Language Contact and Cultural Contact along the Silk Road: Insights from Translation Studies
Rachel Mairs
（University of Reading, UK）
“Why are the Jesuits wrong when they use the Chinese word Tian Zhu, the Lord of Heaven, for God? The Dominicans do the same. There may be a Chinese idol called by that name, but the missionaries have only to define their use of the term. Is it not much better than to speak of‘Deus’, as some missionaries do in Japan, introducing a rather strange syllable which has no meaning to the natives? I cannot see why it should be more dangerous to use the term Tian Zhu, or just Tian... I do not claim this to be a settled matter, but as long as the entire controversy has not been clarified, I do think that we ought to leave out any prejudice and stop slandering these people and all their forefathers for their impiety.”
(Bayer 1730, quoted in Lundbaek 1986, 139.)
Translation is a crucial element in communication between cultures. As I write this paper in English, I am conscious that many of the participants in the conference for which it is written‒and many of the readers of the publication which will result‒will read it in Chinese translation.(1) The agency of the translator is critical. Translators make numerous choices in rendering information between languages. Their task is complicated not just by linguistic matters (e.g. word order, syntax, grammatical gender) but by cultural differences between the communities which speak the languages in question. The modern scholarly literature on translation is a rich one, and tends to focus on the translator’s role as decision-maker, and the constraints under which they work. Recent topics of interest include translation in the Egyptian Revolution (Mehrez 2012;Baker ed. 2015). The idea that translation is a transparent process of the precise rendering of the full content, tone and resonances of one utterance into another is long outdated (see e.g. Venuti 1995).
Along the Silk Roads, numerous cultures and polities came into contact with one another. There were linguistic differences both between and within these cultures and polities. When we examine processes of cultural interaction, and the spread of philosophical and religious ideas, along the Silk Roads, we are therefore also dealing with scenarios in which translation played a key role. Although for many periods the raw data (texts and evidence of spoken language) is lacking, it is important also to analyse cultural transmission as a linguistic transaction.
The quotation with which I opened this paper comes from the work of a German, Gottlieb (or Theophilus) Siegfried Bayer (1694-1738), who worked at the Academy of Sciences in St. Petersburg (Lundbaek 1986). Bayer was a scholar with wide-ranging interests. Like his educated, Christian peers in early eighteenth-century Europe, he knew Latin, Greek, Hebrew and several contemporary European languages. His published written works are in Latin, which was still a common international auxiliary language among scholars in Europe at that time. He was interested in numismatics, and wrote the first modern study of the Graeco-Bactrian Kingdoms, on the basis of coins in St. Petersburg (Bayer 1738;see Browne 2003 and Mairs 2013/2014). He was also interested in the Chinese language, but had a major obstacle to his researches: he could not visit China himself, had no access to native Chinese-speakers, and had very limited written materials at his disposal in Europe. What Bayer could do was to correspond with European Jesuits in Beijing, a tortuous process which could take years between letters, as they were carried across Siberia. Bayer’s studies of Chinese are incomplete and contain many errors, as might be expected. He was very aware of this. He nevertheless displays a self-effacing, ‘can-do’ attitude, which to a modern reader is very appealing.
Bayer’s views on translation are practical, and in some ways strikingly modern. The translation of the name of the Christian God was a problem which Jesuit missionaries in China struggled with. As Bayer describes it, Jesuits took two different approaches. The first was to preserve the Latin word for god (deus) in the translation, transliterating it into a different script, and adapting it to the phonology of the host language where necessary. The second approach was to find a word in the host language which approximated the concept of the Christian God, and adapt it to a new use (天主, Tiānzhǔ). Bayer favours the second approach. In taking the time to discuss these matters, he draws attention to the range of options available to a translator, and the ways in which their choices can affect the reception of a new concept in a host language.
Since Bayer’s time, the translator’s choices and independent agency in transposing ideas between languages have been discussed with increasing sophistication, with reference to both modern and ancient languages(e.g. McElduff 2013; McElduff and Sciarrino eds. 2011). The following flowchart is adapted (and translated) from Frédéric Colin’s essay on‘translating cultural alterity in the civilisations of antiquity’ (Colin 2015). It outlines the options available to a translator who needs to transpose a culturally-specific notion from one language into another.
The two most basic choices are between translation and borrowing. If the translator chooses to translate the word (e.g. deus) into the host language, they can either find an equivalent (e.g.天主) or create a new word. If they use a translation equivalent, they may choose‒or be compelled‒to superpose it onto an existing concept, or to shift the meaning of the word in the host language to encompass new associations. Creation of a new term may involve creating a calque, or explaining the concept in a more roundabout way. Lexical borrowing‒using a loanword‒can be carried out in the script of the host language, or the script of the donor language. In both cases, there is a choice between carrying the word over directly, or adapting it to the morphology and/or phonology of the host language.
It is important to examine the options available to the translator when we examine how texts containing culturally-specific concepts (such as religious or philosophical terminology) are transmitted and translated along the Silk Roads. This gives an insight into how ideas were altered as they moved between cultures, and how different cultures received and reinvented foreign concepts. I do not assume that the translator in every case was able to exercise free choice: political and ideological forces, and the structure and phonology of the language involved, will always have constrained this choice.
I would like to explore a concrete example of these processes at work, in the transmission of Buddhism along part of the Silk Roads. I take my example from Afghanistan, near the beginning of Buddhism’s long journey from India to China, but the concepts I introduce here apply equally to texts in other languages along the Silk Roads (such as Sogdian, Tocharian, Tibetan or Chinese), in which I have no training. The Aśokan Edicts in Greek, Aramaic and Prākrit at Old Kandahar were part of the Mauryan emperor Aśoka’s new Buddhist ethical programme. The region had passed from Seleucid to Maurya control towards the end of the fourth century BC(on Hellenistic and Mauryan Kandahar, see Bernard 2005; Mairs 2014,110-113). One of the Aśokan inscriptions (bilingual Greek-Aramaic) was on a rockface (Rougemont 2012, No. 82) and another (Greek only) on a stone block used originally in a building (Rougemont 2012, No. 83). Another short Aśokan text is known from the site, which alternates Prākrit and Aramaic. All these inscriptions are translations‒direct or indirect‒of Prākrit inscriptions from elsewhere in Aśoka’s empire.
Prākrit loanwords are uncommon in these translations: there are just four. Aśoka is referred to as Piodasses, a transliteration of the personal name or Prākrit epithet piyadasi ‘of benevolent sight’ (Νo. 82, line 2;No. 83, line 11). The Indian place name Kalinga also occurs (83, line 12). These are simply renderings of proper nouns, given Greek case endings. The only real items of Indian vocabulary are bramenai ‘Brahmins’ and sramenai ‘ascetics’ (83, line 17). The masculine Sanskrit nouns, brāhmaṇa and śramaṇa are given the Greek masculine singular ending most closely approximating their original phonology: -es, plural-ai.
What options did the translator have in transposing these terms into Greek? First, he had to make a phonological choice in how to render phonemes not conventionally written in Greek. Foreign sibilant and nasal consonants are replaced with the closest Greek equivalent. The alternative to taking the Indian terms over, with phonological and morphological adaptation, would be to find or create Greek translation equivalents. In accounts of Alexander the Great’s encounters with Indians who lived by particular ethical-philosophical codes, some Greek authors use the term gymnosophistai‘naked philosophers’ (e.g. Strabo 16.2.39; Plutarch, Alexander 64).
It is perhaps the places where the Greek Aśokan Edicts choose not to borrow culturally-specific Indian Buddhist concepts that are most interesting. As has been noted frequently in discussions of the Edicts, this is the case with the ethical terminology employed (see the commentary to Rougemont 2012, Nos. 82 and 83). There are several places where the Prākrit originals offered the translator the opportunity to make more than one linguistic choice. I focus here on the selection of the Greek eusebeia for the Prākrit dhamma (Sanskrit dharma).
The meaning and significance of dhamma in Prākrit is itself problematic. Etymologically, it comes from the Sanskrit dharma, which Monier-Williams’ Sanskrit-English Dictionary gives as ‘that which is established or firm, steadfast decree, statute, ordinance, law’. It is not commonly encountered in pre-Aśokan texts (for example, Vedic literature), but assumes great importance in the Edicts. Olivelle has argued that Aśoka builds on the appropriation of dharma/dhamma by growing ascetic religious movements in northern India in the mid-first millennium BC, using it as“the central concept to construct a theology that gave legitimacy to his rule and a religious/moral foundation to his empire” (Olivelle 2011, 128).
The Buddhist connotations inherent in the term are clear, and themselves pose problems for the Greek translator (Hiltebeitel 2011, 36-37). But above and beyond its ethical significance (holding up such virtues as vegetarianism, respect for one’s elders, etc.), the Aśokan dhamma is a unifying imperial ideology, which regularises social behaviour (Olivelle 2011, 129; Olivelle 2012, 174). It is in this context that his ‘Buddhist missionary activities’ need to be considered: “Ashoka was not confining himself merely to spreading the message of the Buddha; he was reflecting his understanding of the centrality of social ethics to kingship and to a king’s relationship with his subjects” (Thapar 2006, 303).
And Aśoka did seek to explain the dhamma to Greek speakers, both within his empire and outside its borders‒as well as to a range offoreign peoples. In the Thirteenth Major Rock Edict, Aśoka is stated to have obtained ‘the victory of dhamma’ even “where the Greek king Antiyoke is, and beyond that Antiyoke, four kings: Turamaye, Antikini, Maka and Alikasandare” (trans. adapted from Norman 1997-1998; names transcribed as in the Shahbazgarhi text of the edicts). The Prākrit text here adapts the phonology of the Greek names Antiochos, Ptolemaios, Antigonos, Magas and Alexander, just as Greek does with Prākrit. No further testimony survives to Aśoka’s claimed mission to the Hellenistic kingdoms, so it is impossible to say whether and in what manner the dhamma was explained there. One should not assume that the same linguistic strategies were adopted as in the Greek edicts at Kandahar. In particular, the political force of dhamma may have been treated differently in regions outside Aśoka’s political control. What little we know of the linguistic landscape at Kandahar indicates a plurality of written languages in use for different purposes in the Achaemenid period, before the conquests of Alexander the Great (the evidence of the multilingual administration of Achaemenid Kandahar is summarised by Mairs 2014, 40). Still less can be said about the speech community. The only written languages attested are Aramaic, Elamite and Greek, none of which are native to the region. In the Aśokan edicts, and all surviving inscriptions from Achaemenid, Hellenistic and Mauryan Kandahar, expression of ideas is potentially constrained by the written registers available (Mairs 2014, 142-143). A multilingual community, the majority of whose members did not have Greek as their first language, may have provoked different translation strategies than a mission to Greek courts.
What meanings, then, did eusebeia communicate to those literate in Greek? The basic meaning is ‘piety’, in the sense of reverence towards the gods, or sometimes parents. In this sense, it represents certain aspects of Aśoka’s dhamma very well. It may not represent the full range of meanings‒of a term which was in any case still in the process of directed evolution in Prākrit‒but it was the choice the translator made.
The Aramaic Edicts from Kandahar provide an interesting comparison to the Greek versions. It should first be noted that the surviving Aramaic material from the site is scant: only two inscriptions. Aramaic was used as a language of administration throughout the Achaemenid Empire. As such, it stood in a diglossic relationship to local spoken and written languages. The presence of the language at Old Kandahar and, significantly, an Elamite tablet from the site, indicate that the administration at the city was run along similar lines to that of better-documented sites such as Persepolis, or even Bactra. The Aramaic used in Central Asia contained many Iranian words. This pattern is common to the documents from the satrapal archive at Bactra (Naveh and Shaked 2012, 51-54), the Aśokan Aramaic texts (Mukherjee 1984, 46), and the single Aramaic-script ostrakon from Ai Khanoum (Rapin 1992, 105). It is difficult to tell in the current state of the evidence whether these were considered as intrusive to the language, or particular to the usage of bilinguals (marked code-switches). My inclination‒because of their pervasiveness‒is to view them as naturalised loanwords, at least in as far as the formal register of imperial Aramaic is concerned.
One of the Aramaic inscriptions from Kandahar‒the Aramaic section of the Greek-Aramaic bilingual‒conforms to this pattern of standard Achaemenid Aramaic with some Iranian vocabulary. Like its Greek complement, it opens with a statement about Aśoka’s promulgation of the dhamma. The Greek uses eusebeia for dhamma and states that people have become eusebesterous ‘more pious’. The Aramaic uses an Iranian word, represented in Aramaic script as ptytw, and later in the same sentence the Aramaic phrase qšyṭ’ mhqšṭ. The interpretation of this passage, and the text as a whole, is controversial (competing views are summarised by Teixidor 1969, 347-349). I present it here in Altheim and Steihl’s (1958) transcription, with Aramaic words in capitals and non-Aramaic words in lower case:
ŠNN 10 ptytw ‘BYD ZY MR’N prydrš MLK’ QŠṬ’ MHQŠṬ
For ten years ptytw was carried out, because our lord the king Priyadarš qšyṭ’ mhqšṭ.
Ptytw may be taken to mean ‘expiation’ or ‘equalization of guilt and punishment’ (discussed by Garbini in Pugliese Carratelli and Garbini 1964,43-46, following Levi Della Vida in Pugliese Carratelli, et al. 1958 and Altheim and Stiehl 1958). Qšyṭ’ mhqšṭ means ‘has promoted truth’ or ‘who promotes truth’ (Pugliese Carratelli and Garbini 1964, 47-48). The latter is usually taken as a translation of dhamma, but it is possible that the Iranian term is also a translation of this, or a similar, Prākrit term.
Examining the strategies available to translators, and the reasons they made the choices they did, can grant insights into cultural contact along the Silk Roads. Like much later European Jesuit missionaries in China, those who wished to spread Buddhism along the routes between India and China were faced with the challenge of explaining new concepts in languages to which these concepts were alien. As with modern translators, they had a range of options available, including borrowing (e.g. transcribing deus in Chinese characters), or changing the meaning of an existing word in the host language (e.g. using天主, Tiānzhǔ). In the Aśokan Edicts at Kandahar, specifically Buddhist terminology is rendered into two languages (Greek and Aramaic) belonging to communities which had no previous experience of Buddhism. The translators’ decisions to introduce Prākrit loanwords, or adapt an existing Greek or Aramaic term, reveal processes which must have taken place further along the Silk Roads, as Buddhist teachings were spread through multiple cultures, polities and languages.
Bibliography
Altheim, Franz and Ruth Stiehl 1958: “The Aramaic Version of the Kandahar Bilingual Inscription of Aśoka,” East and West 9, 192-198.
Baker, Mona ed. 2015: Translating Dissent: Voices from and with the Egyptian Revolution.(Critical Perspectives on Citizen Media.) London: Routledge.
Bayer, Theophilus Siegfried 1730: Museum Sinicum in quo Sinicae Linguae et Litteraturae ratio explicator. St. Petersburg: Ex Typographia Academiae Imperatoriae.
——. 1738: Historia regni graecorum bactriani: in qua simul graecarum in India coloniarum vetus memoria explicatur. Petropolis [St. Petersburg]: Ex typogr. Acad. scient.
Bernard, Paul 2005: “Hellenistic Arachosia: A Greek Melting Pot in Action,” East and West 55, 13-34.
Browne, Gerald M. 2003: “Bayer’s Coin of Eucratides,” Zeitschrift für Papyrologie und Epigraphik 145, 212.
Colin, Frédéric 2015: “Traduire l’altérité culturelle dans les civilisations de l’Antiquité:le paradigme de la transposition lexicale,” in Frédéric Colin, Olivier Huck and Sylvie Vanséveren (eds.), Interpretatio. Traduire l’altérité culturelle dans les civilisations de l’Antiquité, 35-64. Paris: De Boccard.
Hiltebeitel, Alf 2011: Dharma: Its Early History in Law, Religion, and Narrative. (South Asia research, New York, N.Y.) New York; Oxford: Oxford University Press.
Lundbaek, Knud 1986: T. S. Bayer, 1694-1738: Pioneer Sinologist. (Scandinavian Institute of Asian Studies Monograph Series 54.) London: Curzon Press.
McElduff, Siobhán 2013: Roman Theories of Translation. London: Routledge.
McElduff, Siobhán and Enrica Sciarrino (eds.) 2011: Complicating the History of Western Translation: The Ancient Mediterrannean in Perspective. Manchester: St. Jerome.
Mairs, Rachel 2013 (2014): “The Reception of T. S. Bayer’s Historia Regni Graecorum Bactriani (1738),” Anabasis: Studia Classica et Orientalia 4, 255-262.
——. 2014: The Hellenistic Far East: Archaeology, Language and Identity in Greek Central Asia. Berkeley: University of California Press.
Mehrez, Samia (ed.) 2012: Translating Egypt’s Revolution: The Language of Tahrir. Cairo:American University in Cairo Press.
Mukherjee, B. N. 1984: Studies in Aramaic Edicts of Aśoka. Calcutta: Indian Museum.
Naveh, Joseph and Shaul Shaked 2012: Aramaic Documents from Ancient Bactria (Fourth Century B.C.E.) from the Khalili Collections. (Studies in the Khalili Collection.)London: The Khalili Family Trust.
Norman, K. R. 1997-1998: “Aśoka’s Thirteenth Rock Edict,” Indologica Taurinensia 23-24, 459-484.
Olivelle, Patrick 2011: Language, Texts, and Society: Explorations in Ancient Indian Culture and Religion. London: Anthem Press.
——. 2012: “Aśoka’s Inscriptions as Text and Ideology,” in Patrick Olivelle, Janice Leoshko and Himanshu Prabha Ray (eds.), Reimagining Aśoka: Memory and History,157-183. New Delhi: Oxford University Press.
Pugliese Carratelli, Giovanni, Giorgio Levi Della Vida, Giuseppe Tucci and Umberto Scerrato 1958: Un editto bilingue greco-aramaico di Asoka. La prima iscrizione greca scoperta in Afghanistan. Rome: Istituto italiano per il medio ed estremo Oriente.
Pugliese Carratelli, Giovanni and Giovanni Garbini 1964: A Bilingual Graeco-Aramaic Edict by Asoka. The First Greek Inscription Discovered in Afghanistan. (Serie Orientale Roma 29.) Rome: Istituto italiano per il medio ed estremo Oriente.
Rapin, Claude 1992 : Fouilles d’Aï Khanoum VIII: La trésorerie du palais hellénistique d’Aï Khanoum, l’apogée et la chute du royaume grec de Bactriane. (Mémoires de la Délégation Archéologique Française en Afghanistan 33.) Paris: de Boccard.
Rougemont, Georges 2012: Inscriptions grecques d’Iran et d’Asie centrale. (Corpus Inscriptionum Iranicarum II: Inscriptions of the Seleucid and Parthian Periods and of Eastern Iran and Central Asia. Vol. I: Inscriptions in Non-Iranian Languages.) London: School of Oriental and African Studies.
Teixidor, Javier 1969: “Bulletin d’épigraphie sémitique (1969),” Syria. Archéologie, Art et histoire 319-358.
Thapar, Romila 2006: “The Mauryan Empire in Early India,” Historical Research 79,287-305.
Venuti, Lawrence 1995: The Translator’s Invisibility: A History of Translation. London:Routledge.
(1)此文保持外文原版，并未译成中文。——编者注