Overview

Egypt in the Hellenistic and Roman periods was a profoundly multilingual society, where users of different languages (including Egyptian, Greek, Latin and Aramaic) came into contact with one another. It is curious, however, that relatively few individuals in the papyrological record use «interpreter» or «translator» (ἑρμηνεύc) as a professional title. Widespread individual bilingualism is one possible reason for the absence of interpreters. But it is also likely that the role of interpreters and translators was, consciously or unconsciously, obscured. This might be for a variety of reasons: lack of professional identity and prestige, the inconvenience or embarrassment of mentioning that an interpreter had to be used, or because using an interpreter was too commonplace to require mention. To further complicate the picture, in a large number of instances where the title «interpreter» does appear, it is with reference to individuals engaged in commercial, rather than explicitly linguistic, mediation.

Those cases where interpreters or translators between languages are explicitly mentioned are usually in some way exceptional – special cases where it is important to record that interpreting or translation has taken place. In several Roman period transcriptions of legal proceedings, for example, it is noted that the presiding official questioned a witness or other party «through an interpreter» (δι’ ἑρμηνέωc). Similarly, in some written translations of Latin legal documents into Greek, the translator not only signs the document with his name and the title ἑρμηνεύc, but also certifies to the accuracy of the translation and its validity under Roman law. Interpreters and translators are frequently only recorded with their professional title when the task in which they are engaged has additional legal ramifications.

The research presented in this article is a preliminary report on one part of a larger project on interpreters of foreign languages throughout the Graeco-Roman world. Other case study regions provide interesting parallels for the Egyptian evidence, in particular for the frequent «invisibility» of interpreters in the epigraphic and documentary record, and for the use of the title «interpreter» to refer to a wider spectrum of professional activities than simply language mediation. This project, a collaboration between the present author and Dr. Maya Muratov (Adelphi University), aims to examine the semantic range of the various ancient terms rendered in modern languages as «interpreter», and to contextualise epigraphic and papyrological references to ancient «interpreters», both geographically by reference to other regions of the Graeco-Roman world, and theoretically within the scope of current research on ancient sociolinguistics¹. Our case study regions are: 1) Egypt and the Levant; 2) the Roman provinces of northern and eastern Europe (where almost all of our evidence comes from contexts associated with the Roman army); 3) the Bosporan Kingdom; and 4) the city of Rome. Our focus is on the documentary record, although we intend to publish some subsidiary studies on the perception of interpreters in Greek and Latin literary and historical sources². The number of references to interpreting, translation and their agents in the papyrological and epigraphic record as a whole is very small. Our Egyptian database contains at present 110 entries; the non-Egyptian database, 20.

The Egyptian Evidence

The papyrological material from Egypt – and also Arabia – naturally rather dominates our study. As well as the advantages of producing a new overview of the papyrological evidence on interpreters and translators as agents (and to a lesser degree on translation and interpretation as processes), a conspectus and reexamination of the Egyptian evidence may also allow us to highlight some of the common themes which recur in our case study regions. In summary, our common findings from all of the geographical contexts we examine are that references to interpreters (Greek ἑρμηνεύc, Latin interpres) are rather infrequent in the literary and the documentary records; and that there is a degree of terminological ambiguity in the use of these terms. The more copious documentary material from Egypt offers an excellent opportunity to explore these commonalities, and to develop some hypotheses about why they recur.

Although several studies have already been devoted to the question of interpreters and translators in Hellenistic and Roman Egypt, none have so far been comprehensive or systematic, and there is room for new critical appraisal of the evidence. The Egyptian sections of Wiotte-Franz (2001) are, unfortunately, incomplete in their coverage and do not offer full analysis of the material. The most useful and nuanced studies of the papyrological evidence are Peremans (1983), Rochette (1994) and Rochette (1995), with older treatments in Taubenschlag (1951) and Calderini (1953). Online tools now allow us to search more efficiently for occurrences of various terms for «interpreter» in the papyri and to identify new occurrences (see e.g. Mairs [2010]), but there is more work to be done on the topic than simply expanding our corpus of papyrological interpreters. Since there is not space in the present discussion for a comprehensive assessment of the papyrological evidence, I shall restrict my discussion to two topics – the «interpreter» as commercial mediator, primarily in material of the Ptolemaic and early Roman periods; and the necessity of recording translation and interpreting in legal documents and court cases, mostly of the Roman period. In many other cases, we have little or no further information which enables us to say anything about an individual interpreter’s professional role or social position, especially where someone is named with the profession «interpreter» in a list or register. Such is the case, for example, with Iakob the ἑρμηνεύc, named in a list of debtors (SB XXII 15599; provenance unknown, V / VI AD); Kronion the ἑρμηνεύc who appears in a γραφεῖον register from Tebtunis (P.Mich. II 124, AD 46); or the anonymous ἑρμηνεύс who paid 4 drachmae for an unknown service in a private account from Theadelphia (P.Berl.Leihg. II 39 recto, AD 160). In other cases, even additional evidence does not serve to clarify an individual’s role, and may even multiply the number of interpretations which we might bring to bear (e.g. the Apollonios «ἑρμηνεύс of Trogodytes» of UPZ II 227; see Rochette [1995]).

Interpreters as Commercial Agents

Neither the Greek ἑρμηνεύc nor the Latin interpres means solely «interpreter of a foreign language». This degree of terminological breadth or ambiguity is also present in the English word «interpreter», which can mean an expounder or mediator of any kind of difficult, obscure or otherwise incomprehensible information. There are, in fact, many instances in which the Greek term ἑρμηνεύc in the papyri can be demonstrated to refer to someone whose principal occupation is not mediation between languages, whether or not such linguistic interpreting may also have been part of his professional remit³.

In many cases, it is difficult to discriminate between linguistic and non-linguistic interpreters, and we should assume that, in the Egyptian context, this terminological ambiguity reflects a very real degree of overlap between the two roles. The semantic – and professional – kinship between language mediation and commercial mediation is a natural one. Rather than attempting to identify any individual «interpreter» as primarily a middleman or a language technician, as has been the goal of most previous studies, we understand this ambiguity as central to the perception and professional roles of «interpreters» in many of the chronological and geographical contexts we examine.

The evidence from the Zenon archive (Pestman [1981]) is especially interesting in this regard, since it provides examples of «interpreters» in explicitly commercial roles as early as the third century BC. There have been some efforts in the past to force these interpreters into a more linguistic role, but, although language skills will have been useful, or even essential, to such a agent, this is not necessary. The editors of P.Col.Zen. II 63, for example, «see no reason to regard these ἑρμηνεῖc [in the Zenon papers] as anything else than interpreters, conversant with the Egyptian language, who were useful in the daily dealings between the natives and those Greeks who did not speak the native tongue.»⁴ But the «interpreter» in P.Col.Zen. II 63, one Glaukias, is in fact clearly another middleman, paid 12 drachmae «for purchases» (εἰc ἀγοράcματα: a point made by Peremans [1983], 15). The positions of the ἑρμηνεῖc Limnaios (P.Cair.Zen. III 59394) and Apollonios (P.Cair.Zen. I 59065; P.Ryl. IV 563; PSI IV 409: discussed by Crawford [1973], 352) in the archive are rather more ambiguous. Knowledge of more than one language might have put Glaukias and his colleagues in a good position to be able to facilitate commercial transactions, but here the sense of «language-interpreter» is evidently no longer paramount, whether or not language skills were still needed.

It is interesting that one of our other broad case study areas – inscriptions and documents associated with the Roman army in the northern and eastern European provinces – presents a similar ambiguity or crossover between language mediation and commercial mediation. The verb interpretari may be used in a commercial sense (T.Vindol. 213; Bowman / Tomlin / Worp [2009]), and military interpretes might have concurrent or subsequent careers as negotiatores (AE [1978] 635).

The papyrological record also furnishes cases where the role of an «interpreter» has strayed very far from any linguistic association at all. In one restricted context – the Arsinoite nome in the early Roman period – «interpreter» becomes a term for a regulator of weights and measures (discussed in Mairs [2010]). In twelve papyri, dated between AD 14 and AD 151, dry goods are measured in a «six-choinix measure of the ἑρμηνεύc»⁵. The ἑρμηνεύc in question is never named, but is stated to belong to a particular village. As well as employment in the services of large estates, such as that managed by Zenon, there are also examples, closer in date to the Arsinoite regulators, where «interpreters» can be seen in commercial business for themselves (e.g. P.Col. II 1; P.Graux III 30 [Theadelphia, AD 155]). The Arsinoite regulator-interpreters are our clearest instance of commercial interpreters in an official position.

Interpreting and the Law

The ambiguity of the Greek and Latin terms for «interpreter», and their frequent employment in a commercial or regulatory context, do not compromise the terms’ additional, complementary meaning of a mediator between languages. The verb ἑρμηνεύειν and its variants are the terms by which written translations are referred to at all periods. In translations from Demotic into Greek, the standard formula used to introduce a translation is «copy of an Egyptian document translated into Greek as far as possible» (ἀντίγραφον сυγγραφῆс Αἰγυπτίαс μεθηρμηνευμένηс Ἑλληνιсτὶ κατὰ τὸ δυνατόν). This formula recurs with only very minor variations in documents from the latter part of the second century BC (when new legislation clarified the relationship between Greek and Egyptian tribunals; see Pestman [1985]), right through into the early Roman period. Examples are listed and the terminology discussed in Mairs (forthcoming). There are at least two cases where we have both the Demotic original and the Greek translation preserved⁶. Similar formulae also introduce later translations of Latin legal documents into Greek⁷.

Aside from the terminology for «translation» (derivatives of the verb ἑρμηνεύειν), another persistant and integral part of these formulae is the phrase κατὰ τὸ δυνατόν «as far as possible»⁸. As well as perhaps recognising the impossibility of utterly faithful translation, accurate in content and equivalent in sense, it is clear that this phrase is above all a legal caveat. The majority of our restricted number of references to interpreters, interpreting and translation come from legal papyri: legal implications make language mediation stand out in a papyrological context. Inaccurate transmission of information might have severe consequences for any one of the parties to a dispute or transaction, and all those involved – parties, translators and interpreters, attorneys, presiding officials – might seek to absolve themselves of any personal liability for inaccurate translation. In stressing that his translation is κατὰ τὸ δυνατόν, the writer cautions those involved that translation is not an exact science, and also excuses himself of responsibility.

This is true not only of written translation of legal papers, but in the oral examination of parties and witnesses who do not speak the language in which proceedings are being conducted. Several accounts of hearings before Roman officials record that someone was questioned and answered δι’ ἑρμηνέωс «through an interpreter»⁹. In only one of these cases is the interpreter named (P.Col. VII 175: Anoubion), and this is because he is also one of the presiding officials.

There are very few cases in which a legal written translator is named, and this is invariably because he is also taking responsibility for a document’s legal validity, in addition to its accurate rendition into Greek. Two named translators translated from Latin into Greek. In BGU I 326 (Karanis, AD 194), one Gaius Lucius Geminianus, an «expert in Roman law» (νομικὸс Ῥωμαϊκόс) certifies that he has translated the preceding document and that «it is in conformity with the original will» (ἐсτὶν сύμφωνον τῇ αὐθεντικῇ διαθήκῃ). In SB VI 9298 (Oxyrhynchus, AD 249) an Aurelius Aigyptos describes himself in the same way and also certifies that the document he has translated conforms to that previously registered in Latin. The concern here is evidently accountability, and the reason the translators – who are also legal experts – name themselves is to certify that their translations are accurate representations of the original Latin legal papers and valid in a court of law.

Visible and Invisible Interpreters

In Hellenistic and Roman Egypt, language mediation had possible repercussions in matters such as the accuracy of testimony, the ability of parties to understand the proceedings, or the accuracy of written legal papers. There is a certain irony in the fact that the most visible «interpreters» in the papyrological and epigraphic records tend to be those who are not principally linguistic interpreters, but more general mediators, and those whose language mediation is of interest and import because of its further legal ramifications. Those who are only language technicians are far more more likely to be anonymous, and they lack visibility.

The interpreter’s invisibility is a recurring theme in the history of translation and interpreting (Venuti [2008]). This invisibility may be a product of the inconvenience and even embarrassment of noting the presence of an interpreter in diplomatic or other negotiations, or of underestimation or devaluation of the skills involved, but it is also the product of common perceptions about the very goal of mediation between languages, the illusion that a translation may offer a transparent window on the original.

In studies of interpreting in more recent colonial societies, such as early modern Ireland or the early colonial Americas (Palmer [2003]; Fausz [1987]), as well as in current interpreting practice, the professional role of the interpreter is to be «invisible»: to facilitate communication without him or herself becoming openly involved in the transaction. This is, of course, a fiction. Interpreters make all sorts of linguistic and diplomatic choices which involve them profoundly in the conversations which they mediate. Monolinguals may be ignorant of the skills and linguistic and non-linguistic decisions involved; an interpreter, for example, may decide not to pass on information which he or she knows will cause unintentional offence. An element of linguistic and cultural chauvinism may even be present, as in the New Kingdom Egyptian Tale of Wenamun, where the Egyptian envoy arrives in Cyprus (Alasiya) and expects to be able to communicate in Egyptian:

«He had me board and sent me off from the harbor of the sea. And the wind drove me to the land of Alasiya. Then the town’s people came out against me to kill me. But I forced my way through them to where Hatiba, the princess of the town, was. I met her coming from one of her houses to enter another. I saluted her and said to the people who stood around her: “Is there not one among you who understands Egyptian?” And one among them said: “I understand it.” I said to him: “Tell my lady that I have heard it said as far away as Thebes, the place where Amun is: ‘If wrong is done in every town, in the land of Alasiya right is done.’”»¹⁰

The bystander who helps Wenamun speak to Hatiba is not specifically called an «interpreter» – he is simply «one who understand Egyptian» – and after the beginning of the dialogue it may be noted that he ceases to be mentioned any more, even though he is presumably still interpreting the conversation. This conversation is now presented as unmediated speech between the princess and Wenamun. It is often more convenient for the parties to an interpreted conversation to ignore the decisions and personal agency of the interpreter in this way. Certain contexts, however, force these decisions and their implications on the attention of the parties, and make recognition of the interpreter’s presence and the limitations of his capabilities important. It is such circumstances in which Egyptian legal documents record that their translation has been accomplished κατὰ τὸ δυνατόν, and specific individuals take responsibility for a translation and its legal validity.