"If mankind and the jinn were to come together with a view to producing a Koran like this one they would not produce the like of it, not even if each were to support the other" (Koran 17:88).

It is in the context of this challenge that Kees Versteegh's opening paragraph must be understood: "Since its earliest appearance as a world language in the seventh century, Arabic has been characterised by an opposition between two varieties:... standard... and... vernacular..." The standard is ultimately, of course, the language of the Koran, its beauty making it revered as perfect and therefore inimitable. Ask most native speakers about the rich variety of dialects, on the other hand, and one will generally be told that the vernacular language is devoid of rules. Under such circumstances, Arabic speakers, despairing ever since its revelation of attaining the perfection found in the Koran, appear to have settled, at least in everyday speech, for the more realistic goal of emulating more manageable prestige forms, now sedentary, now bedouin.

The Arabic Language is intended as a textbook introducing the historical, linguistic and sociolinguistic development of the Arabic language. Versteegh has marshalled the published data well, and each chapter ends with a digest of the sources for suggested further reading. The book begins with a brief history of Arabic studies in Europe, correctly stressing that from the 18th century onwards, after the earlier attraction first for medical and other scientific purposes then for reasons biblical and polemic, "most of the interest was genuine and without ulterior motives". The ensuing chapters describe Arabic within its Semitic context, then in its earliest and pre-Islamic forms, and the development and structure of the classical norm. Next Versteegh launches into discussion of the emergence of "New Arabic" as the Islamic Empire was poised to take shape. It is this Ur-Dialekt and not the Koran, we learn, which acted as the source of all later vernacular Arabics. Veering back in the direction of the standard before describing the dialects proper, Versteegh tackles the phenomenon of so-called "Middle Arabic", the data on which comprise texts in Judeo-Arabic and Karshuni, as well as those early papyri and medieval manuscript codices containing the hypercorrections, colloquialisms and malapropisms of authors and scribes only partially trained in the classical standard.

There will always be pitfalls for the linguist when basing conclusions on the evidence of isolated utterances. Indeed, Versteegh recognises that "written dialect is not identical with spoken dialect". Occasionally, the selected samples are no more indicative of what a speaker might deem natural and correct usage than are the contributions to the Colemanballs column of Private Eye. And so I found myself disagreeing strongly with a few statements made in relation to the Egyptian dialect, with which I am most familiar. For example, when exploring "contemporary Middle Arabic", it is claimed that Egyptian speakers, in order to appear educated, may produce qurqan for qur'an as a hypercorrect form.

ÁñÁ«ÊÓƵ

Another concern, since this is intended as a textbook, is the use of highly technical linguistic terminology throughout, often without explanation. This factor makes the sixth chapter, on the structure of classical Arabic, uncharacteristically hard going. A glossary, in addition to the comprehensive bibliography of European-language publications and index, would be one way of addressing this. Curiously the book lacks a proper transliteration table. As a result readers must deduce for themselves the rules of the scheme used in the copious quotations, which are given only in romanised form. Aside from the inevitable slip-up (eg qaddis and qiddisa), occasionally the adopted scheme is ambiguous, as in the examples l-malaka and l-malika, where, had the rules been more consistent, one would have expected either the former to read l-malakatu or the latter, l-malik. These few reservations apart, the structure of the work is clear and apt, and its content well argued and thought-provoking.

Indeed, challenges to ingrained assumptions occur throughout. In analysing the sparse epigraphical evidence for Arabic prior to Islam Versteegh enumerates the salient features in the orthography and pronunciation of the Koran, which suggest that its language "reflects the eastern usage whenever differences between eastern and western Arabic existed". Then, much nearer the end of the book, in the invaluable and insightful chapter on diglossia in the Arab world, Versteegh mentions the speech delivered by President Sadat immediately prior to his assassination, the text of which appeared in the press the next day "with a note from the publisher that there had been no time to 'translate' it (from Egyptian dialect) into standard language". Though seemingly unrelated, these two points beg an interesting question: since the prophet Muhammad was a native of the western region of the Hijaz, whose dialect is characterised, inter alia, by its "loss of the glottal stop (hamza), which was retained in the Eastern dialects", how is it that the language of the Hadith appears so similar to the prestigious form found in the Koran? Either i) the theories relating to Arabic dialects in the pre- and early Islamic periods are flawed; or ii) the Prophet Muhammad spent his entire ministry effecting a dialect other than his own; or iii) the journalistic practice of "upgrading" the reported speech of Arab leaders to a more prestigious form rather than quoting them verbatim is not new but dates back to the first centuries of Islam.

ÁñÁ«ÊÓƵ

It is perhaps still too early to reflect on the likely impact of the world wide web on Arabic, or to predict the outcome of recent Jordanian publications promoting the use of simplified Arabic (luga muyassara) as a language of instruction at university level. Versteegh concludes therefore with a chapter on the influence of Arabic on the non-Arabophone world, on Spain, on Africa, Iran, Turkey on South, East and Southeast Asia.

The germ of another conclusion appears in the middle of the work. Here Versteegh refers to the paradox that, in spite of the spread in literacy since the end of the colonial era, when "people write Arabic, they tend to make the very same mistakes which one finds in the Middle Arabic texts of the Classical period". To the believer, perhaps this is inevitable: after all, to err is human. In the 20th century only one major Muslim scholar, Taha Hussein, has questioned the inimitability of the Koran, though his suggestion immediately met with opprobrium. It was therefore never realised; to have done so would have undermined the above-quoted basis of the Koran's miraculous nature. Nevertheless, for more than 1,400 years, classical Arabic has continued to survive unsullied the influence of the colloquial "mother tongues". It is tempting therefore to ask whether, far from constituting a threat, the existence of this expanse between two opposing poles - one divine, one demotic - is not in fact the well on which the vitality of this unique language depends.

Roderic Vassie is Middle East country manager, the British Library Document Supply Centre.