact that phenomena are socially constituted, and even how they are constituted, may lose much of its interest. With Sacks, Lynch (2022, 6) argues that there is ‘order at all points’ in social life, which indicates that practical actors don’t need the assistance of ethnomethodologists in solving the problem of social order. Long ago, Garfinkel (1967, 4) acknowledged that what ethnomethodology studies is ‘uninteresting’ to lay people. This would also be true of many other scientific fields, so there has to be some rationale for why investigations in these fields are worthwhile that goes beyond the simple fact that they can be done. In discussing the Greiffenhagen, Mair and Sharrock's (2011, 2015; Mair, Greiffenhagen and Sharrock 2016) study of social science practices, I suggested that, since these authors do not express scepticism about the findings those practices produce, their work could only have value if linked to the concerns of research methodology. I think this problem of value emerges for much, but not all, ethnomethodology once the whiff of radical constructionism has been blown away.