In: James Clifford and George E. Marcus (eds.). Writing Culture. The Poetics and Politics of Ethnography.
Reading Other Cultures.
This inequality in the power of languages, together with the fact that the anthropologist typically writes about an illiterate (or at any rate non-English-speaking) population for a largely academic, English-speaking audience, encourages a tendency I would now like to discuss: the tendency to read the implicit in alien cultures.
According to many social anthropologists, the object of ethnographic translation is not the historically situated speech (that is the task of the folklorist or the linguist), but “culture,” and to translate culture the anthropologist must first read and then reinscribe the implicit meanings that lie beneath/within/beyond situated speech. Mary Douglas puts this nicely:
The anthropologist who draws out the whole scheme of the cosmos which is implied in [the observed] practices does the primitive culture great violence if he seems to present the cosmology as a systematic philosophy subscribed to consciously by individuals… So the primitive world view which I have defined above is rarely itself an object of contemplation and speculation in the primitive culture. It has evolved as the appanage of other social institutions. To this extent it is produced indirectly, and to this extent the primitive culture must be taken to be unaware of itself, unconscious of its own conditions. (1966:91)
One difference between the anthropologist and the linguist in the matter of translation is perhaps this: that whereas the latter is immediately faced with a specific piece of discourse produced within the society studied, a discourse that is then textualized, the former must construct the discourse as a cultural text in terms of meanings implicit in a range of practices. The construction of cultural discourse and its translation thus seem to be facets of a single act. This point is brought out in Douglas’s comments on her own translations of the meanings of the pangolin cult among the Lele:
There are no Lele books of theology or philosophy to state the meaning of the cult. The metaphysical implications have not been expressed to me in so many  words by Lele, nor did I even eavesdrop a conversation between diviners covering this ground… What kind of evidence for the meaning of the cult, or Of any cult, can be sensibly demanded? It can have many different levels and kind. of meaning. But the one on which I ground my argument I. the meaning which emerge. out of a pattern in which the parts can incontestably be shown to be regularly related. No one member of the society is necessarily aware of the whole pattern, any more than speakers are able to he explicit about the linguistic patterns they employ. (1966:173-74)
I’ve suggested elsewhere (Asad 1983a) that the attribution of implicit meanings to an alien practice regardless of whether they art acknowledged by its agents is a characteristic form of theological exercise, with an ancient history. Here I want to note that reference to the linguistic patterns produced by speakers does not make a good analogy because linguistic patterns are not meanings to be translated, they are rules to be systematically described and analysed. A native speaker is aware of how such patterns should be produced even when lit- cannot verbalize that knowledge explicitly in the form of rules. The apparent lack of ability to verbalize such social knowledge does not necessarily constitute evidence of unconscious meanings (cf. Dummett 1981). The concept of “unconscious meaning” belongs to a theory of the repressive unconscious, such as Freud’s, in which a person may be said to “know” something unconsciously. The business of identifying unconscious meanings in the task of “cultural translation” is therefore perhaps better compared to the activity of the psychoanalyst than to that of the linguist. Indeed British anthropologists have sometimes presented their work in precisely these terms. Thus David Pocock, a pupil of Evans-Pritchard’s, writes:
In short, the work of the social anthropologist may be regarded as a highly complex act of translation in which author and translator collaborate. A more precise analogy is that of the relation between the psychoanalyst and his subject. The analyst enters the private world of his subject in order to learn the grammar of his private language. If the analysis goes no further it is no different in kind from the understanding which may exist between any two people who know each other well.[!] It becomes scientific to the extent that the private language of intimate understanding is translated into a public language, however specialized from the layman’s point of view, which in this case is the language of psychologists. But the particular act of translation does not distort the private experience of the subject and ideally it is, at least potentially, acceptable to him as a scientific representation of it. Similarly, the model of Nuer political life which emerges in Professor Evan s-Pritchard’s work is a scientific model meaningful to his fellow-sociologists as sociologists, and it is effective because it is potentially acceptable to the Nuer in some ideal situation in which they could be supposed to be interested in themselves as men living in society. The collaboration of natural scientists may from this point of view be seen as developing language enabling certain people to communicate with increasing subtlety about a distinct area of natural phenomena which is defined by the name of the particular science. Their science is, in the literal meaning of the term, their commonsense, their common meaning. lo move from this common sense to the “common sense” of the wider public involves again an act of translation. The situation of social anthropology, or sociology in general, is not at this level so very different. The difference lies in the fact that sociological phenomena are objectively studied only to the extent that their subjective meaning is taken into account and that the people studied are potentially capable of sharing the sociological consciousness that the sociologist has of them. (1961: 88-89; emphasis added)
I have quoted this remarkable passage in full because it states very lucidly a position that is, I think, broadly acceptable to many anthropologists who would otherwise consider themselves to be engaged in very different kinds of enterprise. I have quoted it also because the nature of the collaboration between “author and translator” is neatly brought out in the subsequent reference to the psychoanalyst as scientist: if the anthropological translator, like the analyst, has final authority in determining the subjects meanings-it is then the former who becomes the real author of the latter. In this view, “cultural translation” is a matter of determining implicit meanings-not the meanings the native speaker actually acknowledges in his speech, not even the meanings the native listener necessarily accepts, but those he is “potentially capable of sharing” with scientific authority “in some ideal situation”: it is when he can say, for example, with Gellner, that vox Dei is in reality vox populi, that he utters the true meaning of his traditional discourse, an essential meaning of his culture. The fact that in that “ideal situation” he would no longer be a Muslim Berber tribesman, but something coming to resemble Professor Gellner, does not appear to worry such cultural translators.
This power to create meanings for a subject through the notion of the “implicit” or the “unconscious,” to authorize them, has of course been discussed for the analyst-analysand relationship (e.g., recently in Malcolm 1982). It has not, to my knowledge, been considered with regard to what the cultural translator does. There are, of course, important differences in the case of the anthropologist. It may be pointed out that the latter does not impose his translation on the members of the society whose cultural discourse he unravels, that his ethnography is therefore not authoritative in the way the analyst’s case study is. The analysand comes to the analyst, or is referred to the latter by those with authority over him, as a patient in need of help. The anthropologist, by contrast, conies to the society he wants to read, he sees himself as a learner, not as a guide, and he withdraws from the society when he has adequate information to inscribe its culture. He does not consider the society, and neither do its members consider themselves to be, sick: the society is never subject to the anthropologist’s authority.
But this argument is not quite as conclusive as it may seem at first sight. It remains the case that the ethnographer’s translation/representation of a particular culture is inevitably a textual construct, that as representation it cannot normally be contested by the people to whom it is attributed, and that as a ‘‘scientific text’’ it eventually becomes a privileged element in the potential store of historical memory for the nonliterate society concerned. In modern and modernizing societies, inscribed records have a greater power to shape, to reform, selves and institutions than folk memories do. They even construct folk memories. The anthropologist’s monograph may return, retranslated, into a “weaker’’ Third World language. In the long run, therefore, it is not the personal authority of the ethnographer, but the social authority of his ethnography that matters. And that authority is inscribed in the institutionalized forces of industrial capitalist society (see page 158 above), which are constantly tending to pus the meanings of various Third World societies in a single direction. This is not to say that there are no resistances to this tendency. But “resistance” in itself indicates the presence of a dominant force.
I must stress I am not arguing that ethnography plays any great role in the reformation of other cultures. In this respect the effects of ethnography cannot be compared with some other forms of representing societies — for example, television films produced in tie West that are sold to Third World countries. (That anthropologists recognize the power of television is reflected, incidentally, in the increasing number of anthropological films being made for the medium in Britain.) Still less can the effects of ethnography compare with the political, economic, and military constraints of the world system. My point is only that the process of “cultural translation’’ is inevitably enmeshed in conditions of power — professional, national, international. And among these conditions is the authority of ethnographers to uncover the implicit meanings of subordinate societies. Given that that is so, the interesting question for enquiry is not whether, and if so to what extent, anthropologists should be relativists or rationalists, critical or charitable, toward other cultures, but how power enters into the process of ‘‘cultural translation,” seen both as a discursive and as a nondiscursive practice.
For some years I have been exercised by this puzzle. How is it that the approach exemplified by Gellner’s paper remains attractive to  so many academics in spite of its being demonstrably faulty? Is it perhaps because they are intimidated by a style? We know, of course, that anthropologists, like other academics, learn not merely to use a scholarly language, but to fear it, to admire it, to be captivated by it. Yet this does not quite answer the question because it does not tell us why such a scholarly style should capture so many intelligent people. I now put forward this tentative solution. What we have here is a style easy to teach, to learn, and to reproduce (in examination answers, assessment essays, and dissertations). It is a style that facilitates the textualization o other cultures, that encourages the construction of diagrammatic answers to complex cultural questions, and that is well suited to arranging foreign cultural concepts in clearly marked heaps of “sense” or “nonsense.” Apart from being easy to teach and to imitate, this style promises visible results that can readily be graded. Such a style must surely be at a premium in an established university discipline that aspires to standards of scientific objectivity. Is the popularity of this style, then, not a reflection of the kind of pedagogic institution we inhabit?
Although it is now many years since Gellner’s paper was first published, it represents a doctrinal position that is still popular today. I have in mind the sociologism according to which religious ideologies are said to get their real meaning from the political or economic structure, and the self-confirming methodology according to which this reductive semantic principle is evident to the (authoritative) anthropologist and not to the people being written about. This position therefore assumes that it is not only possible but necessary for the anthropologist to act as translator and critic at one and the same time. I regard this position as untenable, and think that it is relations and practices power that give it a measure of viability. (For a critical discussion of this position as it relates to Islamic history, see Asad 1980.)
The positive point I have tried to make in the course of my interrogation of Gellner’s text has to do with what I have called the inequality of languages. I have proposed that the anthropological enterprise of cultural translation may be vitiated by the fact that there are asymmetrical tendencies and pressures in the languages of dominated and dominant societies. And I have suggested that anthropologists need to explore these processes in order to determine how far they go in defining the possibilities and the limits of effective translation.