Arabic Grammar in Context. (Lan- Guages in Context)

411 BOEKBESPREKINGEN — ARABICA 412

ARABICA

ALHAWARY, M.T. — Arabic Grammar in Context. (Lan- guages in Context). Routledge, London & New York, 2016. (25,5 cm, 264). ISBN 978-0-415-71595-9. ₤ 90.00. Mohammad Alhawary (hence MA) is associate professor of Arabic linguistics and second language acquisition at the Uni- versity of Michigan. He is editor of the Journal of Arabic Lin- guistics Tradition, and of Al-‘Arabiyya: Journal of the Ameri- can Association of Teachers of Arabic. He is also the author of Modern Standard Arabic Grammar: A Learner’s Guide, and Arabic Second Language Acquisition of Morphosyntax. The book under review (abbreviated here as Context) pre- sents according to the Preface (p. vii) mainly an exercise tool for intermediate and advanced students of Modern Standard Arabic (MSA). This characteristic feature is elaborated in 70% of the 22 chapters which are assigned to exercises while the remaining 30% is devoted to the introduction of the required theoretical foundation in a condensed form. Context leans heavily on the forementioned Learner’s Guide of the same author; the Bibliography in Context mentions even only Learner’s Guide. Learner’s Guide which presents a theoretical framework of Arabic grammar in a more ampli- fied form, is meant for beginners and also, like Context, for more advanced students, so the readereship of these books is largely the same, but the presentation of grammatical items is different. Because of the inextricable bond with Context I will present an additional review of Learner’s Guide at the end of this article. MA states explicitly (p. viii) that Context “does not pre- sent an exhaustive coverage of (Modern, WvT) Standard Arabic Grammar”.1) Nor does Learner’s Guide, even put

1) More complete reference works include Modern Written Arabic. A Comprehensive Grammar (Elsaid Badawi, Michael G. Carter and Adrian Gully, 2nd ed., 2016, Routledge) and A Reference Grammar of Modern Standard Arabic (Karin C. Ryding, First published 2005, Cambridge Uni- versity Press). 413 BIBLIOTHECA ORIENTALIS LXXV N° 3-4, mei-augustus 2018 414 together with Context, I would like to add here already. Nev- of the chapters is concluded with ‘Other points to note’ ertheless the most important grammatical issues are dealt which contains not only issues accessory to the previous with in these two books. theoretical essentials but also references to the precise find- Context consists of 22 chapters, followed respectively by ing places in Learner’s Guide. Additional texts for review, Keys (to the exercises), a Bibli- The Arabic language used in the Introductory texts and in ography, a Glossary of Vocabulary and Expressions, and the Exercises (especially in the last two of them) is roughly finally an (English) Index (of grammatical terms). The Addi- 85% prosaic and for the remaining 15% of poetic origin. For tional texts for review are intended to practice the gained and a non-Arab student who has to find his way in the Arabic developed grammatical skills in all of the preceding chapters. ‘desert’ of grammatical difficulties I am of the opinion that The Keys contain elaborations of all exercises, i.e. those in the poetic part in the exercises, meant to achieve the pro- the chapters individually as well as those included in the posed learning objectives, should be 5% at most. This is Additional texts. The ensuing Glossary is a very convenient because grammatical analysis of poetic constructions tool for students in order to get a flavour of the right transla- demands additional skills from students which makes the tion of the Arabic texts. However, I am astonished that the intended learning process unnecessarily difficult for them. contained words and expressions are arranged alphabetically, not according to the root-arrangement of most Arabic dic- Now I will make some remarks on the treatment of gram- tionaries. For the student of MSA it is of utmost importance matical items throughout the chapters. to learn mastering to look up such words in that way. It is In Ch. 1, section “Form”, about the definite article, MA the more surprising because in the very Learner’s Guide MA starts with the sentence “Apart from pronouns and proper unfolds a separate part to pick up this skill.2) The final Index names which are definite, words in Arabic (nouns and adjec- is restricted to grammatical terms in English. I would expect tives) are marked for definiteness or indefiniteness.” The here an Arabic Index, just as MA does in Learner’s Guide, possible carrying of nunation of some proper names has the more so because throughout Context grammatical terms nothing to do with indefiniteness; proper names (and also in Arabic are used, although mostly combined with their pronouns) are definite by nature. The only reason for this English counterparts. Unfortunately the Glossary does not nunation lies in the fact that such proper names are triptotes, furnish these Arabic grammatical terms. so his mentioning of definiteness in this spot is irrelevant and Each chapter concentrates mostly on one main grammati- may even confuse the reader. cal matter, which leads to a clear-cut arrangement. However, Also in this chapter the broken plural accusative pause this positive claim is impaired because the theme of diptotes form should be supplied with an alif᾽ , not without it as MA has been postponed to the very last chapter. The very central does and this ʾalif is pronounced an, both in pause form and and characteristic Arabic grammatical feature of case-end- in the corresponding full form. The only difference here with ings is now just rendered in a very compressed scheme in the full form concerns the nunation that is not visible in Chapter 2, not systematically as a topic or chapter on its own, pause forms. Besides, the author did not take the chance to as the treatment of this feature would deserve. mention the fact that words ending with hamza or tā ᾽ marbūṭa Chapter 6 combines ʼinna and kāna, together with their never receive a final alif᾽ in accusative forms.3) so-called ‘sisters’. This concurrence in one chapter is justi- In the scheme on p. 2 the author presents four possible fied because they share particular predicates and subjects forms according to the heading of form: pause and full form, only named in relation to them. The learning point here is to and to the heading of number: singular and plural (irregular enable the students to distinguish subjects and predicates in broken). Other plurals and the dual are treated in Ch. 2. In nominal sentences starting with the particle ʼinna as well as this scheme only indefinite forms are represented; their defi- to make such a distinction in verbal sentences with the verb nite counterparts are sorely missing here. So I consider this kāna. scheme and the accompanying explanation rather defective Each chapter consists of the following parts: Introductory and incomplete. Arabic text(s) with highlighting of the forthcoming main In Ch. 2 about noun-adjective and iḍāfa᾽ phrases I was very grammatical matter, Form (equalling morphological ele- surprised that MA does not explicity4) mention the important ments), Use (resembling syntactical elements), Other points characteristic Arabic grammatical rule that non-human plural to note (including references to topics and also sections of nouns behave as singular feminine. This rule should be noted Learner’s Guide), and finally Exercises (always five of in Context as Learner’s Guide does (on p. 64). them). In Ch. 3 about present and future tenses of the verb it It’s my opinion that the organization of the chapters is strikes me that MA, while he introduces here the term ‘nomi- very logical and analytical. The ‘Introductory text’ is not nal sentence’ (in ‘Other points to note’), doesn’t mention the only drawing the reader’s attention to the following gram- term ‘verbal sentence’. It would have been easy to do so after matical topic but also preludes on the strengthening of the the last bullet of this part (p. 18), starting with “If the verb recognition skills to be executed by students in this chapter, precedes the subject (…)”. The accompanying examples are which reaches its climax in the final exercises. In the subse- restricted to humans, singular and plural only, not dual, while quent part ‘Form’ the main grammatical point is explained non-humans are not accounted for at all here. Besides, the in a very lucid way which must be easy to grasp for the target grammatical terms ‘nominal sentence’ and ‘verbal sentence’ readership of this book. The following section ‘Use’ enables the student to build further on the fundamentals laid before, keeping their attention to the main point. The theoretical part 3) MA could have done so, for example by making use of the second text of this chapter on p. 2, line 6, where the accusative form masāʾan (‘in the evening’) occurs, ending with Hamza but without final ʾalif. 2) Learner’s Guide, Appendix D ‘Identifying the Root and Looking up 4) Implicitly application of this rule is shown in this chapter, in the text Words in the Dictionary’, pp. 363-371. on p. 8, line 6, in al-ʾansām-u l-qawiyyat-u (‘the strong breezes’). 415 BOEKBESPREKINGEN — ARABICA 416 are hardly to be found in the Index; you will not catch them I miss examples of geminated verbs and also of verbs with under ‘sentence’, as you might expect, but surprisingly under hamza as first radical. ‘subject’. On both p. 97 and p. 99, the author uses in the examples In this chapter I would have appreciated that MA had verbs with t as third radical. Such verbs may hamper the included the generally known interchangeability of the sub- student in identifying the verbs in the right way, because they junctive particle an᾽ together with the following verb in sub- can easily be confused with verbs in the past tense. junctive mood on the one hand, with the verbal noun of that On p. 98 sub d. MA remarks “Other verb forms, such as verb on the other. derived verb Forms II through X, require no hamza in initial In Ch. 5 “The nominal/verbless sentence” I miss exam- position (…)”. This remark is not correct for derived Form ples of gender and number agreement for human feminine IV which retains hamza in initial position, carried by alif᾽ [alif᾽ plurals, and for non-human plurals. On the other hand I do al-qaṭʽ, WvT]. appreciate in this chapter the examples where MA makes a In Ch. 15 about the passive voice of verbs, the author clear difference between noun-adjective phrases and nomi- shows schemes of the passive voice in present and past nal/verbless sentences. tense7) (in the ususal third person masculine singular form). On p. 32 the verb laysa is only introduced as a particle of The schemes could have been more advantageous if they negation; in Ch. 6 MA makes implicity clear that laysa is were confronted directly with the corresponding tenses in the conjugated as a verb, belonging to the ‘sisters’ of the verb active voice because the student will then be more aware of kāna. the changes in vocalization. In this scheme (p. 105) there is In Ch. 6 about ʼinna and kāna, together with their so- a typo in Form I: for fuʽilu one should read fuʽila, and on called ‘sisters’, MA could have used the sentence “We p. 107 iḵtīra for iḵitīra. entered the evening while far from home” (p. 42) to refer to In this chapter I miss the alternative construction to Ch. 8 where other examples of such ḥāl-constructions are express the passive voice of a verb by means of the auxiliary dealt with more extensively. verbs jarā (or tamma), in active form, followed by the verbal In Ch. 7 about the verb group of kāda and its sisters MA noun of the verb in question.8) gives in ‘Other points to note’ two examples of their usage I would like to add that there is a possible choice to as intransitive verbs, but according to the renowned diction- express the active voice of a verb by means of the active ary of Hans Wehr both of them can also be transitive. In the auxiliary verb qāma (active form) followed by the preposi- examples this transitivity is ‘hidden’ slightly because both tion bi-, with attachment of the verbal noun of that verb.9) verbs (i.c. kāda and šaraʽa) are combined with prepositions. This possibility is not recorded by the author. In the first example the verb kāda is used with the root k-y-d, Ch. 17 “Conditional sentences” sub b., juncto 3 (p. 125) (not with root k-w-d as in the other examples of this chapter) contains a typo: for ‘present/imperfective part’ read ‘present/ with a totally different meaning, i.c. ‘plot (against)’. imperfective form’. Verbs from the group kāda can also be used in ḥāl- The author suggests in Ch. 19 about the vocative, the pos- constructions to describe simultaneous actions or states, sibility that an indefinite noun may be followed by a personal a possibility that is lacking now.5) pronoun, i.c. ‘my’. However, in such a case the noun under Ch. 8 “Relative pronouns and sentence adjective” has a consideration must be definite according to the remark of typo on p. 57: for m-na one should read ma-n (‘whom’). Ch. 1 (p. 2) where it is stated: “Naturally, use of a pronoun In Adverbs of time and place, the topic of Ch. 10, the suffix marks nouns for definiteness (such as ‘my book’ and author could easily have added the non-derived adverbs qaṭṭu ‘her book’).” ‘never’ and ṯumma ‘then, thereupon’. The two regular patterns for exclamation, the issue of In Ch. 13 about Adverb of emphasis/cognate accusative Ch. 20, are wrongly vocalized: ᾽afaʽil bi – should be ᾽afʽil the author depicts in a very clear way the process of transla- bi – and for mā afaʽala᾽ one has to read mā afʽala᾽ . tion for non-Arabs, in several examples, e.g. on p. 89 starting In Ch. 21 ‘Apposition’ MA shows in a quite refreshing with literal word-by-word translation ‘I stand with him the way the adjectival use of ibn ‘son’, and ibna, bint ‘daughter’, standing of the helper’ directly followed by the correct Eng- namely as adjective of the preceding noun. Grammatically lish phrase ‘I stand by him as a helper’. I have seen many spoken the use of the term ‘adjective’ in these cases is not examples of this pedagogical working method througout the correct, but here as a rule of thumb it turns out right. book. In Ch. 22 (the final chapter) about Diptotes the author In the chapter about the (negative) imperative (Ch. 14) the states (p. 171): ”Diptote rules do not apply when diptote ample scheme of imperative verb forms (pp. 95-6, Forms words occur definite (i.e., either by means of the definite I-X6), for all second persons) strikes me as compared with article or the ᾽iḍāfa structure)”. MA should have bracketed the very sparing use of schemes of other verb forms (II-X, here also the possibility of definiteness by means of a per- for all persons) in the book. Verbal nouns (II-X) and Partici- sonal suffix as he correctly does in Ch. 1. It would have been ples (I-X) are missing altogether. Such scarcity contributes desirable if the sentence was concluded with something like: heavily to my view regarding the book under review as less “thus behave then as ordinary triptotes.”. suitable for references. For schemes of this kind the reader should consult Learner’s Guide or another basic grammar.

7) Forms VII and IX are excluded from the scheme, because they are 5) See also A Reference Grammar of Modern Standard Arabic (Karin already passive. However, this reason for exclusion is not mentioned by C. Ryding, First published 2005, Cambridge University Press), pp. 451- MA, neither in Context, nor in Learner’s Guide. 454, about the different uses of kāda c.s.. 8) E.g. in Media Arabic (Julia Ashtiany, Reprint 2007, Edinburgh Uni- 6) Form IX is excluded from the scheme because of its passive meaning, versity Press), pp. 29-30, 45. as explained by MA in Learner’s Guide (p. 218). 9) Media Arabic p. 29. 417 BIBLIOTHECA ORIENTALIS LXXV N° 3-4, mei-augustus 2018 418

This chapter could have been a proper place to propose a But then I consider the following remark in the same exer- resuming overview of the various Arabic noun types includ- cise to be very applicable in general for students: “Focus on ing their declensions (for case, number, and definiteness), but the most essential words [in the nominal/verbless sentence, the author has not seized upon this opportunity. WvT] that comprise the subject and the predicate.”. Ch. 11, Exercise 2, contains examples of non-human plu- Now I will focus on the structure of exercises, the greater rals that must behave grammatically as singular feminine part of Context. (this rule is not stated in Context, see my remark about Exercises 1 of each chapter require a direct application of Ch. 2). Furthermore, in the Key to the exercises the author the previously investigated topic. So these are, naturally, overlooks the required helping vowel, which is -i, after a exercises on the lowest level. verb in past tense, for singular feminine third person, when In Exercises 2 the student is asked to translate a given followed by the definite article al-. English text, of course focusing on the main grammatical Ch. 14, Exercise 4 (sub 5) contains an imperative of the matter of the corresponding chapter, into Arabic. Fortunately geminated verb ʽadda (‘to count’). In the Key only ʽudda is he/she can check the produced translation by means of the, mentioned, but not the other possibility of uʽdud. to this purpose, very satisfactory Key (to the exercises). Ch. 17, Exercise 3, contains conditional sentences with Exercises 3 are always introduced with the phrase: “Iden- only the jussive mood of the verb in the result clauses, at tify the errors, if any, and provide the necessary corrections.” least according to the Key. The alternative usage of the indic- With this exercise the student is forced to reflect further upon ative mood here is absent in the Key, while the chapter pro- the investigated item of the chapter. vides this possibility explictly in some examples (like sub b., Generally spoken Exercises 4 ask the student to recognize juncto 3 on p. 125). and/or to produce alternatives of the previously presented The exercises are concluded with the chapter “Additional issue(s). texts for review”, consisting of seven texts including one Finally, the most sophisticated Exercises 5, demand flaw- poem, which are spread over ten pages. These texts and the less recognition of the highlighted grammatical topics in a posed questions about them require a sufficiently improved text, or rather context, comparable with the introductory proficiency of the main grammatical matters in the previous text(s) of the chapter. 22 chapters of the book, both theoretically in the grammati- Regarding the contents of the exercises I want to mention cal parts and practically in the Exercises. in advance that it disturbs me seeing a considerable number The following “Keys” provide an indispensable means of of exercises where grammatical matters occur that are dealt control for the exercises contained in the chapters as well as with only in later chapters and therefore blur the core matters in the “Additional texts for review”. The ability to control is of the concerning chapter. I give here a small selection of further strengthened by rendering the Arabic elaborations examples of such blurring: with full vocalization. For Exercises 2 of each chapter the Ch. 1 and 2: verbs and participles included in exercises Keys supply correct translations of the English texts, which while they are not treated yet in Context. is really useful to the reader. Ch. 4: diptotes occurring here while it is the main gram- In my opinion the Keys would have been really invaluable matical matter of Ch. 22. if MA had given here correct translations of all Arabic texts Ch. 5 contains already participles (only as of Ch. 9). including those of the Introductory part of the chapters, Ch. 6, i.c. Exercise 1 contains Apposition (Ch. 21); in because non-Arab students, the target readership of the book, Exercise 5 Adverb of manner (Ch. 9), Adverb of specifica- experience as one of the major difficulties in the process of tion (Ch. 11), Adverb of cause (Ch. 12), Conditional Sen- translation the transposition of a word-by-word translation tence (Ch. 17). into a smoothly running one. This is certainly true in the case Ch. 7 in Exercise 5: Adverb of manner (Ch. 9), Exceptive of occurring verbal nouns (=maṣdars), relative adjectives10), sentence (Ch. 18). participles (both active and passive), and also for Arabic Ch. 9 in Exercise 5: Passive voice (Ch. 15), Exceptive expressions (or their absence) that must be translated with ‘to sentence (Ch. 18). be’ or ‘to have’. Ch. 10 Exercise 5: Cognate accusative (Ch. 13). Conceived as an exercise tool, Context serves that objec- This blurring is a needless obstacle for students in their tive sufficiently. However, the blurring of grammatical mat- endeavour to master Arabic grammar. That is a pity because ters in exercises has made my overall judgment of that appli- the exercises are generally very constructive to this purpose. cation less positive. The capability as a grammar is rather On the other hand, it is not that bad because the student must restricted by showing too many imperfections, and as a refer- already have obtained some familiarity with these grammati- ence tool it is less appropriate because of incompleteness of cal topics in Learner’s Guide or another basic introduction the indexed grammatical terms. Furthermore Context meets to Arabic grammar. only partially the necessities of a translation device for non- Now I move on to other points to notice with respect to Arab students. In the end I regard the reviewed hardcover the exercises: version as very expensive; the paperback version (about Ch. 5, Exercise 5, a poem, contains a number of sentences ₤ 30) is much better affordable. with subject solely (e.g. if it is accompanied by question words or negation particles) without predicate while the stu- As I promised I will now make some remarks on the dent is asked here to “identify the two constituents (subject referred work of the same author, in short Learner’s Guide, and predicate) in nine nominal/verbless sentences.”. I am of the opinion that this formulation of the question is rather 10) E.g. In Arabic a relative pronoun must be omitted, when the noun confusing, and, again, usage of a poem here complicates the modified/described by a sentence adjective is indefinite (examples on exercise unnecessarily. p. 55). 419 BOEKBESPREKINGEN — ARABICA 420 or completely: ALHAWARY, M.T. — Modern Standard — Nice elaboration of the writing rules of hamza in Ch. 1, Arabic Grammar: A Learner’s Guide, Wiley-Blackwell, very well completed with Appendix B (see above). Chicester (UK), 2011. (24,5 cm, 418). ISBN 978-1-405- — Very good and comprehensive elaboration of Arabic ver- 15501-4. ₤ 30.00 (Paperback). bal tenses in Ch. 4. MA states that the target readership for Learner’s Guide — The overview of nominal sentences in Ch. 4 is also far- covers Arab second language students “from beginners to reaching. advanced proficiency levels” (p. xvii). With this statement — Relative pronouns including the dropping of them in MA gives support to the view that Learner’s Guide is meant Ch. 5 are evaluated in a very clear way. as an introductory tool for grammar as well as a reference On the negative side I have also made some remarks: tool for Context, but with regard to grammar Context adds — As in Context MA seems to show in Learner’s Guide hardly anything to Learner’s Guide. And so it is, because the some reluctance in giving sufficient attention to the basic Table of Contents encloses 22 chapters, exactly as many as issue of declension, i.e. the different case-endings of Context and covers globally the same range of topics, but in nouns and adjectives, because as late as the last chapter Learner’s Guide the grammatical issues are dealt with more deals with this issue, and then only to a limited extent. profoundly. The organization of the chapters is very well laid — A basic grammar as Learner’s Guide should have down in this Table with a detailed subdivision of all chapters deserved an exhaustive analysis of the circumstantial which enhances the referential quality of the book. Attach- factors determining the different endings including ments to the core grammar in the chapters are added as their morphology in a straightforward way. Then ele- follows: mentary questions as: ‘why should one use in situation — ‘References and further reading’ (containing only Arabic X the nominative (or accusative or genitive)?’ would titles). be answered while now only the case-endings are — ‘Appendix A: The geometric basis of Arabic numerals’. showed scattered throughout the book without such Despite this intriguing title the content (taking just one motivation. page) does not appeal to me in any way. Therefore I — The very important grammatical issue of circumstantial regard this appendix as superfluous. clauses, the so-called ḥāl-constructions, is looked upon in — ‘Appendix B: The writing of the hamza’. This attachment a very restricted way, confined to the description is very useful for the target groups, with a very clear, “Adverbs of manner, expressed by a word, phrase, verbal systematic analysis of this – usually seen as difficult – sentence, or nominal sentence” (Ch. 9, p. 154). issue. — The first paragraph of Ch. 19 “The Vocative”, concern- — ‘Appendix C: The Phoenician alphabet’ (taking just like ing the vocative particle yā, receives too much attention Appendix A one page) does not add anything to the core in relation to more important grammatical matters. matters. — The important verb kāda and its ‘sisters’, which gets — ‘Appendix D: Identifying the root and looking up words some attention in Context, is not even mentioned in in a dictionary’. This appendix provides real additional Learner’s Guide. value. The explanations are very understandable, and, — To the grammatical item of connectors/conjunctions, again, very systematic. It is a pleasure to read this part. which often poses problems for students of Arabic, MA — ‘Appendix E: The number phrase’. Traditionally this dedicates no distinct attention, probably supposed to be topic is regarded as very difficult and boring for students. understood in his treatment of other grammatical It seems to me a wise decision of the author to pay atten- topics.11) tion to this matter in a separate appendix, as he does, in — As noted already in Context, MA neglects the need of order not to aggravate the treatment of the more central helping vowel which is -i, after a verb in past tense for grammatical items. singular feminine third person, when followed by the — ‘English index’ referring to the pages of mentioning of definite article al-. I also will note here the omission of the grammatical terms. the helping vowel ū required after a verb in plural mas- — ‘Arabic index’ of the grammatical terms (arranged alpha- culine second person (ending with -tum), when followed betically, not on root). Both indices do not cover all of by an object pronoun (by means of a suffix). Further- them. For instance a very obvious notion like ‘definite more, no attention is paid to the deletion of ᾽alif if the article’ (al-ta‘r‘īf) is unexpectedly not included. verb in plural masculine third person is followed by a I consider it a shortcoming that a Glossary of grammatical pronoun suffix. Because MA does not demonstrate any terms is missing here; in the forementioned reference works specific consideration to the necessity of suffixation of of Ryding and Badawi, Carter and Gully such glossaries are verbs in past tense, standing on their own e.g. ‘they (m) recorded. The taking up would have greatly improved the wrote’, he presumably has overlooked this feature in referential capacity of Learner’s Guide. extended sentences like ‘they wrote it’. The approach to Arabic grammar in Learner’s Guide has — There are very few typing errors throughout the book. been described by MA in a very systematic, analytical and However I noticed on p. 275 (first line of 15.5: “A pas- clear way. There is an abundance of examples which are very sive participle (…)” that should be: “An active participle helpful for a good understanding of the grammatical topics. For the target readership of non-Arab students, or rather Eng- lish speaking students, the illustrated comparisons on some 11) According to the Introduction of The Connectors in Modern Stand- topics with the corresponding English grammar will probably ard Arabic by Nariman Naili Al-Warraki and Ahmed Taher Hassanein (1994, American University in Cairo Press) mastering of writing skills of provide them with a pervading knowledge of Arabic. Fur- connectors is indispensable for intermediate and advanced students who are thermore, on the positive side of the scale, the following studying Arabic as a foreign language [exactly the same target readership points deserve also to be mentioned in my opinion: of Context and Learner’s Guide!, WvT]. 421 BIBLIOTHECA ORIENTALIS LXXV N° 3-4, mei-augustus 2018 422

(…)”. On p. 288 (third line of 16.1.8) “every” should be 1840, “The truth is, that at this present writing, the ‘Arabian replaced by “all”. On p. 294 in the second example for Nights’ is the most popular book in the world.”1) “you/yourself came” read “she/herself came”. Edward William Lane’s The Thousand and One Nights began to appear in serial parts in 1838 and was published in In general I consider this book a suitable learning device three volumes, 1839-41.2) Though cruelly derided by both for students that start to learn Modern Standard Arabic up to John Payne and Richard Burton, Lane’s Nights is a reason- intermediate levels of proficiency. More advanced students ably accurate, closely annotated translation, albeit an incom- are only partially served well by Learner’s Guide, whether plete one. Contemporary readers found its biblical mode of or not in combination with Context. Gaps could or rather English easily readable and suitably Oriental in flavour. Its should be filled with one of the cited reference works and/or copious wood engravings, made by some of the most skilled additional lessons from a professional teacher. practitioners of the day, were widely admired, although they had little direct relevance to the text. Lane’s translation Zuidhorn, Wim van Thes remained paramount during most of the rest of the nineteenth April 2018 century, and it still has admirers today. Four decades after the appearance of Lane’s translation John Payne published The Book of the Thousand Nights and * One Night.3) He followed that with presentations of more * * stories from the Nights in Tales from the Arabic (1884-89)4) and Alaeddin and the Enchanted Lamp (1889).5) A some- what eccentric Englishman who operated socially on the HORTA, Paulo Lemos — Marvellous Thieves. Secret periphery of the Pre-Raphaelite circle, Payne established his Authors of the Arabian Nights. Harvard University literary reputation in 1878 with a well-received translation of Press, London, 2017. (24 cm, VII, 363). ISBN 978-0- François Villon. A poet of at least marginal merit, he did an 674-54505-2. $ 29.95; £ 23.95; € 27,-. exceptionally good job in presenting the poetry that is The origin of Alf Layla wa-Layla, A Thousand Nights and embedded throughout the Nights. Though Payne’s translation a Night, lay in clusters of stories that circulated around the was soon eclipsed by Burton’s—something Payne came to Middle East and beyond for centuries before being compiled resent—Payne’s was well received in its moment. The fact into surviving manuscript versions. The tales were organized that its limited print run was quickly oversubscribed undoubt- within an endlessly adaptable and highly suspenseful framing edly motivated Burton, ever on the lookout for literary and tale. Shahriyar, king of India and China, having discovered financial success, to complete his own translation. his wife’s sexual infidelity, resolved to take a new, virginal Richard Francis Burton’s The Book of the Thousand bride every night, then have her executed the following Nights and a Night (1885)6) is by far the most famous of all morning to ensure that he would never again be so betrayed. English translations of the Arabian Nights. Its ten volumes With the populace reduced to despair and the supply of eli- were followed by six more volumes of Supplemental Nights gible brides running low, the young maiden Shahrazad to the Book of the Thousand Nights and a Night (1886-88).7) offered herself in marriage to the king. Once in the nuptial Burton’s translation was both popular and controversial with bed she spun stories so enchanting that he repeatedly post- its affected archaic English and explicit sexual passages. poned her execution to hear their outcomes and continua- Like Lane, Burton extensively annotated his translation. tions, for one story always led into another. Eventually he Despite his oft-repeated loathing for Lane’s translation, Bur- spared her life entirely and the kingdom was relieved from ton wrote, “The student who adds the notes of Lane to mine his deadly marital practice. Such a flexible framing tale read- ily incorporated new tales and variations. The structure of the frame even allowed interpolations of stories within stories. 1) Leigh Hunt, “New Translations of the Arabian Nights,” London and As the tales circulated through different times and places, Westminster Review 33 (1840):106. 2) Edward William Lane, The Thousand and One Nights, Commonly they picked up local colour and detail. That was what made Called, in England, the Arabian Nights’ Entertainments. A new translation them so irresistible to Edward William Lane and Richard from the Arabic, with copious notes, 3 vols. (London: Charles Knight and Francis Burton. Co., 1839–41). Shahrazad’s persistent voice acquired new resonance in 3) John Payne, The Book of the Thousand Nights and One Night, 9 vols. (London: Villon Society, 1882-84). the eighteenth and nineteenth centuries when her tales were 4) John Payne, Tales from the Arabic of the Breslau and Calcutta translated into European languages, for to translate the sto- (1814-18) editions of the Book of the Thousand Nights and One Night not ries is to tell them anew. First came Antoine Galland’s occurring in the other printed texts of the work, now first done into English French translation Les mille et une Nuits, published in twelve by John Payne. (Alaeddin and the Enchanted Lamp; Zein ul Asnam and the King of the Jinn: two stories done into English from the recently dis- volumes between 1704 and 1717 to an immediate, sensa- covered Arabic text by John Payne.), 4 vols. (London: Villon Society, tional reception. Within two years an anonymous Grub Street 1884-89). writer produced a pirated translation from Galland entitled 5) John Payne, Alaeddin and the Enchanted Lamp (London: Villon The Arabian Nights Entertainments, a persistent title that lin- Society, 1889). 6) Richard F. Burton, The Book of the Thousand Nights and a Night: A gers today as The Arabian Nights, and sometimes just the Plain and Literal Translation of the Arabian Nights Entertainments, Now Nights. More English versions appeared over the course of Entituled The Book of the Thousand Nights and a Night with Introduction the century and beyond, perhaps the most notable being the Explanatory Notes on the Manners and Customs of Moslem Men and a one by Jonathan Scott in 1811. These were all translations Terminal Essay Upon the History of the Nights, 10 vols. (Benares: Kama Shastra Society, 1885). from Galland’s published text, not from an original Arabic 7) Richard F. Burton, Supplemental Nights to the Book of the Thousand manuscript; nevertheless, they were intensely popular, so Nights and a Night with Notes Anthropological and Explanatory, 6 vols. much so as to prompt the writer Leigh Hunt to proclaim in (Benares: Kama Shastra Society, 1886-88). 423 BOEKBESPREKINGEN — ARABICA 424 will know as much of the Moslem East and more than many least of what is most wanted, the dialect of Egypt and Syria. Europeans who have spent half their lives in Orient lands.”8) His prose is so conscientious as to offer up spirit at the shrine Burton’s notes are, however, more anthropological and per- of letter; and his verse, always whimsical, has at times a sonal than Lane’s, and sometimes less pertinent to the main manner of Hibernian whoop which is comical when it should text. Despite and because of its controversial reception, Bur- be pathetic.”10) Lane’s assessment of Torrens was much ton’s became the classic English translation of the Arabian higher than Burton’s. News that Torrens was embarking on Nights—a term Burton hated—though, like many classics, it a translation caused him great anxiety. “I have two rivals; is sometimes more praised than read. one in Germany [Gustav Weil], & another, a very formidable Horta’s primary objective is not to examine the mechanics one, in India [Torrens],” he wrote to his friend Robert Hay or accuracy of those translations. Instead, he writes in his in January 1838.11) But Torrens also considered Lane to be Introduction, “Marvellous Thieves examines three of the a formidable rival and stopped after publishing the first of most influential European translators of the story collection what was intended to be a nine or ten volume work.12) with reference to their unacknowledged reliance on other sto- Although it is a slight exaggeration to say, as Horta does, rytellers, translators, and cultural insiders. … I trace the pro- that “Torrens’ translation has been largely forgotten in the cess by which the tales of Shahrazad were claimed as a ter- history of the Arabian Nights (91),” it has never been ritory of European authorship by an erasure of the many accorded a level of recognition on par with the translations collaborative relationships that made these three versions of of Lane, Payne, and Burton. Horta believes that both Payne the stories so distinctive (2-3).” Horta means the versions and Burton were heavily influenced by Torrens and that they of Galland, Lane, and Burton, but he might well have said drew on his translation in preparing their own. The latter “four” because he also devotes considerable attention to point should not come as a surprise in the case of Burton John Payne, without whose translation Burton’s cannot be who confessed in “The Translator’s Foreword” to his trans- fully assessed—or even “five” because of the amount of lation that he made “ample use” of it, along with Lane’s space he gives to Henry Whitelock Torrens who made a and Payne’s translations.13) Horta insinuates a degree of notable but incomplete one-volume translation. resentment and jealousy toward Torrens by Burton (252) but Some of the most interesting and informative passages in there is no evidence for that. On the contrary, although the Marvellous Thieves are those about Antoine Galland (1646- two men served at opposite ends of India and never met, they 1715) and his translation of the Nights, or to be more precise had much in common with their manifold interests that about the Syrian traveller Hanna Diyab and his contributions sometimes ran too far afield and their problems with author- to that translation. Until recently Diyab was a shadowy fig- ity. Just as Burton’s Indian career was scarcely a success, ure, but during the past two decades or so both Galland and Torrens’ ended with a dead-end posting at Murshidabad. His Diyab have been the subjects of increasing attention with a death from dysentery at age 46 ensured that he and Burton number of new discoveries and publications. These illumi- would never meet or correspond. nate many aspects of a remarkably adventurous life that When Horta moves to Edward William Lane (1801-76) impacted not only Galland but also other French scholars and and his translation, he is on less sound ground, and his text writers such as Paul Lucas. Among the new finds are Diyab’s begins to display the numerous factual errors that mar much travel diary, discovered in the Vatican Library and published of Marvellous Thieves. Horta is repeatedly and consistently in 2015. Drawing on those newly available sources, properly wrong (8, 136, 179, 196) about the dates of Lane’s second acknowledged, Horta suggests that Diyab’s role in Galland’s trip to Egypt, a crucial time in his career when he completed translation was much larger than previously recognized. That composition of his important work, An Account of the Man- was especially so in regard to the so-called “Orphan Tales.” ners and Customs of the Modern Egyptians (1836), and gath- Those are stories included in Galland’s later volumes, after ered materials that would appear in his translation of the he met Diyab, that have no identifiable Arabic manuscript Nights. According to Horta (124) the first volume of Lane’s source. They include two of the most famous tales, Ali Baba Nights appeared in 1838. That is incorrect: Lane’s translated and Aladdin. Diyab, a master story-teller himself, recounted tales began to appear in monthly parts on 1 May 1838. The them to Galland who added touches of his own. Horta is first volume was published in 1839. The sensational exhibit justified in stating, “The mixture of different cultural tradi- by Giovanni Battista Belzoni in Piccadilly that may have tions and narrative strategies within the text [of Galland’s inspired Lane was at the Egyptian Hall, not the “Egyptian Nights] is the product not just of ‘translation’ but of the House (135).” Astonishingly, Horta writes “According to his reworking of stories by figures like Diyab, who already own diary entries, Lane was decidedly unprepared for the embodied the overlapping worlds of East and West (21).” role of Egyptian explorer when he arrived in Cairo in 1825 Horta’s pages about Henry Torrens9) (1806-52) provide (134).” That was simply not the case. Although Lane’s diary an informative picture of a man of literary talent and schol- entries often express surprise at the new environment, it has arly bent who served as secretary and vice-chairman of the long been established that he conducted a prolonged and Asiatic Society of Bengal—not yet the Royal Asiatic Society of Bengal as Horta gives its name (105, 109, 227) because the institution was not so renamed until 1936. Burton affected 10) Burton, The Book of the Thousand Nights and a Night, 1:xi. 11) Edward William Lane to Robert Hay, 15 January 1838, Bodleian not to take Torrens and his translation altogether seriously, Library, Department of Western Manuscripts, Ms Eng. lett d 165, f. 99. writing that “the plucky author knew little of Arabic, and Quoted in part by Horta on 159. 12) Henry Whitelock Torrens, The Book of the Thousand Nights and One Night: From the Arabic of the Ægyptian Ms. as edited by Wm. Hay 8) Burton, The Book of the Thousand Nights and a Night, 1:xix. Macnaghten, Esq., B.C.S. Done into English by Henry Torrens, B.C.S., 9) His full name of Henry Whitelock Torrens should be provided to B.A., and of the Inner Temple (Calcutta: W. Thacker & Co.; London: distinguish him from his similarly named father and son who also did W.H. Allen & Co., 1838). important military and imperial service. 13) Burton, The Book of the Thousand Nights and a Night, 1:xiii. 425 BIBLIOTHECA ORIENTALIS LXXV N° 3-4, mei-augustus 2018 426 intense program of preparation in England that enabled him of the manners and customs of the Arabs, and particularly to settle quickly into Cairo and begin collecting materials for those of the Egyptians.”15) It was also an opportunity to use his intended book. Within eight months of his arrival in ethnographic material that he had gathered in Cairo but could Egypt he was able to set off on the first of his two voyages not fit into Modern Egyptians, a work that had already pro- of exploration into Upper Egypt and Nubia. Some of Horta’s vided him with ample authority to publish a translation of the mistakes, and there are others, may seem minor, but collec- Nights. Having suggested that Lane’s annotations played a tively they undermine confidence in reliability of research peripheral role, Horta then asserts that “As Lane proceeded and accuracy of presentation. with his version of the Arabian Nights, the translation was In an uncomfortably large number of instances the sources increasingly treated as a function of his goals as a commenta- Horta cites turn out to have only tangential pertinence to his tor, rather than the reverse (163).” Again that gets it back- text, and occasionally none. As someone who has written wards. From its profuseness in the first volume, Lane’s anno- about Lane I was surprised to see one of my articles cited tation steadily decreases through the second and diminishes following the statement “Social behavior that diverged from even more in the third. Much of the “Notes” in that third what Lane understood as appropriate for a Muslim society volume are stories, not notes, that Lane presented in that was particularly difficult for him to process (147).” That arti- smaller format, probably to save space. Then, in a further cle, which is about passages that Lane’s publisher urged him inconsistency, Horta writes that “Scholars of Lane’s work to omit from Modern Egyptians, does not support that state- have seldom acknowledged just how comprehensively he ment in any way. Further down (155) Horta cites my book prioritized the production of commentary over the literary about Lane when he states that Lane “had been forced to labor of translating (166).” While there have been few schol- take employment with his friend [Robert] Hay,” but my ars of Lane’s work, the importance he placed on his notes book says no such thing; on the contrary, it makes it clear has assuredly and explicitly been acknowledged.16) Many are that Lane was helping Hay gratis, for friendship, not remu- remarkable ethnographic essays in themselves, not the neration. Horta makes the same mistake in describing the “indistinguishable mixture of fact and fiction” concocted artist Pascal Xavier Coste as an employee on Hay’s team in from a “promiscuous mixture of sources” that Horta presents Egypt (168). Coste never worked for Hay; he merely sold them to be (166). Hay a portfolio of drawings. Then he cites (174) another of But Horta’s primary focus on Lane is the “suppressed my articles following the statement that Lane “began to voices” in his published work. That Lane occasionally under- rewrite his diary entries to conceal the importance of those stated the roles of his assistants, informants, and helpful friends and acquaintances who had made his life and work friends is a well-established fact, but he never consigned in Cairo possible.” Nothing in my article supports that them to oblivion, and he usually mentioned them favourably. assertion. That was certainly the case with Sheikh Ibrahim Abdul There is an element of the straw man in Horta’s depiction Ghafar al-Dasuqi, Lane’s valuable language assistant, manu- of Lane as someone excelled only by Richard Burton as a script transcriber, and cultural informant in Cairo during his “consummate self-mythologizer (10)”. That was something third trip to Egypt, 1842-49, when he gathered materials for the shy, retiring man who declined nearly all honours offered his Arabic-English Lexicon. Al-Dasuqi came regularly to to him—perhaps even a knighthood, if family tradition is work with Lane five afternoons—not six, as Horta states correct—certainly was not. Other tendentious statements (172)—every week for years as they collated lexicographical such as “the myth of Lane as Orientalist scholar (134)” set manuscripts and discussed fine points of meaning as well as up Lane as an easier, more subjective target. Horta’s charac- the cultural context of words. Horta writes that Lane would terization of Edward William Lane should be carefully and “hide his collaborative relationship with al-Dasūqī” with the critically examined with cognizance of the three existing bio- result that “The Arabic-English Lexicon would be considered graphical studies of Lane14) because it varies from them Lane’s great legacy, and al-Dasūqī would be remembered in widely while offering no new evidence for variation. Horta Orientalist circles merely as his assistant (173).” But Lane’s shares Payne’s and Burton’s distaste for Lane’s translation assistant is precisely what al-Dasuqi was, and while Lane and diminishes his role and stature as a translator, but a cru- may never have been profuse in giving credit for contribu- cial distinction is left unsaid: Lane made his own translation tions, there is no real evidence that he attempted to hide al- whereas Horta goes on to suggest strongly that Payne and Dasuqi’s. On the contrary, he stated on the first page of the Burton freeloaded off the translations of their predecessors. Preface to volume one of the Arabic-English Lexicon “that Horta presents Lane’s scholarly annotation to his text of a person better qualified for the services that I required of the Nights as contrived cover for “establishing his authority him, than the sheikh Ibráheem Ed-Dasooḳee, could not have as its translator (158)” and avoiding “associating himself been found by me in Cairo; and I had no occasion to employ with the imaginative world disclosed by the tales (157).” any other assistant, except, occasionally, transcribers, under That is getting it backwards. For Lane the annotation was one of the primary justifications for making a translation of a work that he believed presented “most admirable pictures 15) Edward William Lane, An Account of the Manners and Customs of the Modern Egyptians: Written in Egypt during the Years 1833, -34, and -35, Partly from Notes Made during a Former Visit to That Country in the Years 1825, -26, -27, and -28. 5th rev. ed., Edward Stanley Poole, ed. 14) These are A.J. Arberry, “The Lexicographer: Edward William (London: John Murray, 1860), xiv, n. 2. Lane,” in Arberry, Oriental Essays: Portraits of Seven Scholars (London: 16) E.g., “It is no exaggeration to say that in the first half or more of Allen & Unwin, 1960); Leila Ahmed, Edward W. Lane: A Study of His Lane’s translation of the Arabian Nights, the text often serves as an accom- Life and Works and of British Ideas of the Middle East in the Nineteenth paniment to the endnotes. … These annotations are much more than schol- Century (London: Longman, 1978); and Jason Thompson, Edward William arly explications of the text: they are literary doors into deeper perception Lane 1801-1876: The Life of the Pioneering Egyptologist and Orientalist of the details, the textures, the scenes, the manners and customs of the (Cairo: American University in Cairo Press, 2010). Middle East.” Thompson, Edward William Lane, 415. 427 BOEKBESPREKINGEN — ARABICA 428 his supervision.”17) At least two more positive references to through “private study.”22) Horta notes that Payne prepared al-Dasuqi are to be found in that Preface. In his Thousand his Nights in what might be considered a suspiciously short and One Nights Lane wrote a substantial acknowledgement time and, following Poole, writes that “a likely explanation” of the contributions by Sheikh Muhammad al-Tantawi, his is that Payne “was relying on other translations published in principal assistant in that project. “Without the valuable aid English and German to provide the substance of his version which he afforded me, I would not have attempted the of the stories (220).” But when Horta plainly states that translation.”18) Lane also mentioned his friend Abraham Payne “was not sufficiently proficient in Arabic” to make Salamé who offered “his valuable aid, of which I have the translation (219) the appended endnote has no clear bear- availed myself in several cases when I found it desirable to ing on the assertion. He proceeds to make a fairly good case obtain the opinion of a learned native of Egypt.”19) that Payne emphasized literary quality over accuracy, that his Having written that his intention was to expose “unac- translation sometimes significantly differed from the impor- knowledged reliance on other storytellers, translators, and tant Calcutta II source for the original Arabic text, and that cultural insiders (2-3),” Horta should have been cautious he appeared to rely heavily on Lane, whose translation he about his own unacknowledged reliance on the findings of affected to despise, and on Torrens. But nowhere, either with others. That is particularly true in his treatment of Edward Payne or Burton, does Horta actually compare translations William Lane. Horta asserts that “Marvellous Thieves mines with the original sources. The question of whether Payne new evidence to chart these forgotten chapters of cross-­ somehow acquired Arabic more or less in his spare time or cultural encounter, collaboration, and theft. … A careful if he made a pseudotranslation of the Arabian Nights is never study of Lane’s notebooks and correspondence reveals the resolved, yet it has a major bearing on the validity of Bur- importance of a series of intermediary figures who estab- ton’s translation. lished crucial links between the social world of nineteenth- Horta’s mention of Reverend Badger raises another trou- century Cairo and the European travellers and scholars who bling question about contextual accuracy. George Percy pursued its manuscripts and monuments (14).” It might be Badger was a very important person in Richard Burton’s more correct to say that Horta mined the footnotes of other life. An accomplished Arabist, and a most interesting person scholars because most of what he presents about Lane, his in his own right, Badger deserves to be better remembered lifestyle, his informants, his support network, and his work- than he is today, but he should be remembered with correct ing habits has been published before, with full citations. information. Horta identifies him as “a professor of Arabic Horta is clearly aware of these publications because he men- at Oxford (239),” but the self-taught Badger, though one of tions them at various places in his endnotes, but usually in the most accomplished Arabists of his day, never had an regard to peripheral or contextual concerns; the archival dis- institutional connection with Oxford, much less a professor- coveries are presented as new and as his own. To paraphrase ship there. Horta also errs in referring several times to a Richard Burton about attribution, one should not enrich one’s “Terminal Essay” at the end of the ninth volume of Payne’s pages at the expense of others,20) although Burton has occa- translation. Payne’s title for his extended concluding essay sionally been accused of doing precisely that, and he is so was “The Book of the Thousand Nights and One Night: Its accused in Marvellous Thieves. History and Character.” It was Burton who authored a noto- John Payne (1842-1916) turned his attention to the Ara- rious “Terminal Essay” at the end of volume ten of his bian Nights in 1878 after his success with Villon. Early the translation. following year he published a notice announcing his inten- And it is Richard Francis Burton (1821-90) who occupies tion to publish a complete translation and bitterly denouncing the most prominent place among Horta’s marvellous thieves. the translation by Edward William Lane who had died less In a concluding chapter entitled “The False Caliph” Horta than two years earlier. Horta tells how that provoked a stren- ranges across Burton’s life, career, and the various personae uous response from Lane’s nephew, Reginald Stuart Poole, that he adopted. Burton is assigned the role of false caliph who resented the attack on his uncle’s work, and from the because, Horta writes, “Burton found in the figure of Caliph Reverend George Percy Badger, an Arabist with impeccable Harun al-Rashid an analogue for his own restlessness (284).” credentials and a good friend of Richard Burton who also Regrettably, this chapter, like some of the preceding ones, is became involved in what became a bitter dispute that condi- disfigured by a number of mistakes and imprecisions. One tioned much of Payne’s and Burton’s translations.21) Badger occurs when Horta writes, “Perhaps Burton’s greatest feat of especially questioned Payne’s expertise in Arabic, wondering self-invention was to persuade his contemporaries that he how and when he had acquired it. Payne’s biographer (and had translated the Arabian Nights in ten volumes in a single enemy of Burton) recorded only that he learned his impres- year (1885) at his consular residence in Trieste (10).” That sive array of “a dozen or more languages,” including Arabic, undocumented statement is wrong. Both Richard and Isabel Burton made no secret that composition extended over at least two and a half years, and Burton claimed an even longer 23 17) Lane, An Arabic-English Lexicon, 1:v. compositional time, ) a claim some of his detractors have 18) Lane, The Thousand and One Nights, 1:xii–xiii. doubted. Also, the dated dedications and other prefatory mat- 19) Lane, The Thousand and One Nights, 1:xxi. ter in those ten volumes are sufficient to show that he was 20) Richard F. Burton, The City of the Saints, and across the Rocky Mountains to California (London: Longmans, 1861). American edition: (New York: Harpers, 1862), x. 21) Horta writes, “The debate took place in three issues of the Acad- emy” and provides their dates of publication (330, n. 49). Apparently he 22) Thomas Wright, The Life of John Payne London: T. Fisher Unwin, replicated inaccurate information provided by Burton in the final volume 1919), 12. of the Supplemental Nights (6:335-36). By following Burton’s text instead 23) Isabel Burton, The Life of Captain Sir Richd. F. Burton, K.C.M.G., of the original publications Horta got both the number of issues wrong F.R.G.S., 2 vols. (London: Chapman & Hall, 1893), 1:274; Richard Burton, (there were four, not three) as well as the dates of their publications. The Book of the Thousand Nights and a Night, 1:ix. 429 BIBLIOTHECA ORIENTALIS LXXV N° 3-4, mei-augustus 2018 430 not only in Trieste but also in London, Oxford, and Tangier his exam in Arabic (252).” In fact, Burton performed well as he worked on the Nights. on that examination, but it was ruled invalid on a largely Later, when Horta reproduces one of Burton’s famous spurious technicality by one of his rivals in the Bombay quotations in regard to his translation—“I intend to publish Army and the result never entered into the official record. (at Brussels) some day all the excised parts in plain English. The key issue in Horta’s chapter about Burton is whether It will be nice reading for babes and [“&” in Burton’s manu- Burton plagiarized John Payne, a question that has long been script letter] sucklings”—he states that it “can be found in a a festering sore on Burton’s literary reputation. It is now letter to his publisher (240).” That letter is not to any pub- fairly well established that Burton’s Nights contain extensive lisher but to the poet Gerald Massey, not “George” Massey passages that are identical to Payne’s, or nearly so. Burton as Horta writes (246), a well-known literary figure during his had a cavalier attitude toward handling the manuscript mate- time and one occasionally remembered today. Burton and rials of others when they were entrusted to his care. He lifted Massey corresponded for several years about their respective passages from other authors on more than one occasion. Sev- works in progress, especially Burton’s translations of the eral questions persist about Burton’s Book of the Thousand Portuguese epic poet Camões to which Massey affixed a and One Nights, which his subtitle stated was “a plain and poem; but apart from the instance above the Nights are literal translation.” When did Burton begin his translation? scarcely mentioned in their extant correspondence, now To what degree did he translate directly from the Arabic mostly in the British Library. Massey was certainly not “the manuscript sources? Was his translation “plain and literal?” publisher of his [Burton’s] translations of Luís de Camões And, to return to the question at hand, how much did Burton (246).” All three of Burton’s books about Camões were pub- take from Payne and others? lished by Bernard Quaritch, as their title pages clearly state. Horta overstates his role as a trailblazer in the Burton- Burton had just finished the third, The Lyricks, when he Payne plagiarism controversy when he writes that his analy- turned his full attention to the Nights on 1 April 1884.24) A sis of Burton “exposes a chain of plagiarism that has only fourth remained in manuscript. Careful study of these is been hinted at until now (13).” He finds it “puzzling that this essential in evaluating Burton’s attitude toward translation fact is not common knowledge (219).” That is nonsense. In and presentation. 1898, just eight years after Burton’s death, none other than Lack of attention to original sources allows Horta to rep- John Payne wrote in the introduction to his Quatrains of licate an error by an earlier scholar when writing about Bur- Omar Kheyyam of Nishapour that Burton’s translation ton during his consulship at Damascus when he occasionally closely followed his own and that Burton “again and again went into the city in Eastern disguise. Horta states: “One borrowed whole pages in difficult passages, such as are of Arab acquaintance reported that ‘his attempts to pose as a frequent occurrence in the work.” (Horta commits a serious native were a constant source of amusement to all with error by incorrectly dating the publication of Quatrains of whom he came in contact (287).’” That statement is a com- Omar Kheyyam of Nishapour to 1889. That is much more pound of errors. First, the ‘Arab acquaintance,’ who wrote than an inconsequential typo because in 1889 both Sir Rich- under the pseudonym ‘Salih,’ was, as is well known, one of ard and Lady Burton were still very much alive. The result Burton’s friends in Damascus, a British missionary named would have been explosive had they read that passage.) William Wright. Second, Wright was writing not about Bur- Shortly thereafter Thomas Wright publicized the accusation ton’s time in Damascus; he was wondering if Burton could of plagiarism against Burton in his widely read and highly have successfully concealed his Englishness when he made influential Life of Sir Richard Burton (1906) in which he the hajj to al-Medina and Mecca years earlier. Third, the asserted that “Burton, indeed, has taken from Payne at least words Horta quotes did not even apply to Burton but to three-quarters of the entire work. He has transferred many another famous explorer of Arabia, William Gifford ­Palgrave, hundreds of sentences and clauses bodily. Sometime we as Wright’s full statement makes clear: “I never conversed come upon a whole page with only a word or two altered.27) with a Muhammedan who had accompanied Burton on that The following year Wright baldly and publicly proclaimed journey, but I have seen Arabs who saw Palgrave on his way that “Burton’s ‘Arabian Nights’ is the most bare-faced and to Nejed, and his attempts to pose as a native were a constant stupendous piece of plagiarism in literature.”28) He reiterated source of amusement to all with whom he came in contact.”25) his accusations in his 1919 Life of John Payne. In 1938 There are other errors as well and imprecisions that con- D.B. MacDonald plainly stated (Supplement to the Encyclo- vey false impressions; for example, Horta writes that “Burton­ paedia of Islam, lemma “Alf laila wa-laila”, p. 20a) that claimed to speak thirty languages … (10)” Not exactly. Burton’s translation “often reproduces Payne verbatim.” What Burton actually claimed toward the end of his life was Fawn Brodie’s epochal biography of Burton, The Devil that over the course of his career he had learned 29 (the exact Drives, devoted a full appendix to “The Burton-Payne Con- number varies) languages, not that he spoke them all fluently troversy” in 1967. Others have published about it. Those are at any given time. To the contrary, he stated in his Supple- much more than scarce hints. The Burton-Payne controversy mental Nights that “when a language is not wanted for use continues; Horta offers nothing new to elucidate or resolve my habit is to forget as much of it as possible, thus clearing it. He is, however, probably on target when he writes “Bur- the brain for assimilating fresh matter.”26) In regard to Bur- ton would have felt that his contributions to Payne’s project ton’s linguistic prowess Horta slips into another imprecision justified a sense of joint ownership over it (243).” by following an earlier writer and stating that Burton “failed

24) Isabel Burton, The Life of Captain Sir Richd. F. Burton, 2:274. 27) Thomas Wright, The Life of Sir Richard Burton, 2 vols. (London: 25) “Salih” (William Wright), “Burton at Damascus,” The Bookman 1 Everett & Co., 1906), 1:xii. (October 1891-March 1892): 23-25. 28) Thomas Wright, “John Payne and Walter Pater: To the Editor of 26) Burton, Supplemental Nights, 6:331. The Academy,” Academy 72 (6 April 1907): 349. 431 BOEKBESPREKINGEN — ARABICA 432

The “recent archival discoveries (248)” mentioned by general sense of meaning of the overall text. Horta writes Horta in regard to the Burton-Payne controversy are less that Lane’s organizational choices involve “duplicity (173- recent and novel than he suggests. He writes that “there have 74),” an accusatory word with its meanings of deceitfulness long been rumours among collectors and archivists that and double-dealing that is very strong indeed. Sophia Poole’s ­Burton owned a second set of Payne’s work that he dismem- reworking of portions of Lane’s then unpublished “Descrip- bered and used as the basis for his own edition of the story tion of Egypt” for her book The Englishwoman in Egypt is collection. A recent acquisition of the British Library now dismissed as a “new forgery (216).” Yet the Preface to that offers scholars an example of Burton’s method: two small book, to which Poole added much of her own text, tells booklets made from pages cut from Payne’s edition of the precisely how it was composed, not forged, with Lane’s Arabian Nights and covered with Burton’s handwriting participation.30) (248).” Never properly or fully identified, these are actually The endnotation for Marvellous Thieves is sparse for a some page proofs for the third volume of Payne’s Tales from work that can draw on such rich sources. Horta mentions the Arabic. Payne regularly sent his material to Burton at the “the Cambridge Arabist Robertson Smith (who would later proof stage for Burton’s corrections and comments before deny that Payne translated the Arabian Nights) (241).” He going to press. Burton routinely requested two sets of proofs does not give a source for that intriguing statement, but the when he was seeing his own books through press. That connection with Smith, whose name was properly William would have been the case with those at the British Library. Robertson Smith, would be worth pursuing. Smith and Burton received two sets of Payne’s proofs, returned one set ­Burton met in Egypt in 1880, travelled together, and later with his annotations to Payne, and kept the other for his own corresponded. This is likely to have had some bearing on further use. Fawn Brodie wrote about them in the 1960s Burton’s Nights. In another instance Horta refers to “the when they were part of the magnificent collection of Burto- Romantic assumption, espoused by the Orientalist Owen niana belonging to the late Quentin Keynes.29) Several other Jones, that the essential spirit of a people might be found in proof pages from Tales from the Arabic were donated to the an authentic literary text—even more so if it was a work of British Museum years earlier when the British Library was poetry (116).” The uninstructed reader would welcome a still part of that institution. citation for this, and it would dispel any lingering suspicion Although Horta’s characterization of Edward William that Horta may have confused the Victorian architect and Lane as someone who “attempted to disappear into a con- designer Owen Jones with the eighteenth-century Orientalist structed Egyptian identity (215)” is unconvincing, there is Sir William Jones. The endnotes also contain inaccuracies, considerable substance to what he calls “the mythology of some noted above, as when Horta states that “As British Burton (217).” Burton was notable during his lifetime for consul in Damascus, Burton’s failure to serve the interests of projecting different impressions of his personality, and he the local British populace, especially the many missionaries took fiendish delight in portraying himself as worse than among them, doomed his career as a diplomat in the Middle he actually was. These differences have reverberated through East (276).” But one of the secondary sources that he cites the years, resulting in widely different interpretations of a in support of that overly generalized assertion—Burton also complex, many-faceted person. Though one of the most had strong British supporters in Syria, even among mission- remarkable and remarked-on Victorian people, Burton still aries—has nothing to do with Syria at all.31) eludes full definition, despite many biographical attempts. In Marvellous Thieves does not contain a bibliography, a dealing with such an intriguingly complex person who was very useful feature in a work of this kind, and such biblio- involved in so many controversial issues it is almost impos- graphic information as is provided in the endnotes contains sible not to take sides for or against, but partisan impulses glaring omissions. In his list of what he considers the three should be kept in check. Horta reveals his partisanship when or four outstanding biographical studies of Burton, including he mentions in passing Burton’s “famous attempt to take a fictional one (328, 333), Horta does not include Mary credit for the discoveries of his travel companion John Han- S. Lovell’s dual biography of Richard and Isabel Burton, ning Speke in their search for the sources of the Nile (297).” A Rage to Live (1998). Lovell was not a professional aca- This is no place to enter the Burton-Speke controversy demic, and she discounted the notion that Burton plagiarized which, like the one concerning Burton and Payne, will prob- Payne; but no student of Burton can ignore her explosion of ably never die, but Horta’s statement misrepresents Burton’s several major Burton myths, such as the widow’s burning role in it as expedition leader, leaving unsaid Speke’s ques- of all his manuscripts, or the immense new archival material, tionable actions. Like Lane, Horta’s Burton is in some now housed in the Wiltshire and Swindon History Centre at respects a straw man. Chippenham, that she discovered and that that has opened Another fallacy to avoid is history by insinuation. The new dimensions into Burton’s life and career. Horta is aware word “seems” (which in at least one instance occurs in two of Lovell’s book because he cites her twice in passing, once consecutive sentences) and its variations appear dozens of incorrectly, as noted above, and in another instance (265) times in Marvellous Thieves, as does “claimed,” allowing where he implies that she was overly credulous. Further wide play for authorial subjectivity and for speculation to ­reference to Lovell would have saved Horta at least one extend beyond evidence. Surely some of these instances could have been presented on a more conclusive basis. Loaded words and phrases are sprinkled liberally throughout 30) Sophia Poole, The Englishwoman in Egypt: Letters from Cairo, the text with terms such as “admits,” “boasted,” or “he slyly Written during a Residence There in 1842, 3, & 4, 3 vols. (London: Charles inserts” when less charged language would better suit the Knight and Co., 1844-1846), 1:v-vi. 31) This reference is to Mary S. Lovell’s A Rage to Live: A Biography of Richard and Isabel Burton (New York: Norton, 1998), 427, where she 29) Fawn M. Brodie, The Devil Drives: A Life of Sir Richard Burton writes about Burton not at Damascus but in Fernando Po in western Africa, (New York: Norton, 1967), 342. several years before he went to Damascus. 433 BIBLIOTHECA ORIENTALIS LXXV N° 3-4, mei-augustus 2018 434 misreading of the Burton manuscripts in regard to Burton’s The question Andrzej Zaborski (who sadly passed away in attitude toward homosexuality.32) 2014) tries to answer, is whether or not Classical Arabic is A useful insight to take away from Marvellous Thieves is archaic (pp. 35–50). While earlier Semitists considered Clas- that the history of the translations of the Arabian Nights is a sical Arabic to be the most conservative of the Old Semitic tale of thousand and one nights in itself. One story leads to languages, this view later shifted, and it came to be seen as another, with stories nested within stories of adventure, crea- one of the most innovating. While Akkadian is considered by tivity, and occasionally appropriation. New chapters in that some to be the most archaic of the Semitic languages because ongoing process remain to be written. Existing ones are there Akkadian texts are the oldest we have, Zaborski points out for careful, attentive study. that the old age of the records of a language does not neces- sarily mean that the structures of that language are archaic. Copper Hill, Virginia, Jason Thompson Zaborski studies the verbal systems and their components, February 2018 and demonstrates that Classical Arabic has conservative forms as well as innovations. His conclusion is that although he considers Classical Arabic and Akkadian to be the most * archaic Old Semitic languages, “Classical Arabic is * * per saldo even a bit more conservative” than Akkadian (p. 45). The section contains three AL-JALLAD, A. (ed.). — Arabic in Context. Celebrating Arabic in its epigraphic context chapters. The first of these, by Manfred Kropp (pp. 53–74), 400 years of Arabic at Leiden University. (Studies in proposes a new reading and translation of the much-studied Semitic Languages and Linguistics, 89). Brill Academic ῾Ayn ῾Abada inscription, which was found in 1979 and first Publishers, Leiden-Boston, 2017. (24 cm, XIX, 507). published in 1986. The inscription is written in two lan- ISBN 978-90-04-34303-0. ISSN 0081-8461. € 132,-; guages: the introductory formula and protocol are in Naba­ $ 152.00. taean, and the remainder in Arabic. The text is an incantation This volume was published on the occasion of the 400th to the god Obodas, written in highly formal speech. Kropp anniversary of the chair of Arabic at Leiden University. It proposes a new reading of the Arabic text and provides a contains papers delivered at the colloquium held in Novem- new translation. The chapter ends with the presentation of ber 2013 to celebrate this memorable occasion and therefore translations proposed by other scholars. The great variation not only contains papers penned by staff working at Leiden in these translations shows just how difficult the text is to University, but by prominent Arabists from other universities interpret, and Kropp concludes that a better drawing of the as well. The book starts with a foreword by the current inscription is needed. holder of the chair, Professor Petra Sijpesteijn, and a preface The next chapter, by Laïla Nehmé (pp. 75–98), deals with by the editor, Ahmad Al-Jallad. As the latter points out, the inscriptions from the 3rd–5th centuries, written in the transi- study of the Arabic language is in need of a wider, more tional script between the Nabataean script and the Arabic interdisciplinary perspective. The volume therefore presents one. These inscriptions were found in both northwest and sixteen studies, divided into five sections, which write the southern Arabia. The texts are divided into two groups: those history of the Arabic language by “looking beyond the tra- dated prior to AD 275, with evolved y, h, ḥ, and g, and those ditional sources and methods” (p. x). from AD 275–475, which contain more evolved letters and The book starts with the section What is Arabic? This is letter combinations. Nehmé reserves the label Nabataeo- also the question that John Huehnergard attempts to answer Arabic for the latter group. She explores Arabic influences in the first chapter (pp. 3–34). He argues that the innovations on the Nabataeo-Arabic texts by tracing phonological and that (Proto-)Arabic does not share with the other Semitic lan- grammatical features that are characteristic of Arabic, such guages, is what defines a language as Arabic. Huehnergard as the use of the broken plural. Her conclusion is that the takes as his starting point the now commonly accepted divi- people who wrote the texts almost certainly spoke Arabic, sion as described by Hetzron in the 1970s, which argues that although the area “was probably still characterised by a cer- Arabic is a separate branch within the Central Semitic sub- tain bilingualism” (p. 94). group. Huehnergard states that both Classical Arabic and the In chapter 5, “Graeco-Arabica I: The Southern Levant”, colloquial varieties must have a single common ancestor, with almost 90 pages by far the longest of this volume, labelled Proto-Arabic. He gives examples of innovations of Ahmad Al-Jallad discusses the linguistic features of Arabic Proto-Arabic that it did not inherit from its Central Semitic written in the Greek script (pp. 99–186). This is the first ancestor, as well as features that are found in the modern article in a series of four, each covering a different geo- dialects but not in Classical Arabic. Huehnergard concludes graphic area. The chapter starts with a presentation of the that inscriptional Ancient North Arabian cannot be consid- criteria used to identify the Arabic material, after which fol- ered Arabic. Safaitic, however, does share a number of inno- lows the actual study, containing three sections, on the pho- vations with Arabic, and therefore they must share a com- nology of the consonants, the phonology of the vowels, and mon ancestor. Based on this evidence, he suggests a new morphology. The phonology part discussing the consonants subdivision, which instead of having Arabic directly branch- gives a very thorough study of all the possible ways of tran- ing off Central Semitic, has Proto-Arabic branching off Cen- scribing Arabic consonants in Greek, and their implications tral Semitic and then branching into Arabic and Safaitic. for the pronunciation of these sounds in that period of time. As for the phonology of the vowels, one of the advantages of the Greek script is its ability to represent the Arabic vowel 32) Horta takes seriously some manuscript pages that are actually a spoof system, which makes features such as the raising of *a and diary that Burton wrote to relieve the tedium of a transatlantic voyage. the syncope of unstressed short vowels visible. The section Horta, 296; Lovell, A Rage to Live, 343-46. on morphology contains, among other topics, the feminine 435 BOEKBESPREKINGEN — ARABICA 436 ending -at, the case endings, the definite article, the diminu- Arabic dialects to see what range of colour ᾿aṣfar could indi- tives, the ᾿af῾al pattern of the elative, and the broken plurals. cate there. We are looking forward to the other three instalments of the Francesco Grande’s (very technical) chapter on terminative-­ “Graeco-Arabica” article series. adverbial and locative-adverbial endings (pp. 271–316) The third section, Classical Arabic in Context, starts with investigates the declensional paradigms of pre-Classical Ara- a chapter by Daniele Mascitelli (pp. 189–211), which dis- bic and Akkadian, the two Semitic languages with fully pro- cusses the Classical Arabic quadriliteral roots starting with ductive case systems, as well as Ugaritic and Biblical šīn or sīn that have triliteral roots as their counterparts. His Hebrew. Grande reconstructs a ‘dualist’ system of case end- hypothesis is that these quadriliteral roots may have been ings in Early Semitic based on internal structural clues. With derived from triliteral ones by adding a causative prefix š- or regard to pre-Classical Arabic, Grande concludes that two s-. The roots are compared with the corresponding roots of relic forms of locative-adverbial endings existed side by Ancient South Arabian and Modern South Arabian, as both side: those that ended in u, as in qablu ‘before’, and those these languages have a productive causative s1- prefix in that ended in un or um as in ladun ‘at’ or ḏalikum ‘that’, their verbal system. The chapter provides a list with 78 roots which are older than their counterpart with u. He also finds beginning with š- and 90 beginning with s-, discussing some evidence for some relics of terminative-adverbial endings. of these in depth, such as sulaḥfā ‘turtle’ which is linked to The last contribution in the section on Classical Arabic the triliteral root LḤF ‘to cover’. is Johnny Cheung’s chapter on Middle Iranian borrowings Lutz Edzard (pp. 212–226) discusses the phenomenon that in the Qur᾿ān (pp. 317–333). He discusses the Qur᾿ānic the relative pronouns allaḏī and illi can be used as subordi- words of probable Iranian origin found in Arthur Jeffery’s nating particles meaning ‘that; for that; because’. An exam- The Foreign Vocabulary of the Qur᾿ān (1938), explaining ple from Classical Arabic is al-ḥamdu li-llāhi llaḏī saqaṭa their etymology, as well as the phonological and morpho- min-nā dirhamun fa-῾awwaḍa-nā llāhu dīnāran ‘Praise be to logical changes that the words have gone through in the God that we lost a dirham and God compensated us with a borrowing process. Some of the words have not been bor- dinar’. Earlier interpretations such as by Diem (2007), stated rowed directly from Iranian, but have come into Arabic via that the relative pronouns in these examples are reinterpreted another language, most often Aramaic. Cheung sometimes as subordinating markers. Edzard proposes an alternative offers alternative etymologies, such as for arā᾿ik ‘couches’, explanation, in which allaḏī / illi as subordinators are an which might have been borrowed from Greek rather than archaic feature, as “the general direction in grammaticaliza- Iranian. It is interesting to note that most borrowings men- tion processes tends to lead from demonstrative elements that tioned are related to the description of paradise, or, as also serve as subordinators towards relative elements” Cheung summarises it: “items & products related to luxury (p. 222). To reinforce his theory, he mentions parallels in and refinement” and “intangible (spiritual, religious) ideas” Indo-European languages, such as the English demonstrative (p. 332). and relative markers that. The section on Qur᾿ānic Arabic in Context contains two The next chapter, by Jordi Ferrer i Serra (pp. 227–270), articles. The first of these, by Guillaume Dye, presents some discusses the translation of the word ᾿aṣfar. Franciscus data evidencing bilingualism or even multilingualism in ­Raphelengius (1539–1597), Professor of Hebrew in Leiden, Qur᾿ānic Arabic (pp. 337–371). He points out that it is not translated this colour term in his Lexicon Arabico-Latinum unlikely that both the audience and author(s) of the Qur᾿ān with ‘red, brown, orange’. He based this on Q 2:69, where were bilingual or multilingual to a certain extent, especially he identified the ‘yellow’ cow mentioned as the red heifer in having a command of Aramaic. Dye presents cases of phrases Num 19:2 and interpreted ᾿aṣfar as a translation of Hebrew in the Qur᾿ān that are literal translations of Bible texts. He -Thomas Erpenius (1584–1624), who wrote an adden- also looks at the use of foreign words, loanwords, and syn .אדם dum to Raphelengius’ dictionary, changed the translation tactic structures and concludes that some stylistic peculiari- into flavus ‘yellow, light brown’, which then established ties of the Qur᾿ān could be explained by taking into account itself as the standard translation of ᾿aṣfar. Ferrer i Serra dis- interference of the style of other, Jewish and Christian, reli- cusses several old sources that indicate that ᾿aṣfar should be gious works. He also stresses the role of scribes, who were seen as a shade closer to orange, red, or reddish brown. He highly literate in both Arabic and Aramaic, in the composi- discusses the appellation Banū al-᾿Aṣfar for the Romans, tion and transmission of the Qur᾿ān. which was related to Esau/Edom’s red skin colour, and must Martin Baasten’s chapter (pp. 372–392) takes as its start- have referred to a darker, rather than a pale, complexion. He ing point Luxenberg’s Die Syro-aramäische Lesart des also refers to a 19th-century source that mentions that in the Koran (2000), and looks specifically at Sūrat al-Kawṯar, Sudan, ᾿aṣfar was used to connote darker skin. I would like which has baffled both medieval and modern exegetes. Baas- to add two references from contemporary Egyptian Arabic ten analyses Luxenberg’s proposed new reading of this verse that support his theory, and both point to a more brownish/ word by word. For instance, the word kawṯar, which is tra- reddish interpretation of the colour ᾿aṣfar. The first one is ditionally translated with ‘abundance’ or, alternatively, from the Western Delta: “ṣafra hellbraun ohne Flecken explained as one of the rivers in Paradise, is identified by (Kuh)”;1) the second “ṣafār(i) iššams the moment before Luxenberg as the Syriac kuttārā ‘persistence’. Baasten finds sunset” 2) (when the sun is red). It would be a useful addition this theory plausible, but does not exclude the possibility that to Ferrer i Serra’s study to have a look at other modern kawṯar could come from another Arabic dialect rather than from Syriac. In other cases, such as the noun šāni᾿ ‘adver- sary’, Baasten shows that there is no reason to assume influ- 1) See Peter Behnstedt and Manfred Woidich, Die ägyptisch-arabischen­ Dialekte: Band 4 Glossar Arabisch-Deutsch, Wiesbaden 1994: 264b. ence from Syriac. Baasten therefore rejects Luxenberg’s con- 2) See Martin Hinds and El-Said Badawi, A Dictionary of Egyptian clusion that Sūrat al-Kawṯar was the first piece of evidence Arabic, Beirut 1986: 505b. for Christian epistolary literature in the Koran. 437 BIBLIOTHECA ORIENTALIS LXXV N° 3-4, mei-augustus 2018 438

The last section, Middle and Modern Arabic in Context, today, which must have come from a stratum other than the contains four chapters. The first of these, by Geoffrey Khan current North-African Arabic dialects. (pp. 395–404), deals with vocalised Judaeo-Arabic texts Generally, the volume has been carefully edited, except (Arabic written by Jews in Hebrew script), which are an for some typos and minor errors (e.g. p. 462, ex. 42: ‘water’ important source for the history of the Arabic language. He should be ‘hundred’; p. 377 naḥara should be naǧara). compares the Judaeo-Arabic reading tradition with two non- However, Kropp’s chapter would have benefited from more Jewish texts: a Greek transcription of the Arabic translation careful proofreading (see, e.g., p. 54 “they developed till to of Psalm 77, dated to the 9th–10th century, and a Coptic a stage”; p. 63, fn. 15 “the verb […] fits well into the con- transcription dated to the 13th century. The Judaeo-Arabic text and fives a required meaning”). On p. 79, Laïla Nehmé manuscripts discussed are from the Cairo Genizah and date refers to the inscription of ῾Ēn ῾Avdat (῾Ayn ῾Abada), stating from the High Middle Ages. The texts are written in Classi- that the most recent publication about it is Macdonald cal Judaeo-Arabic orthography with Tiberian Hebrew vowel (2015); it would have been useful if she had referred to the signs, which provide important information on the phonol- chapter about the inscription by Kropp that precedes hers in ogy of the language. Some features of modern Arabic dia- this book. Apart from these minor points, this is an excellent lects found in these texts are pronominal suffixes without collection of articles, which contains a wealth of information case inflection, the raising of the vowel a, and the prefix and shows a high level of scholarship. The book does exactly yi- with kasra in the verbs. The texts also exhibit pseudo- what it promises on the cover: to look at Arabic from an Classical features, such as treating initial hamzatu l-waṣl as interdisciplinary point of view, combining methodology from hamzatu l-qaṭ῾. Most of these characteristics are found not the fields of historical linguistics, epigraphy, dialectology, only in the Judaeo-Arabic texts, but also in those in Greek and history. This book is a must-read for everyone interested and Coptic script. in the history of the Arabic language. Alexander Magidow’s chapter (pp. 405–440) addresses the problem of applying a traditional tree-based model for a University of Amsterdam, Liesbeth Zack genealogical reconstruction of the Arabic dialects, as it is 17 May 2018 difficult to address the merger of dialects and language con- tact in such a model. As an alternative strategy, he proposes reconstructing information about speech communities, which he defines as “a group of people bound by social network ties as well as a sense of social allegiance, and having at least some part of their linguistic repertoires in common” (p. 416). He combines linguistic data from modern Arabic dialects and historical data about the Arab migrations to reconstruct the Arabicisation process. As a case study he takes the demon- stratives and uses these, in combination with information about historical events, to reconstruct the Arabicisation of the Nile Valley and of the Levant. Na῾ama Pat-El (pp. 441–475) takes the features discussed by Huehnergard in the first chapter to demonstrate that some archaic features of the modern Arabic dialects are not found in Classical Arabic, and therefore preserve a linguistic state earlier than Old Arabic. One of the features she discusses is the definite article, giving examples of dialects in which only the adjective, and not the noun, takes the article, such as sū᾿ ǝl-῾atī᾿ ‘the old market’. There are also examples from pre- Islamic Arabic, the Qur᾿ān, and other Central Semitic lan- guages, which makes her conclude that it is an old feature. She points out that the “existence of a certain feature in Clas- sical Arabic cannot by itself serve as a proof that it is archaic” (p. 470) and that Classical Arabic is not necessarily conservative, while the dialects may have preserved archaic features that cannot be found in Classical Arabic. The last article in the collection, by Marijn van Putten and Adam Benkato, looks at traces of Arabic in the Berber lan- guage of the Libyan oasis of Awjila (pp. 476–502). One of the phonological features investigated is the reflex q of qāf in some Arabic loanwords, a reflex not encountered in the Arabic dialects of Libya today, where it is realised as g. Another feature is that of loanwords in which the ῾ayn is missing, such as in lāfīt ‘health’ (from al-῾āfiya), which points to an early borrowing from Arabic. Other features dis- cussed are the morphology of loan verbs and the structure of Arabic nouns. There are indications that the language of Awjila was influenced by Arabic varieties not found in Libya