Thursday, August 13, 2015

What is wrong with abstract social theory?

A lot of archaeologists like to use high-level, abstract social theory. I think that such theory is not only a waste of time, but harmful to archaeology. It is harmful because such theory does not explain variation and change in past human societies, and it diverts attention away from the kind of middle-range theory needed to explain past social developments in a causal framework. I am NOT talking about Binford's concept of middle-range theory; I refer instead to the standard social-science meaning as described by Robert Merton). See Smith (2011) for discussion.

Here is a passage from Smith (2011):

“High-level theoretical schemes describe how the social world works on a very abstract, philosophical level, and as a result their utility in the analysis of particular empirical cases is rather limited (Ellen, 2010). In the words of [sociologist C. Wright] Mills, grand theory is “so general that its practitioners cannot logically get down to observation. They never, as grand theorists, get down from the higher generalities to problems in their historical and structural contexts” Mills, 1959:33). In their empirical studies, archaeologists who enjoy high-level theory typically cite such authors in their introductions, and perhaps again in their conclusions, but rarely during the course of their analyses of data"

Abstract social theory, "grand theory" in Mills's terms, is fine for archaeology IF:
  • If one conceives archaeology as more in the humanities than the sciences.
  • If one is and idealist and not a materialist.
  • If one is only interested in particular sites/cultures/regions, but not interested in comparisons with broader spatial and social contexts.
  • If one is not concerned with creating a body of reliable empirical knowledge about the past.
  • If one has no concern for causality and explanation of past social patterns and changes.
  • If one thinks that archaeology constitutes the total relevant scholarly universe, and thus one is not interested in other disciplines.
  • If one doesn’t care whether scholars in other social science disciplines find archaeological data useful or not.
  • If one thinks that archaeological findings have no relevance to understanding, explaining, or solving the problems of the contemporary world.

 I've written a lot about this topic previously. Check out my urban theory paper (Smith 2010), or a paper due out in the SAA Archaeological Record next month (Smith 2015). Or look at some of my prior posts in the blog:

"How would you know if you are wrong?"

"How do archaeologists make arguments?"

"Why do I dislike archaeological theory?"

"Do grad students have to know social theory?"

"Problems with Bourdieu? We can help! Call now"

"Why don't archaeologists talk about causality, explanation, and epistemology?"

"Theory, theory theory. What do we mean by theory?"

Ellen, Roy
    2010    Theories in Anthropology and "Anthropological Theory". Journal of the Royal Anthropological Institute 16:387-404.

Mills, C. Wright
    1959    The Sociological Imagination. Oxford University Press, New York.

Smith, Michael E.
    2011    Empirical Urban Theory for Archaeologists. Journal of Archaeological Method and Theory 18:167-192.

    2015    How can Archaeologists Make Better Arguments. The SAA Archaeological Record (in press).


Anonymous said...

I'm just a lowly grad student, but I feel compelled to say that I agree with some of this, but not all. I don't find that "middle-range theory" is always any better at some of these things. To give an example that's fresh in my memory, when I read in a recently published edited volume on comparative archaeology that, "Resistance is the effort by provincials to reduce imperial control of local affairs and may be expressed in diverse forms," I find myself thinking Joan Robinson's response to her hypothetical Marxist: "Naturally — what else did you think it could be?" Did we really need multiple case studies to get to that? (Yes, sure, it's part of a typology of strategies and it gives us a framework for comparative study, I'm cherry-picking, etc. It seems like a lot of people would defend the use of various "grand theories" in the same way, though. To (mis)quote Chomsky, "what do[es this] explain that wasn't already obvious"?)

Don't get me wrong. I'm not saying that "middle-range theory" doesn't have anything to contribute to archaeology, in the same way that I wouldn't say that about "grand theory." I'm just not sure that it's always contributing something more novel or useful than, say, Bourdieu's analysis of the "mismatch" between the economic systems brought together in the wake of the Algerian War. Know what I mean?

Beyond that, I also don't know if I find the "grand theory"/"middle-range theory" distinction to be all that useful. In part, this is because I'm actually not sure, if we take C. Wright Mills seriously, that he would place all of (since we're already talking about him anyway) Bourdieu's work in the realm of "grand theory," or if all of it really fits his reasons for rejecting grand theory. Sure, some probably does, and of course it's impossible to read everything, but this seems like an excuse to say "Bourdieu=French=grand theory=useless" without actually reading him and move on (though for Derrida, that might be a good idea. . .).

I actually agree with a lot of (most of, even?) what you're saying: theory shouldn't just be window dressing, we should be trying to reliably explain what happened in the past, comparative studies are good, etc. I just can't get behind the "if you read Foucault you're killing archaeology!" side of it.

Michael E. Smith said...

@Ian - There is an epistemological hierarchy that runs from high-level, abstract theory, through middle-level empirical theory, down to data and finally to the empirical world. Virtually all philosophers of science accept this notion, but Hodder and Johnson deny it (see my urban theory paper for citations). "Grand theory" was C. Wright Mills's label for abstract philosophical theory. If one is interested in explaining social phenomena in causal terms, high-level abstract theory is of little if any use, whereas middle-range theory provides the causal mechanisms that explain human society and behavior. This is basic social-science epistemology. If we want to be social scientists (as many of us do), then we should look at social science methods, theories and epistemology. John Gerring's 2012 textbook is the best place for this. Anthropology and archaeology are of little help here.

For me, one of the biggest mysteries of professional archaeology today is: Why don't we discuss these things? Why don't graduate classes in archaeology devote attention to epistemological issues of the place of causality, explanation, comparison in archaeological reasoning and practice? Why don't archaeologists write papers on these things? One positive attribute of the New Archaeology was an interest in scientific epistemology, and a willingness to discuss it and write about it. Too bad they barked up the wrong tree (logical positivism with its "covering law" explanations) and got the nature of explanation all wrong, thereby holding back progress.

I despair of the state of archaeology when both "archaeological scientists" and social archaeologists write that processual archaeology (with its logical positivism) and postprocessual archaeology are the only major approaches in archaeology today (Martinon-Torres, M, and D Killick, n.d., Archaeological theories and archaeological sciences. In Oxford Handbook of Archaeological Theory, edited by A. Gardner, M. Lake and U. Sommer, Oxford University Press, New York; this is posted on; Susan Gillespie 2013, in Azania 48: 301-314).

Sorry, this topic depresses me and sends me off to read sociology or to talk to an economist.

Anonymous said...

@Michael — Apologies if this comes off as a bit hostile; it isn't meant to be, but I think I was hedging too much to be clear before. Yes, I am aware of and understand this "epistemological hierarchy," I've read Mills, etc. It, likewise, matters rather little to me whether or not "Hodder and Johnson deny it." I'm not taking the stance that the distinction is meaningless — and I actually think Mills has rather a good point — but I think I might have been attempting a bit too much deference here before. When I say that I don't find "the 'grand theory'/'middle-range theory' distinction to be all that useful," it should really read "your . . . distinction." When reading Mills, one does not get the impression that he has only a passing familiarity with Talcott Parsons' work (though I have only a passing familiarity with it, so perhaps Mills was bluffing); he appears to have read it, and is offering a critique of it as grand theory. Your use of this hierarchy seems to be primarily geared toward allowing you to do the opposite. You can dismiss Bourdieu (etc.) without ever having read any of his work, because you once had a bad grad student who tried to use him, and Hodder seems to like him OK, so he must have written nothing but grand theory. My problem with this is that it lacks the rigor you're asking other archaeologists to maintain. You insist on empirical evidence to back up archaeological claims, but seem perfectly happy to rely on "grand theorizing" rather than empirical evidence to sort work into an "epistemological hierarchy," and to evaluate "Bourdieu" as an abstract concept, rather than a scholar who wrote a lot of not entirely identical stuff. I'm not suggesting that you need to read Bourdieu, that you would find him useful, etc. I'm just saying that "archaeologists use Bourdieu's work in a way that isn't useful" isn't a good reason to lump everything he wrote into the category "grand theory." Archaeologists have used all kinds of theory in ways that aren't useful. This doesn't strike me as a compelling method of evaluation.

I'm also somewhat taken aback at your dismissal of your own edited volume as being "of little help here," but . . . uh . . . I don't know. I'd say something snarky, but I'd imagine I buried the reference a bit too well.

I'd say archaeologists do write papers on these things, but you're right that it isn't many. I don't know what things are like at ASU, but way back in the mists of time, my first-year grad archaeological theory course certainly covered these issues. I will also say that I don't really read Martinon-Torres and Killick as claiming that these are the only major approaches in archaeology. I haven't read the Gillespie paper, but a quick PDF search doesn't reveal any matches for the term "processual" so I'll have to take your word for it.

Michael E. Smith said...

@Ian - I apologize if my off-the-cuff remarks about theory came off as an attempt at a scholarly analysis. This is a blog, and I use it to shoot from the hip. I do attempt some scholarly analysis of these issues in my urban theory paper, where I rely on citations to the methodological literature from various fields to back up my points. Also, I outline some of the pitfalls of high-level theory in my paper on arguments.

Bourdieu is fine for those who like that sort of thing. I am unaware of useful insights into ancient sociality from Bourdieu, but maybe I have just missed them. The rational-choice sociologists (Boudon, Hedstrom, Elster) have written stronger critiques of Bourdieu than I could ever produce. For philosophically-minded scholars interested in the human condition, Bourdieu may be the cat's meow. But as a materialist interested in explaining past social processes using causal mechanisms, I can't find anything useful in his work.

Anonymous said...

@Michael — To be honest, I had sort of hoped that one of the advantages of the "scholarly internet" was the ability to talk seriously to people without publishing a response in JAMT.

More seriously, though, I'm not trying to hold your work as a blogger to an unreasonably high standard. I have in fact read your published work (in particular your recent papers), and the same tendency is there, if somewhat muted. Yes, in your papers, you do "outline some of the pitfalls of high-level theory." Assume, for a second, that I'm not trying to defend the use of high-level theory. I haven't been, as far as I know, but it seems like I must be coming off that way, so I have to ask. Again, it seems to me that for all of this discussion about the pitfalls of grand theory, we're basically left with "how have archaeologists used that scholar's work?" as the only real metric for evaluating whether something is grand theory. Case in point: Turner is middle-range, Goffman is high-level. Why? Because Turner has been used to generate archaeological middle-range theory in a way you consider successful, while other scholars have criticized archaeological uses of Goffman as too removed from archaeological data. That's fine as a positive evaluation, but I would suggest that to make the negative assessment (i.e. "it's all just airy grand theorizing") we might have to go back and consult that scholar's actual body of work. Have you never seen airy, grand theoretical uses of Turner? (I have.) To go back to Bourdieu (my citations of whom, incidentally, are generally also of the "here is why I'm ignoring Bourdieu" variety), though, should we assume that it is impossible that he wrote anything that isn't grand theory, or should we assume that most archaeologists who cite him are probably only familiar with about 20 pages of his most famous book? I think this is an important way of thinking about the problem, maybe. When you're interested primarily in someone's very French phrase about "structuring structures" and not their boring analyses of economic systems or house forms or whatever else, you're probably not going to get much that is actually applicable to anything out of their work. That isn't to say that everyone's body of work contains something useful, of course. I'd just argue that "Ian Hodder doesn't use this very well" is not the same thing as "this is all nonsensical grand theory."

"Bourdieu is fine for those who like that sort of thing." I'd say "what sort of thing?", but fair enough. This is what's kind of weird to me, though. I would otherwise have guessed that Turner would, likewise, not be your sort of thing. And yet. . .

Also, we might have to agree to disagree on rational-choice theory. . . There's a reason Bourdieu and those people argued so well with one another. (I think I might have just undermined myself. . .)

Michael E. Smith said...

I really dislike discussing theory about theory. For me, theory is instrumental. It helps you figure out things about the past. When a body of theory provides all sorts of insights and ideas about the phenomena I work on (e.g., collective action models of governance, or Sampson's models of neighborhood organization, or Rapoport's models of the built environment), I use it. When a body of theory provides no such insights (such as practice theory and agency theory) then I don't use it.

I spent years banging me head against the wall, trying to figure out why I couldn't see the value in social theory, when many smart and respected colleagues evidently thought it was important and valuable. When I started reading the analytical sociologists, and Charles Tilly, and Robert Sampson, and John Gerring in political science, I finally understood why I never really got the stuff that passes as social theory in archaeology. It is philosophy, it is not theory about the real world and how it works. As Raymond Boudon pointed out, Bourdieu's work provides "pseudo-explanations" and is mostly rhetorical, pedantic, and nebulous (I am citing from Tena-Sanchez here). Tena-Sanchez likens Boudon's view of Bourdieu to GA Cohen's concept of "academic bullshit."

Actually, one epiphany for me was during a discussion with a postmodernist about the "social production of space" (Henri Lefebvre's concept). I realized that there is no such thing as (human related) space that is NOT socially produced. I asked my colleague, "So is space always socially produced? Is there non-socially produced space in human societies?" The answer was negative. Space is always socially produced. OK, then how is that a useful concept? It is always true, by definition. This concept does not help me explain why Aztecs lived in small houses while Teotihuacanos lived in big apartment buildings, or why nobles lived in bigger houses than commoners, or why urban and rural space were remarkably similar in the Aztec period, yet very different in Teotihuacan times. I can draw on other theories that help me explain these and other questions of space and settlement that interest me. But the concept of the social production of space is of no help at all. The reason I get worked up about this stuff is that I can't believe I wasted years of my career trying to get "with it" on social theory, when that was a useless task.

Cohen, G. A.
2002 Deeper into Bullshit. In Contours of Agency: Essays on Themes from Harry Frankfurt, edited by Sarah Buss and Lee Overton, pp. 321-339. MIT Press, Cambridge, MA.

Tena-Sánchez, Jordi
2014 Raymond Boudon: An Analytical Social Theorist. Papers: Revista de Sociologia 99(4): 423-431.

Anonymous said...

Sure, and I can definitely see that. After some flirtation with practice theory as a much younger graduate student, I have come to largely the same conclusion about it. I'm just more reluctant to say, "Well, X scholar cites Y scholar who dismisses Z scholar's entire body of work as crap, so their ethnographic work surely produced absolutely nothing of any use to anyone."

Michael E. Smith said...

Sorry, this is not productive. You keep putting words into my mouth.

Anonymous said...

I apologize if that's the case. That certainly wasn't my intention, but I have to thank you for indulging me, nonetheless.