Bryn Mawr Classical Review 2004.02.19
J. N. Adams, Mark Janse, Simon Swain, Bilingualism in Ancient Society: Language Contact and the Written Text. Oxford: Oxford University Press, 2002. Pp. x, 483. ISBN 0-19-924506-1. $98.00.
Contributors: J. N. Adams, Frédérique Biville, Claude Brixhe, Philip Burton, Penelope Fewster, Pierre Flobert, Mark Janse, D. R. Langslow, Martti Leiwo, Zeev Rubin, Ian Rutherford, Simon Swain, David G. K. Taylor, and Kees Versteegh
Reviewed by Joshua T. Katz, Princeton University (firstname.lastname@example.org)
Word count: 3699 words
Mel Gibson, the son of a fiery opponent of Vatican II and believer in the Latin mass, is, as I write this review, putting the final touches on his potentially explosive new movie, "The Passion of the Christ," which depicts the crucifixion of Jesus. The dialogue is in Aramaic and Latin; there are to be no subtitles. If the controversy surrounding the film puts a damper on the apocryphal (?) story we have all heard about the teacher/principal/parent/governmental lackey who objects to the teaching of foreign languages in school on the grounds that "If English was good enough for Jesus, it's good enough for me," that is just as well. It is perhaps too much to ask that anyone except scholars puzzle over the emphasis on Latin on Golgotha one spring day around the year 30.
In fact, however, questions of who spoke or read, or could speak or read, which language(s) where, when, and for what purposes rarely have simple answers, and it is the aim of this terrific new collection to provide a framework, based on both methodological reflection and careful case studies, for examining bilingualism in the ancient world. Bilingualism is, as most of the contributors point out, a hot topic in linguistics and other social sciences -- in Langslow's words, "one of those subjects, like medicine, in which nearly every kind of specialist has an interest" (24). This volume, along with Adams's massive Bilingualism and the Latin Language (Cambridge: Cambridge University Press, 2003), will ensure that linguistic juxtaposition (to make up a term to cover multi-lingualism, diglossia, code-switching, language contact, "alloglottography" [see pp. 4, 37, and 44-46], and much else) gets its due in Classical and Mediterranean studies as well, where as Adams and Swain note, the "evidence ... is immense, and [yet] the subject is underexploited" (1). A recent reviewer commented in BMCR that she "like[s] collected volumes and [is] not overly concerned about whether they are 'coherent', as long as they offer interesting papers loosely centered on an interesting theme";1 I feel the same way, and yet it cannot be denied that there is something especially satisfying about the rare book -- and this is one -- in which most of the papers really do seem on any account to belong between the same covers, despite dealing with different cultures and genres, times and places. The display of historical, philological, and linguistic learning here is dazzling, and every student of Classics would profit from going through it. "The book springs from a conference held at the University of Reading in 1998" (1), write Adams and Swain; it must have been quite an event.
The volume opens with Adams and Swain's well-organized and not merely summarizing "Introduction" (1-20) and continues in four parts: "I. Introduction" (two papers), "II. Greek-Latin Bilingualism" (four), "III. Greek and other Languages" (six), and "IV. Latin and other Languages" (two); there follow a substantial bibliography and a not especially useful index. Typographical errors pop up more than occasionally, but I noticed fewer than one might expect in so lengthy and complicated a book. Perhaps the most embarrassing mistake is that the subtitle on the dust jacket, "Language Contact and the Written Word," is different from the one on the title page, "Language Contact and the Written Text." With few exceptions, most notably the short and refreshingly witty paper by Flobert on Latin and Frankish in sixth-century Gaul, the emphasis throughout is on texts rather than individual words -- a good thing, all in all, as it leads readers away from the common "gee-whiz" assessment of etymological pursuits ("Gee, isn't it neat that the Greek word for 'bathtub' is 'substratal'?!"; "Yeah, sure, but what's the payoff?") and presses historians and literary critics to realize that they must pay attention to both obvious and covert linguistic interactions in the passages they study.
The first paper, by Langslow, "Approaching Bilingualism in Corpus Languages" (23-51), gets the book off to an excellent start. Langslow has modest aims -- "I propose merely to touch on a few of the more important themes, issues, and parameters in recent and current research on bilingualism, and to offer a few illustrations, ... nearly all from the published work of others, ..." (23) -- but his is one of the most elegantly written of the essays and arguably the most generally useful one. It also provides, in what the author calls "in many ways the most important [section] of the paper" (46), a spirited and thoroughly reasonable defense of the theory that underlies all serious research in historical linguistics, the so-called Comparative Method, which, it is true, does not in its traditional tree diagrams of linguistic relationships typically take account of the fact that languages can be in contact and exert mutual influence. I strongly recommend that all serious undergraduate and graduate students of Classics read this piece, and I plan to assign it in the courses on comparative grammar I expect to teach in coming years.
The other paper in the introductory section is Versteegh's "Dead or Alive?: The Status of the Standard Language" (52-74), which makes the point, at perhaps greater length than necessary, that when one deals exclusively with written records, it is particularly important to keep in mind that they are "always affected by the norm of a written standard" (53): because people do not typically provide exegetical commentary on the "history of the metalinguistic attitude" (72) that underlies their own writings, it becomes all the more difficult for scholars today to assess a given language's registers and document linguistic change with authority. Referring to earlier publications, Versteegh, an Arabist, suggests that "drastic linguistic changes only take place when there is a break in the process of language acquisition, such as the breaks that involved the mass acquisition of Greek and Latin during the Hellenization in the east and the Roman conquests in the west, respectively"; in his opinion, the "linguistic changes that affected Greek and Latin started at the very beginning of the contacts with speakers of other languages and took place within a short period of time. Such a scheme runs counter to what is commonly assumed about the history of Greek and Latin on the basis of the written record, since the written record seems to suggest a gradual change" (71). I do not know what sorts of changes count as "drastic," a critical term that is never defined; in addition, although his paper is largely about the fact that linguistic change can take place without being recorded ("the written language can never be taken as evidence of the terminus ante quem of a certain change" ), Versteegh is not careful about distinguishing between actual change, which is necessarily instantaneous (someone makes a speech error), and the necessarily gradual success of that change (the wider community adopts the error with the result that it truly becomes part of the language).
The four contributions that concentrate on Greek and Latin are all very fine. I am particularly taken with Swain's "Bilingualism in Cicero?: The Evidence of Code-switching" (128-67), which is that special paper that covers its subject both so magisterially and with such a light touch that the reader new to the material learns a great deal while the specialist is never bored. Swain interweaves his conclusion about Cicero -- in short that his code-switching to Greek, as exemplified notably in his letters, is "not testimony of his bilingualism: it is first and foremost a discourse strategy within his Latin" (164) -- with an excellent general account of the status of Greek in Rome, where, he claims, "there is ... no evidence to prove that any educated Roman did regularly hold conversations in Greek" (147). Like Langslow's paper, it deserves to be the centerpiece of many a classroom discussion.
Very different in style is Adams's typically intense piece, "Bilingualism at Delos" (103-27), which presents a wealth of data in much sparer terms. (Don't blink or you'll miss something -- though in fact you have a second chance since a fuller account appears as Chapter 6 of Adams's 2003 monograph, mentioned above.) Adams describes as apparently unique the Greco-Roman community of negotiatores on Delos in the second century B.C.: in their numbers and comparable status, the Italian traders and the Greek ones "set up a potential competition between the two public languages on a roughly equal footing" (103). His is a persuasive account of the image the Italian members of the guild were trying to present of themselves in the bilingual inscriptions that they set up -- monuments whose linguistic and epigraphical quirks show "accommodation" of both Greek in Latin and Latin in Greek. Their "repeated use" of the self-designation Italici suggests that the traders were striving for a sort of "corporate identity" (108), and Adams is especially interesting when he describes how this group of mobile people on Delos and elsewhere in the Greek-speaking world employed the term.
The other two papers on Greek and Latin combine specific case studies with more widely applicable methodological observations. Biville's "Graeco-Romans and Graeco-Latin: A Terminological Framework for Cases of Bilingualism" (77-102) concentrates on Hellenic influence on Latin: her well-documented discussion moves nicely from largely ordinary material about mono- and bilingual speakers of "Greek and Latin" to the phenomenon of "Graeco-Latin," which she characterizes as "one culture, two languages," following Suetonius, Claud. 42.1 utroque ... sermone nostro (92). Particularly valuable are her remarks on interpreters,2 "Roman Greek," and the concepts of the "hybrid" and the "contact language." This last beast she defines as an "entity which need not have resulted from the fusion ... of the two founding languages, but rather was located at their point of intersection" (100): a good example of a Greco-Latin "interlexeme" is the nonce-verb morari with a long o in Suetonius, Nero 33.1, Nero's succinct and jokey description of the life and death of Claudius as a cross between the Latin verb morari 'linger' (with a short root vowel) and the Greek adjective μῶρος / μωρός 'foolish.' I also especially enjoyed her accounts of the words Graeculi and (see fn. 2, below) semigraeci (vs. Graeci proper; see pp. 90f.) and of Graecolatini in the Itinerarium Egeriae (see pp. 84f.; compare also Livy's Gallograeci on p. 99).
Leiwo's paper, "From Contact to Mixture: Bilingual Inscriptions from Italy" (168-94), treats Greco-Latin contact in Italy from the third to the sixth century A.D. as it can be seen in the writings of the Christian and Jewish communities. Truly fascinating are his linguistically and philologically astute remarks on the Greek-cum-Latin-cum Hebrew funerary inscriptions of the Jews of Venusia (note that the contemporary pagan and Christian inscriptions from Venusia are typically just in Latin); as Leiwo admits, the conclusions one is to draw from them are not entirely clear, but he wrestles with a number of possibilities. A lasting merit of Leiwo's paper will be his 11-part typology of the "features of contact [that] can at least be detected from inscriptions" (173): on pp. 172-79 he carefully classifies and then exemplifies a range of types, from "Inscriptions which have the same text written in two languages" (L/G or G/L) to "Inscriptions which have two clearly different texts (e.g. a funerary inscription + an epigram) written in different languages" (L+G or G+L) and from "Latin in Greek characters or Greek in Latin characters" (Lg or Gl) to "Phonological code-mixing, i.e. isolated foreign phonological elements in an otherwise Latin or Greek text" (LGphon or GLphon). The abbreviations he uses are sometimes unaesthetic (not that I have been able to come up with better ones); the ideas, however, are very much not.
The four papers on "Greek-Latin Bilingualism" can be read as a block in a way that the six papers in the next section cannot, for the "other languages" of the header "Greek and other Languages" range from Phrygian (Brixhe) to Middle Persian and Parthian (Rubin) via Lycian (Rutherford), Egyptian of various sorts (Fewster), and Syriac and local Aramaic dialects (Taylor). Rutherford's "Interference or Translationese?: Some Patterns in Lycian-Greek Bilingualism" (197-219) provides a good overview of what we know of the Lycian language and tries to establish a few specific points about linguistic interactions between Lycian and Greek (and, in the case of the Xanthos Trilingual, also Aramaic). His most interesting proposals concern syntax and especially "aesthetic symmetry in translation" (216), the maintenance of a certain syntactic or stylistic pattern proper to one language (Lycian, say) when it is rendered, otherwise apparently unidiomatically, into another (Greek) that appears alongside it: not "translationese," but "part of a strategy that Greek and Lycian texts should be in a relationship of formal symmetry, just as the physical texts are in symmetry on either face of the stone" (217). Rutherford's paper should be required reading in any graduate class on the east face of Helicon.
A number of languages were in use in Roman Egypt, but Fewster's "Bilingualism in Roman Egypt" (220-45) concentrates on the interaction between Greek, the language with the highest status, and Egyptian, especially Demotic. She presents a clear picture of the kinds of evidence we have for bilingual activity and has many interesting things to say about language and power -- without, however, wishing to accede to a facile interpretation of Egyptian language and script as a means of protest by the indigenous non-elite. She begins with some ostraca from the Narmouthis archive that are in Egyptian but have Greek words scattered here and there, pointing out, though, that while "there is clearly an important level of bilingualism at this modest village temple," these documents are "unique" in Roman Egypt (224). A striking example of how the side-by-side use of two languages does not necessarily indicate a carefree diglossic situation comes from a mummy label from Sohag: the dead woman, Tamosis, is described in the Egyptian section of the label as t3 Wynn.t, that is to say, 'the Greek'; but the person who wrote the Greek "seems baffled by the term and laboriously transliterates it as Ταουαειαναεινε. He perhaps knew no Egyptian; alternatively, his Greek was so limited that he was unable to translate the name. If one scribe was responsible for the whole label his bilingualism was surely limited, and if two scribes were involved there can have been little communication between them" (231).
"Interactions between Greek and Phrygian under the Roman Empire" (246-66) by the dean of Phrygian studies, Claude Brixhe, may strike some readers as linguistically over-technical, especially in its second half. Nevertheless, the paper provides a concise summary of the two languages' mutual influence, with particular and welcome emphasis on Phrygianized Greek. It is to be hoped that more Hellenists -- not just Indo-Europeanists and experts on Greek dialects -- will mine the comparatively small corpus of Neo-Phrygian inscriptions (currently 114, dating from the end of the first to the middle of the third century A.D.) for cultural and historical information about this very interesting area of the Hellenistic world.
The next paper, Rubin's "Res Gestae Divi Saporis: Greek and Middle Iranian in a Document of Sasanian Anti-Roman Propaganda" (267-97), is explicitly historical rather than linguistic, though the author does anything but shy away from philological detail. This is an ambitious essay about an extraordinary document, a trilingual inscription that the Sasanian king Shâpûr I (ruled 240-71 A.D.) set up at the foot of the rock reliefs of Naqs-i Rustâm, a few miles north of Persepolis, with the obvious purpose of trumpeting his might and military successes. It is clear why the king would have commissioned such an inscription in Middle Persian (his own language) and Parthian (the language of the Arsacid dynasty and still an important one under Sasanian rule) -- the two languages of the only comparable document from the period, the Paikuli inscription -- but why, Rubin asks, should there be a version in Greek (albeit in "what might be called 'learner's Greek'" [275, with thanks to Adams]) as well?: "[W]ho were the Greek-reading destinees of Sasanian royal propaganda in the district of Stakhr supposed to be?" (269). Rubin spends considerably more time telling his readers about the wonderful inscription than he does answering this question, though he makes two tentative suggestions on pp. 289-91, namely that it might impress Roman diplomats (compare Lactantius, De mort. pers. 5.6 on legati ... nostri to Shâpûr's court) and that Greek-reading merchants might sometimes be on site (Rubin offers no corroborating evidence for this). In his attempts to understand the document, and in particular to determine the relationships among the three versions, Rubin sometimes comes to conclusions different from those of Philip Huyse, whose imposing two-volume edition of the inscription,3 appeared only once Rubin's paper had already been finished. I lack the competence to judge the issues and simply note the existence of a long Appendix in smaller-than-usual type, "The Problem of the Genesis of the Greek Text" (291-97), in which Rubin attempts to counter Huyse's analysis and bolster his own opinion, namely that a Parthian scribe, having taken instructions in Middle Persian from Shâpûr and his flunkies, first composed the Parthian text before turning to the Middle Persian and Greek versions.
Taylor's "Bilingualism and Diglossia in Late Antique Syria and Mesopotamia" (298-331) is a nuanced account of Greco-Aramaic interaction. Taylor makes a strong case that the standard sociolinguistic designations "H" (high) and "L" (low) -- used to indicate relative prestige when a community employs two or more modes of speech -- are largely inadequate for describing the status of Greek and Aramaic among the Syro-Mesopotamian population; indeed, he goes so far as to claim that at Palmyra, Greek and Aramaic are both H, with "equal public status, although [each was] associated with particular specialist functions and linguistic domains" (320). Of particular interest to me -- and, I would be glad to think, to Mel Gibson -- is Taylor's attention to varieties of Aramaic and how they interact with Greek: in not very many pages he neatly demonstrates why it is important to examine local phenomena closely before making sweeping claims about a large area, and he does an excellent job of making readers see why "some care should be taken to distinguish between different Aramaic dialects and not to presume that they are identical, or to label them all as, for example, 'Syriac'" (303). Although it is not a principal point of his paper, it seems worthwhile to take the opportunity to highlight in this review, for those who may wish to follow up on it, the existence of what Taylor calls an "extraordinary hybrid variety of Syriac for use in translating Greek texts that is perhaps the most extreme example of the interference of one language in another to be found in the ancient world" (328).
The longest and, in my opinion, least persuasive paper in the volume is Janse's "Aspects of Bilingualism in the History of the Greek Language" (332-90). Janse does not wear his learning lightly, and the individual points he wishes to make are often lost in the forests of unnecessary and sometimes downright unhelpful detail. Examining three very different linguistic features -- relative clauses, causatives, and clitics -- in two situations of language contact that are "complete opposites in almost every respect" (337) -- the composition of the Septuagint (translated in Hellenistic times from Hebrew) and the formal characteristics of the Cappadocian Greek dialect (spoken until recently in Turkey) -- yields a great deal of interesting data but not a coherent paper. Janse writes that his studies are "intended to be illustrative of the difference between conscious interference in religious translation and unconscious interference in language maintenance under strong cultural pressure and long-term bilingualism" (361), but surely it would be more fruitful to compare apples and pears, say, than apples and mushrooms. Indeed, the incoherence of Janse's approach is clear from his very next sentence: "It should once again be noted, however, that whereas the two types may be contrasted as being complete opposites, they cannot be properly compared."
The final part of the book, "Latin and other Languages," is the bastard child, being blessed with only two papers, Burton's "Assessing Latin-Gothic Interaction" (393-418) and Flobert's "Latin-Frankish Bilingualism in Sixth-century Gaul: The Latin of Clovis" (419-30). Burton begins by summarizing the kinds of (limited) evidence that have been adduced in support of influences between Latin and Gothic; he then turns to a discussion of individual examples of such linguistic phenomena as loan-shifts and "underdifferentiation"4 in an effort to determine the nature of the language of the Bible as translated in the 4th century by the Visigothic bishop Wulfila and to assess the likelihood that this translation, our principal source of the Gothic language, might have become Latinized a century or two later in Ostrogothic Italy. He is cautious in his conclusions but suggests, among other things, that G. W. S. Friedrichsen may have been right in supposing, some 75 years ago, that the sixth-century Codex Argenteus introduced further Latinisms into the Gothic Bible beyond any due to Wulfila's own knowledge of Latin.
The final paper, an altogether pleasing read, is Flobert's on the Latin of Clovis. "Did Clovis even know Latin?" (419), the author asks. Probably he did know some, and Flobert is convincing when he concludes that "[i]n short, while Franks knew a bare minimum of Latin [but generally not enough to write it; see p. 426], Latin speakers were seldom bilingual" (428). Historians will wish to read Flobert's paper in the context of the growing and increasingly nasty debate over the viability of the so-called "ethnogenesis-model" for European societies and beyond.5 From the linguist's point of view, of course, there can be no doubt but that Frankish had an especially successful "posthumous revenge" (429), as for that matter did Gaulish: is there any expression more Gallic than "Vive la France!"?
Adams, Janse, and Swain have put together an outstanding and admirably coherent volume that is unlikely -- since Oxford University Press has saved money for some time by not sewing bindings -- to withstand the assaults of all the Classicists and linguists who will be running to make use of it.
[[For a response to this review by Anthony Alcock, please see BMCR 2004.02.35.]]
[[For a response to this review by John Ma, please see BMCR 2004.02.36.]]
1. C. S. Kraus, Review of D. S. Levene and D. P. Nelis (eds.), Clio and the Poets: Augustan Poetry and the Traditions of Ancient Historiography (Leiden: Brill, 2002), BMCR 2003.09.01.
2. In her discussion on pp. 85f. of the technical terms uertere and interpretari 'adapt, translate' and of Livius Andronicus and Ennius as semigraeci (Suet. Gram. 1.2) I miss mention of Stephen Hinds's elegant point about the reflexive nature of uersutum in the conventional first line of Latin literature, Livius' "translation" of Homer's Odyssey (Allusion and Intertext: Dynamics of Appropriation in Roman Poetry [Cambridge: Cambridge University Press, 1998] 61f.).
3. Die dreisprachige Inschrift Sâbuhrs I. an der Ka'ba-i Zardust (SKZ) (London: School of Oriental and African Studies [= Corpus Inscriptionum Iranicarum III/1], 1999).
4. "The phenomenon of underdifferentiation is a very familiar one to anyone who has tried to speak a foreign language, but it has received surprisingly little attention in the literature of language contact. It is what happens when speakers of a second language fail to reproduce in that language a difference not found in their own; as, for instance, when an English speaker fails to distinguish in Italian between tempo and volta, or an Italian fails to distinguish between to hear and to smell (sentire)" (415). Penetrating remarks on this phenomenon in the context of translation are now to be found in Umberto Eco's Mouse or Rat?: Translation as Negotiation (London: Weidenfeld & Nicolson, 2003), whose title refers to some consequences (e.g., for Hamlet III.4) of the usual underdifferentiation of mouse (German Maus, French souris, Spanish ratón) and rat (Ratte, rat, rata) in Italian, where both animals are normally referred to as topo (see pp. 32-34).
5. See, e.g., The Myth of Nations: The Medieval Origins of Europe (Princeton: Princeton University Press, 2002), a popular work of scholarship by the leading American proponent of the model, Patrick J. Geary. Geary, whose book features on its dust jacket François-Louis Dejuinne's painting "Le Baptême de Clovis à Reims," notes that the infamous French politician Jean-Marie Le Pen declared in 1991 that "the French people [was] born with the baptism of Clovis in 496" (7, with reference to Le Monde in 175 n. 4).