Ethnography on an Awkward Scale

…talking to the natives is evidently a dangerous experiment. Violetta Lee [1890] 1998:407

More than thirty years ago we met a madman in Mafeking, now the hyphenated capital of the North West Province in the “new” South Africa.1 Or, to be more precise, we met a prophet in polythene robes who had been incarcerated in a mental asylum by the apartheid state. We spoke of him in a scholarly essay (Comaroff and Comaroff 1987): outside of his extravagantly colored costume, what had marked his presence on the local scene before his “admission” to hospital was a fondness for standing, hour af- ter hour, as a silent witness near the local railway depot. It was from here that genera- tions of black men were transported nightly to the cities of Makgoweng, the Place of Whites, to toil in its mines and factories. It was from here, too, that the capillaries of ra- cial capitalism, South Africa-style, became visible to anybody who cared to gaze upon the twilight movement of migrating males across a cloven landscape. Anybody troubled enough. Or mad enough.

Three decades on, after the demise of the ancien régime, we passed the very spot, in Station Road, where the mute madman used to linger. He had died, anonymou- sly, some years before. It was early afternoon on a Saturday, a sparkling winter day in July. As we crossed the street on our way to the local police station we noticed a small knot of men-in-blue nearby. They had surrounded a decidedly strange figure: an adult male, nude except for a pair of threadbare boxers, covered in white paste. Emaciated, his eyes showed no animation whatsoever. With a measure of gentleness not usually associated with the law here, he was taken to the Mafikeng Community Service Center – police stations are now “community service centers,” just as the old South African Po- lice Force has become the South African Police Services – where he was fed and al- lowed to wander around unhindered. Which he did, every now and again climbing on a chair or a desk, every now and again curling up in fetal repose. All the while, like the madman of earlier vintage, he uttered not a word. We asked the officers on desk duty who, or what, the figure was.
“A zombie,” we were told.
“What is to become of him,” we asked.
“We hope that his people, maybe a maternal uncle (malome), will come for him,” said one officer.
“How did he come to be wandering in Station Road?”
“Who knows? Perhaps his owner lost him or let him go by mistake.”

As we have noted (Comaroff and Comaroff 1999c; cf. Ralushai et al 1996:5), there are a fair number of living-dead about these days. Termed dithotsela or dipoko (sing. sepo- ko; from the Afrikaans spook, “ghost”), they are thought to be the creatures of witches who by nefarious means, have sucked away their human essence and turned them in- to brute labor power; this to make them toil away at night in the fields. Indeed, the (then) acting Vice Chancellor of the local University of the North West, himself a scholarof the white Afrikaans occult, casually promised to introduce us to one about whom he had long known.

He did not have to. We encountered many more in the course of our own resear- ch.2 Some of them appeared in circumstances much less benign, much more violent and troubled, than those that brought the frail phantom to the attention of the Mafikeng police. One such circumstance ended in the murder of a well-known personage in the province, “Ten-Ten” Motlhabane Makolomakwa. Sometime middle-level state emplo- yee, owner of a local football team, successful farmer, and the chairman of the tribal council of Matlonyane village, he was set alight by five youths who insisted that he had killed their fathers and turned them into spectral field hands.3 Another, in 1995, involved striking workers at a coffee plantation in nearby Mpumalanga Province: they refused to work for three supervisors whom they accused of killing employees and turning them into zombies for their private enrichment.4 A third case – immortalized in a play, Ipizom- bi, well-known in local cultural circles and beyond – was sparked by a taxi accident in Kokstad in which twelve schoolboys were killed. Much discussed all over South Africa at the time, it involved the murder of two elderly “witches” who were said to have stolen the corpses and conjured them into living-dead.5

Cases like this are often reported in matter-of-fact terms by the national media (cf. Fordred 1998) and – along with Hollywood horror movies, local telenovelas of witch killings, and other iterations of spectral death-and-dread – are widely consumed. Signifi- cantly, they are sometimes invoked, either before the fact or in the act, by those who perpetrate occult-related violence in the South African countryside. On occasion they have also become the stuff of cybertalk, not least among southern Africans abroad, whose anxious internet exchanges, intermittently filtered through EuroAmerican urban legends, have flowed back onto local soil, there to be fabricated into new kinds of fact. Thus it is that reality and its representations become confounded in one another, at once both cause and effect, each inseparably a part of the phenomenology of everyday life in the postcolony. Thus do imported and domestic spirits infuse each other; all being signs of both the local and the translocal, here and elsewhere, now and then, the con- crete and the virtual. Thus it is that the national population of living-dead is thought, in some parts of South Africa, to have been joined by transnational zombies, entering the country from Mozambique and other places (see n.2), just as they did in earlier times (Harries 1994). Thus it is, too, that home-grown phantoms bear more than passing, if culturally inflected, resemblance to images originating in Haitian Voodoo, to the celluloid freaks that haunt such films as Night of the Living Dead (George Romero 1968) or The Serpent and the Rainbow (Wes Craven 1988), and to ghouls that rise to the rhythms of various popular musics.

These specters, in turn, evoke others: most obviously, a trade in human beings and body parts at once local and transnational, real and imagined, legitimate and illicit, more or less coerced. It is a trade, as we all now know, that stretches from the import and export of sex-workers, domestic workers, and mail order brides (these often being hard to tell apart); through the sale and adoption of babies (also difficult to distinguish, the latter often being an ethicized, affectively-acceptable euphemism for the former); to transactions in blood, genes, eyes, hearts, kidneys, and the like, transactions in which the medical may run into the magical (Ehrenreich and Hochschild 2003; Scheper- Hughes and Wacquant 2002). Some of this trade, when it entails fully sentient persons, evokes the horrors of slavery; where less than whole persons are involved, it extends the logic of commodity exchange to ever more divisible components of homo sapiens. Almost everywhere it is regarded, by those whose populations are being harvested, as a new form of Empire erected under the increasingly contested sign of global free trade and its highly inequitable flows of wealth; a curious footnote, this, to Hardt and Negri (2000).6 Elsewhere, we argue that these phenomena are all interrelated features of an “occult economy,” itself spawned by a brand of neoliberal capitalism that attributes to the free market an ineluctably salvific, redemptive, even messianic quality. By “occult economy” (Comaroff and Comaroff 1999a) we intend a set of practices involving the (again, real or imagined) resort to magical means for material ends; or, more expansi- vely, the conjuring of wealth by inherently mysterious techniques. Of course, what counts as “magic” varies, although it is always set apart from habitual, more transparent forms of production. This arcane economy has other, well-known manifestations: among them, an alleged rise in many parts of the world of witchcraft and satanism (J. Comaroff 1997; Geschiere 1977; La Fontaine 1997), of “fee for service” faiths (Weller 2000; Comaroff and Comaroff 2002; cf Kramer 1999), of enchanted financial practices that, like pyramid schemes and lotteries, promise fabulous wealth without work.

All this enchantment, tellingly, is making itself felt at just the moment when the global triumph of modernity was supposed to put an end, once and for all, to such puta- tively premodern things. The iron cage, so feared by Max Weber, turns out to have been a cage of ironies. To be sure, if ever there was a figure that typified the magical production of wealth without work, of the occult grounding of neoliberal capitalism tout court, it is the zombie: all surplus value, no costly, irrational, troublesome human needs. This kaleidoscopic figure, the ultimate embodiment of flexible, “non-standard,” asocial labor, comes to us in a range of ethnographic, historical, and literary accounts that point both to subtle differences and to noncoincidental similarities. Zombies appear, simultaneously, as antemodern and postmodern, simultaneously supralocal, translocal, and lo- cal, simultaneously planetary and, refracted through the shards of vernacular cultural practices, profoundly parochial. Which is why the living-dead now regularly cross inter- national borders; why, say, a South African doctor of Indian origins could claim to have been turned into a ghostly automaton by a Nigerian satanist.7 And why zombification, the stuff of much urban legend across the world right now, has become an allegorical touchstone for describing the ostensible alienation, loss of individuality, and corporate mastery of an epoch, as yet in its infancy, already being described as Post-Human (Halberstam1995; Fukuyama 2002) As it did, albeit in somewhat different guise, with the rise of Fordism and the mode of human abstraction (dis)embodied in its production lines8 – and, before that, on the plantations and in the mines of far-flung colonies.

Our concerns here, we stress, are not theoretical or conceptual.9 We came ac- ross zombies, recall, through an empirical conjuncture: it was by force of historical fact, rather than by way of abstract analytical interest, that we found ourselves compelled to make sense of them in situ. Consequently, what detains us here is much more immediate, much more modest, much more, well, methodological. By what ethnographic means does one capture the commodification of human beings in part or in whole, the occult economy of which it is part, the material and moral conditions that animate such an economy, the new religious and social movements it spawns, the modes of produc- ing wealth which it privileges, and so on? Inherently awkward of scale, none of these phenomena are easily captured by the ethnographer’s lens. Should each of them none- the less be interrogated purely in their own particularity, their own locality? Or should we try to recognize where, in the particularity of the local, lurk social forces of larger scale, forces whose sociology demands attention if we are to make sense of the worlds we study without parochializing and, worse yet, exoticizing them. Geertz (1973), for whom ethnography defined the generic practice of anthropology, once remarked famously that we do not study villages, that we study in villages. The point was well-taken. But how – given that the objects of our gaze commonly elude, embrace, attenuate, transcend, transform, consume, and construct the local – do we arrive at a praxis for an age that seems post-anthropological? Of an age in which we are called upon not to study inplaces at all, indeed not to trust “anthropological locations” (Gupta and Ferguson 1997), but rather to study the production of place (Appadurai 1996)? If we are not sure where or what “the field” is, or how to circumscribe the things in which we interest ourselves, wherein lie the ways and means by which we are to make the knowledges with which we vex ourselves?

Of course, the question of Method, in the upper case, is not new. It has been with us throughout the life of the discipline, if in different forms and formulations. Nor, right now, are we alone in this. Postcolonial historians, for example, seem to be angui- shing a lot these days about the death of history. Not The End of History as proclaimed a decade ago, somewhat infamously, by Francis Fukuyama (1992), but an altogether new kind of death: death by diffusion into memory, biography, testimony, heritage tour- ism, and other expressions of history-as-lived rather than history-as-learned (Minkley, Rassool and Witz n.d.; cf. J. Comaroff n.d.). In times past we anthropologists plagued ourselves over the epistemic, ethical, and political dimensions of what we do: over whe- ther ours was not an endemically colonizing enterprise – a preemptive seizure of autho- rity, of voice, of the right to represent and, incidentally, to profit – or, worse yet, an acti- vity founded, voyeuristically, on the violation of “the” other. Now, like those postcolonial historians, we worry whether our subject matter is ours at all or whether it has forever dispersed itself beyond our privileged dominion. Once we were told that we would be out of business just as soon as our natives were no longer authentically native (a.k.a. primitive, colonized). Today we are undermined by the fact that those very “natives” have seized the terms of our trade, terms in which we once described them, terms that seem not to work very well any more as analytic constructs, terms that, now essentiali- zed and commodified by “others” one and all, return to haunt us. Add to this two other considerations, themselves intimately connected: first, the aforementioned fact that al- most everything which falls within the discursive purview of contemporary anthropology exists, in the phenomenal world, on a scale that does not yield easily to received anth- ropological theories or methods; and second, that our “subjects” no longer inhabit social contexts for which we have a persuasive lexicon, not least because abstract nouns like society, community, culture, and class have all been called into question in this ever more neoliberal age (cf. Stoller 1997:82), this age of the scare quote-around-everything, this age of ironic, iconic detachment. What, in the upshot, are we left with? A very stark question: Has ethnography become an impossibility? Have we finally reached its end?

ETHNOGRAPHY AND ITS GLOBAL DESTRACTIONS

…what actually happened, the facts of the case, who said what,..all that is incidental. The real truth is behind all that. The real truth may be swimming in a completely different direction…And that’s what you have to get to…Forget the appearances.

Neil McCarthy, The Great Ourdoors10

Not surprisingly, in light of this Big Question, there has been a fair bit of debate, over the past few years, about the fate of ethnography in the age of globalization. We have addressed the matter ourselves, most pointedly in our Max Gluckman Memorial Lecture of 1998 (Comaroff and Comaroff 1999a). Its title, “Occult Economies and the Violence of Abstraction,” sought to invoke the stark dislocations wrought on the every- day lives of ordinary citizens in the northerly provinces of South Africa by material for- ces of ever more planetary scale, dislocations about which many of them spoke with both anxiety and passion. We also meant to underscore the challenge involved in gras- ping, ethnographically, the processes by which those world-historical forces were being made meaningful and tractable by the human beings in question: how they labored to condense and personalize values and relations in conditions which they presumed to be labile, difficult to understand, inherently mysterious in their effects. Among our objec- tives, in sum, was an effort to reflect on the interplay of theory and method in the treat- ment of an anthropological location of changing proportions. Although of pressing con- cern at the moment, this is a problem as old as the discipline itself. Our essay, after all, was written to commemorate a scholar who tried long ago to subject the broad sweep of the colonial encounter to the ethnographic gaze.

In the post-Maxist age, the strongest suit of anthropology, in the eyes of most of its practitioners, remains its “ability to get inside and understand small-scale communi- ties, to comprehend local loyalties and systems of knowledge” (Graeber 2002:1222). Our disciplinary concerns may alter, our genres may blur, our theories may come and go, but ethnography remains “the anthropologist’s muse” (Lewis 1973), the source of solace to which we turn in the face of epistemic or political doubt. An extended spell of “participant-observation” is still the irreducible minimum of professional credentials re- quired in the discipline, Sherry Ortner (1997:61) notes. This in spite of the ambiguity that attaches to each of the two terms, not to mention the oxymoron built into their hy- phenation. This in spite, also, of the fact – illustrated by Ortner’s own account of study- ing the “postcommunity” – that contemporary anthropological practice deviates, as it probably always has done, from the foundational fiction of fieldwork: the conceit, now long criticized (cf. Gupta and Ferguson 1997), that it is possible to access “the totality of relations” of a “society,” or the essential workings of “a culture,” in any one place.11

And yet the axiom that lies behind this fiction, that any knowledge derived at first- hand by proximity to natives has an a priori privilege, continues to shape the analytical vision of the discipline. “Ethnography,” says George Marcus (1994:44), “functions well and creatively without a sense that it needs a positive theoretical paradigm – that is, conventional social theory – to guide it. Instead, it breeds off the critique of its own rhe- toric.” As a result, anthropology has, for the most part, remained unrelentingly positivist in spirit. Much of its shared wisdom consists in generalizations about the particular that are also particularizations of the general; empirical aggregates, in short, not abstract propositions or explanatory schemata. The role of this species of knowledge, like its po- litics (Graeber 2002), has been to show that, even in the act of accommodating to ine- luctable macrocosmic forces, different peoples do things differently, be it because of their distinctive cultures, their social situations, or their will to resist (cf. Marcus 1994). The epistemic consequences that follow are plain enough: a committed realism, and a form of relativism that sits uneasily with “general” theory grounded in history, philoso- phy, political economy, or whatever. True, there have always been counter tendencies: those who have espoused evolutionary, Marxist, sociobiological, or psychoanalytic ap- proaches, for instance, have been more partial to higher-order abstraction, generaliza- tion, explanation. But this minority has tended to be the exception that proves the rule.

The epistemic foundations of anthropology’s empiricism received somewhat less scrutiny than they might have done during the “reflexive moment” of the 1980’s. But in practice, ethnography was already undergoing a metamorphosis. The discipline was co- ming face to face with the consequences of what had begun to make itself felt in the 1960s: that “local” systems – or, to be more precise, the signs and practices observable within any given social world, however it was constituted – could no longer be studied, or accounted for, with reference to conventional geographies; that the fiction of sover- eign cultures, however deftly described or ethnographically authenticated, could no lon- ger be sustained; that established modes of representation were no longer sufficient unto the political and ethical demands of “writing culture” (Clifford and Marcus 1986). Yet, in the absence of “an explicit paradigm for experimentation” (Marcus 1994:46), the methodological revolution one might have expected to flow from these shifts of pers- pective – themselves sharpened with every passing year by the complex, uneven eff- ects of processes of planetary integration – has not been forthcoming. Per contra, not- withstanding some creative efforts to author new kinds of anthropology, the reaction in many quarters, in Europe as well as in North America, has been conservative. There has been a tendency to batten down the hatches in fervid defense of the particular, the local, and the parochial against the onslaught of “the global” (e.g. Sahlins 1999; Kapfer- er 2000, 2001; Englund 1996; Rutherford 1999), the latter, in anthropology-talk, having become a generalized, under-motivated sign of the changing universe in which we live and work.

Why? One consequence of globalization for the human sciences, argues Appa- durai (1997:115), has been to instill an anxiety that the “space of intimacy in social life” will be lost; the very space of intimacy that has always been the ethnographer’s stock- in-trade. Whether or not this is a sufficient explanation for the anthropological angst of the present moment, it certainly is the case that our latest “crisis of representation” has been transposed into a methodological key – as if the survival of the discipline depended entirely on preserving its established modes of producing knowledge. Note how, in some quarters, ethnography is being depicted as an endangered species. Englund and Leach (2000:238), for instance, appear to believe that “it” is engaged in a mortal struggle with “generalizing perspectives” whose powerful, if unnamed, proponents have allegedly decreed that “localized fieldwork has had its day.” For Englund and Leach, the enemy is the “meta-narrative of modernity,” a somewhat ill-defined construct which, de- spite their protestations to the contrary, seems suspiciously like a synonym for “Theory” in the upper case.12 And for an ensemble of “familiar sociological abstractions,” among them commodification, space-time compression, individualization, disenchantment. This “metropolitan” meta-narrative, they argue, “undermine(s)…what is unique in the ethnographic method – its reflexivity, which gives subjects authority in determining the context of their beliefs and practices” (Englund and Leach 2000:225). The apprehensiveness about the future of fieldwork palpable here seems to stem, above all, from a crisis of identity, from sacred boundaries breached, and, concomitantly, from the desire to pre- serve a unique scholarly patrimony from the encroachment of an ever more generic so- cial science. It cannot have gone unnoticed, in this regard, that other disciplines have lately laid claim to ethnographic methods. Thus Englund and Leach (2000:238) insist that “[t]he uniqueness of the ethnographic method is at stake in the current fascination with multiple modernities…Sociocultural anthropology merges into cultural studies and cultural sociology, and ethnographic analyses become illustrations consumed by metropolitan theorists.” How unlike an earlier, brilliantly iconoclastic Leach (1961), who encouraged anthropologists to move, by “inspired guesswork,” beyond hide-bound empiri-

cism. Thereareseriouspoliticalissuesatstakeinargumentslikethis.Intheeffortto privilege “the local,” however worthy it may be, we risk slighting or misrecognizing the global forces that – increasingly, if with varying degrees of visibility – are besetting “little guys” (Graeber 2002:1223) all over map. Many of those among whom we work, appa- rently unlike Englund and Leach’s “natives,”13 are very anxious about the effect of those forces, which, they tell us, are putting their social and material survival at risk. In the faux egalitarianism of these neoliberal times, it is easy to become mired in trivial argu- ments over whether “meta-narratives of modernity,” or “Theory,” removes from “others” the capacity to represent themselves or to determine their own futures. All this while the masters of the market, and powerful political pragmatists, fashion new modes of extrac- tion, abstraction, and explanation. We would do well to ponder, in this respect, why it is that so many “native” intellectuals have been distrustful of even the most sensitive, os- tensibly other-centered knowledge produced by our discipline, why they believe that this knowledge is intrinsically inimical to their own authority and interests (cf. Banaji 1970; Magubane 1971, Asad 1973). Mafeje (1998:67; cf. Sharp 1998), for one, holds that eth- nography, to be true to itself, needs to be liberated entirely from anthropology, thus to become – without even the most reflexive of ethnographers – a source “of social texts authored [solely] by the people themselves.” The logical end point of reducing our prac- tice to the elicitation of narratives of local experience is not a unique anthropology at all. Nor is it a politics of positive engagement. Quite the opposite. In a postcolonial age in which “natives” everywhere speak for themselves, it is, simply, redundancy. The alter- native, patently, is to argue for a theoretically and politically principled social science.

For our own part, we continue to have confidence in ethnography and the forms of insight – both reflective and reflexive, both imaginative and empirical – to which it gives access. There is a proviso, however: that, instead of fetishizing method, instead of romancing the idea that it might itself yield up naked truths, we face up to the episte- mic challenge of what it takes to “commit social science” in the postcolonial world: in a world in which “globalization” is an increasingly contested, troubling reality, in which “modernity” is an increasingly contested, troubling ideological formation (Knauft 1997). Those anthropologists who have chosen to take on this challenge have tended not to decry “localized” ethnography, but to insist on its unique value in plumbing the nature and effects of large-scale social, economic, and political processes (e.g., Appadurai 1997, 1996; Geschiere 1997; Meyer 1999; Weiss 1996). Their work points to the fact that our modes of producing knowledge demand critical review – even “redesign” (Marcus 1994:46) – in the face of history; especially the history of a time such as this, when popular discourses across the planet posit that the world is undergoing changes of ma- jor proportions. This perception, after all, does not exist only in the imagination of anth- ropologists afflicted with “the meta-narrative of modernity.” What is more, we need to concede that our craft is not, and never has been, analytically self-sufficient. Part of a shifting division of labor within the human sciences, it is engaged in dialogue with other ways of making sense of the present in both its macro- and microcosmic dimensions (Stoller 1997; Sharp 1998). This is all to the good, since it is only by broadening our frames of reference that we may address some of the awkward questions that have come to confront us about our methodology: can we be sure, for example, that “the par- ticular” we seek to study, or the cultural worlds we presume to exist, may actually be empirically bounded? Is “the local” not the constantly refashioned product of forces well beyond itself (Appadurai 1996; also 1997)? Does it not exist only as part of a sociopoli- tical geography of multiple scales and coordinates (Ortner 1997)? Is it not true that the singularity of places, just like the singularity of “traditions,” “customs,” and “cultures,” is being fashioned ever more in response to the market? Surely, neat antinomies between the local and the global, between field and context, between ethnography and metanar- rative, beg the very questions that we should be asking.

These questions have also been at the core of a friendly exchange we have had with Sally Falk Moore (1999) over the susceptibility of large-scale analytic claims to eth- nographic proof. Her critique of our Gluckman Lecture hinges on a methodological poi- nt: the unverifiability of its central thesis, namely, that the rapid expansion of an occult economy in postcolonial South Africa has been a by-product of the material and experiential impact on rural populations of the cumulative effects of a globalizing capitalism – specifically, of the processes of abstraction and alienation built into it. The “imaginative sociology” by means of which we arrive at this thesis may be illuminating, concedes Moore. But it does not offer sufficient evidence either to substantiate or to falsify a claim of cause-and-effect. Moreover, by ascribing the growth of a local occult economy to world-historical forces, we “turn general context into particular explanation” (l999:306). We also confuse the general and the particular. How so? At times, she suggests, we deny that resort to the magical, and to its associated forms of violence, is unique to South Africa; at other times we imply that there is something special about its deployment here.

Allow us to recall what the disagreement is about. Our objectives in the Gluck- man lecture were twofold. One was to make sense of some highly visible, much discus- sed, old-yet-new practices in postcolonial South Africa. Taken together, these practices, themselves rooted in variously defined “localities,” appeared to constitute a discernible phenomenon: an occult economy. As we have implied, this term describes an empirical- ly-grounded abstraction, an abstraction derived from, but not reducible to, the narrated experience and social activities of a large number of diversely positioned human be- ings. In short, it is an analytic concept based in the concrete. Located between the glo- bal and the local, subsuming them in a four-dimensional geography,14 that concept is mobilized to arrive at “thick,” moving portraits of peoples’ lives and labors; also to eluci- date the motivation, the meaning, and the consequences of their actions. It is a tool that enables the dialectic of deduction and induction on which, in our view, all principled social science ought to be founded.

The other objective of our analysis of “occult economies” was to explain why that enchanted economy should manifest itself so palpably now, when conventional wisdom would have us expect otherwise; why it calls forth received cultural practices, yet trans- mutes them into virulently altered forms; why, while clearly a domestic product, it bears close resemblance to similar economies in other places, most of all in post-totalitarian contexts, where neoliberal reform has suddenly and simultaneously liberated and dis- empowered, enriched and impoverished. These parallels are striking and yet hard to pin down. They bear witness to the play of large forces (i) that, although volatile and only partly visible, are not random; (ii) whose existence may be inferred only through their ef- fects; (iii) whose workings vary across the axes of the planetary map, making them im- possible to grasp at only one site; (iv) which, because they have not yet fully run their course, elude proof by ordinary means. The problem that we set ourselves, then, was to account for the workings of a metamorphosing capitalism that is both global in its reach and localized in its protean manifestations. Built into that problem is an effort to engage at once with the general and the particular, with variance and similarity, with continuity and rupture. Far from being a confusion yielded by our method, it is a necessary re- quirement thereof. Respectful of the empirical without being empiricist, we seek to open up new angles on a world-historical process of awkward, shifting scale.

At issue here, then, are alternative ideologies-of-method, alternative epistemes. The differences that flow from them, not least over what it takes to prove an argument or to verify a theory, are substantial. Which is why we stand accused, in this exchange, of not having provided enough evidentiary support for claims about some very general transformations in South African economy and society; even more, about their location in the broad sweep of the history of capital. Even if we agreed that we ought to render as “provable propositions” our analysis of these transformations – or of the ways in which they are locally inhabited, experienced, narrated, acted upon – we find it hard to see how to do so without resorting to reduction ad absurdum. But we do not believe that this is what we should be doing; indeed, we resist the positivist reflex that would encour- age us to do so. After all, if they were held to the demands of empiricist validation, or subjected to the blinding lights of Western science, some of the most enduring insights of modernist social thought would not pass muster. We have in mind, inter alia, Marx’s analysis of the commodity, Weber’s elective affinity between Protestantism and the rise of capitalism, and Durkheim’s theory of the elementary forms of religious life.

Marx, Weber, and Durkheim, of course, all argued against both ungrounded historical conjecture and theory deduced purely from philosophical first principles; although each of them indulged in these things on occasion. More to the point here, each sought to take the measure of the difficult relation between the experience of social phenome- na and the forces and facts, the rhymes and reasons, that lay behind it. Each exercised a fertile sociological imagination, seeking, in the Great Outdoors of their changing wor- lds, to “forget the appearances,” the better to discern the “real truths” swimming behind them (McCarthy, cited above). Each knew that social action, like the fabrication of social meaning, is not pursued by human beings just as they please; that its determinations have to be explained; that the job of the social scientist is to construe the processes by which realities are realized, objects objectified, classes of persons and things classified, and so forth. All of which returns us to the dialectics of deduction and induction – to the co-production of fact and sociological imagining – implicated in “doing ethnography.”

It also returns us to a very basic question: Precisely what kinds of methodological operations are entailed in “doing ethnography” as we envision it? That question is not, we suggest, best answered in the abstract. Just as method is always profoundly theore- tical in its provenance, so its substance ought always to be practice-based and context- sensitive.

CONFRONTING THE GREAT OUTDOORS

Our time: 1989, near the end of apartheid.

Our place: the North West Province of South Africa where, long ago, we did doctoral research.

We returned to the Mafikeng District after an enforced absence of some twenty years; our research, in the interim, had crossed over into Botswana and into the colonial past. Driving in from across the veld, we crested the foothill to the south of the Tshidi- Rolong capital to behold a strikingly discordant landscape. The contours of the old Tswana town – weathered red-clay walls, desiccated thatch roofs, giant boulders, cattle-trodden trails, spry camelthorns – had been dwarfed by a skyline of altogether different scale. The precocious, postmodern outlines of a new city, its architecture a bold pastiche of various international styles of the 1970’s and 1980’s, proclaimed an as- sertive, upstart governmentality. History and Hubris, both capitalized, had consummat- ed a brazen, quick and dirty affaire on this arid terrain: on it, one of apartheid’s most elaborate ethnic “homelands,” had been put in place. The illegitimate insta-polity of Bophutatswana, and the simulacra of its bastard sovereignty, had been erected on land long owned and inhabited by the Tshidi, subjecting them and scores of other Tswana chiefdoms across the northwest to the violent authority of a puppet-state empowered by the material, military, and ideological might of the apartheid regime.

What met our astonished gaze, in sum, was the enactment, in concrete, of that regime’s version of indirect rule: the tight, closely-policed integration of local polities, un- der their “traditional” rulers, into an ostensibly independent ethno-nation. Herein lay the completion of the process, endemic to colonialism, by which those polities – now designated “tribal authorities” – were relegated to the peripheries of a nation-state predicated on difference. The running together of humble adobe and soaring plate glass made visible another juxtaposition: the affirmation, on one hand, of a sense of Tshidi cultural particularity, and, on the other, its encompassment within a wider, multi-ethnic state that was itself a maelstrom of powerful economic, social, and moral currents. It hardly seem- ed accidental that the independent-minded Tshidi chief, Kebalepile, had died in Mafikeng in the early 1970’s, allegedly as a result of witchcraft at the hands of the recently- installed president of Bophuthatswana, Lucas Mangope.15 Mangope, a subaltern sovereign if there ever was one, was seen by the citizenry of Mafikeng as the new colonizing cuckoo in their nest. By blaming him for the occult killing of their traditional leader, Tshidi sought to name the spirit of a spiritless age, the Zeitgeist of late-colonial history.

This magical murder, refracted through the local moral imaginary, might have opened a new chapter in the unfolding confrontation between the late Tshidi world and the wider universe that embraced it. But the history of which it was part went back a long way. As we have noted before (e.g., Comaroff and Comaroff 1991, 1997), setswa- na, the more-or-less open, more-or-less labile ensemble of signs and practices taken locally to constitute vernacular “culture” – the term is used as freely by black South Afri- cans these days as it is by others – was itself the offspring of a protracted colonial en- counter. Mafikeng bore all the scars of that encounter, of earlier struggles, of earlier conjunctures. In the mid-1800’s, for example, it was at the nub of a frontier along which white settlers and African chiefdoms fought over land, labor, and sovereignty; along which, too, evangelists fought for souls and civilization, Later, at the turn of the twentie- th century, during the South African War, it became an imperial battleground on which heroes and villains of all races vied for national gains and personal glory. More recently, it has been branded as a commodity, a heritage site on the newly-wrought tourist map of the postcolony. And for all this time it has lain at the crossroads of an intricate web of exchange relations: relations among the various Tswana polities of the region, relations between them and diverse “strangers,” relations that fan out, today, across the globe. The embarrassment of historical traces we found here stubbornly resist the foreshorten- ed lens of the ethnographic here-and-now.

Consequently, in order to account for the social archaeology of the place, and for the ebullient memories of its people, we were forced from the first to historicize our me- thods; this, in the early 1970’s, at a time when there was a great deal of antipathy within anthropology toward history. We had no alternative but to develop an ethnography of the archives to discern the processes by which the past and the present had construct- ed each other; an ethnography that, among other things, entailed scouring the records – images, inventories, accounts, material shards, documents, linguistic residues, even silences and absences – for the constellation of ordinary practices, the passions and interests, that produced and reproduced this site as an empirical fact, a named-and- known locale (Comaroff and Comaroff 1992). Often this meant trawling texts for what they were not, putting into conversation pieces of paper that, in the cold storage of the archives, languish as solitary objects. It also necessitated our transposing inert verbs and nouns into depictions of living things, of vibrant ritual activities, of expressions of collective affect, effort, effect.

If the ethnography of the archives proved anything, it was that Mafikeng, “Place of Stones,” had, from the start, been situated between a rock and a hard place. The town was established by the ruling Tshidi chief, in the 1850’s, with two ends in mind: to ward off the seizure of his land by white settlers and to quarantine the rise of Christianity, along with its Eurocentric forms of civility. In time, and for complicated historical reasons, Mafikeng would become the capital of the chiefdom. It was here that Tshidi asserted their autonomy as fully as they could from the colonial state, the settler economy, and the British missions; here that they fashioned an ethnically-marked localism – refer- red to, explicitly, as setswana, “Tswana ways and means” – that quietly fused into itself the cultural practices of various others. For their part, the Protestant converts, original residents of the place, were also to make common cause with a national black petite bourgeoisie anxious to proclaim its modernity.

We hardly need insist here that, to be read ethnographically, these economies of signs and practices have to be situated in the intimacy of the local contexts that gave them life. At the same time, they require to be inserted into the translocal processes of which they were part ab initio: processes – commodification, colonization, proletarianization, and the like – composed of a plethora of acts, facts, and utterances whose very description demands that we frame them in the terms of one or other Theory of History. The emerging substance of Tshidi religious, legal, literary culture, their styles of costu- me and senses of self, all deployed images and materials at once fresh and familiar, autochthonous and imported. Each, in its own idiom, replayed, and sought to redress, the mutually-constitutive antimonies of the colonial world: in marking the contrast bet- ween magic and faith, custom and reason, folk dress and fashion, the living forms of setswana recycled and remade the contrast between the culturally particular and the universal, between ethnic subjects and modern persons. Between Africa and Europe.16

For much of its modern existence, anthropology has been trapped inside this set of antinomies. Its ethnographic habitat has, conventionally if not always, been the first term of each: the particular, ethnic subjects, Africa. Conventionally, too, these terms have been taken to signify analytic domains that may be treated as self-sufficient unto themselves. And, for heuristic purposes at least, as hermetically, hermeneutically clos- ed. This was certainly the orientation that framed our first fieldwork among “the Tshidi- Rolong” of Mafikeng in the late 1960’s, when the proclivities of a British structural-func- tionalist training seemed perfectly reflected in the ethnology of African tribes invoked by high apartheid. Yet our field-site – chosen because it gave us an alternative vantage across the Bostwana border if we were expelled by a regime hostile to research on the “wrong” side of the color bar – proved stubbornly intractable to this perspective. Wheth- er in respect of political or religious life, of kinship relations or healing rites, there simply were no “customary” practices that did not bear the imprint of long-standing engage- ment with various elsewheres, with (often coercive) embodied social and material forc- es beyond themselves. The production of the local here was always also entailed in the effort to fabricate some measure of existential coherence and closure against the cross- currents of history, a history of overrule and economic expropriation, of colonial evange- lism, of apartheid, of the ravages of deliberately exploitative labor markets. Of prophets and profiteers, madmen and migrants.

For all their discordant hyper-modernity, then, the built-forms of the bantustan were but an increment in a drawn-out dialogue between the local and the translocal, here and elsewhere – these tropes being understood not as antonyms but as imaginative axes on maps of shifting scale. As it turned out, for all its concrete confidence, this edifice of apartheid was in its death-throes. The long colonial history that had spawned it was coming to an abrupt end, swept away by the changes that marked the close of the Cold War and the realignment of the old international order. So, too, was the nation- al economy that underpinned the ancien régime, its industrial infrastructure and its sovereign autonomy recalibrated by the cumulative effects of neoliberal capitalism. By the time we next visited Mafikeng, two years after South Africa’s first free democratic elections, its civic structures had been inhabited by functionaries of a new provincial government. The old white town, once set off from its black counterpart by the railway-line, and by equally caste-iron cultural and legal barriers, had been significantly integrated.

Other auguries also suggested that Mafikeng had entered a new era – or, rather, that the proportionate relationship between rupture and continuity had, for the moment, tilted somewhat toward the former; history, in our view is never all one or the other, al- ways a complex analytic equation-to-be-resolved. Unfamiliar forces, emanating less from the old international order than from the global economy, were making themselves felt as never before. Some of them promised the infusion of cargo that black South Afri- cans had expected with their liberation: an army of NGOs, of “universal” Neoprotestant churches, of distance-learning corporations, of internet services had opened up around town. Almost immediately, locals tried to capture the bounty promised by these techni- cians of twenty-first century “development.” Not only did satellite dishes mushroom ac- ross the veld. One mud-brick building, nestling beneath a thorn tree on an otherwise barren stretch of land, sported a rough, hand-painted sign: “We teach in English, in step with the global age.”

At the same time, less sanguine signs gave Tswana cause for anxiety. Many pointed out – to us, in letters to newspapers, on local TV – that the old migrant labor system had collapsed and that this collapse, along with a severe recession, had made jobs extremely scarce, especially for young black males. An unusual number of people appeared to be dying in accidents, to be committing suicide, to be victimized by brutal crime, to be ill, to be depressed. Public facilities and welfare services were receding by the month. There was a growing population of “black people” on the streets, immigrants from elsewhere in Africa who were drawing much suspicious and scandalized talk: hav- ing eluded state regulation, it was said, they were plying their wares noisily on once pristine sidewalks, thus usurping the trade of South African merchants. Not only that: they had brought drugs and AIDS with them, and had taken the few available jobs on the surrounding farms. And yet, despite all this pessimism, notwithstanding all this apo- calyptic talk, in the midst of this economy of genuine hardship, some locals seemed, mysteriously, to be prospering. As we note elsewhere (Comaroff and Comaroff 1999a), it is this that has fed the raw underside of the occult economy: the killing of alleged wit- ches and zombie conjurers.

Zombie-conjurers. This brings us back full circle to where we began. To the strangely dissociated man in the police station, to the youths who killed because they believed their fathers to have been turned into phantom laborers, to popular representa- tions of the violent abstraction entailed in witchcraft. Recall what we said at the outset: that the zombie is a figure metonymic of the playing out of world-historical forces in the northerly reaches of South Africa right now; also of the domestication of a form of neoliberal capitalism thought to enable the production of wealth without work. Recall, too, the question that followed: how are we to make sense methodologically of this figure, of those forces, of their determinations, of the unfolding connections between them? That the question demands a sociological imagination at once local and translocal, empirical and analytic, was brought into sharp relief for us in a context part pedagogic, part ethnographic. During a history class at the University of the North West, a graduate student broke suddenly across the discussion: “Do Americans believe in diphoko, in magical medicines? ” he asked. “Is it like here? Is there also trouble with zombies in America?”

By what methodological means, then, did we actually address the question of the living-dead in the late Tswana world?

DISCURSIVE FLOWS AND THE DIALECTICS OF DISCOVERY

[Writing a novel is] like playing chess in three dimensions.

David Lodge 1999:52

So, too, is doing ethnography. Four dimensions, actually, if one includes the ter- rain of the virtual: the electronic commons that has interpellated itself – as a medium of translocal communication, as a vehicle for the flow of money and other kinds of capital, as a mechanism of the market, as an instrument for the establishment of public spheres of different scales – into even quite remote social worlds.

Unlike chess, however, ethnography-as-practice has, in the first instance, to con- struct its own field of play, its heuristic landscape. Strategically, it has always seemed logical to us to locate the center of that field around one or more focal points to which the anthropological senses are drawn because they are the crucibles in which contem- porary vernacular concerns – whatever they may be, whatever their phenomenal scale – are construed, enacted, played out, socially contextualized. Given that our anthropo- logy seeks to be empirically-grounded without being empiricist, our objects of research have invariably been defined with reference to the prevailing preoccupations of the times and places in which we have worked, whether they be the politics of chiefship or ecstatic religious movements, agrarian development and its undersides, the colonial en- counter, occult economies, or, most recently, crime, policing, and the metaphysics of disorder. In the dialectic of the concept and the concrete, it is the latter that sets metho- dology in motion, serving as the fons et origo of the operations by which we set out to apprehend the existential processes of everyday life. Our ethnography, in other words, takes off not from theory or from a meta-narrative, but from the situated effects of see- ing and listening. Of course, the way in which we see, what we pay attention to, and how, is not empirically ordained; that, ineluctably, depends on a prior conceptual scaf- folding, which, once the dialectic of discovery is set in motion, is open to reconstruction.

In the late 1990’s, the zombie, and the enchanted economy of which it was part, provided just such a focal point at which the preoccupations of the period had taken tangible shape. How did we know this? It came at us, insistently, from a number of diverse sources, some of them already alluded to: in such episodes as the encounter with the almost-naked man on Station Road, in what followed at the police station, and in the sheer ordinariness of the whole thing to the men in blue; in the murder of alleged corpse-conjurer, Motlhabane Makolomakwa, in its avidly-consumed press coverage (see Figure 1), in its courtroom arguments, and in the conversations to which it gave rise, many of them about the “epidemic” of occult violence afflicting the northerly provinces; in “mob” attacks, committed by local youths in the name of their communities, against those suspected of practicing the arcane arts; in personal stories of the sort told to us by the scion of a ruling dynasty – a man with a first-class graduate degree in development studies, an excellent job in government, and a large following as a DJ in a large city nearby – who had lost a beloved sibling, snatched away secretly by a witch for whom he worked until rescued many months later; in a remarkable incident in which po- lice tried to save a young boy from the persistent attacks of a vicious tikoloshe, a trans- local witch familiar,17 first by calling in the local television channel in the hope that its cameras might immobilize the creature and then by eliciting the help of several technicians of the sacred (see Figure 2); in the reactions of the state to outbreaks of witch- craft killings, which included tough law enforcement, high level conferences on the top- ic, and the appointment of a commission of enquiry; in discussions on the internet, in national and regional TV dramas, documentaries, news broadcasts, and talk shows, in local genres of cultural production (see Figure 3); and, most of all, in our everyday ex- changes in homes and schools, stores and shebeens, taxi ranks and churches across the length and breadth of the Mafikeng district.

This was not all, not by any means. But it gives a sense of the way in which a flow of narratives, incidents, activities, dramas, material exchanges, conversations, and representations embedded in the “natural” discourse of different and complementary public spheres may come to organize the ethnographic gaze – and, thereby, to set me- thod in motion. Discursive flows, although having focal centers, are inherently open, fle- xible in scope, and shifting in both their content and their constituents. Determining what, exactly, falls within the purview of any such flow is itself a product in part of pay- ing careful attention, in part of inspired guesswork, in part of theoretical and philosophi- cal predilection; making sense of its substance depends on what, previously, we have spoken of as an “imaginative sociology.” We use “imaginative” here in two senses. It re- fers to (i) doing ethnography by plumbing – through whatever resources of the analytic imagination are available to us within the political and ethical imperatives of our practice – the phenomenal worlds in which we situate ourselves; this by (ii) seeking to grasp the manner in which those worlds are indigenously imagined and inhabited by people vari- ously positioned within them. Note all the plurals. They point to an anthropological cli- ché, albeit an important one: that most of the signs and practices with which we con- cern ourselves are either contested or, if not, are the object of a polyphony of percep- tions, valuations, means and ends.

To the extent that doing ethnography necessitates, in the first instance, tapping into focal discursive flows – and, lest we be misunderstood, we reiterate that this inclu- des not “just” talk or texts but practices as well, not “just” the meaningful but also the material – it demands three critical methodological operations. Each is a condition of the others’ productivity.

The first is the pursuit, in respect of any given discursive flow, of points of articu- lation among the various spheres in which it manifests itself; this by tracing the co-pre- sence of persons, texts, images, or arguments (and especially arguments of images) across them. Thus, for example, the imaginative sociology we were able to construct surrounding the figure of the zombie – and that was to sediment into our ethnographi- cally-rooted abstraction, the “occult economy” – took shape when we began to hear si- milar words and see similar pictures over and over: when, among many other things, the accused youths in the Makolomakwa murder trial claimed that the deceased had “killed their fathers and put them to work”; when stories about zombification kept return- ing to the “fact” that the witches in question, invariably sexual “perverts,” had “turned people into tools,” thereby preventing ordinary citizens from earning a living or starting a family; when an old woman, said to have amassed “mysterious” wealth, was told, as she was set on fire by the “boys” of her village, that they had no income because of her; when so much local opinion, from the most intellectual to the most humble, blamed the living-dead for the absence of employment, for denying young black males the opportu- nity to graduate to adulthood, for the despoliation of community. This is not to say that all representations of, or explanations for, the postcolonial (re-)appearance of zombies are the same. Nor that they are ascribed the same social salience by everyone. However, where there is argument about the matter – be it in courts of law or over quarts of beer, on soccer fields or in the maize fields, around backyard fires or among fired work- ers, in university classrooms, church meetings, or the electronic media – it usually turns back to the connection of witchcraft to the dearth of work and the impossibility of securing the future; the last being what we, in theory-speak, might refer to as a crisis of social reproduction. This, in short, is the animating vernacular around which the discursive flow is organized. It turns out to be crucial in the dialectic of the concept and the conc- rete, of theory and ethnography.

The first methodological operation, then, is to map the substance of the pheno- menal landscape on which any discursive flow is grounded, thus to identify its animating vernaculars and to chart the object world in which it interpolates itself. The second is to follow the traces of that discursive flow, of its various signs and images, tracking the mi- gration of the latter from their densest intersections to wherever else they may lead.

Let us give a couple of instances from the situation with which we were concern- ed here. One is the allusion to the sexual perversion of witches, a submerged theme in many zombie narratives. At face value, this allusion seems, in itself, to have little to do with the workings of the occult economy or the figure of the zombie, more with the figuration of the witch as “standardized nightmare” (Wilson 1951), the epitome of anti-sociality and immorality. But in pursuing the allusion, in posing questions about it, in seeing where else it turned up, we found ourselves drawn into a meaning-maze that took in AIDS, the sexualization of death, bad blood, compromised masculinity, and drought – and culminated, by fusing all of these things, in the clear and present threat to the fu- ture of communities everywhere attendant on the fact that young men cannot find work or make families. As sexual pervert, the witch, in short, embodies social destruction, fertility abused, social reproduction violated.

The other instance also arose out of a recurrent theme in zombie narratives: What precisely is the reach of the occult economy? What is the reach of the modern witch? Is corpse-conjuring, or the arcane fabrication of wealth without work, purely a parochial matter? Or does it somehow extend beyond? One night, the local TV channel held a phone-in talk show in which the special guests were a pair of young “reformed” satanists, each with his spiritual advisor. Asked about the difference between witchcraft and satanism, one answered, in a fluent mix of Setswana and English: “Satanism is high-octane witchcraft. It is more international” (Comaroff and Comaroff 1999a). This comment called forth a flood of responses from the virtual community constructed by the program. Tse station switchboard was overrun. Audiences across the province were fascinated. Satanists were said, by and large, to be youthful, male, and black, just the social category most under threat of joblessness, most likely to dabble in nefarious new technologies. Witches, by contrast, were held to be motivated, more often, by local con- flicts, framed in long-standing idioms of kinship and community; although they, too, app- ear to be widening their horizons and their range of techniques. As the “high octane” petrochemical image suggests, what “satanic” youth bring to the occult economy is a capacity to “ride the tiger” – actually, in these parts, a leopard – “of time-space comp- ression” (Harvey 1990:351): to move across vast distances instantly, thus to accumu- late riches, without visible effort, by means unknowable to ordinary persons. The sym- bolic references in this are too dense to unravel here: they extend from the “fast” wealth being produced in the postcolony by control over the transportation of people, signs, and things to the changing salience of borders and transnational elsewheres in neolibe- ral South Africa. Above all, however, what became plain, listening both to the partici- pants in the show and to those with whom we watched it, was the fact that the occult economy is understood to link the most local of concerns, activities, and relations – un- derstood in the most local of terms – to inscrutable forces arising out of an equally in- scrutable world beyond, a world ever more “global.” This last, we stress again, is not our gloss. Recall that sign on the mud-brick school, the one that promised an education “in step with the global age” (above, p.00).

In short, the second methodological operation involves mapping the extensions of the phenomenal landscape, the four-dimensional geography (see n.14) with referen- ce to which any discursive flow constitutes itself. Self-evidently, this, like mapping its substance, demands more than “multi-sited” ethnography. It demands an ethnography that, once orientated to particular sites and grounded issues, is pursued on multiple di- mensions and scales: an ethnography as attentive, say, to processes occurring in virtu- al space as to those visible in “real” places-under-production; to the transnational mass- mediation of images as to ritual mediations between human beings and their ancestors; to the workings of state bureaucracies or international courts as to the politics of “tradi- tional” chiefship and customary moots; to the flow of commodities across the planet as to marriage payments between lineages; and so on and on. Often it turns out that there are intimate, if invisible, connections across dimensions and scales: just as planetary commodity flows may, these days, determine bridewealth in an African village, so bride- wealth in an African village may have an impact on the planetary flow of labor, cash, and goods; similarly, just as the purview of local chiefs and their “traditional” courts may be decided by global human rights jurisprudence, global human rights jurisprudence is being challenged by demands for the recognition of “traditional” cultural imperatives.

The third methodological operation is to trace the passage of a discursive flow over time; this to establish what, precisely, is new about it and what is not, what are the relative proportions of rupture and continuity to which it speaks, what is unique and what is merely a local instance of a wider phenomenon. How? By means of a counter- point: by (i) eliciting a local genealogy of cultural precursors and (ii) running it up against a comparative archaeology of similar signs and practices to ascertain where else, and in what circumstances, parallel discourses might be found. In respect of the zombies of the North West Province, and the occult economy of which they are part, local genealo- gies make it clear that they have not been around for much more than a decade; regar- ded thus, they signal a rupture. But there did exist a foreshadowing: sefifi, observed by missionaries in the nineteenth century (Comaroff and Comaroff 1991:143), a condition – in which “manhood is dead, though the body still lives” (Brown 1926:137) – brought about by the eclipse of a person by another, more powerful than s/he. This condition, it seems, provided a semantic frame within which the zombie has been accommodated. As to a comparative archaeology, there is evidence of at least two broadly parallel his- torical situations in Africa – in Mozambique and Cameroon earlier this century – in which living-dead have appeared (Comaroff and Comaroff 1999c). In both instances, their presence was intimately tied to radical changes in colonial labor conditions, to the disruption of received connections between persons, production, and place, to the pre- cariousness of wage employment, and to the alienation attendant on new forms of work. Put all this together and the point becomes clear: once historicized and interpella- ted into its local cultural context, the discursive flow surrounding the figure of the zom- bie has most immediately to do with labor history, with a burgeoning fear of the eclipse and commodification of people and social relations, with a sense of lost control over the means of producing value, with threats to the survival of local worlds under the impact of enigmatic forces from outside, with the unmooring of horizons and expectations oc- casioned by shifts in the workings of capital.

CONCLUSION

This brings us back, one last time, to the dialectic of induction and deduction, of theory and ethnography, of the concept and the concrete.

When we resumed our work in post-apartheid South Africa in the 1990’s, as we have said, we had no idea that we would run into a fully fledged occult economy; or, to be more precise, into the phenomena captured in this ethnographically-grounded abs- traction. Nor could we have known how that economy had become a public preoccupa- tion. The appearance of a new breed of witches and zombies, and the anxieties they heralded, might have been interpreted purely as an expression of parochial conflicts and relations gone bad. What is more, in the hands of a cultural anthropologist with only the pristine horizons of the particular in view, a case could no doubt have been made for the idea that the living-dead of the present are a transformation of the sefifi of old; that the mystical evil of the here-and-now is an extension of “traditional” notions of witchcraft and sorcery. However, once we had traced out the discursive flow in which zombification is caught up – made manifest, methodologically, by charting the land- scape on which it had taken shape, rendered decipherable by recourse to local genealogies and comparative archaeologies, mediated by our own conceptual categories and commitments – it became obvious that this kind of explanation would have been woe- fully incomplete. For one thing, it would have left unaccountable the fact that similar phenomena have appeared in very different cultural contexts at roughly the same time and in response to the same broader historical conditions. For another, it would also have paid scant respect to the real-world concerns of Tswana living in the North West: to their arguments about the impossibility of social reproduction, about arcane means of producing wealth, about new forms of labor, commodification, and alienation, about witchcraft, satanism and globalization.

In seeking to take account of those arguments and their social motivation, and to grasp the phenomenology of the lived, material world out of which they arose, we brought to bear an explicit theoretical orientation; it is one about which we had written a fair amount over the previous decade, one which contained within it a particular under- standing of the contemporary history of capital. That orientation primed our early read- ings – and misreadings – of the “new” South Africa. But it did not take long for its insuf- ficiencies to make themselves plain. Apart from all else, our take on the workings of modern industrial capitalism and its colonial extensions did not prepare us for the post- colony, for its postmodern zombies and unemployment-related witch killings, for its “cri- sis” of masculinity and generation, for the complex absent-presence of the state. It was, in other words, the incompleteness of our theoretical scaffolding – incomplete, that is, in the face of the concrete world which we were encountering – that set the dialectic in motion anew, altering our conceptual repertoire just as that repertoire was being mobili- zed to make sense of the unexpected landscape on which we found ourselves.

Ethnography is like much else in the social sciences; indeed, more so than anthropologists often acknowledge. It is a multi-dimensional exercise, a co-production of social fact and sociological imagining, a delicate engagement of the inductive with the deductive, of the real with the virtual, of the already-known with the surprising, of verbs with nouns, processes with products, of the phenomenological with the political. Robert Foster (2002:247) has recently remarked, as we have ourselves (Comaroff and Coma- roff 1999a), that the key problem of doing ethnography “is ultimately a question of scale.” For him, that question boils down to the avoidance of “dissolving local particularities in the uniform sameness of global conditions without treating the radical distinctiveness of the local as if it stood against or apart from the global.” For us, the challenge goes yet further. It is to establish an anthropology-for-the-present on an ethnographic base that dissolves the a priori breach between theory and method: an anthropology, of multiple dimensions, that seeks to explain the manner in which the local and the trans- local construct each other, producing at once difference and sameness, conjuncture and disjuncture. An anthropology that takes, as its mandate, the need to make sense of the intersecting destinies of human lives, wherever they may happen to be lived out.