1.

“Local” clearly is a “relative” term. In the Solar System, Earth is local (as has been brought home, in good anthropological manner, by leaving it at least temporarily to look back at it from the Moon and other orbits); in the Galaxy, the Solar System is local (Voyager should help with that); and in the Universe, the Galaxy is local (a while to wait, perhaps, for this). To a high energy physicist, the particle world—or zoo—is, well, the world. It's the particle, a thread of vapor in a cloud of droplets, that's local.

2.

Thus the opposition, if we must have one (and I am not persuaded an opposition—another opposition—is what we need or ought to want, rather than a shifting focus of particularity), is not one between “local” knowledge and “universal,” but between one sort of local knowledge (say, neurology) and another (say, ethnogra/ phy). As all politics, however consequential, is local, so, however ambitious, is all understanding. No one knows everything, because there is no everything to know.

3.

The failure to see this shining truth by people otherwise ap/ parently rational is the result, in part, of an endemic confusion in the social (or human) sciences (or scientists) among: (a) universals (“everybody has,” to quote a false, or at least a highly misleading example, “the incest taboo”); (b) generalizations, which may be prob/ abilistic, have exceptions or contradictions without fatality, or may be mere ceteris paribus, “as a rule” approximations that are instru/ mentally useful (“Horticultural societies are more peaceful than pas/ toral ones”—but consider the Maya, regard the Lapps); and (c) laws. (It is hard to produce an example—“group marriage to ma/ triliny to patriliny”—in cultural anthropology, or indeed anywhere in the human sciences, that is not laughable or outmoded. Perhaps a proposal from a few years back that cultural traits diffuse—that is, migrate across the globe—on an average of plus or minus two miles a year conveys some of the comic effect involved.)

4.

My own view, merely to give it, because in a short compass I can hardly defend it, is that either: (a) most (conceivably all) uni/ versals are so general as to be without intellectual force or interest, are large banalities lacking either circumstantiality or surprise, pre/ cision or revelation, and thus are of precious little use (“People ev/ erywhere have views about differences between the sexes”; “All so/ cieties have systems of social ranking”; “Powerlessness tends to corrupt, absolute powerlessness tends to corrupt absolutely”—this last, which is of course my transformation, exemplifies another characteristic of many universals: like reversible raincoats, they can be worn either way); or (b) if universals do have some degree of non/ triviality, circumstantiality, and originality, if they actually assert something interesting enough to be wrong (the ubiquity of the Oedipus complex, the functional necessity for psyches and societies of mourning customs, the solidarity-making force of the gift), they are ill-based. We are ethnographically acquainted with only a very small proportion of the societies that have existed; of those, only an even smaller proportion have been systematically studied, and those that have been systematically studied have not been studied evenly or comprehensively. We may or may not know something about Oedipus notions in the Trobriands or Sri Lanka; I don't know that anyone has even thought to look into the matter for the Havasupai—or, if perchance (I haven't checked) someone has, then the Montenegrins, the Incas, or the Kabyles. There is a tremendous unevenness, as well as a tremendous instability, in anthropological attention. Nothing gets studied everywhere or for very long. Until not long ago, there wasn't much of use on the kinship system of the Navajo, though kinship is one of our most obsessively investigated subjects and the Navajo are one of our most thoroughly researched groups.

This is not remediable—not by setting up Notes-and-Queries/ type schedules, standardized research training programs, or what/ ever. Nor should we, in my view, try thus to remedy it. The search for universals leads away from what in fact has proved genuinely productive, at least in ethnography (I don't think only in ethnogra/ phy, but I will let others argue the other cases)—that is, particular “intellectual” obsessions (Malinowski's about exchange, Lévi/ Strauss's about animal symbolism, Evans-Pritchard's about divina/ tion)—toward a thin, implausible, and largely uninstructive com/ prehensiveness. If you want a good rule-of-thumb generalization from anthropology, I would suggest the following: Any sentence that begins, “All societies have …” is either baseless or banal.

5.

Generalities of the “not in the South” sort can, of course, be had and can, of course, be useful; but more as heuristic starting points for deeper-going local inquiries than as bankable conclusions fit for textbooks. (“Funeral rites are a good thing to look into if you are interested in a people's conceptions of the self.” “In Southeast Asia, status differentiation tends to be unusually important, gender contrast rather less so; in North Africa, contrariwise,” “Child raising practices have a lot to do with adult personality.”) Most of the more valuable of these are conceptual generalities of a proof-of-the-pud/ ding sort: if they really get you somewhere, fine; if not, the hell with them. Linguistic ones, back in favor of late, in part as a result of the Chomskyan revolution (or, as I think, counterrevolution, but let that pass for the prejudice it is), tend to be like this: noun/verb distinctions, markedness regularities, etc. They do indeed seem to have broad applications, though claiming they are universally appli/ cable is either dogmatical, tautological, or a regress to the vacuities I discussed just above. As surface signs of more deeply-lying matters, however, they are the shale (it is hoped) above the oil field.

All this is not the same as saying that the search for broad generalities is the obvious or best way to go, though there is admittedly something about anthropology—its up-from-the-ape, study-of-man sweep, perhaps—that seems to encourage it. To put the matter another way, even the generalizations of so-called cogni/ tive anthropology—the ethnobotany work, Berlin and Kay's color work (often misread, perhaps even by its authors, in “universals” terms)—are surely to some degree cosmopolitan in nature, though how cosmopolitan is not always clear. Reading them into the world in a “realist” manner, as part of the very furniture of things, is an/ other matter, one I also cannot get into here, save to say that I think it a dubious proposition. “Species” are “real,” so far as they are, precisely in the way that (so far as it is) “power” is.

6.

As for laws, I have already suggested that I can't think of any serious candidates in my field with which to contend. One of the most irritating things in my field is people who say you're not doing “real science” if you don't come up with laws, thereby suggesting that they themselves have done so, without actually telling you what these laws are. On the rare occasions they do tell you—two miles a year, cannibalism and protein shortage—the situation is worse. Scientism, and here I will talk of the human sciences overall, is mostly just bluff. It's one thing to call the spirits from the misty deep, quite another to make them come when you do call them. But it is not just imposture that's involved: the utopianism induced by a misconceived view of pre-twentieth-century physics (the world before Maxwell) that was imported into the human sciences has led not to the gates of paradigm-land, but to a great deal of wasted motion and high proclamation.

7.

So much for negatives. What are the virtues of a “local knowledge” sort of tack?

a.

Limits. The title of this discussion seems to assume that the existence of limits is a counterargument to something. (Why isn't it called “‘Universal Knowledge’ and Its Limits”? Possibly because to do so would raise the possibility that, being universal, it hasn't got any, and therefore isn't knowledge.) To my limited mind, direct and open acknowledgment of limits—this observer, in this time, at that place—is one of the things that most recommends this whole style of doing research. Recognition of the fact that we are all what Re/ nato Rosaldo has called “positioned (or situated) observers” is one of its most attractive, most empowering features. The renunciation of the authority that comes from “views from nowhere” (“I've seen reality and it's real”) is not a loss, it's a gain, and the stance of “well, I, a middle-class, mid-twentieth-century American, more or less standard, male, went out to this place, talked to some people I could get to talk to me, and think things are sort of rather this way with them there” is not a retreat, it's an advance. It's unthrilling perhaps, but it has (something in short supply in the human sciences) a certain candor. (Views from nowhere can be imaginatively con/ structed, of course. If they are done well they can be, and in the natural sciences have been, immensely useful. But thus constructed, they are in fact a particular variety of view from somewhere—the philosopher's study, the theorist's computer.)

b.

Circumstantiality. We can at least say something (not of course that we always do) with some concreteness to it. I have never been able to understand why such comments as “your conclu/ sions, such as they are, only cover two million people [Bali], or fifteen million [Morocco], or sixty-five million [Java], and only over some years or centuries,” are supposed to be criticisms. Of course, one can be wrong, and probably, as often as not, one is. But “just” or “merely” trying to figure out Japan, China, Zaire, or the Central Eskimo (or better, some aspect of their life along some chunk of their world line) is not chopped liver, even if it looks less impressive than explanations, theories, or whatnot which have as their object “History,” “Society,” “Man,” “Woman,” or some other grand and elusive upper-case entity.

c.

Of course, comparison is both possible and necessary, and it is what I and others of my persuasion spend most of our time doing: seeing particular things against the background of other particular things, deepening thus the particularity of both. Because one has located, one hopes, some actual differences, one has something gen/ uine to compare. Whatever similarities one might find, even if they take the form of contrasts … or incomparabilities … are also gen/ uine, rather than abstract categories superimposed on passive “data,” delivered to the mind by “God,” “reality,” or “nature.” (Otherwise, Santayana's comment that people compare when they can't get to the root of the matter becomes all too true.) Theory, which is also both possible and necessary, grows out of particular circumstances and, however abstract, is validated by its power to order them in their full particularity, not by stripping that particularity away. God may not be in the details, but “the world”—“everything that is the case”—surely is.

8. 

But the critical issue involved in the local versus universal tension in the “human sciences” (and I have already indicated my discomfort with that way of putting things—“versus” should be left to prize fights, elections, wars, and the law courts) is: What do we want from those “sciences”? What does, or should, “science” mean here? There is not much point in arguing about whether to involve ourselves with matters inextricable from “this time or that place,” or to look past such matters to ask how everything, everywhere, always is, unless we are clear about what we expect to gain by taking one tack or the other. The dispute, which seems to be about the worth of different paths to an agreed destination, is really about the worth of alternative destinations, however arrived at. We are divided less by method—one uses what avails—than by what we are up to.

The contrast here is familiar, but not less important for that: between those who believe that the task of the human sciences (though they are more likely to call them “behavioral”) is to dis/ cover facts, set them into propositional structures, deduce laws, pre/ dict outcomes, and rationally manage social life, and those who be/ lieve that the aim of those sciences (though sometimes they will not agree to call them “sciences”) is to clarify what on earth is going on among various people at various times and draw some conclusions about constraints, causes, hopes, and possibilities—the practicalities of life.

Whether the first view is, as some people have said, a bit like wanting to know where you will die so as never to go near the place, or the second, as others have said, is like blowing out the candle and cursing the darkness, is perhaps less important (though it is hardly unimportant) than which enterprise we in fact see our/ selves as pursuing. If advances in the technical, fine-tuning control of social life (Bentham's dream, Foucault's nightmare) is what you are after, then universality talk is, I guess, the talk to talk. If you are after refinements in our ability to live lives that make some sense to us and of which we can on balance approve (Montaigne's skeptical hope, Weber's desperate one)—moral skills, not manipulative ones— then something less vaulting would seem to be called for.

Those of us who take the second view (a growing number, I think, now that the ideologies of disincarnate knowing have been somewhat shaken) have, of course, much to make clear and even more to make persuasive. But we are working on it and need not be unduly worried, except perhaps politically, about measuring up to standards emerging from the first one, fishing for other fish, perhaps inedible, in other seas, perhaps unpopulated. What Stephen Toul/ min has called “the recovery of practical philosophy” has its own agenda and its own ideas about how to advance it. What he lists as “the return to the oral” (he really means to rhetoric, utterances, speech acts, discourse, narrative, conversation, and language games—not the literally oral, but the linguistically so), to “the par/ ticular,” to “the local,” and to “the timely,” is a movement not a doctrine and, like any movement, needs achievement not dicta to sustain it. What we need (to give a dictum) are not contemporary reenactments of old debates nomothetic and ideographic, erklären and verstehen, but demonstrations, on the one side or the other, of either an effective technology for controlling the overall directions of modern social life or the development and inculcation of more delicate skills for navigating our way through it, whatever directions it takes. And when it comes to that, I am reasonably confident both as to which is the more desirable and the more likely actually to occur.

Who knows the river better (to adopt an image I saw in a re/ view of some books on Heidegger the other day), the hydrologist or the swimmer? Put that way, it clearly depends on what you mean by “knows,” and, as I have already said, what it is you hope to accom/ plish. Put as which sort of knowledge we most need, want, and might to some degree conceivably get, in the human sciences any/ way, the local variety—the sort the swimmer has, or, swimming, might develop—can at the very least hold its own against the gen/ eral variety—the sort the hydrologist has, or claims method will one day soon provide. It is not, again, a matter of the shape of our thought, but of its vocation.

I don't know if this is an adequate “response to the critical claims of universality and authority” against work which emerges from “historical point(s) in time or … geographical point(s) in space” (as the charge to this discussion puts it), or even what would count as “adequate” here. But, like all “local knowledge,” it is sub/ stantive, somebody's, and will do for the moment.