THE LETTERS OF THE GREEK ALPHABET

The first task in learning Greek is to learn the alphabet, which consists of twenty-four letters. Many are similar to English ones, and you may already be familiar with some others (for example, pi π and theta θ from mathematics). Learning the alphabet has three parts.

Learning how to write each of the Greek letters

In Greek, just as in English, different people will have different styles of handwriting. Also, printed Greek often looks a little different from handwritten Greek. That is fine – the aim is not to win prizes for the artistic quality of your lettering.What matters is for the different letters to be clearly distinguished from each other. In practice, you will probably copy the style of your teacher.

Learning which sounds the different Greek letters make

Greek has been spoken for over three thousand years, and in many different dialects. This means that there is no single right way to pronounce Greek.What matters is to make each letter have its own distinctive sound. It is also useful if your pronunciation is similar to that of other biblical scholars (and your teacher and classmates) so you can understand each other.

You may wonder why pronouncing the letters is important at all, since your desire is to read Greek, not speak it.

Students are the ones who matter. Students, and more generally all those wishing to learn, are the only reason for teachers and academic books to exist. For ninety years those wishing to learn to read the New Testament in Greek have been ably served by The Elements of New Testament Greek published by Cambridge University Press. First, in the book of that name by H.P.V. Nunn published in 1914, and then in its 1965 replacement by J.W. Wenham. So successful was John Wenham's book that for much of its forty-year history it has been the standard first-year Greek course not only in the UK but across large parts of the English-speaking world. For generations of students, ‘Wenham’ was synonymous with Greek.

Wenham's success was that he cared about students and did everything possible to make learning ‘the elements’ of New Testament Greek as simple and painless as possible. The most striking example of this was his handling of Greek accents. The scholarly tradition behind the use of accents went back many centuries, though not, as Wenham was keen to point out, back to the time of the New Testament itself. Nevertheless Wenham dispensed with accents. Or at least he dispensed with most of them – keeping only the few cases where they were useful to the student in distinguishing between otherwise identical words. Even today many scholars and teachers find this regrettable, if not even scandalous. I have never met a student, though, who shares that opinion.

INTRODUCTION

Latin had only a marginal place in Egypt; it has been estimated that only about I per cent of documents from the Roman period that have survived are in Latin, proportionate to those in Greek. Even in military finds only about 10 per cent of texts are in Latin. Bagnall (1986:5) notes that included in the 76 publication numbers of the texts from Quseiral-Qadim (military documents of early Imperial date from the Red Sea coast, eight km north of the modern town of Quseir) there are three Latin papyri (18–20), three Latin ostraca (44–6) and perhaps one Latin inscription (66). He observes: ‘Latin thus makes up about ten percent of the corpus of Greek and Latin texts. The same ratio is found in the Florida ostraka, and no doubt in both cases we have at least a general reflection of the limited but real role of Latin in the administration of the Roman army in Egypt.’ To talk of the Latin of Egypt would be misleading, because it might imply that there was an established Latin-speaking population such as might have developed over the generations its own regional features in that language. In fact the Latin which has turned up in Egypt represents a considerable diversity. If a Latin text is discovered in Egypt, that does not mean that its author was of Egyptian origin.

INTRODUCTION

In this final chapter I refer briefly to some of the themes scattered throughout the book without attempting to be comprehensive. Other topics, such as accommodation, regional variation determined by language contact, language mixing etc. are discussed particularly in the concluding sections to various chapters, and can be followed up through the index.

IDENTITY

It is often argued in linguistic literature that language is the most important marker of identity that there is (see e.g. Crystal (2000:40)). Even monolinguals living in a monolingual society where they have little or no contact with other languages may be conscious that it is their language which gives them a special identity in the world. The English, for example, constantly express pride in their language, and in doing so they are implicitly presenting the English language as one of the markers of their achievements as a people. Attempts by the French to eliminate ‘Franglais’ from the French language are reported with alacrity in the British press, with the implication that the English language is now doing what the English people have ceased to do, that is colonising foreign territory. The fading of French as an international educated language in the face of English is reported from time to time with satisfaction. Articles in the press regularly speak of the ‘richness’ and ‘subtlety’ of English, usually without acknowledgment that the ‘richness’ of its vocabulary is largely due to its reception of foreign loan-words over a long period.

INTRODUCTION

The imperfect bilinguals or second-language learners who have been identified so far belonged largely to groups resident in areas in which two or more languages were in contact, such as the city of Rome. But language contact with its consequent bilingualism did not only take place in static populations.One environment, for example, in which speakers of different languages mixed together was the army; but language learning in the army is a topic in itself (see above, 5.VII.3).

Another group in any population who come into contact with foreign language speakers are traders purveying goods across language boundaries. In many cultures the trading classes have proved adept at acquiring enough of the foreign languages needed to carry on business, and there is no reason to think that Roman traders and their trading partners will have been any different. Language learning was no doubt a two-way process, in the sense that Latin-speakers wishing to exploit a province would have come into contact with provincial languages, and on the other hand provincial traders moving into Latin-speaking areas or dealing with Latin speakers will have picked up some Latin.

We have already seen one example of this second form of contact. The Greek-speaking Milesian slave-trader Aeschines (see 1.VII.2.3), while trading in the west, was required to write out in his own hand a legalistic document in Latin. It can be deduced from some of his misspellings that he was familiar with the sounds of substandard varieties of the Latin language.

INTRODUCTION

In a previous chapter code-switching was defined, and distinguished from borrowing and interference (I.V). Some testimonia were cited which established that the phenomenon was recognised in antiquity. Code-switching is common both in literary texts and primary material, particularly inscriptions. I turn in this chapter to its determinants.

There is now a variety of opinion about the nature and motivation of code-switching. An old view was that the ‘ideal bilingual switches from one language to the other according to appropriate changes in the speech situation (interlocutors, topics, etc.), but not in an unchanged speech situation, and certainly not within a single sentence’ (Weinreich (1953: 73), my italics). Weinreich goes on to ‘visualize two types of deviation from the norm’ (i.e. the norm whereby the ideal bilingual is resistant to switching). The second of his ‘deviations’ is ‘in the direction of insufficient adherence to one language in a constant speech situation’ (1953:74). He observes that this ‘tendency (abnormal proneness to switching) has been attributed to persons who, in early childhood, were addressed by the same familiar interlocutors indiscriminately in both languages’. Switching is thus acknowledged to exist, but is seen as an aberration, or, as Weinreich puts it (1953:74), a ‘deviant behaviour pattern’. Such claims have now been rejected as a result of study of bilingual communities in which in unchanged speech situations speakers have been observed to switch languages with considerable freedom even within sentence or clause boundaries.

THE COMMUNITY OF NEGOTIATORES AT DELOS

Delos provides an interesting case study of bilingualism in the public domain in a trading community in which Romans and Italians were prominent in the last centuries of the Republic. Trade is an activity in which cross-language communication is essential. Two separate trading or commercial communities, in which Roman linguistic attitudes differed markedly, are the subject of chapters in this book. In the pottery at La Graufesenque there is no sign of Italians learning Gaulish. Gaulish went on being used by Celts, but there are indications that they were also learning Latin. But on Delos the Roman/Italian attitude to the other language was far more deferential. Italians were happy to have themselves presented as Greek-speaking in formal texts, though the situation was complex, as we will see.

Delos was the site of the earliest and largest Roman–Italian commercial community in the Greek world. After 167 the senate expelled the Delians and the island became a free-trade centre under nominal Athenian supervision. Many inscriptions, in Greek, Latin and both languages, distributed for present purposes from the second century BC onwards into the early Imperial period, attest the activities of Latin-speaking, or, one should more accurately say, bilingual, negotiatores, and raise fundamental questions about the relationship between the two languages, the motivations of language choice, if only in public documents (for ‘private’ language use, see the discussion below (VI) of the Myconos curse tablet), and the character of the bilingualism of the traders.

A TRANSLATION OF BABRIUS

It was noted earlier that the acquisition of literacy in the script of a second language can be divided into stages (see 5.VII.3), several of which are identified through various types of errors. Learners also commit different categories of errors in the acquisition of a second language, whether they are learning that language by contact at the level of speech, or by formal instruction with the help of writing. Some research has been devoted to the question whether sequences can be identified in second-language acquisition. I would not wish to claim that I have any contribution to make on the general issue of whether such sequences may be detected across a broad spectrum of languages; indeed it seems to me that where formal instruction is involved, the sequence will be determined by the nature of the instruction. Nevertheless, there is one Latin text extant which is relevant to the matter. I refer to a Latin translation of parts of two fables of Babrius which is preserved at P. Amh. II.26. The document is dated by the editors (p. 26) to the late third or early fourth century. In this chapter I offer a discussion of the Latinity of the translation, paying attention not only to its substandard features, but also to its numerous errors, many of which cannot be explained merely as displaying typical substandard or vulgar usages. There can be no doubt that the author was a Greek learner of Latin as a second-language.

INTRODUCTION

Works on historical linguistics have tended to concentrate on factors internal to a language as determining linguistic change, but external factors may also be influential, though perhaps mainly in the short term (see below, VI.4). This chapter will be about language contact as a cause of language change, in either or both of the languages in contact. I make the assumption that most bilinguals will have a first or ‘native’ language and a second language. There are ‘balanced’ bilinguals, but these are probably in a minority except in some special communities. In the material available from antiquity it is usually possible to identify a writer's first language.

The chapter has several aims. First, in general accounts of bilingualism and language change evidence from antiquity has been neglected, though it is extensive. Thomason and Kaufman (1988), for example, range all over the world but pass over antiquity in silence. This chapter will present evidence relevant to the general debate. Secondly, discussions of cross-language influence in the Roman world (such as that of R. Coleman (1975)) have dealt above all with the literary languages, and non-literary evidence has not been fully exploited. The present chapter will not be restricted to the non-literary evidence, but that will occupy a significant place. Thirdly, in keeping with a general theme of the book I will be attempting to identify the effects of bilingualism on the regional diversification of Latin.

I first began working on contact between Latin and other languages in an organised way when I had the good fortune to be Visiting Senior Research Fellow at St John's College, Oxford, in 1994–5. The project was given impetus by the invitation to deliver the J. H. Gray Lectures in the Faculty of Classics, University of Cambridge, in May 1999. The title of the lectures was the same as that of the present book. The subject turned out to have such ramifications, and the material relevant to it to be so scattered, that I might never have finished the book had I not had the even greater good fortune to be elected to a Senior Research Fellowship at All Souls College, Oxford, in 1997.

An account of the full range of bilingualism in the ancient world across the whole of the Mediterranean and Middle Eastern areas and at all recorded periods would be virtually unmanageable, unless a team of collaborators was assembled. I have restricted myself to the Roman period, from the early Republic to the late Empire (approximately the fourth century). I have not adopted a fixed cut-off point, but on the whole have avoided entering into the period of the barbarian invasions in the west. In the western Empire Latin came into conflict with a number of vernacular languages and eventually effected their death.

INTRODUCTORY REMARKS; SOME ISSUES IN THE STUDY OF BILINGUALISM

It is thought that bilingualism is more common than monolingualism, and yet linguistics has traditionally operated as if the monolingual were the normal speaker. Bilingualism across the Roman world cannot be quantified, but numerous languages survive in the written record (usually in a fragmentary state) or are attested in contact with Latin (Gaulish, forms of Hispanic, Oscan, Umbrian, Venetic, Etruscan, Hebrew, Aramaic, Egyptian Demotic and hieroglyphics, Coptic, Punic, Libyan (?), Thracian, forms of Germanic, as well as Greek), and others were spoken without leaving any trace in our sources. In the vast expanses of the Roman Empire, where mobility was high among such groups as the army, administrative personnel, traders and slaves, language contact was a fact of everyday life. To survey bilingualism in the whole of the ancient world would be an immense task, but the Roman domain, particularly during the Empire, offers more manageable data.

Bilingualism has traditionally been of interest not only to linguists, but also to anthropologists, social and cultural historians and students of literature. As found in the Roman period it has received a good deal of attention, explicitly in some of the works of (e.g.) Dubuisson, Holford-Strevens, Horsfall, Leiwo, Millar, Neumann and Untermann, Rochette and Wenskus, and implicitly in virtually any work on Latin literary genres with Greek forerunners.