You are on Érudit's new platform. Enjoy! Switch to classic view

Lessons from Chinese History: Translation as a Collaborative and Multi-Stage ProcessLeçons de l’histoire chinoise : la traduction comme pratique collective aux étapes multiples

  • James St. André

…more information

  • James St. André
    University of Manchester, Centre for Translation and Intercultural Studies, School of Languages, Linguistics and Cultures, Oxford Road, Manchester M13 9PL UK

Article body


There has always been a tendency for theorists to construct models which simplify complex processes while at the same time laying claim to universal validity. This tendency is understandable, given the fact that a theory which does not successfully explain all phenomena in any field can hardly be conceived of as comprehensive, while an overly detailed model is impractical to apply. The rise in prominence of scientific discourse from the eighteenth century onward, the invention of modern statistical methods in the nineteenth century, and the emergence of centralized funding bodies along with the peer-review process in the twentieth century have all exacerbated this tendency.

This tendency, when combined with longstanding Orientalist attitudes toward “the Rest” of the world by “the West” has led to a situation whereby most, if not all, theoretical models and paradigms in a wide range of humanities and social science disciplines have been developed based on a narrow range of data and case studies gathered from a handful of European and North American countries. To take only one region as an example, scholars studying East Asia have for decades complained about under-representation of languages, cultures, societies, and histories from these long and rich traditions in major journals in many fields. A study of a large number of sociological journals over a period of decades, for example, found only a handful of articles relating to Japanese society (see McSweeney, 1999). Zhang Longxi (1998) has written in a similar vein regarding Comparative Literature, and the Bernheimer report, commissioned in the 1990s by the American Comparative Literature Association, came to similar conclusions (see Bernheimer, 1993). Likewise, statistically speaking, the bulk of articles in journals such as Target, META, TTR, and The Translator concern translation between a handful of major Western languages, with little representation of other cultures and societies. While the past five years have seen increased representation of a broader range of studies,[1] much more could be done.

The result of the confluence of these two factors is that theoretical models, which are generalizations based on European society, history and culture, are often applied rather mechanically to other peoples and places. Recently Maria Tymoczko has discussed this trend as it pertains to Translation Studies in great detail (2007, pp. 3-8).

As someone who is both an Asianist and an historian, I would like to use case material from two important periods of translation activity in China, that of the translation of Buddhist texts and the late Qing, to challenge two commonly held assumptions in Translation Studies today. These assumptions are first, that translation is essentially a solitary act; and second, that relay translation is an evil that must be tolerated, and therefore is seldom discussed and certainly never to be praised. I will show how, historically speaking, translation has often been a collaborative act in China, and that relay translation has played a crucial role as well. I will also suggest that, although studies of such phenomena are relatively rare, if we look more carefully at translation in “the West,” these two trends are also observable in many periods, for example, Spain in the twelfth and thirteenth centuries (Foz, 1998, pp. 83-95). The Chinese examples should therefore make us reconsider these two assumptions and problematize current models of translation and historical studies. Finally, I suggest that there are also pedagogical implications.

Chinese Translation as a Collaborative Act

I am not the first person to notice the fact that translation in Chinese history has often been a collaborative act. Eva Hung, for example, has recently described the elaborate process of Buddhist scripture translation in English (Hung, 2005, pp. 84-91; see also Hung, 2006).

However, most scholarship on the topic has been published in Chinese, and so it has perhaps not been impressed upon a Western audience both the percentage and scale of collaborative translation in China before the twentieth century. Due to limitations in space, I will only be discussing two cases: the translation of Buddhist texts over a period of roughly ten centuries (150 CE-1150 CE), and the much shorter, but equally obvious case of the late Qing (1860-1911 CE).

In the approximately ten centuries during which Buddhist sutras were translated into Chinese, tens of thousands of texts were translated and re-translated. A wide variety of texts were translated, including cautionary tales, chantefables, charms, and poetry, although the main category was sutras (religious texts). The introduction of this material had an enormous impact on, not just religion, but virtually every other aspect of Chinese culture as well; including, intriguingly, theories of language and theories of translation (Ma, 2006, pp. 114-123). It is commonly accepted that there were no theories of translation before the introduction of Buddhism; that it was through a consideration of some of the problems arising from translating sutras into Chinese that, for the first time, the Chinese language is described as being tonal in nature; and that Chinese scholars first developed a method of recording the sound of written characters in response to the need to reproduce as closely as possible the sound of Buddhist proper names, to name just a few facets of Buddhism’s far-reaching influence (Ma, 2006, pp. 115-116).

Although there were variations over the course of the centuries, notably swings between more literal and more free translation as in the West (Hung, 2005, p. 91), methodologically speaking, there was one constant: translation was virtually always a cooperative undertaking. Translation was a communal act, normally of monks in a monastery setting, consisting of several steps. First, a monk who understood the source text would recite the sutra, followed perhaps by explication of its meaning. This would be followed by an interpretation of the sutra in Chinese (not necessarily by the same person), a discussion among all those present, and the recording of a written draft of the translation that resulted from the discussion. This written text would then be subject to revision, confirmation with the original interpreter, and sometimes additional rounds of group discussion and revision. In later periods, when the text being translated had often already been translated by previous generations, comparison with earlier translations was also part of the process, as was collation and comparison of multiple copies of the original sutra if available. The process could, in some cases, take several years. At a minimum, there would be the interpreter, an audience (which might include laypersons as well as monks), and a recorder/scribe. These sutra translation sessions at times included up to one thousand people, all of whom had the right to participate in the translation process for a single sutra (see Ma, 2006, pp. 65-137; Hung, 2005, pp. 84-91).

The collaborative nature of sutra translation may have originated in the fact that the interpreter in the first few centuries was usually a foreigner. The two most famous interpreters in this early period, for example, An Shigao (d. 172 CE) and Kumarajiva (350-409 CE) were both from Central Asia. However, many scholars attribute the process, not to issues of linguistic competence, but rather to the Buddhist tradition of oral explication of sutras in public, with debate and discussion of the meaning of the sutra as part of the process (Ma, 2006, p. 93). Even in later centuries, someone as famous as the Chinese monk Tang Xuanzang (602?-664) (who journeyed to India, spent eighteen years studying Sanskrit and Pali texts there, and eventually returned to China to spend the rest of his life translating) also worked with the committee format. Ma Zuyi notes that the translation process employed by Xuanzang was actually more elaborate than that used by the non-native Kumarajiva; records indicate eleven separate tasks, many of which could involve more than one person, whereas Kumarajiva had listed only eight (Ma, 2006, pp. 98-99). Thus even the leading figure of his day in a much later period, fluent in both Sanskrit and Chinese, did not feel he could work alone on sutra translation. This elaborate process, or slightly simplified versions of it, continued in use right to the end of translation from Sanskrit in the twelfth century. Whatever its origins, then, collaboration became and remained the norm in China. Moreover, in terms of cultural interaction, the translation of Buddhist texts had by far the most lasting impact on China; all other translation activity until the nineteenth century pales in comparison.[2]

After China’s defeat in the Second Opium War (1860), the Treaty of Tianjin stipulated that all diplomatic correspondence with Europe and the United States was to be carried out in English or French. This led to the Qing court establishing a new bureau to handle foreign affairs, the Zongli Yamen, which included a language college, the Tongwen Guan, to train Chinese scholars and diplomats in Japanese and European languages. The Zongli Yamen commissioned numerous translations of texts by employees of both the Jiangnan arsenal in Shanghai and the Tongwen Guan in Beijing, and both institutions also carried out extensive translation activity on their own initiative. These and other factors led to China experiencing another wave of translation in a wide variety of material, including mathematics, the sciences, engineering, the social sciences, and literature (Ma, 2006, pp. 44-51; Wong, 2005, pp. 114-115).

Although this material was quite different in subject matter from the Buddhist sutras, and translated under very different historical circumstances, teamwork was also in fact the norm for most translation in this period, albeit on a smaller scale. Many translators worked in pairs consisting of one native speaker and one foreigner; or worked as part of a small committee or group within a bureau, checking each other’s translations. We know, for example, that all of the scientific texts translated by John Fryer while employed at the Jiangnan Arsenal were checked and revised by native speakers, sometimes as a two-step process, sometimes as a collaborative one (Wong, 2005, p. 120).[3]

A more interesting case was that of Lin Shu, who “translated” over one hundred and fifty books in a period stretching from 1899 until his death in 1924. Lin Shu was not affiliated with either of the government bureaus mentioned above; he translated only literary texts, including perhaps most famously La Dame aux Camélias (his first effort), Uncle Tom’s Cabin, several works by Dickens, and a host of other nineteenth and early twentieth century novels. Yet Lin Shu knew no foreign languages! Instead, he relied on another native Chinese speaker who could read the original work and provide him an oral interpretation of the novel, which he would then revise and write down in Classical Chinese. Like the Buddhist translators before him, Lin Shu acknowledged the team approach publicly, and his publisher included the name of the interpreter on the title page of the published translations. Although strictly speaking, by modern standards, his role was that of recording and revising the oral interpretation, no one in China at the time felt it odd to describe what he did as “translation.” Lin Shu’s name was in fact synonymous with translation in this time period. Everyone read his translations, and one of the fiercest debates concerning the proper language to use in the translation of Western works (Classical Chinese or the Vernacular) involved him as the champion for continued use of Classical Chinese (Chan, 2004, p. 18).

Lessons from Chinese History

The classic model of translation in Western theory, from ancient times down to the twentieth century, centers around a lone individual working on a single text. Some of the earliest recorded statements on translation in Europe clearly envision the translator as a solitary individual working on a text that he transforms with no aid from others. Cicero, for example, says:

I translated the most famous orations of the two most eloquent Attic orators, […] I did not hold it necessary to render word for word, but I preserved the general style and force of the language […] The result of my labour will be that our Romans will know what to demand from those who claim to be Atticists and to what rule of speech, as it were, they are to be held.

my italics; quoted in Robinson, 2002, p. 9

The use of “I” and “my labour” clearly indicates that Cicero envisioned the act of translation as involving a single individual who, unaided, produced a finished translation.

It would be tedious to list quotations from all writers in Translation Studies showing that they assume translation is essentially a solitary act,[4] but it may be instructive to choose a few representative samples from different schools that have emerged since the Second World War. Among the philosophic writers, Willard Quine’s essay “Meaning in Translation” is constructed around the hypothetical situation of a “jungle linguist” who, thrown into a group of foreigners speaking an unknown language, must attempt single-handedly to produce competent translations from that language (2000 [1959]). In the linguistic camp in Translation Studies, it is easy to find statements such as “translators can choose from two methods of translating […]. But translators may also notice gaps, or ‘lacunae,’ in the target language (TL) which must be filled by corresponding elements […]” (Vinay and Darbelnet, 2000 [1958], p. 84); “Once the translator has decided in favour of one of the alternatives, he has predetermined his own choice in a number of subsequent moves […]” (Levy, 2000 [1967], p. 149); “the translator, who becomes a secondary sender […]” (Reiss, 2000 [1971], p. 160); “The discovery of textual equivalents is based on the authority of a competent bilingual informant or translator. Thus, to find the French textual equivalent of the English text My son is six, we ask a competent translator to put this into the TL, French. He supplies Mon fils a six ans” (Catford, 1965, p. 27). Judging from the way these sentences are constructed, it seems clear that linguistic theories of translation are based on communication models of language, where a speaker and a listener are the most commonly envisioned situation, and would naturally lead to the assumption that translation is the act of a single individual.

However, this viewpoint is by no means restricted to the “linguistic camp” of Translation Studies. George Steiner’s After Babel contains a detailed discussion of “the hermeneutic motion,” wherein “The translator invades, extracts, and brings home” (Steiner, 1975, p. 298). Furthermore, virtually all of his examples of translations are presented as the work of a single individual. In systems theory, Toury talks of how the phenomenon of the “bilingual speaker may be said to gain recognition in his/her capacity as a translator has hardly been studied so far” (Toury, 1995, p. 53). In feminist approaches, we find statements such as “In fact, translation as we know it today depends on the security of bounded identities: the boundaries of authorship, language and text. At the same time, translation serves historically as a means to fix and consolidate these boundaries” (my italics, Simon, 1996, p. 45). Simon then goes on to discuss a series of individual female translators through history who worked alone to produce translations (1996, pp. 46-58). Spivak also considers the female translator as an individual acting alone, as seen in such statements as “The task of the feminist translator is to consider language as a clue to the working of gendered agency” (Spivak, 1992, p. 176). Virtually all other theories share similar assumptions, and contain statements such as “the translator does this,” or “the translator does that.” One of the leading journals in the field, The Translator, also proclaims the singularity of the translator in its very title.

When using translation theory in related disciplines, it should come as no surprise that again the basic metaphor used is that of the solitary translator. Paul Bandia, in applying translation theory to postcolonial writing situations, speaks of the postcolonial author (singular) as an individual who acts as a bridge between two cultures, like a translator (2008, p. 31). The “translator as bridge” metaphor again assumes a single, unaided agent, despite the fact that elsewhere he emphasizes the importance of oral culture on the development of African literature.

Eugene Nida’s diagram representing the translator as a black box between a source and a target text from the 1960s is a particularly interesting example because, as someone centrally interested in the translation of the Bible into as many languages in the world as possible, he must have been aware of the fact that most translations of the Bible in the twentieth century (and indeed, before that), have been accomplished by committee. Just as the Buddhists did not feel that the translation of a sutra should be entrusted to a single individual, the number of original languages, the length of the text, and the importance of the Bible for Christian missionaries has led to large-scale, well-funded, and regulated societies which employ people of various expertise working together in rounds of translation, discussion, and revision which often take years or even decades to complete.[5]

Nida’s model for translation does not acknowledge the complexity of this process, even as Nida himself was participating in the activities of groups such as the Wycliffe Bible Translators and the American Bible Association. The current principles of the Wycliffe International make it clear, for example, that they view the translation process as involving teamwork, testing on target audience members (which can be seen as analogous to the Buddhist practice of oral interpretation to an audience), and revision, sometimes over a period of many years (anon., 2008).

One might object that use of the singular is merely a linguistic convention, and that researchers in translation studies do not thereby preclude group activity. However, if they do not preclude it, they certainly do not foreground it. Moreover, this argument is reminiscent of early objections to feminists’ call for inclusive language regarding the use of masculine pronouns to designate a person of unknown sex, or a group of people of mixed sexes. The convention of representing everyone with masculine pronouns erases the presence of the feminine, and feminists have persistently lobbied for the use of inclusive language to rectify this situation. In translation studies, the fact remains that very few studies explicitly treat collaborative translation, while most either explicitly reject it (Quine’s model, for example) or adopt a linguistic convention that implicitly erases it.

This lack of attention to teamwork as a common component of translation has meant that historical studies of earlier periods tend to neglect this facet as well. For example, a 1967 study of John Fryer’s work for the Chinese in the nineteenth century speaks of Fryer as translating texts into Chinese with no mention of the fact that none of these translations were accomplished alone (see Bennett, 1967); Fryer, as a non-native speaker of Chinese, always worked with one or more native speakers. Studies of Lin Shu also show a tendency to talk of Lin Shu’s achievements as a translator, when in fact his role was closer to editor or reviser; the inter-linguistic component of the process, which we would normally label as translation or interpretation today, was always accomplished by his assistant. If they are ever mentioned, the interpreters are likely to be blamed for “faults” in Lin Shu’s translations.

Paying closer attention to the actual facts involved in the history of translation in China, and historical and theoretical reflections by Chinese scholars, might therefore help us to rethink our models of translation. Not that I am trying to make an essentialist argument about Orientals who value the collective versus Occidentals who value the individual; for although teamwork may not have received much attention in the theoretical modeling of the translation process, it is certainly not the case that it is unknown. Indeed, once we begin to look for teamwork in the modern period, it is easy to find examples. In China, Yang Hsien-yi and his wife, Gladys Yang, were among the most prolific and respected translators of Chinese literature in the post-revolution period (1949-); they always worked as a team. In both the Chinese and the Western context, translation agencies often employ multiple translators, either because a given job is especially large, deadlines are particularly tight, or quality control is an issue. In addition, translation agencies and government translation bureaus often have an elaborate system of translation, checking, revision, re-checking, and editing.

In this respect, the emergence of skopos theory has been a step in the right direction, as it introduces the idea of multiple roles in the translation process. In particular, the introduction of the role of commissioner has had a salutary effect on discussions of the translation process as multi-layered (see Vermeer, 1989). However, skopos theory still tends to theorize the translator as a role that is played by one person, with other roles in the process being treated as ancillary and therefore separate from the translation process. Some good work has also been done on the revising process (see Mossop, 2001), but little attempt has been made to discuss the interaction between translator and editor when they are not the same person. Isolated papers have noted the need for a team approach, especially in translation for theatrical performance (Bassnett, 1985, pp. 91-92), and the tricky and destabilizing effect of author-translator collaboration (see Fraser, 2004).[6] Still, much remains to be done.

The Importance of Relay Translation in the History of Chinese Translation

Another facet of the translation process that has been neglected in Western theory, but which comes to prominence when we examine the history of the transmission of Buddhism to China, is relay translation.

Relay translation is perhaps the most understudied phenomenon in translation studies today (see St. André, 2009). When it is mentioned, it is usually in a negative sense, such as Walter Benjamin’s famous dictum that relay translation is impossible because the relation between language and text has been changed in the translation process from being like skin and fruit to royal robes loosely encompassing the content, “overpowering and alien. This disjunction prevents [relay] translation and at the same time makes it superfluous” (Benjamin, 1996, p. 258).

Relay translation played a vital role in the transmission of Buddhism to China. Buddhism did not arrive in China directly from India; rather, it moved through several intermediate kingdoms in Central Asia along the Silk Road in a complex process that relied on one or more relay translations of texts (see Matniyaz, 1996).

Central Asia has an extremely rich and complex mix of peoples and cultures. Four major language families are represented (Indo-European, Altaic, Sino-Tibetan, and Semitic), and most major religions (Zoroastrianism, Shamanism, Buddhism, Manichaeism, Judaism, Islam, and Christianity) have had followers and recruited converts in this region.

It was through this region that Buddhism proceeded in steps to China; during the first wave of translation (65 CE - 300 CE), virtually all Chinese translations of Buddhist texts were produced from relay translations. Sogdalian was one of the languages used, but there were several others. We know that An Shigao, mentioned above, was not able to read the Sanskrit originals of the sutras he translated (Ma, 2006, p. 79). Even in later periods when working with original texts, the Chinese often relied on the expertise of Buddhists from Central Asia who were trilingual in an Indian language (Sanskrit or Pali), their own mother tongue (into which the sutras had often already been translated), and Chinese. In other words, without relay translation, Buddhism, which became one of the three main religions of China and had a deep influence on all facets of Chinese culture, would never have reached China.

Relay translation has continued to play a vital role in the history of Chinese translation. Eva Hung has described the importance that relay translation played in the history of diplomatic relations between the Chinese and various states in premodern times. During much of this period, relay translation was actually a sign of prestige for the rulers of China, because it demonstrated that even distant kingdoms that needed to rely on relay translation in order to communicate with the Chinese were eager to do so (Hung, 2005, pp. 74-75). Various scholars have described in great detail the importance of Japanese as a pivot language in the late nineteenth and early twentieth centuries (see Lackner et al., 2001; Liu, 1995). In the twentieth century, both Russian and English have also played the role of pivot language. For example, Ibsen’s A Doll’s House, which had an enormous effect on the development of Chinese drama and fiction in the 1930s, was introduced through relay translation (see Tam, 2001). In the 1980s and 1990s, the introduction of a wide variety of theoretical texts in the social sciences and humanities were mostly translated from English versions, for example Walter Benjamin’s “Task of the Translator.”

Again, if we begin to look at the history of translation in other regions, bearing the Chinese case in mind, it is not difficult to find periods where relay translation has played an important role, and isolated studies give us glimpses of this. Van Gorp (1985), for example, details the importance of French as a pivot language for the spread of the picaresque novel in Europe from Spanish into English, Dutch and German, while Foz also mentions Arabic-Latin-Spanish relay in thirteenth-century Spain (1998, pp. 93-94). On a more theoretical level, Luise von Flotow has argued that feminism opens the way to seeing translation as a serial and ongoing process. (1997, p. 47) Such studies are unfortunately few and far between to date.

During the age of exploration and colonization (between the sixteenth and twentieth centuries), relay translation was important in the introduction of non- Western material to Europe. Any text translated from an “exotic” language into a European language was much more likely to be relayed into other European languages based on the first translation rather than be re-translated directly from the original. Thomas Percy’s Pleasing History (itself partly a relay translation from Portuguese), the first English translation of a Chinese novel published in 1761, was quickly relayed into French, Dutch, German and Italian. Direct translation of the novel into French was not undertaken until 1929, whereupon an English relay translation of this French text promptly appeared, although there had also been a second translation directly into English in 1829 (see St. André, 2000).

In the same time period, relay translation also played a vital role in the introduction of European material into colonized regions, with the colonizer’s language serving as the pivot language for most European material being translated into the local languages. Thus in the Philippines, Spanish acted as a pivot language, while in Indonesia it was Dutch, in Vietnam it was French, and in Malaysia it was first Portuguese and then English. Yet recently, while writing the entry on relay translation for the new edition of the Routledge Encyclopedia of Translation Studies, I was surprised to find very little secondary literature on the phenomenon of relay.

I cannot help but see a parallel between the historic mistrust of the translation process among general readers and a mistrust of relay translation within the translation profession. In other words, I believe that the neglect and disparagement of relay translation represents the internalizing of the general mistrust of translation within the translation community; if translation always involves loss, as is generally believed, then relay translation must involve more loss and is therefore undesirable. The continued neglect of relay translation, even as translation theorists speak of empowering translators and treating translation as re-creation, means that the translation community has not yet fully come to terms with these issues. Yet just as we cannot expect everyone in the world to learn every language, and therefore we need translation, so too, we cannot expect every possible language combination to be covered consistently, and so we need relay translation.

If there is no origin, if all texts are intertexts, and if translation really is re-creation or re-writing, then those new texts produced by translators and interpreters should (theoretically) be just as translatable as their immediate source text. We need more studies of relay translation precisely in order to determine what, if any, differences are noticeable between them and “original” translations, and what they might teach us about the translation process.

Recent Developments

At the 2008 annual conference of the Canadian Association for Translation Studies, it was heartening to observe the number of presentations which, directly or indirectly, touched upon relay translation. Rachel Lung’s paper on the role of translation in East Asian diplomacy mentioned the importance of relay translation in that sphere; Sherry Simon discussed the importance of relay in the Bengali Renaissance; Clara Foz mentioned relay translation of Greek into Latin through Arabic; Christopher Larkosh mentioned relay in the Lusophone world; and Diptiranjan Pattanaik mentioned the role of Bengali as a pivot language for other parts of India in the nineteenth century.

There was relatively little mention of translation as a collaborative process, although Judy Wakabayashi did mention collaborative translation in Japan. Li Chun’s paper on the translation of the Bible into Chinese by a single individual was instructive as a counter-example; this solo translation has never gained popularity, whereas the earlier twentieth century “standard” translation that continues to be popular today (the Union Bible) was the result of teamwork.

All of these developments are encouraging, although more work remains to be done, not just in the history of translation, however important that might be. It is also important that theoretical models of translation take into account the complexity of the translation process, and the frequency of relay translation in everyday translation practice. Already today as East European countries join the European Union, the number of official languages expands, and the number of possible interpreting combinations increases exponentially, relay interpretation will need to play an increasing role in the functioning of the European Union, for both logistic and economic reasons. In the business world, it is also clear that Japanese companies are making the decision, for economic reasons, to employ relay translation for technical manuals through English, not because Japanese-German, Japanese-French, or Japanese-Spanish translators are unavailable, but because it is cheaper to use English as a clearing house (see Álvarez, 2005). These translations are being produced increasingly by group or collaborative processes, managed by translation companies that routinely employ an increasingly complex combination of translators/editors/proofreaders/project managers for maximum efficiency, quality, and profit. All of these phenomena and more deserve our attention, as they indicate that both relay and collaborative work will continue to play an important role in the profession for the foreseeable future.

Indeed, I believe that the importance of relay translation and collaborative translation, both in Chinese history and in contemporary practice, hold important implications for translation pedagogy as we help prepare to train future generations of practitioners.

First and foremost, we need to re-assess the tendency to give students individual assignments of discrete blocks of text that they simply need “to translate.” While such tasks are certainly useful in training basic skills, they do not prepare students for the reality of working in the increasingly fragmented work environment many translators face today. For example, in a process where a text is going to be post-edited, understanding that what they as “the translator” will be asked to do does not include stylistic editing and formatting can save unnecessary labour.

If translation is in fact often a collaborative process, then training exercises should reflect this reality. One possible way forward is assigning more group work and more long-term projects which include several steps in the translation process, from evaluation of the text to be translated through post-editing and preparation for publication. While teaching translation at the National University of Singapore, I designed a final-year unit around group work wherein the students actually published their own work as an edited volume of fiction, essays, and poetry translated from the works of local Singaporean writers. Students were asked to take charge of all facets of the work, from choosing the pieces and preparing a bid to overseeing the layout and cover design. Working in teams starting from week one, the students learned just how many individual steps were involved in producing a “finished” translation; each team member participated in several different roles, and were invited to reflect upon the process. This group work did of course include translating the texts, but that task was set within the broader context of the entire process. More recently at the University of Manchester, I have been experimenting with a course module in professional development which focuses on everything but the “actual translation”: evaluating the text, conceptual and terminological research, preparation of a job bid, and consideration of other aspects of the translation process, such as editing, as a potential career path. In both cases, students often make comments such as “I never realized that translation was such a complex process.” Hopefully this expanded awareness will allow them to function better in the industry, no matter what part or parts of the job they eventually choose to take on.

As for the lessons which the history of relay translation can teach us, I think that we should also consider assigning students translated texts to translate into a third language so that they can experience first-hand whether or not such texts pose different challenges for the translator. It might also be useful to have students working in pairs, where a text translated by one student is then passed to another student for translation into a third language.

Finally, the lesson regarding relay translation might also very profitably be applied to interpreter training. Anecdotal evidence from interpreters suggests that relay interpreting may lead to differences both in the way in which the first interpretation is carried out, as well as the way in which the second interpretation is handled. The most common point I have heard is that relay interpretation puts greater strain on the first interpreter, who often feels more pressure to produce a quality interpretation that can then be relayed by her or his colleagues; in particular, long pauses may be replaced by filler. Again, more research is needed so that the situation can be better understood, and interpreters better trained.

I hope that this brief overview of some of the possible implications of lessons from Chinese translation history for pedagogy will stimulate others to conduct further research and improve training for translators and interpreters.