This interesting study of notarial documents from the Thebaid in Upper Egypt provides an overview of the linguistic situation in Hellenistic Egypt. In addition to linguistic analysis of the documents, it touches on core topics such as bilingualism, diglossia, onomastics and ethnicity. Discussions of the koine and its historical context have traditionally been dogged by a failure to understand key concepts, starting with koine itself, but including related issues such as standard language, diglossia and even the difference between script and language (meaning, spoken language is often conceived and discussed in a framework suitable for, and derived from, writing). This book is written in a sound sociolinguistic framework, and based on clear expertise in Ptolemaic Egypt, with the result that the discussion and conclusions are useful and meaningful.

The book has seven chapters followed by ‘General Conclusions’ and appendixes. The first chapter is an introductory overview of the material, the location (Pathyris in the Thebaid) and the sociolinguistic method. The corpus is a relatively homogeneous corpus of notarial documents (contracts, wills, etc.) from 174 to 88 b.c., deriving from the offices of a limited number of civil servants (agoranomoi) in Upper Egypt about 30 km south of Thebes. The formulaic nature of the corpus allows close comparison of scribal practice, and may occasionally be part of the explanation of linguistic surprises in the documents. It is good to see V. tackle immediately the term ‘non-standard’ Greek; it is used as a neutral term by linguists for features which, from the perspective of Lysianic Attic, would be classified as mistakes. She asks the important question ‘With what stage of Greek are we comparing the language of the notaries? What is the so-called “standard language”?’ This is a question that could be asked more frequently of Greek. It would sound strange to refer to features of Sophoclean language as ‘non-standard’, since we do not suppose that tragedians were seeking to reproduce the syntax of Attic prose. Each period and genre has its own koine, and as V. notes, administrators (then as now) have a ‘bureaucratic standard’ of their own. So we need to distinguish, if possible, mistakes made by an individual which stem from imperfect competence in the local or generic standard, and features of the local standard itself which differ from classical literary prose.

In spite of the title, the book is a study of diglossia rather than bilingualism. Diglossia is a property of a community (society), while bilingualism is a property of an individual. A bilingual person has native competence in two languages, and may (of course) be illiterate in one or both. We cannot know whether notaries of Upper Egypt were sensu stricto bilingual – V. very reasonably supposes that most were not – but we do know that Hellenistic Egypt was diglossic in a complex way. The term diglossia came into use in the mid-twentieth century to describe the situation in the modern Arabic-speaking world (and much of the research was done in Egypt), in which the speaker's native language is the vernacular they learned to speak as infants (the ‘Low’ variety), while the prestige standard language (the ‘High’ variety) is mastered through education. Modern Standard Arabic is based on the classical language, and educated speakers from across the Arab world can read and communicate in it, while the spoken varieties are not necessarily mutually intelligible. In diglossic cultures where the High and Low varieties are forms of the same language, the vernacular is typically conceived as a debased or impure version of the standard, rather than its most recent form.

The Greek-speaking world was diglossic in the Hellenistic and Roman periods, since the literary standard followed classical models, while the vernacular, in the way of all languages, continued to change and develop. The term koine denotes the continuum of varieties, including the highest (though language users, especially after the first century a.d., would have been affronted by this description of their attempts at literary Attic). The reason that koine is a tricky term is that within a diglossic culture speakers have peculiar views about language, which need to be understood, but which do not match the arid realities of the linguist.

Egypt is doubly complicated, because figures such as the notaries, who were competent in Demotic and Greek, were part of a culture that was doubly diglossic: since they were Egyptians (pp. 48, 105–6), they were competent in both vernacular Egyptian and in Demotic, which denotes a writing system and a specific written form of the language. (Egyptian had another level, hieroglyphic, which notaries in the Thebaid at least are quite unlikely to have been able to write, but which adds another level to the linguistic culture of the country.) Then they must have been able to speak some vernacular Egyptian koine, and wrote a good version of the bureaucratic standard. V. touches on these issues in Chapter 2, ‘Linguistic Landscape of Hellenistic Egypt’, in which there is also a good discussion of onomastics and ethnicity. The third and fourth chapters (‘Language Use in the Pathyrite Area’ and ‘Notaries at Work’) give further detail about the linguistic and socio-historical background of the region. These insightful chapters are a rich source of information for the non-expert in Ptolemaic Egypt. The next three chapters are linguistic, and deal respectively with phonology, morphosyntax (the inflection of nouns in the sentence) and ‘syntactic transfer’ (from Egyptian, the presumed first language of the notaries, to Greek). Just occasionally the discussion of phonology and spelling in Chapter 5 gives the impression of confusion. Vowel change in Egyptian koine is difficult: V. wavers over whether to accept S.-T. Teodorsson's early dating of key changes (The Phonology of Ptolemaic Koine [1977]). Most linguists now seem to accept it, in spite of some mistakes in his data. Μηθένα for μηδένα (p. 116) is a widespread feature of koine Greek and cannot easily be blamed on Egyptian phonology, at least not without qualification (explanation in Buck, Greek Dialects [1955], p. 61). For the scribes in question it was probably a feature of the orthography, anyway. V. disagrees, unwisely, with G. Horrocks (Greek: a History of the Language and its Speakers [1997]) on the ‘alleged weakness of word final sibilants and nasals’ (p. 117). She is wrong, however. For one thing, that /-s/ is maintained in Modern Greek (by which she means the standard variety based on the southern dialects), even if correct (and there are instances of loss in attested modern dialects), would not prove that loss did not happen in this or that social/regional variety of Greek in the Hellenistic period. Many varieties of Greek disappeared: standard Modern Greek is the reflex of a specific socio-historical process, and is based on specific modern varieties. Most of V.'s analysis is interesting and persuasive, however, and shows how her expertise in the history and society of the period can illuminate the linguistic and orthographic choices of the scribes. In the chapter on syntactic transfer she makes a good case for the interference of ‘Language 1’ in the practice of at least some of the notaries.

V. finishes with some important conclusions, which would not be obvious without a study of this nature. Greek, for example, ‘did not enjoy any particular prestige status’. Her analysis of the notarial relationship with Greek (their approach to writing high-quality koine reflected their training in Demotic, which was equally remote from vernacular) strengthens, to my mind, the case for Teordorsson's analysis of the phonology of Egyptian koine: standard spellings reflect scribal culture in Egypt, not scribal competence in the spoken language. The book gives an excellent account of the socio-historical situation for linguists, and of the linguistic situation for historians.