Tuesday, April 10, 2012

The clarity of our concepts (or the lack thereof)

Many important concepts used by archaeologists are nebulous and ill-defined, and this is not good for the discipline. In the new revised edition of his excellent textbook on social science methodology, John Gerring (2012) has a chapter on concepts, and it has lots of go0d advice that would benefit archaeology.

Gerring lists seven "criteria of conceptualization" that can be applied to the concepts we use (Table 5.1, page 117):


  1. Resonance.  How faithful is the concept to extant definitions and established usage?
  2. Domain.  How clear and logical is (1) the language community(ies) and (b) the empirical terrain that a concept embraces?
  3. Consistency.  Is the meaning of a concept consistent throughout a work?
  4. Fecundity.  How many attributes do referents of a concept share? (coherence, depth, richness, etc.)
  5. Differentiation.  How differentiated is a concept from neighboring concepts? What is the contrast-space against which a concept defines itself?
  6. Causal utility. What utility does a concept have within a causal theory and research design?
  7. Operationalization.  How do we know it (the concept) when we see it? Can a concept be measured easily and unproblematically?
 After discussing each of these, he goes on to list several "strategies of concepualization" (Table 5.2, page 131):
  1. Survey of plausible concepts
  2. Classification of attributes
  3. Definition
    1. Minimal
    2. Maximal
    3. Cumulative
 Gerring concludes with the observation:

“Even with the most complex concepts, carefully crafted definitions in the minimal, maximal, or cumulative mold should provide a common scaffolding upon which the work of social science can rest in a reasonable stable and consistent manner.” (Gerring 2012:140).

Gerring uses the concept of "democracy" in political science as an example of how these criteria and strategies work. For archaeology, we have many concepts that could use a good meta-analysis of this sort.  Consider "identity."

The concept of identity has become a real fad in some archaeological circles, with fanatic devotion to the concept. By a fad with fanatic devotion I refer to concepts or terms which get tossed around all the time, typically without definition, often in settings where the concept really has nothing to contribute (think about practice theory or "agency and structure," another fad with fanatical devotion in archaeology). Given that there are many types of identity (race, ethnicity, place, gender, etc.) and radically different theoretical approaches (primordialist vs. instrumentalist, and others), it is not professionally responsible to be tossing the term around without definition and without contextualization. I have read quite a few shockingly bad decontextualized treatments of "identity" in the past (and I often avoid reading anything with the term "identity" if I can help it, so I have only sampled a small portion of the literature). I find the sloppy usage itself less distressing than the fact that journals and book editors allow this stuff to pass.

If there ever was a concept in need of conceptual hygiene, identity in archaeology is it. Now it turns out that there is a very very highly-cited paper from more than a decade ago that does this work of clarification for identity (Brubaker and Cooper 2000). This paper is highly-cited, but not by archaeologists. I did some searching in Google Scholar. Yes, a few archaeologists do cite the work in their bibliographies, but I have yet to find an archaeologist who actually engages with the ideas in Brubaker and Cooper.

These authors identify five distinct meanings of the term "identity" in the social science literature (Brubaker and Cooper 2000: 6-8)

  1. “often opposed to “interest” in an effort to highlight and conceptualize non-instrumental modes of social and political action” (p.6)
  2. As a collective phenomenon, to denote a fundamental sameness among members of a group or category.
  3. As a core aspect of (individual or collective) selfhood. Something allegedly deep, basic, abiding, or foundational.
  4. As processual, interactive development of solidarity or “groupness” that can make collective action possible.
  5. “Understood as the evanescent product of multiple and competing discourses, ‘identity’ is invoked to highlight the unstable, multiple, fluctuating, and fragmented nature of the contemporary ‘self.’ This usage is found especially in the literature influenced by Foucault, post-structuralism, and post-modernism. In somewhat different form, without the post-structuralist trappings, it is also found in certain strands of the literature on ethnicity—notably in ‘situationalist’ or ‘contextualist’ accounts of ethnicity.” (p.8).
 They go on to point out strengths and weaknesses with strong and weak conceptions of identity, and then they propose three distinct concepts to organize the diverse meanings that identity has taken on in the literature:
  1. Identification and categorization.
  2. Self-understanding and social location.
  3. Commonality, connectedness, groupness.
I use this example not because I am interested in the concept of identity for itself. Much about the concept still seems slippery and difficult even after reading Brubaker and Cooper and other related works. I am not opposed to the concept per se; I particularly like Charles Tilly's use of identity in his causal mechanisms concerning inequality and political contention. Rather, this is an example of a concept in archaeology that is in dire need of clarification and healing. Why don't archaeologists cite Brubaker and Cooper? This either reflects ignorance and disciplinary myopia (inexcusable in the age of Google Scholar), or, more sinister, it could reflect deliberate avoidance of rigorous scientific approaches to social phenomena. Gerring's criteria and methods, if applied more widely, would go a long way toward conceptual clarification in our field.

Brubaker, Rogers and Frederick Cooper
2000    Beyond "Identity". Theory and Society 29:1-47.

Gerring, John
2012    Social Science Methodology: A Unified Framework. 2nd ed. Cambridge University Press, New York.

Tilly, Charles
1998    Durable Inequality. University of California Press, Berkeley.

2004    Social Boundary Mechanisms. Philosophy of the Social Sciences 34:211-236.

2005    Identities, Boundaries, and Social Ties. Paradigm Publishers, Boulder, CO.



16 comments:

Anonymous said...

The "problematize x" taste of this post is just as tiresome as any bite of identity in archaeological conceptualization. Context matters. Let's move on. Don't mean to be rude. I do love reading your blog. Just saying . . . Maybe I need a snack.

Michael E. Smith said...

I am baffled at this comment. Did you think I want to "problematize" the concept of identity? I can't stand the concept of "problmatizeation" (and I can't spell it either), and it is certainly not something I want to to do identity. To problematize s concept, I think, is to take something that is generally agreed upon and break it down to show that it is really more complicated than anyone thought. I can't imagine that anyone would consider identity an agreed upon concept. It is a sloppy, multi-meaning concept that needs clarification. I guess my call for conceptual clarification went in the opposite of the intended direction. Forget about identity and read Gerring!

Anonymous said...

I apologize for my previous comment. Really, no ill will here. I will check out Gerring . . .

Anonymous said...

I don't think identity is necessarily sloppy, but it is a fluid concept. It is hard to pin down, but that would seem to be part of the point. I always think of the old ideas by Barth on ethnic boundaries: focus on the boundaries and the boundary making as a process of interaction, not the "stuff" that falls in (i.e., don't come up with a trait list). This is useful but also difficult, especially archaeologically. How do you decide what past people thought was a marker of identity? Can we identify identity even if what we are using as a material signature was not significant for past people? What kinds of identity? I like to look at the example of race and the archaeological study of race in historical archaeology. We can study race archaeologically (not ID "races" but study its permutations and insanity) with historical documents to contextualize what we find. I ask my students, can we do this without written records? How do we know what kind of identity? I typically argue, in the case of race, that it is a moot point as race is a historically recent social construct that did not exist prior to historical documents (well, not this insidious concept of "race" that plagues US society today). But, in the absence of historical documents, aren't we just projecting either ourselves or our fads onto the past? Sorry for the rant.

cm

Michael E. Smith said...

@cm- You seem to be talking as if identity is a thing, a fluid thing, but a thing nevertheless. I prefer Tilly's view of identity as a system of relationships, not as attributes of individuals:

“Once we recognize that relevant identities consist not of individual conditions but of relations across boundaries, that multiplicity—admittedly mystifying for any individualistic theory of action and identity—loses its mystery.” (Tilly 1998:99)

“Identities in general are shared experiences of distinctive social relations and the representations of those social relations. Workers become workers in relation to employers and other workers…” (Tilly 1998:217)

(Charles Tilly, 1998, Durable Inequality, Univ. Cal press)

Anonymous said...

No not a thing, but part of a process of interaction and hence fluid and changing, which seems to be what Tilly is saying in these quotes. Of course it is about attributing but it occurs, I agree, via a system of (perceived) structured relationships. But who is attributing? What attributes? How does one find oneself in such a position? Identities are systemically structural, as you write, qualitatively different, overlapping/nested as well as ideological. Tilly's definitions don't sound too far from most perspectives on identity formation. That said, things can be in such systems of relations and still not be recognized as such by the people doing the identifying. Then there is the problem of how to do this archaeologically. Anyway, thanks for the chance to discuss. cm

I am not Anonymous. I'm Robert! said...

"I don't think identity is necessarily sloppy, but it is a fluid concept. It is hard to pin down, but that would seem to be part of the point."

No offense, but this is an example of the type of flawed thinking that is so prevalent in anthropology. If you are trying to pin down some aspect of the world to understand it, you are going to simplify, project it, and distort it. At a cost we are able to see relationships and patterns. Simplifying assumptions are present in all models and concepts, formal or not.

To say that things (anything!) is complex and difficult to capture in words, or quantities, or whatever is not saying much.

Michael E. Smith said...

@Robert the non-Anonymous - I agree completely. Although if one is taking a humanities-type approach, then complexity and fluidity are positive values (and how about "hybridity," another of these terms). But for a scientific epistemology, you are right on the mark. A suggestion for those post-processualist-leaning archaeologists out there: why don't you take a look at how positivist social scientists talk about identity. Check out the Brubaker/Cooper paper I cite, or the works by Charles Tilly. Or check out people who cite those papers (interesting fact - Brubaker and Cooper is probably cited more times than ALL archaeological works on identity).

dogscratcher said...

Perhaps I am just naive, but it seems to me that one of the implicit goals of "science" is to clarify and de-mystify aspects of empirical reality, whereas it seems as if an implicit goal of all the "posts" is instead to obscure and mystify their subject matter. For this reason, I don't think "consilience" between the two cultures is possible even if principle, a la D.S. Wilson, or Mark Collard and Edward Slingerland.

Michael E. Smith said...

@Dogscratcher- I agree with your observation, which makes sense when considering the polar views, or ideal-type views, of those two approaches. But somehow, in archaeology, post- ideas have become integrated with a scientific, or quasi-scientific approach to archaeology. This allows archaeologists to be postmodernly vague and make claims about contingency and negotiation and social construction, but still manage to publish in mainstream journals and get funding from scientific agencies. I see it as a contamination of archaeological thought more than a take-over by postmodernist ideas, and it is very frustrating.

So, while some more vigorous postmodern archaeologists might relish and vigorously promote the vagueness of, say, identity in archaeology, others would probably think the term ought to have an understandable definition. Yet they don't seem to want to make the effort to achieve that goal. So it is hard to make hard-and-fast contrasts between the approaches. Again, very frustrating to an empirical kind of guy.

dogscratcher said...

"This allows archaeologists to be postmodernly vague and make claims about contingency and negotiation and social construction, but still manage to publish in mainstream journals and get funding from scientific agencies."

Agreed, but I keep thinking at some point the funding agencies (especially) will see the emperor has no clothes, and the free ride will be over. But as you say, at least until then, it is frustrating. In my opinion, "contamination" is a great way to put it, though I prefer the adjective "infected."

Anonymous said...

@ Robert. You seem to be willfully misinterpreting what I mean. Saying identities are fluid and changing does not mean that they cannot be studied empirically. People are lumped into groups and lump themselves into groups constantly and constantly change how and why they do so. To ignore that dynamism is to mischaracterize the phenomena one seeks to study. To suggest that such dynamism is beyond the reach of archaeology is to promote an unfortunate pessimism of the ability of archaeology to address complex processes using, well, empirical data. If anything, my comments point to the intrinsic flaws and problems with archaeologies of identity that fail to take into account this complexity and, thus, are not in disagreement with this discussion's overall sentiment. Rather than just coming up with hyper-simplistic proxies of, say, ethnicity and maybe gender, as you seem to suggest, would it not be better to try to figure out if those identities even existed or mattered or maybe how they shifted and overlapped with other identities as a product of social relationships across time? I imagine one could identify with an ethnic group and with a gender and maybe as elite, etc., etc. That is how identity is fluid, changing, and overlapping. So, I suppose I am not entirely correct when I wrote that the concept of identity is fluid. I mean mainly that identities themselves are and, therefore, the concept should not bury such complexity. And, as those Tilly quotes suggest, the ways in which people identify are relational. If identities are relational they are dynamic and changing. Saying identities change does not say that changing identities cannot be studied. Simplifying one's assumptions to create static identities does not mesh at all with the notion of identity as "shared experiences of distinctive social relations and the representations of those social relations." cm

Michael E. Smith said...

@anonymous (CM) - Sorry, but "relational" in Tilly's scheme does not imply "dynamic and changing." Relational (for Tilly) means that his basic ontological model of society is built on social relations, as opposed to models built on individuals (methodological individualism), and models based on larger structures.

One of Tilly's big contributions, in my view, is showing that "relational" phenomena, such as inequality (or identity, I guess) can be extremely durable and hard to change. Hence his book title, Durable Inequality.

Anonymous said...

I know the word "relational" is getting used and abused by folks with the postmodern condition in archaeology, but I certainly do not intend it to mean that. I imagine you have an identity as a man, a professor, an archaeologist, perhaps American, perhaps more. That is a pretty dynamic assortment of identities that you move around in. Why (which is something I am not getting here) deny it? My observations are not postmodern and are not post-processual. They are rooted in basic materialism, albeit one with a historical bent. If identities are built from social relations then they are relational, regardless of the ugly baggage people attach to this word. I'm not talking semiotics, network STS, and I am not talking relational with the "ity" at the end. Saying that identity is dynamic, often changing, based on social relations, and nested does not mean that it is considered "mysterious" and can't be studied comparatively. It is also not a postmodern or postprocessual perspective. I have only read a bit of Tilly, but I never found his work disagreeable. He wrote very favorably of the work of Eric Wolf, one of the best comparative anthropologists. But he is not the only individual with a useful, materialist-informed, perspective on these matters. One thing that Tilly says too is that the extent to which these things stay durable is due to underlying social relations where some people have somethings to gain and others have less to gain. They are historically durable, not analytically durable. This is an interesting discussion, but I just don't think we really disagree (or at least I don't think I do). cm

Michael E. Smith said...

I don't find anything objectionable in your comments or perspective. I can't say anything about whether we might agree or not about identity - I don't have much of an opinion about the concept, other than the observation that is is used far too often without clear definition and without citing the relevant literature. I don't really know what identity is all about in general terms. In fact I have my doubts about whether it is a high-level concept with any useful content at all, but then I haven't thought much about it and can't offer an informed opinion.

Anonymous said...

Speaking of postmodernism, can somebody explain this to me. How does one propose such a session and how does this advance archaeological theory? I'm all about celebrating doing archaeology. But.......

"The 34th Annual Conference of the Theoretical Archaeology Group is returning to Liverpool for the first time since 1996. TAG 2012 is hosted by the School of Archaeology, Classics and Egyptology at the University of Liverpool and will take place from Monday, 17 December to Wednesday, 19 December. Our theme for TAG 2012 is “live archaeology”, and throughout the conference we shall be having a number of activities celebrating becoming an archaeologist and doing archaeology.

The TAG 2012 Committee now welcomes proposals for sessions, individual papers and posters for the conference. If you are interested in proposing a session, paper or poster, please go to our Submission of proposals page for information about guidelines and deadline dates."