Theories and Tools

by johnmccreery

The following is something I wrote for a thread on Savage Minds. The topic is the relation of theory to ethnography. In it I recommend that we take seriously the metaphor that likens anthropological theories to tools and see training in anthropology as preparing people with a toolkit filled with a rich diversity of tools, some of which may prove useful at the sites where we do fieldwork.

Once we get past the “love a theory and apply it” trap — the one with the consequences implicit in the maxim that to someone who only has a hammer everything looks like a nail—the virtues of entering the field with a diverse toolkit of theories and methods quickly become apparent. Whatever else ethnography is, it is clearly exploratory research, an attempt to get oriented in a space that at first is largely unknown. Every fieldworker rapidly discovers both unanticipated opportunities and unexpected barriers to doing what their grant application says they are planning to do.

Could I have anticipated that a Daoist healer I met in Taiwan would pull me aside one day and tell me about a vision in which the Jade Emperor had told him that I should become his disciple? No way. Should I have foreseen that, while Victor Turner worked in Africa with a people who live in villages with an average population of a couple of dozen people, I would be working in a Chinese market town with a population of 35,000, with people who keep much of their lives private behind the brick walls of their houses? Probably, but nothing in my training had taught me to think like that.

And it wasn’t just me. I remember a seminar in which Terry Turner told us about going to Brazil intending to do the kind of extended case studies of social dramas that Victor Turner had done in The Drums of Affliction. He quickly discovered that, while the Ndembu might have long memories and be ready to tell you who did what to who going generations back, this wasn’t at all true of the people he found himself studying. What they would rattle on about was myth.

I also remember hearing something that made the opposite point, someone remarking on how the African peoples studied by British anthropologists all had complex social structures but the African anthropologists studied by French anthropologists all had complex cosmologies.

The point of all these anecdotes is a recommendation that we avoid looking for the Theory with a capital T that will be our key to understanding everything and, instead, see theories as tools that direct our attention to some particular aspects of whatever we happen to be studying. Take the toolkit metaphor seriously. No one gets much work done by staring endlessly at the hammer, screwdriver or wrench that catches their eye when they look in the toolbox. The work begins when we recognize that, for this or that particular problem, this or that tool (or, more likely, combination of tools) is what we need to use. Some of us may recognize that for the problem they are working on, none of our tools works very well, and invent a whole new tool. But trying to do that without first becoming familiar with the uses and limitations of the tools already in the toolkit is foolish, indeed.


18 Comments to “Theories and Tools”

  1. Bricolage is still a useful concept, for me. Not because I’m a Lévi-Strauss disciple (I’m more of a neo-structuralist), but because it allows for creative use and integration of diverse tools. If Lévi-Strauss had writtem his book om “wild thinking” after geek culture hit the mainstream, “hacking” might have been the concept he’d have used, instead of «bricolage». As Perl crestor Larry Wall says, “There Is More Than One Way.” No surprise Wall has a background in anthro.
    In the US, I’ve noticed a rather strong “What? You haven’t read so-and-so?” trend, in discussions among ethnographers. And people tend to look for a theoretical framework to put in their research. In Francophone contexts I’ve known, it’s merely assumed that people have read the same thing and it’s fairly accepted to fake knowledge of an author, if you can pull it off. People also tend to take all sorts of liberties with theories, including adopting concepts in passing, often with an adjective. A friend even played with «bourdivin» as a way to talk about a concept from Bourdieu.

  2. Alexandre, welcome to the conversation. Could you say a bit more about what bricolage means to you? I ask because my French is petit peu seulement and the sense I have of bricolage is based on my, probably idiosyncratic, reading of Lévi-Strauss in translation. On the one hand, the idea of working with what comes to hand is very attractive, not least because it fits so well with the haphazardness of my own intellectual career. It also evokes “found art,” suggesting that remix can be an authentic expression of creativity. On the other hand, however, my philosophical disposition inclines me to order and system; my toolkit needs a toolbox nicely divided into spaces for different kinds of tools. From this perspective, bricolage, as I understand it, seems perhaps a bit too haphazard. The license for creative integration is lovely. Does it, however, go too far toward positioning scholarship as an art form dependent on historical accident and the mystery we label individual genius?

  3. Hi Alexandre! I thought a bourdivin was a wine-storage service, but what do I know.

    Speaking of system and bricolage, I have sort of a two-part position. On the one hand, I agree with Durkheim that to understand what goes into theory there’s no substitute for an apprenticeship studying one master theorist in depth. On the other hand, it’s only by studying a variety of systems that one can see how any one approach purchases the understanding it enables by constraining other possible understandings.

    I’m also mindful of a conversation I had in an airport with the translator of Gramsci’s Prison Notebooks, philologist Joe Buttigieg, in which he gently (as always) disparaged what he called the “sausage machine” approach to theory in which bits and pieces of this and that are cut out of context, dumped into a hopper and ground up into a whatsis. Of course sausage can be tasty and nourishing, but it’s true that ‘the essence of Gramsci’, or whoever, is lost by this procedure.

  4. Carl, we should, I suggest, consider separately the two projects conflated in your remark.

    If the project is to understand a theorist, the bits and pieces approach will not work. If, however, the project is to understand some piece of the world in which we are interested, we may need several bits and pieces taken from different theorists to get closer to our goal. The problem of the blind men and the elephant isn’t solved by deciding that the guy touching the tail has the most interesting things to say and obsessing over how the various parts of the rope he perceives fit together.

  5. I’m attracted to your distinction, but no theorist thinks the only purpose of reading her is to understand her and not the world. She thinks she’s got some stuff figured out and would like you to follow her whole thought rather than taking a nibble and grazing on. Durkheim’s point was actually that each theoretical system is a comprehensive worldview, in itself an attempt to see and understand the whole elephant. That is certainly true with Marx, Durkheim, Gramsci, and where I agree with Joe is that Gramsci, say, has much more to offer as a comprehensive theory than he does if you’re just peeling off conceptoids like hegemony and passive revolution.

  6. no theorist thinks the only purpose of reading her is to understand her and not the world.

    Irrelevant and immaterial. Has no bearing on what others do with the theorist’s work. Speaking only for myself, the privileging of prima donnas is not why I read theory.

  7. Since I’m watching World Cup right now another metaphor for all this occurs to me, via G.H. Mead. There are various levels at which a game like soccer might be perceived and theorized. For each position on the field (goalie, defender, midfielder, striker) there is a characteristic set of ways the game looks and thinks. That position relates to the others and shares them to some degree as part of a whole, but is otherwise quite ‘partial’, like the guy with the elephant’s tail.

    Some players play more than one position, or otherwise learn to perceive and think positions other than their own. Often those guys play midfield, where they’re responsible to managing the whole field and creating opportunities for other players. The very best players see the whole field and flow of the game at once, imaginatively entering the perspectives of all of the other players and using this greater theoretical scope to gain advantage of understanding and action. But they are still limited to the perspectives offered by the game of soccer itself.

    Understanding that soccer is just a game, and one among many, is not a productive perspective for its specialists, but it is available for many other people, who are therefore candidates to theorize sport in general and its place in more comprehensive cultural, social, economic, political fields. The best of these theorists will see the whole field of fields, imaginatively entering the perspectives of all of the various games being played and using this greater theoretical scope to gain advantage of understanding and perhaps action.

    These are the guys it’s worth studying in some depth, and not because they’re massaging an elephant’s tail. Opinions differ on who those guys are, of course, but are usually ignorant and worthless because the scope of perspective required to make this judgment is so rare. And of course virtuousos of particular positions in particular sports are also worth studying for different reasons.

  8. All I’ve got is nails. Looking to borrow a hammer, please.


    All your nails are belong to us.

    Take your pick. Neither seems particularly principled, at least in the sense that in the former case one doesn’t care particularly how the nail gets hammered, as long as it gets done (and dissertation completed, or whatever), and in the latter, we see the poverty of stubbornly applying the same tool no matter the details of the actual problem at hand. Both are failures of laziness. (Obviously John is not advocating either of these positions).

    It should go without saying that different representations have different affordances; neither are their perspectiveless views on the world (even the simultaneity of events depends on the perspective of observers!). What seems interesting to me is the metatheoritical problem of connecting different representations or perspectives in a common framework. In formal logic, for example, we have a multitude of different formal logics. An important branch of study in logic is metalogical, in the sense that logicians attempt to understand the relations that stand between different logics, how statements in one logical language can (or can’t) be translated into another, and what will be lost in the translation. It seems to me that is in such a metatheoretical framework that Theory with a capital T has the most chance of thriving.

    Jerry Seligman has done some interesting work on a logic of perspectives; but it is pretty abstract, and very distant from the practical and empirical concerns of most social scientists.

  9. What seems interesting to me is the metatheoritical problem of connecting different representations or perspectives in a common framework.

    Me, too. This issue has been a lifelong obsession with me. I suspect it began by trying to figure out what mom and dad were talking about when they seemed to be saying different things. Later came philosophy, anthropology, advertising, now social network analysis. For my money, Andrew Abbott’s Methods of Discovery is the best overview. Some reviews niggle that it may be too simplistic (no, duh), but the overall framework works well for me.

    A slightly different take on the multiple perspectives problem that Carl’s soccer example raises. In my seminars on advertising, I used to illustrate the meaning of thick description by dividing the class into four sections and showing them a TV commercial. The first section got the easy job, the ordinary consumer’s perspective; all they were to do was to jot down whatever came to mind as they watched the commercial. The second section got the hardest job; they were to be behaviorists and describe all and only what they saw without attributing meaning to it. The third section was to describe the business case for the commercial; they had a briefing form to work from—proposition, audience, desired action, a brief rationale explaining why that proposition communicated to that audience would produce the desired action. The fourth section was assigned the creative perspective; they were to focus on aesthetics, the music, the colors, the casting, the costumes, the voices, the overall tone and manner and how they worked together. In the best examples of the exercise, the commercial was shown twice, each section given a few minutes to figure out what they wanted to say, and, then, after cycling through the reports in the indicated order, we returned to the consumers who were asked what they now thought. Then we turned to discussing the critical implications from a social/political perspective.

  10. Replying to Joel Wright on OAC, where I also posted this provocation, I wrote something relevant to Carl’s concerns,

    However, what you are proposing is going to be an ever-mounting and very prodigious task, giving that the body of literature isn’t going to shrink anytime soon.

    Not the problem you think it is. The clue lies in what you write here.

    Also, where do we draw the line in what we stick in that toolbox? Certainly figures like Malniowski and Levi-Strauss go in there. Does Marx and Durkehim? What about Roland Barthes and Walter Benjamin? Sartre, Proust and Baudelaire?!

    If we wanted to include everybody, that would be impossible. But we don’t. We don’t want to include everybody. We don’t, in fact, want to include anybody, except for grateful acknowledgements of where ideas came from.

    Check out a basic physics textbook. Yes, we see a few names. “Newton” points to classical mechanics, “Boyle” points to basic thermodynamics, “Einstein” points to relativity. But the book isn’t about Newton, who wrote more pages about astrology than he did about gravity or mathematics, or Boyle, who was, says Wikipedia, “also noted for his writings in theology,” or Einstein, who, again from Wikipedia wrote “over 150 non-scientific works….and commentated prolifically on various philosophical and political subjects.” The focus is on the big ideas, of which, when you come down to it, there aren’t that many.

    I may be prejudiced here, having once had a philosophy teacher, Hal Walsh at Michigan State, circa 1962-66, suggest that there were no more than 75 significant ideas in the whole history of Western philosophy. That said, my proposition is that if we focused on the ideas instead of what amounts to a dumbed down intellectual history of the discipline, we might actually get somewhere.

  11. John, do you think a gun is the same tool when you use it to shoot someone, when you use it to hammer in a nail, and when you use it to keep papers from blowing off your desk? If no, then I suggest there’s more meat on the bone of systematic theorizing than you’ve addressed.

    I am a bricoleur, and a huge fan of bricolage. In fact I think that all theories no matter how rigorously systematic are configurations of found materials. So I agree with the point you’re pushing here, as I’ve said upthread, but I think it’s one-sided. The particular configurations of those materials matter for what they have to offer, like the configurations of the chemical bases, of which there are only four (ACGT), make some difference in the lifeform (or monster, or nothing) the resulting DNA produces. And some people do more productive configurations than others. Learning how they do that is not a matter of hero worship but of entry-level theoretical competence.

    It is certainly possible to read the generally very poorly theorized literature of the human studies and find many instances where a theoretical system is being used mechanically like a procrustean bed. Agreed! But the other half of the bad work out there is being done by people who think they can just noodle around with whatever concepts strike their fancies.

  12. Carl, we are, I believe, largely in agreement. That said, your gun example offers more support for the view I’m pushing than yours. There is no need to understand in a more than rudimentary way how a gun works or how to make one to kill someone, hammer a nail, or use the gun as a paperweight. The refinements of design and construction that are the professional gunsmith’s province are simply irrelevant to any of these uses. In a similar way, if your purpose is to understand Gramsci, you have a scholarly obligation to learn all that you can about him and what he has written. If your purpose is to understand how political elites persuade the masses to adopt positions contrary to their own best interests, you may find something in Gramsci that contributes to your understanding. You don’t want to miss something important; but there’s no need to know the corpus in depth. An historian is properly interested in Newton’s astrology and how the thinking it displays relates to the work in physics and math. The physicist is not a historian; she learns basic physics and moves on to more advanced topics that presuppose or challenge what she learned in Physics 101.

    One more real-life example. Georg Simmel is venerated by social network analysts for having noticed the sociological difference between dyads and triads and the importance of overlapping social circles, a.k.a. networks, in shaping social life. But we don’t spend a lot of time pondering the difference between Simmel and Weber. Instead we are more concerned with the mathematical formalization of the concepts he suggested and their implications for empirical social processes like rumors and epidemics. The history has become, as the late, great Joseph Levenson put it, of merely historical interest.

  13. my proposition is that if we focused on the ideas instead of what amounts to a dumbed down intellectual history of the discipline, we might actually get somewhere

    Which presupposes that the ideas in question can be separated from and adequately understood in isolation from their context in intellectual and social history. This is largely true of the ideas of physics 101, which are transhistorical in their content. It is not true of the ideas of social analysis, the content of which is intrinsically historical, and can only adequately be analysed in a historically informed way. This should include a historically informed understanding of the context and production of our own conceptual resources.

    Put another way: the ideas used in social theory and analysis fall within the scope of the discipline’s field of empirical investigation.

  14. Duncan, two thoughts: first, your observation is only partially valid; second, it misses the point.

    First, there are transhistorical ideas relevant to the social sciences. Serendipitously, I am working with some right now. There are rock-hard findings in network analysis that apply to social as well as other types of networks (power grids, protein cascades, the Internet). Did you know, for example, that above a certain scale networks naturally take a form in which there is one giant component with a smattering of smaller components around the edges (a component is a maximal subnetwork in which every node is connected by at least one path to every other).

    Second, much of social science, but especially ethnographic and historical research consists of case studies and, of course, the circumstances of the case must be considered. This is, however, no different from the situation in other forms of case analysis. The biologist investigating a coral reef, the geologist exploring a geological feature, and the physicist interpreting results from the Hadron Collider must all be highly sensitive to the context in which the evidence they point to is collected. There is nothing about the importance of context per se that differentiates the social from the natural sciences. What I am talking about here is what Victor Turner calls the flashes of insight delivered in the field by ideas that may be most illuminating when “ripped from the logical sludge in which we find them embedded.” An idea like unconscious motivation or the importance of ownership of the means of production may be useful to someone who has only sketchy notions of the rest of Freud or Marx.

  15. there are transhistorical ideas relevant to the social sciences

    Well of course – physics still applies, so does chemistry, there are basic features of the human animal that won’t differ across the populations studied in social analysis, etc. I know almost nothing about the social network analysis sub-field: I’m provisionally sceptical that it’s going to be producing much that’s a) genuinely transhistorical, and b) contentful enough to be of much use in concrete social analysis, but I’m open to learning something if you’ve got a link or a reference.

    There is nothing about the importance of context per se that differentiates the social from the natural sciences.

    But the social context is the object of sociological study (along with other social stuff), whereas the social context is not the object of study for physicists, etc. So for instance Weber’s Science as a Vocation is part of his social-scientific oeuvre, whereas Hardy’s A Mathematician’s Apology is not part of his mathematical oeuvre. There’s a difference, even if it is, as always, a muddy one.

    Anyway, I think the Freud and Marx examples illustrate one of the other points in play here quite well, in that both figures have extremely complicated analyses of the phenomena they’re discussing, and even the content of the famous ideas about the unconscious and property relations will be and are hugely distorted and oversimplified unless the appropriator has a sense of the larger theoretical context from which those ideas take their meaning in the figures’ work. So it’s a bad idea to just grab the ideas and run with them without regard for the larger conceptual space from which the ideas are drawn (a conceptual space that may exceed the bounds of the original figures’ texts), because that larger space often in fact contributes to the ideas’ meaning and force. [This is, it goes without saying, a separate point from those made immediately above.]

    That last point could I think be rephrased, along the lines of your original post, in terms of the importance of “becoming familiar with the uses and limitations of the tools already in the toolkit

  16. even the content of the famous ideas about the unconscious and property relations will be and are hugely distorted and oversimplified unless the appropriator has a sense of the larger theoretical context from which those ideas take their meaning in the figures’ work.

    Duncan, I take your point with a grain of salt. “Oversimplified” for what purpose? I’m inclined to say that for deep understanding of Freud or Marx? Of course. For developing an understanding of a particular set of cases? Maybe. Science always works that way. For many purposes, e.g., balls rolling down inclined planes, classical mechanics provides an adequate understanding. To understand apparent aberrations in the orbit of Mercury, you need relativity. The question would be whether there is in Freud or Marx something missed by the simplification that applies to the case in question. If there is, and the critic’s noting this fact leads to a better explanation, no question about it, that something should be taken into account. Even then, however, the notion that you need to know everything to know anything useful seems implausible.

  17. I appreciate Duncan’s point and John’s reply. I agree with both. My point has never been that theoretical systems cannot under any circumstances be disarticulated into standalone tool-concepts, even if there’s a translation-loss in doing so. But I trust someone to do that more mindfully and effectively after they’ve apprenticed and mastered a particular theoretical system, so they have a sense of responsible craft, know how context works and see what’s gained and lost by particular configurations of concepts. In fact, I think this is self-evidently of value and given that no one has disagreed with the point you’re making, John, and I know that in other contexts you appreciate the value of coherent theoretical systems (Daoism, for example, from which yin and yang are often extracted and stood alone with perturbing results), I’m actually at a loss what this conversation is for.

  18. Daoism, Carl, is nothing like a coherent theoretical system. As for the conversation, the reason for any conversation is to see what can be learned or to enjoy a bit of sociable interchange. If this one no longer provides either, let it end.

    For my part, I am feeling a bit overheated and overstuffed, the result of summer weather and a too large Indian dinner and also a bit anxious about leaving Ruth behind with the grandkids while I’m off to Italy on Sunday for Sunbelt XXX and still none too happy with my presentation. Babbling on for twenty minutes won’t be a problem, and I have a few neat tricks to demonstrate, but I don’t have the sense of solid progress that I had achieved before last year’s Sunbelt. The lack of a clear way forward and a sense that declining stamina is making it hard to maintain focus while also multitasking business, family and other interests has me in a bit of a funk.

    What do you think of George Dyson’s take on our intellectual predicament in the age of the Internet, the bit I quoted in “Theories and Canoes”?

Leave a Reply!

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: