Discourse is fashionable, discourse is in. Half social sciences and humanities ‘do discourse’, I suspect some physics and engineering might soon be doing it, too. Versions of discourse analysis are spreading like wildfire. For me, I must admit that I am somewhat overwhelmed by how often I hear about discourse. So, today, I want to suggest that not everything is a discourse.
I have thought of writing this post for a long time, I am walking on very thin discourse analytic ice. And so, very hesitantly, I venture into the murky field of discourse analysis of much varied sorts. I want to write about references to discourse which, in my view, are too often made with no evidence to back them up. Needless to say, as in the previous post, I want to make some reservations. I only speak of my impressions, I do not wish to pass my impressions as truths, I do not speak or wish to speak about anyone in particular. The examples I refer to below are genuinely ones that came to my mind, please do not read anything into them. Their source is in my own interests and not anyone’s research.
What has always struck me is that I don’t actually use the word ‘discourse’ all that often (apart from when it is attached to ‘analysis’), much like my fellow linguists-discourse analysts. In contrast, when you go beyond linguistics, I think it all changes. All of a sudden, discourse is everywhere. Time and again, I hear statements that an utterance, expression, a clause or something yet different is part of a discourse or draws on a discourse or simply is a discourse. And I keep wondering – what on earth does it mean?
So, here come my impressions. Let me start with two situations. First is one in which the researcher is talking about a set of data, finds something of interest and claims that he or she found a/the discourse of something. It’s like if someone apologises, so we have the discourse of apology, someone says they’re scared and, lo and behold, we found the discourse of fear, someone says they considered suicide and, hey, presto, you guessed it, we just glimpsed the discourse of suicide.
The other situation is when the researcher is also looking at the data, but this time s/he is saying that the extract or extracts reflect or draw upon ‘the discourse of’ suicide, apology, pain or depression. Or they are anchored in it, or something yet different.
In both situations the researchers offer neither any definition of discourse they use, or, indeed, any evidence for any such discourses they claim to exist. They postulate certain phenomena on the basis of an utterance or two.
But there is more here, I think. The word ‘discourse’ can also become a way in which to claim that ‘we’ have just discovered something important about the society, another crucial underlying social principle. In some cases it becomes a whip with which we can flog those we (really) disagree with by pretending that we have more than an opinion. We offer ‘discourse’, which sounds cool and scientific. It’s a way through which one interview, or perhaps even 10, offers us a way into full-on generalisation of what happens outside the (meagre) corpus we have.
Interestingly, although it’s rarely clear whether what is claimed is just a social practice (like in Critical Discourse Studies which I subscribe to) or the existence of a Foucaultian macro-discourse, I actually think it is the latter which is implied. Yes, with our corpus of interviews we are beginning to write a new History of madness, except it will be on the discourse of pain or suicide, as if there were just one homogeneous way of speaking about them and as if a couple of interviews could possibly offer us any insight….
Let me just briefly summarise what I am saying: just because you spot something, just because someone says something interesting, even if you spot a pattern, it doesn’t really necessarily mean that you have spotted a discourse. It really doesn’t. For example, I recently co-authored a paper on children’ s stories about their fathers who had killed themselves. There were many patterns, but not a discourse.
Now, here I wanted to write something about the newly (I think) labelled FDA, the Foucaultian Discourse Analysis. Then I decided not to, then I decided to write, then not…And then, I thought, oh well, I have been ‘this guy’, the one people point at so many times, one more will not make much of a difference. So, I really would like to see more papers anchored in the FDA which consist of more than, shall I say, general impressions on language (mostly the lexical material) which are packaged as evidence for generalisation about the discourse of X.
Now, I would like to flag up one more situation, admittedly, quite rare. This is a situation in which the researcher claims the absence of a discourse and so they say something like: It turns out there is no discourse of pain (in narratives of illness). And immediately one would like to say: how can you possibly say that? You have, I don’t know, 20 interviews, and on that basis you claim an absence of a discourse? How?! We, the story-tellers, can postulate that certain discourses exist, but, really, we can say absolutely nothing, like zilch and nada combined, about whether a discourse doesn’t exist. I’m not even talking about the impossibility of such claims in inductive disciplines like ours, but for goodness sake, 20, 30, 50 interviews really do not tell us what reality is NOT like. In my view, what we can claim at best is that our informants didn’t say something, which for one reason or another, we expected to find there. But let’s be very clear as to what exactly we mean by “say something”. For just because a person didn’t use the word ‘suicide’ or ‘pain’ doesn’t mean they were not talking about pain or suicide. It really doesn’t.
So here are my impressions about the use of the word ‘discourse’. I think ‘we’ use it way too often and by doing it, we postulate things that might not actually exist. And just to be clear. I am not claiming at all that there aren’t such things as, for example, a discourse of psychiatry, though I doubt very very much that there is only one, the discourse I often hear about. The same applies to other discourses. I’m just appealing for considerably more caution in the way we use the word. Let’s make sure that we know exactly what we mean when we talk about such discourses, and let’s make sure that we actually have the evidence to back our claims up. And so, no, this blog post is not the discourse of discourse.