One Saturday morning in 1980, I found myself addressing a large, mainly black working class audience in Detroit on the subject of the Atlantic slave trade. I was claiming that, before the industrial revolution, relations between Europeans and Africans on the Guinea Coast were relatively equal. The Africans supplied the slaves and the Europeans bought them. At this point, a huge man sitting on the front row suddenly stood up:
Man: You sayin’ we slaved our own people, motherfucker?
Keith: Well, yes, I suppose…yes.
Man: How you know?
Keith: Oh, historical records – Portuguese, Dutch, French, British records.
Man: Any black people write them records?
Keith: Ah yes, good point. We do rely mainly on what white people wrote then.
He sat down and, rather shaken, I continued with my lecture. I start with this story because, although the contributors to this workshop have brought up the scientific, forensic and self-referential dimensions of evidence in a variety of media, I miss an explicit engagement with the politics of evidence, outside as well as inside the universities.
Why does evidence matter? It clearly does to the young anthropologists who convened this workshop (see note at the end of this post) and to others who did much the same thing not long before (Engelke 2008). Here I will try to explain why, for reasons of institutional politics, a group of apprentice anthropologists from around the world, temporarily inhabiting the shell of what was once British social anthropology, might be pre-occupied with questions of evidence at this time. But there are more general considerations too. What links the two levels is the problem of intellectual authority in a world where respect for it is dwindling fast (Grimshaw and Hart 1993).
I was once asked to launch a series of Key Debates in Anthropology (Ingold 1996) by proposing the motion “Anthropology is a generalizing science or it is nothing”. Wanting to win and knowing my audience, I cast around for the most ecumenical definition of “science” I could find, since any hint of the men in white coats would be a disaster. (Our side lost in any case). I argued that the eighteenth-century Enlightenment invented modern anthropology as a means to a revolutionary political end, the institution of “democracy” at the expense of the Old Regime. People power could only succeed if it was based on sound knowledge of the world; so “science” was promoted as an alternative to religion, myth, superstition, etc., all of which sustained unequal society. The first reason evidence matters is that it is central to the scientific project; this in turn is indispensable to social and material improvement of the human condition. Science is about getting something right over and over again within an acceptable range of error; and that means systematically juxtaposing the rules of knowledge and the evidence of practice. Otherwise the bridge collapses, you miss the moon or whatever. The consequences for human life of getting medical research right are obvious and this does sit uneasily with conventional ethnography.
My Detroit encounter was an object lesson in the truism that the rules of engagement normally observed in the academy don’t hold outside it. My life as an anthropologist has been shaped by the politics of race, often in situations where I was not only unprotected by my academic expertise, but actually suspect because of it (and hopelessly outnumbered). Evidence matters under these circumstances because a claim to objectivity made in ordinary language may be the best means of self-defence. (Convincing people of one’s humanity helps too). The founders of twentieth-century ethnography – Boas, Rivers and Mauss – all deliberately engaged with political questions of the day through their anthropological knowledge. Mauss in particular was absolutely clear about how sociological science supported his career as a political activist (Hart 2007). It is symptomatic of what happened next that the leaders of British social anthropology declined to take part in the UNESCO book on race (UNESCO 1956) on the grounds that their discipline’s academic standing as a social science would be compromised by getting involved in “controversial” topics (Verena Stolcke personal communication).
The forensic uses of evidence often have legal consequences that can be particularly severe for individuals. At the same time, this rhetorical practice has wide application inside and outside the courtroom, with effects across a continuum ranging from the catastrophic to the trivial. Marilyn Strathern explores, through examples like academic promotion procedures and women’s credibility as witnesses in Mount Hagen, how evidence actually works in discourse and social practice. Here I would just make the point that the burden placed on evidence is relative to the consequences of the decision made or the story told. For instance, readers of a newspaper story about “man bites dog” will demand a stronger line of evidence than for the commonplace reverse.
The same goes for issues implicating professional authority. Thus, as part of his unsuccessful campaign to impress the Freudians, Malinowski (1929) asserted that the Trobrianders were ignorant of physiological paternity. This was later refuted in Man by a colonial administrator who had spent sixteen years in Papua and cited various kinds of evidence, including the Trobrianders’ practices of contraception and animal breeding (Rentoul 1931). Malinowski replied with fury, mentioning his being described as a “visiting anthropologist” seven times in five pages and complaining haughtily that Rentoul “sweepingly disposes of the qualifications of the scientific specialist” (Malinowski 1932:34). At least the world cared about the truth value of Malinowski’s story. The vast bulk of ethnography produced today is of interest only to professional insiders and their students. I think this should be made explicit when discussing questions of evidence.
But of course life carried on within academic seclusion has its politics too (Cornford 1908); and this, it seemed to me then and now, played a significant part in the dynamics of the original workshop. In short, a small network of anthropology PhD students and postdocs in London’s hinterland (Cambridge in particular) conceived of a one-day workshop organised by and for people like themselves. The response of young and established scholars alike was larger than expected and the workshop expanded into a two-day weekend event. By the time my slot arrived at the very end, the audience had reverted to its original expected character. So I tried to address the question at issue from the perspective of people writing a doctoral thesis in anthropology or having just completed one. I was particularly sensitive throughout the workshop to clues about why evidence mattered to them.
Almost all the younger participants had been forced by departmental convention to write a thesis employing the methods of fieldwork-based ethnography. The founders of the British school they had joined so recently produced a stream of exemplary monographs in this genre between the wars and just after. I have long been convinced that, when the history of twentieth-century thought is written, the ethnographies of Evans-Pritchard, Firth, Fortes, Lienhardt, Nadel, Richards, Turner and many others will be celebrated for the extraordinary achievements that they were. Their successors—trained in the 1960s and early 70s, blessed with the golden age of university expansion and still broadly in charge, although now on the edge of retirement—maintained the tradition, while also branching out into several other genres of research and writing. (I must confess that I never managed to complete an ethnographic monograph myself).
Since the 1980s, the institutional context of anthropology in Britain and the world more generally has been transformed in ways that are both too familiar and complex to summarise here (Grimshaw and Hart 1995, Hart 2005). What anthropologists do professionally today is almost unrecognisable when compared with those classical monographs. Yet they still cling to a guild identity founded on fieldwork and an increasingly loose invocation of the term “ethnography”. I believe that a concern for “evidence” arises from a shared experience of the dilemmas these young anthropologists encounter when trying to establish themselves within such a tradition.
Let me be clear from the outset. Choosing to live with the people in order to discover what they do and think was a radical political move, fully compatible with anthropology’s birth in the Enlightenment’s drive to apply science to democratic ends. Read again the “Introduction” to African Political Systems (Fortes and Evans-Pritchard 1940) to get a sense of this politics. Hitherto, they say, political philosophy has been the province of European elites with no direct experience of how the mass of humanity lives. We have to abandon our libraries, laboratories and seminar rooms (if only temporarily) to discover what is really going on out there and to develop concepts honed in empirical investigation. This is anthropology’s deep political purpose, to shake up ruling ideas in the name of a wider human agenda. Marx (in Capital, 1867) refers to ideology as a camera obscura that turns the image upside down. Ideas come from life, but it is the task of ideologues to persuade people that it is the other way round, that their lives depend on ideas produced and controlled by an expert elite. It was Malinowski’s supreme achievement to embody a model of ethnographic practice where ideas have to be derived from life, from systematic observations made in all corners of the world. That is why the classical British school has a unique part in the history of modern anthropology. No-one has ever done it better. The premise is of course naïve and epistemologically problematic (Grimshaw and Hart 1995). But as a riposte to the lingering legacy of the Old Regime, the move was priceless and should be conserved at all costs.
Writing a doctoral thesis has a claim to being the worst job in the world. Its demands throw students onto their own resources in lives of lonely isolation; yet they are subject to a formal hierarchy that might literally stop them from ever submitting their work or might reject it utterly when they do. There is nothing new about that. What is new is that the intellectual practice anthropology students are expected to follow is now contradicted by the conditions of the job market they hope to enter and by the prevailing activities of the profession at large. For the same teachers who invoke the legacy of Malinowski and Firth routinely advance their own careers by literary means that owe almost nothing to that legacy.
“The field” for most anthropologists occupies a small part of our working lives. The number of truly ethnographic monographs we produce is even smaller. We spend any time spared from teaching and administration cranking out conference papers, journal articles and book chapters, usually with an upper limit of less than 10,000 words. The whole aim of this writing is to come up with a new argument relevant to “the literature” we presumptively have in common. There is precious little room here for anything like ethnographic description. Instead the camera obscura of ideology rules in an inverted pyramid: scattered soundbites drawn from life illuminate baroque arguments of unfounded generality, all delivered with a long list of references to shore up the author’s claim to erudition. Even while students are labouring to write a thesis on the Malinowskian model, they are being encouraged to emulate this institutional practice, since, without publications, their prospects of academic employment would be even lower than they already are.
The literary turn in contemporary anthropology (Clifford and Marcus 1986) feeds into how students approach writing a thesis. Many of them believe they are writing a book, whereas the highest value of a thesis is its methodological transparency. The few people who ever read it will want to extract bits they can use for their own research purposes. They need to know if they can trust these snippets, which means that how they were gained should be made unusually explicit. This privileges bad writing over good and it would never do for a book, where literary convention favours elegance. I was told as a graduate student that a good thesis or monograph should not be driven by its theoretical ideas. Rather the reader should be able to reach conclusions independent of those of the author, drawing on an abundance of empirical material that has not been shaped merely by the need to illustrate an abstract argument. According to Bruce Kapferer (2005), Max Gluckman took this principle even further to a pedagogical practice where students were encouraged to critique the published ethnographies of their teachers, who would then be forced out into the field to make good the damage and try again. However idealised a portrait, this does stand in contrast to the spirit of anthropological scholarship today.
The importance of evidence for apprentice anthropologists operating within the leftovers of the British school should perhaps be seen in this light. Why should they sacrifice their abstract intellectual ambitions to a realist paradigm when none of their teachers do? Here again, Marilyn Strathern illustrates what we all do when we reach a position of some security. If asked to contribute to a theme, we assemble a few ideas, string them together into an argument, and illustrate it with examples drawn from our whole life experience, whether based on “fieldwork” or not. This Afterword is another example of the genre. I actually think that the idea of ethnography as a practice drawing on all of life, not just some artificial compartments labeled “research”, has much to be said for it. Strathern’s essay is proof of the genre’s potential. But we would never allow a PhD student to try the same thing, and postdocs won’t get far by employing this strategy either. Could this contradiction be one source for the current fascination with questions of evidence?
The workshop’s organisers, Timm Lau in particular, let the cat out of the bag when he spoke in some introductory remarks about a dialectic of “evidence” and “protocols”. This last term is significant. It comes from the politics of the internet and computing practice. Some now argue (e.g. Lessig 1999) that control is no longer established by making the rules, but by writing the code, the “protocols” that invisibly shape what practitioners can actually do. The “evidence/protocols” pair seemed to me a metaphor for the student/teacher dialectic. There was a time when any ethnographer had a huge chunk of the world to him – or herself. The world these students encounter is crowded with burdensome precedents, established ways of “knowing” the Dogon or whoever. Their teachers are closer to these traditions, more retentive of the past than the students. Lau was explicit in raising the possibility that “evidence” (“I have been there more recently than you and it isn’t like that any more”) could be used to break out of restrictive “protocols” that often seemed arbitrary and unjust. The workshop was dotted with other examples of this move. It is linked to the emphasis on reflexivity and the use of new media. If truth is in fact continuously renewed on a reflexive basis or transformed by different technologies, the power of tradition to constrain present action is thereby reduced. But can they get away with it? An appeal to the evidence, whatever that is these days, might be one way of fighting the underdog’s corner.
I make this inference partly because my own doctoral research involved studying the Tallensi made famous by the head of my department, Meyer Fortes (e.g. Fortes 1949). He began his fieldwork in Northern Ghana in 1934 and I in 1967, as part of a rural-urban study of migrants. Apart from being shamed by the sheer virtuosity of his fieldwork technique, I found it almost impossible to place my findings in some framework of comparison with his. Suppose I observed a ritual that he described in one of his writings and found a difference. Was this to be accounted for as a change in practice through time, or as empirical error or bias on the part of either ethnographer? Such questions were fascinating in a recursive way, but I found them to be ultimately fruitless (but see Hart 1978). They raise the thorny issue of the relationship between ethnography and history, as touched on in my opening anecdote. Anthropologists do far more than fieldwork (Hart 2004) and it is time that we abandoned the ostrich-like pretence that our methods have not moved on since Malinowski founded the modern British school.
Our ethnographic methods have always been substantially occult. The practice of hoarding “fieldnotes” in private places where they can never be inspected by other professionals is indefensible. We all know it, but we keep quiet. The pretense that we “learn the local language” in a year or so covers up tacit acknowledgment that our linguistic competence might reach that of a 9-year-old, but rarely more. Those who commit to a discipline called “ethnography” live in constant fear of being found out; and this is exacerbated by a revolution in transport and communications that is rapidly collapsing the walls anthropologists once erected between their academic base and “the field”. How much more intolerable it is then for apprentice ethnographers to learn to live with the contradictions of this inherited norm under social conditions that by most measures are deteriorating, when compared with the heyday of the universities in the 60s and 70s. I offer, without much evidence, the suggestion that one impulse to this collective effort lies in such pragmatic dilemmas.
However we explain the metaphysics of contemporary anthropological practice, the factors that sometimes lend urgency to the search for reliable evidence were not the main focus of concern on this occasion. Rather the object of the exercise seems to have been a mixture of concern for professional authority and for existential truth. The former is defensible, even laudable in apprentice scholars; but I find it less interesting than the latter. There has always been an element of the “lone ranger” in anthropologists, a sense of making a personal journey in the world that offers infinite scope for reflection. We move inwards as we get older and I would certainly not deny the value of such an exercise. Edward Said once suggested (in a 1992 TV programme) that life gives us so many cultural fragments and our task is to make a story out of them. It is not quickly done. My gloss on this sentiment would be Durkheimian: fieldwork is one of many opportunities to internalise experience of society and our job is to get some of it out eventually in a coherent form that we can share with others. My own explorations of the “informal economy” (Hart 1973) came from interacting with development economists who were then obsessed with the problem of “urban unemployment”. This term did not sit well with my fieldwork experience in an Accra slum, but it took some time to figure out why and even longer to find an alternative vehicle to convey the results of my ethnography to economists. Questions of evidence were definitely secondary in this process of excavation.
A preoccupation with self and the world has its roots in humanism, but it is not science and it lacks the political force of good science. I have argued here that anthropologists should be more explicit in confronting the politics of evidence. Maybe it would help if this generation of Malinowski’s heirs tried being more honest than their predecessors. In the process they might renew anthropology’s mission to engage with the deepest political questions concerning humanity’s past, present and future.
Clifford, James and George Marcus, ed. 1986. Writing culture: the poetics and politics of ethnography. Berkeley: University of California Press.
Cornford, Francis. 1953 (1908). Microcosmographia academica: Being a guide to the young politician. London: Bodley Head.
Engelke, Matthew. 2008. The objects of evidence. In The objects of evidence: anthropological approaches to the production of knowledge, ed. Matthew Engelke. Journal of the Royal Anthropological Institute Special issue:S1-S21.
Fortes, Meyer. 1949. The web of kinship among the Tallensi. London: Oxford University Press.
Fortes, Meyer and Edward Evans-Pritchard (editors). 1940. African political systems. London: Oxford University Press.
Grimshaw, Anna and Keith Hart. 1993. Anthropology and the crisis of the intellectuals. Cambridge: Prickly Pear Pamphlets No. 1
1995. The rise and fall of scientific ethnography. In The future of anthropology: Its relevance to the contemporary world, ed. Akbar S. Ahmed and Cris Shore. London: Athlone, 46-64.
Hart, Keith. 1973. Informal income opportunities and urban employment in Ghana. Journal of Modern African Studies 11.3: 61-89.
1978. The economic basis of Tallensi social history in the early twentieth century. In ed. Research in Economic Anthropology, Vol. 1. ed. George Dalton. Greenwich, CT: JAI Press, 185-216.
2004. What anthropologists really do. Anthropology Today 20.1: 3-5.
2005. The hit man’s dilemma: Or business, personal and impersonal. Chicago: Prickly Paradigm Press.
2007. Marcel Mauss: In pursuit of the whole. Comparative Studies in Society and History 49(2): 473-485.
Ingold, Tim (editor). 1996. Key debates in anthropology. London: Routledge.
Kapferer, Bruce. 2005. Situations, crisis and the anthropology of the concrete. Social Analysis 49.3: 87-125.
Lessig, Lawrence. 1999. Code: And other laws of cyberspace. New York: Basic Books.
Malinowski, Bronislaw. 1929. The sexual life of savages in North-western Melanesia. London: Routledge and Kegan Paul.
1932 Pigs, Papuans, and police court perspective. Man 32, 33-38.
Marx, Karl. 1970 (1867). Capital: The critique of political economy, Volume 1. London: Lawrence and Wishart.
Rentoul, Alex. 1931. Physiological paternity and the Trobrianders. Man 31, Aug.: 152-154.
UNESCO. 1956. The race question in modern science. New York: UNESCO/Wm. Morrow.
‘Questions of evidence: ethnography and anthropological forms of knowledge’, Cambridge, January 2007. This is a lightly edited version of an Afterword to the conference publication, How Do We Know? edited by Liana Chua, Casey High and Timm Lau, Cambridge Scholars Publishing, Newcastle, 2008.