Metaphors and practices of translation in anglophone anthropology

Abstract

This article argues that while translation as a metaphor was prominent in anglophone anthropology for most of the second half of the 20th century, the practices of interlingual translation that are often central to the production and circulation of disciplinary knowledge have tended to attract much less attention. The first main section begins by briefly reviewing debates about the idea that anthropology crucially involves ‘the translation of culture’ or ‘cultural translation’. This use of translation as a metaphor to characterize the discipline emerged within both UK and US anthropology in the 1950s, but subsequently became subject to sustained criticism. The focus of the article then shifts in the second and third sections to practices of translation in, respectively, anthropological fieldwork and the international circulation of anthropological texts. More specifically, it examines the difficulties associated with interlingual translation in fieldwork and ethnographic writing; the hidden or invisible ‘translation work’ undertaken in the past by the assistants and spouses of ethnographers; and the translation of anthropological work into other languages. The article concludes by calling on more anthropologists to publish detailed accounts of their own translation practices, arguing that in so doing they will be able to draw on valuable historical antecedents from within the discipline, important collections of anthropological reflections on translation that have appeared over the past 30 years, as well as on relevant work by translation studies scholars and sociologists of translation.

In a thought-provoking article, Anthony Pym (2009) draws a distinction – one that he suggests is not limited to the specific issue of translation – between two kinds of study. The first of these ‘goes out into the world to see what is happening’, while the second ‘sees the world through the authoritative insights of others, mostly as recycled certitudes of theory’ (Pym, 2009: 1). Pym argues in favor of the former and against the latter. In a similar way, I propose in this article that two different ways of addressing translation have tended to be apparent, both historically and to a certain extent still today, within particularly the UK and US branches of anglophone anthropology. ¹ On the one hand, anthropologists have used translation as a metaphor, notably when in the 1950s prominent figures such as E.P. Evans-Pritchard (1951) and Godfrey Lienhardt (1954) presented their discipline as centrally concerned with ‘the translation of culture’ or ‘cultural translation’. Although this approach was subsequently criticized on a number of grounds, traces of it are still to be found in current work. On the other hand, anthropologists have focused on the nature of what Roman Jakobson (1959) referred to as ‘translation proper’ (p. 233), that is, interlingual translation; they have sought to develop procedures and methods relating to translation as a practice. The argument I develop here is that the first of these approaches has tended to be much more visible within anglophone anthropology than the second, and that attempts to combine reflection on metaphors or theories of translation with discussion of practical issues and examples have been surprisingly rare (notable exceptions include Pálsson, 1993b; Rubel and Rosman, 2003b; Severi and Hanks, 2015; Sturge, 2014 [2007]). Building on both recent work on translation within anglophone anthropology and important historical antecedents, and drawing on key contributions from translation studies and the sociology of translation, I call for a theoretically informed discussion of translation practices to become much more widespread within the discipline.

More specifically, the article explores metaphors and practices of translation within anglophone anthropology in the following way. In the first section, I briefly review the idea of anthropology as involving ‘the translation of culture’ or ‘cultural translation’, considering possible reasons why it became popular in the 1950s and also how it was later subjected to sustained criticism. I then introduce some recent work that has informed the argument developed in the rest of the article and mention a few other contributions to theoretical and conceptual reflection about translation in anthropology. This leads me, in the two other sections of the article, to discuss at greater length key practices of translation in anglophone anthropology. I begin with the centrality of interlingual translation to much historical and contemporary anthropology, highlighting claims that, despite this, anthropologists have often remained silent about the specific translation procedures they have adopted in their work. After discussing how this might be explained, I focus on the role of informal and professional translators in anthropological fieldwork, drawing on both historical examples and the results of a recent survey of US-based anthropologists and sociologists. In the final section, I turn to the essential part played by translation in the international circulation of anthropological texts and, more specifically, in the project of ‘world anthropologies’, one of the core objectives of which is to challenge the global hegemony of the English language in disciplinary exchanges. I reflect on examples of the translation of work both from and into English, emphasizing the various challenges involved but also ending by highlighting some encouraging recent initiatives in this connection.

Metaphors of translation in anglophone anthropology

For much of the second half of the 20th century, metaphors of translation occupied a prominent place in anglophone anthropology. In the 1950s, British social anthropologists started to describe their discipline as centrally concerned with ‘the translation of culture’ or ‘cultural translation’. This metaphor proved to be ‘a particularly compelling one’ (Pálsson, 1993a: 18) for US anthropologists too. However, it was to become the focus of sustained criticism in the 1980s and 1990s, leading to the emergence of new metaphors, such as the concept of ‘culture as translation’. All these developments have already been extensively analyzed by other scholars, notably Kate Sturge (2014 [2007]) in the second chapter of her wide-ranging and insightful book Representing Others: Translation, Ethnography and the Museum. It is not my intention, therefore, to provide a very detailed account of them here. Instead, I propose in this first section to highlight the points most relevant to the argument I pursue in the rest of the article about practices of translation.

As Conway (2019: 129) has noted, the term cultural translation has come to be used widely in popular speech as well as in fields such as postcolonial and translation studies. Its genealogy within anglophone anthropology can be traced back to two talks by the British social anthropologists Evans-Pritchard and Lienhardt that were broadcast on BBC radio in the 1950s. In the first of these, Evans-Pritchard (1951) presented his discipline to the listeners in the following way:

As I understand the matter, what the social anthropologist does can be divided into three phases. In the first phase, as ethnographer, he goes to live among a primitive people and learns their way of life. He learns to speak their language, to think in their concepts, and to feel in their values. He then lives the experience over again critically and interpretatively in the conceptual categories and values of his own culture and in terms of the general body of knowledge of his discipline. In other words, he translates from one culture into another. (p. 61)

A few years later, Lienhardt gave a talk on BBC radio entitled ‘Modes of Thought’ in which he also used translation as a metaphor in order to explain the aims and methods of anthropology. After echoing several of the points made by Evans-Pritchard in the passage quoted above, Lienhardt (1954) went on to argue that

[t]he problem of describing to others how members of a remote tribe think then begins to appear largely as one of translation, of making the coherence primitive thought has in the languages it really lives in, as clear as possible in our own. (p. 97)

In other words, both anthropologists characterized anthropology as involving the translation of culture, not simply language.

Over the next three decades, many other anglophone anthropologists followed Evans-Pritchard and Lienhardt in using the metaphor of cultural translation to define their discipline. In a very perceptive comment, however, Talal Asad (1986) later drew attention to the fact that ‘despite the general agreement with which this notion has been accepted as part of the self-definition of British social anthropology, it has received little systematic examination from within the profession’ (p. 143). Evans-Pritchard had of course previously addressed specific questions of translation in his work on the Nuer (see Sturge, 2014: 47–56). There are also several more general passages on the topic in his Theories of Primitive Religion (Evans-Pritchard, 1965: 12–14, 109–110), but it is worth noting that he brings the first of these to a rather abrupt end, declaring, ‘Nor do I therefore discuss the more general question of translation any further here, for it cannot be treated briefly’ (Evans-Pritchard, 1965: 14). ² Although a quotation from the same passage is used as an epigraph in The Translation of Culture (Beidelman, 1971), a collection of essays by other anthropologists published as a tribute to Evans-Pritchard, even this volume contains no detailed presentation and analysis of the metaphor. Rodney Needham’s (1972) study Belief, Language and Experience stands as one of the very few attempts on the part of British anthropologists during this period to explore in some depth the issues raised by the notion of cultural translation.

Despite being the focus of surprisingly little direct discussion initially, the translation of culture metaphor nevertheless quickly became popular among both UK and US anthropologists. How can the idea’s appeal at this time be explained? Noting its origins in public talks that Evans-Pritchard and Lienhardt gave on the BBC, Ulf Hannerz (1993) has suggested that it probably began ‘almost as an off-the-cuff metaphor’ (p. 44), a convenient and accessible way of presenting the discipline to non-anthropologists. For his part, Gísli Pálsson (1993a) has argued that other factors were at work too, ones internal to the discipline, notably a tendency to view culture as ‘something with the properties of language’ (p. 19). Whatever the precise combination of reasons, the metaphor attracted widespread support within anglophone anthropology until the 1980s.

One of the earliest attempts to examine critically the use of the translation metaphor within the discipline was Talal Asad’s (1986) ‘The Concept of Cultural Translation in British Social Anthropology’. It is a rich, complex text that remains an extremely valuable resource for contemporary reflection on translation issues in anthropology. For the purposes of the present discussion, I will focus on the key points Asad makes about power relations and relations between languages. He emphasizes, first, that ‘the process of “cultural translation” is inevitably enmeshed in relations of power – professional, national, international’ (Asad, 1986: 163). More specifically, it is shaped in significant ways by the relationship between the anthropologist and the people with whom they are working, as well as by that between Western and non-Western countries and by industrial capitalism as ‘a dominant force’ in the world. Asad (1986) draws attention, second, to the existence of what he terms ‘the inequality of languages’ (p. 156). Some languages are ‘weaker’ than others – due to wider political-economic relations between the societies concerned as well as to the degree to which they are vehicles for ‘desired knowledge’ – and this leaves them more susceptible to ‘forcible transformation’ in the translation process (Asad, 1986: 157–158). Given this, Asad (1986) suggests 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’ (p. 164). He concludes by calling on anthropologists to consider how these power relations and inequalities might affect ‘the possibilities and limits of effective translation’ (Asad, 1986: 164). This is an issue I will explore in the other two sections below with respect to practices of translation in fieldwork and in the international circulation of anthropological texts.

Many other scholars have criticized the cultural translation metaphor in anglophone anthropology (see, for example, Bachmann-Medick, 2014 [2006]; de Pina-Cabral, 1992; Pálsson, 1993b). Rather than dwell on this work here, I propose instead to draw the first part of my discussion to a close by quickly mentioning some subsequent developments in anthropological research on translation. In addition to the important study by Sturge (2014) mentioned earlier, the past 30 years have in fact witnessed the publication of three substantial edited collections on the topic: Beyond Boundaries: Understanding, Translation and Anthropological Discourse (Pálsson, 1993b); Translating Cultures: Perspectives on Translation and Anthropology (Rubel and Rosman, 2003b); and, most recently, Translating Worlds: The Epistemological Space of Translation (Severi and Hanks, 2015). I draw on all of these sources in the next two sections.

Before turning to examine different practices of translation in anglophone anthropology, which will be the focus of the rest of the article, it is worth noting, even just in passing, that anthropologists are also currently elaborating new theoretical and conceptual approaches to translation. ³ William Hanks and Carlo Severi (2015), for example, have drawn attention to the multidimensional nature of translation and argued that it should be defined ‘not only as a key technique for understanding ethnography, but also as a general epistemological principle’ (p. 16). Building on this work, Anne-Christine Taylor (2022) has recently presented an original case for viewing ethnography not as the translation of ‘a culture’, but instead as ‘the translation of a translation’, that is, as a translation of the various kinds of translation activity that are a fundamental part of the everyday lives of the people anthropologists study. These are stimulating suggestions, which have a certain amount in common with attempts to develop ‘a translational sociology’ that is ‘centrally defined by an investigation of the key role of translation in society’ (Bielsa, 2023: 1). In short, while the cultural translation metaphor is much less prominent in contemporary anglophone anthropology than it was in the middle part of the 20th century, theoretical and conceptual work on translation is ongoing within the discipline. At the same time, increasing attention has been devoted to different practices of translation, and I will now examine these, in relation first to anthropological fieldwork and then to the international circulation of anthropological texts.

Practices of translation in anthropological fieldwork

Only a few years after the prominent British anthropologists Evans-Pritchard and Lienhardt first used the metaphor of translation in relation to their discipline, an article entitled ‘Problems of Translation and Meaning in Fieldwork’ (Phillips, 1959) appeared in the journal Human Organization. Its author was Herbert P. Phillips, who subsequently became a professor in the Department of Anthropology at the University of California, Berkeley, but at the time was a PhD student at Cornell University. ⁴ Emphasizing the ‘central importance’ (Phillips, 1959: 192) of translation to anthropology, Phillips called, crucially, for a much more explicit discussion of actual translation practices within the discipline. Although rarely cited today by anthropologists working on translation, the article is therefore an appropriate starting point for the examination of fieldwork practices that will be the focus of this second section.

Phillips conducted field research in a Thai village from 1956 to 1958 for a culture-and-personality study that involved thematic apperception tests (TATs), sentence completion tests (SCTs), observation, and the collection of life histories. As he explains, it was the use of SCTs in his particular study that necessitated careful attention to ‘translation problems’ (Phillips, 1959: 184), but, he emphasizes, the latter arise in all research where different languages are involved. He goes on to note, however, that

Despite the fact that cultural anthropology is the one discipline whose practitioners must almost all translate, comparatively little attention has been given to analyzing the translation process as such. [. . .] for those ethnographers who do not work on linguistic problems, and they are the majority, the question of translation procedures has often seemed wrapped in a conspiracy of silence. (Phillips, 1959: 186)

Almost immediately, Phillips acknowledges that Malinowski, Kluckhohn, and a few other anthropologists had already discussed the process of translation at some length, and he includes a very interesting review of their work. Nevertheless, he suggests that ‘avoidance’ (Phillips, 1959: 186) of translation issues appears to be the approach generally adopted within the discipline.

As a doctoral researcher at the start of his academic career, Phillips was perhaps wary of provoking a negative reaction on the part of more senior colleagues by reflecting directly on the possible reasons for the ‘conspiracy of silence’ he had identified. Instead, he proceeds to provide a meticulous account of the translation procedures he followed during his own field research in Thailand. Before doing so, he observes that a number of factors can affect ‘accuracy’ of translation, not all of which may be under the anthropologist’s control (Phillips, 1959: 184). Among these are the researcher’s own degree of proficiency in the language(s) concerned, the amount of time they are able to spend in the field, the size of their budget, and the availability in the field site of skilled interpreters and translators. Phillips (1959: 187–191) then describes in impressive detail how he worked with several Thai translators (who are, however, not named) on the translation of the sentence completion test items and responses, specifying who undertook which specific task and estimating how long this took to complete. After providing a more general summary of the ‘translation instructions’ (Phillips, 1959: 191–192) that guided his research, he concludes by stating that ‘If this paper stimulates other field workers to be more explicit in describing their translation procedures, or better, to develop more systematic and precise methods of translation, it will have achieved its end’ (Phillips, 1959: 192). It does not seem, as I will now discuss, that Phillips’ aim was realized to a great extent in the subsequent period. Nevertheless, his article remains an exemplary attempt to provide a full account of the translation process during field research, including a clearly stated rationale for each of the specific procedures and practices adopted. As such, it is still likely to be of considerable value to contemporary anthropologists interested in translation.

Translating Cultures: Perspectives on Translation and Anthropology (Rubel and Rosman, 2003b), one of the few edited collections on the topic available in English, appeared 40 years after the publication of Phillips’ article. It is striking that in their introductory chapter, the editors echo Phillips’ earlier assessment, writing that despite its centrality to the methods and aims of the discipline ‘curiously, the role that translation has played in anthropology has not been systematically addressed by practitioners’ (Rubel and Rosman, 2003a: 1). Already in the 1990s, other scholars had noted the relative lack of sustained, explicit discussion on the part of anthropologists of what Roman Jakobson (1959: 233) had termed ‘translation proper’, that is, interlingual translation (see, for example, Finnegan, 1992: 194; Pálsson, 1993a: 14). Significantly, the most in-depth and comprehensive study to date of translation in British anthropology (although it also covers relevant developments in US branches of the discipline) was written not by an anthropologist but by a translation studies scholar. This is Sturge’s (2014 [2007]) book Representing Others: Translation, Ethnography and the Museum, which was mentioned in the previous section.

Almost half a century after Phillips, Sturge (2014 [2007]: 14) too observes that there is a remarkable ‘silence’ about matters of language and translation in anglophone anthropology. After exploring how anthropologists have used the ‘translation of culture’ and ‘cultural translation’ metaphors, she makes the following important points:

But something has gone largely unmentioned in the account for far, especially in the more metaphorical explorations – something, in fact, one wouldn’t expect to find missing in a discussion about translation, and that is language difference. This is where translators may well baulk at the metaphorical uses of the world ‘translation’, which sometimes proliferates as a general label for any kind of mediation, change or confrontation with difference, and at worst for any kind of communication breakdown, failure and loss. There is a striking lack of attention in practical handbooks of anthropology to the requirements of translation. Even when issues of representation are covered in detail, such textbooks often subsume translation under their advice on transcription technique or do not mention it at all. (Sturge, 2014: 13)

An up-to-date survey of textbooks and handbooks of anthropology would be needed to establish whether Sturge’s final comments here remain true. ⁵ It would also be interesting to investigate the extent to which, if at all, detailed, practical guidance on how to approach questions of translation is included in training courses for field workers. Although not too much should be read into a single survey, a recent analysis of 107 syllabi for ethnographic methods courses at US universities contains not a single mention of translation (Ruth et al., 2022). Given the frequently multilingual nature of contemporary ethnographic research, this is somewhat surprising.

According to Sturge (2014: 14), silence about translation acts as a ‘repression’ of discussion not only of the effects of the specific translation choices made by the anthropologist, but also of wider questions of power and inequalities between languages. These are also the issues that Asad, with whose work Sturge engages, raised in his criticism of the idea of cultural translation in British social anthropology, as covered in the previous section. Like Phillips before her, Sturge (2014) acknowledges that there are ‘major exceptions to the silence’ (p. 14) about translation in anglophone anthropology. As well as Malinowski’s essay on translation in the second volume of Coral Gardens and Their Magic (1935: 11–22), which Sturge analyses in depth (2014: 24–29), she points to relevant work in the ethnography of speaking (pp. 14–15) and in studies of Native American oral literature (pp. 100–128). All of these are valuable resources for anthropologists interested in reflecting on different translation practices, in ways that Sturge helpfully outlines. Nevertheless, with the notable additional exceptions of the collections edited by Pálsson (1993b), Rubel and Rosman (2003b) and, more recently, Severi and Hanks (2015), not forgetting Sturge’s own study, there still appears to be a rather surprising lack of in-depth discussion in anglophone anthropology of translation practices and procedures. Instead, as Brinkley Messick (2003: 181) has observed, commentary on such matters tends to be confined, where it exists at all, to the ‘preface’ of a monograph or to a brief ‘note on translation’ separated from the main body of the text. A further illustration of this point is the fact that the rather acrimonious exchanges that took place in the late 1960s over the first English translation of Claude Lévi-Strauss’ (1962) book La pensée sauvage were conducted not in the main pages of the journal Man, but in its ‘Correspondence’ section (see Bialecki, 2021; Leavitt, 2021).

At regular intervals over the past 60 years, then, scholars have pointed to a tendency on the part of anglophone anthropologists to avoid describing and analyzing their own translation practices. How can the disciplinary silence about such matters be explained? Rubel and Rosman (2003a) suggest that it may be a consequence of ‘the ongoing internal dialogue about the nature of the discipline’ (p. 1), that is, whether anthropology belongs with the sciences or with the humanities. For her part, Sturge seems to imply in the passage quoted earlier that the prominence of translation as a metaphor within ‘classic’ British social anthropology resulted in comparatively less attention being paid to the more practical aspects of the translation process. Interestingly, Sherry Simon has developed an argument along these lines with respect to Anglo-American gender and cultural studies. In these fields, she maintains, the ‘material realities of translation’ have tended to be obscured by the ‘highly metaphorical language’ used to describe it (Simon, 1996: 135). This may also have been the case for anthropology, although as I indicate in the next section, important attempts have been made over the past two decades to bring such neglected issues to the fore.

It is also possible, as Sturge (2014: 14) notes, that anglophone anthropologists have tended to remain silent about translation practices in order to avoid potentially awkward questions about their own degree of socio-linguistic competence or proficiency that could undermine their professional status and authority. A similar argument has been made in recent years to explain the comparable silence that exists within the discipline over the learning of new languages (see, for example, Borchgrevink, 2003; Gibb and Danero Iglesias, 2016; Gibb et al., 2019; Tanu and Dales, 2016), although the issue had of course been previously raised by scholars such as Edwin Ardener (1971: xiv–xviii) and Maxwell Owusu (1978). Concern about the effects of acknowledging limited fluency in a field language might well have made anthropologists reluctant to discuss their own translation practices in an open and honest manner.

As far as British social anthropology is concerned, finally, an additional factor may be the relatively marginal place of language and linguistics within the discipline, historically and arguably still today. Hilary Henson (1971: 3) showed, for example, that language remained ‘a peripheral interest’ during the discipline’s formative years from 1850 to 1920. On the basis of a detailed study of work produced in this period, she concluded that ‘[e]arly British anthropology [. . .] never produced anything that remotely resembled a sociological theory of language’ (Henson, 1971: 24) and that it was characterized by ‘a general ignorance of all the deeper implications of language for a study of culture’ (Henson, 1971: 25). Despite the importance subsequently attached to language by Malinowski, Edwin Ardener (1971) was later to report that when in 1969 the Association of Social Anthropologists of the Commonwealth organized a symposium on language ‘the number of full members of the Association who felt qualified to offer formal papers was still very few’ (p. xi). ⁶ Even as recently as 2007, Ben Rampton (2007) could point to a range of evidence suggesting that it was still the case that ‘British anthropology tends to overlook language’ (p. 584). Whether or not this is true today, it may have contributed in the past to a lack of prominence within the discipline of questions of language and translation.

Whatever the precise reasons for the silence that has tended to surround practices of translation in anglophone anthropology, one of its undoubted effects has been to prevent full recognition of the ‘translation work’ that is commonly undertaken in the field by people other than the anthropologist. In ‘Anthropology’s hidden colonialism: Assistants and their ethnographers’, Roger Sanjek (1993) highlights the crucial roles played by local assistants and key informants in the history of the discipline. As he shows, anthropologists and other ethnographers have often relied on members of the societies they have studied for many different kinds of assistance, including interpreting and translating (Sanjek, 1993: 13, 14). He argues that ‘[e]thnographers and assistants together made anthropology’ (Sanjek, 1993: 16), but that this is rarely reflected in anthropology textbooks and other accounts of the development of the discipline. In a similar vein, Akhil Gupta (2014) has written more recently that the lack of attention to the role of research assistants in the field ‘leaves a large gap in anthropology’s self-reflexivity’ (p. 394). In addressing this, as Townsend Middleton and Jason Cons (2014) have emphasized, there is an opportunity to reflect on the potential transformation of traditional ethnographic practice.

Against this background, it is instructive to consider an example of how the translation work of a local assistant was progressively rendered invisible in an anthropological ‘classic’. In an illuminating account of translation practices during the ‘colonial’ period of British social anthropology, Sturge (2014: 45) discusses A.R. Radcliffe-Brown’s (1922) monograph The Andaman Islanders: A Study in Social Anthropology. As she explains, Radcliffe-Brown briefly acknowledges in the preface to the first edition of the book that an Andaman Islander acted as an interpreter for him during his field research. The man is not named, however, and although two photographs of him are reproduced in the text the captions to these do not indicate the role he played as an interpreter. Even this rather partial or incomplete acknowledgment of the translation work undertaken by the Andaman Islander, though, disappears in the second edition of the book. The man’s picture remains, but the preface to this later edition does not mention that Radcliffe-Brown had the assistance of an interpreter. As Sturge (2014) concludes, ‘The relationship between anthropologist and informant is structured in such a way that this translator (i.e. the Andaman Islander) is almost entirely invisible – so much so that no-one reading the second edition could more than guess at his existence’ (p. 45, italics in original). This is unlikely to have been an isolated case.

If the failure to acknowledge the essential contribution made by local assistants to anthropological research in the past was, in Sanjek’s words, a form of ‘hidden colonialism’, this could be accompanied by a hidden sexism too. Discussing research conducted between the 1910s and 1970s, Barbara Tedlock (1995) has highlighted the fact that the wives of (male) anthropologists sometimes worked with their husbands in the field. These women were usually not trained ethnographers themselves initially, although some later turned professional. In many cases, accounts of the research subsequently published by their husbands failed to acknowledge explicitly the substantial assistance they had provided or even at times their presence in the field. One of the women Tedlock mentions is Carobeth Laird, who was married to the linguist John Peabody Harrington. As Tedlock (1995) observes, ‘Although he never officially acknowledged her help, Caroline materially assisted him by hand copying and typing native vocabulary, translating from the Spanish, gathering plant samples, and cooking, as well as driving for both him and his informants’ (p. 270). Laird was most probably far from being the only wife of a male ethnographer during this period who acted as a translator while they were both in the field, and potentially afterwards too, but whose contribution failed to be recognized in reports on the research.

The historical examples discussed by Sanjek, Sturge, and Tedlock are important reminders that in the past the anthropologist was frequently not the only translator involved in the research. What is known about current fieldwork practices and relationships? I will end this section by briefly discussing two recent publications that provide some initial answers to this question. The first is ‘Emergent Collaborations: Field Assistants, Voice and Multilingualism’ (2019) by Susan Frohlick, an Anglo-Canadian anthropologist, and Carolina Meneses, the bilingual Costa Rican field assistant Frohlick recruited to work with her on a research project about youth, sexual health, and global tourism. Drawing partly on the article by Gupta (2014) mentioned earlier, this is a fascinating exploration of questions of socio-linguistic competence, translation, and forms of collaboration between anthropologist and field assistant. A co-written piece of work, it explicitly sets out to challenge the frequent ‘voicelessness’ of field assistants (and translators) in conventional ethnographic writing. One notable way it achieves this is by including long extracts from Meneses’ fieldnotes and reflections. The authors write that ‘While initially hiring a bilingual field assistant was a remedy for the anthropologist’s limited Spanish, without recognizing it at the time we ultimately became collaborators through our entangled co-presence in the field’ (Frohlick and Meneses, 2019: 31). Central to this, as they explain, were ‘the embodied politics and emotions of language’ (Frohlick and Meneses, 2019: 32). Illustrating one of the points emphasized by Sanjek (1993), Frohlick and Meneses (2019: 39) also show directly how anthropologist and field assistant co-produced knowledge, with Meneses shaping the research through ‘her continual feedback’ as well as her role as translator/interpreter. In all these different ways, the authors make visible aspects of fieldwork that have often been obscured in more conventional accounts, and in so doing offer a valuable, reflexive account of their deliberate attempt to develop a new kind of research relationship.

Quantitative as well as qualitative data about current research practices is provided by the second publication I will consider here: a recent survey of US-based anthropologists and sociologists that investigated levels of proficiency in field languages and the use of interpreters/translators. Analyzing the 913 anonymous responses they received, Katarzyna Sepielak et al. (2023) report that ‘over 65.3% of our respondents, and especially 68.9% of anthropologists [. . .] used a translator at least in one of their fieldsites’ (p. 512). One of the reasons given by respondents for needing the assistance of interpreters/translators in the field was involvement in short-term comparative projects, where there was insufficient time for the researcher to attain the necessary level of proficiency in the language(s) concerned. Interpreters/translators were often also necessary when research was conducted in fieldsites where multiple languages were spoken (Sepielak et al., 2023: 515). In addition, the survey contained a question designed to throw light on the ‘profile’ of the person who interpreted/translated for the researcher. The 558 answers provided by anthropologists revealed that over a half had relied on the paid or non-paid services of a local person who was not a professional interpreter/translator; just under one in ten had used a paid professional (mostly a local person but in a few cases a member of the outside research team); and in over a third of the cases interpreting/translating was carried out by a member of the research team who was proficient in the language but not a professional interpreter/translator (Sepielak et al., 2023: 519, Table 5).

To illustrate their findings, Sepielak and her colleagues reproduce a series of lengthy extracts from written responses to their survey. These include one that suggests that the wives of some (male) anthropologists may still contribute in crucial ways to their husband’s work today. In the section of the article entitled ‘Profile and evaluation of interpreters’, the authors attribute the following remarks to a respondent they identify as ‘Female–White–Anthropology–30+–Full_Professor–Ecuador–Peru’:

The worse in my opinion is the white male researcher who does not know the local language, but has married a woman who is a native of the country (but not the region) and uses her as his translator, project manager, etc. It’s very common in some subfields, and it establishes a race/gender/nationality hierarchy that reinforces local power dynamics and guarantees that the researcher is only going to hear hegemonic points of view (i.e., older males in the community probably approve, young women are only going to say and do what they think powerful men want to hear). (Respondent comment quoted in Sepielak et al., 2023: 520)

It is not clear whether this kind of practice was reported by other respondents or how widespread it might be. Nevertheless, as the authors state, this concerned respondent’s comments undoubtedly raise important ethical and methodological issues.

The survey conducted by Sepielak and her colleagues, although restricted to US-based academics, provides valuable insights into the use of interpreters/translators in contemporary anthropological research. One of the main conclusions the authors draw from their findings is that ‘translation services are employed in a haphazard manner, frequently relying on non-professional interpreters, non-paid locals but also colleagues and family frequently not even recognized in their roles’ (Sepielak et al., 2023: 522). In other words, the survey highlights the existence of an ‘informal economy in translation’ (Sepielak et al., 2023: 522), and the power relations, possible conflicts of interest, potential (or real) exploitation and concern over the quality of translation associated with this. It paints a worrying picture of the ‘material realities’ of translation, drawing attention to practices that should give serious cause for concern.

Similar kinds of survey could usefully be conducted in other countries and extended to include a number of key questions not covered by the US survey. For example, it would be interesting to ask if and how translation work undertaken in the field by assistants is subsequently acknowledged in research reports and other publications. Also, how often are interpreters/translators involved as co-authors? In addition, more detailed information could be collected about the kinds of translation provided by those carrying out such work, the different contexts in which they translated, and the specific translation practices they adopted and the rationale for these. Research of this kind would help to build up a fuller picture of the use of interpreters/translators in anthropological research today.

Translation practices and the international circulation of anthropological texts

In a recent article, Esperança Bielsa (2022) has argued for ‘a translational sociology’, pointing out, among other things, that translation is central to the production and circulation of sociological knowledge. As far as the production of knowledge is concerned, translation can be a crucial part of field research in sociology, as the previous section showed is also the case for anthropology. Another way translation becomes part of ‘the ordinary work’ of many sociologists, Bielsa explains, is when they translate their own writing or that of other sociologists into other languages. She discusses the example of David Frisby, whose activity as a translator of Simmel’s work into English significantly shaped the development of his own ideas (Bielsa, 2022: 413–414). Although this kind of translation work is essential for the international circulation of ideas and knowledge and has contributed to the existence of a transnational ‘sociological tradition’, it tends not to be particularly visible or valued. Instead, as Bielsa (2022) states, ‘Translating sociology is still primarily viewed as second-order, derivative work not worthy of attracting critical attention or of being considered as sociological work in its own right’ (p. 413). This is true also of ‘self-translation’, a practice whereby multilingual authors translate their own work into other languages, which increasingly often means English, given the dominance of the latter as the language of international scientific exchange and communication (Bielsa, 2022: 414).

All these points are valid with respect to the production and circulation of anthropological knowledge too. As I explored in the previous section, translation usually plays a crucial role in the process of knowledge production during anthropological fieldwork and is bound up with key issues such as the anthropologist’s degree of socio-linguistic competence and their frequent reliance on the assistance of professional or informal interpreters/translators. However, it has also been at the heart of debates about how to transform anthropology by undermining epistemological, political, organizational, and institutional hierarchies within the discipline on a global scale. A key initiative in this regard has been the ‘World Anthropologies’ project (Ribeiro and Escobar, 2006), which has criticized Eurocentric perspectives within the discipline and sought to challenge the dominance of Anglo-American anthropology. One of its founders, Gustavo Lins Ribeiro (2005), emphasized the importance of translation at the outset:

We need to foster the visibility of non-metropolitan works of quality and enhance our modes of exchanging information. Translation of different anthropological materials into English is important to help diversify knowledge of the international production of anthropology. But unidirectional translation is not enough. If we want to avoid linguistic monotony, we also need to increase the quantity of heterodox exchanges and translations. German anthropologists should be translated into Japanese, Mexicans into German, Australians into Portuguese, Brazilians into Russian, and so on. (p. 5)

Starting from a distinction between ‘dominant anthropologies’ (i.e. France, the United Kingdom, and the United States) and ‘other anthropologies/anthropology otherwise’, the project has attempted to develop ‘new conditions of conversability’ (Ribeiro, 2014: 494) of a more equal or horizontal nature between anthropologists located in different parts of the world and coming from different traditions. It has also sought to promote ‘a plural landscape of world anthropologies’ (Restrepo and Escobar, 2005: 100) in opposition to the idea of a single tradition emerging out of the West. To this end, translations of work have been made available in different languages, notably through an electronic journal on the website of the World Anthropologies Network (http://www.ram-wan.net/en_US/).

The practical difficulties associated with translating anthropological texts into other languages are the subject of a wide-ranging discussion in the ‘World Anthropologies’ section of a recent edition of the journal American Anthropologist. In the main article, entitled ‘Publishing Translations’, Michael Chibnik (2003, 2019) provides a thought-provoking account of the process of publishing a Spanish translation of his book Crafting Tradition: The Making and Marketing of Oaxacan Wood Carvings (2003) in Mexico. As he explains, he wanted the results of his research to be available in Spanish so that they would be accessible to various audiences: the Mexican artisans whose work he had studied; Oaxacans and other Mexicans interested in the topic; and university teachers, students, and museum directors in Mexico who might not be able to read the English-language edition (Chibnik, 2019: 207). The translation and publication of a Spanish version of the book proved, however, to be a time-consuming, expensive, and complicated undertaking. In the end, he was only able to publish the Spanish translation in 2017, several years after he had retired and having spent US$10,000 of his own money to cover translation and book subvention costs (Chibnik, 2019: 209).

Chibnik (2019) acknowledges that this is just one specific case, and that the translation of ‘canonical anthropological books’ with high sales in their original language is likely to be less difficult than that of ethnographic monographs with more ‘limited sales and readership’ (p. 209). In reflecting on his experiences, however, he makes the point, also highlighted by Bielsa, that translating social scientific texts into other languages is not generally regarded as a valuable activity. More specifically, he notes that

I would have received few, if any, professional benefits from the translation if it had been published before my retirement in 2015 [. . .]. When I was a faculty member at the University of Iowa, criteria for promotion and tenure placed almost no emphasis on my making my research results more available to Latin American audiences. Working on this project would have been regarded as a distraction from the more important work of publishing in prestigious Anglophone journals and university presses. (Chibnik, 2019: 209)

In other words, the current orientation of institutional priorities and career structures within universities creates little incentive for anthropologists to spend time self-translating their work or trying to get it translated into languages other than English – in fact, it often actively discourages this. The implications of Chibnik’s experience are drawn out further by Jorge Hernández Díaz (2019), one of the other anthropologists invited by the journal to comment on the article. Díaz (2019: 213) argues that trying to make a text available and accessible to the people with whom an anthropologist worked is ‘a political action’, but one that clearly depends on forms of both institutional and economic power – seniority/security of tenure and financial resources, respectively – if it is not to have ‘negative repercussions’ on the professional career of the individual concerned. This is a serious problem, he continues, because it ‘inhibits the political commitment of the discipline as a whole’ (Díaz, 2019: 213).

Díaz (2019: 212) also draws attention to the combination of skills required to undertake translation work, and this is an issue raised by several other anthropologists in their comments on Chibnik’s article. Ulf Hannerz, for example, points out that while Chibnik was in a position to check the Spanish translation of his book, those whose work is translated into languages they do not speak lack such control. This creates the potential for errors to pass unnoticed, especially if the translator has the necessary linguistic skills but only ‘a limited knowledge of anthropological concepts or of the wider context of what the writing is about’ (Hannerz, 2019: 211). Another interesting perspective on this is offered by Peter-Jazzy Ezeh in an interview with Virginia Dominguez and Emily Metzner that is included in the same ‘World Anthropologies’ section as Chibnik’s article. Responding to a question from Metzner about whether anthropologists have ‘something of a professional duty’ to use their skills to translate anthropological and non-anthropological texts, Ezeh (2019) remarks,

I think that translation skill is important to anthropological practice. [. . .] Do anthropologists have a professional duty to carry out or encourage translations of important works that are related to their practice? Yes. Does everyone have translation skill? Of course, no. [. . .] And if I had my way, I would make acquisition of this skill an integral part of the training in anthropology. (p. 217)

Ezeh’s last point here about anthropological training is, in my view, extremely important. Although care should obviously be taken not to draw overly firm conclusions from a single survey, Ruth et al.’s (2022) study of ethnographic methods syllabi referred to in the previous section strongly suggests that training in translation skills is not currently a prominent component of such courses. If this is the case then it is, I think, a missed opportunity. The kind of training Ezeh calls for here could make possible the translation of more anthropological texts into other languages, assuming, of course, that the other obstacles identified by Chibnik can be overcome. Crucially, though, it could also enhance reflection on practices of translation in fieldwork and help to make explicit and in-depth discussion of translation procedures and choices more common than it tends to be today.

As Ayşe Çağlar (2019: 213) comments, Chibnik’s account is particularly valuable as it documents the problems that seeking to translate from rather than into English may encounter, something which, she claims, has been ‘rather neglected’ in debates about world anthropologies. Together with all the other contributions published in the same ‘World Anthropologies’ section in American Anthropologist, it raises many important issues about language, translation, and inequalities and hierarchies within anthropology as both a global discipline and a profession. As Virginia Dominguez (2019) writes in her ‘Foreword’ to the section: ‘That language inequality is so palpable these days in anthropology, despite its progressive and anticolonial commitments, is sobering. What to do about it is for all of us to figure out’ (p. 205). In addition to the initiatives already launched as part of the World Anthropologies project mentioned earlier, other actions could be envisaged. For instance, the development of translation skills could be incorporated as a core component of anthropological training much more widely than it currently appears to be. At an institutional level, criteria for promotion and tenure in anglophone universities could be changed so that translation of work into languages other than English, especially but not limited to the languages of the places where the research was conducted, is valued – and rewarded – as a key aspect of an anthropologist’s professional activity. Among other things, this would provide an additional incentive for anglophone anthropologists to try and publish original work in non-English language journals, something they tend to avoid doing at the moment (Saillant, 2015: 146).

There are in fact encouraging signs already of the emergence of what has been described as ‘a broader trend towards cross-linguistic anthropological practice’ (JASO, 2022). In 2022, the Journal of the Anthropological Society of Oxford (JASO) announced the launch of a new translation series in order to make available in English work originally written in other languages. The rationale for this initiative is clearly stated as follows:

The global hegemony of English as a scholarly language, in addition to the everyday Anglo-centrism of Anglo-American anthropology departments, poses a serious epistemic barrier to an even scholarly dialogue within the worldwide anthropological community. We are well aware that translation, on its own, cannot be the panacea to cure all structural imbalances in academic knowledge-production. However, it is a small step towards acknowledging the limitations of Anglo-American conversations that rarely feature non-Anglo-American scholars. (JASO, 2022)

In a similar way, the editors of the Cambridge Journal of Anthropology wrote recently that ‘we do not currently have the resources to publish in anything other than English – but we are aiming to shift the balance in various ways’ (Chua et al., 2023: v). They explain that among the latter is the introduction of new section of the journal entitled ‘Re-Reviewed’, which will include reviews of books not yet translated into English. Other English-language journals, notably HAU: Journal of Ethnographic Theory, already publish translations of articles periodically. It is worth noting here too that in association with the Society for the Anthropology of Europe (SAE), a section of the American Anthropological Association (AAA), Berghahn publishes a book series called ‘European Anthropology in Translation’ (https://www.berghahnbooks.com/series/european-anthropology-in-translation). This focuses on texts relating to the anthropology of Europe previously published in languages other than English. Twelve volumes have been published in the series since it was launched in 2007.

Conclusion

This article began by referring to a distinction drawn by Pym between two kinds of study. Defending an approach to translation centered on ‘actual practice’ (Pym, 2009: 3) as opposed to ‘the repetition of theoretical propositions without empirical application’ (Pym, 2009: 1), he encourages us to ‘go out into the world to see what people think translation is, who translators are, how translations are actually used, [. . .] what the political effects of translation have been, and what the actual alternatives to translation are’ (Pym, 2009: 9). Anthropologists have of course often done exactly that over the course of the discipline’s history, as well as using translation as a metaphor to describe the nature of their work and contributing to the development of theories and concepts of translation. However, I have suggested here that the use of translation as a metaphor has tended to be more visible in anglophone anthropology, historically, than discussion of actual translation practices. Even today, anthropologists are, with some notable exceptions, often silent about their own procedures and practices of translation during fieldwork and in ethnographic writing. It may be true of anglophone anthropology, as Simon claimed with respect to Anglo-American gender and cultural studies, that attention to ‘the material realities’ of translation has been obscured by more metaphorical approaches. A continuing reluctance to discuss such matters explicitly may also be attributable, at least in part, to the wider question of the anthropologist’s degree of socio-linguistic competence and concerns about losing professional status or authority.

Be that as it may, it is to be hoped that more anthropologists will start to publish detailed accounts of their own translation practices, relating these to specific theoretical and conceptual perspectives. In so doing, as I have aimed to show here, they will be able to draw on valuable historical antecedents, the important collections of anthropological reflections on translation that have appeared over the past 30 years, as well as on relevant work by translation studies scholars and sociologists of translation. One of the most damaging effects of the silence about translation within anglophone anthropology historically has been to make invisible the crucial translation work often undertaken by assistants – and sometimes also by the spouses of male anthropologists – in the field. Over the past three decades, though, the key role played by assistants and others in the production of anthropological knowledge has increasingly been the focus of explicit attention and recognition. This is a welcome development, particularly given evidence indicating that many contemporary US-based anthropologists (and the same is likely to be true elsewhere) are reliant on interpreters and translators when conducting field research. Finally, I have emphasized here how practices of translation are central to the international circulation of anthropological texts, and to attempts to challenge the global hegemony of the English language in disciplinary exchanges. Unfortunately, the issues Asad raised in his criticism of the cultural translation metaphor in British social anthropology – power relations and inequalities between languages – remain all too real in this connection too, despite some recent positive initiatives concerning translation in anglophone journal and book publishing.