93397_JOSS_2010/2_Cv
24-06-2010
15:19
Page 1
ISSN 0022-4480 (PRINT) ISSN 1477-8556 (ONLINE)
365 385
391 407 423
451 479 497 509 539 589 621
Semitic Studies
SHORT NOTICES
347
JOURNAL OF
REVIEWS
325
AUTUMN 2010
ALEXANDER ANDRASON ‘The Akkadian Iprus from the Unidirectional Perspective’ CYNTHIA L. MILLER ‘Vocative Syntax in Biblical Hebrew Prose and Poetry: A Preliminary Analysis’ WILLIAM MORROW ‘“To Set the Name” in the Deuteronomic Centralization Formula: A Case of Cultural Hybridity’ AARON MICHAEL BUTTS and HUMPHREY HILL HARDY II ‘A Revised Reading of a Nabataean Inscription from Umm al-Jimal’ TZVI NOVICK ‘Naming Normativity: The Early History of the Terms Sûrat ha-dîn and Lifnîm mis-sûrat ha-dîn’ FRANÇOISE BRIQUEL CHATONNET and ALAIN DESREUMAUX ‘A Study and Characterization of the Syro-Malabar Script’ PATRICIA CRONE and ADAM SIVERSTEIN ‘The Ancient Near East and Islam: The Case of Lot Casting’ MIRIAM GOLDSTEIN ‘“Arabic Composition 101” and the Early Development of JudaeoArabic Bible Exegesis’ OLIVER KAHL ‘Two Antidotes from the “Empiricals” of Ibn at-Tilmi∂’ GEORGE A. BEVAN and M. BARBARA REEVES ‘A New Funerary Inscription from Humayma’ ABDULRAZZAK PATEL, ‘Language Reform and Controversy in the Nah∂a: Al-Shartuni’s Position as a Grammarian in Sahm SADOK MASLIYAH ‘The Folk Songs of Iraqi Children: Part Two’
JOURNAL OF
Semitic Studies VOLUME LV. NO. 2 AUTUMN 2010
VOLUME LV. NO. 2
93397_JOSS_2010/2_Cv
24-06-2010
15:19
Page 2
SUBSCRIPTIONS A subscription to Journal of Semitic Studies comprises 2 issues. Prices include postage by surface mail, or for subscribers in the USA and Canada by airfreight, or in India, Japan, Australia and New Zealand, by Air Speeded Post. Airmail rates are available on request. Annual Subscription Rate (Volume LIII, 2 issues, 2008) Institutional Print edition and site-wide online access: £166//249/US$315 Print edition only: £152//228/US$289 Site-wide online access only: £138//207/US$262 Personal Print edition and individual online access: £54//81/US$103 Please note: £ Sterling rates apply in UK, / rates apply in Europe, US$ elsewhere There may be other subscription rates available, for a complete listing please visit www.jss.oupjournals.org/ subinfo. Full prepayment, in the correct currency, is required for all orders. Orders are regarded as firm and payments are not refundable. Subscriptions are accepted and entered on a complete volume basis. Claims cannot be considered more than FOUR months after publication or date of order, whichever is later. All subscriptions in Canada are subject to GST. Subscriptions in the EU may be subject to European VAT. If registered, please supply details to avoid unnecessary charges. For subscriptions that include online versions, a proportion of the subscription price may be subject to UK VAT. Personal rate subscriptions are only available if payment is made by personal cheque or credit card and delivery is to a private address. The current year and two previous years’ issues are available from Oxford University Press. Previous volumes can be obtained from the Periodicals Service Company, 11 Main Street, Germantown, NY 12526, USA. Email:
[email protected]. Tel: +1 (518) 537 4700. Fax: +1 (518) 537 5899. For further information, please contact: Journals Customer Service Department, Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, UK. Email:
[email protected]. Tel (and answerphone outside normal working hours): +44 (0)1865 353907. Fax: + 44 (0)1865 353485. In the US, please contact: Journals Customer Service Department, Oxford University Press, 2001 Evans Road, Cary, NC 27513, USA. Email:
[email protected]. Tel (and answerphone outside normal working hours): 800 852 7323 (toll-free in USA/Canada). Fax: 919 677 1714. In Japan, please contact: Journals Customer Services, Oxford University Press, 1-1-17-5F, Mukogaoka, Bunkyo-ku, Tokyo, 113-0023, Japan. Email:
[email protected]. Tel: (03) 3813 1461. Fax: (03) 3818 1522. Methods of payment. (i) Cheque (payable to Oxford University Press, to Oxford University Press, Cashiers Office, Great Clarendon Street, Oxford OX2 6DP, UK) in GB£ Sterling (drawn on a UK bank), US$ Dollars (drawn on a US bank), or EU/ Euros. (ii) Bank transfer to Barclays Bank Plc, Oxford Group Office, Oxford (bank sort code 20-65-18) (UK), overseas only Swift code BARC GB 22 (GB£ Sterling to account no. 70299332, IBAN GB89BARC20651870299332; US$ Dollars to account no. 66014600, IBAN GB27BARC 20651866014600; EU/ Euros to account no. 78923655, IBAN GB16BARC20651878923655). (iii) Credit card (Mastercard, Visa, Switch or American Express). Journal of Semitic Studies is published two times a year in April and October by Oxford University Press, Oxford, UK. Annual subscription price is £166//249/US$315. Journal of Semitic Studies is distributed by Mercury International, 365 Blair Road, Avenel, NJ 07001, USA. Periodicals postage paid at Rahway, NJ and at additional entry points. US Postmaster: send address changes to Journal of Semitic Studies (ISSN 0022-4480), c/o Mercury International, 365 Blair Road, Avenel, NJ 07001, USA. PERMISSIONS For information on how to request permissions to reproduce articles/information from this journal, please visit www.oxfordjournals.org/permissions. ADVERTISING Inquiries about advertising should be sent to Helen Pearson, Oxford Journals Advertising, PO Box 347, Abingdon OX14 1GJ, UK. Email:
[email protected]. Tel: +44 (0)1235 201904; Fax: +44 (0)8704 296864. DISCLAIMER Statements of fact and opinion in the articles in Journal of Semitic Studies are those of the respective authors and contributors and not of Journal of Semitic Studies or Oxford University Press. Neither Oxford University Press nor Journal of Semitic Studies make any representation, express or implied, in respect of the accuracy of the material in this journal and cannot accept any legal responsibility or liability for any errors or omissions that may be made. The reader should make his/her own evaluation as to the appropriateness or otherwise of any experimental technique described. © The University of Manchester 2008 All rights reserved; no part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise without prior written permission of the Publishers, or a licence permitting restricted copying issued in the UK by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1P 9HE, or in the USA by the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923. Typeset by Drukkerij Peeters, Herent, Belgium Printed by Bell & Bain Ltd, Glasgow, UK
INFORMATION FOR AUTHORS Contributions, books for review, and other editorial communications should be addressed to: The Editors Journal of Semitic Studies Middle Eastern Studies School of Languages, Linguistics and Cultures University of Manchester Manchester M13 9PL UK Article Length Editorial policy is normally to favour shorter articles, i.e. no longer than 10,000 words. Typescript layout Articles and reviews should be typed on one side of the paper (A4 or 8.5 x 11 inches) with double spacing and generous margins on one side of the paper only. Authors of articles should submit TWO COPIES, and disk, if available. Non-Roman script is to be written as clearly as possible and used as sparingly as possible. Notes should be numbered consecutively, double spaced and placed at the end. The Editors will not accept for publication any typescripts which do not conform to these standards. For detailed instructions please see the style sheet. Abstracts Authors of articles are required to provide an abstract, which should be a single paragraph, not exceeding 150 words. Submission We will also normally require the final revised version to be supplied in hard copy and electronically. Whilst we can accept most wordprocessor formats the preferred option is Word. The files should be submitted in the wordprocessor's native format and as an RTF (Rich Text Format). Illustrations Electronic submission of illustrations. We can accept 3.5-inch floppy disks, CDROMs and email attachments. High-quality print-outs of all figures are still essential, as they will be scanned for reproduction in the event of problems with the electronic files. Figures must be saved as high-resolution TIF files (at least 300 pixels per inch for photographs, and 1200 pixels per inch for black and white line drawings). Offprints and Proofs Authors of articles will receive one copy of the issue containing their article and twenty-five offprints of their article free of charge. Alteration on the proofs, other than the correction of printer's errors, may be disallowed. Copyright and Permissions Manuscripts submitted will be expected to contain original work and should not have been published in abridged or other form elsewhere. It is a condition of publication in the Journal that authors grant an exclusive licence to The University of Manchester. Any requests from third parties to reproduce articles are handled on behalf of the journal by Oxford University Press. This will also allow the article to be as widely disseminated as possible and will protect the rights of the author and OUP. In granting an exclusive licence, authors may use their own material in other publications provided that the journal is acknowledged as the original authority for publication, and Oxford University Press is notified in writing and in advance.
JOURNAL OF
Semitic Studies V OL UME LV. NO. 2 AUT UM N 2010
EDITORIAL COMMITTEE P. S. ALEXANDER G. J. BROOKE R. BUCKLEY A. CHRISTMANN J. F. HEALEY P. C. SADGROVE
JOURNAL OF
Semitic Studies Editorial Committee P.S. Alexander G.J. Brooke R. Buckley A. Christmann J.F. Healey P.C. Sadgrove
Editorial Advisers C.E. Bosworth S.P. Brock K.J. Cathcart R. Gleave M. Kropp Y. Matras A.R. Millard M. Morgenstern M.E.J. Richardson P. Schäfer A. Shivtiel G. Rex Smith Y. Suleiman E. Ullendorff
The Journal of Semitic Studies publishes a book supplement series. For more information see www.jss.oxfordjournals.org. The editors welcome proposals for further titles in the series. For more details please write to the editors, Journal of Semitic Studies, Middle Eastern Studies, School of Languages, Linguistics and Cultures, University of Manchester, Manchester M13 9PL, UK
JOURNAL OF
Semitic Studies V OL UME LV. NO. 2 AUTUMN 2010
Contents articles ALEXANDER ANDRASON, ‘The Akkadian Iprus from the Unidirectional Perspective’ CYNTHIA L. MILLER, ‘Vocative Syntax in Biblical Hebrew Prose and Poetry: A Preliminary Analysis’ WILLIAM MORROW, ‘“To Set the Name” in the Deuteronomic Centralization Formula: A Case of Cultural Hybridity’ AARON MICHAEL BUTTS and HUMPHREY HILL HARDY II, ‘A Revised Reading of a Nabataean Inscription from Umm al-Jimal’ TZVI NOVICK, ‘Naming Normativity: The Early History of the Terms Sûrat ha-dîn and Lifnîm mis-sûrat ha-dîn’ FRANÇOISE BRIQUEL CHATONNET and ALAIN DESREUMAUX, ‘A Study and Characterization of the Syro-Malabar Script’ PATRICIA CRONE and ADAM SIVERSTEIN, ‘The Ancient Near East and Islam: The Case of Lot Casting’ MIRIAM GOLDSTEIN, ‘“Arabic Composition 101” and the Early Development of Judaeo-Arabic Bible Exegesis’ OLIVER KAHL, ‘Two Antidotes from the “Empiricals” of Ibn at-Tilmi∂’ GEORGE A. BEVAN and M. BARBARA REEVES, ‘A New Funerary Inscription from Humayma’ ABDULRAZZAK PATEL, ‘Language Reform and Controversy in the Nah∂a: Al-Shartuni’s Position as a Grammarian in Sahm SADOK MASLIYAH, ‘The Folk Songs of Iraqi Children: Part Two’
325 347 365 385 391 407 423 451 479 497 509 539
reviews Billie Jean COLLINS, The Hittites and their World (Charles BURNEY) Wilfred G. LAMBERT, Babylonian Oracle Questions (Alasdair LIVINGSTONE)
589 591
Israel FINKELSTEIN and Amihai MAZAR, edited by Brian B. SCHMIDT, The Quest for the Historical Israel: Debating Archaeology and the History of Early Israel (Philip R. DAVIES) Lester GRABBE (ed.), Israel in Transition: From Late Bronze II to Iron IIa (c. 1250-850 B.C.E.) Volume 1. The Archaeology (Matthew WHINCOP) Dörte BESTER, Körperbilder in den Psalmen (Sue GILLINGHAM) Stephanie DALLEY, Esther’s Revenge at Susa: From Sennacherib to Ahasuerus (Roger TOMES) Ehud NETZER, The Architecture of Herod, the Great Builder (Kay PRAG) Karel JONGELING, Handbook of Neo-Punic Inscriptions (Edward LIPINSKI) Andrzej S. TURKANIK, Of Kings and Reigns: A Study of Translation Technique in the Gamma/Gamma Section of 3 Reigns (1 Kings) (J.M. DINES) Edward M. COOK, A Glossary of Targum Onkelos: According to Alexander Sperber’s Edition (Aaron D. RUBIN) Robert PHENIX, The Sermons on Joseph of Balai of Qenneshrin: Rhetoric and Interpretation in Fifth Century Syriac Literature (John W. WATT) Robert GLEAVE, Scripturalist Islam: The History and Doctrines of the Akhbari Shi’i School (Andrew J. NEWMAN) Geoffrey KHAN, Arabic Documents from Early Islamic Khurasan (C. Edmund BOSWORTH)
593 595 597 599 601 603 609 612 614 616 618
short notes Simo PARPOLA, Robert WHITING et al. (eds), Assyrian-English-Assyrian Dictionary (Jonathan TAYLOR) Hermann GUNKEL (edited and translated by K.C. Hanson), Israel and Babylon: The Babylonian Influence on Israelite Religion (Roger TOMES) G. BOCCACCINI (ed.), Enoch and the Messiah Son of Man: Revisiting the Book of Parables (Maurice CASEY) Brian A. CATLOS, The Victors and the Vanquished: Christians and Muslims of Catalonia and Aragon, 1050 1300 (Anthony John LAPPIN)
621 621 622 623
Books Received
625
Index
629
Journal of Semitic Studies LV/2 Autumn 2010 doi: 10.1093/jss/fgq001 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
THE AKKADIAN IPRUS FROM THE UNIDIRECTIONAL PERSPECTIVE ALEXANDER ANDRASON UNIVERSITY OF STELLENBOSCH
Abstract This paper is aimed at describing the Akkadian construction iprus from the perspective of the unidirectional theory of grammaticalization. First, general notions of grammaticalization, unidirectionality and source determination are given. Next, in the first part of the article, the author presents the development of the resultative construction as posited by Dahl (1985, 2000) and Bybee, Perkins and Pagliuca (1994). These famous definitions will be modified in view of observations of Maslov (1988) and Andrason (2009a and 2009b). In the second part, the author proceeds to the analysis of the iprus gram. Firstly, the core and peripheral uses of the iprus are presented. Secondly, employing the unidirectional model of the development of resultative constructions established in the first part, the author explains all meanings and functions of the iprus as realizations of different stages of the regular resultative evolution. In consequence, from the unidirectional perspective, the synchronic set of chaotic and unrelated values displayed by the iprus turns into a homogeneous and regular picture — a single resultative diachrony, i.e., one original resultative input that follows a determined diachronic path. Finally, the author shows how this proposal fits into the proto-Semitic model and how other Semitic languages may support the unidirectional resultative interpretation of the iprus.
There exists a wide range of theories explaining the verbal system of Akkadian. Some of them emphasize its aspectual nature, while others stress the temporal or the perfectal component. What unifies all these theories is the fact that they describe elements of the Akkadian verbal organization from the synchronic perspective, having based the model a priori on the alleged dichotomy between two constructions: the iprus and the iparras — and, depending on the scholar’s view, this opposition consists of aspect, tense or relative tenses. In consequence, due to the radical synchronic perspective, both the iprus and the iparras are usually described as static items that form a static contrast. 325
THE AKKADIAN IPRUS
However the author of this paper, who admits the supremacy of the diachronic method, intends to present the iprus as a dynamic gram, and consequently to explain its nature and values from the unidirectional perspective of grammaticalization and universal paths (cf. Dahl 1985 and 2000, and Bybee, Perkins and Pagliuca 1994). Following the laws of grammaticalization, any verbal formation must be understood as a developing gram that moves from the periphery of the system into the core area, acquiring and losing determined meanings. This means that different uses (both typical/core and atypical/peripheral) of a construction are the results of its gradual, unidirectional and ordered development. Thus, if the theory is correct, the dynamic view of the iprus should enable us to group the synchronically chaotic and unrelated values displayed by the Akkadian gram into a homogenous and regular set — one original input that follows a single diachronic path. Moreover, it should be possible to explain the complex opposition between the iprus and the iparras in terms of two distinct functional trajectories without, at the same time, reducing it to a bipolar contrast.1
1 The author is aware of the unusual application here of the diachronic and synchronic method. Normally these terms refer to earlier and later forms (diachronic) or contemporary forms (synchronie). However, the author uses the universal diachrony (paths) of a given construction in order to explain contemporary data. Put differently, the diachronic approach provides a schematic explanation for forms that are all viewed as contemporaneous. In this way, the diachronic model enables us to relate all superficially incompatible values of a construction (in this case of the iprus) and define it as a homogeneous form, i.e., as a synchronic expression of a prototypical diachrony (the resultative path). This means that every grammatical formation at a given moment is a synchronic manifestation of a diachronic development that is consistent with predetermined universal paths. Certainly, such a view does not explain conditions that determine different uses of the construction in question. However, it does explain why the construction displays various and almost opposite meanings (for instance the past and the future in the case of the iprus), and furthermore it reduces all synchronic values to a single input and its subsequent regular diachrony. This method of explanation coincides with methodology used in Indo-European studies. For instance, in Germanic languages the same verbal morphology serves to provide both the simple past (strong preterite) and present meaning (preterite-present verbs or modal verbs). In order to explain this paradox, one must resort to the concept of the resultative (perfect) that was the PIE source of both groups of verbs (cf. Nedjalkov and Jaxontov 1988). In a similar vein the value of the verb o÷da is explained — in Classical Greek this verb had a clear present meaning while the same morphological pattern in the immense majority of verbs showed prototypical perfect uses. Yet again, both types originated in the PIE resultative and followed either of the two drifts that resultative constructions universally follow (on resultative paths and drifts see part 1).
326
THE AKKADIAN IPRUS
In the next part of the article, general notions of grammaticalization, unidirectionality and source determination will be given. Then, we will focus on the development of the resultative construction as posited by Dahl (1985, 2000) and Bybee, Perkins and Pagliuca (1989). These famous definitions will, however, be modified in view of observations provided by Maslov (1988) and Andrason (2009a and 2009b). After that, the author will proceed to the analysis of the iprus. Firstly, the core and peripheral uses of the Akkadian formation will be presented. Secondly, applying the unidirectional model of grammaticalization, we will explain all meanings and functions of the gram as manifestations of different stages of a regular functional evolution, i.e. as a single diachrony. Finally, it will be demonstrated how this proposal fits into the proto-Semitic model and how other Semitic languages may support the unidirectional interpretation of the iprus. 0. Introduction – Grammaticalization Grammaticalization is a change that attributes a grammatical character to a formerly independent word or a semantically transparent periphrasis (cf. Meillet 1948: 132), or put differently, ‘grammaticalisation consists in the increase of the range of the morpheme advancing from a lexical to a grammatical or from a less grammatical to a more grammatical status’ (Kury¥owicz 1965: 52). This means that grammaticalization is a movement whereby one unit acquires a grammatical function and loses its phonetic substance, semantic independence, transparency and complexity, syntactic freedom, and pragmatic significance (cf. Heine and Reh 1984: 15). In consequence, a semantic item or a periphrasis becomes an element of the grammatical (morphological, analytic or synthetic) system of the language travelling from the peripheral area to the more central sphere (cf. Bybee, Perkins and Pagliuca 1994 and Dahl 2000). According to Dahl (2000), there are two types of grams, i.e. peripheral grams and core grams. This distinction reflects differences in grammaticalization. Non-grammaticalized peripheral grams are usually lexical and analytical. Core grams, on the other hand, are fully grammaticalized and may be inflectional (cf. Dahl 2000: 15–17). Moreover, a distinction may be made between coming grams, i.e. those which are invading the core area of the system, and leaving grams, i.e. those which have already achieved their functional apogee and are consequently moving out of the centre zone back to the periphery. The leaving grams, even though peripheral, are usually synthetic. It must be also emphasized that grammaticalization 327
THE AKKADIAN IPRUS
is a complex process in which various phenomena participate and converge, e.g. generalization, decategorialization, specialization, increased frequency, morphologization and phonological reduction (cf. Hopper and Traugott 2003: 99–174). Grammaticalization theory claims that the functional paths are universal, unidirectional and input-output determining (cf. Bybee 1994: 9–15). As far as the the verbal system is concerned, Dahl (1985, 2000) and Bybee, Perkins and Pagliuca (1994) establish three possible evolutionary processes: from locative constructions into imperfective (and present), from modal expressions into modality and future, and finally from resultative-completive periphrasis into perfective past and simple past. In this article, we will deal especially with the third trajectory, which may be applied to the Akkadian preterite iprus. In the next paragraph, we will present the resultative path as posited by Dahl (1985, 2000) and Bybee, Perkins and Pagliuca (1994). However, these definitions will be slightly modified in light of the evidence concerning the development of the resultative constructions provided by Maslov (1988) and Andrason (2009a and 2009b). The precise description of the resultative path will enable us to reconstruct the history of the Akkadian iprus and explain its uses in Old Babylonian from the unidirectional perspective. 1. Resultative and its development Dahl (2000: 15–17) determines the resultative path as a diachronic process which begins with lexical periphrases that include items with the meaning of ‘already’ or ‘finish’. The lexical inputs that correspond to the meaning of ‘already’ generate the resultative proper, while those that have their source in the ‘finish’ lexemes lead to the creation of the perfect. The resultative consequently develops into the perfect, and then after a possible convergence with the ‘finish’ perfect, it turns into a perfective or past. This development may be illustrated by the following picture (cf. Figure 1): INPUT
PATH
OUTPUT
finish
perfective perfect
already
resultative
past
Figure 1. Resultative path according to Dahl (1985, 2000) 328
THE AKKADIAN IPRUS
According to Dahl (1985) and Bybee and Dahl (1989) the perfective equals in fact the perfective past since the typical system that includes the timeframe perfective aspect is tripartite and the perfective is restricted to the past. In consequence, the transformation of the perfect into the perfective involves the further change into the perfective past. Finally, the aspectual distinction may be lost and the gram acquires pure temporal past significance. A highly similar trajectory (called here anterior path) has been posited by Bybee, Perkins and Pagliuca (1994). According to their model, the evolution originates in three possible types of expressions: first, in periphrases with the verbs ‘have’ and ‘be’, together with participles (as commonly used in Romance and Germanic languages). These constructions give the resultative proper, and consequently the anterior. They may also lead to evidential uses and functions. Secondly, the anterior path may have its source in verbs that convey the idea of ‘coming’. This type will directly generate the anterior. Thirdly, the anterior may be a descendent of completive locutions that, on its own, can derive from ‘finish’ constructions and directional expressions. The completive, as we have said, turns regularly into the anterior. However, some completives may become what Bybee, Perkins and Pagliuca call ‘derivational perfective’ (as for example in Slavic). All resultatives, completives and direct anteriors (i.e. those that arise from ‘come’ constructions) evolve into perfects (anterior proper), and consequently turn into the perfective, finally becoming the simple past. Bybee, Perkins and Pagliuca follow Dahl’s opinion on the perfective aspect and its temporal implication. Thus, the perfective will usually equal the perfective past. They observe that the perfective past is more specific and it ‘interacts with the imperfectivity by contrasting with it’, while the simple past is more general and may be used to describe not only perfective events but also situations viewed as imperfective (Bybee, Perkins and Pagliuca 1994: 84). In consequence, in languages where there is an imperfective past, the new anterior will develop, by contrast, into the perfective past, while in languages where the imperfect does not exist, the anterior will directly generate the simple past. Bybee, Perkins and Pagliuca (1994: 85) also note that derivational completives, differently from other formations which correspond to preanterior stages situated on the anterior path, do not evolve into the perfective past, but rather yield the derivational perfective. The derivational perfective is not restricted to past events and situations; quite the contrary, it may occur in any temporal sphere of a finite verb, cf. the present perfective that indicates future (Polish) or 329
THE AKKADIAN IPRUS
in subordinate clauses expresses the idea of purpose (Serbo-Croat). It can also appear in the imperative or in non-finite verbal forms (cf. Classical Greek or again Slavic languages). In fact, such a derivational perfective aspect is not restricted to verbs, but it may be present in nouns, cf. verbal nouns in Polish which can be aspectually marked expressing imperfective or perfective actions. The entire anterior trajectory, and its sub-paths may be summarized in the following figure (cf. Figure 2): INFERENCE FROM RESULTS
‘be/have’
INDIRECT EVIDENCE
RESULTATIVE
‘come’
ANTERIOR
PERFECTIVE/SIMPLE PAST
COMPLETIVE
‘finish’
directionals
DERIVATIONAL PERFECTIVE
Figure 2. The anterior path according to Bybee, Perkins and Pagliuca (1994: 105)
Bybee, Perkins and Pagliuca (1994) maintain that the completive and the resultative, i.e. the constructions which reflect pre-anterior stages of the development, are highly similar. Both present an event as composed of two temporal and consecutive planes: the previous cause and the posterior result. However, while the resultative emphasizes the resulting present state of a prior action, the completive stresses the previous action. As shown in the figure (cf. Figure 2), the resultative can also acquire an evidential character. This change is itself a gradual development that passes through the following stages: a) information via direct evidence, b) via indirect inferred evidence; and c) via reported (second-hand) evidence; cf. Bybee, Perkins and Pagliuca (1994: 95). As we have said, the resultative is a semantically complex and bipolar formation. According to Maslov (1988: 64) ‘[t]he term “[resultative]2” include[s] two temporal planes: that of precedence, and that of sequence. The situations corresponding to these planes are in one way or another related, as cause and effect’. In consequence, the internal structure of a resultative construction can be represented in the following way (cf. Figure 3):
2
Maslov uses the term ‘perfect’. 330
THE AKKADIAN IPRUS
gZhjaiVi^kZ D& D' D&
D'
eg^dg XVjhZ
edhiZg^dg gZhjai
Figure 3. The structure of a resultative construction
Maslov (1988: 64) adds that ‘[u]sually, one of the two situations seems to be semantically more important, while the other serves, as it were, as a background, often barely suggested’. This inherent semantic instability leads to a focus either on the precedence or on the consequence. Gradually, one of the two planes becomes dominant and the other disappears. Here the terms prior and posterior are used to describe the internal correlation between two planes of the resultative. In relation to the time of the context (the main time of sentence, text, dialogue, discourse etc.) the labels anterior and simultaneous will be used. This means that the emphasis on the prior plane (cause) of the resultative construction will surface as anterior drift to the main time of the context, while the emphasis on the posterior plane (result) will surface as simultaneous drift to the main time. Though the resultative is originally shaped and grammaticalized in the present tense context (in the spoken language, and in discursive genres that are closely related to it), it is not a tense expression. In fact, the resultative does not carry any exact time reference by itself. The tense is frequently provided by the auxiliary verb or by the context if there is no verb in the initial periphrasis. The development following the anterior drift (resultative → perfect → past) corresponds to Dahl’s resultative path (1985, 2000) and to the anterior path as posited by Bybee, Perkins and Pagliuca (1994). On the other hand, scholars have paid less attention to the simultaneous path (resultative → stative → present). However, this development is extremely common and may be observed for example in Latin (cf. the verbs like memini and odi), Greek (cf. the verb o÷da) and in Germanic languages (cf. the preterite-present verbs like kunna in Old Icelandic and Gothic). Moreover, as demonstrated by the history of the Romance and Germanic families, the two drifts may be active in a single language (cf. Andrason 2009a and 2009b). In consequence, the complete model of the development of a resultative construction based on Dahl (1985, 331
THE AKKADIAN IPRUS
2000), Bybee, Perkins and Pagliuca (1994) and Andrason (2009a and 2009b) may be summarized in the following way (cf. Figure 4):3 EVIDENTIAL ANTERIOR
PERFECT
PERFECTIVE
PAST
RESULTATIVE STATIVE
PRESENT
Figure 4. The resultative path
The trajectory described above applies to the present temporal reference. Since grammatical items usually start their grammaticalization in the present time frame, the development in this temporal context is more advanced and better marked (cf. the difference between the passé composé — originally the present perfect vs. the plus-queparfait — the past resultative in French). Furthermore, the anterior drift is regulary stronger than the simultaneous drift, which is frequently limited to stative verbs (this may have its roots in the fientive nature of verbs which describe an action rather than a state). However, the resultative may also occur in the past or future time context. The evolution, though governed by the same principles, i.e. by the anterior and simultaneous drifts, is not identical with the resultative path in the present time frame. The outcomes of the simultaneous drift in the past are first the past stative and later the simple past. Similarly with the future temporal reference, the same drift produces the future stative and subsequently the simple future (cf. the memineram and meminero in Latin). The anterior drift in the past gives the past perfect (the pluperfect) and then, usually, the remote past tense. In the future temporal environment, it leads to the creation of the future perfect (future resultative) that may finally generate the simple future (cf. Polish by¥em (z)robi¥ y bπdπ robi¥), cf. Figure 5:
3
The objections that have been targeted at the grammaticalization and path theory have led to its use for heuristic purposes rather than as the prediction of linguistic developments. However, the author claims that even if the future shape of a language is not entirely predictable, the grammaticalization laws are deterministic in nature. In this respect, languages are examples of chaotic systems — they are governed by deterministic laws but their shape remains to a certain degree unpredictable. 332
THE AKKADIAN IPRUS PAST STATIVE
PAST
PAST PERFECT
REMOTE
FUTURE STATIVE
FUTURE
FUTURE PERFECT
FUTURE
PAST CONTEXT
RESULTATIVE
FUTURE CONTEXT
Figure 5. The anterior and simultaneous drifts in the past and future time context
It is important to stress that the growing resultative gram when acquiring new semantic values does not have to lose its previous uses. It may preserve them as typical and primary functions (cf. passé composé in French which still conveys resultative meaning even though it may be used as a perfective or narrative past) or as secondary and peripheral uses (cf. the preterite-present verbs in Germanic languages — this original stative present function of the resultative is preserved only with approximately ten verbs while the perfect of all other strong predicates acquired the past meaning). Having explained the universal evolution of resultative constuctions, in the following section, we shall analyse the typical and atypical uses of iprus, This inventory of functions will subsequently enable us to define the Akkadian formation from the grammaticalization perspective as a homogeneous phenomenon, a diachronic path. 2. Akkadian iprus The core of the Akkadian verbal system is formed by the opposition between two verbal forms, i.e. the iprus and the iparras (called durative or present, cf. Knudsen 1986, Kienast 2001, Malbran-Labat and Vita 2005) — this opposition is usually defined in terms of aspectuality (perfective aspect vs. imperfective aspect, cf. Malbran-Labat and Vita 2005: 97–8, 102) or relative tenses (anteriority vs. non-anteriority, cf. Streck 1999). According to Landsberger (1926) and Huehnergard (2005: 19) the iprus indicates a punctual event ‘seen as occurring or having occurred at a single point in time’. It frequently corresponds to the simple past or, in temporal clauses, to the pluperfect. In Akkadian grammar the iprus formation is usually labelled ‘preterite’ and consequently defined as the tense of narration of past events (cf. the following example in Huehnergard 2005: 119): 333
THE AKKADIAN IPRUS
(1) Erib-Sîn u Nur-Samas tappûtam ipusu-ma ana bit Samas irubu-ma †emsunu ipusu-ma kaspam babtam amtam u wardam sa Ìarranim u libbi alim mitÌaris izuzu ‘Erib-Sin and Nur-Shamash entered into a partnership; they entered the Shamash temple and carried out their intention: they divided equally the silver, outstanding goods, (and) female and male slaves of (both) business trip(s) and within the city’ (CT 2 28 = Schorr, VAB 5 no. 172).
According to Malbran-Labat and Vita (2005: 102) the iprus ‘se opone al inacabado [i.e. the iparras] sobre el plano del aspecto [y] expresa una acción considerada como acabada’. However, they affirm that both the iprus and the iparras may have not only aspectual but also temporal values (cf. Malbran and Vita 2005: 97). Loesov (2006: 111) rejects Landsberger’s view, and claims that the primary function of the iprus is the expression of a general past and not of punctual or momentary events (cf. the example 2). Even though Loesov criticizes the theory of the aspectual opposition between the iprus and the iparras, he admits that the aspectual interpretation of the iprus is predominantly perfective4 (cf. the following example in Loesov 2006:111). (2) eqlam mari PN assur-ma MU 15.KAM akul ‘I cut off of the field belonging to the sons of PN and lived from it for fifteen years’ (AbB 4, 69:7 ff.).
Streck (2003: 425) is another modern scholar who does not agree with the traditional definition of the iprus construction as a punctual past. His hypothesis of relative tenses seems to work better than the traditional aspectual view; for instance, this model can more accurately explain the use of the iparras with the conjunction lama when the time of the main clause is the past. There also exists a deictic tense approach that relates the iprus and the iparras in independent clauses with clearly temporal interpretation, respectively as past (cf. the interpretation of the iprus in Lipinski 2001) and non-past. This hypothesis seems to be weak when it comes to explaining situations which 4 The aspectual theory may be also supported by the fact that the iprus of stative verbs has the inchoative meaning. This corresponds to the situation in Slavic languages where the perfective aspect of stative verbs has not the terminative or telic value (as in the case of regular verbs), but rather indicates the ingressive meaning, cf. the imperfective bac siπ ‘be scared’ vs. the perfective przestraszyc siπ ‘become scared, get scared’ or miec ‘have’ vs. dostac ‘get’.
334
THE AKKADIAN IPRUS
occur simultaneously with the speech time and especially various uses in temporal clauses. Kienast (2001: 296–7) does not reduce the iprus to a single semantic domain, but enumerates its functions. The primary values of the iprus are perfect (the expression of an anterior and completed event), pluperfect (the same but in the past context), and narrative past tense. Kienast also identifies two other uses which we will present in detail when describing the atypical (or peripheral) functions of the formation, i.e. the performative value (Koinzidenzfall, cf. Kienast 2001: 297) and the future perfect meaning (futurum exactum, cf. ibid.). Whichever model is chosen, the iprus seems to refer to past events or situations, viewed, depending on the scholar’s position, as perfective, punctual, anterior, or simple. However, there are uses of the iprus which do not directly indicate the past. In consequence, one may detect examples where the gram cannot be labelled as a straightforward narrative past, preterite, perfective past or anterior to present. We shall now introduce these atypical uses. According to Loesov, one of the peripheral functions of the iprus, would be the expression of performative meaning (cf. Loesov 2006: 115–17) which frequently corresponds to the present tense in IndoEuropean languages. The performative iprus is commonly used with the verb saparum ‘to send, write’ in the conventional greeting formula of OB letters (cf. Example 3), or in other less formulaic phrases, cf. atam ‘I swear’, ú-na-Ìi-i-id-ka ‘I call your attention, I order you’ (cf. Loesov 2006: 117). (3) ana sulmika aspur-am ‘I wish you well-being’ (Sallaberger 1999: 87–92).
In Loesov’s view, the performative use of the iprus was still productive in Old Babylonian since even non-utterance verbs, i.e. verbs that are not explicitly performative (e.g. sakanum ‘put’, ‘place’ in as-ku-un-ka ‘I demand [this] of you’ in Goetze 1958, No. 47:27–9), could be performativized (cf. Loesov 2006: 117). On the contrary, Denz (1982), Müller (1986), Streck (1995) and Metzler (2002) relate the performative nature of the iprus to its basic perfective aspect or anterior value Consequently, the performative meaning would be consistent with the prototypical domain of the construction (perfect, perfective and past functions). There are other uses of the iprus that do not have any deictic past reference — in such cases, the formation does not equal a preterite or a past. In the majority of temporal clauses the idea of explicit 335
THE AKKADIAN IPRUS
anteriority in reference to the future (or general present) is conveyed by the perfect iptaras. However, in subordinated phrases introduced by the conjunction adi…la, when the time of the main clause is present or future (in fact, this is the only possibility in the case of this conjunction), the iprus (and not the perfect iptaras) expresses the anteriority (cf. Huehnergard 2005: 285): (4) adi abi la illikam ul târ ‘I will not return before my father has come’
Additionally, the iprus may be found in temporal clauses with the particle lama. If the main clause describes a future event, the iprus is used to indicate immediate and certain future actions (cf. Huehnergard 2005: 286): (5) lama ipsurusu alkim ‘Come before they sell / have sold it’
Another atypical meaning of the formation may be found in the case of two verbs, i.e. edûm and isûm, which Testen (2000) labels ‘prefixed statives’. These predicates also show some morphological irregularities. Firstly, in the form iprus they do not display the expected preradical vowel (i and a), but the vowel is invariably i for all persons. Secondly, the verbs in question do not offer, as regular verbs do, three stem configurations (iprus, iparras and iptaras), but employ only one form, i.e. the iprus. The meaning of the iprus of these two verbs is also peculiar. Huehnergard affirms that, even though preterites, they do not have any specific temporal value, and translates them as presents, respectively ‘know(s)’ and ‘have/has’ (cf. Huehnergard 2005: 282). This situation recalls the Indo-European verb *ºoid- / ºid- ‘see’ — its perfect (resultative) in all daughter languages, even though formally perfect or past, has a present meaning (cf. Classical Greek o÷da). Another parallel may be provided by Germanic languages and the preterite-present verbs like kunna or ∫urfa which may be understood as stative. The preterite (formerly perfect) of these verbs is used in all Germanic languages (even already in Gothic) uniquely with the present meaning (Birkmann 1987 and Andrason 2009b). In consequence, the iprus construction can function as a past, both perfective and simple with no perfective or punctual marking (this second possibility does not mean that the aspect is imperfective!). It may be employed as a past tense of narration, i.e. as a pret336
THE AKKADIAN IPRUS
erite. And in main clauses, it usually indicates the temporal context that is anterior to the present. These three uses seem dependent on the perspective adopted by a given scholar — but as we will see, they may be understood as three different contextual realizations of one and the same meaning. Furthermore, in subordinated temporal, and sometimes in principal clauses when the main time reference is the past, the formation equals the past anterior (i.e. anterior to the past or past perfect) — the pluperfect. In clauses with adi…la and lama it appears with the present-future time reference, and indicates respectively anteriority to the future, and immediate future events. Finally, the iprus form of two verbs edûm and isûm has stative meaning with no explicit temporal information; in the present or general time context, it corresponds to the Indo-European present tense. Additionally, the iprus may appear in modal expressions as the precative and vetitive (cf. Huehnergard 2005: 144–7, MalbranLabat and Vita 2005: 102). The uses in the injunctive constructions may additionally support the aspectual theory of the iprus since in this function the gram refers to the future time sphere.5 It is possible to summarize the typical and atypical values of the iprus in the following table (cf. Table 1): Value
Uses
Preterite (simple past and past narrative)
Main clauses
Past perfective (punctual)
Main clauses
Anterior to present
Main clauses
Performative
Mainly, but not only with performative verbs
Pluperfect, i.e. past perfect
Temporal clauses and main clauses
Anterior to the future
Temporal clauses with adi … la
Immediate future
Temporal clauses with lama
Present (past) stative (atemporal)
Prefixed stative edûm and isûm
Precative and vetitive
With prefixes (li-,lu-, i- and ayy-/e-)
Table 1. The uses of the iprus 5
The uses of the iprus in injunctions may in fact reflect the Proto-Semitic situation. According to Huehnergard (1988:22) ‘[b]oth the zero-form yaqtul and […] the imperfect, Central Semitic yaqtulu, were probably unmarked for mood: both could occur in both statements and injunctions. What distinguished the two forms, accordingly, was not mood but primarily a matter of aspect, and secondarily of tense: yaqtul is a perfective or punctual form, temporally a specific past; yaqtulu is an imperfective or durative form, temporally a future’. 337
THE AKKADIAN IPRUS
If one tries to relate the above mentioned uses and to find one synchronic formula covering them all, one faces a hard if not impossible task of explaining such a complex amalgam of values. How is the future perfect connected to the present stative or to the narrative past? In what sense is it possible to relate the stative meaning of edûm and isûm to examples of many fientive verbs that in the iprus show aspectually perfective value? And in what way is the discursive nature of performative expressions cognate to the narrative? Is the iprus form an amalgam of random meanings and uses? The formation certainly appears chaotic if one limits the description to the synchronic perspective. However, if we analyse the iprus from the unidirectional perfective, as a developing gram, the construction stops being semantically heterogeneous and instead of displaying a set of random and unrelated meanings becomes the expression of a single growing grammatical unit (cf. the Figure 6, see p. 339); in the case of the iprus we are dealing with a resultative gram that progresses in three temporal contexts following the resultative path which itself consists of two drifts (anterior and simultaneous).6 This understanding of the Akkadian formation coincides with the regular evolution of the resultative as posited in part 1 above. The main functions match the anterior drift in the present time context. The construction may be used as anterior, perfective past or simple past of narration. This means that, the gram has acquired all possible values on the scale of the anterior drift. However, even if it provides uses that reflect all stages of the anterior drift, its main function corresponds to the perfective past and narrative past. Furthermore, one must also take into consideration other resultative constructions which, being aggressive younger grams, reduce the old semantic domain of the iprus, i.e., the stative and perfectal values (cf. the doughnut gram phenomenon). The usual expression of the perfect is the iptaras, and the regular stative is conveyed by the permansive parsaku. 6 It is important to note that one gram may follow several drifts (or sub-paths) at the same time. For example, a resultative construction can simultaneously follow three drifts established for the resultative path, i.e., anterior drift, simultaneous drift, and evidential drift. There are numerous examples of such a multi-faced development, e.g. the PIE perfect that in the majority of verbs became a past (e.g. in Greek, Latin and Gothic). However, some stative verbs had followed the simultaneous drift and thus became presents (cf. Latin memini, Greek o÷da, and preterite-present verbs in Germanic languages). As observed by Graves (2000) and Lindstedt (2000), in Macedonian and in some other Slavic languages the same morphology has followed both the anterior and evidential drifts (evidential drift is the third possible drift that constitutes the resultative path, cf. Bybee, Perkins and Pagliuca 1994: 95).
338
THE AKKADIAN IPRUS edûm and isûm PAST STATIVE
PAST
PAST PERFECT
REMOTE
PAST CONTEXT temporal clauses, main clauses in the past time reference main clauses regular uses
ANTERIOR
PERFECT
PERFECTIVE
PAST
RESULTATIVE PRESENT CONTEXT STATIVE
PRESENT
edûm and isûm
edûm and isûm FUTURE STATIVE
FUTURE
FUTURE PERFECT/IMMEDIATE
FUTURE
FUTURE CONTEXT
clauses with adi … la and lama
Figure 6. The iprus as a developing resultative gram7
Consequently, this situation (i.e., the coexistence of three resultative diachronies) leads to the phenomenon of layering. The anterior drift in non-present time spheres is more conservative and shows clear perfectal uses as pluperfect and future perfect. As usual, the simultaneous drift is less active being restricted to two stative verbs edûm and isûm. With the present time reference it conveys a present stative or simply present meaning. Less commonly, it appears in the past and future time context giving respectively a past (stative) and future (stative). It is also possible that in the case of the iprus one deals with a zero gram (cf. Dahl 1985 and Bybee, Perkins and Pagliuca 1994: 90–1), and the perfective value of the iprus is a consequence of the development of the iparras. Originally, the iprus was a resultative gram with no aspectual value. However, due to the development of the imperfective iparras (cf. imperfective path in Bybee, Perkins and Pagliuca 1994 and Dahl 2000), the non-aspectual iprus became a default per7 It should be noted that the author does not attempt to determine a discrete location on the evolutionary path for each and every one of the meanings of the iprus. These graphic locations are simplifications. It is obvious that the semanticfunctional development of the iprus reflects a gradual and non-discrete progress on the resultative path.
339
THE AKKADIAN IPRUS
fective category. It must be stressed that the aspectual meaning was not a prototypical characterization of the Akkadian formation — the iprus took up the space left over by the overtly marked iparras. Nevertheless, the iprus may still provide a general perspective of an event, which is not limited to the perfective aspect (cf. the example given by Loesov above). A similar phenomenon, but with a marked perfective and an unmarked imperfective (the situation opposite to the relation between the iprus and the iparras) may be found in Slavic languages, for example in Polish, where some imperfective verbs can still (of course only in determined contexts) provide a perfective view of events. In consequence, the simple past meaning of the iprus — less common than the perfective value — may stem from its origin (the resultative was not marked aspectually) as well as from the universal development of the past perfective into a simple past. However, in the case of the iprus the gram is far from being grammaticalized as an expression of all past events since the language includes a core category that describes imperfective past events, i.e. the iparras. The simple past meaning of the iprus does not mean that the aspect is imperfective — it means that the form is aspectually unmarked, i.e., neither perfective nor imperfective. The development of the iprus in terms of a zero gram may be summarized in the following way, cf. the Figure 7: origin
Akkadian
peripheral use?
core use
NON-ASPECTUAL RESULTATIVE
ASPECTUAL ZERO GRAM
‘objective’ peripheral use?
SIMPLE PAST
Figure 7. The iprus as a zero gram
From the typological perspective, the Akkadian example is not isolated, and a parallel, both synchronically and diachronically, situation may be found in French in the case of the passé composé. Let us first present all possible meanings of the French gram according to Grevisse 1975 (cf. Table 2, see p. 341). Like the iprus, the passé composé seems to be synchronically an amalgam of unrelated meanings. However, diachronically one deals with a single resultative gram that has its roots in the Latin possessive periphrasis habeo scriptum. The passé composé shows all possible values situated on the scale of the anterior drift: resultative, anterior, perfec340
THE AKKADIAN IPRUS
Values of the passé composé with the auxiliary in the present tense
Uses
Past – anterior to the present Past perfective opposite to the imperfect Narrative past – preterite Present perfect Present resultative, or present Immediate future Future perfect Present of the general truth
Main clause Main clause Main clause – orally (seldom literally) Main clause Main clause, cf. il est mort Main clause – discursive Temporal clauses Proverbs
Table 2: The meanings of the passé composé with the auxiliary in the present tense
tal, perfective past, simple past and narrative. Since, the present tense (the tense of the auxiliary in forms like j’ai terminé) in French may also indicate future events, the construction can provide future perfect or future immediate information. The passé composé has not lost its resultative and perfect uses because there is no grammaticalized new resultative or perfect. The only new resultative expression (venir de + infinitive) has still a lexical force, or at least has not been grammaticalized entirely. As in the case of the Akkadian iprus, the anterior drift in the past and future temporal context (the form is then respectively j’avais terminé and j’aurais terminé) is less advanced and more conservative — this means that the uses with the non-present time reference are practically reserved to perfectal or resultative ones. 3. Iprus and Proto-Semitic yaqtul If the description of the Akkadian iprus as a dynamic category (i.e. as a resultative diachrony) is correct, we should be able to relate this fact to Proto-Semitic and to other Semitic languages. In consequence, two questions must be answered. First, how does this model fit into the description of the Proto-Semitic verbal system? And secondly, is the resultative character of the iprus preserved (at least residually) in other Semitic languages? According to Lipinski (2001) the proto-Semitic yaqtul 8 was the most unmarked category situated outside the system of aspect. Lipinski defines it as a simple past — a preterite. Huehnergard (1988:21) 8
The Proto-Semitic form yaqtul is the ancestor of the Akkadian iprus. 341
THE AKKADIAN IPRUS
affirms that the yaqtul ‘was a single form that expressed the perfective aspect of verbal actions […] and injunctions’. He adds that the yaqtul could also have a temporal value functioning as ‘a specific past’ (Huehnergard 1988: 22). Kienast (2001: 334) claims that the yaqtul was the basic form in the Proto-Semitic verbal system and sees its origin in an agglutination of the proclitic personal prefix and the nominal predicate. In his view, the contrast between the primary form yaqtul and the ‘Steigerungsform’ yaqattal consists of aspect, respectively the perfective vs. the imperfective. The same aspectual opposition between the yaqtul and yaqattal was posited by Diakonoff (1965: 95, and 1988: 89) and Bubenik (1998: 44). However, Diakonoff and Bubenik reconstruct the imperfective as yaqatal — in their opinion the reduplication of the second root consonant was limited to the derived D conjugation (cf. Bubenik 1998: 45). In all these models the yaqtul is the unmarked basic form, while the yaqat(t) al is a secondary, derived and overtly marked imperfective (a dynamic imperfective gram in its development according to the imperfective path; cf. Bybee, Perkins and Pagliuca 1994, Dahl 2000, Bertinetto 2000, Bertinetto, Ebbert and Groot 2000, and Ebbert 2000). This coincides with our definition of the yaqtul (iprus) as an originally unmarked (aspectually and temporally) form — the perfective and past marking would be acquired by default due to the development and grammaticalization of the yaqattal (cf. zero gram). Nevertheless, the past and perfective values could also stem from the development determined by the resultative path (cf. the development from the resultative through the perfective to the past). The yaqtul cannot be reduced in Proto-Semitic to the notion of preterite, since it still shows non-past meanings in Akkadian (cf. part 2). Furthermore, the non-past values of the yaqtul are not limited to Akkadian. They may also be found in Classical Arabic — cf. the use of yaqtul in conditional clauses. Moreover, the Arabic construction lamma yaqtul has clearly a present perfect meaning. The perfectal (both present and past) values of the yaqtul can also be found in the Biblical Hebrew wayyiqtol. Even though this construction usually functions as a preterite, it can sometimes (in fact very seldom) refer to future events presenting them as a single perfective whole (cf. Waltke and O’Connor 1990: 557). Additionally, according to Waltke and O’Connor (1990), the wayyiqtol can describe a present persistent perfective situation. Finally, in Phoenician the proto-Semitic yaqtul may function as the future perfect (futurum exactum, cf. Kienast 2001: 314). There is also another fact that supports the view of the yaqtul as a dynamic phenomenon. The opposition between the yaqtul and yaqat342
THE AKKADIAN IPRUS
tal in Semitic languages is not simply perfective vs. imperfective but may consist of three gradually intensified elements (aspect, tense, and resultativity9-simultaneity). In consequence, the yaqtul is more resultative, aspectually perfective and temporally past-oriented than the yaqattal. Conversely, the yaqattal is more simultaneous, imperfective and non-past-oriented. However this opposition is far from being discrete. One rather deals with two contrastive diachronies: the resultative path vs. the imperfective path. Each one of these trajectories is a gradual development — in consequence, the two grams do not form a perfect dichotomy, but rather two different diachronic movements that at different stages of their developments may surface as an aspectual, resultative-simultaneous, or temporal opposition. In fact, at the same time, the yaqtul and the yaqattal may involve a set of contrasts reflecting achieved, but not necessarily lost, stages of their development. Thus, it is no longer surprising that in one language, even in one context, this opposition may be complex and consist of three segments, i.e. aspect, tense, and resultativity-simultaneity. 4 Conclusion In light of the presented evidence, we may state that the iprus is a resultative gram that grows in accordance with the unidirectional development codified by the resultative path and grammaticalization. This observation enables us to claim that, in the case of the Akkadian formation, one does not deal with a single semantic category (past, perfective aspect, anterior etc.) but rather with a single diachrony. In consequence, from the unidirectional perspective, the synchronic set of chaotic and unrelated values displayed by the iprus turns into a homogeneous and regular picture — one original resultative input that follows a single diachronic path. We have also observed that the thesis of the resultative nature of the iprus is supported by data provided by other Semitic languages, and furthermore, it perfectly fits into the proto-Semitic model. The positive verification of the thesis whereby the iprus should be understood as an originally resultative construction that undergoes the regular development in terms of the resultative paths leads to another conclusion. According to evolutionary linguistics, meanings that correspond to less advanced stages on the anterior drift (a sub-path of the resultative path) should be older while those that reflect more advanced 9
Resultativity in a broad sense as was posited in part 1. 343
THE AKKADIAN IPRUS
stages are supposed to be younger. This means that, the oldest uses of the iprus should correspond to more resultative and less preterite values, while the youngest ones are expected to have a more temporally explicit character (in particular, they should indicate the past). The corroboration of this hypothesis constitutes the future research project of the author. Address for Correspondence: University of Stellenbosch, Department of Ancient Studies, Private Bag XI, 7602 Matieland, South Africa REFERENCES Andrason, A. 2009a. ‘Comparando diacronías. El resultativo en las lenguas románicas y germánicas’. Tendencias actuales en la investigación diacrónica de la lengua. Actas del VIII congreso Nacional de AJIHLE (Barcelona). 183–95 —— 2009b. ‘Pretérito germánico como un resultative – la situación en el gótico’. Interlingüística XIX (Gerona). 452–65 Bertinetto, M. 2000. ‘The progressive in Romance, as compared with English’, in Östen Dahl (ed.), Tense and Aspect in the Languages of Europe (Berlin). 559–604 Bertinetto, M, K. Ebert and C. de Groot. 2000. ‘The progressive in Europe’ in Östen Dahl (ed.), Tense and Aspect in the Languages of Europe (Berlin). 517–58 Birkmann, T. 1987. Präteritopräsentia. (Tübingen) Bubenik, V. 1998. ‘Grammatical and lexical aspect in Akkadian and Proto-Semitic’, in Monica Schmid (ed.), Historical Linguistics. Amsterdam Studies in the Theory and Historical Linguistics vol. 164. (Amsterdam, Philadelphia). 41–56 Bybee, J., R. Perkins and W. Pagliuca. 1994. The Evolution of Grammar. (Chicago) Bybee, J. and Ö. Dahl. 1989. ‘The creation of tenses and aspect systems in the languages of the world’, Studies in Language 13: 51–103 Dahl, Ö. (ed.). 2000. Tense and Aspect in the Languages of Europe. (Berlin, New York) Dahl, Ö. 2000. ‘The tense-aspect systems of European languages in a typological perspective’ in Östen Dahl (ed.), Tense and Aspect in the Languages of Europe (Berlin). 3–25 —— 1985. Tense and aspect systems. (Oxford) Denz, A. 1982. ‘Die Struktur des Klassischen Arabisch’, in W. Fischer (ed.), Grundriß der Arabischen Philologie (Band I: Sprachwissenschaft, Wiesbaden). 58–109 Diakonoff, I. 1965. Semito-Hamic Languages. (Moscow) —— 1988. Afrasian Languages. (Moscow) Ebert, K.. 2000. ‘Progressive markers in Germanic languages’ in Östen Dahl (ed.), Tense and Aspect in the Languages of Europe (Berlin). 605–53 Goetze, A. 1958. Fifty Old Babylonian Letters from Harmal. Reprinted from Sumer vol. 14. (Baghdad) Grevisse, M.. 1975. Le bon usage. (Gembloux) Graves, N. 2000. ‘Macedonian – a language with three perfects?’ in Östen Dahl (ed.), Tense and Aspect in the Languages of Europe (Berlin). 479–94 Heine, B. and M. Reh. 1984. Grammaticalization and Reanalysis in African Languages. (Hamburg) 344
THE AKKADIAN IPRUS
Hopper, P. and E. Traugott. 2003. Grammaticalization. (Cambridge) Huehnergard, J. 2005. A Grammar of Akkadian. (Winona Lake) —— 1988. The Early Hebrew Prefix-Conjugations. Hebrew Studies 29. 19–23 Kienast, B. 2001. Historische Semitische Sprachwissenschaft. (Wiesbaden) Knudsen, E.E. 1986. ‘Innovation in the Akkadian Present’, Orientalia Suecana 33–5, 231–9 Kury¥owicz, J. 1965. ‘Zur Vorgeschichte des germanischen Verbalsystems’, in Beiträge zur Sprachwissenschaft, Volkskunde und Literatürforschung: Wolfgang Steinitz zum 60. Geburstag. (Berlin) 242–47 Landsberger, B. 1926. ‘Die Eigenbegrifflichkeit der babylonischen Welt’, Islamica 2, 355–72 Lindstedt, J. 2000. ‘The perfect – aspectual, temporal and evidential’ in Östen Dahl (ed.), Tense and Aspect in the Languages of Europe (Berlin). 365–83 Lipinski, E. 2001. Semitic Languages Outline of a Comparative Grammar (Uitgeverij Peeters en Departement Oosterse Studies Leuven. Paris) Loesov, S. 2006. ‘Akkadian Sentences about the Present Time — Part One’, in Leonid Kogan, (ed.), Babel and Bibel 2: Memoriae Igor M. Diakonof. (Annual of Ancient Near Eastern, Old Testament, and Semitic Studies. Winona Lake). 101–48. Malbran-Labat, F. and J.-P. Vita. 2005. Manual de lengua acadia vol.1. (Saragossa) Maslov, J. 1988. ‘Resultative, Perfect and Aspect’, in Vladimir Nedjalkov (ed.), Typology of Resultative Constructions. (Amsterdam). 63–85 Meillet, A. 1948. ‘L’évolution des formes grammaticales’, Linguistique historique et linguistique générale. (Paris) Metzler, K.A. 2002. Tempora in altbabylonischen literarischen Texten (Alter Orient und Altes Testament 279). (Münster) Müller, H.P. 1986. ‘Polysemie im semitischen und hebräischen Konjugationssystem’, Orientalia 55, 365–9 Nedjalkov, V and S. Jaxontov. 1988. ‘The Typology of Resultative Constructions’, in Vladimir Nedjalkov (ed.), Typology of resultative constructions (Amsterdam). 3–63 Sallaberger, W. 1999. “Wenn Du mein Bruder bist,…”. Interaktion und Textgestaltung in altbabylonischen Alltagsbriefen. (Leiden) Streck, M.P. 2003. ‘Sprache und Denken im Alten Mesopotamien am Beispiel des Zeitausdrucks’, in L. Kogan (ed.). Studia Semitica. (Orientalia. Papers of the Oriental Institute. Issue 3, Moscow). 424–31 —— 1999. ‘Das “Perfekt” iptaras im Altbabylonischen der Hammurapi-Briefe’, in N. Nebes (ed.) Tempus und Aspekt in den semitischen Sprachen (Jenaer Kolloquium zur Semitischen Sprachwissenschaft = Jenaer Beiträge zum Vorderen Orient 1, Wiesbaden). 101–26 —— 1995. Zahl und Zeit. Grammatik der Numeralia und des Verbalsystems im Spätbabylonischen. (Cuneiform Monographs Band 5. Groningen) Testen, D. 2000. ‘Conjugating the “prefixed stative” verbs in Akkadian’. Journal of Near Eastern Studies 59 no. 2. 81–92 Traugott, E. and B. Heine (eds). 1991. Approaches to Grammaticalization. (Amsterdam) Waltke, B. and M.P. O’Connor. 1990. An Introduction to Biblical Hebrew Syntax. (Winona Lake)
345
Journal of Semitic Studies LV/2 Autumn 2010 doi: 10.1093/jss/fgq002 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY: A PRELIMINARY ANALYSIS1 CYNTHIA L. MILLER UNIVERSITY OF WISCONSIN–MADISON AND UNIVERSITY OF THE FREE STATE
Abstract Vocatives are syntactically interesting items. On the one hand, they are intonationally part of a host sentence. On the other hand, they are only loosely linked syntactically with their host sentence and are not integrated into it as a clausal constituent. Previous descriptions of the vocative in Biblical Hebrew have largely considered the internal syntax of the vocative (e.g. the kinds of constituents that can be used as vocatives). The external syntax of the vocative, however, has been largely ignored. This article considers the external syntax of the vocative with respect to the locations in the host sentence that can serve as a niche for the vocative as an extrasentential element. Special attention is paid to the significant ways in which prose and poetry differ in this respect.
Introduction The vocative is a syntactically peculiar entity. 2 On the one hand, the vocative is phonologically related to its host sentence and may even appear in the middle of it, as illustrated in (1): 1
This paper was originally presented in the Linguistics and Biblical Hebrew section of the 2006 annual meeting of the Society of Biblical Literature in Washington D.C. A revised version was presented to the Oberseminar of the Department of Ancient Studies at the University of Stellenbosch, May 29, 2007. I am grateful to the participants on both occasions for their helpful comments, and especially to Christo van der Merwe, Jacobus A. Naudé, John A. Cook, and Robert D. Holmstedt. The research was supported by the Graduate School of the College of Letters and Sciences of the University of Wisconsin-Madison and by the Ettinger Family Foundation. At various stages of the research, three graduate student assistants, Timothy P. Mackie, Wendy L. Widder and Jared A. Henson, assisted in checking the vocative data. 2 Definitions of the vocative differ. Revell equates the vocative with a ‘free term of address’ (E.J. Revell, The Designation of the Individual: Expressive Usage in Biblical 347
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
(1) 1 Kgs 12:28 הנה אלהיך ישראל אשר העלוך מארץ מצרים ‘Behold your gods, O Israel, which brought you up from the land of Egypt’
In this case, the vocative appears between a head noun and the relative clause which syntactically depends upon it. On the other hand, the vocative is not syntactically integrated into the sentence as a constituent; that is, the proper noun ישראלis a term of address which refers to the collective addressee of the speech event, but it plays no grammatical role in the host sentence, nor does it play a grammatical role within the object noun phrase which it dissects.3 At the same time, the vocative does not require a host sentence, but can stand alone as a complete utterance, as in (2): (2) Gen. 22:1 ויהי אחר הדברים האלה והאלהים נסה את־אברהם ויאמר אליו אברהם After these things God tested Abraham and said to him, ‘Abraham!’
The precise relationship, then, of a vocative to its host sentence is not easy to characterize. It is phonologically integrated with its host sentence on a prosodic level, but only loosely associated on a syntactic level. As a result, the vocative renders the constituent structure of the host sentence discontinuous: ‘between its constituents there Narrative [Contributions to Biblical Exegesis and Theology 14, Kampen 1996], 325). Leech, however, considers the vocative to be a specialized type of term of address (Geoffrey Leech, ‘The Distribution and Function of Vocatives in American and British English Conversation’, in Hilde Hasselgård and Signe Okselfjell [eds], Out of Corpora: Studies in Honour of Stig Johansson [Language and Computers: Studies in Practical Linguistics 26, Amsterdam 1999], 107). I follow Zwicky in considering vocatives to be constituents that refer to the addressee, but do not serve as an argument of the verb (Arnold M. Zwicky, ‘Hey, Whatsyourname!’, in Michael W. Lagaly, Robert A. Fox, and Anthony Bruck [eds], Papers from the Tenth Regional Meeting, Chicago Linguistic Society, April 19–21, 1974 [Chicago 1974], 788–801). As Zwicky notes, vocatives may be used as calls (to get someone’s attention) or as addresses (to establish or reinforce contact between speaker and addressee). 3 For linguistic arguments that the vocative phrase ‘does not belong to the thematic grid of a predicate,’ even in cases where it may be cross-referenced with a pronoun within the thematic grid of the predicate, see Andrea Moro, ‘Notes on Vocative Case: A Case Study in Clause Structure’, in Josep Quer et al., (eds), Romance Languages and Linguistic Theory 2001: Selected Papers from ‘Going Romance,’ Amsterdam, 6–8 December 2001 (Amsterdam Studies in the Theory and History of Linguistic Science, Current Issues in Linguistic Theory, 245, Amsterdam 2003), 247–61. 348
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
is something that is not a constituent of it’.4 Because of the vocative’s relationship to ‘the locutionary act that the speaker performs in uttering the host sentence’, some linguists have seen the structural tie between the vocative and its host as based in ‘the structure of action rather than in syntactic structure per se’.5 But even if vocatives are viewed as members of extrasentential discourse units, they nonetheless exhibit both internal and external syntax, which must be described.6 Previous descriptions of Biblical Hebrew vocatives have largely focused on the internal syntax of vocatives. Grammarians have enumerated the kinds of constituents which can be used as vocatives (personal names, nouns and noun phrases, participial clauses, headless relative clauses, etc.) with special consideration of the definite article with vocative nouns.7 Attention has also been paid to the sociolinguistic features of vocatives, that is, the ways in which vocatives provide explicit indications of the social relationships between speakers and addressees.8 But far less attention has been paid to 4
James D. McCawley, The Syntactic Phenomena of English 2 (Chicago 1998), 751. See also Dirk Panhuis, ‘The Vocative is Outside the Sentence’, Studies in Language 10 (1986), 443–7. 5 McCawley, The Syntactic Phenomena of English, 752. 6 Leech uses the ‘c[ommunicative]-unit’ as the unit of analysis, noting that the c-unit is the ‘spoken analogue’ of the sentence; see Leech, ‘The Distribution and Function of Vocatives in American and British English Conversation’, 108–9. McCarthy and O’Keeffe use the conversational ‘turn’ as the unit of analysis, following conversational analysts; see Michael J. McCarthy and Anne O’Keeffe, ‘“What’s in a Name?”: Vocatives in Casual Conversation and Radio Phone-in Calls’, in Pepi Leistyna and Charles F. Meyer (eds), Corpus Analysis: Language Structure and Language Use (Language and Computers: Studies in Practical Linguistics 46, Amsterdam 2003), 153–85. Both of these articles examine the use of vocatives in corpus studies of spoken English. In terms of the Biblical Hebrew data, we could use the c-unit or the turn as our unit of analysis of narrative, since speakers and their utterances are delimited there. However, in poetry, it is very difficult to determine speakers, much less c-units or turns. Therefore, to facilitate the comparison of narrative and verse, I will follow McCawley (and effectively, Leech) in using the host sentence as the unit of analysis. 7 See Bruce K. Waltke and M. O’Connor, An Introduction to Biblical Hebrew Syntax (Winona Lake 1990), §§9.1a, §19.4b, and p. 128 n 11; W. Gesenius and E. Kautzsch, Gesenius’ Hebrew Grammar, trans. A. Cowley (Oxford 1910), §126e, Paul Joüon and T. Muraoka, A Grammar of Biblical Hebrew2, (Rome 2006), §137g; Ronald J. Williams, Hebrew Syntax: An Outline2, (Toronto 1976), §§34, 89, and Cynthia L. Miller, ‘Definiteness and the Vocative in Biblical Hebrew’, JNSL 36 (2010), 43–65. 8 See especially Revell, The Designation of the Individual, 325–40. 349
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
the external syntax of vocatives, and especially to a consideration of the locations in the host sentence which can serve as a niche for a vocative constituent.9 It is that question which I will examine in this essay with particular attention to the ways that prose and poetry differ in this respect. The prose corpus for analysis consists of Genesis, Joshua, Judges, Samuel, and Kings, with 221 vocatives. The poetic corpus consists of the book of Psalms and the inset poems from the prose corpus, with 475 vocatives.10 A rough comparison of the relative density of vocatives within the prose sample as compared to the poetic sample is intriguing, given the fact that biblical narrative is permeated with direct speech. And yet, vocatives apparently do not occur as frequently in narrative depictions of speech as in the performance of speech acts in the liturgical poetry of the Psalms. 9 Revell considers where the vocative may occur in the host sentence by differentiating three locations: before the clause, after the first word or constituent in a clause of two or more constituents, and after the clause (Revell, The Designation of the Individual, 335–40). He then correlates the position in the clause with the social relationships of speaker and addressee. Rosenbaum describes the position of the vocative in his corpus in terms of whether it precedes (18%), follows (82%), or interrupts the predication (Michael Rosenbaum, Word-Order Variation in Isaiah 40-55: A Functional Perspective [Studia Semitica Neerlandica, Assen 1997], 116–32). Fassberg describes the position of the twenty-eight instances of vocatives within Ben Sira with respect to whether they occur before an indicative clause (eighteen examples), at the beginning of the sentence after the particle ( הויone possible example), before a negated clause (three examples), before a conditional sentence (three examples), within the conditional sentence (two examples), after the initial verb (six examples), and at the end of the clause (two examples); see S.E. Fassberg, ‘On Syntax and Style in Ben Sira: Word Order’, in T. Muraoka and J.F. Elwolde (eds), Sirach, Scrolls, and Sages: Proceedings of a Second International Symposium on the Hebrew of the Dead Sea Scrolls, Ben Sira, and the Mishnah, held at Leiden University, 15–17 December 1997 (Studies on the Texts of the Desert of Judah 33, Leiden 1999), 117–31, esp. 125–9. For other external syntactic features of vocatives, see Carl Brockelmann, Hebräische Syntax (Neukirchen 1956), §10; Waltke and O’Connor, An Introduction to Biblical Hebrew Syntax, §§4.7d, 8.3d; and Christo H.J. van der Merwe, Jackie A. Naudé, and Jan H. Kroeze, A Biblical Hebrew Reference Grammar (Biblical Languages: Hebrew, 3, Sheffield 2002), 249. 10 Two electronic databases of Biblical Hebrew tag vocatives: syntactic database in the Logos system (on the entire Hebrew Bible) produced by Francis I. Andersen and A. Dean Forbes and the Stuttgart Electronic Study Bible (for Genesis–2 Kings) produced by Eep Talstra. There are some theoretical differences among us concerning the identification of vocatives. A. Dean Forbes in personal communication on August 30, 2007, reports that in the Psalms we differ in our analysis of Pss 4:9, 29:9, 59:6, 71:5; 81:9, 91:2, 91:9, 109:27.
350
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
Before presenting an analysis of the external syntax of vocatives, there are three features of their internal syntax which I will mention but not solve. First, I will not deal with the question of vocative lamed ,11 vocative kî 12 vocative waw,13 or vocative yah.14 Second, I will not include in the analysis apparently discontinuous vocatives, as identified by Andersen-Forbes in their electronic database, and illustrated in (3): (3) Ps. 80:2 רעה ישראל האזינה נהג כצאן יוסף ‘O shepherd of Israel, hear, [who] leads Joseph like a flock’
In cases like this, where a participle has no overt subject in the sentence proper, Andersen-Forbes usually understand a noun phrase within a vocative phrase to provide the subject, even if other sentential constituents intervene.15 Such an analysis is certainly possible, especially given the fact that a relative clause can be separated from its head noun in cases where a vocative is not involved, as in (4): (4) Gen. 24:15 והנה רבקה יצאת אשר ילדה לבתואל And behold, Rebekah was coming out who was born to Bethuel
Third, following an emerging consensus among linguists as well as some Hebraists, I do not consider vocatives to be the grammatical subjects of imperatives.16 In Hebrew, we can see this most clearly in 11
See e.g. Mitchell Dahood, ‘Vocative Lamedh in the Psalter’, VT 16 (1966), 299–311; idem, ‘The Grammar of the Psalter’, in Psalms III: Psalms 101–150 (Anchor Bible 17A, Garden City 1970), 407–8; idem, ‘Vocative Lamedh in 1 Kings 19,10.14’, Biblica 54 (1973), 407–8; idem, ‘Vocative Lamedh in Psalm 74,14’, Biblica 59 (1978), 262–3; idem, ‘Vocative Lamedh in Exodus 2,14 and Merismus in 34,21’, Biblica 62 (1981), 413–15; Marvin Pope, ‘Vestiges of Vocative Lamedh in the Bible’, UF 20 (1988), 201–7; and cf. the critique of Patrick D. Miller, Jr., ‘Vocative Lamed in the Psalter: A Reconsideration’, UF 11 (1979), 617–37. 12 Mitchell Dahood, ‘Vocative kî and wa in Biblical Hebrew’, in Mélanges offerts au R. P. Henri Fleisch (Mélanges de l’Université Saint-Joseph 48, Beirut 1973–4), 51–63. 13 See e.g. Mitchell Dahood, ‘Vocative waw in Psalm 30,9’, Biblica 58 (1977), 218; idem, ‘Vocative kî and wa in Biblical Hebrew’, in Mélanges offerts au R.P. Henri Fleisch (Mélanges de l’Université Saint-Joseph 48, Beirut 1973–4), 51–63. 14 On vocative yah, see recently Josef Tropper, ‘Die Vokativpartikel ya h im Hebräischen’, ZAH 15–16 (2002[2003]), 168–71. 15 See e.g. Pss 97:10; 102:25; 104:1–4. 16 James D. McCawley, ‘Speech Acts and Goffman’s Participant Roles’, in Gloria Alvarez, Belinda Brodie and Terry McCoy (eds), East States Conference on Linguistics 351
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
cases where deictic indicators within the vocative phrase are thirdperson rather than second person, as in (5): (5) Mic. 1:217 שמעו עמים כלם הקשׁיבי ארץ ומלאה ויהי אדני יהוה בכם לעד אדני מהיכל קדשו ‘Hear, O peoplesi, all of themi (*you); listen, O earthj and itsj (*your) fullness, so that the Lord GOD may be youri+j accuser, the Lord from his holy temple’
The relationship of an independent second-person pronoun to an imperative, however, requires additional considerations. Since Hebrew is a pro-drop language, an overt, independent subject is always optional. Like other inflected verbs, the imperative may optionally have an independent subject pronoun, as in (6): (6) Gen. 9:7 ואתם פרו ורבו שׁרצו בארץ ורבו־בה ‘And you be fertile and increase; abound on the earth and increase on it’
And, like other inflected verbs, the subject of the imperative may optionally be followed by a vocative phrase, as in (7):18 (7) Ps. 109:21 ואתה יהוה אדני עשה־אתי למען שׁמך ‘and you, O GOD, my Lord, deal with me according to your name’
We can understand (7) as essentially the same as a sentence with a second-person pronoun that is the overt subject of a perfect verb, as in (8):
Held at the Ohio State University, September 28–30, 1984 (Columbus 1985), 260– 74; Leech, ‘The Distribution and Function of Vocatives in American and British English Conversation’, 107–9; Revell, The Designation of the Individual, 325–6; T. Muraoka, Emphatic Words and Structures in Biblical Hebrew (Jerusalem 1985), 47–59. For some of the linguistic complexities of subjects of imperatives, see Simon Mauck et al., ‘Imperative Subjects: A Cross-linguistic Perspective’, in C. Brandstetter and D. Rus (eds) Georgetown University Working Papers in Theoretical Linguistics (Georgetown 2005), 135–52. 17 See also 1 Kgs 22:28; Isa. 44:23, 54:1; Ezek. 21:30; Mic. 1:2. 18 Waltke and O’Connor argue that ‘vocatives stand in apposition to the secondperson pronoun, expressed or unexpressed’ (An Introduction to Biblical Hebrew Syntax, §4.7d). 352
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
(8) 2 Sam. 7:24 ואתה יהוה היית להם לאלהים ‘and you, O LORD, became their God’
Here the vocative intervenes between the subject and the perfect verb. Geoffrey Leech has argued that you in English is never a vocative when another vocative occurs in the same sentence, as indicated by the ungrammaticality of reading both you and son as vocatives in (9a). In (9b) and (9c), you is the subject of the imperative and has the normal intonation that a subject would have in that position; in (9d), you is a vocative and has the intonational contour of a vocative.19 (9a) (9b) (9c) (9d)
*You, be quiet, son. You be quiet, son. You be quiet. You, be quiet.
Although אתהcan be understood as the subject of the imperative and not a vocative in (6) and (7), example (10) is different: (10) Ps. 86:2 הושׁע עבדך אתה אלהי הבוטח אליך ‘Save your servant, you my God, who trusts in you’
In this example, the independent pronoun אתהfollowed by the vocative אלהיintervenes between the object of the imperative עבדך and the participial phrase which modifies it, הבוטח אליך. Therefore, the pronoun cannot be the subject of the imperative (since it occurs in the middle of the object constituent). Instead, the pronoun must be a vocative term that is appositional to the vocative אלהי. I will therefore assume that when a second-person pronoun followed by a vocative occurs before an imperative, as in (7), the pronoun is the subject of the imperative. I will also assume that a second-person pronoun before an imperative, as in (6), is the subject of the imperative (analogous to English [9c]) rather than a vocative (analogous to English [9d]).20
19
Leech, ‘The Distribution and Function of Vocatives’, 109. Revell considers a second person ‘free pronoun’ with an imperative to be equivalent to the use of a free pronoun in apposition to the subject of a verb form (The Designation of the Individual, 326). 20
353
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
Vocatives in Prose In prose, vocatives are found within direct representations of speech and thus the boundaries of the utterances are clearly delineated. As already noted, a vocative may comprise the entire utterance, as in (2) above,21 or it may commonly occur at the margins of the utterance, either at the beginning or at the end.22 However, I am most interested in those cases where the vocative finds a niche within the utterance and it is those cases to which we now turn. Clause-initial Position In the sense that vocatives are only loosely integrated into the sentence in which they occur, they are very similar to another linguistic category, exclamatives or interjections. Not surprisingly, then, a vocative commonly occurs after an exclamative, as in (11): (11) Judg. 11:35 ויהי כראותו אותה ויקרע את־בגדיו ויאמר אהה בתי הכרע הכרעתני When he saw her, he tore his clothes and said, ‘Alas, my daughter, you have brought me very low…’
Besides the exclamative אהה,23 vocatives in the prose sample appear after הוי,24 בי,25 and the oath formulation חי נפשׁך.26 Example (12) illustrates vocatives after ביand after חי נפשׁך:
21
See also Gen. 22:1, 7; 27:1, 18; 31:11; 46:2; 1 Sam. 3:6, 10, 16; 24:9 For examples in which the vocative occurs at the beginning of the utterance, see Gen. 4:23, 15:2, 8; 16:8; 18:3; 20:4; 23:15; 24:12, 42; 29:4; 32:10; Josh. 7:19; Judg. 11:36; 16:28; 1 Sam. 1:8, 11; 20:30; 23:10; 2 Sam. 19:1, 27; 1 Kgs 1:17, 24; 13:2; 17:20, 21; 18:26, 36; 22:15; 2 Kgs 1:9, 11, 13; 2:4, 12; 5:13; 6:17, 20; 8:5; 13:14; 19:15. For examples in which the vocative occurs at the end of the utterance, see Gen. 22:8; 24:18; 27:18, 20, 26, 34, 37, 38; 43:29; Judg. 3:19; 6:12; 16:9, 12, 14, 20; 1 Sam. 4:16; 17:55, 58; 22:12; 24:17; 26:14, 17; 2 Sam. 2:20; 13:11; 14:4; 15:31; 16:4, 7; 19:1, 26; 20:9; 1 Kgs 12:16; 13:30; 18:7, 17; 19:9, 13; 21:20; 22:28; 2 Kgs 2:23; 4:40; 5:25; 6:21, 26; 9:5, 22, 23. 23 Other examples include: Josh. 7:7; 2 Kgs 6:5, 15. 24 1 Kgs 13:30. 25 Other examples after ביinclude: Gen. 43:20; 44:18; Josh. 7:8; Judg. 6:13, 15; 13:8; 1 Sam. 1:26; 1 Kgs 3:17; 3:26. 26 See also 1 Sam. 17:55; 2 Sam. 14:19; cf. 1 Sam. 25:26, where יהוה חיfollows the vocative. 22
354
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
(12) 1 Sam. 1:26 ותאמר בי אדני חי נפשך אדני אני האשה הנצבת עמכה בזה להתפלל אל־יהוה She said, ‘Oh, my lord! As you live, my lord, I am the woman who was standing with you here to pray to the LORD’
Vocatives also find a niche in clause-initial position after sentential modifiers such as interrogatives,27 sentential adverbs,28 the presentative הנה,29 and negatives.30 As one example, consider (13): (13) Judg. 21:3 ויאמרו למה יהוה אלהי ישראל היתה זאת בישראל להפקד היום מישראל שבט אחד They said, ‘Why, O LORD God of Israel, has this happened in Israel that one tribe in Israel is missing today?’
Vocatives, then, often occur at the beginning of the sentence.31 When not absolutely at the beginning, they frequently occur immediately after exclamatives and sentence-modifiers and before the sentence proper. Clause-final Position In clause-final position vocatives typically appear between syntactically related clauses.32 In (14), the vocative appears at the end of the matrix clause before the dependent clause introduced with פן:
27
See 1 Kgs 1:13 (cf. 1 Kgs 1:17) with ;הלא אתהJudg. 21:3 with למה. For examples with ועתה, see Gen. 27:8; 27:43; 1 Sam. 25:26; 2 Sam. 7:25; 7:28; 13:20; 24:10; 1 Kgs 3:7; 8:25, 26; 2 Kgs 19:19. See also 2 Kgs 20:3 with ;אנה1 Kgs 19:4 with ;רב עתהand 2 Kgs 19:17 with אמנם. 29 See Gen. 19:2; Judg. 20:7. 30 See Gen. 19:7; 19:18 with ;אל נאGen. 23:11; 42:10; 1 Sam. 1:15; 2 Kgs 6:12 with ;לאGen. 48:18 with לא כן. The negative particle אלmay both precede and follow the vocative; see Judg. 19:23; 1 Sam. 2:24; 2 Sam. 13:12, 25; 2 Kgs 4:16. 31 Often the beginning of the sentence is also the beginning of the utterance; see n. 22 above for examples of vocatives at the beginning of the utterance. In some cases, however, the vocative is the beginning of a sentence which does not begin the utterance (e.g. 1 Sam. 23:11). 32 For examples in which the vocative is at the end of a clause and the subsequent clause is syntactically independent, see Gen. 15:1; 21:17; 23:6; 24:31; 27:13, 38; 33:9; Josh. 7:13; Judg. 4:18; 9:7; 16:28; 1 Sam. 3:6; 2 Sam. 7:18, 19, 20, 22; 18:22; 1 Kgs 12:16; 14:6; 17:18; 18:37; 2 Kgs 2:23. 28
355
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
(14) 1 Sam. 4:9 התחזקו והיו לאנשים פלשתים פן תעבדו לעברים כאשר עבדו לכם ‘Take courage and be men, O Philistines, lest you become slaves to the Hebrews just as they have been slaves to you…’
In (15), the vocative ends the matrix clause before a dependent infinitival clause that begins with לשׁמע: (15) 1 Kgs 8:28 ופנית אל־תפלת עבדך ואל־תחנתו יהוה אלהי לשמע אל־הרנה ואל־התפלה אשר עבדך מתפלל לפניך היום ‘… and turn to the prayer of your servant and to his plea, O LORD my God, to listen to the cry and to the prayer which your servant prays before you today.’
We can now explain the position of the vocative in (1) above — the vocative intervenes at the end of the matrix clause before a dependant relative clause.33 This explanation also helps us to understand (16): (16) 1 Sam. 22:16 ויאמר המלך מות תמות אחימלך אתה וכל־בית אביך The king said, ‘You will certainly die, Ahimelech, you and all your father’s house.’
The coordinate phrase ‘you and all your father’s house’ is co-referential. The vocative intervenes between the matrix clause and a coordinate noun phrase (‘you and all your father’s house’) that is (partially) coreferential with the subject of the matrix clause, but is not adjacent to it.34 Clause-medial Position In the prose corpus, only four constructions involve a vocative that finds a niche within the centre of the clause, at least on the surface. None occurs more than five times. 33 Other examples in which the vocative is in clause-final position of the matrix clause with a dependent clause following include: Gen. 27:21; 48:19 (the matrix verb is repeated after the vocative and before the dependent clause); 2 Sam. 14:22; 19:23; 23:17. 34 The analysis of this verse follows Naudé in assuming that a coordinate noun phrase that is co-referential with an independent subject pronoun is an adjunct; see Jacobus A. Naudé, ‘Syntactic Aspects of Co-ordinate Subjects with Independent Personal Pronouns’, Journal of Northwest Semitic Languages 25 (1999), 75–99.
356
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
In the first construction, the independent second-person pronoun occurs in initial position with the vocative following, as in (17):35 (17) 2 Sam. 7:27 כי־אתה יהוה צבאות אלהי ישׂראל גליתה את־אזן עבדך ‘… for you, O LORD of hosts, the God of Israel, revealed to your servant…’
Since the usual word order after כיis verb-subject, the fact that the order here is subject-verb must mean that the pronoun אתהis highlighted. In the context of the passage, the pronoun in initial position is a focused constituent — David reiterates that it is the LORD and no one else who has revealed to him that he would build a house for God. The placement of the vocative between the subject pronoun and the remainder of the sentence further serves to highlight the fact that the pronoun is in contrastive focus.36 In the second construction, the vocative intervenes between the subject and predicate of a verbless clause, as in (18): (18) 1 Sam. 25:24 ותפל על־רגליו ותאמר בי־אני אדני העון She fell at his feet and said, ‘On me alone, my lord, be the guilt…’
Again, the prepositional phrase that serves as the predicate is in contrastive focus — Abigail asks that the guilt be on her as opposed to her husband. The pronoun אניin apposition to the first-person suffix on the preposition provides further evidence of the focused nature of the predicate. In each of the other three instances in our prose sample in which a vocative intervenes between the subject and predicate of a verbless clause, the constituent in initial position likewise seems to be in contrastive focus.37 In the third construction, the vocative intervenes between a verb and its object: 35
See also 2 Sam. 7:29; 1 Kgs 1:20 (reading with the MT). For a discussion of basic word order in Biblical Hebrew and the notion of contrastive focus, see Robert D. Holmstedt, ‘Word Order and Information Structure in Ruth and Jonah: A Generative-Typological Analysis,’ JSS 54 (2009), 111– 39. 37 See 1 Kgs 18:37; 2 Kgs 19:19 for the vocative intervening between subject and predicate; see 2 Sam. 14:9 for the vocative intervening between predicate and subject. 36
357
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
(19) 2 Kgs 19:16 הטה יהוה אזנך ושמע פקח יהוה עיניך וראה ‘Incline, O LORD, your ear and hear. Open, O LORD, your eyes and see.’
The two examples in this verse are the only examples of this construction in all of the prose corpus, but it is a very common construction in the Psalms in petitions to God. It is interesting, then, that our only prose examples occur within a prayer and never in petitions to kings or other superiors.38 In the fourth and final construction in which the vocative seems to occur in the centre of a clause, there are two examples of a clause-final prepositional phrase that is separated from its clause by a vocative:39 (20) 2 Sam. 13:4 ויאמר לו מדוע אתה ככה דל בן־המלך בבקר בבקר He said to him, ‘Why are you thus haggard, O son of the king, morning after morning?’
The vocative occurs at the end of the core of the clause and before two juxtaposed prepositional phrases that are not arguments of the predication. However, because a vocative phrase has an intonational contour which sets it apart from its host sentence, a constituent that follows the vocative would naturally be highlighted. It is possible, then, that the vocative is used to draw rhetorical attention to the iterative nature of Amnon’s distress that the prepositional phrases describe. This type of construction is also quite common in poetry. Summary of Prose Constructions In prose, then, we have seen that in most instances the vocative finds a niche within the utterance at a clause boundary — either at the beginning of a clause boundary or at its end. The exceptions fall into two types. When a clausal constituent at the beginning of the utterance (or sentence) is partitioned off by the vocative, the vocative seems to highlight the informational status of the initial constituent as contrastive focus. When a clausal constituent at the end of the utterance (or sentence) is partitioned off by the vocative, the vocative seems also to provide rhetorical highlighting, though of a less specific nature. 38 The syntax of Josh. 22:25 is difficult. If בני־ראובן ובני־גדis considered a vocative phrase, then את־הירדןas an object is separated from its verb. 39 See also 1 Sam. 30:23.
358
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
Vocatives in Poetry In poetry, all of the features that we identified in prose are found with much greater frequency and more variety. In clause-initial position, for instance, a somewhat different inventory of exclamatives and sentence adverbials is found with the vocative, but no new syntactic features are evident.40 In clause-final position, a vocative may still separate a main clause from a dependent clause, but again we encounter greater variety. For example, in poetry the dependent clause may precede rather than follow the matrix clause:41 (21) Ps. 40:9 לעשׂות־רצונך אלהי חפצתי ‘To do your will, O my God, I desire’
In clause-medial position we find the greatest variety. The sentence constituents may occur in a bewildering variety of orders and the vocative may find a niche between any of the clausal constituents. For example, the vocative may separate a clause-final noun phrase or prepositional phrase from the rest of the clause.42 The vocative may separate an initial prepositional phrase from the rest of the clause.43 The vocative may separate an initial noun phrase that is an adjunct from the remainder of the clause (Ps. 69:14). The vocative may commonly intervene between verb and object,44 or less frequently between subject and verb (Pss 138:4; 145:10). But although the vocative appears in many more locations, in these examples there are no essential differences in its external syntax. 40
See, for example, the frequent use of אנהin clause-initial position with a vocative following in Pss 116:4; 116:16; 118:25 (2≈). 41 See, for example, Pss 25:11; 71:19; 137:5; 143:11 for instances where the dependent clause precedes. For cases where the dependent clause follows, see Pss 40:14; 103:20; 114:5. 42 For a noun phrase separated from the rest of the clause, see Ps. 45:4. For a prepositional phrase separated from the rest of the clause, see, e.g., Pss 7:7, 9; 21:14; 33:1, 22; 51:3; 60:12; 65:2; 79:9; 86:12; 92:5; 93:5; 97:12; 103:22; 106:4; 108:12; 116:7; 119:65; 125:4; 130:7; 131:3; 132:1, 8; 135:9; 137:7; 140:2, 5. 43 Pss 25:1; 28:1; 30:9; 31:2; 38:16; 60:10; 71:1, 18; 76:7; 86:4; 88:14; 119:89. 44 Pss 9:21; 17:1; 24:7, 9; 25:22; 27:7, 11; 35:1; 39:5; 48:10; 55:2; 61:2; 64:2; 66:8, 16; 68:17; 78:1; 85:2; 86:1, 6, 11; 89:51; 103:2, 22; 104:1, 35; 119:33; 126:4; 139:19; 140:7, 9; 141:3; 146:1; 147:12 with word order verbobject; and 25:4; 27:8; 51:19; 94:5; 119:64; 139:21 with word order object-verb. 359
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
There are, however, new features of the external syntax of the clause-medial vocative in poetry. In prose, a vocative does not divide a clausal constituent, except that it can intervene between a head noun and its relative clause. In poetry, however, a vocative may sometimes find a niche between phrasal components of a single clausal constituent. In (22), for example, the vocative divides a coordinate constituent: (22) Ps. 25:6 זכר־רחמיך יהוה וחסדיך Remember your mercy, O LORD, and your loyalty.
The object constituent, רחמיך וחסדיך, is divided by the vocative. Similarly, in (23), the vocative divides two appositional phrases, which together make up a constituent, namely, the prepositional adjunct of the verb ויגלו:45 (23) Ps. 18:16 ויגלו מוסדות תבל מגערתך יהוה מנשמת רוח אפך The foundations of the world were laid bare at your rebuke, O LORD, at the blast of breath from your nostrils.
Conversely, in (24), the vocative divides two prepositional phrases both headed with the same preposition עלwhich should not be understood as appositional. (24) Ps. 57:6 (see also 57:12; 108:6) רומה על־השמים אלהים על כל־הארץ כבודך Exalt yourself over the heavens, O God! May your glory be over all the earth.
The two prepositional phrases belong to two separate clauses and fulfil two different grammatical functions within them — the first one, על־השמים, is a prepositional complement to the verb in the first line and the second one, על כל־הארץ, is the predicate of the verbless clause in the second line. The vocative provides a break between the two prepositional phrases, so that they should not be heard or read as appositional phrases belonging to the same sentence. In fact, if the vocative were not present and a reader initially parsed the two prepositional phrases as appositional, the final word כבודךwould be stranded 45 See also Ps. 86:2, where the vocative separates a head noun phrase from a participial clause that modifies it.
360
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
without a syntactic role in the sentence. The reader would then need to re-analyse the string of words in order to assign the second prepositional phrase to the predicate role of a verbless clause whose subject is כבודך. We can see, then, that the position of the vocative assists the hearer/reader also in correctly ascertaining where the first poetic line (and sentence) ends and where the second line (and sentence) begins.46 It appears, then, that the vocative is frequently used to assist the hearer/reader in inferring line ends. Accepting the argument of the linguist Nigel Fabb that poetic lineation is a matter of implied form and not inherent form,47 we can see that the vocative is especially used as a means to effectively delimit the ends of clause boundaries and thus the ends of lines. This poetic function of the vocative is especially apparent in couplets involving ellipsis. In (25), the vocative occurs at the end of the first line; ellipsis of the verb plus object has occurred out of the beginning of the second line. The direction of ellipsis is thus forwards:48 (25) Ps. 119:41 ויבאני חסדך יהוה תשועתך כאמרתך Let your steadfast love come to me, O LORD, [ ] your salvation according to your promise.
The vocative helps to prevent us from reading חסדךand תשועתךas appositional phrases. Instead, we must consider them to be constituents of different clauses. It is even more common, however, for the vocative to occur between lines involving ellipsis when the direction of ellipsis is backwards; that is, when ellipsis has occurred out of the first line, as in (26):49 (26) Ps. 94:3 עד־מתי רשעים יהוה עד־מתי רשעים יעלזו ‘How long the wicked [ ], O LORD, how long the wicked will exult?’ 46 Alternatively, we could consider the second line to exhibit ellipsis of the verb from the first line, as illustrated in the following examples. In that analysis, the vocative also assists in ascertaining the end of the first line. 47 Nigel Fabb, Language and Literary Structure: The Linguistic Analysis of Form in Verse and Narrative (Cambridge 2002), 136–77. 48 See also Pss 35:23; 68:25; 114:7. 49 For other examples involving backwards ellipsis and a vocative strategically placed between lines, see Pss 29:1; 55:10; 89:52; 93:3; 96:7 (with the object deleted backwards); Pss 92:10; 115:1 (with the verb deleted backwards).
361
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
In this verse, the verb has been elided backwards from the second line.50 Assigning the correct structure to the two lines is facilitated by the vocative. Additionally, the extensive repetition of the interrogative עד־מתיand the subject רשׁעיםmakes the line end unmistakable and allows the restoration of the verb in the first line without difficulty. In (27), the vocative assists in delimiting the lines, but now the vocative is in initial position in both lines: (27) Ps. 70:2 אלהים להצילני יהוה לעזרתי חושה O God, to save me [ ] / O LORD, to help me hurry.
Backwards ellipsis of the main verb has occurred and all that remains in the first line is the infinitival complement of the verb.51 We can explain the use of the vocative as an indicator of clause ends and thus of line ends as follows. The vocative usually occurs at the margins of clauses. The vocative also has an intonational contour that sets it off from the speech that surrounds it. When ellipsis occurs within poetic lines, the fragmented surface structure must be restored by the hearer/reader, who accomplishes this feat only by properly aligning the two lines, matching their structures, and determining what belongs in the gap. If the lines cannot be correlated, the fragmented structure cannot be restored. A vocative contributes to the implied lineation in two ways. First, a vocative between the lines sets the second line off phonologically. And second, the vocative implies that the second line should be set off syntactically from what precedes. The implied division into two lines is thus assisted by the presence of the vocative between them. Much less frequently in couplets involving ellipsis, the vocative does not occur between the lines, as in (28):
50
For additional explanation and examples of backwards ellipsis in Biblical Hebrew, see Cynthia L. Miller, ‘Constraints on Ellipsis in Biblical Hebrew’, in Cynthia L. Miller (ed.), Studies in Comparative Semitic and Afroasiatic Linguistics Presented to Gene B. Gragg (Studies in Ancient Oriental Civilization 60, Chicago 2007), 165–80; and ‘The Relation of Coordination to Verb Gapping in Biblical Poetry’, Journal for the Study of the Old Testament 32 (2007), 41–60. 51 For an analysis of this verse as elliptical and its relationship to Ps. 40:14, see Cynthia L. Miller, ‘A Linguistic Approach to Ellipsis in Biblical Poetry: Or, What to Do When Exegesis of What is There Depends on What Isn’t’, Bulletin of Biblical Research 13 (2003), 251–70. 362
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
(28) Ps. 146:10 ימלך יהוה לעולם אלהיך ציון לדר ודר The LORD will reign forever; Your God, O Zion, for ever and ever.
We can schematise the order of constituents of the two lines as follows: Verb Subject
Subject Vocative
Prepositional phrase Prepositional phrase
The surface order of constituents, however, raises the following problem: What was the original location of the elided verb? Did the second line originally have the same order as the first line (i.e. Verb Subject Prepositional phrase)? Verb [Verb]
Subject Subject
Prepositional phrase Prepositional phrase
Or, was the order of the verb and subject reversed in the second line to form a partially-chiastic structure (i.e. Subject Verb Prepositional phrase)?52 Verb Subject
Subject [Verb]
Prepositional phrase Prepositional phrase
While it is impossible to be absolutely certain, a plausible argument can be made that the proper noun ציוןoccurs in the place where the verb would be. This location relates to the cross-linguistic tendency for a small pause to occur in the location where an item is deleted. Since vocatives are surrounded by a small pause when they are inserted into a sentence, it makes sense that a vocative would find a niche in place of the deleted item. Again, the insertion of the vocative in the gap left by ellipsis assists the hearer/reader in accurately processing the sentence fragment. Conclusions In conclusion, we have seen that the status of the vocative as an extrasentential unit bears directly on its external syntax. The vocative occurs most frequently at clause/sentence boundaries. Otherwise, the 52 The notion of partially chiastic lines is debatable. For an analysis that accepts partially chiastic lines, see M. O’Connor, Hebrew Verse Structure (Winona Lake 1980), 394–5.
363
VOCATIVE SYNTAX IN BIBLICAL HEBREW PROSE AND POETRY
vocative always occurs at phrasal boundaries. In other words, the linguistic pieces left by the insertion of a vocative are never less than a phrase — the vocative never divides a construct phrase, a noun and its adjective, or a preposition and its complement. In prose, the vocative usually appears in clause-initial or clause-final position. Clause-medial position is extremely limited in occurrence. In poetry, however, the vocative may appear between any phrasal constituents, seemingly without any restrictions, and constructions that were deemed unusual in prose are quite common in poetry. The only dramatic syntactic difference in poetry is that the vocative can dissect a clausal constituent, so long as both of the remaining pieces are themselves phrases. Poets, however, apparently used the vocative for an additional literary function related to poetic lineation. Since the vocative most commonly occurred at a clause boundary, the poets used the vocative as an additional cue for the delimitation of poetic lines, especially in cases involving ellipsis. In utterances involving ellipsis, it is important for the hearer/reader to know the end of the first clause and the beginning of the second, in order to process the sentence fragment. This is especially true in backwards ellipsis in which the initial line contains the sentence fragment and the following line has the complete sentence from which the fragment must be restored. By placing a vocative between the lines, the poet provided his hearers with both a phonological and a syntactic clue to his intended lineation. Address for Correspondence: Dept of Hebrew and Semitic Studies, University of Wisconsin – Madison, 1344 Van Hise Hall, 1220 Linden Drive, Madison, WI 53706-1558, USA
364
Journal of Semitic Studies LV/2 Autumn 2010 doi: 10.1093/jss/fgq003 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA: A CASE OF CULTURAL HYBRIDITY WILLIAM MORROW QUEEN’S THEOLOGICAL COLLEGE
Abstract Derivations of lskn smw sm from Akkadian suma sakanu in the Deuteronomic centralization formula generally assume that Judah’s scribes were fluent readers of cuneiform. It is difficult, however, to construct a plausible path of transmission for the Akkadian phrase to Iron Age Judah. Moreover, Judah’s scribes may have had limited knowledge of Akkadian. This paper proposes, therefore, that the Deuteronomic centralization formula rests on meanings of ‘to set the name’ native to Biblical Hebrew. The alteration of the expected Hebrew expression to imitate an Assyrian parallel can be explained with reference to the phenomenon of ‘hybridity’ described in post-colonial studies.
A number of scholars suggest that the Biblical Hebrew construction lskn smw in the Deuteronomic (Dtn) centralization formula was derived from the Akkadian phrase suma sakanu.1 This article supports a connection of the Hebrew phrase to the Akkadian; however, its interpretation of this association rests on three assumptions: 1. the phrase ‘to set the name’ is a native West Semitic expression; 2. the relevant Akkadian parallels are found in the Neo-Assyrian period; 3. the Dtn writer(s) may have had a limited knowledge of Akkadian. 1 Roland de Vaux, ‘Le lieu que Yahvé a choisi pour y établir son nom’, in F. Maass (ed.), Das ferne und nahe Wort: Festschrift L. Rost (BZAW 105, Berlin 1967), 221; Dean S. McBride, ‘The Deuteronomic Name Theology’, unpublished Ph.D. dissertation (Harvard University 1969); Moshe Weinfeld, Deuteronomy and the Deuteronomic School (Oxford 1972), 193–4; Norbert Lohfink, ‘Zur deuteronomischen Zentralisationsformel’, Biblica 65 (1984), 327–8; G. Vanoni, ‘sym’, TDOT 14, 105; A.S. van der Woude, ‘sm’, TLOT 3, 1361; Eckart Otto, Das Deuteronomium: Politische Theologie und Rechtsreform in Juda und Assyrien (BZAW 284, Berlin 1999), 344–5; Sandra L. Richter, The Deuteronomistic History and the Name Theology: l¢sakken s¢mô sam in the Bible and the Ancient Near East (BZAW 318, Berlin 2002), 208–9; John Van Seters, ‘The Formula l¢sakken s¢mô sam and the Centralisation of Worship in Deuteronomy and DH’, JNSL 30/2 (2004), 6–8.
365
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
The presence of the phrase lskn sm in Deuteronomy can be explained with reference to the concept of ‘hybridity’ used in post-colonial studies. It was a neologism coined as part of a polemical rejection of Neo-Assyrian claims in favour of YHWH, the god of Israel. 1. The West Semitic Context The long form of the Deuteronomic (Dtn) centralization formula has two variants in the Masoretic Text. ‘The place that YHWH your God will choose’ is qualified by the infinitive clause lskn smw sm (Deut. 12:11; 14:23; 16:2, 6, 11; 26:2) and by the infinitive clause lswm smw sm (12:21; 14:24). The two phrases are conflated in 12:5, which contains both the infinitive clause lswm smw sm and lsknw (vocalized as a noun).2 The phrase lswm sm is a later, Deuteronomistic (Dtr) substitution for the original lskn sm.3 Nevertheless, both clauses have the same meaning, ‘to set his name there’.4 Their equivalent semantic value is suggested not only by their placement in similar contexts in Deuteronomy but also by their treatment in the ancient versions, especially the Septuagint and the targums.5 Evidence for the use of the Piel of skn in Biblical Hebrew outside passages in the Dtn centralization formula is slim. Besides texts which seem to be drawing on the Dtn centralization formula (Jer. 7:12; Ezra 6:12; Neh. 1:9), there are four occurrences of skn in the Piel (Num. 14:30; Jer. 7:3, 7; Ps. 78:60). The presence of the Piel in Jer. 7:3, 7 and Ps. 78:60 is dubious textually and there are arguments for construing the stem of skn as Qal in all three cases.6 As Num. 14:30 is the only clear instance of the root skn in the Piel outside of apparent Dtn influence, it serves to highlight the unusual usage of the verb in the centralization formula. If there is attestation for a factive meaning of skn in the Piel (i.e. ‘to make dwell, settle’), why has the Dtr tradition construed the formula as equivalent to the connotations of ‘set, put’? 2 lsknw may have originally been a Piel infinitive; see Baruch Halpern, ‘The Centralization Formula in Deuteronomy’, VT 31 (1981), 23–4; Richter, Name Theology, 46–8. 3 Lohfink, ‘Zentralisationformel’, 298; Richter, Name Theology, 207. 4 Lohfink, ‘Zentralisationformel’, 327–8; Richard D. Nelson, Deuteronomy: A Commentary (OTL; Louisville 2002), 152; Yair Zakovitch, ‘To Cause His Name to Dwell There—To Put His Name There [Hebrew]’, Tarbiz 41 (1972), 338–40. 5 Léo Laberge, ‘Le lieu que YHWH a choisi pour y mettre son nom’, EstBib 43 (1985), 212–13. The Samaritan Pentateuch has collapsed the verbs in 12:21 and 14:24 to the dominant form with skn. 6 A.R. Hulst, ‘skn’, TLOT 3, 1327; Laberge, ‘Le lieu’, 223–4.
366
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
Extensive philological work on the Dtn use of skn appears in Sandra Richter’s 2002 study. Richter demonstrates convincingly that the root skn does not possess the meanings of ‘to place’ or ‘to set’ in West Semitic languages including Phoenician, Aramaic, and Biblical Hebrew. Her solution is that the unusual meaning of the formula lskn sm in Deuteronomy is best explained on the assumption of a ‘loanhypothesis’ from Akkadian. Methodologically, such a hypothesis is defensible only if a number of conditions can be met. These entail establishing a history of language contact, a distinctive geographic and lexical distribution of meanings, and the ad hoc nature of the semantic usage in question. All of these conditions apply in the case of relating the Dtn formula lskn smw to suma sakanu .7 It is clear that there was a history of language contact such that a number of Akkadian expressions were appropriated by languages such as Biblical Hebrew and Aramaic.8 Secondly, the meanings ‘place, put’ belong to the root skn in Akkadian, but not in West Semitic languages. Finally, there is evidence that a translation of Akkadian sakanu into a West Semitic tongue would use the root swm/sym. This is indicated by the Aramaic bilingual inscription from Tell Fakhariyeh:9 Akkadian ll. 15–17
Aramaic ll. 10–12
mannu arkû anÌussu luddis sumima liskun Whoever in the future will restore its ruin, may he (re)place my name. mannu sa sumi unakkaru u sumsu isakkanu… Whoever removes my name and puts his name… (curses) wmn }Ìr kn ybl lknnh Ìds wsmym lsm bh Whoever afterwards takes (it) to erect it anew, may he put my name on it. wzy yld smy mnh wysym smh… And whoever effaces my name from it and puts his name… (curses)
These observations underscore the ad hoc nature of the Dtn formula lskn sm and make a strong case for the claim that Dtr writers perceived the Dtn phrase as a foreign usage that required commentary. 7
Richter, Name Theology, 98–121. Paul V. Mankowski, Akkadian Loanwords in Biblical Hebrew (HSS 47, Winona Lake 2000), 175–6. 9 Richter, Name Theology, 202–3. The Akkadian transcription follows Jonas C. Greenfield and Aaron Shaffer, ‘Notes on the Akkadian-Aramaic Bilingual Statue from Tell Fekherye’, Iraq 45 (1983), 109–16. The Aramaic transcription follows Stephen A. Kaufman, ‘Reflections on the Assyrian-Aramaic Bilingual from Tell Fakhariyeh’, Maarav 3/2 (1982), 137–75. 8
367
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
One qualification of Richter’s philological work is necessary, however. According to Richter, while lskn sm in the Dtn centralization formula is a loan adaptation of suma sakanu, Dtr lswm sm is to be explained as a calque of the Akkadian phrase.10 But the fact that the expression swm/sym/st sm is independently attested in three West Semitic languages does not suggest derivation from Akkadian. Richter’s belief that all instances of ‘to set the name’ in West Semitic languages represent a dependency on Mesopotamian usage gives too much power to Akkadian influence in West Asia in the Late Bronze (LB) and early Iron Ages (see below in §2). The phrase has the following distribution in extant Old Aramaic, Biblical Hebrew (excluding expressions related to the Dtr centralization formula),11 and Phoenician texts: 1.1 Setting a name on an object
KAI 10:13; 26: A iii.14, 16; C iv.16, 18; Tell Fakhariyeh, 10–12
1.2 Acquiring fame or power cf. {sh sm
2 Sam. 7:23 (= 1 Chron. 17:21) e.g. Gen. 11:4; 2 Sam. 7:9; Isa. 63:12
1.3 Giving someone/thing a name KAI 26: A ii.9–10, 17–18; Judg. 8:31; 2 Kgs 17:34; Dan. 1:7; Neh. 9:7 cf. qr} sm e.g. Gen. 2:20; 5:2; Josh. 5:9; 1 Sam. 1:20 1.4 Acquiring progeny
KAI 226:3 (cf. ll. 4–5); 2 Sam. 14:7
1.5 Bestowing a blessing
Num. 6:27
2. The Neo-Assyrian Context Suggestions of Mesopotamian influence on biblical literature typically invoke claims involving one of three eras. First, there is the well-known phenomenon of Akkadian literacy in LB Canaan.12 Second, there is the period of Neo-Assyrian hegemony over Israel and Judah in the late monarchical era. Third, there is contact with Neo-Babylonian 10 Richter, Name Theology, 201–3. A difficulty with Richter’s position is that she does not survey all the cases of lswm sm in Biblical Hebrew in order to substantiate her claim. Lohfink (‘Zentralisationsformel’, 327–8) also derives both forms of the purpose clause in the Dtn centralization formula from Akkadian. 11 The biblical Hebrew evidence is presented in Vanoni, ‘sym’, 105. Texts outside Deuteronomy showing reflexes of the Dtr centralization formula in Deut. 12:5, 21; 14:24 include 1 Kgs 9:3; 11:36; 14:21; 2 Kgs 21:4, 7; 2 Chron. 6:20; 12:13; 33:7. 12 See, e.g. David M. Carr, Writing on the Tablet of the Heart: Origins of Scripture and Literature (New York 2005), 60–1.
368
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
culture during the exile.13 There is no need to evoke the exilic period, however, since the book of Deuteronomy contains evidence of NeoAssyrian influence, especially in treaty rhetoric.14 The hypothesis of Bronze Age influence deserves more discussion than space permits. However, in a recent article I dispute the possibility that there was a line of transmission (either in writing or orally) for Akkadian literature from LB Canaan to Iron Age Judah.15 My conclusions are based on what is known about scribal education in LB Canaan and the contents of the archives of Ugarit. Clearly, West Semitic-speaking scribes had the opportunity to encounter Akkadian texts at that time. But the archives from Ugarit do not support the assumption that Canaanite scribes appropriated Mesopotamian literature into their native traditions.16 There is no reason to imagine the situation was different in the less educated scribal circles of southern Canaan. Of course, it is possible that technical terms in Akkadian were appropriated by Canaanite scribes even if Mesopotamian literary traditions were not.17 It is unlikely, however, that the presence of lskn sm in Deuteronomy can be explained through cultural defusion of a technical term (e.g. via Aramaic or Phoenician). In that case, one might have expected a wider attestation of the phrase in Biblical Hebrew. Moreover, if lskn sm had been adopted into Hebrew as a technical term, there would be no need for the Dtr gloss. I assume that the origins of Deuteronomy as a written document were in the southern kingdom of Judah during the period of NeoAssyrian hegemony.18 The Neo-Assyrian empire made significant 13
See, e.g. Ronald Hendel, ‘Genesis 11 and Its Mesopotamian Problem’, in E.S. Gruen (ed.), Cultural Borrowings and Ethnic Appropriations (Oriens et Occidens 8, Stuttgart 2005), 26. 14 See, e.g. Udo Rüterswörden, ‘Dtn 13 in der neueren Deuteronomiumforschung’, in A. Lemaire (ed.), Congress Volume: Basel 2001 (VTSup 92, Leiden 2003), 190–4. 15 William Morrow, ‘Resistance and Hybridity in Late Bronze Age Canaan’, RB 115 (2008), 321–39. 16 Jean Nougayrol, ‘L’Influence babylonienne à Ugarit, d’après les textes en cuneiformes classiques’, Syria 39 (1962), 28–30; Florence Malbran-Labat, ‘Akkadien, bilingues et bilinguisme en Élam et à Ougarit’, in J. Teixidor (ed.), Mosaïque de langues, mosaïque culturelle: Le Bilinguisme dans le Proche-Orient ancien (Antiquités sémitiques 1, Paris 1996), 57; Seth L. Sanders, ‘What was the Alphabet For? The Rise of Written Vernaculars and the Making of Israelite National Literature’, Maarav 11 (2004), 33–46. 17 See, e.g. Avi Shveka, ‘A Trace of the Tradition of Diplomatic Correspondence in Royal Psalms’, JSS 50 (2005), 297–320. 18 Moshe Weinfeld, Deuteronomy 1–11 (AB 5, New York 1991), 17; Nelson, Deuteronomy, 6–8. 369
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
incursions into West Asia and the Levant beginning with the reign of Shalmaneser III in the ninth century BCE. The northern kingdom of Israel was completely absorbed into the Assyrian province system by 720. The kingdom of Judah became a vassal state around 734.19 After the disastrous revolt of Hezekiah was suppressed in 701, Judah remained under the domination of Assyria until the late seventh century.20 It follows, therefore, that Akkadian texts and literature most relevant for comparison with the formula lskn sm were written between the ninth and the seventh centuries.21 The following catalogue of uses of suma sakanu is based on a survey of various Akkadian sources from the ninth to the seventh centuries, including RIMA 2–3, RIMB 2, SAA 1–18, inscriptions of Tiglathpileser III,22 Sargon II,23 Sennacherib,24 Esarhaddon,25 Assurbanipal,26 and some miscellaneous texts.27 Since this evidence is more widely distributed (and somewhat less accessible) than the comparable West Semitic expressions, it is set out here in full. Only cases where the meaning of suma sakanu can be recovered from the context are
19
Baruch A. Levine, ‘Assyrian Ideology and Israelite Monotheism’, Iraq 67 (2005), 417. 20 c. 630 according to Mordechai Cogan and Hayim Tadmor, 2 Kings (AB 11, New York 1988), 293; but later according to Nadav Naaman, ‘The Kingdom of Judah under Josiah’, Tel Aviv 18 (1991), 40–1. 21 Van Seters, ‘The Formula’, 6–7. 22 Louis Levine, Two Neo-Assyrian Stelae from Iran (Royal Ontario Museum Art and Archaeology Occasional Papers 23, Toronto 1972), 11–21; Hayim Tadmor, The Inscriptions of Tiglath-pileser III, King of Assyria: Critical Edition with Introduction, Translations, and Commentary (Jerusalem 1994). 23 Hugo Winckler, Die Keilschrifttexte Sargons: Nach den Papierabklatschen und Originalen 1 (Leipzig 1889); Hayim Tadmor, ‘Fragments of a Stele of Sargon, King of Assyria, from the Excavations of Ashdod [Hebrew]’, EI 8 (1967), 241–5; Levine, Two Neo-Assyrian Stelae from Iran, 25–45; Andreas Fuchs, Die Inschriften Sargons II. aus Khorsabad (Göttingen 1994). 24 Daniel David Luckenbill, The Annals of Sennacherib (University of Chicago Oriental Institute Publications 2, Chicago 1924). 25 Rykle Borger, Die Inschriften Asarhaddons Königs von Assyrien (AfO Beiheft 9, Berlin 1956). 26 Rykle Borger, Beiträge zum Inschriftenwerk Assurbanipals: Die Prismklassen A, B, C=K, D, E, F, G, H, J, und T sowie andere Inschriften. Mit einem Beitrag von Andreas Fuchs (Wiesbaden 1996). 27 Leroy Waterman, Royal Correspondence of the Assyrian Empire: Translated into English, with a Translation of the Text and a Commentary 1–4 (University of Michigan Humanistic Series 17–20, Ann Arbor 1930–1); Wayne Horowitz, Takayoshi Oshima, and Seth Sanders, Cuneiform in Canaan: Cuneiform Sources from the Land of Israel in Ancient Times (Jerusalem 2006). 370
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
listed.28 While I expect that future researchers will add to this list from new or additional sources, these texts seem representative of meanings for suma sakanu during the Neo-Assyrian era (cf. the entry for sakanu in CAD). There are correlations between the uses catalogued below and meanings of the comparable West Semitic expression in terms of putting names on objects, securing fame, giving names to persons and things, and acquiring descendents. 2.1 Setting a name on an object i) Tell Fakhariyeh, 15–17 (cited above) ii) RIMA 3.A.0.104.2012, 9’–11’ narâya al†ur u sumi ana daris altakan I wrote my monumental inscription and established my name for eternity.
The author was likely Samsi-ilu, a major Assyrian official during the time of Adad-nerari III (810–783).29 iii) sumiya misiÌti dursu askunma I made my name the measure of its enclosure and…
Sargon II used the numerical value of his name to determine the measurements for the enclosure of his fortress/palace at Dur-Sharrukin (Khorsabad).30 iv) RIMB 2.S.O.1002.9, iii.12–13 suma attua i[tti s]umi attusu [as]kun [I pla]ced my name w[ith] his [na]me
A stele of Ninurta-kudurri-uÒur (eighth century ruler of SuÌu) reflects the use of suma sakanu in foundation deposits.31 v)
Annals of Sennacherib vi.72–3 musarê si†ir sumiya ina qirbisa askun In it I set up a stela with my name inscribed.
28 Excluded are texts in damaged contexts such as Nimrud letter 45, ll. 9-10 (see H.W.F. Saggs, ‘The Nimrud Letters 1952—Part IV’, Iraq 20 [1958], 196–7) and letter #918: ll. 13–14, published in Waterman, Royal Correspondence, 2, 138–9 (both discussed in Richter, Name Theology, 177–8). Other broken contexts likely containing the expression ‘to set down the name’ occur in Assurbanipal’s Prism E, 19–24; SAA 10.349, r. 26–27e; and SAA 17.21, r.15–16. 29 RIMA 3, 235. 30 The formula occurs on the clay cylinder inscription §65, 3–5 (Fuchs, Inschriften Sargons II, 42) and on the bronze tablet inscription §§47–8 (ibid., 47). Both texts were hidden from public view. 31 Discussed in Richter, Name Theology, 173.
371
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
Setting down (sakanu) a ‘written name’ (si†ir sumi) is a close variant of the expression suma sakanu.32 Sennacherib records that he set up this stela after he refurbished a palace in Nineveh.33 vi) RIMB 2 B.6.31.11, 19 musarû si†ir sumiya samnu lipsus niqu liqqi ina asrisu liskun May he anoint my royal inscription with oil, offer a sacrifice, (and) set (it) back in its place.
Wishes such as this of Esarhaddon, to (re)place a former king’s inscribed name on a building or foundation inscription, are common in this time period.34 2.2 Acquiring fame or power i)
RIMA 3.A.0.102.2, i.4 sumi kabtu zikri Òira eli napÌar bele madis iskunu}inni (the gods who) have richly established for me my honourable name (and) my lofty command over all lords
This description comes from the prologue of the inscriptions of Shalmaneser III (see also RIMA 3.A.0.102. 6, i.10). ii) sume rabû ina puÌri sa mat Akkadî luskunka a great name in the assembly of the land of Akkad I shall certainly establish for you.
The letter is from an Assyrian king, probably Esarhaddon, to a highly placed person in Babylon. Here the idea of ‘establishing the name’ is joined with the idea of reward.35 iii) SAA 10.74, e.22–3 [su]mu u isdiÌu issakkansu [fa]me and prosperity will (then) be granted to him
On a given date when the crown-prince prostrates himself before Marduk, the god will give him a favourable omen. 32
Ibid., 184–8. Luckenbill, Annals of Sennacherib, 130. 34 From the time of Esarhaddon, the formula is found in RIMB 6.31.12, 37; 15, 38; 16, 21; 17, 21 and in the Nineveh inscription §25, 69–70 (Borger, Inschriften Asarhaddons Königs, 64). Uses of similar formulas in inscriptions of Assurbanipal are found in RIMB 6.32.1, 25–6; 6, 28–29; 12, 24–5; 13, 25 and in Borger, Beiträge zum Inschriftenwerk Assurbanipals, B §61:84–7; C §86:109, 113; D §61:92; F §37:68; T §20: vi.17. 35 From the Akkadian facsimile in Robert Francis Harper, Assyrian and Babylonian Letters 4 (Chicago 1896) #539, ll. 23–4. 33
372
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
iv) SAA 13.181, r.5–6 anaku ina ramniya anamdinma sumiya asakkan I will give of my own accord and establish my name.
The passage quotes a Chaldean sheikh who has been intimidating the inhabitants of towns responsible for provisioning Marduk’s cult, while refusing to give an offering of dates. According to the writer, the sheikh shows no respect for the king and wants to establish his own reputation instead. 2.3 Giving a name SAA 12.1, 5 Qibit-Assur sumsu askun I named it Qibit-Assur
The tablet records a grant of lands by Adad-nerari III to Assur for the provision of offerings. The same formula is also found in ll. 10, 12, but the name of the place is not preserved. 2.4 Acquiring progeny One of the most common ways to establish a name was by having children.36 This is reflected in the use of suma sakanu in personal names, e.g.37 i) ii) iii) iv)
Adad-sumu-iskun Bel-sumu-iskun Nabû-sakin-sumâti Nabû-sumu-iskun
Adad has provided a name Bel has provided a name Nabu is the establisher of names Nabu has provided a name
Nabû-sumu-iskun is a particularly common name, borne by no less than 33 persons in the documents surveyed by the Neo-Assyrian Text Corpus project.38 One bearer of this name was a son of Mardukapla-iddina II (Merodach-Baladan), who was an ally of Hezekiah according to 2 Kgs 20:12–13. 2.5 Expressions of status/loyalty i)
SAA 17.27, 23 sum(u) sa sarri ina muÌÌisunu sak[nu] the king’s name is sett[led] upon them 36
F.R. Kraus, ‘Altmesopotamisches Lebensgefühl’, JNES 19 (1960), 130. These examples come from Karen Radner and Heather D. Baker (eds), The Prosopography of the Neo-Assyrian Empire 1–2 (Helsinki 1998, 1999, 2001). 38 Baker, Prosopography of the Neo-Assyrian Empire 2:2, 888–90. 37
373
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
This text guarantees the loyalty of a certain group of people. ii) SAA 18.158, 10 sumsu ana kidin sakin Its name is set up for protection
A statement of the privileged status of Babylon. 2.6 Setting down the written name in omen rituals sa sumsu ina ni}ari/nibzi annâ sa†ruma ina maÌar ilutika rabiti saknu whose name is written on this papyrus/tablet and placed before your great divinity
It was common practice in Neo-Assyrian liver divination procedures to write down the name of the person or city in question and set it before a symbol of the god, Samas. The majority of examples concern royal appointments, but the same ritual was also used for military purposes and medical cases.39 3. The Inter-Cultural Context The comparative method used here follows Jeffrey Tigay in assessing claims regarding literary borrowings in ancient Near Eastern literature. Two forms of evidence must combine. First, there must be a reasonable degree of probability that borrowing has taken place. Second, there must be a probable channel of transmission.40 The philological arguments set out in §1 suggest a high degree of probability that borrowing from Akkadian has taken place in the Dtn centralization formula. Yet to be addressed is the channel of transmission. How could knowledge of cuneiform usage have come to the writers of Deuteronomy? As §2 illustrates, the phrase suma sakanu was used in royal monuments and building inscriptions, diplomatic correspondence, personal names and in omen rituals. But not all of these genres are plausible vectors of transmission to the chancery of Iron 39 The formula occurs or is restored forty-seven times in SAA 4. For example: queries about official appointments are found in 290, r. 6–7; 306, 17–r.1; 307, r.1–3; and 310, r.1–3. Enquiries about cities as potential targets for attack are found in 105, 2; 106, r. 8–9. Subjects of medical enquiries are mentioned in 311, r.2-3 and 312, r.1–3. 40 Jeffrey H. Tigay, ‘On Evaluating Claims of Literary Borrowing’, in M.E. Cohen, D.C. Snell and D.B. Weisberg (eds), The Tablet and the Scroll: Near Eastern Studies in Honor of William W. Hallo (Bethesda 1993), 255.
374
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
Age Judah. Mesopotamian building inscriptions are an unlikley source because many of these (though not all) were buried in walls and foundation deposits, and their distribution does not seem to have penetrated into the western part of the Neo-Assyrian empire. The biblical record shows that Mesopotamian personal names were not usually translated. This leaves two categories for consideration: politically motivated communications and omen literature. In a previous discussion of Akkadian literarcy in Iron Age Judah, I concluded on different grounds that its scribes were motivated to become familiar with Akkadian religious texts and political rhetoric.41 This suggestion accords with an analysis of the text-types in §2 from the viewpoint of transmission to Judah. Remnants of Assyrian monumental inscriptions have been found in Israel and the Levant.42 Stelae were a common means of NeoAssyrian propaganda through the ninth and eighth centuries, though use tailed off in the seventh century (perhaps because imperial expansion had more or less reached its peak).43 Nevertheless, attestation of suma sakanu on Neo-Assyrian monuments is sparse. I have been able to find only one example after the ninth century, and that not set up under Assyrian aegis (§2.1, iv). It would be rash to entirely rule out this vector of transmission, but the available evidence is not strong. Of course, there could be a source of transmission from a political context that has been lost. Judah’s scribal elite was familiar with NeoAssyrian propaganda and it is probable that diplomats from Jerusalem visited Assyria where they would have been exposed to palace inscriptions,44 some of which (like the stela of Sennacherib, §2.1, v) could have contained the expression suma sakanu. While Assyrian religious influences on seventh-century Judah were informal, they were by no means insubstantial.45 Material evidence
41 William Morrow, ‘Cuneiform Literacy and Deuteronomic Composition’, BO 62 (2005), 210. 42 Horowitz, Oshima, and Sanders, Cuneiform in Canaan, 20. 43 Though dated, the appendix of Louis Levine (Two Neo-Assyrian Stelae from Iran, 51–7) is still representative. Out of twenty-nine royal inscriptions on stelae and related monuments, only four are as late as the time of Esarhaddon. 44 Peter Machinist, ‘Assyria and Its Image in the First Isaiah’, JAOS 103 (1983), 730–1. 45 Mordechai Cogan, Imperialism and Religion: Assyria, Judah and Israel in the Eighth and Seventh Centuries B.C.E. (SBLMS 19, Missoula 1974), 86–93; C.D. Evans, ‘Manasseh, King of Judah’, ABD 4, 498; J.M. Miller and J.H. Hayes, A History of Ancient Israel and Judah (Philadelphia 1986), 371–2.
375
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
includes a Lamashtu amulet from the vicinity of Lachish.46 It is possible, therefore, that transmission came through acquaintance with Akkadian religious techniques. But this is less probable than transmission through political rhetoric. The expression ‘to place the name’ was used in Akkadian rites of liver divination (cf. §2.6). Hypothetically, Judah’s diplomats and officials may have had the opportunity to observe practices of extispicy by Assyrian experts attached to the occupying administration. Diviners were attached to the army,47 and it is probable that an Assyrian garrison was stationed in Judah after Hezekiah’s abortive rebellion.48 Hermann Spieckermann has suggested that liver divination was practiced in the Jerusalem temple of Manasseh’s time.49 But there are no references to extispicy in the biblical accounts. It is noteworthy, however, that Manasseh is the first king said to have introduced the worship of the heavenly host into the Jerusalem temple.50 Consultation of the stars reached unprecedented importance in Neo-Assyrian religion during the seventh century.51 It is possible, therefore, that some form of astral divination inspired by Assyrian practices was conducted in the First Temple. There is no connection, however, between Assyrian astral divination and the Akkadian formula being discussed. The speculative element involved in the two most likely vectors for transmitting Akkadian knowledge to Judah’s intellectuals should not be overlooked. Neither Neo-Assyrian political rhetoric nor astrology shows frequent use of the expression suma sakanu. This is an important datum, because it undermines attempts to locate knowledge of the use of suma sakanu with a particular Akkadian text-type. It is easier to assume that Judah’s scribes were familiar with the formula suma sakanu through a general knowledge of Akkadian. But this assumption carries its own difficulties. Given the limits of the evi46 Mordechai Cogan, ‘A Lamashtu Plaque from the Judaean Shephelah’, IEJ 45 (1995), 155–61. 47 Walter Farber, ‘Witchcraft, Magic, and Divination in Ancient Mesopotamia’, CANE 3, 1904. 48 Moshe Elat, ‘The Political Status of the Kingdom of Judah within the Assyrian Empire in the 7th Century B.C.E.’, in Y. Aharoni (ed.), Investigations at Lachish: The Sanctuary and Residency (Tel Aviv 1975), 63–4. 49 Hermann Spieckermann, Juda unter Assur in der Sargonidenzeit (FRLANT 129, Göttingen 1982), 245, 256. 50 Mordechai Cogan, ‘Judah under Assyrian Hegemony: A Re-examination of Imperialism and Religion’, JBL 112 (1993), 410. 51 Spieckermann, Juda unter Assur, 256–7.
376
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
dence, it is not easy to reach firm conclusions on the degree of cuneiform literacy that scribes in Iron Age Judah may have possessed. There has been a tendency in recent studies on biblical law to assume a significant degree of Akkadian literacy among the scribes of seventh-century Judah. For example, Eckart Otto thinks they were familiar with the Vassal Treaties of Esarhaddon (VTE), the Esarhaddon succession oracles and the Middle Assyrian Laws.52 David Wright claims that Codex Hammurabi was a major source for the composition of the Covenant Code (Exod. 20:23–23:19) in Judah during the Neo-Assyrian era.53 Bernard Levinson finds evidence in the Aramaic treaties from Sefire (which show Neo-Assyrian influences) and the bilingual Tell Fakhariyeh inscription that there were trained scribes in Syro-Palestinian states able to read and work with cuneiform documents.54 While a full discussion of Akkadian fluency among Iron Age scribes in Jerusalem is beyond the scope of this paper, it is worth noting that the biblical scholarship just cited stands in tension with Assyriologists skeptical about Akkadian literacy in the Western periphery of the Neo-Assyrian empire.55 During the eighth and seventh centuries the Western part of the Neo-Assyrian empire increasingly relied on Aramaic instead of Akkadian as the administrative language of choice (cf. 2 Kgs 18:26).56 The preference for Aramaic indicates reduced 52
Otto, Deuteronomium, 68–9, 83–5, 300–2. For a critique of Otto’s position, see Morrow, ‘Cuneiform Literacy and Deuteronomic Composition’, 204–13. 53 David Wright, ‘The Laws of Hammurabi as a Source for the Covenant Code (Exodus 20:23–23:19)’, Maarav 10 (2003), 11–88. For a critique of Wright’s position, see Bruce Wells, ‘The Covenant Code and Near Eastern Legal Traditions: A Response to David P. Wright’, Maarav 13 (2006), 85–118. 54 Bernard M. Levinson, ‘Is the Covenant Code an Exilic Composition? A Response to John Van Seters’, in J. Day (ed.), In Search of Pre-exilic Israel (London 2004), 295–6. A difficulty with Levinson’s thesis, however, is that he interprets a passage from the Khorsabad cylinder inscription of Sargon II (published in Fuchs, Inschriften Sargon II, 43–4, §296, ll. 72–3) as indicating a programme for the spread of Akkadian literacy where the reference is probably to encouragement of the use of Aramaic in the NeoAssyrian empire; cf. William Schniedewind, ‘Aramaic, the Death of Written Hebrew, and Language Shift in the Persian Period’, in S.L. Sanders (ed.), Margins of Writing, Origins of Culture (Oriental Institute Seminars 2, Chicago 2006), 138–9. 55 See, e.g. Simo Parpola, ‘A Letter from Marduk-Apla-Usur of Anah to Rudamu/ Urtamis, King of Hamath’, in R.J. Riis and M.L. Buhl (eds), Hama: Fouilles et recherches de la Fondation Carlsberg 1931–1938 2:2. Les objets de la période dite Syro-Hittite [Âge du Fer] (Copenhagen 1990), 264; Machinist, ‘Assyria and Its Image’, 732–3; J.S. Cooper, ‘Cuneiform’, ABD 1, 1217. 56 Hayim Tadmor, ‘The Aramaization of Assyria: Aspects of Western Impact’, in H.J. Nissen et al. (eds) Mesopotamien und seine Nachbarn (Berlin 1982), 449–54; 377
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
opportunity for native scribes in Judah to encounter cuneiform documents and reduced incentive to read them for diplomatic purposes. The sparse material remains of cuneiform documents found in sites associated with ancient Judah and its environs do not suggest that Akkadian was a major language of communication during the era of Neo-Assyrian occupation.57 Even the Assyrian treaty tradition (as exemplified by the VTE) may have reached Judah through the medium of Aramaic.58 Without a clear indication of the degree of Akkadian knowledge on the part of Judah’s scribes, a conservative approach to the question makes it preferable to follow the prevailing opinion among Assyriologists and assume a limited degree of Akkadian fluency among Judah’s intellectuals. An explanation for the Assyrianism in the Dtn centralization formula does not necessarily require the hypothesis of extensive familiarity with Akkadian. Suggestive here is the pun found in Isa. 10:8, in which the Assyrian king boasts, ‘Are not my commanders (saray) all kings?’ There is a wordplay between Hebrew sar and Akkadian sarru(m).59 But, as anyone who has begun studying cuneiform will attest, one does not require a great deal of fluency to learn the word for ‘king’ in Akkadian. By the same token, it would not have taken much explanation or learning to discover the overlapping semantic fields of Akkadian sakanu and Hebrew sym. 4. The Ideological Context Since Judah’s scribes may have learned basic meanings of sakanu without knowing a great deal of Akkadian, I propose that previous discussion about the relationship of the Dtn phrase lskn smw sm to Akkadian texts be turned on its head. Rather than assume that Judah’s S. Parpola, ‘Assyrian Royal Inscriptions and Neo-Assyrian Letters’, in F.M. Fales (ed.), Assyrian Royal Inscriptions: New Horizons in Literary, Ideological, and Historical Analysis (Orientis Antiqui Collectio 17, Roma 1981), 123. 57 Horowitz, Oshima, and Sanders (Cuneiform in Canaan, 20) count fifteen texts (excluding cylinder seals). The majority of these were not found inside the borders of seventh-century Judah. 58 Manfred Krebernik, ‘M. Weinfelds Deuteronomiumskommentar aus assyriologischer Sicht’, in G. Braulik (ed.), Bundesdokument und Gesetz: Studien zum Deuteronomium (Herders Biblische Studien 4, Freiburg 1995), 35–6; Christoph Koch, ‘Neue Perspektiven auf das Verhältnis von altorientalischen Vertragsrecht und Bundestheologie’, in D. Prechel (ed.), Fest und Eid: Instrumente der Herrschaftsicherung im Alten Orient (Würzburg 2008), 89–107. 59 Machinist, ‘Assyria and Its Image’, 735–6. 378
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
scribes were well informed about the meaning of suma sakanu in its Mesopotamian contexts, the meaning of the Dtn phrase is best explained on the basis of the native expression lswm sm. The reason why the Dtn writer(s) substituted the infinitive lskn as an Assyrianism should be sought after the meaning of the underlying Hebrew expression has been clarified. The survey of swm/sym/st sm in §1 shows that the phrase has a range of connotations. Applied to the Dtn centralization formula, it may indicate that the place where YHWH chooses to set his name symbolizes the presence of the deity, the special status of the place, and/or a concern for preserving the divine reputation. Each of these meanings has a prominence in previous scholarship, as the following survey indicates. Name and presence were closely related in the ancient Near East.60 Previous scholarship has often discussed the notion of divine presence in connection with a Dtn ‘name theology’. Advocates of a name theology contend that the idea of setting the name in the Dtn centralization formula signifies a departure from a traditional vision of numinosity towards a more abstract conception of YHWH’s relationship with sacred space. Proposals of a Dtn name theology have come under criticism in recent years;61 but a resolution of this debate is not directly pertinent to the interests of this article. Important for the present discussion is to note that the dispute itself points to a widespread opinion that what is connoted by setting the name in Deuteronomy is related to the concept of divine presence. Recently, John Van Seters has suggested a variant on this interpretation by connecting the Dtn formula with the deposit of a lawbook bearing YHWH’s name in the temple (cf. Deut. 31:24–6).62 Van
60 Karen Radner, Die Macht des Namens: Altorientalische Strategien zur Selbsterhaltung (SANTAG Arbeiten und Untersuchungen zur Keilschriftkunde 8, Wiesbaden 2005), 17. 61 Ian Wilson, Out of the Midst of Fire: Divine Presence in Deuteronomy (SBLDS 151, Atlanta 1995), 213–17; Richter, Name Theology, 36–9. 62 Van Seters, ‘The Formula’, 9–12. Van der Woudes (‘sm’, 1361) also attempts to connect the Dtn centralization formula with divine instruction through an assocation with the revelatory formula in Exod. 20:24 (}zkyr }t smy). His use of Akkadian evidence to explain the meaning of lskn sm is unconvincing, however, because he appeals to the similar syntax of names like Nabû-suma-iskun and Sîn-suma-izkur to establish the case for a convergence between the phrases suma sakanu and suma zakaru in Akkadian. There are two problems with this thesis. First, it is unlikely that Akkadian names were the medium by which Judah’s scribes learned the meaning of sakanu. Second, the contexts in which the phrase suma sakanu appear in Neo-Assyrian
379
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
Seters’s theory runs into difficulty because of his explanation for the provenance of lskn smw sm, which he thinks was coined by analogy with Akkadian building inscriptions marking temple reforms.63 As noted above, there is little reason to think this usage penetrated as far west as Judah. Nevertheless, though somewhat speculative, Van Seters’s suggestion merits consideration. Akkadian nuances for setting the name include depositing a document with a written name (§2.6). In the West Semitic world, §1.1 shows that setting the name may entail writing a name on an inscription. According to Deut. 27:2–8, the text of the Torah is to be used for a monumental inscription written on altar stones.64 Deposit of a spr twrh in the Jerusalem temple as setting YHWH’s name in sacred space would be an extension of these analogies. Setting the name may also indicate a special status or relationship with the agent involved. A number of scholars explain the Dtn centralization formula on the assumption that putting a person’s name on a place can be tantamount to a claim of ownership (cf. 2 Sam. 12:28).65 This interpretation, however, is resisted in Karen Radner’s recent study of suma sakanu, which concludes there is no basis for supposing that the Akkadian expression connotes ownership.66 Even so, although the semantic range of the Hebrew and Akkadian expressions overlap significantly, they may not do so entirely. There is a peculiarity to the Hebrew usage which is not paralleled in Akkadian. As opposed to the Mesopotamian context, Dtn and Dtr employment of ‘to set the name’ is unique in its attribution of this action to a deity. Even in Mesopotamia, where the motivation for setting a name is primarily for self-preservation,67 the action has various nuances. For example, Sargon II used the numerical value of his name to determine the measurements of his fortress/palace at Dur-Sharrukin (§2.1, iii). As Radner notes, in Mesopotamia there is an inseparable unity between a name and a name bearer.68 This initiative not only served to preserve Sargon’s name for posterity, therefore, but also established an intimate inscriptions do not allow for the substitution of zakaru, because they refer specifically to the act of setting down the name, not pronouncing it. 63 Van Seters, ‘The Formula’, 16. 64 Hans Ulrich Steymans, Deuteronomium 28 und die adê zur Thronfolgereglung Asarhaddons: Segen und Fluch im Alten Orient und in Israel (OBO 145, Freiburg, 1995), 25. 65 De Vaux, ‘Le lieu que Yahvé a choisi’, 225; Laberge, ‘Le lieu’, 218–19; Nelsen, Deuteronomy, 153; Richter, Name Theology, 182–3. 66 Radner, Macht des Namens, 70, 177. 67 Ibid., 74. 68 Ibid., 68. 380
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
relationship between the king and his new capital city. A nuance of ownership or a claim of possession may well be implied when a god sets his name on a sacred place. There is also an analogy between setting a name on a public monument and gaining fame or a reputation for great deeds. Of course, connections between the centralization formula and the monumental tradition in Akkadian have been made previously;69 but the Dtn expression does not have to assume extensive knowledge of the Mesopotamian context. Native Hebrew usage also has this connotation. According to 2 Sam. 8:13, David constructed (¨sh) a sm to commemorate a victory over Aram. Another indication that Hebrew sm may connote a monument is found in the hendiadys yd wsm in Isa. 56:5.70 The use of ‘to set the name’ in 2 Sam. 7:23 implies that YHWH’s redemptive efforts on behalf of Israel enhanced his reputation. If the place the deity chooses has his name set upon it, it may be understood as a monument to his sovereign power and might.71 Multiple connotations of the phrase ‘the place where YHWH chooses to set his name’ are one of the reasons why there are various explanations for the significance of the name in the Dtn centralization formula. The survey above has shown that several inferences can be derived from native Hebrew usages. While some are more convincing than others, a certain ambiguity about the phrase’s signifying function needs to be acknowledged. In part, this is due to the range of meanings implicated in the expression ‘to set the name’ itself. However, the ‘place’ which contextualizes the phrase contributes to the lack of clarity. In the Dtr History, e.g., the centralization formula may refer to the temple (1 Kgs 8:16, 29; 9:3; 2 Kgs 23:27), the city of Jerusalem (1 Kgs 11:36; 14:21), or both (2 Kgs 21:4, 7).72 Despite these ambiguities, one may conclude that common to all of the interpretations surveyed is the inference that YHWH sets his name in the place he chooses in order to promote his divine presence and his claims to sovereignty. Granted that these notions could be expressed perfectly well by the native expression lswm smw, why did the primary layer of the Dtn centralization laws opt for the Assyrianism lskn smw? In my opinion, the appearance of the phrase lskn smw in Deuteronomy is connected with processes described in post-colonial 69 Nelson, Deuteronomy, 153; Sandra L. Richter, ‘The Place of the Name in Deuteronomy’, VT 57 (2007), 343–4. 70 Weinfeld, Deuteronomy and the Deuteronomic School, 193. 71 Richter, Name Theology, 209. 72 Ibid., 51.
381
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
studies. These concern the phenomena of ‘hybridity’ or ‘colonial mimicry’, which involve an ambivalent process of ‘cleaving’ away from colonial definitions and, at the same time, a ‘cleaving to’ or appropriating textual forms from the dominant culture. In the very act of mimicking the dominating culture’s linguistic forms, there is an effort to make an ideological expression that serves the interests of the colonized, not the colonizer.73 Post-colonial theory, therefore, helps to explain the motivation for Assyrianizing the idiom for ‘placing the name’ in the Dtn centralization formula. It arose from awareness of the power of the dominant culture and the challenge that its imperial pretentions posed to Judah as the kingdom and people of YHWH. The expression lskn smw simultaneously acknowledges the reality of NeoAssyrian hegemony while also subverting it. Loyalty to the Neo-Assyrian empire required the payment of imposts and taxes, and the performance of services for the crown.74 It is no coincidence that the Dtn centralization formula using lskn sm occurs precisely in passages that address the obligations of Israel to pay its cultic dues to YHWH (12:11), including tithes (14:23), the Passover sacrifice (16:2, 6), grain offerings (16:11) and first fruits (26:2). In a period of Neo-Assyrian hegemony, Deuteronomy is subtly reminding its readers that YHWH remains the true king of Judah.75 The place where the deity has chosen to set his name is a sign of YHWH’s continuing dominion over Israel. It is not to be assumed that most people would have understood the subversive associations of the phrase lskn smw. In the pre-exilic period, literature like Deuteronomy was the province of a small, educated elite.76 Only a rather select group would have appreciated the bilingual pun implicated by minting the neologism lskn smw. But the insertion of such abstruse knowledge is hardly exceptional in ancient Near Eastern scribal practice. There are many examples in Mesopotamian literature of obscure references that would only make sense to the especially learned.77 73 Bill Ashcroft, Gareth Griffiths, and Helen Tiffin, The Empire Writes Back: Theory and Practice in Post-Colonial Literatures2 (London 2002), 205–6. 74 Seymour Gittin, ‘Israelite and Philistine Cult and the Archaeological Record in Iron II: The “Smoking Gun” Phenomenon’, in W.G. Dever and S. Gittin (eds), Symbiosis, Symbolism, and the Power of the Past: Canaan, Ancient Israel, and their Neighbours from the Late Bronze Age through Roman Palaestina (Winona Lake 2003), 288. 75 Otto, Deuteronomium, 74. 76 Carr, Writing on the Tablet of the Heart, 166. 77 See, e.g. the commentary on ludlul bel nemeqi in BWL, 26. The same phenomena also appear in later Jewish literature, see ‘Piyyut’, EJ 13: 601.
382
‘TO SET THE NAME’ IN THE DEUTERONOMIC CENTRALIZATION FORMULA
In summary, the meaning of the Dtn centralization formula can be determined from the native Hebrew connotations of ‘to set the name’. This expression is well attested in West Semitic languages including Aramaic, Hebrew, and Phoenician. Its application to the centralization formula entails claims of divine presence and sovereignty. But an early layer of Deuteronomy chose to render the idea of setting the name by using a Piel form of skn in imitation of Assyrian usage. While debate continues about the degree of cuneiform literacy in Iron Age Judah, Jerusalem scribes could have learned about the equivilance of Assyrian sakanu and Hebrew swm/sym on the basis of relatively little acquaintance with Akkadian. The substitution of the expected Hebrew verb for an unusual Piel form of skn is related to processes of hybridity described in contemporary post-colonial studies. The phrase lskn sm was coined for subversive reasons connected to the realities of NeoAssyrian hegemony and the need to affirm the autonomy and prestige of YHWH. Dtr revision of Deuteronomy abandoned the vocabulary of lskn sm and returned to the Biblical Hebrew expression lswm sm. This transformation suggests that the Neo-Assyrian reference was either no longer understood or not considered important by later Dtr writers.78 Address for Correspondence:
[email protected] Queen’s School of Religion, Kingston ON, Canada K7L 3N6
78
So also Richter, Name Theology, 210–11. 383
Journal of Semitic Studies LV/2 Autumn 2010 doi: 10.1093/jss/fgq004 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
A REVISED READING OF A NABATAEAN INSCRIPTION FROM UMM AL-JIMAL AARON MICHAEL BUTTS AND HUMPHREY HILL HARDY II UNIVERSITY OF CHICAGO
Abstract A new reading and interpretation are proposed for line two of a Nabataean inscription from Umm al-Jimal that was recently published in this journal by S. Said and M. al-Hamad (JSS 48 [2003], 29–34).
In a recent issue of this journal, S. Said and M. al-Hamad presented a new Nabataean inscription from Umm al-Jimal in the southern Îawran.1 In their edition, Said and al-Hamad noted several outstanding difficulties with the analysis of this inscription, but they published it ‘in the hope that new solutions may be found by other scholars’ .2 It is in response to that invitation that we present this new reading and interpretation. Said and al-Hamad transcribe and translate the inscription as follows: Transcription 1 [d]} nps} dy bnh shymw 2 [br] ¨lt bn} lh }bwhm 3 [..l]rb}l mlk} mlk nb†[w]…
Translation This is the tomb which SHYMW son of ‘LT built. He built (it) for himself ’BWHM … of king Rabel, king of the Nabataeans … We would like to thank D. Pardee and L. Van Rompay for their helpful comments on this article. All errors are, of course, ours alone. 1 Said and al-Hamad 2003. For an overview of Nabataean inscriptions from the Îawran, see Starcky 1985. Other Nabataean inscriptions from Umm al-Jimal can be found in CIS ii 190–93; Littmann 1914: nos. 38–68; Knauf 1985; Said and al-Hamad 1994; Khairy 2000; al-Muheisen and Ajlouni 2005; Graf and Said 2006. 2 Said and al-Hamad 2003: 29, see also 34. 385
A REVISED READING OF A NABATAEAN INSCRIPTION FROM UMM AL-JIMAL
The principal difficulties with this interpretation are found in line 2. Said and al-Hamad analyse bn} as a 3ms suffix-conjugation of the third weak verb √BNY ‘to build’.3 While the orthography with final <}> is occasionally attested in Nabataean Aramaic,4 the expected form is bnh as occurs in line 1 of this inscription.5 In addition, though lh is occasionally attested, lnpsh is the usual way of expressing the reflexive ‘for himself’ in Nabataean (as elsewhere in Aramaic).6 Finally and most importantly, the syntactic analysis of }bwhm remains unresolved in the interpretation of Said and al-Hamad.7 These morphosyntactic problems result from an unsatisfactory reading of line 2 of the inscription. This is most clearly demonstrated by the fact that Said and al-Hamad do not take into account a vertical stroke between what they read as
in line 2 and the top of an extended from line 3. Based on the published photograph (reprinted here as Fig. A),8 we propose the following reading for line 2: -¨brhb}lh}bwhm. The traces at the beginning of this line could possibly be read as , but this is tentative. The first fully visible letter is <¨>, which is followed by . Then, there is , which is composed of two strokes — a full length, slightly angled vertical line and a short angular stroke connected to the top of this line. The fourth character corresponds to one of the two types of found in this inscription.9 3
Said and al-Hamad 2003: 33–4. See, e.g., CIS ii 333.2 5 See Cantineau 1930–2, I: 83. Said and al-Hamad (2003: 32) attempt to explain the contrasting orthographies of bnh (l. 1) and bn} (l. 2) by parsing bnh as a suffix-conjugation verb plus a 3fs pronominal object suffix. This analysis of bnh is, however, unlikely. According to Cantineau (1930–2, I: 56), a pronominal object suffix is rarely attached to a verb, being more commonly found with a nota objecti, such as yt. It should also be noted that a resumptive pronoun is not necessary when it would be the accusative in a relative clause, as is demonstrated by locutions such as dnh kpr dy ¨bd PN … ‘this is the tomb which PN made …’ (Healey 1993: 3.1). 6 See Cantineau 1930–2, I: 57. In the inscriptions from Mada}in ∑aliÌ that were (re-)edited in Healey 1993, for instance, the reflexive is expressed by lnps- plus a pronominal suffix a total of twenty-two times (1.2; 3.1; 4.2; 7.1; 8.2; 9.1; 10.1; 11.2; 14.2; 16.2; 19.1; 25.2; 26.1; 27.2; 28.2; 29.3; 30.2; 32.2; 34.1; 35.2; 36.1; 38.2) as compared to only four examples of l- plus a pronominal suffix (5.2; 12.2; 20.2; 22.2). 7 Said and al-Hamad (2003: 33–4) admit as much. 8 The handcopy of Said and al-Hamad (2003: 31) is representative of their interpretation of the inscription, but it appears not to reflect accurately what is visible in the photograph published conveniently on the same page. It should be noted that our reading is based solely on the photograph and so should (ideally) be verified epigraphically at the source. 9 Compare the two other uses of this letter in shymw (l. 1) and }bwhm (l. 2). This is the ‘second type’ of in the terminology of Said and al-Hamad (2003: 32). 4
386
A REVISED READING OF A NABATAEAN INSCRIPTION FROM UMM AL-JIMAL
Fig. A: Nabataean Inscription from Umm el-Jimal (Said and al-Hamad 2003: 31)
This correction, thus, eliminates the possibility of reading the small stroke above the right side of as , and what remains is the extended left side of connecting with and <}> successively. With this new reading of line 2, the inscription becomes more intelligible. Revised Transcription 1 [ ]} nps} dy bnh shymw [ ] 2 [ ]¨ brh b}lh }bwhm [ ] 3 [ ]rb}l mlk} mlk nb†[ ]
Revised Translation (1) [Th]is is the tomb which SHYMW … built … (2) … [for P]N, his son, through (the help of) the god of their father … (3) … king Rabel, king of the Nabataeans …
According to this interpretation, the initial letter of line 2 is the end of the name of the person for whom the tomb was built.10 This is fol10
Several names ending in <¨ >, such as myd¨ and mn¨, are attested in Nabataean inscriptions from the Îawran. For Nabataean prosopography in general, see Negev 1991 with important corrections in Macdonald 1999. Names ending in <¨ > are also not infrequent in Palmyrene Aramaic (see Stark 1971). 387
A REVISED READING OF A NABATAEAN INSCRIPTION FROM UMM AL-JIMAL
lowed by the appellation brh ‘his son’ in apposition; the referent of the suffix is SHYMW. The preposition b- is, then, to be understood as ‘through the help of’ or simply ‘by’.11 The final two words of line 2 consist of a construct phrase ‘the god of their father’ according to which ‘father’ refers to the paternal progenitor of SHYMW and his son.12 Because it is uncertain how far the left side of the inscription originally extended, one can only speculate as to the exact composition of the inscription.13 At the very least, the present interpretation requires space for at the beginning of line 1, for the preposition l- plus the beginning of a personal name between the end of line 1 and the beginning of line 2, and for a minimal dating formula composed of (b)snt plus a numeral between the end of line 2 and the beginning of line 3.14 This could, however, be expanded with various formulae if the left side of the inscription was originally larger.15 Thus, while the exact composition of this inscription will in all likelihood remain conjectural, it is our hope that the new reading and interpretation [ ]¨ brh b}lh }bwhm [ ] ‘…PN, his son, through (the help of) the god of their father…’ provides a plausible solution to the previous morphosyntactic difficulties. Address for Correspondence: [email protected], [email protected] Department of Near Eastern Languages and Civilizations, University of Chicago, 1155 East 58th Street, Chicago, IL 60637, USA REFERENCES Brock, S.P. 1992. ‘Some Notes on Dating Formulae in Middle Aramaic Inscriptions and in Early Syriac Manuscripts’, in Z.J. Kapera (ed.), Intertestamental Essays in Honour of Józef Tadeusz Milik (Kraków). 253–64 11
Cantineau (1930–2, II: 69) gives ‘par, grâce à’ as a gloss. To our knowledge, this formula is not otherwise attested in Nabataean; nevertheless, a similar expression does occur in Palmyrene Aramaic: … wqrbw … lsms }lh byt }bwhn ‘… and they dedicated … to Shamash, the god of their father’s house’ (Hillers and Cussini 1996: C3978). 13 While there is a noticeable space at the end of line two, this space is not so great, in our opinion, as to rule out the possibility that additional text could be missing from the left side of the inscription (contra Said and al-Hamad 2003: 30). 14 According to Brock’s study of dating formulae in Middle Aramaic, snt plus a numeral is the shortest dating formula attested in Nabataean (Brock 1992). 15 A hypothetical expanded reading could look something like the following: [d]} nps} dy bnh shymw [br PN lnpsh wl-PN]¨ brh b}lh }bwhm [DN byrÌ X bsnt Y l]rb}l mlk} mlk nb†[w dy }Ìyy wsysb ¨mh] ‘[Th]is is the tomb which SHYMW, [son of PN,] built [for himself and for P]N, his son, through (the help of) the god of their father [DN. In the month of X in year Y of] king Rabel, king of the Nabatae[ans, who provides life and salvation for his people]. 12
388
A REVISED READING OF A NABATAEAN INSCRIPTION FROM UMM AL-JIMAL
Cantineau, J. 1930–2. Le Nabatéen. 2 vols, (Paris) Graf, D.F. and S. Said. 2006. ‘New Nabataean Funerary Inscriptions from Umm al-Jimal’, JSS 51, 267–303 Healey, J.F. 1993. The Nabataean Tomb Inscriptions of Mada}in Salih. (JSS Supplement 1. Oxford) Hillers, D.R. and E. Cussini. 1996. Palmyrene Aramaic Texts. (Baltimore, London) Khairy, N.I. 2000. ‘New Nabatean Inscriptions from the 1996 Survey in the Umm el-Jimal Area’, in L. Stager, J.A. Greene and M.D. Coogan (eds), The Archaeology of Jordan and Beyond. Essays in Honor of James A. Sauer (Winona Lake). 255–65 Knauf, E.A. 1985. ‘A New Nabatean Inscription from Umm el-Jimal’, AION 45, 17–18 Littmann, E. 1904. Semitic Inscriptions. (New York) —— 1914. ‘Nabataean Inscriptions’, in Publications of the Princeton University Archaeological Expeditions to Syria in 1904-1905 and 1909 (Leiden). Division IV, Section A Macdonald, M.C.A. 1999. ‘Personal Names in the Nabataean Realm: a Review Article’, JSS 44, 251–89 Al-Muheisen, Z. and A. Ajlouni. 2005. ‘A Nabataean Inscription from Northern Jordan Survey (Umm al-Jimal area)’, Syria 82, 167–72 Negev, A. 1991. Personal Names in the Nabatean Realm. (Jerusalem) Said, S. and M. al-Hamad. 2003. ‘A New Nabatean Inscription from Umm el-Jimal’, JSS 48, 29–34 —— 2004. ‘Three short Nabataean inscriptions from Umm al-Jimal’, Proceedings of the Seminar for Arabian Studies 34, 313–18 Stark, J.K. 1971. Personal Names in Palmyrene Inscriptions. (Oxford) Starcky, J. 1985. ‘Les inscriptions nabatéennes et l’histoire de la Syrie méridionale et du nord de la Jordanie’, in J.-M. Dentzer (ed.), Hauran I. Recherches archéologiques sur la Syrie du Sud à l’époque hellénistique et romaine (Bibliothèque Archéologique et Historique 124.1, Paris). 167–81
389
Journal of Semitic Studies LV/2 Autumn 2010 doi: 10.1093/jss/fgq005 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
NAMING NORMATIVITY: THE EARLY HISTORY OF THE TERMS SÛRAT HA-DÎN AND LIFNÎM MIS-SÛRAT HA-DÎN TZVI NOVICK NOTRE DAME
Abstract One of the most important terms for supererogation in Jewish sources from the Babylonian Talmud onward is lifnîm mis-sûrat ha-dîn. This article examines the usage of this and related terms in tannaitic literature, and discerns four developmental stages that culminated in the Babylonian Talmud’s ‘canonical’ usage.
Introduction In the medieval period, the term lifnîm mis-sûrat ha-dîn came to serve as a marker for the whole or the greater part of the supererogatory realm in Jewish practice, i.e., of behaviour in excess of what the law requires, and it has retained this usage in some academic discussion.1 A number of scholars have observed that in classical rabbinic sources it generally has a narrower sense, namely, the decision to renounce or My thanks to Professors Elitzur and Michal Bar-Asher Siegal, and to Professor Christine Hayes, for their critiques of earlier versions of this essay. This essay was made possible in part by a grant from the Memorial Foundation for Jewish Culture. 1 Thus, for example, Bernard Jackson, ‘Judaism as a Religious Legal System’, in Andrew Huxley (ed.), Religion, Law and Tradition: Comparative Studies in Religious Law (New York 2002), 43: ‘Jewish law is familiar with the notion of supererogatory standards, termed middat hasidut (“the standards of the pious”) or lifnim mishurat hadin’. Aharon Lichtenstein (‘Does Jewish Tradition Recognize an Ethic Independent of Halakha?’ in Menachem Marc Kellner [ed.], Contemporary Jewish Ethics [New York 1978], 115) ventures that ‘[l]ifnim mishurat hadin … is the sphere of contextual morality’. Aaron Kirschenbaum (Equity in Jewish Law: Beyond Equity: Halakhic Aspirationalism in Jewish Civil Law [Hoboken, New Jersey 1991], 109) uses lifnîm mis-sûrat ha-dîn as ‘the generic term for supererogatory acts’, though in his chapter on the term he aims ‘to delineate the exact usage of this phrase in tannaitic literature, the amoraic broadening of its application, and its ever-widening connotation as an operative force in Jewish legal thinking’. 391
NAMING NORMATIVITY
waive a property or right to which one is legally entitled.2 Thus, for example, R. Yehudah, walking after Shmuel in a certain market, asks him what one who finds a purse in such a market must do with it. Shmuel responds that he may keep it, on the assumption that the owner, having lost the purse in so public a place, will have despaired of finding it again, and thus will have renounced ownership of the purse. R. Yehudah presses him further: ‘What if a Jew came and provided identifying characteristics?’ ‘Then he is obligated to return it.’ ‘Both?’ (That is, if the purse belongs to the finder, how can he be obligated to return it in the latter case?) ‘It is lifnîm mis-sûrat ha-dîn.’3
In other words, the finder is entitled, under the law, to keep the purse, but if the owner does come to claim it, he should waive this entitlement and return it. To explain how the words lifnîm mis-sûrat ha-dîn convey this principle of waiver or renunciation, scholars usually analyse the term into parts. Thus, for example, Saul Berman isolates three categories: a basic legal obligation (dîn) (in the above case: to return a lost object), a limitation thereof (sûrat ha-dîn) (here: that the lost object need not be returned if one may reasonably assume that the owner has given up hope of recovering it), and a (typically) moral imperative to renounce the limitation and take on the basic obligation (lifnîm mis-sûrat ha-dîn).4 Boaz Cohen, by contrast, sees only two stages: sûrat ha-dîn, literally ‘the line of the law’, or strict law, and lifnîm mis-sûrat ha-dîn, literally ‘within the line of the law’, or equity.5 2
See, e.g., Louis E. Newman, ‘Law, Virtue and Supererogation in the Halakha: The Problem of “Lifnim Mishurat Hadin” Reconsidered’, JJS 40 (1989), 72: ‘It would appear that the concept of lifnim mishurat hadin parallels most closely notions of waiver in Anglo-American jurisprudence. … [T]he term lifnim mishurat hadin designates a willingness to waive voluntarily some benefit or right to which one is entitled by law’. See likewise Felix Böhl, Gebotserschwerung und Rechtsverzicht als ethisch-religiöse Normen in der rabbinischen Literatur (FJS 1; Freiburg im Breisgau 1971), 59: ‘Die Verzicht auf ein Recht wird gekennzeichnet durch die Formel “innerhalb der Rechtslinie handeln”’. 3 b. B.M. 24b (MS Hamburg 165). 4 ‘Lifnim Mishurat Hadin’, JJS 26 (1975), 86–104; 28 (1977), 181–93. Berman’s view is anticipated by Moshe Silberg (Talmudic Law and the Modern State [trans. B.Z. Bokser; New York 1973 (1961)], 123): ‘its distinguishing mark is that the one who volunteers to heed it … foregoes the qualification which relieves him and returns to the essence which obligates him, or leaves the peripheral thought and penetrates to the central issue’. Kirschenbaum (Equity, 110 and passim) adopts Berman’s view, while Newman (‘Law’, 63 n. 7) appears to reject it. 5 Jewish and Roman Law: A Comparative Study (New York 1966), 1.52. 392
NAMING NORMATIVITY
Cohen’s analysis is, I think, more or less correct as an account of the Bavli’s understanding of the terms sûrat ha-dîn and lifnîm mis-sûrat ha-dîn. But this essay argues that the Bavli’s usage represents a late development, whose earlier stages we may trace in tannaitic sources by careful philological analysis of a network of other related terms. These sources demonstrate that the term lifnîm mis-sûrat ha-dîn reflects a reinterpretation of sûrat ha-dîn that is inconsistent with the latter term’s original meaning; in origin, sûrat ha-dîn has little to do with the boundary of one’s legal duty. Diachronic study of lifnîm mis-sûrat ha-dîn has, in prior scholarship, taken the Bavli as its starting point, and moved forward from it into the medieval period and beyond. The current study, by tracing the prehistory of the Bavli’s usage, recovers a narrative of growing conceptual and terminological sophistication about the outer boundaries of legal obligation. Lifnîm mis-sûrat ha-dîn and Sûrat ha-dîn in Tannaitic Sources The term lifnîm mis-sûrat ha-dîn occurs in parallel passages in Mek. R. Ish. Yitro 2 (Horowitz ed., 198) and Mek. R. Sh. Exod. 18:20 (Epstein-Melamed ed., 133). In the biblical context, Jethro advises Moses to convey the people’s cases to God, and to return to the people with instructions as to what they should do. R. Joshua and R. Elazar of Modiin debate about the meaning of Jethro’s words. ‘And inform them of the way in which they should walk’: this is study of Torah; ‘and the deed’: this is good deeds. These are the words of R. Joshua. R. Elazar of Modiin says: ‘And inform them of ’: inform them of their life dwelling; ‘the way’: this is visiting the sick; ‘they should walk’: this is burial of the dead; ‘in which’: this is kindness; ‘and the deed’: this is sûrat ha-dîn; ‘that they should do’: this is lifnîm mis-sûrat ha-dîn.
The Mek. R. Sh. parallel involves a number of minor changes, among them the interpretation of ‘and the deed’ as dîn, not sûrat ha-dîn.6 R. Elazar does not elaborate on the meaning of sûrat ha-dîn or lifnîm mis-sûrat ha-dîn, though the general context suggests that the latter term may have something to do with kindness or piety. The words lifnîm mis-sûrat ha-dîn do not occur elsewhere in tannaitic literature, save in one very dubious instance in Midrash Tannaim.7 Given that 6
The two Bavli parallels (b. B.Q. 100a; b. B.M. 30b) also have dîn rather than sûrat ha-dîn. On the relationship between sûrat ha-dîn and dîn see below. 7 See Midr. Tann. Deut. 22:4 (Hoffmann ed., 134). A superfluity in the biblical verse teaches that the general obligation to assist a neighbour’s falling animal does 393
NAMING NORMATIVITY
R. Elazar, at many other points in his recorded debates with R. Joshua, adopts substantive positions and an exegetical methodology strikingly at odds with the tannaitic mainstream, the fact that the lone occurrence of lifnîm mis-sûrat ha-dîn comes in a quotation from R. Elazar suggests that the term’s marginality in tannaitic literature is no accident.8 Outside of this source, the phrase sûrat ha-dîn, without lifnîm min, occurs in seven other pericopes, one in the Mishnah, four in the Tosefta, and two in parallel passages in Mek. R. Ish. and Mek. R. Sh. I take up first the Mishnah passage, m. Gi†. 4:4. A slave whose master made him collateral, then freed him; in sûrat ha-dîn the slave is not liable for anything, but because of ‘establishing the world’ (tiqqûn ha-{ôlam) we compel his master and he makes him free and he writes a deed for his value. R. Shimon b. Gamaliel says: he does not write but frees.9
This pericope is obscure and much-debated.10 I offer the following interpretation, though my general argument here can accommodate not apply to individuals such as elders for whom corralling animals would be indecorous. But: ‘if he is pious (Ìasîd) and does lifnîm mis-sûrat ha-dîn, this is praiseworthy’. This final line is almost certainly not genuinely tannaitic. First, it is missing from the two parallel passages in Sifre Deut. 222, 225 (Finkelstein ed., 256–8), and the many parallel baraitot in the Bavli (b. Ber. 19b; b. B.M. 30a; b. San. 18b). Second, and more importantly, outside of Midr. Tann. Deut. 22:4, the association of the Ìasîd with the category of lifnîm mis-sûrat ha-dîn occurs only in the post-talmudic period. Finally, as Michal Bar-Asher Siegal pointed out to me, the relevant line in Midr. Tann. Deut. 22:4 occurs almost verbatim in Maimonides’ Mishneh Torah (Laws of Murder and Preserving Life, 13:4; Or Vishua ed., 757): ‘and if he is pious and does lifnîm mis-sûrat ha-dîn, even if is the high prince … he unburdens and loads with him’. Here, as elsewhere, Midrash ha-Gadol, from which Hoffmann drew, does not quote a genuine tannaitic midrash, but reworks an earlier source in light of Maimonides’ code. The editorial tendencies of Midrash ha-Gadol are described further in Michal Bar-Asher Siegal’s M.A. thesis (כט[ במכילתא לדברים- כב:פרשת מעשרות ]דב‘ יד בירור טקסטואלי ופירוש:[ דר“יHebrew University, 2006]), to be published soon. 8 On the R. Joshua/R. Elazar debates see Menahem I. Kahana, The Two Mekhiltot on the Amalek Portion: The Originality of the Version of the Mekhilta d’Rabbi Ishma}el with Respect to the Mekhilta of Rabbi Shim}on ben Yohay (Jerusalem 1999), 288–320. On the extreme pietism demanded by R. Elazar’s positions see Daniel Boyarin, Intertextuality and the Reading of Midrash (Bloomington 1990), 75. 9 The text is from MS Kaufman, as are other quotations from the Mishnah in this essay. 10 Part of the confusion may stem from the similarity between the final line of the anonymous view in this pericope, ‘we compel his master and he makes him free and he writes a deed for his value’, and the resolved law of the next pericope, ‘we compel his master and he makes him free and he writes a deed for half his value’. The echo is rhetorically effective, but appears to come at the expense of clarity. 394
NAMING NORMATIVITY
others: A master borrows money, and pledges his slave as collateral for the debt. Then he frees the slave. The Mishnah indicates that ‘in sûrat ha-dîn’, the slave is not liable for the debt. However, on grounds of ‘establishing the world’ — i.e., from fear lest the lender nevertheless attempt to seize the freed slave as collateral, or more likely, from a desire to protect lenders against such devious practices — we force the borrower to confirm the slave’s freedom and to write out a document pledging the value of the slave as collateral in place of the slave himself .11 Here, the norm characterized as ‘in sûrat ha-dîn’ is supplanted, on policy grounds, by a different norm that is more demanding of the master. The four Tosefta pericopes (t. Ter. 2:1–3; t. Pes. 4:7) — they amount to three, since t. Pes.4:7 replicates t. Ter. 2:3 — involve permutations of a situation in which, in the aftermath of a sale of food, the seller informs the buyer that the food is in fact prohibited from consumption. I quote as an example t. Ter. 2:1. One who sells fruit to his friend and tells him, ‘the fruit that I sold you is †evel (that is, food from which tithes have not been separated)’, … sûrat ha-dîn, he is not believed. R. Yehudah says: Jews are not suspected of this. Rather everything depends on the person.12
The first, anonymous view is that the buyer may, sûrat ha-dîn, disbelieve the seller and treat the food as altogether permissible. R. Yehudah objects to this blanket leniency: it cannot be assumed, across the board, that a Jewish seller would lie about such a matter. Instead R. Yehudah holds that the particular seller’s character must be taken into account. Saul Lieberman claims that the first speaker, by characterizing his own position as sûrat ha-dîn, implies that the buyer would in fact do well to believe the seller, and treat the food as prohibited. Or in other words, sûrat ha-dîn carries a trumping implication: to call a rule sûrat ha-dîn is to say that it should be trumped by a different course of action. Lieberman suggests that the sequence in t. Ter. 2:1–3 is continuous with the pericopes at the end of chapter 1 of t. Ter., which address cases in which the expected stringent rule is displaced on the ground of tiqqûn ha-¨ôlam (t. Ter. 1:12 [bis]; 1:13): ‘These baraitot come among a group of teachings that deal with cases in 11 On the early history of the term tiqqûn ha-¨ôlam, from which the current pericope diverges, see Sagit Mor, ‘Tikkun Olam – Implications of the Concept on Tannaitic Divorce Law’, Mo¨ed 15 (2005), 24–51. 12 Saul Lieberman, The Tosefta (New York 2001 [1955]), 1.111. Unsourced citations to the Tosefta elsewhere in this essay also use Lieberman’s text.
395
NAMING NORMATIVITY
which on account of tiqqûn ha-{ôlam they rejected sûrat ha-dîn (or in which one ought be strict beyond sûrat ha-dîn)’.13 The contrast in m. Gi†. 4:4 between sûrat ha-dîn and tiqqûn ha-¨ôlam strengthens Lieberman’s case. As Lieberman’s parenthetical remark implies, he sees a difference between the cases in t. Ter. 1, where the technically correct rule is rejected, and t. Ter. 2:1–3, where the technically correct rule (that the seller is disbelieved) remains in place, with only the expectation that the buyer will act more stringently than the rule requires. In fact, however, a related discussion in t. Ma¨as. Sh. 3:8 indicates that in the produce cases, too, the technically correct rule is displaced. Thus t. Ma¨as. Sh. 3:8: ‘One who sells fruit to his friend and tells him, “the fruit that I sold you is †evel ”, and the buyer says, “you sold me tithed fruit”, we compel the seller to tithe the fruit.’14 Lieberman, cross-referencing t. Ter. 2:1–3, writes: ‘the buyer is conscientious and does not wish to rely on sûrat ha-dîn’.15 But there is no indication that the buyer believes the seller: on the contrary, he explicitly contradicts the seller. Rather, the shift from the rule characterized as sûrat ha-dîn (t. Ter. 2:1–3) to the rule of t. Ma¨as. Sh. 3:8 simply involves a transfer of the loss consequent upon the seller’s revelation from the buyer to the seller. Under the rule of t. Ter. 2:1–3, the buyer bears the loss: he must either tithe the produce, or risk eating †evel. Under t. Ma¨as. Sh. 3:8, the seller bears the loss: it is he who must tithe the produce (whether or not the buyer believes him). The question of the buyer’s piety does not arise. There is a structural analogy between m. Gi†. 4:4 and the cases of produce in the Tosefta: the technically correct rule (sûrat ha-dîn), which would disadvantage the victim of a questionable commercial practice (the borrower in m. Gi†. 4:4, the buyer in the Tosefta) is replaced by a different rule that protects the victimized party. In both cases, the new rule is backed by coercion (‘we compel’). The final case of sûrat ha-dîn in tannaitic literature occurs in connection with Exod. 17:1–2, where the Israelites, lacking water, complain to Moses. The rabbinic commentaries to Exodus (Mek. R. Ish. Vayassa¨ 6 [Horowitz ed., 173] and Mek. R. Sh. Exod. 17:2 [Epstein-Melamed ed., 117]) remark: ‘they transgressed sûrat ha-dîn’. Mek. R. Ish. continues:
13 14 15
Lieberman, Tosefta Ki-Fshu†ah, 1.307–8. See also t. Ma¨as. Sh. 3:12. Lieberman, Tosefta Ki-Fshu†ah, 2.755. 396
NAMING NORMATIVITY
It is customary practice (derek }ereÒ) that when a person is angry in his home, he sets his sight on his youngest son, but these set their sights on the oldest. Therefore it says, ‘And the people contended with Moses’: they transgressed (¨aberû ¨al) sûrat ha-dîn.16
The midrash, apparently punning ‘they contended’ (wa-yareb) in the verse on rab ‘great’ or on ¨-b-r ‘to transgress’, takes the Israelites to have violated customary practice by imposing their frustrations on a great man. While the term derek }ereÒ can simply describe standard behaviour, without any implication such behaviour is desirable, it seems clear that the midrash intends to criticize Israel for departing from derek }ereÒ, and thus for transgressing sûrat ha-dîn.17 In the above sources from tannaitic literature, sûrat ha-dîn describes a rule or norm. The characterization of a rule as sûrat ha-dîn can signal that it is to be trumped by another rule — we have called this a trumping implication — but this tendency must be qualified in various ways. First, the trumping implication does not always arise: in Mek. R. Ish. Vayassa¨ 6 = Mek. R. Sh. Exod. 17:2, sûrat ha-dîn indicates the prevailing norm (not trumped, in the end, by any other). A similar usage of sûrat ha-dîn occurs in a later text, Deut. Rab. Re}eh 1 (Lieberman ed., 92): ‘“I am with him in distress” (Ps. 91:15): it is not sûrat ha-dîn that my children should be cursed while I am blessed’. Second, in the 16
This continuation does not occur in MS Oxford 151, nor in the Genizah fragment MS Oxford Hebrew d 62.82 (Menahem I. Kahana, The Genizah Fragments of the Halakhic Midrashim: Part I [Jerusalem 2005], 80 ll. 2–3), nor in the Mek. R. Sh. parallel. But it does occur in MS Munich and in the editio princeps. In the absence of characteristically late features, the continuation should not be dismissed as late, as its omission is easily attributed to homoioteleuton. For another case in tannaitic literature where the Israelites are criticized for complaining to Moses rather than consulting with lesser leaders, see R. Joshua’s comment to Exod. 16:2 in Mek. R. Ish. Vayassa¨ 1 (Horowitz ed., 155). Language similar to that in Mek. R. Ish. Vayassa¨ 2 is employed in Sifra SheraÒim 1:3 (Weiss ed., 47b) in observing that Moses in Lev. 10:16, though unhappy with a decision made by Aaron and his sons, directs his anger at the sons alone. In later tradition, the Israelites in Exod. 17:2 and Moses in Lev. 10:16 are set off as counter-types. See }Abot R. Nat. A chapter 37. 17 On the range of meanings of derek }ereÒ see Shmuel Safrai, ‘The Term Derekh ereÂ’, Tarbiz 60 (1990), 147–62; David Flusser, ‘“Which is the Right Way that a Man Should Choose for Himself?” (Sayings of the Fathers, 2:1)’, Tarbiz 60 (1991), 163–78; Avraham Walfish, ‘Creative Redaction and the Power of Desire: A Study of the Redaction of Tractate Qiddushin: Mishnah, Tosefta, and Babylonian Talmud’, JSIJ 7 (2008), 13 n. 29. Boaz Cohen (Law and Tradition in Judaism [New York 1969], 233) has sûrat ha-dîn here indicate ‘the rule of expediency’, without any normative connotation. 397
NAMING NORMATIVITY
two informative contexts where the trumping implication does arise, m. Gi†. 4:4 and t. Ter. 2:1–3/ t. Ma¨as. Sh. 3:8, the dynamic is altogether different from that in the Bavli example given above. The trumping rule has the force of law (‘we compel’), and does not in any way involve supererogation or waiver of an entitlement. Finally, the trumping implication, where it occurs, has no settled form: the implication is either left unstated (as in t. Ter. 2:1–3, although it is spelled out elsewhere, in t. Ma¨as. Sh. 3:8), or grounded in the consideration of tiqqûn ha-¨ôlam (m. Gi†. 4:4), or classified as lifnîm mis-sûrat ha-dîn (Mek. R. Ish. Yitro 2 = Mek. R. Sh. Exod. 18:20). In the next section, I survey terms that occur in the same context as sûrat ha-dîn and do not carry the trumping implication. The evidence collected in this section supports the conclusion dictated by Mek. R. Ish. Vayassa¨ 6 = Mek. R. Sh. Exod. 17:2, that the trumping implication is not an obligatory feature of the usage of sûrat ha-dîn. Then I examine the semantics of sûrat ha-dîn and lifnîm mis-sûrat ha-dîn, and show that the latter is lexically awkward, and likely reflects a reinterpretation of sûrat ha-dîn against its original sense. This conclusion coalesces with the above observations on the marginality of lifnîm mis-sûrat ha-dîn in tannaitic literature, namely, first, that it occurs only once, and that in the mouth of a figure who is in many ways an outlier; and second, that when sûrat ha-dîn is trumped elsewhere, the trumping implication does not involve the term lifnîm mis-sûrat ha-dîn, and has nothing to do with piety. In the final section, I focus on the verbs that take lifnîm mis-sûrat ha-dîn as their object, and use them to isolate another stage in the term’s development. Dîn, Sûrâ, Derek }EreÒ and Derek As sûrat ha-dîn can signal either the governing rule or a rule supplanted (in practice or ideally) by other considerations, so likewise can dîn standing alone, particularly in the form ba-dîn.18 Ba-dîn indicates 18 For the vocalization of ba-dîn see m. Mid. 3:4 in MS Kaufman. The range of meanings in the word dîn, from ‘judgment’ to ‘apt’, is also found in Greek âzíwma. In the sources that follow, dîn sometimes occurs as a MS variant in place of ba-dîn. I think it probable that ba-dîn is the original form in the relevant phrases ‘it is/was ba-dîn’ and ‘it is not ba-dîn’ (phrases not involving a fortiori argumentation), and was corrupted to dîn due to the prevalence of the latter as a term of art in a fortiori argumentation. But it is altogether possible that dîn and ba-dîn occur in more or less free variation. For my argument I need not decide between these possibilities.
398
NAMING NORMATIVITY
the governing rule in Mek. R. Ish. Kaspa 20 (Horowitz ed., 333). Interpreting Exod. 23:15 ‘and my face shall not be seen emptyhanded’ as a reference to whole-offerings, the midrash offers this reasoning: ‘it is not appropriate (ba-dîn) that your table should be full (from sacrificial meat) and your creator’s empty’.19 Likewise, in Mek. R. Ish. Yitro 1 (Horowitz ed., 195), the Sages protest that it is not appropriate (ba-dîn) that R. Gamaliel should stand and serve them.20 Ba-dîn indicates the supplanted norm in the phrase ba-dîn hû}, as in t. San. 11:6 (MS Vienna), where R. Shimon b. Elazar observes that the law of the rebellious son ought in fact to apply only to the daughter (because disobedience is to be expected from sons?): ‘it ought to be (ba-dîn hû} lit. “it is ba-dîn”) the daughter and not the son, but such is the decree of the king’. Likewise in t. So†. 15:10 (MS Erfurt), R. Ishmael asserts: it is appropriate (ba-dîn hû} ) that after the Temple’s destruction, no wine or meat should be consumed, but we do not impose upon the congregation an impossibly harsh decree.21 The term sûrâ, too, standing alone, can indicate propriety. In tannaitic literature this usage occurs only once, in m. B.B. 8:5, where R. Shimon b. Gamaliel encourages fathers to disinherit sons who do not act as they ought: One who transfers his property to others and not to his sons, the transfer is effective, but the Sages are displeased with him. R. Shimon b. Gamaliel says: if his sons do not act ba-sûrâ then [one who thus disinherits them] should be mentioned for good.22
19 The text is taken from MS Antonin 239 ll. 20–1 (Kahana, Genizah Fragments, 146). In MS Oxford 151 dîn occurs in place of ba-dîn. The midrash here probably takes ‘shall not be seen’ (yire}û) paranomastically as ‘appropriate’ (ra} ûy) so that the verse says: ‘it is not appropriate that my presence be empty’. 20 Ba-dîn occurs in MS Oxford 151 and is confirmed by the parallels in Mek. R. Sh. Exod. 18:12 (Epstein-Melamed ed., 131) and Sifre Deut. 38 (Finkelstein ed., 74). Other cases of ba-dîn describing the prevailing norm are: m. Mid. 3:4; Sifra AÌare Mot 8:3 (bis) (Weiss ed., 86a, 92d); Sifre Zu†a Num. 12:13; 31:21 (bis) (Horowitz ed., 277, 329). 21 In MS Vienna, ‘it is dîn’ occurs instead of ‘it is ba-dîn’. Other examples are t. Hor. 2:10; Mek. R. Ish. Neziqin 8 (Horowitz ed., 274); Sifre Num. 4 (Horowitz ed., 7). In t. Shevu. 3:1–3, absence of liability in a series of cases is identified as min ha-dîn (not ba-dîn), and in each case qualified with the observation: ‘but the heavens do not forgive him until he pays’. 22 Thus MS Kaufman. MS Lowe and possibly MS Parma have ka-sûrâ. In MS Lowe, the Mishnah adds bô ‘with respect to him’, thus: ‘if his sons did not act with respect to him ka-sûrâ’.
399
NAMING NORMATIVITY
The same usage is attested (always with ka-sûrâ) in baraitot cited in the Bavli (b. B.M. 73b [bis]; b. B.B.133b), and is very common in later sources.23 The three terms triangulated above, sûrat ha-dîn, dîn and sûrâ, all interchange with derek }ereÒ, the first two in tannaitic texts and the third in an amoraic text. Sûrat ha-dîn and derek }ereÒ interchange in Mek. R. Ish. Vayassa¨ 6, quoted above. Dîn interchanges with derek }ereÒ in Mek. R. Sh. Exod. 3:8 (Epstein-Melamed ed., p. 2): R. Elazar b. Arak said: why did the holy one, blessed be he, reveal himself from the heavens to Moses in a thornbush? He should have (spoken) from the cedars of Lebanon or from the tops of mountains and hills! But God made low his presence and acted in a courteous fashion ({asa devarô kederek }ereÒ), lest the nations of the world say: because he is God, and lord of the world, he acted in an improper fashion ({asa devarô se-lô} ka-dîn).
It is just possible that derek }ereÒ here indicates not proper behaviour but human behaviour: God spoke as a human, from the ground, rather than from the heights. But in light of the precise verbal parallels between the two phrases as a whole, it is more likely that derek }ereÒ is the contextual equivalent of dîn. God speaks with Moses properly, courteously, so as not to give the appearance of a strongman. Derek }ereÒ interchanges, finally, with sûrâ in Lev. Rab. 26 (Margulies ed., 600). In this pericope, R. Isaac, reflecting on 1 Sam. 28:8, where Saul visits the witch with two men, remarks that the Torah teaches derek }ereÒ, a behavioural norm, namely, that one should not take to the road with fewer than two men as escorts. R. Ibo follows immediately by observing that two biblical figures acted ka-sûrâ: Saul, in the aforementioned incident, and Abraham who takes two servants with him on his journey to Moriah (Gen. 22:3). The same norm is described as derek }ereÒ and sûrâ.24 The terms derek and dîn interchange again as objects of the verb ¨-b-r pi¨el or hif ¨il. According to t. B.M. 11:27, one who provides a public service (e.g., a baker) may not quit without providing a replacement unless he so stipulated in advance before a court or unless ¨îbberû ¨alayw }et ha-derek.25 The same phrase is also attested in Sifre 23
E.g., y. Mo¨ed Q. 3:1 (81d); Gen. Rab. 31, 36, 49, 50, 67, 93 (TheodorAlbeck ed., 279, 336, 512, 520, 755, 1159). Note the interchange of ka-sûrâ with ya}ût in Gen. Rab. 67, and with ke-hôgen in Gen. Rab. 93. 24 In some MSS of Lev. Rab. and in the Gen. Rab. parallel (Theodor-Albeck ed., 594), bi-derek }ereÒ occurs in place of ka-sûrâ in R. Ibo’s statement. 25 For ¨-b-r in the pi¨el form here see Lieberman, Tosefta Ki-Fshu†ah, 10.324. That ha-derek occurs here rather than derek }ereÒ should give little pause. For derek 400
NAMING NORMATIVITY
Deut. 29 (Finkelstein ed., 46). Here Moses’ account of God’s refusal to entertain Moses’ plea to enter Canaan is thus described: ‘as one says to his friend: so-and-so ¨îbber derek ‘alay’. The same phrase is also attested once in an amoraic source (b. San. 31b), where a certain Uqvan the Babylonian is recorded as lodging a complaint with the court in Tiberias to the effect that Jeremiah his brother ‘he¨ebîr ¨alayw }et ha-derek’.26 Here ¨-b-r is in the hif¨il rather than the pi¨el, but in all three cases, the phrase seems clearly to involve unfair dealing or mistreatment. A variant of this idiom where dîn substitutes for derek occurs in tannaitic and amoraic sources. According to the anonymous voice of m. Mid. 2:2, one who encounters a banned individual in the Temple comforts him by saying: ‘may he who dwells in this house place it in their hearts to draw you near’, i.e., to revoke the ban. R. Yose challenges this view: ‘you have made it as though ¨îbberû ¨alayw }et ha- dîn’. That is, the blessing recommended by the anonymous view implies that the Sages acted wrongly in banning the addressee. In Deut. Rab. Va}etÌanan s.v. rab lak (Lieberman ed., 49), which parallels the passage from Sifre Deut. 29 quoted above, Moses’ account is paraphrased as: ‘there was ¨îbbûr ha-dîn ¨alay, on your account’.27 Likewise, in a baraita cited in y. B.B. 8:1 (15d–16a), R. Ishmael characterizes inheritance by daughters rather than sons as ¨îbbûr ha-dîn.28 Hanoch Yalon has established that ¨îbber/he¨ebîr in the above idiom means ‘to ruin, to pervert’.29 The idiom is thus more or less alone carrying the same sense as derek }ereÒ, see Mek. R. Ish. Amaleq 2 (Horowitz ed., 181) per MS Oxford 151: ‘let everyone learn a lesson (derek) from Amaleq’. The reading of MS Oxford 151 is confirmed by MS Munich, and the gap in a Genizah fragment in Kahana, Genizah Fragments, 84, seems to have room only for derek. Other MSS have derek }ereÒ. For another case of derek alone indicating what one ought to do, see R. AÌa’s statement about the blessing of the bloodletter in b. Ber. 60a. 26 See MS Yad R. Herzog. 27 The qi††ûl noun class, to which ¨îbbûr belongs, derives from the pi¨el stem. The Deut. Rab. passage has important text-critical implications. In the Sifre Deut. version, the comment attaches to the lemma rab lak, but the connection is obscure. In the Deut. Rab. version, the comment attaches to the word wa-yit¨abber, and evidently puns on it. See Finkelstein’s comment in Sifre Deut. ad loc., where he conjectures that the midrash is based on the word wa-yit¨abber but does not advert to the Deut. Rab. parallel. 28 See also Mek. R. Sh. Exod. 23:6 }e¨abber ¨alayw }et ha- dîn (Epstein-Melamed ed., 215). In the corresponding line in Mek. R. Ish. Kaspa 20 (Horowitz ed., 346), the root n-†-y, taken from the lemma (Exod. 23:6), is employed instead of ¨-b-r: }anî ma††eh ¨alayw }et ha- dîn. 29 Hanoch Yalon, Introduction to the Vocalization of the Mishna (Jerusalem 1964), 142–5; 202–3. 401
NAMING NORMATIVITY
equivalent to qilqel + dîn, as in Sifra Qedoshim 2:2 (Weiss ed., 88d), which speaks of ‘the judge who perverts justice (miqalqel et ha-dîn)’.30 In amoraic sources, qilqel can also govern sûrâ. Thus, for example, in Gen. Rab. 55, R. Shimon b. YoÌai voices astonishment at the fact that Abraham saddled his own ass when setting out to sacrifice Isaac (Gen. 22:3): did he not, after all, have many slaves who might have performed this task? His answer: ‘love perverts (meqalqelet) the sûrâ’.31 Origins of Sûrat ha-dîn and Lifnîm mis-sûrat ha-dîn The overlapping usage of the terms sûrat ha-dîn, dîn, sûrâ, derek }ereÒ, and derek supports the hypothesis that sûrat ha-dîn simply conveys, like these others terms, a norm or rule. Its usage can but need not imply that some other rule does or should govern. The term sûrat ha-dîn may have arisen by the combination of the rough synonyms sûrâ and dîn in a construct phrase, a phenomenon known in Hebrew poetics as gîbûb ‘piling’.32 The term derek ha-dîn, which occurs in one passage in tannaitic literature, Sifre Num. 8 (Horovitz ed., 14), and seems to mean roughly the same thing as sûrat ha-dîn — the terms derek and sûrâ overlap here again — likely originated in the same way.33 Given that dîn in the sense of ‘the appropriate rule’ occurs far more often than the other terms discussed above, we may suppose that the construct phrases derek ha-dîn and sûrat ha-dîn were designed to select for dîn in this sense, in contrast with other common meanings of dîn, especially ‘lawsuit’ and ‘an a fortiori argument’. Possibly, though I think less likely, sûrat ha-dîn is a calque on Aramaic sedar dîn. The Aramaic phrase regularly renders BH mispa† 30 This latter idiom appears never to occur with derek. Like ¨îbber (see Yalon, Introduction, 142–5), qilqel can govern Òûrâ ‘shape’, as in Gen Rab. 32:5 (TheodorAlbeck ed., 292). 31 See also Gen. Rab. 95 (Theodor-Albeck ed., 1232). 32 For the term and examples in the Qumran Rule Scroll see Jacob Licht, The Rule Scroll: A Scroll from the Wilderness of Judaea: 1QS, 1QSa, 1QSb (Jerusalem 1965), 32. 33 On the Sifre Num. 8 passage see Moshe Halbertal, Interpretative Revolutions in the Making: Values as Interpretative Considerations in Midrashei Halakhah (Jerusalem 1999), 19–20. In this passage, it is observed that the initial interpretation of the verse would have the consequence that derek ha-dîn, likely referring in this case to the principle that God’s ‘good’ aspect outweighs his ‘destructive’ aspect, is ‘struck (lôqe) through it’. This interpretation is therefore modified so that derek ha-dîn is not ‘struck’. Is the phonological resemblance between the root l-q-y, used here, and q-l-q-l, used in connection with dîn and sûrâ, coincidental?
402
NAMING NORMATIVITY
in Neofiti.34 Sedar is also the standard translation equivalent of BH †ûr ‘line, row’ in Onqelos and Neofiti (Exod. 28:17–20; 39:10–13) and in Jonathan (1 Kgs 6:36; 7:2–4, 12, 18, 20, 24, 42).35 Given the synonymy of MH sûrâ and BH †ûr, sûrat ha-dîn may constitute the Hebrew equivalent of sedar dîn. If there is a relationship between the two terms, then the direction of influence was probably from Aramaic to Hebrew, since sedar in Neofiti also occurs as nomen regens in other common constructions, including sedar berkatâ, sedar ¨ovad bere}sît, and sidrê qeravâ. Ordinarily, the Hebrew equivalents of such expressions use seder not sûrâ — thus, for example, seder berakôt in m. R. ha-Sh. 4:5 and seder bere}sît in b. San. 108b — but one does find sûrôt ha-milÌamâ in Mek. R. Ish. Amaleq 1 (Finkelstein ed., 151), evidently the counterpart to Neofiti’s sidrê qeravâ .36 Whatever the origin of sûrat ha-dîn, the extended phrase lifnîm mis-sûrat ha-dîn seems to concretize the metaphor latent in the word sûrâ, ‘line, row’. The compound preposition lifnîm min governs nouns indicating either an area (such as he-ÌaÒer, ‘the courtyard’) or a boundary (such as ha-Ìômâ, ‘the wall’).37 Something lifnîm min an area is usually inward from it (e.g., one courtyard behind another, i.e., farther from the street), while something lifnîm min a boundary is within it.38 34
Cf. the equivalent terms in Greek (táziv toÕ nómou in Plato, Laws, 925B) and Latin (ordo legis in Pliny the Younger, Epistles, 3.ix.31), both indicating a legal provision. 35 See also Targum Jonathan to Ezek. 28:13, where sedar renders BH mesukâ in evident allusion to the rows of the high priest’s breastplate. Cf. BH sederôt and Akk. sidirtu ‘row’; sadara ‘to array, to set in a row’. 36 Here MS Oxford 151 has ke sûrôt ba-milÌamâ, but this version, too, clearly echoes the Neofiti phrase. The parallel in Mek. R. Sh. Exod. 17:13 (Epstein-Melamed ed., 123) also refers to sûrôt of warriors. Kahana (Two Mekhiltot, 305) supposes that the midrashim refer to the Greek phalanx, but if sûrôt ha-milÌamâ is a calque on sidrê qeravâ, Kahana’s claim becomes less probable. The correspondences between sedar in Neofiti and sûrâ in the rabbinic sources would constitute counter-examples to B. Barry Levy’s observation (Targum Neophyti 1: A Textual Study [Lanham 1986], 27) that sedar in Neofiti occurs ‘in a number of idioms but rarely, if ever, as the equivalent of an independent Hebrew term’. Boaz Cohen (Jewish and Roman Law, 1.52) traces sûrat ha-dîn, which he renders literally as ‘the line of the law’, to Isa. 28:17, ‘I shall make law (mispa†) the line (qaw), and justice the plummet’. ‘Equity is, in a way, a special rule, adapted to specific circumstances. Similarly Aristotle compares the special ordinance made to fit the circumstances of the case to the leaden rule used by the Lesbian builders (hosper kai tes lesbias oikodomes ho molibdinos kanon)’. The reference to Aristotle is puzzling: Aristotle is speaking of equity, which for Cohen corresponds to lifnîm mis-sûrat ha-dîn, not sûrat ha-dîn. In any case, were Isa. 28:17 the source of the term, we should expect something like qaw ha-mispa†, not sûrat ha-dîn. 37 The preposition is itself a calque on Aramaic le-gô min. See Richard C. Steiner, ‘New Light on the Biblical Millo from Hatran Inscriptions’, BASOR 276 (1989), 21. 38 See Steiner, ‘New Light’, 21, 22 n. 2. As an exceptional case where lifnîm 403
NAMING NORMATIVITY
As it is difficult to imagine what it might mean to be behind sûrat ha-dîn, the phrase presumably signals, rather, the area within sûrat ha-dîn, with the latter term signifying the outer boundary of dîn. sûrâ is, however, an awkward lexical choice to describe an outer boundary; the phrase sûrâ ÌîÒônâ can indicate the outermost of a series of rows (see, e.g., m. Dem. 7:8; t. Ber. 2:11), but sûrâ is never associated, per se, with the boundary demarcating an area. The lexical awkwardness of lifnîm mis-sûrat ha-dîn thus further confirms the conclusion reached above, that the phrase encodes a later, secondary interpretation of the term sûrat ha-dîn, originally ‘the norm’, as ‘the line of the law’.39 The terminological independence of sûrat ha-dîn from lifnîm mis-sûrat ha-dîn also emerges from the pattern of attestation in the tannaitic corpus: lifnîm mis-sûrat ha-dîn is attested only once (in two parallel sources), while sûrat ha-dîn occurs alone five times. Two Stages in the Usage of Lifnîm mis-sûrat ha-dîn The verb signalling comportment with the lifnîm mis-sûrat ha-dîn standard is not attested in tannaitic literature, but in talmudic sources it is either niknas ‘to enter’ or {asa and its Aramaic equivalent ¨abad ‘to do’.40 While niknas governs lifnîm min in many contexts, the usage of ¨asa with lifnîm min, or ¨abad with le-gô min, is unattested other than with lifnîm mis-sûrat ha-dîn. Usage of ¨asâ/¨abad with lifnîm mis-sûrat ha-dîn thus depends on the existence of lifnîm mis-sûrat ha-dîn as a technical term, to which the latter’s occurrence in Hebrew in otherwise Aramaic sources also attests. A closer look at the distribution of niknas and ¨asa/¨abad provides some insight into a development in the contextual meaning of lifnîm mis-sûrat ha-dîn. The genuine tannaitic attestations provide no information about the phrase’s meaning. In the later sources, as noted above, to act min governing an area indicates ‘within’, Steiner cites only one doubtful example in m. B.B. 3:7, but the phrase lifnîm min ha-maÌaneh clearly indicates ‘within the camp’ in Sifra MeÒora 1:1 (Weiss ed., 70b bis); 2:1 (Weiss ed., 71b bis); Sifre Deut. 256 (Finkelstein ed., 281 bis). In one of these cases, the phrase is governed by hayâ ‘was’ and in all the others, by likkanes ‘to enter’, and it may be that likkanes, in particular, licenses the ‘within’ interpretation even for areas. 39 Another possible solution, to my mind much less likely, is that lifnîm min governing sûrâ is used loosely or metaphorically to mean ‘less than’. This meaning is attested in t. ¨Uqs. 1:2, where lifnîm mik-ke-sî{ûr ‘less than the prescribed amount’ contrasts with yôter mik-ke-sî{ûr ‘more than the prescribed amount’. Cf. the use of êlattoÕsqai at n. 42 below. 40 Niknas occurs in b. Ber. 7a (bis); Deut. Rab. Re}eh 3 (Lieberman ed., 93). ¨asa/¨abad is used in b. Ber. 45b; b. Ket. 97a; b. B.Q. 99b; b. B.M. 30b (bis); b. ¨A.Z. 4b (bis). 404
NAMING NORMATIVITY
lifnîm mis-sûrat ha-dîn is usually to waive a personal exemption. Let us take another example, concerning a story in which Rav instructs R. Îiyya to compensate a woman for the loss she incurred because R. Îiyya had mistakenly counselled her to accept a buyer’s debased coin. The Talmud continues: And how does this differ from Danbo and Isur, who were exempt (from paying for their mistakes)? For they (were so expert that they) did not need to study. But R. Îiyya also did not need to study! Let us therefore say that R. Îiyya acted lifnîm mis-sûrat ha-dîn.41
The stam concludes that R. Îiyya was not obliged to compensate the woman; he acted lifnîm mis-sûrat ha-dîn. In all cases involving a dynamic of this sort, the verb ¨abad is used. Cases with niknas operate quite differently. In Deut. Rab. Re}eh 3 (Lieberman ed., 93), God not only provides the Israelites two paths, of life and of death, but also ‘enters’ lifnîm mis-sûrat ha-dîn and urges them to choose life. In b. Ber. 7a, God prays, and also is asked, that he ‘enter’ lifnîm mis-sûrat ha-dîn and judge his people mercifully. The association of lifnîm missûrat ha-dîn with God’s merciful judgment also occurs in b. ¨A.Z. 4b, with ¨asa, but otherwise, the distinction is sharp: ¨asa/¨abad + lifnîm mis-sûrat ha-dîn signals waiver of an exemption, whereas niknas + lifnîm mis-sûrat ha-dîn indicates something more like mercy, a kindness from an individual in a position of power. The image of ‘entering’ within the line of the law, thus contextualized, recalls the classical Athenian commonplace, expressed by both Aristotle and Demosthenes, that the prosecuting party should seek less (êlattoÕsqai) than what it is by right entitled to lest it appear filódikov ‘fond of lawsuits’ or âkribodíkaiov ‘exact in suing’.42 To enter within the line of the law is likewise to settle for less than one might extract. One retreats from the outermost bound of one’s power.43 The cases 41
b. B.Q. 99b (MS Hamburg 165). Dem., 56.14; Aristotle, E.N.1138a. Both passages are quoted in C. Carey, ‘Nomos in Attic Rhetoric and Oratory’, JHS 116 (1996), 41 n. 31. On the Aristotle passage see also Nancy Sherman, Fabric of Character: Aristotle’s Theory of Virtue (New York 1989), 23–4; Albrecht Dihle, The Theory of Will in Classical Antiquity (Berkeley 1982), 136. 43 It is not impossible that this more judicial sense of lifnîm mis-sûrat ha-dîn is intended in the one tannaitic instance (Mek. R. Ish. Yitro 2 = Mek. R. Sh. Exod. 18:20). The midrash attaches its distinction between sûrat ha-dîn and lifnîm mis-sûrat ha-dîn to Exod. 18:20, a context in which Jethro is advising Moses about establishing a judicial system. It is true that Jethro only proposes the appointment of other judges in Exod. 18:21, but the topic has already been broached in Exod. 18:20. See Berman, ‘Lifnim’, 186–7. It is clear, in any case, that Pseudo-Jonathan to this verse, 42
405
NAMING NORMATIVITY
of lifnîm mis-sûrat ha-dîn with niknas thus not only make grammatical sense but also work sensibly as idioms (the lexical awkwardness of sûrâ aside). The cases with ¨abad, by contrast, present not only a grammatical challenge — lifnîm min/le-gô min is not otherwise governed by ¨abad¨asa — but also an idiomatic one: why should waiver of a personal exclusion, essentially a proactive decision, be described in terms of retreat from the law? Conclusion The evidence marshalled above allows us to trace four stages in the prehistory and early history of the phrase lifnîm mis-sûrat ha-dîn. 1. sûrat ha-dîn occurs by itself, and indicates a rule. The trumping implication (that some other rule should govern) is nonexistent or weak. (Mek. R. Ish. Vayassa¨ 6 = Mek. R. Sh. Exod. 17:2) 2. sûrat ha-dîn acquires a more consistent trumping implication, which, however, has nothing to do with renunciation. (m. Gi†. 4:4; t. Ter. 2:1–3/Ma¨as. Sh. 3:8) 3. The trumping implication of sûrat ha-dîn comes to be associated specifically with renunciation and to be expressed as [niknas] lifnîm mis-sûrat ha-dîn. (Mek. R. Ish. Yitro 2 = Mek. R. Sh. Exod. 18:20; cases involving judgment in the Bavli) 4. lifnîm mis-sûrat ha-dîn becomes a technical term, so that it can be governed by the verb ¨asa/}abad. (standard Bavli usage) The first two stages predominate in tannaitic literature. The third stage has roots in one tannaitic source, but clear evidence for it and for the fourth stage becomes available only in the Bavli. The changes in the usage of the terms sûrat ha-dîn and lifnîm mis-sûrat ha-dîn attest to the growing sophistication of legal thought and the growing specialization of legal language in progressive strata of rabbinic literature. Address for Correspondence: 130 Malloy Hall, Department of Theology, Notre Dame, IN 46556, USA which depends on Mek. R. Ish., understands sûrat ha-dîn and lifnîm mis-sûrat ha-dîn as standards of judicial behaviour: ‘and the deed of sûrat ha-dîn and that they should do within the sûrta for the wicked’. See Böhl, Gebotserschwerung, 63 n. 182. The fact that the other areas of behaviour listed in Mek. R. Ish. Yitro 2 = Mek. R. Sh. Exod. 18:20 concern the general populace (e.g., visiting the sick) does not preclude sûrat ha-dîn and lifnîm mis-sûrat ha-dîn from attaching specifically to the forensic context, since the terms could, particularly given the Greek precedent, envision not only judicial indulgence but also indulgence by the plaintiff. 406
Journal of Semitic Studies LV/2 Autumn 2010 doi: 10.1093/jss/fgq006 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT1 FRANÇOISE BRIQUEL CHATONNET
AND
ALAIN DESREUMAUX
CNRS, PARIS
Abstract The purpose of the paper is to describe the Kerala type of writing as one of the Syriac scripts in the Syriac Oriental branch. It describes the characterization of the ductus. Already found in sixteenth century manuscripts, when an influence of Latin Portuguese style of writing is probable, this kind of script developed in the nineteenth century, mainly as a result of the isolation of the Syro-Malabar communities with regard to the mainstream Syriac tradition.
According to the typology found in Syriac language manuals, there are three forms of Syriac script: Estrangela, Serto and Syro-Oriental. Introductions often leave aside the Syriac Melkite script, which was illustrated by Hatch.2 Recent studies have allowed progress in the understanding of the relationship between Serto and Estrangela3 and of the regional characteristics of Syriac Occidental scripts.4 Differences and evolutions of the Syriac scripts in the Oriental branch are, on the contrary, mostly ignored. It is now possible to study Syriac scripts from Central Asia and China, thanks to recent research collecting epigraphic and manuscript documents from these regions.5 Moreover, the Syriac script from Kerala (India), which was already identified in the Encyclopédie of Diderot and d’Alembert6 in 1
The authors thank warmly Robert Hawley who corrected their English. Hatch 1946. 3 Healey 2000; Briquel Chatonnet 2001. Ayda Kaplan has made very interesting suggestions about a transitional phase of Syriac script, intermediary between Edessean and Serto, in her diploma at Leuven University: we are grateful for permission to mention it. It is now fully developed in her unpublished Ph.D. See already . 4 Desreumaux 2004. 5 See the contributions about Kirghizstan and China in Briquel Chatonnet, Debié and Desreumaux 2004. 6 In the commentary of the plates concerning Syriac script. 2
407
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT
the eighteenth century and by Lenormant in 1873,7 is generally forgotten. The preparation of the Kerala volume of the Recueil des inscriptions syriaques8 allowed an opportunity to examine several inscriptions written in this very peculiar Syriac script. We were also able to identify this script-type in several manuscripts of different collections around Kerala.9 The purpose of this paper is to describe and characterize this script and to offer a tentative history and explanation of the context of its development.10 It was a surprise to find, first in the Saint Joseph monastery in Mannanam, a series of manuscripts whose script, definitely Syriac and with a neat and elegant calligraphy, proved nevertheless very difficult to read (plates 1 and 2). It has the appearance of little separated oblique strokes with some curiously curled letters. In Mannanam also, a painted inscription on the altar perplexed us for a moment (plate 3: Insc. Mannanam 2). Other inscriptions in various Syro-malabar churches, in Muttuchira, Kaduthuruthy, Champakulam (plate 4), Pallikara, Kudamaloor, Pummattam, also presented forms of this particular script, which could also be seen in collections of manuscripts outside of Mannanam (plates 5 and 6). Taking into account the main forms present in all these documents (table 1), it is evident that there is a common structure of the script, which may be described as follows: – letters which normally join are here segmented due to the ductus; – some individual letters are even divided into separated elements: it is a broken script, something like the German ‘Fraktur’ ductus; – there are different ways of drawing the alaph, taw and the ligature between taw and alaph; – lines are very thin and embellished with little balls or curls on the top; this thinness is clearly a distinctive characteristic with respect to the Syro-Oriental script of Mesopotamia which is generally drawn with thick lines; – down-strokes have exaggerated curls; – he and taw are drawn like volutes and can be very similar. The broken structure makes this script difficult to read: segments are all of the same size and look the same: Ìet, yudh, nun, ¨ayn. The 7
Lenormant 1872–3, t. II, p. 40 and pl. VI. Briquel Chatonnet, Desreumaux and Thekeparampil 2008. The inscriptions cited below have sigla derived from this volume. 9 Briquel Chatonnet, Desreumaux and Thekeparampil 1996. 10 This paper was first presented in the IX Symposium Syriacum in Kaslik (Lebanon) in 2004. 8
408
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT
lamadh is a little longer but it is not always distinguishable. On the contrary, zayn is definitely distinct: originally shaped like a droplet in the Oriental script, only its outline is here preserved in the form of a curl and the ending has been lengthened. The semkhat is also segmented and can be formed of two completely separated and superimposed circles. Beth, mim, quph and sin have the same ductus and can only be differentiated through their open or closed aspect and the addition of small segments. The three letters dalath, kaph and ris have a basically identical form and can be differentiated through the dot and, sometimes, the segmentation of dalath and ris. The main characteristics are the particular forms of zayn and semkhat and the volute form of he and taw. This last letter also occurs in a segmented form, as is visible in MS SEERI 18 (paper, dated 1891).11 Is it possible to draw a chronological sketch of the appearance of this script? It is clearly visible in documents, manuscripts and inscriptions from the nineteenth century onwards. Yet some considerations may also be drawn from earlier manuscripts. In the Vatican Library are preserved a few manuscripts coming from Kerala and copied before the Synod of Udayamperoor (Diamper) in 1599, where it was decided to destroy Syriac heretical writings. Vat. Sir. 2 dated 1558 and Vat. Sir. 88 copied in 1557 in Angamaly have a script generally consistent with Mesopotamian tradition. They were copied at the moment when the church of Malabar separated from its mother church. Some characteristics are nonetheless specific, like the zayn in the form of a curl, clearly visible in MS Vat. Sir. 88 as well as in the inscription on the bell of Kuravilangad (dated 1584).12 This script, as we saw, was already identified in the commentary on the plates displaying Semitic scripts in the Encyclopédie of d’Alembert and Diderot. This identification had certainly been made on the basis of the oldest manuscripts which arrived in Europe and are dated from the sixteenth to the eighteenth centuries. But its characteristics become really distinctive in the nineteenth century, when it definitely becomes a different script. The most representative and numerous exemplars of this script are from the nineteenth and twentieth centuries. During this period, the manuscripts of the Mannanam collection were produced, as well as other testimonies preserved in the library of the Saint Ephrem Ecumenical Research Institute in Kottayam and in the Syro-Malabar church in Thrissur. 11 12
Briquel Chatonnet, Desreumaux and Thekeparampil 1997. See Recueil, Kuravilangad 1. 409
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT
The foundation inscription of Mannanam13 is dated 1831 CE and the invocation to Joseph (see supra) is certainly more recent. There must have been a regular development from the sixteenth century onwards, evolving gradually towards a more and more distinctive form, culminating in the nineteenth century. A more precise periodization would require the inventory of all known testimonies of this type of script and the identification of a significant number of dated manuscripts. The need for catalogues of Keralan collections is evident. Where does this script come from? Can we see characteristics which reveal the influences which were present at the origin of this particular tradition? In the context of Kerala, one first thinks, of course, of the local Indian script, Malayalam. It is a very rounded script, with wide curls. Yet it is difficult to see how such a script could have influenced the origin of a broken and segmented ductus such as Keralan Syriac. Its influence may be more evident for the letters characteristic of Keralese garshuni, but for the Syriac script in itself this possibility is not very promising. Another hypothesis would be the use of palm spathes as a support for writing, with an engraving technique. It is on this material that the ancient archives of the churches of Kerala were generally written, in Malayalam as well as in Syriac. The need to engrave the strokes transversally with regard to the fibres could have been at the origin of the way of writing with broken letters and with few horizontal strokes. But as it had no influence on Malayalam script, it is not really pertinent. More suggestive would be an influence of Latin script of the Classical period, as it could have been used by the Portuguese chancery or by Portuguese or Roman missionaries. One has especially in mind the large curls under the ligatures which are particularly developed in this script, for Òade or final nun and kaph. But what is most striking is that this script exaggerates certain of the characteristics of Syro-Oriental script, which are already attested in Mesopotamia, even if to a much lesser degree: zayn as a full loop, sometimes even lightly hollowed; semkhat with the second loop going up; segmentation of strokes, notably for dalath, {ayn and ris, with the thickening of the ends here becoming mere balls. Keralan script is thus typically Syro-Oriental, but developed to a much further extreme (table 2).14 13
Mannanam 1. Mesopotamia 16th cent.: BL Add 7178 (1544–5 CE); Kerala 16th: Vat. Sir. 88 (1557 CE); Mesopotamia 19th: the sketch relies mainly upon MS. BnF syr. 411 (1862 CE) but MSS 430 (1893 CE) and 431 (1823 CE) were also collated; Kerala 19th: different MSS from Mannanam collection. 14
410
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT
The historical context is clear. The Syro-Keralan script is typical of the Syro-Malabar community. All the above mentioned documents, inscriptions as well as manuscripts, belong to the Syro-Malabar Church, the Catholic Church of the Oriental Rite created by the sixteenth century missionaries. While Serto written in Kerala from the Syro-Orthodox missions of the patriarchate of Antioch in the seventeenth century onwards is identical to that of the documents of Tur Abdin and Syria (inscriptions of Qurillos Mar Yuyaqim)15 and while the Syro-Oriental script used by the metropolitan see in Thrissur, the Church created in 1912 by the Catholicos, conforms to the Mesopotamian model,16 it is the Syro-Malabar Church which developed its own script. This fact must be linked to its situation in relation to the community to which it was originally bound. The Antiochan Churches of Kerala regularly went back to their roots through relations with the Middle East: one thinks of the missions attested in the nineteenth century inscriptions of Qurillos Mar Yuyaqim and also of Mar Basilios and Mar Iwannis,17 for example, but one also thinks of the Syrians and Indians who brought manuscripts in Serto, as is attested by colophons and notes at the end. The Chaldean Church of the East in Thrissur was born through an initiative from Mesopotamia and always maintained links with the Middle East. By way of contrast, the Syro-Malabar Church was dramatically cut off from its Syriac roots. The funerary inscription of Alexander de Campo (Kuravilangad 3) is in this respect particularly representative. The decoration in bas-relief is typically European, as is the use of the English language. As a matter of fact, the most surprising thing in this monument is the Syriac script. It is heavily marked by Latin influence. If the Syro-Malabar Church preserved the use of Syriac language, it is through following its own evolution. Documents written in Syro-Malabar script are those which bear the greatest European influence, in the apparent style of the monuments as well as in expression and phraseology: for example, two inscriptions are written on bells,18 which is not a Syriac tradition; one can also highlight the use of the expression ‘saint’ instead of ‘mar’ when mentioning the Virgin or Saint Thomas;19 the mention of the 15 Akaparampu 1, Kandanad 2, Kothamangalam 1, Kadamattam 1, Kurupumpady 1, Rakkad 2, Mulanthuruthy 2. 16 Inscriptions of Kuriachira, Paravattani, Thiroor, Thrissur, …. 17 Inscription Kothamangalam 5. 18 Kuravilangad 1; Koratty 2. 19 Kuravilangad 1; Mulanthuruthy 1.
411
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT
‘last sacraments’20 or of a ‘privileged altar’.21 The text of the inscription of Kudamaloor is particularly revealing in this regard (plate 7): Saint Mart Mary, help of Christians, May you assist all the believers, pray and intercede before God, So that we may be delivered from all sudden fear and from all harms, Present and future. Amen
This script is thus characteristic of the culture of the Syro-Malabar Church. We have insisted mainly on the strong influence of the Latin Church, notably through Portuguese missionaries. What is striking in this context is that this Church developed nonetheless its own culture, deeply rooted in its own heritage. Its script, even if it enjoyed a different development, is a faithful offspring of the SyroOriental script; the texts copied are those of the Syriac heritage: this is attested by the fact that we could, through a quick glance at collections of manuscripts, identify two manuscripts bearing the sugitha of the Cherub and the Good Thief, which must be added to the four collated by F. Graffin.22 This typically Syriac literary genre, and particularly this sugitha, were thus widely copied by the Syro-Malabar copyists. Even through the ecclesiastical and historical separation, the Syro-Malabar script is in itself a testimony to faithfulness to Syriac heritage. Address for Correspondence: UMR 8167 — Orient et Méditerranée, CNRS, 27, rue Paul Bert, 94204 IVRY SUR SEINE cedex, France REFERENCES Briquel Chatonnet, F. 2001. ‘De l’écriture édessénienne à l’estrangelâ et au sertô’, Semitica 50, 81–90 Briquel Chatonnet, F., A. Desreumaux, and M. Debié. 2004. Les inscriptions syriaques. (Etudes syriaques 1. Paris) Briquel Chatonnet, F., A. Desreumaux, and J. Thekeparampil. 1996. ‘Données nouvelles sur la culture des communautés syriaques du Kérala’, Comptes-rendus des séances de l’Académie des Inscriptions et Belles-Lettres, 915–34 —— 1997. ‘Catalogue des manuscrits de la collection du Saint Ephrem Ecumenical Research Institute (Kottayam)’, Le Muséon 110, 383–446 —— 2008. Recueil des inscriptions syriaques 1.Kérala. (Paris, Académie des inscriptions et Belles-Lettres) 20 21 22
Kuravilangad 2 et 3. Kaduthuruthy 2. Graffin 1967. 412
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT
Desreumaux, A. 2004. ‘La paléographie des manuscrits syriaques et araméens melkites: le rôle d’Antioche’, in B. Cabouret, P.-L. Gatier, C. Saliou (eds), Antioche de Syrie. Histoire, images et traces de la ville antique, Actes du colloque de Lyon, octobre 2001 (Topoi suppl. 4, Lyon). 555–71 Graffin, F. 1967. ‘La sôghîtâ du chérubin et du larron (vendredi saint)’, L’Orient syrien XII, 481–90 Hatch, W.H.P. 1946. An Album of Syriac Dated Manuscript. (Boston) Healey, J.F. 2000. ‘The Early History of the Syriac Script. A Reassessment’, JSS 45, 55–67 Lenormant, F. 1872–3. Essai sur la propagation de l’alphabet phénicien. (Paris)
413
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT
Plate 1. Manuscript of Mannanam collection
414
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT
Plate 2. Manuscript of Mannanam collection
415
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT
Plate 3. Inscription Mannanam 2
Plate 4. Inscription Champakulam 3
416
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT
Plate 5. MS. Konatt collection
417
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT
Plate 6. MS. SEERI 25 418
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT
Plate 7. Kudamaloor Inscription
419
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT
Table 1. Examples of letters in Syro-Malabar script
420
A STUDY AND CHARACTERIZATION OF THE SYRO-MALABAR SCRIPT
Table 2. Comparison of Syro-Oriental script in Mesopotamia and Kerala
421
Journal of Semitic Studies LV/2 Autumn 2010 doi: 10.1093/jss/fgq007 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING PATRICIA CRONE
AND
ADAM SILVERSTEIN
INSTITUTE FOR ADVANCED STUDY, PRINCETON UNIVERSITY OF OXFORD
Abstract This article is a contribution to the question how far there was continuity between ancient Near Eastern and Islamic culture. It focuses on the practice of using lot-casting to allocate inheritance shares, conquered land, and official functions, and briefly surveys the history of this practice from ancient through Hellenistic to pre-Islamic times in order to examine its Islamic forms as reflected in historical and legal sources. It is argued that the evidence does suggest continuity between the ancient and the Islamic Near East, above all in the first century of the hijra, but also long thereafter, if only at a fairly low level of juristic interest. The article concludes with some general consideration of the problems involved in the study of the two disconnected periods of Near Eastern history.
In 1993 classical archaeologists made an exciting discovery at Petra. This city, once the capital of the Nabataean kingdom, thereafter a major town in the Roman province of Arabia, had long been assumed to have been destroyed in an earthquake of 551 CE, but this proved to be wrong, and in the church of St Mary the archaeologists found a cache of papyri. Completely carbonized by the fire which had destroyed the church in the early seventh century, these papyri could nonetheless be read by means of sophisticated modern techniques, and an edition is in progress.1 They contained the private archive of a major family of the city, covering the years from at least 537 to 593 CE. The papyri are in Greek but reflecting a community whose native language appears to have been Arabic, and among the papyri is a record of a division of an inheritance between three brothers. The 1 For all this, see L. Koenen, R.W. Daniel and T. Gagas, ‘Petra in the Sixth Century: the Evidence of the Carbonized Papyri’, in G. Markoe (ed.), Petra Rediscovered (New York 2003), 250–61; J. Frösen, A. Arjava and M. Lehtinen (eds), The Petra Papyri, 1 (Amman 2002). Our thanks to Glen Bowersock for referring us to this literature.
423
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
estate, which consisted of land and buildings, was divided into three equal shares and awarded to the sons by a procedure which the editors, with reference to a comparable papyrus from Nessana, take to have been lot casting.2 The Nessana papyrus, written in 562 CE, also records the division of an estate, here among four sons. The property, which consisted of buildings, farmland and personal articles, was divided into four shares of roughly equal value and awarded to the sons by lot in the presence of friends and relatives. Here, as at Petra, the parties concluded the proceedings by swearing by the Trinity and the Emperor’s health that they would abide by the division.3 The interest of this discovery to historians of the Near East lies in the fact that the procedure used for the division of the property in these two papyri is endorsed in Islamic law. It is also extremely ancient and raises the question how far, and in what way, the traditions of the ancient Near East lived on to contribute to Islamic culture. In what follows we briefly survey the attestations of lot casting as an official practice from ancient Near Eastern to Islamic times and discuss what we see as its significance. Assigning land, booty, and other property by lot In the ancient Near East (by which, for the purposes of this article, we mean the ancient Fertile Crescent), lot-casting was much used in the division of inheritances. The standard way of distributing an inheritance in Assyrian and Babylonian Mesopotamia was to divide the property into parcels and then to assign the parcels by lot to the heirs (with variations when the eldest son was privileged).4 The gods themselves are said to have divided the world by this procedure. ‘They took the box (of lots)…, cast the lots; the gods made the division’: Anu acquired the sky, Enlil the earth and Enki the bolt which bars the sea.5 This is 2 Cf. Koenen, Daniel and Gagas, ‘Petra in the Sixth Century’, 251. The papyrus (Inv. 10, P. Petra Khaled and Suha Shoman) is still unpublished. There is no explicit mention of lots in the draft edition and translation that Crone has seen, courtesy of her colleague Glen Bowersock, but the parallels with the Nessana papyrus are certainly striking. 3 C.J. Kraemer, Excavations at Nessana, III (Non-Literary Papyri), (Princeton 1958), no. 21. Compare nos. 16, 31, where lots are not mentioned. 4 A History of Ancient Near Eastern Law, ed. R. Westbrook (Leiden 2003), 1, 57f (general), 395f (Old Babylonian), 542f (middle Assyrian), 2, 939 (Neo-Babylonian). 5 Atrahasis in B.R. Foster, Before the Muses: an Anthology of Akkadian Literature 3 (Bethesda, Maryland 2005), 229; also in S. Dalley (tr.), Myths from Mesopotamia, revised ed. (Oxford 2000), 9.
424
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
famously one of the ancient Near Eastern myths that passed into Greek culture: Zeus, Poseidon and Hades divide the world among themselves by lot in the Iliad, and here as in the Akkadian myth, the three gods are brothers.6 The custom is well attested in the Bible, too.7 God Himself distributed the desolate land of Edom to wild animals by lot (Isa. 34:17), and He also instructed Moses to divide the Promised Land by lot when it had been conquered;8 Joshua duly did so.9 Micah seems to have envisaged conquest as the result of divine or angelic lot casting: he prophesied that Israel would have nobody in God’s assembly to cast lots for land for it (Mic. 2:5). Ezekiel added that the land would be divided up anew by means of arrows in the messianic age (Ezek. 45:1; 47:22). Land and captives taken by the Babylonians and Assyrians were apparently divided up in the same way: the Babylonians entered Israel’s gate and ‘cast lots for Jerusalem’ (Obad. 1:11); but God would punish the nations for having divided up his land and cast lots for his people (Joel 3:3). When the Assyrians conquered Thebes in Egypt in 663 BCE, ‘lots were cast for her nobles’ (Nahum 3:10). The Bible does not refer to inherited land being divided by this method. The idea of allocating new land by lots reappears in Jewish Hellenistic works. In Jubilees, composed by a Palestinian Jew in the second century BCE and later translated from Hebrew into Greek and Syriac, Noah divides the earth by lot between his three sons, Shem, Ham and Japheth; Canaan, the son of Ham, nonetheless settled in Shem’s portion.10 In Maccabees, Antiochus IV (175–63 BCE) is described as sending a Syrian commander with orders to wipe out the residents 6 Cf. W. Burkert, The Orientalizing Revolution (Cambridge, Mass. and London 1992), 90f; id., Babylon, Memphis, Persepolis: Eastern Contexts of Greek Culture (Cambridge, Mass. and London 2004), 36. For the subsequent history of this myth, see A. Silverstein, ‘From Atrahasis to Afridun: on the Transmission of an Ancient Near Eastern Motif to Iran’ (forthcoming). 7 Cf. Th. Gataker, On the Nature and Use of Lots2 (London 1627), modernized and updated by B. Boyle (forthcoming Exeter 2008), ch. 4, §10, an extremely learned work still worth consulting despite its age; J. Lindblom, ‘Lot-casting in the Old Testament’, Vetus Testamentum 12 (1962), 164–78. 8 Num. 26:52ff; 33:50ff (at 54); 34:13; cf. also Josh. 21:4ff; 1 Chron. 6:54ff, where priests and Levites are given certain cities to dwell in by lot. 9 Josh. 18:3ff, 10; 19:51; cf. Josephus, Antiquities, book 5, ch. 1, pars. 22, 24, 26. 10 Jubilees, 8:11ff, 10:30 (tr. O.S. Wintermute in J.H. Charlesworth [ed.], The Old Testament Pseudepigrapha [New York 1983–5]), ii; cf. also his introduction. The detail that the division was effected by lots seems to have been lost in the later Greek, Latin and Syriac translations, but it was apparently known to the Muslims, cf. Silverstein, ‘From Atrahasis to Afridun’.
425
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
of Judaea and Jerusalem and to ‘settle aliens in all their territory, and distribute their land by lot’ (I Macc. 3:36).11 Thereafter, leaving aside mere retelling of the Biblical passages, the theme of lot casting for land and/or its inhabitants seems to disappear from the indigenous sources for a long time. Lot casting must be a universal institution, and not just as a private or ad hoc method of decision making: both land and fortune are things that one is ‘allotted’ in a great many languages. In Greek, too, a piece of land was known as a lot (kleros), reflecting the fact that lots were used to distribute land when colonies were set up in order to ensure that every group received an equal share. Moveable booty was distributed in the same way,12 but whether inherited land was also divided in this way is uncertain.13 The practice is not attested at Athens14 nor, it would seem, anywhere else in Greek antiquity, except in a speech once attributed to Dio of Prusa (in Anatolia, d. c. 120), now held to be by Favorinus (d. mid-second century), a native of Arles: here we are told that ‘brothers also divide their patrimony that way’.15 Wherever the orator may have encountered the practice, it certainly sounds similar to that attested in Petra and Nessana, but it is hard to say more on the basis of a single passage. The Romans, who took over from the Greeks, also used lots for the distribution of land, both at home and in connection with the foundation of colonies.16 Moveable booty, too, was (or might be) distributed by lot.17 But the evidence relating to conquered land and 11 Settling foreigners on land confiscated from the local population was an Assyrian practice later adopted by the Achaemenids and Macedonians alike, but this passage could be inspired by Obadiah on foreigners casting lots for Jerusalem. 12 Cf. G. Wissova, Pauly’s Real-Encyclopädie der classischen Altertumswissenschaft (Stuttgart 1894–1980, hereafter Pauly-Wissova), s.v. ‘Losung’, col. 1463 (Ehrenberg); D. Asheri, Distribuzioni di terra nell’antica Grecia (Turin 1966), 13 (drawn to our attention by D. Roussel). 13 Ehrenberg categorically denies it, against earlier authors (cf. Pauly-Wissova, s.v. ‘Losung’, cols. 1478b). 14 Cf. A.R.W. Harrison, The Law of Athens: the Family and Property (Oxford 1968), ch. 5 (where the possibility is not even discussed). 15 Dio Chrysostom, (attrib.) Oratio, 64, 25, where ‘that way’ refers to ‘by lot’ (klerotas). Adduced by Gataker (Nature and Use of Lots, ch. 4, §12 (pp. 102 of the original work, where the references are given, misprinted as 46.25); cf. The Oxford Classical Dictionary3, ed. S. Hornblower and A. Spawforth (Oxford 1996), s.v. ‘Favorinus’. We are much indebted to Glen Bowersock and Christopher Jones for help with this passage. 16 Pauly-Wissova, s.v. ‘Losung’, col. 1493; D.J. Gargola, Lands, Laws, and Gods (Chapel Hill, NC 1995), 95ff. For examples, see Dionysius of Helicarnassus, Roman Antiquities, II, 16; II, 35; V, 60; X, 32.
426
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
booty peters out in the third century, and the Romans do not seem to have used this method in connection with inherited property either, except in three specific circumstances. First, in actions for the division of an inheritance or common property, or for the regulation of boundaries, it was difficult to decide who was the plaintiff and who the defendant, but the person who appealed to the law was generally considered plaintiff; to this Ulpian (d. 223) adds that if the parties appealed at the same time, the matter was usually decided by lot.18 Secondly, in 428 a law was passed which entitled the curia (city council) to claim one fourth of the estate left by a member of the council to an outsider: the estate was to be divided into four parts, of which the curia would take one by lot.19 Thirdly, in 531 Justinian ruled that when several persons had been given the option, by bequest, to pick an item such as a slave and disagreement arose, they could cast lots: the winner would pick the item and pay the others the value of their share.20 Division of the estate among the heirs by lot as the normal procedure in intestate succession does not seem to be attested. In line with this, it is mostly as a literary theme that lot drawing for land is attested in the Near Eastern literature (Jewish and Christian) from the second century onwards, with no sense of a live practice behind it. The gods cast lots again, this time for the nations of the earth, in the Pseudo-Clementine Recognitions, a Jewish Christian work of the mid-fourth century: Simon Magus, representing heresy, here argues that there are many gods, and that it was to one of the lower gods that the Jews were assigned (a gnosticizing paraphrase of Deut. 32:8f).21 In the same vein, Pirqe de Rabbi Eliezer, a Jewish work of (perhaps) the mid-eighth century, tells us that when seventy angels descended in order to confuse the nations building the Tower of Babel, they cast lots among the nations and Israel fell to God (who is not, of course, a lower God here).22 The nations are also divided up by lot in the Acts of Thomas, but now among the apostles rather 17 Cf. the story of the third-century emperor Probus in Historia Augusta, Life of Probus, 8 (ed. and tr. D. Magie [London and Cambridge, Mass], iii, 319). 18 Justinian, Digest, book 5, tit. 1, 13f (ed. and tr. T. Mommsen, P. Krueger and A. Watson [Philadelphia 1985], i, 167). 19 Justinian, Codex, 10, 35, 2; cf. 10, 35.1; A.H.M. Jones, The Later Roman Empire, 284–602 (Oxford 1964), 2, 747f. 20 Justinian, Codex, 6, 43, 3, 1; cf. id., Institutes, II, xx, 23. 21 Clement of Alexandria (attrib.), Recognitions, ii, 39 (tr. B.P. Pratten, M. Dods and T. Smith, The Writings of Tatian and Thophilus and the Clementine Recognitions [Ante-Nicene Christian Library, iii, Edinburgh 1867], 218f). 22 Pirqe de Rabbi Eliezer, tr. G. Friedlander (New York 1971), 176f.
427
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
than the gods: India fell to Thomas.23 Egypt, Ethiopia, Nubia and the Pentapolis fell to St Mark by lot (qur¨a), as a later Christian adds.24 The story of the father who divides the earth between his three sons by lot may have gone into the Persian tradition, though it is only in Ibn al-Kalbi (d. 204/819 or later) that we see it: according to him, the ancient king Faridun divided his realm (consisting of the entire world) among his three sons by writing the names of the regions on arrows and telling each son to choose an arrow.25 There does not seem to be any attestation of this method of allocating inheritance shares in Persian law or practice, however. At this point one is tempted to conclude that the ancient practice of casting lots for land, whether conquered or inherited, had disappeared, except for some special cases where Roman law applied. But it had not. The rabbis discuss it, apparently as a live institution, with reference to two or three brothers dividing an inheritance among themselves in material from second-century Sephhoris (Tiberias) in Palestine onwards;26 and it now proves to have been practised by Christians in Roman Arabia, too, at Petra and Nessana. Apparently, it was also alive in the Prophet’s Arabia, at least in connection with conquered land and booty. We are told that when the Prophet conquered Khaybar (in the year 7/628), he set aside God’s fifth by lot (using arrows); the rest of the conquered land was divided into eighteen portions and subdivided, according to one tradition, into a hundred plots of roughly the same productive capacity which he distributed to his followers by lot.27 Of the booty from the campaign against B. QurayÂa we are told that it was 23
Acts of Thomas, 1 (tr. A.F.J. Klijn [Leiden 2003], 17). Severus b. al-Muqaffa¨, ‘History of the Patriarchs of the Coptic Church of Alexandria’, ed. and tr. B. Evetts, in R. Graffin and F. Nau (eds), Patrologia Orientalis, i (Paris 1907), 105. 25 Al-™abari, Ta}rikh al-rusul wa}l-muluk, ed. M.J. de Goeje et al. (Leiden 1879– 1901), i, 226f (Ibn al-Kalbi), with further details in Silverstein, ‘From Atrahasis to Afridun’. It is not clear whether the story should be taken to reflect Persian appropriation of the theme, either directly from Mesopotamian sources or via para-Biblical literature such as Jubilees, or simply Ibn al-Kalbi’s own familiarity with the theme. 26 Babylonian Talmud, Baba Bathra, 106a. It is not found in the Jerusalem Talmud. 27 Al-Waqidi, Kitab al-maghazi, ed., M. Jones (London 1966), ii, 680, 692; al-Mawardi, Adab al-qa∂i, ed. Y.H. al-SirÌan, (Baghdad 1971), ii, 196f, no. 2715 (citing Waqidi); al-Shafi¨i in al-Bayhaqi, AÌkam al-Qur}an, ed. M.Z. al-Kawthari (Cairo 1951), 163; cf. also Ibn Sa¨d, al-™abaqat, ed. E. Sachau et al. (Leiden 1904– 40), II/1, 78, 82f; ed. Beirut, 1957–60, ii, 107, 113f (without explicit mention of lots); EI2, s.v. ‘Khaybar’, col. 1141a. 24
428
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
divided into 3072 shares, consisting partly of land and partly of moveable booty, of which a fifth was assigned to God and the rest to the Muslims by lot .28 The Muslims also cast lots for the captives taken at Badr.29 ¨Uthman (644–56) instructed Mu¨awiya to single out God’s fifth of the booty by writing ‘God’ on one of the five arrows used for their allocation.30 When ¨Ali’s followers wanted to divide the captives from the Battle of the Camel among themselves, in 36/656, ¨Ali dissuaded them by first telling them to bring the lots and next, when they brought the arrows, by asking them who might get his (spiritual) mother ¨A}isha in his lot.31 On another occasion he used the lots to divide non-Muslim booty.32 Of the Kufan ¨Abida b. Qays (d. 70s/690s) we are told that he would cast lots to assign the leftover from the division of moveable booty, such as a dirham, saying that this was how it had been done in past campaigns, but this was more controversial: the point of the report is that he was persuaded to stop, on the grounds that it was more equitable to use the dirham to buy something that could be distributed (by lot or otherwise).33 All these reports are prescriptive and hardly to be taken at face value as historical reports. Taken as literature, however, they certainly suggest that Muslims who came out of Arabia took the use of lots for the division of conquered land and booty for granted. This is corroborated by the fact that the standard word for a share of the booty was sahm (literally ‘arrow’). As regards inherited land, a Prophetic tradition reports that two men who had a dispute over inherited property submitted their case to the Prophet without having anything to prove their respective claims: he told them to cast lots and take whatever was assigned to them by this method.34 The two men are not identified as brothers, 28
Waqidi, Maghazi, ii, 522; cited in Mawardi, Adab al-qa∂i, ii, 196, no. 2714. Waqidi, Maghazi, i, 100, 107, 139. 30 al-Sarakhsi, SharÌ kitab al-siyar al-kabir li-MuÌammad b. al-Îasan al-Shaybani, ed. ∑.-D. al-Munajjid (Cairo 1957–60), iii, 889. 31 Ibn Qutayba (attrib.), al-Imama wa}l-siyasa (Cairo 1969), i, 78. 32 Al-Nuwayri, al-Bidaya wa}l-nihaya (Cairo 1975), xx, 219, where he divides the booty from IÒfahan, even including a loaf, into seven portions (one for each of the sevenths into which Kufa was divided at the time) and distributes them by lot. 33 Ibn Sa¨d, ™abaqat, ed. Sachau, vi, 62f; ed. Beirut, vi, 93. He was ¨arif (paymaster) for his tribal group. 34 Abu Dawud, Sunan (Cairo 1982), ii, 295 (K. al-qa∂a, bab fi qa∂a} al-qa∂i idha akh†a}a); cited in Ibn Qayyim al-Jawziyya, al-™uruq al-Ìukmiyya fi siyasat al-shar¨iyya, ed. N.A. al-Îamad (Mecca 1428), ii, 743 (in a useful list of Prophetic traditions on qur¨a), where further references are given. For an Imami Shi¨ite version, 29
429
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
however, and the issue is their dispute in a situation without proof rather than the normal procedure in intestate succession. We are also told that when Aban b. ¨Uthman was governor of Medina in the reign of ¨Abd al-Malik (685–705), a man manumitted the six slaves who were his only property on his deathbed; and since bequests were not allowed to exceed a third of the property, Aban drew lots and manumitted the two slaves who had the lucky draw.35 The Prophet is said to have used the same solution when two earlier Medinese manumitted six slaves who were their only property, but this is presumably a simple reworking of the Umayyad report (though it was of course the Prophetic precedent which became canonical).36 Here too the procedure diverges from that attested at Nessana and Petra, for the lots are not being used to allocate equal shares, but rather to pick out two winners. Though it seems unlikely that the inhabitants of Petra and Nessana should have been the only Arabs to use lots as the normal procedure for the division of inherited land, the practice does not seem to be attested in the material on the rise of Islam. We do however find it in classical Islamic law: here we are told that once the property had been divided into parcels representing the smallest fractions to be distributed, the heirs could draw lots among themselves for the parcels; if the estate consisted of different types of property, such as houses and land, the different types had to be divided up separately; they could not be bundled together as was done at Nessana.37 The Near East is not the only region in which lots have been used for the partition of inherited land. It crops up in Europe, too. Thomas see al-Majlisi, BiÌar al-anwar (Tehran 1357–92), civ, 324. Our thanks to Aron Zysow for help in connection with this tradition. 35 Al-Shafi¨i, ‘K. al-qur¨a’, in his Umm (Beirut 1993), viii, 5; cf. J. Schacht, The Origins of Muhammadan Jurisprudence (Oxford 1956), 201f. For further references, see Bayhaqi, AÌkam al-Qur}an, 162n. 36 Ibn Abi Shayba, al-MuÒannaf, ed. M.A. al-Nadwi (Bombay 1979–83), xiv, 158, nos. 17934f; Shafi¨i, ‘K. al-qur¨a’, Umm, viii, 5 (where one manumitter is a woman, the other an AnÒari male); further references in Bayhaqi, AÌkam al-Qur}an, 162n. 37 Mawardi, Adab al-qa∂i, ii, 194f, nos. 2709ff, cf. also 204, nos. 2746ff; alNawawi, Minhaj al-†alibin, ed. and tr. L.W.C. van den Berg (Batavia 1882–4), iii, 395ff; Ibn Rushd, Bidayat al-mujtahid, ed. M.S. al-MuÌaysin and Sh. M. Isma¨ il (Cairo 1970–4), ii, 298ff; tr. I.A. Khan Nyazee and M. Abdul Rauf (Reading 1996), ii, 319ff (both with further discussion); al-Marghinani, al-Hidaya (Cairo n.d.), iv, 46; tr. C. Hamilton, 2nd ed., (Lahore 1957), iv, 571 (K. al-qisma); al-Mawsu¨a alfiqhiyya, xxxiii (Kuwait 1995), 139 (drawn to our attention by A. Zysow); A. ¨Abd al-¨Aziz, Fiqh al-kitab wa}l-sunna (Nablus 1999), iv, 2305. 430
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
Aquinas (d. 1274) knew of it,38 and English Common Law endorsed it for the partition of land held in coparcenary from medieval down to modern times.39 The solution is likely to have commended itself wherever property had to be distributed among equally entitled claimants, and it could in principle turn up anywhere in unrelated forms. The Near Eastern forms come across as related in that all they treat lot-casting as a standard way of dividing land and other property, not simply as a last resort or special solution, as in Roman or Common law. The same may well have been true among many other peoples in ancient times, however, especially in connection with conquered land, and the Near Eastern forms are not related etymologically: the usual term for a lot in the sense of the object used in the procedure is pur(u) in Assyrian, isqu in Babylonian, goral in Hebrew (where it also stands for the share allotted), and qur¨a in Arabic, with sahm (‘arrow’) as the normal word for the lot awarded. But though they may have originated separately in pre-historic times, by the time we have literary evidence, the Near Eastern institutions stand apart from those of the neighbouring lands in that they still treat lot-casting as the standard mode of division, even in connection with inheritance law, and even, after the coming of Islam, when the heirs were awarded highly unequal shares. It is with reference to this feature that we treat them as so many members of a single family, visible in the cuneiform, Jewish, Greek papyrological, and Muslim records at different times and places thanks to a combination of local conditions and the haphazard manner in which the evidence has survived. One interesting point here is that if it had not been for the chance preservation of the two Greek papyri, one might have taken lot-casting for the distribution of land in early Islamic society and classical law to represent a case of Jewish Fortleben in Islam; for until the papyri were discovered, it was only in rabbinic texts that the practice seemed to be alive in connection with inheritance shares, and the rabbis would of course have had much to say about the Biblical use of lot 38
He describes it as a method used for the division of inheritances in case of disagreement, without giving further details (Thomas Aquinas, Commentary on the Ephesians, tr. M. L. Lamb [Albany 1966], book 1, lecture 4, ad Eph. 1:11). 39 An estate held in coparcenary was taken by several heirs as if they were a single person, for example when the deceased only left daughters (the principle being that there could only be one heir, normally the eldest son, who would take everything in the absence of a will). The use of lots for the partition of such estates is first described by Thomas de Littleton (d. 1481), cited in Gataker, Nature and Use of Lots, ch. 4, §12 (pp. 104 of the original work); it is endorsed in Great Britain, Courts, The Legal Guide, 1 (London 1839), 324f, but is now obsolete. 431
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
drawing in connection with conquered property, had they been asked. But as the papyri show, the inference would have been false. Lot-casting for the allocation of inherited property had remained a live practice in Roman Arabia, too, and also, as the accounts of the Prophet’s procedures suggest, in connection with conquered land and booty elsewhere in Arabia. What the striking similarity between Jewish and Islamic law reflects is not, in this particular case, Jewish Fortleben in Islam, but rather the shared roots of Jewish and Islamic culture in the ancient Near Eastern tradition. We seem to have here a case comparable to that of circumcision, practised by both the Jews and the Arabs (eventually Muslims), not by the one borrowing from the other, but rather by both retaining an ancient custom which had once been widespread in the Near East (notably in Egypt). In the case of circumcision, the Biblical record played a role in endowing the old Arabian practice with a new religious meaning. There is no suggestion that it did so in the case of lot-casting. It is because the Arabs were apt to preserve ancient practices also recorded in the Jewish scripture that Old Testament scholars (Wellhausen prominent among them) used to study Arabia with such interest, with special attention to the bedouin because the ancient Israelites had been pastoralists. It is the townsmen of Arabia that we see at work at Nessana and Petra, but the bedouin continued to furnish parallels into modern times: Musil reports that in what he called Arabia Petraea (former Roman Arabia) agricultural land belonging to the whole tribe would be divided into fields of equal size every year and distributed among the families or tribal groups by lot.40 The continuity with ancient Near Eastern practice in Arabia should presumably be related to the forbidding nature of the peninsula. Difficult to conquer and colonize, it was the only region of the Near East to escape a millennium of Greek, Roman or Persian domination, though parts of it (including Petra and Nessana) fell under foreign rule for periods ranging from centuries to decades. We have to stress, however, that the Jews and the Arabs may not have been the only inhabitants of the Near East to use lot-casting for the division of inherited property in late antiquity, for on the Jewish side it is in rabbinic literature that it is attested, not in the Bible. This suggests that what the rabbis discussed was a practice they shared with their neighbours, or in other words that in this particular case the rabbinic literature should not be seen as evidence 40
A. Musil, Arabia Petraea (Vienna 1907–8), 3, 294. 432
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
for the Jews alone, but also for the larger Aramaic culture of which they formed part. Choosing people by lot It was not only in connection with the distribution of land and its inhabitants that lot-casting was used in the ancient Near East; people were selected for a wide variety of functions by that method, too. The Assyrians used sortition to choose the annual occupant of the ‘office of the year eponym’, a dignitary who had the privilege of having a calendar year named after him.41 The king himself never seems to have been chosen by lot in ancient Mesopotamia,42 nor do priests. But the Bible tells us that Saul was chosen as king by lot,43 and the Samaritan Chronicle has it that the first Samaritan king was chosen by the same method.44 In Pseudo-Philo (c. 50–150) the Israelites also choose Kenaz as their leader against the Philistines by lot, directed by an angel, and repeatedly tried the same method to find a successor to Phinehas without success.45 By Roman times succession to the high priesthood of the Jews had come to be decided in the same way, with explicit reference to ancient practice.46 In Biblical times, lots were also used to single out the groups and individuals who were to serve as temple musicians and gate keepers in ancient Israel (1 Chron. 24:5ff, 25:8ff, 26:13f), and to allocate rotating responsibilities such as the serving as priests and providing wood offering to the temple (Neh. 10:35).47 Zachariah was a priest chosen by lot to officiate at a particular time (Luke, 1:8f ),48 and Peter found a replacement for the apostle Judas by selecting two men and then casting lots (Acts 1:23–6), a procedure which was to be imitated 41
W.W. Hallo, ‘The First Purim’, The Biblical Archaeologist 46 (1983), 19f. M.T. Larsen, ‘The City and its King’, in Le Palais et le royauté, ed. P. Garelli (Paris 1971), 298f (against Oppenheim). 43 1 Sam. 10:19–21. But God’s answer in v. 22 must have been given by a seer or prophet, cf. J. Lindblom, ‘Lot-casting in the Old Testament’, 165n. 44 J. Macdonald, The Samaritan Chronicle, no. II (Berlin 1969), 99. 45 Latin text (originally Hebrew) and English translation in H. Jacobson, A Commentary on Pseudo-Philo’s Liber Antiquitatum Biblicarum (Leiden 1996), 25:1f; 49:1. 46 Josephus, Wars, book 4, ch. 3, pars. 7f. 47 Cf. Josephus’ amplifications, Antiquities, book 7, ch. 14, par. 7. 48 Compare Protoevangelium of James 24:4 in W. Schneemelcher and R. McL. Wilson (eds), New Testament Apocrypha, 1 (Cambridge 1991), 437, where Zachariah in his turn is replaced by Simeon by lot. 42
433
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
by later Christians in the Near East and the West alike.49 Indeed, the word ‘clergy’ is derived from kleros, ‘lot’, the clergy being people allocated to God.50 Lot casting may also have been used to assist the decision who should be admitted as new members of the community at Qumran, but this is disputed.51 Again, the Greeks and the Romans had similar practices. In Greece lot-casting was used for the selection of magistrates, especially in democracies, where it was of fundamental importance as an egalitarian device.52 The Romans would distribute functions among magistrates already chosen by sortition. Consuls and praetors, for example, would cast lots among themselves to determine the assignment of campaigns and provinces (‘What if the casting of lots had allocated you Africans or Spaniards or Gauls to rule over?’, as Cicero asked his brother, then governor of Asia);53 lots were also used to determine voting order and other sequences, to choose officials for special tasks, and in diverse other connections, including (at least on one occasion) that of selecting recruits.54 We even hear of bandits who reputedly used lots to decide which members of the gang should labour or serve the 49
For thirteenth-century nuns choosing an apostle (as patron saint) by lot, see G.G. Coulton (tr.), Life in the Middle Ages, 1 (Cambridge 1928), 69f. Thomas Aquinas held that lot-casting could not be used for ecclesiastical office after the arrival of the Holy Spirit (Commentary on the Ephesians, book I, lecture 4), but the Mennonites of today choose priests by lot (personal communication from Christopher Melchert). In the Middle East known to T. Fahd, monks would decide by lot which novices should receive the habit (EI2, Leiden 1956–2004, s.v. ‘Èur¨a’). 50 This too is discussed in Pauly-Wissova, ‘Losung’, cols. 1466f (and indeed by Aquinas, loc. cit.). 51 Lots figure prominently in the Dead Sea scrolls, but almost exclusively in a metaphorical sense (Y. Licht, ‘The Term Goral in the Writings of the Judean Desert Cult’, Beth Miqra 1, 1956, 90–9 [Hebrew]). For the question of its use in admissions, see W.A. Beardslee, ‘The Casting of Lots at Qumran and in the Book of Acts’, Novum Testamentum 4 (1960), 245–52; S.J. Pfann, ‘The Essene Yearly Renewal Ceremony and the Baptism of Repentance’, in D. Parry and E. Ulrich (eds) The Provo International Conference on the Dead Sea Scrolls (Leiden 1999), 337–52; P.S. Alexander, ‘Predestination and Free Will in the Theology of the Dead Sea Scrolls’ in J.M.G. Barday and S.J. Gathercole (eds) Divine and Human Agency in Paul and His Cultural Environment (London 2007), 27–49. 52 Oxford Classical Dictionary, s.v. ‘sortition’’; Pauly-Wissova, s.v ‘Losung’, cols. 1475ff. 53 Cicero, Ad Quintum fratrem, 1, 9, 27. 54 Pauly-Wissova, s.v. ‘Losung’, cols. 1494ff; Gargola, Land, Laws, and Gods, 95; R.J.A. Talbert, The Senate of Imperial Rome (Princeton 1984), 61, 139, 144, 207f, 347–53 (drawn to our attention by Nathan Rosenstein); N. Rosenstein, ‘Sorting out the Lot in Republican Rome’, American Journal of Philology 116, 1995, 43–75, with the recruits at 44, n7. 434
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
others,55 but whether this can count as an example of official use is another question. On the Greek and Roman side, the official use of lots for the allocation of office and functions seems to have petered out by late antique times, and the evidence is thin on the Near Eastern side as well. Rabbinic literature does admittedly abound in discussions of temple duties and other Old Testament institutions, but it is all academic. Choosing priests, monks and other ecclesiastical personnel by lot is more likely to have continued among the Christians, thanks to the precedent set by Peter’s choice of Matthew by this method. It is reflected in the Protoevangelium of James, where Mary is chosen by lot for the privilege of weaving a particular item,56 but the only attestation relating to real life that we know is modern.57 This undoubtedly reflects our ignorance of the vast mass of relevant Syriac literature. Once again there is some ambivalent evidence on the Iranian side:58 in the account of Arda Viraz} journey to heaven and hell, Arda Viraz is chosen for the journey by three lances (nezag) which are thrown at him. But this procedure was in the nature of an ordeal rather than lot casting, for the lances were meant to confirm or deny the suitability of a man already chosen; there were no other candidates.59 One would be inclined to conclude that the once prevalent practice of choosing people for high office and other functions by lot had died out. Again, however, the practice must have survived in Arabia. Unfortunately, there does not seem to be any documentary evidence for this. Three pre-Islamic inscriptions, one from Allat’s temple at Palmyra and two from Yemen, do refer to lot casting, but they probably refer to divination.60 We are told, however, that the pre-Islamic Quraysh would choose men to lead them in war by lot and accept the candi55 B. Shaw, ‘The Bandit’, in A. Giardina (ed.), The Romans, tr. L.G. Cochrane (Chicago and London 1993), 330 (with ref. to Apuleius, Metamorphoses, 4.8). 56 Protoevangelium of James 10:2 (in Schneemelcher and Wilson, New Testament Apocrypha, 1, 430). 57 Cf. Fahd, above, note 49. 58 The Persians are envisaged as casting lots to fix the day on which the Jews were to be killed in the Book of Esther (3:7). The institution credited to them here is Akkadian, but whether it can be inferred that the Persians had adopted it is unclear. 59 Cf. P. Gignoux, ‘Une ordalie par les lances en Iran’, Revue de l’Histoire des Religions 200, 1983, 155–61. The procedure is construed as lot-casting in S. Shaked, ‘Quests and Visionary Journeys in Sasanian Iran’, in J. Assmann and G.G. Stroumsa (eds), Transformations of the Inner Self in Ancient Religions (Leiden 1999), 73. 60 R.G. Hoyland, Arabia and the Arabs (London 2001), 156.
435
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
date even if he was a minor or very old;61 and the terms qari ¨ and maqru¨ (chosen by lot) were used in the sense of chief, leader and person chosen.62 In line with this we later hear of lot-casting for the selection of caliphs. The Christian astrologer, Theophilus of Edessa, active under the caliph al-Mahdi (d. 169/785), tells us that when Yazid I died, the future Marwan I (64/684–5) proposed to solve the succession dispute which ensued by drawing lots; this was apparently agreed, but when Marwan’s name came up, his rival al-ΔaÌÌak b. Qays refused to accept the result, so the two of them fought it out at Marj Rahi†.63 Al-JaÌi also knew of lot-casting in connection with the choice of caliphs, though he did not think it was necessary: in his view, the rightful claimant would always be known without the need for formal procedures, just as everyone knew who was the most generous man or the best horseman among Qays in the Jahiliyya without discussion of their merits or shura or casting lots (al-iqra¨ wa}l-musahama).64 Lot-casting was endorsed by some jurists for situations in which two candidates for the caliphate were equally qualified, or when two of them had come to be elected by some mishap, but others disagreed.65 ‘In our opinion, lots are required by the law to spare people’s feelings, not to establish rights’ (li-ta†yib al-qulub duna ithbat al-Ìuquq), as alNasafi (d. 508/1114) observed with reference to the second situation, meaning that it could only be used for the random distribution of things to which people had a lawful claim, not to pick out winners.66 No caliph actually seems to have been chosen by this method, but 61 Ibn al-Jawzi, al-MuntaÂam (Beirut 1992–3), ii, 217f, apparently from Ibn al-Kalbi. 62 Ibn ManÂur, Lisan al-¨arab, Beirut 1955–6; Murta∂a al-Zabidi, Taj al-¨arus, ed. ¨A. Shiri (Beirut 1994), both s.v. ‘qr¨ ’. 63 Theophilus as reconstituted by R.G. Hoyland, Seeing Islam as Others Saw it (Princeton 1997), 647, cf. 400ff. 64 Al-JaÌiÂ, al-¨Uthmaniyya, ed. ¨A.-S. M. Harun (Cairo 1955), 266. In Ibn ™awus, FatÌ al-abwab bayna dhawi }l-albab wa-bayna rabb al-arbab fi }l-istikharat, ed. Kh. al-Khaffaf (Beirut 1989), 267ff (chs 20–1), musahama consists of drawing lots from paper with names written on them, whereas a qur¨a is an object such as a pebble or a rosary bead, but it was not necessarily so in JaÌiÂ} time. (Our thanks to Etan Kohlberg for drawing Ibn ™awus’ work to our attention.) 65 Abu Ya¨la, al-AÌkam al-sul†aniyya, ed. M.Kh. al-Fiqi, second printing, Cairo 1966, 25 (where lot-casting is prescribed in the first situation and is one out of two acceptable views in the second); al-Mawardi, al-AÌkam al-sul†aniyya, ed. M.J. alÎadithi (Baghdad 2001), 60.1, 62.–6; tr. W.H. Wahba (Reading 1996), 6, 8, on unnamed jurists (without verdict on the first situation, but with arguments against lot-casting in the second). 66 Abu}l-Mu¨in al-Nasafi, TabÒirat al-adilla, ii, 826f, against al-Qalanisi and al-Ka¨bi. His position is Îanafi, cf. below.
436
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
much later we hear of an Ottoman grand vizier who was chosen by lot (drawn from pieces of paper with the names of candidates written on them).67 This was in 1204/1789f, at the beginning of the reign of Selim III, and its relevance to our present concerns is uncertain. There seems to have been a tradition in Arabia of choosing people for other functions by lot as well. The Prophet is said to have decided which wife should accompany him on his travels by lot-casting;68 the Medinese are said to have used lots to determine who should have the privilege of hosting the Prophet;69 ¨Ali is credited with using lots to settle a case in Yemen in which three men denied paternity of a child that any one of them could have fathered.70 ¨Umar II is said to have included the wives and children of the soldiers in the diwan and cast lots to decide who should receive a hundred and who forty dirhams, i.e. from the income of the immoveable booty which was paid out as stipends.71 All these examples refer to men in official positions, but hardly to lot-casting as a regular, public institution (though all decisions recorded for the Prophet were to assume that character). We do, however, encounter lot-casting as a regular institution in connection with mobilisation. When ¨Uthman permitted Mu¨awiya to conduct campaigns by sea, he stipulated that Mu¨awiya was not to select the men himself or cast lots among them (la tantakhib al-nas wa-la tuqri¨ baynahum), but rather to let them decide themselves whether to go.72 Sortition was apparently among the methods normally used in the army to decide who was to go on duty. Of a Syrian soldier who went on annual summer campaigns in the Byzantine empire in the reign of Mu¨awiya we are told that he had a bad dream predicting that he would be the killer of an eminent Medinese and thereby doom himself to Hell; 67 Ahmed Cevdet Pasha, Tarih-i Cevdet (Dersaadet 1309), v, 18 (on Ruscuklu Hasan Pasha). We owe this reference to ≤ükrü Hanioglu. 68 ™abari, Ta}rikh, i, 1519. Compare Babylonian Talmud, Shabbath, 149b, on how Nebuchadnezzar would cast lots to decide which of his recently acquired (male) captives to have sexual relations with. 69 Ibn Sa¨d, ™abaqat, ed. Sachau, iii/1, 288; ed. Beirut, 396 (s.v. ‘¨Uthman b. Ma¨un’). 70 He imposed two thirds of the blood-money (for the child) on the man picked out as the father, presumably on the reasoning that he had caused the other two men to lose a third of a child each. The Prophet found this solution uproariously funny (Ibn Îanbal, Musnad, Cairo 1313, iv, 373; Waki¨, Akhbar al-qu∂aÌ, ed. ¨A.-¨A.M. al-Maraghi, Cairo 1947–50, i, 91). For a variant involving two men and a slave girl, see al-Majlisi, BiÌar, xl, 244f, cf. also civ, 63. 71 ™abari, Ta}rikh, ii, 1367. 72 Ibid., i, 2824.
437
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
when people were chosen by lot for Yazid I’s campaign against Medina (∂uriba qur¨at ba¨th al-Madina) in 63/682f, this man had the misfortune to be selected (fa-aÒabatni al-qur¨a).73 In these two examples it is the authorities who have used lots, but there are also stories in which it is the soldiers themselves who do so, some set in the Prophet’s time. A Medinese desirous of martyrdom told the Prophet that he had missed the battle of Badr because he drew lots with his son to decide which one of them should go and it was his son’s lot that had come out (kharaja sahmuhu).74 Qur¨a was used to select eighty men from a group of volunteers in connection with another expedition.75 In these stories enlistment is envisaged as voluntary, but only one man can go because one has to stay behind to look after the family, or only eighty men are needed of the many who have volunteered. In another Ìadith, Abu Hurayra invokes the example of a man who goes on campaign with some people and whose lot does not come out (lam yakhruj sahmuhu) because he has not said ‘amen’;76 here the volunteers are already on campaign and the question is who should go on a particular expedition in the course of it. We also hear of men in the mid-Umayyad period who would cast lots among themselves when they were called up to decide who should actually go; those who won would stay at home in return for payment of a sum known as ja¨a}il .77 Here the assumption seems to be that a particular tribal group would be told to supply a specified number of men and that the men could decide for themselves whom to send: they all wanted to stay at home rather than to be martyred. The Ottomans provide a much later parallel for the use of lots in connection with military service, too. Al-Majlisi records that when ¨Umar Pasha (1764– 76), Mamluk governor of Iraq on behalf of the Ottomans, arrived, he ‘imposed harsh lot-casting on them (ishtadda ¨alayhim al-qur¨a)’ and took soldiers from villages and the amÒar, high and low, learned 73 74
Ibn Qutayba, al-Imama wa}l-siyasa, i, 215f. Waqidi, Maghazi, i, 212, on Sa¨d b. Khaythama; cited in Majlisi, BiÌar, xx,
125. 75
Ibid., xxi, 77 (on ghuzat al-silsila). Al-Haythami, Majma¨ al-zawa}id (Beirut 1982), ii, 113 (K. al-Òalah, bab alta}min). 77 M. Bonner, ‘Ja¨a}il and Holy War in Early Islam’, Der Islam 68, 1991, 47f, with reference to T. Nöldeke, Delectus Carminum Arabicorum (Wiesbaden 1933), 77, and other sources where the poet is said to have been called up by Mu¨awiya’s governor of Kufa (but the campaigns in Khwarizm only started in the governorship of Qutayba); ™abari, Ta}rikh, ii, 1029, without the poem, where the expedition is despatched by ¨Abd al-Malik. Exactly how the procedure worked is not clear. 76
438
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
and ignorant, and ¨Alids and others alike.78 When MuÌammad ¨Ali (1805–48) introduced conscription in Egypt, qur¨a was apparently also meant to be used;79 the Ottoman conscription system of 1848 was actually known as Qur¨a niÂamnamesi (regulation on the drawing of lots);80 and lots were also used to draft soldiers in Egypt under Khedive Ismail (1863–79).81 We abstain from the attempt to account for the Ottoman examples. The point of interest to us is that in the period with which we can claim some familiarity (from the rise of Islam to the Mongols), references to the use of lots in an official context are clustered in the first century, where the Prophet, the Rashidun and the Umayyads form a continuum, to fall off rapidly thereafter, except in connection with legal procedure. No doubt more will turn up, but it seems reasonable to infer that the official use of lot-casting for the selection of persons was a practice rooted in Arabia. The Qur}an and the Law Lot-casting figures in the Qur}an, but only as a literary theme, not as a live practice or an object of legislation. Two passages are relevant. The first is Q. 3:44, concerned with Mary. Much of what the Qur} an has to say about her life reflects the Protoevangelium of James, a work written in Greek some time after 150, widely read in the Christian Near East, and translated into Syriac in the sixth century. In this work we read that Mary grew up in the temple and that the priests decided to marry her off when she was twelve years old, lest she pollute the temple by having periods (this passage is strikingly reminiscent of the story of the ‘Mouse-Maiden’ in the Pañcatranta/Kalila 78
al-Majlisi, BiÌar, liii, 331. Kh. Fahmy, ‘The Nation and its Deserters: Conscription in Mehmed Ali’s Egypt’, in E.J. Zürcher (ed.), Arming the State: Military Conscription in the Middle East and Central Asia 1775–1925 (London and New York 1999), 67, citing Sir John Bowring’s report of 1840 on how men would be seized without any order, arrangement, inscription ‘or lot-drawing’. 80 E.J. Zürcher, ‘The Ottoman Conscription System in Theory and Practice, 1844–1918’, in Zürcher, Arming the State, 82f, with a description of the system. Prof. ≤ükrü Hanioglu, to whom we are much indebted for help on Ottoman questions, tells us that the draw of lots for conscription was called qur¨a-i sher¨iyye in the vernacular, military service being a religious duty. According to Fahd, qur¨aya girmek came to mean ‘reaching the age of military service’ (EI2, s.v. ‘Èur¨a’). See also Granquist in Lindblom, ‘Lot-casting’, 169n, where the system is slightly different from that described by Zürcher. 81 J.P. Dunn, Khedive Ismail’s Army (New York 2005), 43. 79
439
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
wa-Dimna). The priests assemble the widowers of the people and tell them to bring a rod, and when a dove flies out of Joseph’s rod, they assign Mary to him.82 In other words, it is a miracle that singles out Joseph as her husband, not lots. But lots appear in other stories in the Protoevangelium, and on a later occasion it even mentions that Joseph himself had won his bride by lot.83 The Qur}an, on the other hand, briefly declares that ‘you (sg.) were not there when they threw their rods (to determine) which of them should take care of Mary’ (idh yulquna aqlamahum ayyuhum yakfulu Maryama, 3:44), seemingly referring to the version with the miracle (and presenting the contest as over kafala, care, rather than marriage).84 But the exegetes generally understood the rods as ‘the arrows with which the lot-drawers (almustahimun) from among the sons of Israel cast lots (istahama) for the guardianship of Mary’, as al-™abari puts it.85 The second passage is in the story of how Jonah came to be thrown overboard from the ship on which he was travelling. In the Bible, Jonah is identified by lots as the sinner on whose account the storm is sent (Jon. 1:7). In the Qur}an there is no reference to the storm, the ship is simply overloaded, so lots are cast to determine who should be jettisoned; but Jonah is a guilty party here too, and this does seem to be what the lots indicate: he has run away (abaqa) and behaved shamefully (wa-huwa mulim), and when he cast lots, his plea was rebutted (fa-saÌama fa-kana min al-mudÌi∂in) (37: 140–2). The fact that lot-casting is mentioned in the Qur}an in connection with venerable figures meant that the procedure had excellent legitimation. It also generated some stories in which MuÌammad’s kafala, like Mary’s, is decided by lots.86 But since it was only in accounts of 82 Protoevangelium of James, 8:2–9:1, in Schneemelcher and Wilson, New Testament Apocrypha 1, 429f; in the Indian story it is her father’s house that should not be polluted (cf. the six versions of the passage, including the old Syriac, in F. de Blois, Burzoy’s Voyage to India and the Origin of the Book of Kalila wa-Dimnah [London 1990], 7ff). 83 Protoevangelium of James, 19:1, in Schneemelcher and Wilson, New Testament Apocrypha 1, 434 (Tischendorf’s version). 84 The Protoevangelium thinks of Mary as a perpetual virgin and accounts for Jesus’ brothers by casting Joseph as an old widower with children by his first marriage when he wins Mary. In the Qur}an, the old man who wins her is Zakariyya (cf. 3:37), the father of John the Baptist, and her husband has completely disappeared, an interesting development which must tell us something about the religious milieu reflected in the Qur}an. 85 ™abari, Tafsir, ad 3:44; similarly Fakhr al-Din al-Razi, Tafsir, ad loc., and Ibn ¨Arabi, below, note 87.
440
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
earlier religious communities that the procedure is mentioned, it did not shape Islamic law on this topic. As Ibn al-¨Arabi observed, in the earlier shari ¨a, lot-casting had been sanctioned for general use, whereas it was only used in specific cases in Islamic law, and not in connection with kafala; nor would using lots to throw a man overboard be acceptable under Islamic rules, as both he and others pointed out.87 As Ibn al-¨Arabi noted, lot-casting did, however, remain acceptable in Islam in other contexts. First, as mentioned already, the jurists accepted that inheritances (and other joint property) could be allocated by lot. They seem to have done so without any controversy, and the method is still prescribed for the partition of joint property in the Ottoman Majalla.88 It is a remarkable example of continuity from the ancient to the modern Near East, if only at a fairly low level of juristic interest. That booty could be allocated among equals by lot seems also to have been widely accepted, at least as long as it was only a method of allocation of the appropriate shares rather than the assignation of things left over. The imam was charged with concern for the feelings of his subjects (mura¨at qulub al-ra¨iyya) and avoidance of preferential treatment, as al-Sarakhsi explained; for this reason division of the booty was done by qur¨a, both in connection with the fifth set aside for the imam and for the distribution of the remaining four fifths. The four fifths would also be assigned to the pay-masters (¨urafa’ ) by lot, and each ¨arif would divide the portion assigned to him among the men of whom he was in charge by qur¨a, too, he said (using terminology from the Umayyad period). He adduced the Prophet’s choice of a wife to accompany him on his travels by lot as the paradigmatic case in that the Prophet had used lots to spare their feelings (ta†yiban li-qulubihinna).89 In connection with partition, the Maliki Ibn Rushd also tells us that the jurists accepted lot-casting ta†yiban li-nufus al-mutaqasimin.90 It was on the same principle that the Shafi¨ites and others accepted that one could choose prayer leaders, naqibs Al-Baladhuri, Ansab al-ashraf, ed. M. Îamidallah (Cairo 1959), 85. Ibn ¨Arabi, AÌkam al-Qur}an, ed. ¨A.M. al-Bijawi (Cairo 1378/1958), iv, 1610f; al-Qur†ubi, al-Jami¨ li-aÌkam al-Qur}an (Cairo 1967), xv, 126; and, before both of them, al-JaÒÒaÒ, AÌkam al-Qur}an (Beirut 1994), iii, 496f; all ad 37:141. 88 Al-Majalla (Mecelle-yi ahkam-i ¨adliye), book x, articles 1151, 1156; cf. also 1180 (available in English at www.iiu.edu.my/deed/lawbase/al-majalle). 89 Sarakhsi, Siyar, iii, 889f. On his handling of the Prophet’s precedent in connection with wives, see also below, note 98. 90 Ibn Rushd, Bidaya, ii, 299.2; tr. Nyazee, i, 320 (translated ‘for the satisfaction of the persons participating in the partition’). 86
87
441
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
and other persons by sortition when the candidates were equally entitled:91 the contenders had to be mustawina fi }l-Ìujja, as al-Shafi¨i said.92 There were situations in which some jurists, above all the Îanafis, deemed lot-casting to amount to gambling (qimar), however. If a man manumitted slaves worth more than a third of his property in death, sickness or by will, the Shafi¨is, Îanbalis, Malikis and Imamis would draw lots and manumit however many could be accommodated within the third in accordance with the Prophetic Ìadith, but the Îanafis held that all the slaves should be set free and obliged to work until they had paid off the value of the unmanumitted parts.93 Similarly, when two men claimed ownership of some property and adduced equally valid proof, the Shafi¨is, Îanbalis and Imamis accepted (among various other solutions) that one could cast lots and give the disputed property to the winner, directly or by having him take the oath which settled the matter; there were Ìadiths in which the Prophet and ¨Ali did so. But the Îanafis (and Malikis) would divide the property, arguing that the Ìadiths dated from the period before the prohibition of gambling.94 There were also traditions in which the Prophet cast lots to decide who should swear first (in the situation in which two parties raise claims against 91 Mawardi, AÌkam, 273 (niqaba), 278 (leadership of prayer), 532.ult. (order on the military roll), 589.ult (retaliation); tr. Wahba, 109, 113, 224, 254; Nawawi, Minhaj, iii, 99f, 102 (Ìa∂ana), 119f (retaliation), 379 (admission to the court room). The Malikis and Îanbalites also accept lot-casting in such situations (Mawsu¨a, xvii, 138ff, 148f ), and the Imami Shi¨ites list many more (see Îusayn al-Karimi alQummi, Qa¨idat al-qur¨a, Qum 1420 [1999], 20f; MuÌammad Jawad Ash¨ari, Barrasi-yi Ìujjiyat-i qur¨a (Qum 1382 [2003]), 106ff, 120). 92 Shafi¨i, ‘K. al-qur¨a’, Umm, viii, 3; Bayhaqi, AÌkam al-qur}an, 158. 93 ¨Abd al-Wahhab b. ¨Ali al-Baghdadi, al-Ishraf ¨ala masa}il al-khilaf, ed. Î. ™ahir (Beirut 1999), ii, 990 (no. 2005); al-™usi, al-Nihaya, Beirut 1970, 105ff; Ash¨ari, Barrasi, 109. Some Malikis rejected qur¨a if the slaves had been freed in death sickness (Ibn Rushd, Bidaya, ii, 405f (K. al-¨itq); tr. Nyazee, ii, 450f). Compare the case of a man who divorces one of his four wives and marries a fifth in death sickness without it being known which of the four he had in mind: YaÌya b. Aktham (eventually classified as a Îanafi) would let all five inherit and observe the ¨idda, the Îanbalis and some Imamis would cast lots for the one who had been divorced (Ibn Qayyim al-Jawziyya, al-™uruq al-Ìukmiyya fi siyasat al-shar¨iyya, ed. N.A. al-Îamad (Mecca 1428), ii, 744, 789; Karimi, Qa¨ida, 21; Ash¨ari, Barrasi, 111. Cf. Mawsu¨a, xvii, for the Maliki and Shafi¨ite solutions. 94 Al-Sarakhsi, al-Mabsu† (Beirut 2001), xvii, 49f; al-KhaÒÒaf, Adab al-qa∂i, 391, no. 452; Baghdadi, Ishraf, ii, 983 (no. 1993); Nawawi, Minhaj, iii, 440ff; Mawsu¨a, xxxiii, 142f; ™usi, Nihaya, 343f; Karimi, Qa ¨ida, 105ff; Ash¨ari, Barrasi, 108; cf. F. Rosenthal, Gambling in Islam (Leiden 1975).
442
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
each other and both have to swear), but the Îanafis held that the judge should decide in most such situations.95 The reasoning is clearly that lots could not be used in situations in which all claimants were entitled, but only some could be satisfied in full, or only one person was entitled, but nobody knew who that person was: picking out the lucky winners by lots amounted to gambling with their legal rights. Al-Shafi¨i also had reservations about lot-casting in the latter case, but Îanbalites endorsed it in both.96 Those who claimed that qur¨a amounted to gambling and had been abrogated were ignorant, foul, or positively evil people, Ibn Îanbal said; they had the temerity to label a Prophetic decision qimar.97 Polemicists who credit Abu Îanifa with the statement al-qur¨a qimar typically cast him as rejecting the use of lots altogether. The Imamis are among them.98 According to them, sortition was acceptable in all matters unknown (kullu majhul fa-fihi }l-qur¨a), a principle they defend to this day.99 Attitudes to lots In the Old and New Testaments, too, all forms of lot-casting are consistently envisaged as an appeal to the divine: God could see differences hidden to the human eye; there are passages in which the outcome of lot casting is explicitly equated with His will (1 Sam. 10:24; Prov. 16:33; Acts 1:23–6). The Greeks may once have thought in similar terms, though it has been argued that they never did so 95
Cf. Mawsu¨a, xxxiii, 147f. Cf. Ibn Taymiyya, ∑iÌÌat uÒul madhhab ahl al-Madina (Beirut n.d. [1980?]), 85f. Our thanks to Aron Zysow for drawing this work to our attention. 97 Ibn Qayyim al-Jawziyya, ™uruq, 742, 744f, 747f. 98 Thus Karimi, Qa¨ida, 18. He later notes that the Mawsu¨a shows Abu Îanifa to have accepted qur¨a in general, only to cite a barrage of stories in which Abu Îanifa rejects the Prophet’s precedent, including the latter’s use of qur¨a for choosing a wife to accompany him on a journey (pp. 101f). Since the Prophet’s use of lots in connection with wives is a situation in which the procedure was used to pick a winner, Abu Îanifa may well in fact have disliked this Ìadith, but according to Sarakhsi (above, note 89), none of the wives had any legal right to accompany him (whereas the slaves did have a legal right to such freedom as the estate allowed by virtue of the bequest). 99 ™usi, Nihaya, 345f; Majlisi, BiÌar, x, 203; xiv, 325; Ibn ™awus, FatÌ alabwab, 272 (citing ™usi); Ash¨ari, Barrasi, 106; MuÌammad Îusayn Fa∂l Allah, al-Qur¨a wa}l-istikhara (Beirut 1417/1997), 24ff, against Abu Îanifa, Ibn Abi Layla and Ibn Shubruma at 27, 29; Karimi, Qa¨idat al-qur¨a, 34f. 96
443
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
in connection with divisory lot casting.100 Divisory lot-casting is an expression coined by Thomas Aquinas for the use of lots to determine who should have or do what, as opposed to consultative and divinatory lot-casting, used to decide what to do and to obtain information about the future respectively.101 From ancient times to late antiquity the Greeks seem to have envisaged lot-casting of the divisory kind as a matter of chance, and the same is true of the Romans.102 It was a matter of fortuna, as Justinian called it in his legislation.103 Their attitude affected their Hellenized Near Eastern subjects. Josephus, for example, famously tells how the rebels at Masada chose ten men by lot to kill the rest of them, and thereafter each other,104 and how he himself had used lots to decide who, of his small band about to be captured by the Romans, should kill whom first (he surrendered as one of the last to survive). He too seems to think of the outcome as a matter of luck. He does put it to the reader that his own survival could have been due to God’s providence rather than to chance, but it sounds like mere self-justification.105 The Sunni jurists generally seem to have thought of divisory lot-casting (qur¨a) in much the same sober vein as their Greek and Hellenized predecessors. Their attitudes must of course have varied in place and time and we cannot claim to have studied them in any detail, but unlike Aquinas who (invoking Augustine) identified all sortition as ‘a questioning concerning realities whose occurrence depends on the divine will’, they convey little impression of seeing the divisory form as an appeal to God. They make no attempt to distinguish it from, or relate it to, consultation (istikhara) or divination (istiqsam, kihana), apparently taking it for granted that they were 100
Cf. N.D. Fustel de Coulanges, The Ancient City, New York n.d. (originally published Paris 1864), 182f (book III, ch. x); Pauly-Wissova, s.v. ‘Losung’, cols. 1461ff, mostly disagreeing with Fustel de Coulanges and claiming that the Greeks distinguished between the lot as a divine oracle and as a tool of equality from the start. 101 Thomas Aquinas, Commentary on the Ephesians, book 1, lecture 4, citing Proverbs 18:18 (‘The lots put an end to dispute’) in justification of the first. He put lot-casting for the selection of people in the consultative rather than the divisory category. For other classifications, see Gataker, Use and Nature of Lots, ch. 3. 102 Rosenstein, ‘Sorting out the Lot’, esp. 51. 103 Cf. Justinian, above, note 20; also Favorinus (Ps.-Dio), above, note 15. Fortuna had once been a goddess, but only in the sense that everything beyond human control could be seen as divine. 104 Josephus, War, book 7, ch. 9, par. 1; cf. Y. Yadin, Masada: Herod’s Fortress and the Zealots’ Last Stand (London 1966), 201. 105 Josephus, War, book 3, ch. 8, par. 7. 444
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
different; and those who classified qur¨a as gambling in some situations evidently thought of it as a matter of chance. Their opponents did sometimes counter this by presenting it as an appeal to the divine: one Ìadith displays the Prophet as casting lots in a situation in which there would be winners and losers with the prayer, ‘O God, give judgement among your servants with truth’, and Ibn Îanbal is credited with the statement that ‘the lot hits the truth’ (al-qur¨a †uÒibu }l-Ìaqq).106 But such statements are rare in the Sunni material we have seen. Even on a superficial reading, the Shi¨ites come across as different. Using lots was indeed a way of delegating matters to God in their view,107 and particularly effective if it was done by the imam: his qur¨a never went wrong, being in the nature of waÌy, they said.108 The seventh/thirteenth-century Shi¨ite scholar Ibn ™awus did think of lot-casting as a form of consultative divination, istikhara;109 and qur¨a and istikhara are also treated together in booklets by contemporary Shi¨ites, including Fa∂l Allah, who repeats that lot-casting is a way of delegating problems to God. He mentions unidentified persons who hold that only the imam can do lot-casting, on the grounds that only he knows the special prayer to be said in connection with it (an argument perhaps designed to eliminate the whole institution), but he rejects it on the grounds that no special prayer is needed. The method is only to be used when there is no other solution, he says, and its purpose is simply to solve a problem, not the unveiling (kashf) of anything; but God does not cheat, as he also says.110 By way of contemporary comment, it may be worth noting that there has been much interest in divisory lots as a political device in both England and America in recent years.111 Most Westerners probably think of the procedure as archaic, not so much because they see Sa¨id b. al-Musayyab’s Ìadith in Sarakhsi, Mabsu†, xvii, 49 (with takhrij); Ibn Îanbal in Ibn Qayyim al-Jawziyya, ™uruq, ii, 745. 107 Thus several traditions in Majlisi, BiÌar, xci, 234; civ, 325. 108 Majlisi, BiÌar, ii, 177; xxvi, 32; xl, 245, 328, 363; liii, 331, 332, etc. 109 Ibn ™awus, FatÌ al-abwab, 267ff. 110 Fa∂l Allah, Qur¨a, 26, 30, 33, 49, 62f, 65. For the question whether lotcasting is the prerogative of the imam (as claimed in some traditions), see Ash¨ari, Barrasi, 56ff. For lot-casting, istikhara and istiqsam in another booklet, see Îusayni, Qa¨ida, 123ff. 111 E. Callenbach and M. Phillips, A Citizen Legislature (Berkeley 1985); K. Sutherland, The Party is Over: Blueprint for a Very English Revolution (Exeter 2004); revised as A People’s Parliament (Exeter 2008); B. Goodwin, Justice by Lottery (Exeter 2005). Our thanks to Anthony Barnett for these references. 106
445
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
it as a form of gambling or divination as because they think they can do better than random chance. (In fact, this seems to have been the Îanafi attitude, too, but since the Prophet had endorsed sortition, it was only via his prohibition of gambling that they could reject it.)112 Even today, however, Westerners usually accept the principle of random selection in connection with juries, which are still chosen by (computerised) lot-casting, and it is precisely this principle that is attracting attention as a way of introducing direct representation and popular control to counter what nowadays goes under the name of the ‘democratic deficit’. As a democratic device, random selection is what one book on the subject calls ‘the Athenian option’,113 heartily disliked by a philosopher such as Ibn Rushd because it took no account of virtue;114 but as an antidote to partiality and special interests in general it was wholeheartedly endorsed in the Islamic legal tradition. Ancient though the practice is, it may still be in for new roles, and not just in the West.115 The Return of the Near East Here, however, our interest is not in modern politics, but rather in the relationship between ancient Near Eastern and Islamic culture. The question has not been much studied, but it has received some attention of late,116 deservedly in our view, because it amounts to asking how far we can reconstruct the cultural and religious history of the Near East as a single, continuous narrative rather than as dis-
112 The explanation offered by Rosenthal, Gambling, 33, does not fit the contexts in which qur¨a was identified as gambling. 113 Cf. A. Barnett and P. Carty, The Athenian Option: Radical Reform for the House of Lords (Exeter 2008); cf. also O. Dowling, The Political Potential of Sortition (Exeter 2008), which examines lot-casting as a political device in both Athens and the Western tradition. 114 Cf. P. Crone, Medieval Islamic Political Thought (Edinburgh 2004), 280 and note 111 thereto. 115 Curiously, a ballot or election is actually iqtira¨ in modern Arabic (see H. Wehr, A Dictionary of Modern Literary Arabic [Wiesbaden 1966], s.v.). Other words may be more common (notably intikhab), but iqtira¨ was used in the Iraqi election in 2005, see http://www.newsday.com/news/nationworld/nation/ny-oiraqelectiongallery, 0,322603.photo gallery?index=7 (photo 2). 116 Cf. S. Dalley (ed.), The Legacy of Mesopotamia (Oxford 1998); M. LevyRubin, ‘On the Roots and Authenticity of Conquest Agreements in the Seventh Century’, Jerusalem Studies in Arabic and Islam 34, 2008; and the MELAMMU Project (www.aakkl.helsinki.fi/melammu).
446
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
jointed parts studied under the rubrics of Biblical, Greek, Roman, ancient Iranian, and Islamic history. Between them, the ancient and the Islamic periods cover most of the history of the region, but not all of it: there is a thousand years in between the two, and this is where the problem arises. The thousand years in question are those in which the Near East was under colonial rule, first under the Achaemenids, next under Alexander and his successors, and thereafter under the Greeks and the Romans in its western part, under the Parthians and the Sasanians in Iraq. As the foreigners moved in with their own cultural traditions, the high culture of the Near East was unseated and increasingly reduced to a local tradition of limited interest to those who mattered. The ancient Near Eastern tradition did not die, of course. It changed when it ceased to be written in cuneiform languages and was expressed instead in Aramaic, but as Aramaic culture it lived on. Unfortunately, very little of it has come down to us. We do have Jewish writings in Aramaic, and from the third century CE onwards also Christian ones, but the pagans who formed the vast majority in the region for most of the period have not left us much. By and large, we are forced to study the Near East through the eyes of its conquerors, who remained outsiders to the region in the sense that they continued to be orientated towards their own cultural centres even after having made themselves thoroughly at home in the land. As ill luck would have it, the bulk of the Persian tradition is also lost, so that for practical purposes we only have one pair of foreign eyes, those of the Greeks and the Romans. Some of those who wrote in Greek were Near Easterners by origin, and some of them did try to make their native tradition available in Greek, adapted to Greek tastes. But the bulk of these writings is also lost, and most of the Near Easterners who wrote in Greek had assimilated the hegemonic culture so thoroughly that they sound no different from people of other origin writing in that language. The Jews are again the main exception. From the third century CE onwards, however, all this begins to change. In 211 all members of the Roman empire were granted Roman citizenship (some minor exceptions apart), with the result that all now had to live by Roman law. Since people could not change their ways overnight whatever the degree of Roman control, inevitably this meant that much of what they actually practised was a mixture of Roman and native law. Often called ‘provincial law’, such native law surfaced in both the eastern and the western parts of the empire, and some of it came to be officially endorsed as Roman 447
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
law.117 What this means for us is that the indigenous tradition begins to be visible in the hegemonic culture. The two Greek papyri from Petra and Nessana are perfect examples: the lot-casting by which the shares were allocated was a provincial practice, not a procedure specified in Roman law. Christianity made for even greater change. It originated as a Near Eastern religion carried by speakers of Aramaic, initially Jews, thereafter Jews and gentiles. A socially inclusive movement in which Greeks and non-Greeks, elite and masses, were brought together in a manner hitherto unknown in the Mediterranean, it gradually converted the entire empire to Near Eastern, if increasingly Hellenised, modes of thought, and in the Near East itself it allowed for a more extensive resurfacing of Aramaic culture as the Christians of Syria and Mesopotamia took to writing in Syriac (i.e. the Aramaic dialect current at Edessa). The establishment of a new capital in Constantinople also contributed to the ‘Orientalisation’ of the Roman empire, to use the term adopted by those who see the process from the Greek or Roman point of view. From our point of view, ‘Orientalisation’ is simply a way of saying that it becomes possible to see continuities outside the sphere of law as well. The return of the Near East continued after the Arab conquest, for if Christianity was a kind of homecoming for the Near Eastern provincials, this was even truer of Islam. The Arabs were Near Easterners who definitively unseated the Greeks from their hegemonic role in the region. By then, of course, Greek culture had served as the high culture of the Near East for close to a thousand years, so that there was no way of shedding it: it had gone into the bloodstream of the local culture. But living by Greek culture under the hegemony of Greeks, who continued to see themselves as its ultimate arbiters even in its Christian form, was quite different from continuing Greek cultural ways on one’s own terms, with or without awareness of their Greek origin. Initially, of course, the Arabs were much like the Greeks in that they saw themselves as arbiters of Islamic culture and they too were prejudiced against Aramaeans. But their hegemonic position was shortlived. As converts to Islam, the Aramaeans assumed the legacy, and eventually also the ethnicity, of the Arab conquerors and became their own cultural masters. When we speak of the Arabs today, it is largely the former Aramaeans (and Copts) that we have in mind. Consequently, a great deal of Islamic culture is Aramaic culture, 117 See J. Mélèze Modrzejewski, ‘Diritto romano e diritti locali’, in A. Schiavone, et al. (eds), Storia di Roma, III/2 (Turin 1993), 985–1009.
448
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
brought into Islam in the form in which it had developed under Greek and Persian rule, to develop in new directions thereafter. This is the overall framework in which the connections between ancient Near Eastern and Islamic culture have to be pursued. Lotcasting as an official procedure provides us with a striking example of such a connection, with a typically uneven distribution of documentation: well attested in the cuneiform record, its only attestation in Aramaic seems to be in Jewish works. This is presumably due to the loss of the pagan Aramaic tradition rather than the disappearance of the practice, though it would help if it turned up in Syriac too. As it is, however, we do have it in Greek, and as good luck would have it, the Greek evidence comes from Petra and Nessana, which puts us in the rare situation of having conclusive evidence for pre-Islamic Arabia. Thereafter the evidence is plentiful, but only for the time of the Prophet, the Rashidun and the Umayyads: as the Arab conquest society wanes, so do the attestations. We do find discussion of the practice in Islamic law, but incidental references to the practice in real life seem to disappear until its curious reappearance under the Ottomans. Even the Jews eventually cease to mention it. Partition by lot-casting is still discussed in the Gaonic literature, dating from c. 700–1050; but there is no reference to it in the Kitab al-mawarith of Sa¨adiya Gaon (d. 942), for all that it covers inheritance issues in detail, nor do we know of any in the Cairo Geniza. In short, the overall impression one gets is that what came out of Arabia was in this case an institution that no longer meshed with the way things were done in the rest of the Near East. It came and it went, leaving behind only some traces. One may contrast this with another institution of ancient Near Eastern origin in Islamic law, the clause requiring a freed slave to remain with his or her master until the latter’s death, i.e. as a servant. Known as paramone (‘remaining by’), it was also found as a labour contract for free people. Originating as a contract of adoption designed to provide for the manumitter in old age, it was transmitted from the Near East to Greece at an early stage, and after Alexander’s conquest of the Near East the indigenous and the Greek forms of the institution interacted, to breed an amazing range of variations. The Romans accepted the validity of such contracts when they were made by nonRoman subjects under their own law, but not as part of Roman law. Inevitably, however, it came to be practised under Roman law after the universal grant of citizenship, and though the guardians of Roman law resisted this development, they may eventually have capitulated. With or without official recognition, the paramone remained a prominent 449
THE ANCIENT NEAR EAST AND ISLAM: THE CASE OF LOT-CASTING
part of provincial practice in the Near East. It was also known in Arabia, where free slaves seem often to have been adopted very much as they had been in the ancient Near East. We encounter the paramone as a free labour contract at Nessana and as an archaic requirement of staying with the master in the Îijaz and elsewhere. After the conquests it is reflected in a wide variety of forms in a large number of Ìadiths attributed to early jurists and the Rashidun, and it formed the raw material of what the Muslims were to systematise as kitaba and tadbir.118 If the contract had not been so important outside Arabia, it would presumably have had much the same history as lot-casting: it would have come and gone, leaving behind some traces. But far from receding into obscurity, it generated massive discussion and two new formal institutions. Manumission was of course of much greater practical importance in daily life than lot-casting, so the examples are not entirely comparable. For all that, it is hard not to suspect that the key transmitters of originally ancient Near Eastern culture will prove to be the inhabitants of the Fertile Crescent, now assisted by the Arabian tradition and now without it, but not usually the Arabians on their own. Addresses for Correspondence: [email protected], [email protected]
118 Cf. P. Crone, Roman, Provincial and Islamic Law (Cambridge 1987), ch. 5, and the literature cited there.
450
Journal of Semitic Studies LV/2 Autumn 2010 doi: 10.1093/jss/fgq008 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT OF JUDAEO-ARABIC BIBLE EXEGESIS1 MIRIAM GOLDSTEIN BAR-ILAN UNIVERSITY
Abstract Jewish participation in the culture of the Arabic-speaking world in the ninth and tenth centuries brought revolutionary changes in many areas of Jewish life and scholarship. One of the areas of change was Bible commentary: during these centuries, Jewish Bible commentary as it is known today was created — compositions that proceed verse by verse, elucidating various and varied aspects of the biblical text, including grammar, context, theology, science and philosophy. This essay focuses on one of these innovations, an approach devoting careful attention to the structure of the Bible and emphasizing its organized presentation. One reason for these new approaches to the Bible was the assimilation of new conceptions of authorship and composition in Arabic. Other possible shapers of this approach include particular needs of the Karaite movement and/or traditional masoretic study of the Bible, as well as the requirements of interreligious and intrareligious polemic.
Many aspects of the history of the Jewish community living in the Eastern Mediterranean in the ninth century are obscure. This obscurity stems from a paucity of sources, which has been alleviated only somewhat by the increasing availability of Judaeo-Arabic and Hebrew material from the Cairo Genizah, especially that preserved in the collections of the Russian National Library. Yet despite this lack of clarity regarding its early history and origins, the Judaeo-Arabic literary production of the tenth century unquestionably demonstrates that the preceding years were a time of revolutionary change within the Jewish community, both Rabbanite and Karaite. In this essay I will identify a new approach to the Bible on the part of Jewish exegetes writing in the tenth century, an approach characterized by careful 1 I thank Bruno Chiesa, David Freidenreich and David Sklare for their comments on earlier drafts of this article.
451
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
attention to the structure of the text, as well as new assumptions regarding the virtues of organized presentation. These new approaches to the Bible emerged in part as a result of one of the important literary developments in the surrounding intellectual environment in the previous century: the assimilation of new conceptions of authorship and composition in Arabic. While earlier Jewish exegetes were not unaware of aspects of biblical structure, the depth and extent of the attention displayed by exegetes writing in Judaeo-Arabic disclose a new approach whose origins deserve investigation. Discussions of scriptural structure in the writings of Karaite exegetes in Jerusalem Karaite scholars living in Jerusalem, those active in the scholarly circle of ‘Mourners of Zion’ established by Daniel al-Qumisi in the late ninth century, were prolific composers of works relating to the study of the Bible, including commentaries, grammars, dictionaries, glossaries and even guides to the correct pronunciation of Hebrew.2 The commentaries authored by a number of these exegetes during the tenth century contain a fixed characteristic, new in the period: they assiduously, and at times tediously, describe the structure of the biblical text. Furthermore, in their descriptions, the exegetes often use terms familiar from the realm of contemporaneous Arabic literary composition. Tenth-century exegetes, such as Yusuf ibn NuÌ, Yefet b. ¨Eli and Sahl b. MaÒliaÌ, as well as scholars from later generations such as Abu al-Faraj Harun, Yeshu¨ah b. Yehudah and ¨Ali b. Sulayman, employ the technique in their commentaries. In contrast, this type of structural description is rarely found in the writings of roughly contemporaneous Rabbanite exegetes such as Sa¨adiah Ga}on and Samuel b. Îofni Ga}on. It is not possible at this point to speak of the use of such description as part of a Karaite exegetical school of thought during this period — such characterization will be possible only gradually as more of the Judaeo-Arabic material preserved in the Russian National Library is analysed and published. Nonetheless, the use of such 2
For a survey of the works written by these Karaite scholars on grammar, lexicography and biblical reading, see Geoffrey Khan, ‘The Contribution of the Karaites to the Study of the Hebrew Language’, in Meira Polliack (ed.), Karaite Judaism: a Guide to its History and Literary Source (Leiden 2003), 291–318. A summary and bibliography of the exegetical works of the community can be found in Daniel H. Frank, ‘Karaite Exegesis’, in Magne Sæbø (ed.), Hebrew Bible / Old Testament: The History of Its Interpretation (Göttingen 2000), 110–28. 452
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
descriptions features so prominently in the works of Jerusalem Karaite scholars that Judaeo-Arabic exegetical works from this period can fairly accurately be identified as Karaite on its basis. An apt example of this interest in structure is provided by the commentary on the Pentateuch composed in the late tenth century by the grammarian Yusuf ibn NuÌ and abbreviated by his student Abu al-Faraj Harun (henceforth, the TalkhiÒ). These two grammarianexegetes were revered teachers in the Jerusalem Karaite community, and thus their composition is critical testimony to the exegetical methodology developed and prized by the community. The following discussion is a brief outline of their approach to the explanation of biblical structure.3 The authors of the TalkhiÒ frequently offer basic description of the structure of the biblical text. The exegetes classify sections and count them, explicitly notifying the reader of the component parts of the text. In this way, they comment on Leviticus 7:28–30: This is the third section (faÒl) devoted to the peace-offering. The first is in Vayyiqra and includes a description of the portion of it that is sacrificed on the altar. The second, in this pericope, includes the description of the laws governing its eating by its owner. And the third includes the [description of the] priest’s portion. (MS Russian National Library Yevr.-Arab. I:4116, 101a)
Here, the exegetes explain that the peace-offering has been discussed in two earlier contexts, apparently Lev. 3:1–17, and Lev. 7:11–21, and is addressed again in the section on which they comment. This type of comment is also quite frequent in the works of other contemporaneous Karaite exegetes. Their comments on the Bible are replete with such descriptions of the structure of sections and the connections between them, such as ‘when it/He [= Scripture or God] finished discussing X, it/He began to discuss Y’. Such comments acquaint readers with the structure of the text, and as in the case above, such structural observations are usually the first of a number of observations on the verse. As such, they are likely primarily intended as an educational service to readers, ensuring that they remain oriented within the text. 3
This work is extant in at least eight different copies, and was nearly invariably attributed to the student alone, who was far better known in later centuries than his teacher. For the original Judaeo-Arabic texts, along with further examples and analysis, see Miriam Goldstein, ‘The Bible exegesis of the Karaites Yusuf ibn Nuh and Abu alFaraj Harun: an examination of method in the context of the contemporaneous literary and exegetical approaches of Jews, Christians and Muslims’, unpublished Ph.D. thesis (Hebrew University of Jerusalem 2006). All translations of texts in this article are mine. 453
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
Structural description such as this is never free of exegetical motivation. For example, in the comment above, the exegetes may be motivated by the desire to show that the Bible has a good reason for discussing the same subject in three different contexts, and thus does not include superfluous description. Other possible exegetical motivations exist in other cases: after all, the purpose of these exegetes was primarily to deal with such problems. Nonetheless, the sheer frequency of such descriptions — and indeed, the nearly automatic tendency of these exegetes to include them — points to another purpose as well: to include a type of textual characterization which had become expected trappings of exegesis within the Jerusalem Karaite scholarly community. Among these exegetes, explicit clarification of the overall structure of the text became a crucial element of the explanation of sets of verses. Some exegetes even took this clarification to its logical end, and claimed, as well as attempted to demonstrate, that Scripture is organized logically and well. Such ordered and structureconscious writing was typical of Arabic composition at this point, as I will discuss below. Not only do the authors of the TalkhiÒ identify divisions between sections, they point out structural elements within those sections. Yusuf b. NuÌ and Abu al-Faraj Harun identify verses that function as introductory and concluding elements which frame biblical passages. For example, in the verse, ‘These are the generations of the heaven and of the earth when they were created’ (Gen. 2:4), they explain that the word ‘these’, elleh, signals that the verse functions as a conclusion to the set of verses preceding it. ‘These’ can be a conclusion to what precedes it, as in ‘these are the commandments and the laws’ (Num. 36:13) and similar cases, and it can be an introduction (Òadr) to what follows it, as in ‘these are the descendants of Noah’ (Gen. 6:9) and similar cases. It is likely that here, it serves as a conclusion (khatima) of the preceding description of the creation of the world in six days. (MS Russian National Library Yevr.Arab. I:4785, 14a)
To be sure, the authors’ identification of such introductory and concluding formulas here is motivated by a basic exegetical requirement: the need to determine the meaning of ambiguous expressions in the Bible, and in this case, the demonstrative pronoun. Earlier rabbinic literature addresses the question of this ambiguous pronoun and its relationship to its context (Bereshit Rabbah 12:3; Shemot Rabbah 1:2), and there the rabbis provide a locally relevant statement explaining the specific referent of elleh. The Karaite exegetes respond to the exegetical 454
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
question in a manner which contrasts with that of the rabbis. Rather than merely identifying the referent of the demonstrative pronoun, they fit the pronoun and the verse it introduces into a broad conception of biblical structure, where verses serve specific structural functions. The Karaites’ structural explanation is drawn, consciously or subconsciously, from their awareness of the introduction as a building block of composition, an awareness reflected in their use of terminology familiar from the world of composition: khatima, conclusion, and Òadr, introduction. Similarly, in their comment above, the authors of the TalkhiÒ describe the sections of the Bible with the term faÒl, the same term used in describing the section or chapter divisions of contemporaneous compositions. All three of these components were becoming required elements of Arabic composition. Yusuf b. NuÌ and Abu al-Faraj Harun enumerate many such introductions and conclusions in the Bible. Moreover, they identify a particular type of structure as specific to biblical style: the inclusio well-known from classical rhetoric and familiar to modern-day Bible scholars. This observation allows them innovative responses to wellknown historical questions. Thus, for example, they explain regarding Gen. 9:7: He [= God] revealed this verse as a conclusion of the section, similar to what He began with, in the verse ‘The Lord blessed Noah and his sons, saying to them, “be fruitful and multiply and fill the land”’ [Gen. 9:1]. [This is] in accordance with the frequent scriptural custom of beginning and concluding in like fashion. (MS Russian National Library Yevr.Arab. I:4785, 46b)
Repeated verses such as Gen. 9:1 and 9:7 raise the spectre of superfluity for the Bible exegete. The problem in these verses is noted explicitly, for example, by Rashi, who differentiates between the two instances first ‘according to the plain meaning’ (le-fi peshu†o), and alternately, by quoting a midrashic source dealing with the exegetical difficulty (BT Yevamot 63b). Yusuf b. NuÌ and Abu al-Faraj Harun find a different way to claim that the repetition is not a superfluity. They explain that the verse in Gen. 9:7 serves as an indicator of conclusion specifically because it is similar to the initial verse, Gen. 9:1, and they identify this technique of mirrored introduction and conclusion as a fixed characteristic of biblical style (sirat al-kitab). Unlike Rashi and the rabbinic sources he cites, these exegetes do not need to identify a specific and unique meaning for each of these apparently similar verses. For them, repetition is not a problem; it is a fixed and 455
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
intentional feature of biblical composition and, moreover, serves as a guidepost to the reader, signalling the conclusion of a section. Their structural awareness enables them to formulate unique exegetical solutions in this and numerous other instances.4 Not only do the Jerusalem Karaites identify verses as serving as introductions and conclusions of sections of the Bible; they identify such components of entire biblical Books. Thus the tenth-century exegete Yefet b. ¨Eli, a contemporary of the authors of the TalkhiÒ, explains that the initial seven verses of the Book of Proverbs serve as an introduction to the entire book, in that they present a concise summary of its content and purpose (jawami¨ ma¨ani hadha al-sifr, jawami¨ aghra∂ hadha al-sifr). Similarly, in his commentary on Ecclesiastes, Yefet identifies the initial two verses of the book as intended to concisely set forth its purpose, and he draws a comparison with the first chapter of Proverbs.5 These verses are the statement of the compiler, and were not said by Qohelet. He [= the compiler] intends to mention the purpose of the book as an introduction, just as it says in Proverbs, ‘To know wisdom and education’, and the rest of that section in which he details the purpose of the book and its utility.
This function was precisely that of the Arabic introduction as written by authors in Yefet’s day,6 a function which developed on the backdrop of Syriac and Greek sources. It is no mere coincidence that Yefet identifies biblical introductory sections as serving in this same capacity: time and time again, he and his compatriots find parallel elements in the Bible and in contemporaneous composition. Introductions and conclusions guide the reader through the Bible, and furthermore, according to the Karaite exegetes of Jerusalem, the text as a whole is well-structured; composed in a logical manner. Yusuf b. NuÌ and Abu al-Faraj Harun frequently discuss the relationships between sections of biblical text. One such example is found in their comment on Exod. 22:15, ‘And if a man seduces a virgin who is not 4 For further examples and analysis of the innovation inherent in this approach, see Miriam Goldstein, ‘Servants’ Small Talk: The Structural Function of Biblical Superfluity in the Bible Exegesis of Yusuf ibn Nuh (Hebrew)’, Sefunot 23 (2003), 349–70. 5 Yefet ben Eli, Der Commentar des Karaeers Jephet Ben ¨Ali Halevi zu den Proverbien, Israel Guenzig (ed.), (Cracow 1898), vii, ix. Richard Bland, ‘The Arabic commentary of Yephet ben ¨Ali on the book of Ecclesiastes, chapters 1–6’, unpublished Ph.D. thesis (University of California at Berkeley 1966), 5. 6 See, for example, Peter Freimark, ‘Mukaddima’, EI 2 (Leiden 1995), 495–6.
456
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
betrothed, and lies with her, he must make her his wife by payment of a bride-price’. The exegetes note the context of the verse and explain the thematic connection with the section of laws preceding it: Since He mentioned aggressions against animals by humans, He linked it with (nasaqa ¨alayhi) his discussion of aggressions that humans inflict on each other. (MS Russian National Library Yevr.-Arab. I:4785, 258b)
Yusuf b. NuÌ and Abu al-Faraj Harun note that the Bible proceeds systematically in its discussion of cases of damages, beginning with cases in which animals and property are the victims and then progressing to cases in which humans are the victims. These exegetes and their contemporaries use words derived from the Arabic root n-s-q, meaning ‘to put in proper order’ or ‘to array’, in their descriptions of such connections. In this instance as in many others, Yusuf b. NuÌ and Abu al-Faraj Harun strive to find the principle governing the juxtaposition of sections. Their explanations are founded on the conception that just as a contemporary author carefully shapes the structure of his work and links the sections, so does the Bible. In their exegetical explanations, they systematically and consistently attempt to demonstrate such links. Yefet b. ¨Eli also devotes lengthy discussions to such questions of placement. Indeed, Yefet’s frequent discussion of such issues is likely what led the twelfth-century Spaniard Abraham b. {Ezra, who was quite familiar with his writings, to criticize interpretation according to hesmekh ha-parashiyyot (s.v. Deut. 24:6) as ‘Karaite’ and profitless, despite adopting the technique in certain cases in his own commentaries.7 Not only does Yefet systematically demonstrate the organizational principles of the biblical text, he states explicitly in a number of instances that the Bible is organized in a superior method, and indeed, exemplifies the best type of composition possible. Yefet takes up the topic of biblical organization in his discussion of Exodus 18–19, a classic case of the Bible’s deviation from strict chronological presentation, in the ordering of Jethro’s visit and the giving of the Law. Yefet addresses the apparent exegetical difficulty, explaining that the Bible deviates from chronological order when there are good reasons to do so, and that, moreover, this is evidence 7 Regarding Ibn Ezra’s Arabic sources in his commentary on Psalms as well as other biblical books, see Uriel Simon, Four Approaches to the Book of Psalms: from Saadiah Gaon to Abraham Ibn Ezra, trans. Lenn Schramm (Albany 1991), 156–61.
457
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
that the Bible shuns scattered presentation (Ìatta la yakuna mabtur), in favour of the ‘best possible form of composition’ (aÌsan ta}lif ). Yefet adduces the organization of the book of Jeremiah as proof. He explains that despite the fact that the prophecies to Israel and those regarding the gentiles were interspersed throughout the prophet’s lifetime, they are grouped together by genre in his book, for the sake of good organization and for later generations’ ease of reference.8 In the case of Exodus 18–19, explains Yefet, the non-chronological presentation was dictated by considerations of comparative length and how best to link the accounts. Yefet’s motivation and ability to find evidence of superior literary arrangement provides him numerous and varied solutions to exegetical difficulties regarding the organization of the Bible. The arrangement that he identifies is often related to the characteristic of orderliness, a prized element in contemporaneous writing style. Is it new? Earlier considerations of biblical organization Of course, to nearly any eye, the Bible is not a perfectly organized work, and lack of order, most obviously chronological order, presents a basic exegetical problem. The rabbis, too, addressed the question of the relevance of the juxtaposition of sections, in a technique later called derishat semukhin or semikhut parashiyyot. Despite some degree of similarity with the method described above, comparison of the two approaches points up the innovation inherent in the Karaite structural descriptions of the Bible. Rabbinic interpretation is often characterized as focused solely on microscopic issues of the text, and oblivious to context. It is clear, however, that the rabbis were aware of biblical structure and context. This is suggested by their reading practices, which apparently included a method of learning in which the Bible was read in order, verse by verse.9 Furthermore, rabbinic statements reveal a concern with order and structure in the Bible: one oft-cited view states that ‘The Bible is not ordered chronologically’ (ein muqdam u-me}uÌar ba-torah); other formulations query ‘Why was this portion juxtaposed?’ as well as ‘What is written before this?’ in a later formulation.10 Other midrashim 8 MS St. Petersburg Institute of Oriental Studies of the Russian Academy of Sciences B221, 91a–92a. See text and discussion in Goldstein, ‘Structural Function’. 9 Jonah Fraenkel, The Method of Aggada and Midrash (Hebrew), 3 vols. (Givatayyim 1991), II, 27–9; III, 42. 10 Fraenkel, Aggada and Midrash, I, 183–5; Isaak Heinemann, Methods of Aggadah (Hebrew) (Jerusalem 1953), 140–61.
458
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
specifically address the more general question of the appropriate conclusions to be drawn from the contiguity of passages.11 Such questions bring considerations of structure to the fore, and propose explanation and justification. That said, the rabbinic attention to the juxtaposition of sections is distinct from the Karaite discussions of structure. For one, they differ in purpose. The technique of analysing juxtaposition of sections, in rabbinic literature, was one of the prominent techniques of midrash — a tool in the rabbinic repertoire, helpful in creating multiple interpretations of the text. While their examination of questions of structure was at times intended to address exegetical problems; in many instances, they were merely attempting to provide support for homilies loosely based on the multivalent Scriptural text. Already in the fifteenth century, the exegete Don Isaac Abarbanel identified the rabbinic attempt to understand the order of the text, and its midrashic result: he comments on Numbers 21 that while ‘It is forbidden to attribute to the Judge of all Flesh the ignominy of disorder in His teachings’, the rabbis, ‘in their attempt to provide some type of justification [to the order of Scripture], would prophecy falsely, each man doing as he pleased (hayu mitnab}im ish kol ha-yashar be-¨einav ya¨aseh), in the form of derash and illusory claims’. For the Karaites, such structural description was a basic component of exegesis, which was often included for its own sake, in the absence of exegetical problems requiring solution. Furthermore, the two bodies of exegesis differ with respect to extent. Structural analysis is applied only selectively in rabbinic literature, in contrast with the comprehensiveness of the approach in the tenthcentury Karaite explanations. Interest in compositional issues in the exegesis of Sa¨adiah Ga}on Explicit structural description of the biblical text is a distinctive hallmark of the exegetes discussed above, and is a significant element in the exegetical method of other Karaite Jerusalemites not discussed here. In contrast, it is barely attested in the works of well-known Rabbanite commentators such as Sa¨adiah Ga}on. However Sa¨adiah did devote consideration to aspects of biblical structure, and the 11 See, for example, Sifrei Num. 131, in reference to Hos. 1:9. I thank Jordan Penkower for this reference.
459
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
questions he raises derive from the same assumptions regarding literary excellence. Sa¨adiah arrived at a conclusion directly opposite to that of the Karaites, as encapsulated by Yefet’s admiring characterization, ‘the best possible form of composition’. For Sa¨adiah rejected this notion and instead, characterized both the Pentateuch and the Prophets as disorganized, and surprisingly so! For Sa¨adiah, the question first arises in the introduction to his commentary on Genesis. He notes a number of characteristics of ordered compositional style, and notes that the proverbial ‘questioner’ is likely to ask why the Pentateuch contrasts with these ideals: the book is not arranged according to what Sa¨adiah labels al-qawl ¨ala jihatihi, ‘well-crafted language’.12 Someone might question, saying, ‘If composition in a well-ordered way is considered most honourable, then why do we not find the laws and commandments of the Book, that is, the Pentateuch, grouped together, and then organized by chapter and divided into sections in an ordered way, but rather, we see that it is scattered and disconnected?’ We respond to him by saying that its Revealer, may He be exalted and praised, intended by this to cause his worshippers to occupy themselves diligently and carefully with the text, so that they would preserve it completely and would transmit it. In addition, so that they would link all of the similar verses, and assign each issue its subject and branch of knowledge, so that this would add to their reward in the World to Come.13
Sa¨adiah explicitly compares modern-day composition and the Bible, applying the principles of order governing the composition of secular text to the Bible. He explains that while ordered presentation is the ideal way in which to compose, God consciously chose to deviate from these norms for a particular pedagogical purpose. He left the task of ordering to his worshippers, who would be compelled to dedicate themselves to its study and would in return gain personal and national reward. Sa¨adiah’s discussion of reward for occupying oneself with the text of Scripture is a typical Mu¨tazilite doctrine, as well as a theme familiar from rabbinic literature.14 12 On this usage, see Joshua Blau, A Dictionary of Mediaeval Judaeo-Arabic Texts (Jerusalem 2006), 751. 13 Saadiah Gaon, Commentary on Genesis, ed. Moshe Zucker (New York 1984), 4. 14 Eleventh-century Muslim Mu¨tazilite sources relay a similar explanation, stating that the ambiguities in the Qur}an provide the incentive to decipher, and the more difficult the act of deciphering, the greater the reward (Goldstein, ‘The Pentateuch Exegesis’, 317–20). Regarding rabbinic literature, see for example, Mishnah Avot 2:15–16.
460
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
Sa¨adiah puts forth a similar question in the introduction to his commentary on Isaiah. This introduction puts forth a view of prophetic promises and threats as effective divine tools for human education, apparently issued in response to polemical arguments challenging their efficacy.15 Towards the end of the introduction, Sa¨adiah sets out a side issue, a problem regarding the organization of the prophetic works: What is the divine purpose in scattering the subjects of this book and other books of scripture rather than collecting them together? That is, the warning sermons are not gathered together, nor are the [descriptions of] punishment organized in one place, nor are the good tidings compiled together, but rather, one finds that he includes a section of warning sermons, follows this with a description of good tidings, then returns to scolding words, follows this with ethical discussion, then establishes a series of promises and good tidings, and then again returns to threats and scolding. That is, he diversifies these subjects and ramifies them unceasingly, rather than creating a coherent continuity and thus bringing each subject to full conclusion.16
Sa¨adiah asks why the words of Isaiah, and indeed, Scripture in general, include juxtaposed and ever-varying topics and genres. That is, the Bible is not organized, and Sa¨adiah’s question indicates that he is judging the Bible according to contemporary literary custom, in which order and organization are prized. Here, again, the Ga}on explains God’s choice — an unexpected one requiring justification — as a considered pedagogical method. The human tendency to impatience as well as to pride and despair means that God can best convey His wisdom by varying topics, avoiding inordinate amounts of praise or blame in any one section — and this is the way prophetic works are composed. Sa¨adiah’s answer is apt, and succeeds in explaining an ostensible defect, when judged by contemporary literary standards, as a virtue. Interestingly, Sa¨adiah and later Rabbanite exegetes provide similar explanations in other contexts, such as in instances of ambiguous or metaphorical language.17 15 The claim polemicized against was apparently put forth by groups of freethinkers, whom Sa¨adiah does not mention explicitly, who challenged the authority of religions founded on the concepts of revelation and prophecy. See Haggai BenShammai, ‘Sa¨adiah’s Introduction to Isaiah as an Introduction to the Books of the Prophets’ Tarbiz 60 (1991), 376–9. 16 Judaeo-Arabic text from Ben-Shammai, ‘Sa¨adiah’s Introduction to Isaiah’, 399, ll.173–9. 17 See Paul Fenton, Philosophie et Exégèse dans Le jardin de la méthaphore de Moïse Ibn ¨Ezra (Leiden 1997), 284–6; Goldstein, ‘The Pentateuch Exegesis’, 313–62.
461
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
Of course, it is not only the scholar, Karaite or Rabbanite, living in an era of ultra-organized literary standards who thinks to question apparent lack of order in the Bible. Sa¨adiah Ga}on’s question is foreshadowed by a rabbinic discussion, which explicitly questions the Bible’s apparent lack of organization, and responds that ‘the sections (parashiyyot) of the Torah were not revealed in order, for were they revealed in order, anyone would be capable of reviving the dead and accomplishing miracles’.18 That is, the disorder in the Bible — and this midrash likely refers to chronological order — is intentional, for the true order of Scripture is an esoteric and dangerous knowledge reserved for God alone. In proposing this question, the rabbis again demonstrate some degree of theoretical consideration of biblical structure. Furthermore, as is frequently the case in Sa¨adiah’s work, the question he asks has rabbinic precedents. Sa¨adiah’s explanation accords to a great extent with that found in the rabbinic literature he so prized: that the Bible is ordered in the best pedagogical way possible, and is not necessarily chronologically or structurally perfect. Sa¨adiah’s own query and discussion could derive equally well from these rabbinic statements or from his own environment. His rendering, though, is unmistakably reflective of his own context. The purpose of the rabbinic statement is hyperbole regarding the miraculous nature of the Torah, as well as polemic against any attempt to reorder the Bible. In contrast, Sa¨adiah formulates his initial assumption (‘Given that composition in a well-ordered way is considered most honourable’) as well as his conclusion, from the standpoint of contemporaneous literary and theological conceptions.19 While their evaluations differ, Sa¨adiah Ga}on, like the Karaites, explicitly evaluates biblical organization in comparison with contemporaneous literary norms. The Arabic composition and the prestige of organization Whence, then, this consistent interest in the structure of the biblical text, on the part of Jewish scholars living in the Mediterranean and 18 Midrash Tehillim 3. I thank Yehudah Liebes for this reference. See also Heinemann, Methods, 140. Estimates regarding the setting of the editing of Midrash Tehillim range from the Talmudic to the Geonic period, however it is clear that the work derives from early material predating Sa¨adiah’s time (William Braude, The Midrash on Psalms [New Haven 1959], xxv–xxxi). 19 For further discussion and examples, see Haggai Ben-Shammai, ‘The MidrashicRabbinic Literature in the Commentaries of Sa¨adiah Ga}on: Continuity and Innovation’, in Tradition and Change in the Judeo-Arabic Culture of the Middle Ages (Ramat Gan 2000), 33–68.
462
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
the East during this period? One motivation for such interest lies in the Arabophone scholarly environment surrounding them, and the literary norms which had come to characterize writing in Arabic. Arabic-language scholarship flourished in the ninth and tenth centuries, benefiting intellectually from foundations rooted in earlier traditions of learning as well as materially from ¨Abbasid caliphal patronage. The ¨Abbasids, in part, were lucky — the conquest of geographical areas with experience in paper production was one factor promoting the scholarly efflorescence which took place during their rule.20 However, these rulers also actively encouraged the translation of many classic works of philosophy, logic, astrology, alchemy, mathematics, natural science and medicine, from Syriac, Pahlavi, Sanskrit and Greek, often on the basis of earlier translations.21 The compositions now becoming available in Arabic were important, to be sure, for their content and practical applications. Their form and structure were no less revolutionary for the development of scholarship in Arabic. The newly translated works were, in the main, systematically organized, with clear introductions, conclusions, and chapter divisions; authors spoke in them with an explicit and clear authorial voice. Some of these works included discussions of style and organization of speech and writing: for example, Aristotle’s Rhetoric and its commentaries were among the first of his works translated into Arabic in the early ¨Abbasid period.22 Increasingly, and in part due to their familiarity with the methods of organization found in translated works,23 authors writing in Arabic began to adopt these methods 20 See Jonathan Bloom, Paper Before Print: the History and Impact of Paper in the Islamic World (New Haven 2001). 21 See Dimitri Gutas, Greek Thought, Arabic Culture: The Graeco-Arabic Translation Movement in Baghdad and Early ¨Abbâsid Society (2nd-4th/8th-10th centuries) (London / New York 1998), Franz Rosenthal, The Classical Heritage in Islam, trans. Emile and Jenny Marmorstein (Berkeley 1975). 22 Richard Walzer, ‘Aristutalis’, EI2 (Leiden 1995), 630–3. Badawi edited and published a paraphrastic translation of Aristotle’s Rhetori, and its translator, as is noted by the editor himself, misses the point of the original in numerous instances: Aristotle, Rhetorica in Versione Arabica Vetusta, ed. Abd al-Rahman Badawi (Cairo 1959), vi–vii. This makes it difficult to trace the usage of specific literary terms used to describe structure. To overcome this problem, I have examined the use of terminology in ninth and tenth-century Arabic compositors as reflective of the terms in use during the period. 23 See regarding the contribution made by internal Muslim developments, already visible in Sibawayhi’s Kitab: Gregor Schoeler, The Oral and the Written in Early Islam, ed. James Montgomery, trans. Uwe Vagelpohl (London 2006), 43, 51.
463
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
and style themselves, in rasa}il, epistles, and increasingly, in the monographs that developed from these epistles.24 Muslim scholarship was undergoing a transformation from being primarily oral, to primarily written,25 and the wealth of written material appearing in the late eighth and early ninth centuries was styled and organized in a new way. As new modes of composition became de rigueur in Arabic writing, authors began to discuss these new compositional modes explicitly. One such example is the genre of educational manuals describing the work of the administrative secretary, the katib. Such works addressed the art of composition from the preparation of writing implements and paper, to the organization of words. Organization in writing had high prestige, and good literary organization was seen as a precursor and requirement for successful administration.26 In one famous work of this type, the ‘Virgin Epistle’ of Ibrahim b. MuÌammad al-Shaybani (d. 911), the author tells his audience, ‘Thus follow these customs and ways, […] And be mindful of the introductions of your books and their chapters, and their beginnings and conclusions, and place every subject in its proper place’. He concludes with definitions of good composition (balagha) as detailed by authorities of various origins, including various structural elements such as ‘the knowledge of where to separate and where to join’ or ‘correct establishment of the sections and choice of words’.27 Tellingly, the latter advice is 24
See Albert Arazi and Haggai Ben-Shammai, ‘Risala’, Encyclopedia of Islam2 (Leiden 1995), 532–9. I do not know of any studies in the history of Arabic literature which trace these changes and innovations resulting from Greek and other literary models. For discussion of the adoption of these formal characteristics of composition in Judaeo-Arabic literature, see Rina Drory, The Emergence of Jewish-Arabic Literary Contacts at the Beginning of the Tenth Century (Hebrew) (Tel Aviv 1988). 25 On this transition in the Muslim milieu, in the religious sciences, see Sebastian Guenther, ‘The Excellence of the Written Word in Medieval Islam’, Jerusalem Studies in Arabic and Islam 32 (2006), 125–43; Michael Cook, ‘The Opponents of the Writing of Tradition in Early Islam’, Arabica 44:4 (1997), 437–530; Gregor Schoeler, ‘Writing and Publishing: On the Use and Function of Writing in the First Centuries of Islam’, Arabica 44:3 (1997), 423–35; Schoeler, The Oral and the Written, 62–86. Regarding this question in the genre of adab, see Shawkat Toorawa, Ibn Abi Tahir Tayfur and Arabic Writerly Culture: a Ninth-Century Bookman in Baghdad (London 2005), 7–17. 26 Paul Heck, The Construction of Knowledge in Islamic Civilization: Qudama b. Jafar and his Kitab al-Kharaj wa-sina}at al-kitaba (Leiden 2002), 26. See also Heck, ‘The Hierarchy of knowledge in Islamic Civilization’, Arabica 49 (2002), 27–54. 27 Muhammad Kurd Ali, (ed.), Rasa}il al-Bulagha}3, (Cairo 1946), 234, 251. See parallel in ¨Amr Abu ¨Uthman al-JaÌiÂ, Kitab al-Bayan wal-Tabyin, ed. {Abd al-Salam Harun (Cairo 1968), 88. 464
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
attributed to ‘the Greek’, and indeed, such explicit discussions of style and organization could easily have derived from the third book of Aristotle’s Rhetoric or its commentaries available in Arabic.28 The literary transformation reverberated in the Jewish scholarly world. Jews were writing in Arabic, and indeed, this was the language of choice for most Jewish scholars, both Karaite and Rabbanite, by the mid-to-late tenth century. One hundred years before, the use of Arabic in religious composition was questionable and even prohibited, and in any case, written compositions were relatively rare. However, during the ninth century and the early tenth, Jewish scholarship had also undergone a shift from being primarily oral to being primarily written, and, as part of this shift, Arabic had largely come to replace Aramaic and Hebrew in written composition. The initial transformation likely occurred outside the Babylonian yeshivot, but was brought to the yeshivot and to official religious institutions by the early tenth century, with the leadership of Sa¨adiah Ga}on.29 The linguistic change brought with it changes both in subject and in form. Jewish scholars writing in Arabic took on many new genres and subjects during this period, whether philosophical, theological, legal, poetic or literary; adopting new concepts from their surroundings and alloying them with inherited tradition. They composed these works according to the structure of the newly-available models described above, as mediated through Arabic texts. Recent studies discuss this literary transformation in general terms as well as highlighting specific aspects of Jewish scholars’ adoption of new compositional styles.30
28
See note 22. David Sklare, Samuel ben Hofni Gaon and his Cultural World: Texts and Studies (Leiden 1996), 37–67, 99–141. 30 On the adoption of the monograph form by Jewish authors, see Robert Brody, The Geonim of Babylonia and the Shaping of Medieval Jewish Culture (New Haven 1998), 156–61, 267–82. See also Drory, Emergence, 97–103; Sklare, Samuel ben Hofni Gaon, 50–67. On specific elements of composition, see Sarah Stroumsa, ‘A Literary Genre as an Historical Document: On Saadia’s Introductions to his Bible Commentaries (Hebrew)’, in ‘A Word Fitly Spoken’: Studies in Qur}an and Bible Exegesis, Presented to Haggai Ben-Shammai, ed. Meir Bar-Asher, et al. (Jerusalem 2007) 193–204, which evaluates Sa¨adiah Ga}on’s use of the introduction in his Bible commentaries and the similar technique found somewhat earlier in Syriac texts; see David Sklare, ‘Scriptural Questions: Early Texts in Judaeo-Arabic’, in ‘A Word Fitly Spoken’, 205–31, for analysis of a number of Judaeo-Arabic compositions from the ninth through twelfth centuries in the genre of question and answer and Greek and Syriac forerunners. 29
465
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
Explicit discussion of compositional method on the part of Jewish scholars Explicit discussion of compositional method on the part of Jewish scholars writing in Arabic provides apt demonstration of the extent to which they had internalized such compositional norms, as well as of the prestige of ordered composition. Their testimony adds Judaeo-Arabic testimony to the trend of self-conscious composition represented by the Muslim ‘secretaries’ craft’ genre discussed above. Sa¨adiah Ga}on, for example, explicitly addresses questions of compositional method in his comment on Prov. 25:11. There he includes a lengthy discussion of methods of writing, speech and composition, instructing his reader in how to achieve the eloquent status of ‘golden apples in silver settings’. He describes four methods of composition, including the focused monograph (al-mukhtaÒara), the associative composition devoted to a broad topic (al-muntashara) and the enumerative composition (al-Òa}ida, al-habi†a), in which polemical claims or cures for maladies are discussed in ascending or descending order. Sa¨adiah devotes particular attention to the requirements for the introduction to a composition, noting that it must include explicit discussion of the contents of the work.31 Sa¨adiah’s near-contemporary, the Baghdadi Karaite exegete and theologian Ya¨qub al-Qirqisani, also discusses authorship and compositions explicitly in new fragments identified from the first chapter of his legal and philosophical work, Kitab al-anwar wal-maraqib. There, Qirqisani explains that any author desiring to compose a reputable work of theology must provide comprehensive and accurate renderings of opponents’ claims, as well as the ordered presentation of the author’s own principles, in the introduction to his composition. Indeed, this is the style adopted by Qirqisani himself, as well as by Sa¨adiah, in many of their works. Apparently not all authors paid due heed to this authorly requirement: Qirqisani contrasts these theoretical demands with the inferior, even cowardly, practice of authors of his own day, who present only their own claims, leaving aside those of their opponents as well as any possible challenges (innama sabil kull aÌad minhum an ya}ti bil-dalil alladhi yaÂunnu annahu yathbutu bihi qawluhu wa-yada¨u dhikr ma li-khaÒmihi wa-laysa hadha Ìaqq al-naÂar).32 31 Saadiah Gaon, Proverbs with the Translation and Commentary of Rabbenu Sa¨adiah ben Joseph al-Fayyumi, ed. Yosef Kafih (Jerusalem 1976), 198–202. See discussion in Stroumsa, ‘A Literary Genre’, 199. 32 These new fragments from the collections of the Russian National Library, which further complete the beginning of Book I published by Nemoy, were discovered
466
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
Qirqisani explicitly discusses fixed aspects of contemporaneous Arabic composition, which by his time were de rigueur, and rarely merited discussion. He even asserts that certain features have biblical parallels. According to him, King David himself encouraged the common practice of the invocation of God’s name in the basmala. Thus, in the introduction to his Thirty-Seven Principles of biblical exegesis, he states: In the name of God, the merciful, the compassionate. To invoke the name of God is the most appropriate way to enter upon any work, as David – peace be upon him – says: ‘Blessed is he who cometh in the name of the Lord [Ps. 118:26]’. Also, Solomon says: ‘The name of the Lord is a strong tower [Prov. 18:10]’, which tells us that the name of the Lord – whose greatness is exalted – is like to a strong and secure tower, whither the just and upright man hastens and is helped. And after beginning with the name of the Creator, whose names are holy, his praise and extolling follow.33
Current Arabic literary method, then, of beginning with the invocation of God’s name, followed by his praise, is merely continuing an age-old practice encouraged by Jewish tradition. Qirqisani’s explicit description of the practice and the parallel he notes in the biblical wisdom literature exemplify his awareness of methods of composition, as well as his desire to anchor current practice in sacred text.34 About a century later, a Karaite theologian writing in Jerusalem employs his awareness of new compositional norms in order to challenge a fundamental Muslim doctrine, the inimitability of the Qur}an (i¨jaz al-Qur}an).35 One mainstay of this doctrine was the ‘challenge’ by Bruno Chiesa, and are being edited together with Haggai Ben-Shammai. I am grateful to both for their generosity in providing me with manuscript images and transcriptions. 33 I thank Bruno Chiesa for providing me with a preliminary edition of these principles. 34 Interestingly, numerous modern sources echo Qirqisani’s attribution, citing the biblical usage, together with pre-Islamic Arabic usage, as the source of the basmala as found in the Qur}an and as used in Arabic discourse. See, for example, Theodor Nöldeke, Geschichte des Qorans (Leipzig 1909–38), I, 116. 35 See excerpts and discussion of the polemical composition in which this comment is found in David Sklare, ‘Responses to Islamic polemics by Jewish Mutakallimun in the tenth century’, in Hava Lazarus-Yafeh, et al. (eds), The Majlis: Interreligious Encounters in Medieval Islam, (1999), 137–61. Polemical works against Islam, especially those written in Arabic, were relatively uncommon amongst the Jews of the East and the Mediterranean, and this important article discusses several such polemical works preserved in the manuscript collections of the Russian National Library. Sklare also identifies al-BaÒir’s discussion as an example of Jewish incorporation of 467
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
found in the Qur}an (ayat al-taÌaddi, cf. Qur}an 10:38, 11:13) in which detractors are challenged to produce verses or chapters similar to those in the Qur}an. The fact that, according to some, no such work was ever produced, is one evidence of the Qur}an’s miraculous nature. In his attack on this proof, the theologian Yusuf al-BaÒir, a contemporary of Abu al-Faraj Harun, casts doubt on the authenticity of these verses by analysing the differences between the compilation of the Qur}an, and that of contemporaneous compositions. For if we say to them, what is the proof that these verses were said by him [= MuÌammad], they say that these verses derive from the Qur}an, and we know via transmitted tradition that the Qur}an originated with MuÌammad, just as we know the authorship of contemporary compositions. For example, we know that al-Jami¨ alKabir originated with Abu Hashim and that Kitab al-Amanat originated with the head of the yeshiva al-Fayyumi. [We say:] Know that the situations are not comparable. For today’s authors compose books themselves, and they order their chapters, and they are distributed amongst a large population to such an extent that additions and subtractions become impossible. For if some of the copies were added to or subtracted from, this would be evident. The Qur}an is not like this, for its conveyer did not record it himself, and in his time there did not even exist equivalent versions of the text.36
Yusuf al-BaÒir does not attack the Qur}an for its structure per se — this would not befit a mutakallim like himself. Rather, he challenges the epistemological status of the Qur}an by noting the difference between the Qur}an and a contemporaneously authored book. For one, modern-day authors compose their works themselves (yuÒannifuha hum) — al-BaÒir emphasizes the personal nature of the action, in a criticism of the post-facto compilation of MuÌammad’s prophecy. Furthermore, such modern compositions have internal order (taÒnif ), and their authors include an ordered set of chapters (yuÌa∂∂iruna abwabaha), likely stating the contents of these chapters in the introduction. Such organization limits the possibility of unauthorized additions and subtractions, as does the book’s wide diffusion in this form. This was not the case for the Qur}an, explains al-BaÒir, for it was fully compiled only a number of decades after MuÌammad’s death, and was only then made more widely available in this compiled literary concepts of composition, in ‘The New Edition of QiÒÒat Mujadalat alUsquf ’ (Hebrew), Pe{amim 75 (1998), 101–8. 36 Cf. Judaeo-Arabic text from MS Russian National Library Yevr.-Arab. I:3022, 10a–10b, published in Sklare, ‘Responses to Islamic Polemics’, 157, n. 67. 468
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
form. Until that point, its compiler, the caliph ¨Uthman, was free to add and subtract without anyone knowing the difference.37 The fact that al-BaÒir chooses to base his epistemological claim on contemporaneous methods of composition demonstrates the extent to which he and those surrounding him accepted these norms as requirements not only of good writing, but also of sound and trustworthy composition. A new look at the Bible Given their familiarity with surrounding literary norms and their own incorporation of these norms, Jewish Bible exegetes of the tenth century, both Karaite and Rabbanite, were bound to ask new questions when they examined the Bible.38 These new questions and interests made their appearance in various forms: in the case of the Jerusalem Karaites, it was to assert the existence of such literary norms in the Bible, touting its perfection and orderliness and justifying them at all costs; in the case of Sa¨adiah, it was to mention such norms and to explain why they were not found in the Bible! What is clear is that both groups subscribed to the same literary assumptions. The import of new organizational methods and their infiltration into the consciousness of Jewish scholars writing in Arabic meant that if the Bible was to be a work worthy of admiration, it should accord with contemporary literary norms. Both the Karaite championing of the structure of the Bible and Sa¨adiah Ga}on’s justification of ‘defective’ biblical organization arise from their authors’ absorption of such literary attitudes. 37 Al-BaÒir explicitly notes the Shiite polemical origin of this view of the possibility of omissions (nuqsan) from the Qur}an by ¨Othman, as well as accounts from the Îadith regarding sections of the Qur}an not accepted by all (ziyada), in the continuation of his claim (see Judaeo-Arabic text in Sklare, ‘Responses to Islamic Polemics’, 151–2, n. 53). The transmission of this claim from early Shiite polemic into polemics by Jews and Christians writing in Arabic is an apt example of the way in which Sarah Stroumsa described these arguments ‘as currency, quickly changing hands’: see Sarah Stroumsa, ‘Jewish Polemics against Islam and Christianity in the Light of Judaeo-Arabic Texts’, in Norman Golb (ed.), Judaeo-Arabic Studies; Proceedings of the Founding Conference of the Society for Judaeo-Arabic Studies (Amsterdam 1997), 241. 38 As is common in any literary culture. See the discussion of this phenomenon throughout the history of Bible exegesis in Joseph Sadan, ‘Identity and Inimitability — Contexts of Inter-Religious Polemics’, Israel Oriental Studies 14 (1994), 196–7, n. 3–4.
469
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
Christian and Muslim Scriptural Analysis Did members of other faiths, though, come up with similar questions when they examined their scriptures? If familiarity with parts of the classical corpus and adoption of aspects of its composition inspired such approaches to Scripture on the part of Jews, presumably similar phenomena can be found in other religious traditions. Christian exegesis from an earlier period presents a suggestive parallel. The Christian absorption and appropriation of Greek scholarly method and literary norms in its early centuries motivated some degree of ‘structural exegesis’ — consideration of questions of composition, authorship and styling in the Bible — as well. This structural exegesis can be found in the approach of Theodore of Mopsuestia and his students, scholars writing in Greek in the area of contemporary Syria and southeastern Turkey during the fourth and fifth centuries. Their method contrasted with existing allegorical approaches, and included discussion of a number of historical issues related to the development of the text, including ascertaining the connections between verses, the intent of various ‘titles’ such as those of the Psalms, and the authorship and the compilation of biblical texts. It is generally well-accepted that such approaches to the text originated in the exegetical methods of the rhetorical schools. That is, it was these exegetes’ Greek education — their exposure to the tradition of exegesis of philosophical and rhetorical texts, as well as their familiarity with component parts of these texts — which motivated them to examine the literary and structural aspects of the Bible.39 While the Christian appropriation of such methods in their works was a considered choice, and the link to early rhetorical schools is evident, the appropriation of such methods and structures in JudaeoArabic Bible exegesis was more subtle and selective. Nonetheless, the phenomena are parallel: Bible exegetes begin to ask new and similar questions about the text as a result of their contact with Greek scholarship. In Christianity, this occurred in the early centuries of the common era; in Judaism, in the ninth and tenth centuries, and mediated via Arabic. 39
Frances Young, Biblical Exegesis and the Formation of Christian Culture (Cambridge 1997); Christoph Schäublin, Untersuchungen zu Methode und Herkunft der antiochenischen Exegese, Theophaneia (Köln / Bonn 1974), 158–70; Robert Charles Hill, Reading the Old Testament in Antioch (Leiden 2005), 108–9; Sebastian Brock, ‘From Antagonism to Assimilation: Syriac Attitudes to Greek Learning’, in Nina Garsoïan, Thomas Mathews, and Robert Thomson (eds), East of Byzantium. Syria and Armenia in the Formative Period (Washington D.C. 1982), 17–34. 470
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
Muslim scriptural analysis is an important contemporaneous test case, given that Muslim exegetes examining the Qur}an were certainly exposed to the new Arabic literary norms discussed above. Indeed, Muslims had been composing both administrative and literary composition in the new structures and styles well before their adoption by Jewish authors writing in Arabic. Granted, the Qur}an is structured differently from the Bible, with a distinct rhetorical style. Nonetheless, the work is susceptible to structural analysis, as is exemplified in intermittent comments that do appear in classical exegetes such as Ibn Qutayba (d. 889), as well as the modern studies of Quranic structure by scholars such as Angelika Neuwirth.40 It does not seem, however, that Muslim exegetes devoted discussion to comprehensive analyses of Quranic structure during this period, at least on the basis of the exegetical compositions currently extant and edited. They normally focused their linguistic analysis on the individual word or verse, and while later writers such as Abu Bakr MuÌammad al-Baqillani (d. 1013) and ¨Abd al-Qahir al-Jurjani (d. 1078 or 1081) did consider structural issues which were somewhat more comprehensive, a feature they called naÂm, their technique was still limited to the syntax of the individual phrase or verse. The same is likely to be true for earlier writings no longer extant, despite their suggestive titles, such as the work entitled NaÂm al-Qur}an by one al-Îasan b. YaÌya al-Jurjani, which is preserved in quotations by eleventh-century Muslim exegetes.41 The field of Quranic exegesis developed on the basis of numerous and varied foundations; whatever the contribution of the works being translated into Arabic in the ninth century, it did not impel authors to examine the Qur}an’s macroscopic structural aspects. Thus familiarity with current literary norms was a necessary, but not a sufficient factor, in causing the development of a consistent interest in the structure of Scripture. This is also apparent upon examination of the work of other Jewish exegetes writing in Arabic, roughly contemporaneous with those discussed above. For example, Ya¨qub al-Qirqisani did not include structural explanations like 40
See Ibn Qutayba’s analysis of the refrain in suras 54 and 55 in Abdallah Ibn Qutayba, Ta}wil Mushkil al-Qur}an, ed. Ahmad Saqr (Cairo 1954), 185–6. See Angelika Neuwirth, Studien zur Komposition der Mekkanischen Suren (Berlin 2007), 175–321. 41 See Walid Saleh, The Formation of the Classical Tafsir Tradition: the Qur}an Commentary of al-Tha}labi (Leiden 2004), 136–7, 250. The above discussion is based on correspondence with Professor Saleh. 471
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
those of the Jerusalem Karaites in his writings, despite having been deeply immersed in the literature of the period.42 Nor does he devote consideration to compositional questions such as those raised by Sa¨adiah Ga}on. Such discussion could have been quite organically included in the introduction and list of exegetical principles which he included at the beginning of his Pentateuch commentary, Kitab al-riya∂ wa-l-Ìada}iq.43 However this is not the case and Qirqisani is a telling exception to the motivating factors which I describe above. Despite Qirqisani’s wide reading and broad familiarity with current literary style, and despite the fact that the list does indeed respond to many current-day issues of polemics and theology, his list includes few principles which could be asserted to deal with biblical structure. Given that other exegetes exposed to the new Arabic literary norms, whatever their faith, did not necessarily develop a structural method of exegesis, there were clearly other factors motivating those who did adopt the approach. Inherent Karaite interest in context-based approaches One such additional motivation, particular to Karaites, was dictated by the Karaite religious platform, namely, their critical approach to the Oral Law, and to rabbinic literature in general. They believed that these weighty and venerated collections of scholarship had to be selectively examined on the basis of rational analysis, in order to determine the acceptability of any particular view on its own merits rather than on those of its author. This rejection of large bodies of biblical interpretation left a vacuum, and in order to remedy the lack, Karaite scholars had of necessity to place increased emphasis on alternate methods by which to derive meaning from Scripture. For this reason, interpretation on the basis of close textual analysis and reliance on context for interpretation became fundamental methods for them. 42 On Qirqisani and his varied reading, see Bruno Chiesa, Creazione e caduta dell’uomo nell’esegesi giudeo-araba medievale (Brescia 1989), 41–4. The relationship between the exegesis of the Karaites of Jerusalem in the late tenth century and that of Qirqisani is complex. While it is clear that the later exegetes put many of the exegetical principles recorded earlier by Qirqisani into use, and quoted his exegetical views on specific verses, their model of exegesis was quite different from his. 43 Neither the introduction nor the commentary have been published; see note 33.
472
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
This emphasis on context is already apparent in early Karaite texts. The emphasis on the importance of context in the works of Daniel al-Qumisi, an important early Karaite and ninth-century exegete, likely served as one of the foundations for what later becomes the explicit structural analysis apparent in the late tenth century. Al-Qumisi originated in the area of the Caspian Sea, spent time in Khurasan, and moved to Jerusalem towards the end of the ninth century, where he established the Karaite scholarly community in Jerusalem.44 In his exegesis, which may have encompassed the entire Bible, al-Qumisi refers consistently to neighbouring verses and sections in his exegetical comments, with Hebrew and Arabic phrases referring to continuity, such as ki katuv aÌarav, maktub fawqahu, ve-hinne katuv and ve-aÌarei khen. Such discussion emphasizes the connections between the verses, and employs context as a method of deriving meaning.45 The emphasis on context-based exegesis was also an important area of polemic for al-Qumisi, a leader active during the coalescence and self-definition of the Karaite movement. He frequently inveighs against those, both Rabbanite and Karaite, who disregard the context of biblical verses and interpret verses in opposition to their literal sense, using the method he terms ta}wil. Indeed, polemical interchanges are one reason for the concomitantly growing emphasis on contextually based interpretation during this period, in both the Karaite and Rabbanite camps.46 Comparison of al-Qumisi’s attention to context with that of the later Jerusalem exegetes reveals a possible line of development. 44
For the most up to date current biography of al-Qumisi, as well as a comprehensive list of all identified fragments of al-Qumisi’s compositions, see Nehemia Gordon, ‘Does Scripture Really Only Have One Meaning? A Study of Daniel alQumisi’s Exegetical Approach in Pitron Shnem ¨Asar’, unpublished M.A. thesis (Hebrew University of Jerusalem 2006). 45 For an overview of al-Qumisi’s ‘linguistic-contextual’ orientation, see Meira Polliack, ‘Major Trends in Karaite Biblical Exegesis in the Tenth and Eleventh Centuries’, in idem (ed.), Karaite Judaism: A Guide to its History and Literary Sources (Handbuch der Orientalistik, Abt. 1: Der Nahe und Mittlere Osten 73. Leiden/ Boston 2003), 363–413, at 372–9. For detailed analysis of al-Qumisi’s use of context to draw exegetical conclusions, see Gordon, ‘Does Scripture Really Only Have One Meaning?’, 52–4. 46 See Gordon, ‘Does Scripture Really Only Have One Meaning?’, 60–4; Haggai Ben-Shammai, ‘The Karaite Exegete and his Rabbanite Environment (Hebrew)’ (paper presented at the Ninth World Congress of Jewish Studies, Jerusalem, 1988), 52–3. For al-Qumisi’s comment on Mal. 2:9, where he interprets the verse as an accusation against Rabbanites and their overly allegorical interpretations see Daniel al-Qumisi, Pitron Sheneim ¨Asar, ed. Isaac Markon (Jerusalem 1957), 78. 473
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
Karaite exegetes in Jerusalem, active some three generations following al-Qumisi, can be said to have ‘abstracted’ or ‘theorized’ his technique: while al-Qumisi explicitly described the connections between verses, verse by verse, they strove to characterize entire sections of text and their interconnections. This approach, initially a practical necessity for a new religious movement, became an entrenched method, and indeed, a distinguishing feature of Karaite exegesis in the tenth and eleventh centuries. In this way Daniel al-Qumisi was responsible for more than just the physical and practical establishment of Jerusalem scholarly community; his emphasis on the importance of context for exegesis provided one of the foundations for a mainstay of their exegetical method. Traditional Bible study and Masoretic occupations The motivation to adopt a structurally-based approach cannot be delimited by internal Jewish ideological platforms such as Karaism: an additional motivation, which can only be mentioned briefly here, can be traced to deep-rooted pan-Jewish trends, and perhaps, the connection of the Karaite movement to such groups. For one, the rabbis had long carried out the custom of reading the Bible continuously, verse by verse,47 a practice which no doubt encouraged and developed a contextual understanding, complementing the more wellknown atomization of the midrashic approach. In providing structural descriptions in written form in their commentaries, the Karaites may merely have been recording a technique of explanation which had long been carried out orally. Furthermore, such techniques were all the more natural given the apparent close involvement of Karaites with Masoretic circles, which were developing in the eighth century in both Babylonia and Palestine.48 While earlier generations had no doubt devoted themselves to study of scripture, the Masoretes were more thorough than ever before — and their thoroughness contrasted to the minor role that Bible study played in the geonic milieu, until the period of Sa¨adiah Ga}on.49 Scholars who occupied themselves with the text to the extent demonstrated by the Masoretes were no doubt highly aware 47
See note 9. Geoffrey Khan, The Early Karaite Tradition of Hebrew Grammatical Thought (Leiden 2000), 11–25. 49 Brody, Geonim of Babylonia, 300. 48
474
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
of Scripture’s macroscopic aspects, all the more so immediate issues of context. It is likely that the thorough Masoretic knowledge of the text, in addition to traditional oral methods of Bible study, also contributed to the structural-contextual understandings conveyed explicitly in written form in the Karaite commentaries. A response to polemic? The need to respond to polemic, always a ‘motivating’ genre, could have provided an additional factor motivating the Karaite structural approach, as well as Sa¨adiah Ga}on’s attention to the topic of order in the Bible.50 Perhaps the evaluation of scripture according to contemporary literary style was a conscious move, and was intended to assert the superiority of the Jewish scriptures as a response to polemic? After all, polemic against both Jewish and Christian scriptures was a well-entrenched, long-lived and protean genre, and during the early centuries of Islam in particular, Muslims and non-Muslims alike devoted attention to the Bible and the New Testament, whether in defensive manoeuvres intended to provide support for their own theological claims, or in active polemical stances.51 Jews living in the Arabophone sphere were familiar with these works. For example, it could be suggested that such discussions were a Jewish response to the Muslim doctrine of the inimitability of the Qur}an, which included polemics against earlier scriptures. That is, that Muslim assertions regarding the beauty or miraculous nature
50 This, in accordance with the statement of Hava Lazarus-Yafeh that ‘We should, however, consider a great part of Judaeo-Arabic medieval literature to be both explicit and implicit attempts to refute Islam’ (Hava Lazarus-Yafeh, Intertwined Worlds, Medieval Islam and Bible Criticism [Princeton 1992], 6–7). 51 The literature on polemic regarding Scripture is vast. For discussion of the pre-Islamic roots of polemic, see Lazarus-Yafeh, Intertwined Worlds, 130–5; Sarah Stroumsa, Freethinkers of Medieval Islam: Ibn al-Rawandi, Abu Bakr al-Razi, and their impact on Islamic thought (Leiden 1999), 142–67. On the polemics of freethinkers, see Stroumsa, Freethinkers, 193–213; Saadiah Gaon, Saadia’s Polemic against Hiwi al-Balkhi, a Fragment Edited from a Genizah ms, ed. Israel Davidson (New York 1915). For Muslim polemics against Jewish and Christian Scriptures, see Lazarus-Yafeh, Intertwined Worlds, 19–47; Camilla Adang, Muslim Writers on Judaism and the Hebrew Bible (Leiden 1996), 192–255. For bibliography, see Moshe Perlmann, ‘A Selective Bibliography: Polemic between Islam and Judaism in Medieval Times and Related Topics (Hebrew)’, in Hava Lazarus-Yafeh (ed.) Muslim Authors on Jews and Judaism (Jerusalem 1996), 155–60. (includes many English entries).
475
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
of the Qur}an impelled Jews to make similar assertions about their own scriptures, and that by asserting that the Bible was ‘arranged in the best possible order’ as does Yefet b. ¨Eli, exegetes were hoping to demonstrate the literary quality of the Bible to be equivalent or superior to that of the Qur}an. Such a claim, in a work of exegesis intended for an internal audience, could have served to strengthen the resolve of Jewish readers, and would have complemented the admittedly spare genre of explicit offensives against the Muslim theological principle.52 This suggestion is difficult to maintain. For one, the characteristic of the Bible conveyed by these Karaite exegetes — inherent organization and order — would not have addressed the polemical claims being put forth at the time, which centred by and large on biblical content rather than form. Nor would they have served as an assertion of equivalence to the inimitability of the Qur}an: during the period discussed, the claim of i¨jaz had little to do with stylistic aspects of the Quran, and was still based largely on claims regarding the Qur}an’s miraculous content, given its prophet’s illiteracy. The claim that it was the literary style of the Qur}an that was inimitable was developed consistently only beginning in the late tenth century,53 by which time the ‘structural approach’ had already become a well-entrenched feature of Karaite exegesis. Furthermore, even once the doctrine of i¨jaz came to be founded on stylistic features of the Qur}an, such stylistic features included by and large microscopic elements of style, such as those found in poetic works, and were quite different from the macroscopic organizing elements pointed out, for example, by the Karaite exegetes, or the type of organized composition intended by the ‘well-crafted language’ (Prov. 25:11) discussed by Sa¨adiah Gaon. Comparison of the structural analysis of the Karaites with, for example, the Kitab al-MuÌa∂ara wa-l-Mudhakara (The Book of Conversation and Discussion) of the twelfth-century Spaniard Moses b. Ezra, demonstrates the type of claim that could be expected in a work on the Bible meant to compete with the claim of i¨jaz. There, Ibn Ezra indeed considers the microscopic elements of the Bible such as imagery and metaphor — similar to what Muslims in the late tenth and eleventh centuries had begun to say about the Qur}an.54
52
See Sklare, ‘Responses to Islamic Polemics’. Gustave von Grunebaum, A Tenth-Century Document of Arabic Literary Theory and Criticism (Chicago 1950), xvi–xvii. 54 See discussion in Goldstein, ‘Structural Analysis’, 365–7. 53
476
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
The role of Muslim polemic in motivating Jewish attention to the structure and organization of the Bible in the tenth century East and Levant, then, seems minor. Comparison with Christian polemical engagement with Muslims provides a parallel which supports this conclusion. Christians had been forced to contend with challenges to literary aspects of their Scripture since their very beginnings in the Hellenistic-Roman period, and many early Christian sources devote themselves at least in part to such discussion.55 Yet despite this earlier interest, Christians writing in the Muslim period evince little interest in literary aspects of the text, and tend neither to relay such discussions from earlier periods nor to compose their own.56 Muslim-Christian polemic regarding Scripture centred on doctrine deriving from the content of scripture, rather than scriptural style.57 In reality, the development of the i¨jaz doctrine likely occurred in the opposite direction, with the Muslim theological doctrine developing in the ninth and tenth centuries on the backdrop of challenges from Christian theologians.58 It is interesting to note, for example, that the ninth-century exegete Theodore bar Koni, living in Baghdad and likely familiar with Muslim theological doctrine, has no trouble discussing the composition of the books of the prophets, explaining that later compilers reconstructed them in random order on the basis of ‘what they found’, and furthermore, admits that ‘alteration over time’ occurred in the books.59 Here Theodore continues a trope found in Christian tradition — the 55
See Eduard Norden, Die antike Kunstprosa, 2 vols. (Darmstadt 1958). One exception to this tendency among Christians writing in Arabic in the Muslim East and Mediterranean is discussed in Joseph Sadan, ‘Biblical Poetics Commented by Jews, Christians and Muslims and the Composition by al-Harith ibn Sinan (Hebrew)’, in R. Zur and T. Rosen (eds), Israel Levin Volume (Tel Aviv 1994), 195–240, at 223–7. 57 On Christian polemical composition in Syriac, see Sidney Griffith, ‘Disputes with Muslims in Syriac Christian Texts: From Patriarch John (d. 648) to Bar Hebraeus (d. 1286)’, in B. Lewis and F. Niewöhner (eds), Religionsgespräche im Mittelalter (Wolfenbütteler Mittelalter-Studien 4, Wiesbaden 1992), 251–73. Regarding Christians writing in both Syriac and Arabic, see Sidney Griffith, ‘The Prophet Muhammad, his Scripture and his Message according to Christian Apologies in Arabic and Syriac from the First Abbasid Century’, in T. Fahd (ed.), La vie du prophète Mahomet. Colloque de Strasbourg, 1980 (Paris 1983), 99–146. 58 Sidney Griffith, The Church in the Shadow of the Mosque: Christians and Muslims in the World of Islam, Jews, Christians, and Muslims from the Ancient to the Modern World (Princeton, NJ 2007), 97. 59 Theodore Bar Koni, Book of Scholia, ed. Addai Scher (Paris 1910–54), I, 234. 56
477
‘ARABIC COMPOSITION 101’ AND THE EARLY DEVELOPMENT
Greek Bible exegesis written in Antioch, for example60 — and conveys it despite the changed context of the statement and the obvious danger, in the Muslim setting, of opening up questions of textual falsification (taÌrif ). *** New approaches to the Bible, then, resulted from the Jewish participation in the surrounding Arabophone culture in the ninth and tenth centuries, a phenomenon which brought revolutionary changes in so many areas of Jewish life and scholarship. Here, I have identified one type of exegetical innovation among many, and I have identified a number of contributing factors. The fact that Jews interpreting the Bible in Arabic read their Scripture in a way which was shaped by contemporaneous literary norms is to be expected: throughout the centuries, Bible exegetes and interpreters have made efforts, more or less justified, to explain the structure and usage of the Bible on the basis of literary techniques with which they were familiar.61 The authors above, whether consciously or unconsciously, only did the same. The issues they discussed were adopted by many later exegetes, and were transmitted into the works of Bible exegetes writing in Hebrew.62 In this way, literary norms imported from classical scholarship and crucial in medieval Arabic composition — the rules of Arabic Composition 101 — have shaped Bible exegesis up until the present day. Address for Correspondence: [email protected] Department of Arabic, Bar Ilan University, Ramat Gan, Israel
60 See Diodore’s discussion of the ‘collection’ of the book of Psalms in similar terms in Diodore of Tarsus, Diodore of Tarsus: Commentary on the Psalms, ed. Thomas Hill (Atlanta 2005), 3–4. 61 See note 38 above. The observation of Meira Polliack regarding the discussion of the mudawwin in tenth-century Judaeo-Arabic exegesis in light of contemporaneous practices of anthologizing could now be added to this list (Polliack, ‘Major Trends’, 401–2). 62 For discussion of the impact of Judaeo-Arabic exegesis, both Karaite and Rabbanite, on later Bible exegesis such as that written in Iberia, see Daniel Frank, Search Scripture Well: Karaite Exegetes and the Origins of the Jewish Bible Commentary in the Islamic East (Leiden 2004), 248–57.
478
Journal of Semitic Studies LV/2 Autumn 2010 doi: 10.1093/jss/fgq009 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ* OLIVER KAHL UNIVERSITY OF MANCHESTER
Abstract This article deals with two antidotes which form part of an hitherto unpublished treatise on ‘empirical drugs’ (mugarrabat), composed around the middle of the twelfth century CE by the master physician Ibn at-Tilmi∂: the so-called faruq theriac and the viper pastilles. These two antidotes, which ultimately depend on Greek prototypes, circulated in different Arabic versions and belonged to the stock of medieval Islamic pharmacy. They are edited and translated here on the basis of Ibn at-Tilmi∂’s ‘empirical’ recension, which is also distinguished by the fact that the faruq theriac shows traces of an earlier revision made in the ninth century CE by the famous Îunain ibn IsÌaq; moreover, the prescription of the viper pastilles is based on early, precanonical antecedents. The article concludes with a glossary of terms and a botanical register.
One of the great preoccupations of ancient and medieval pharmacy was the development of the theriac, a universal antidote originally formulated by the Greeks. The earliest extant texts on the subject, Nicander of Colophon’s (d. 130 BCE) two poems entitled Alexipharmaca and Theriaca, still only deal with different kinds of poisonings and their treatment; but legend has it that Mithridates VI Eupator, king of Pontos (reg. 120–63 BCE), already conceived the idea of a prophylactic panacea, mixing a variety of ‘proven’ antitoxic components into a single preparation which became famously known under the name of mithridatium.1 Andromachos of Crete (fl. mid first century CE) brought the number of components to sixty-four, notably adding viper flesh, and thus provided Galen (d. c. 216 CE) with a prototype which he incorporated, under the title galßnj, into his * The system of transliteration used in this article is that of the Deutsche Morgenländische Gesellschaft. 1 On the subject in general see the excellent monograph of G. Watson, Theriac and Mithridatium. A Study in Therapeutics (London 1966). 479
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
toxicological writings.2 The Galenic recension(s) formed the basis of all subsequent handlings of the theriac, not least among the Arabs who, of course, in the vast field of poison literature also drew heavily from Indian sources.3 So highly valued was the theriac in the Eastern Roman Empire that in 667 CE an embassy from Byzantium presented the Tang emperor Kao-tsung (reg. 649–83 CE) with a sample of it, apparently in pilular form, and the theriac was also known in Canton during the Sung era; the Greek term qjriaká4 appears in the Chinese sources as tê-ya-ka.5 In the Arabic tradition, the term qjriaká occurs already in early Islamic poetry in the form diryaqa as a metaphorical name of wine,6 but Galen’s ™ qjriak® (ântídotov) was only adopted into the Arabic language as a technical term in the form diryaq or, usually, tiryaq (pl. tiryaqat) during the translation movement in the ninth century CE. Îunain ibn IsÌaq (d. 873 or 877 CE) himself translated into Arabic the Galenic writings Perì ântidótwn,7 Perì qjriak±v pròv Píswna8 and Perì qjriak±v pròv Pamfilianón,9 and he also composed a book of his own on the subject.10 In its Arabic garb the theriac is often accompanied by the epithet al-faruq ‘the one that distinguishes’ (scil. between life and death),11 and the ‘canonical’ recension (in three variant transmissions) is given by Ibn Sina (d. 1037 CE).12 The core function of the theriac, as already indicated, was to neutralize 2 See Medicorum Graecorum Opera quae exstant, ed. C.G. Kühn, 1–26 (Lipsiae 1821–33), XIV, 82–105 [hereafter MGO]; cf. also note 65 below. 3 See e.g. M. Ullmann, Die Medizin im Islam (Leiden 1970), 321 and 324–5 [hereafter Ullmann, Medizin]. 4 qjriaká and the related ™ qjriakß are derivatives of qßr ‘bête de proie, bête sauvage’ and hence denote, by extension, ‘qui concerne les bêtes venimeuses [et] s’applique aux antidotes’, see P. Chantraine, Dictionnaire étymologique de la langue grecque (Paris 1968), 435. 5 See J. Needham et al., Science and Civilisation in China, 1– (Cambridge 1954–), I, 205; cf. also now C. Nappi, ‘Bolatu’s Pharmacy Theriac in Early Modern China’, Early Science and Medicine 14 (2009), 737–64. 6 See Ullmann, Medizin 321 note 2. 7 Kitab al-Adwiya al-muqabila lil-adwa}, see Ullmann, Medizin 49 and F. Sezgin, Geschichte des arabischen Schrifttums, 1– (Leiden 1967–), III, 121 [hereafter GaS]. 8 Kitab at-Tiryaq ila Bisun, see Ullmann, Medizin 49 and GaS III, 121; also L. Richter-Bernburg, Eine arabische Version der pseudogalenischen Schrift De Theriaca ad Pisonem (Göttingen 1969). 9 Kitab at-Tiryaq ila Bamfuliyanus, see Ullmann, Medizin 49 and GaS III, 121. 10 Entitled Kitab at-Tiryaq (lost), for some quotations see Ullmann, Medizin 327 note 2. 11 Compare note 63 below. 12 al-Qanun fi †-†ibb, 1–3 (Bairut n.d. [reprint of the Bulaq 1877 edition]), III, 312–3 [hereafter Ibn Sina, Qanun].
480
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
the effects of poisonous substances on the body; it could be taken in response to an incident of poisoning but also on a regular basis to build up ‘immunity’; and the step from inventing a general antidote to proclaiming a cure-all was always a small one. A drug as sophisticated as the theriac was difficult to make, hard to come by, and in most cases surely a prerogative of the ruling classes; its undisputed status in premodern medico-pharmacological literature does, if nothing else, corroborate suspicions about the role which poison played in the political lifes of those kings and rulers.13 Just as in Greek literature, the Arabic transmission of the theriac formula was not uniform: numerous versions circulated in medieval Islamic pharmacy, though on the whole variations were limited to metrological discrepancies and/or substitute components. Ibn Sina’s aforementioned account of the theriac is a fairly representative example of the nature of suchlike variations, both in terms of quality and quantity. Yet there are certain Arabic versions of the theriac which deserve to be dealt with in greater detail, and one of them, that of the master Ibn at-Tilmi∂ (d. 1165 CE), is edited and translated in this paper. Ibn at-Tilmi∂’s theriac recension is important for several reasons: first, because Ibn at-Tilmi∂ was one of the most influential physicians of his century and perhaps the last truly great exponent of Eastern Arabic pharmacology; second, because he incorporated this recension, not into his dispensatory (aqraba∂in),14 but rather into a treatise on empiricals (mugarrabat) which, moreover, has not yet been published; and third, because his ‘empirical’ recension is partly based on an earlier revision made by the famous Îunain ibn IsÌaq and probably taken from the latter’s Kitab at-Tiryaq which regrettably, as already mentioned, is now lost. As regards the other antidote published below, namely the so-called viper pastilles, they, too, belonged to the stock of both Greek and Arabic pharmacy; they represent one of the key components of the theriac, and details about their fabrication are therefore often appended to the theriac formulas: Ibn Sina does this, and so does Ibn at-Tilmi∂. But here again, as in the case of the theriac and much for the same reasons, Ibn at-Tilmi∂’s recension stands out; more specifically, a comparison between Ibn Sina and Ibn at-Tilmi∂ clearly shows that the latter’s version of the viper pastilles is either based directly on the Greek version of Galen or else copied from an early,
13 Compare also P. Johnstone, ‘Summ’, EI 2, 1–12 (Leiden 1960–2004), IX, 872–3. 14 For which see O. Kahl, The Dispensatory of Ibn at-Tilmi∂ (Leiden 2007).
481
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
precanonical Arabic rendition of it, which quite possibly formed part of Îunain’s lost Book on Theriac as well.15 The edition of the Arabic texts containing the two antidotes is based on MS orient. A1996 of the Forschungs- und Landesbibliothek Gotha.16 The undated manuscript includes two works of Ibn atTilmi∂: an incomplete copy of his dispensatory (aqraba∂in) on fols. 1b–27a; and a copy of his empiricals (mugarrabat) on fols. 27b–37a (the two passages edited here cover fols. 28a,9–29b,19). The manuscript is written in medium-sized, stretched, unvocalized nasÌ. The unnamed copyist dots the letter ىeven when it represents a (e.g. الي, علي, اقوي, ;)اثنيonce, he separates a word at the end of the line (i.e. [ قرد | ماناloanword] on fol. 28b,2f.). Finally it should be noted that the following metrological units occur in the texts: mi†qal = 4.46g, uqiya ~ 33g, ra†l ~ 400g, and qis† ~ 600g; semispecific measures include aÒabi¨ ‘fingers’ and gillauza ‘hazelnut’; nonspecific terms are guz} ‘part’ and miqdar ‘amount/ quantity/width’. ١ صنعة الترياق الكبير المعروف بالفاروق صورته الحاصلة عن امتزاج بسائطه17وجميع ضماناته إنما تصدر عن خاصية تتبع بتقويته الحار الغريزي والروح فقيام المفسدات حارة كانت أو باردة ويدفع الفضول 18 وأول أوقات صلوحه للاستعمال بعد سنة فإنه بعد ستة أشهر إنما،الردية ويخففها إلى عشر سنين في البلاد الحارة وإلى عشرين20 إلى سنة واحدة ويزيد19يسمى طفل ويبطل ترياقه21في البلاد الباردة ثم لا يظهر ضعفه مدة عشر سنين في البلاد الحارة قال، في البلاد الباردة فيصير كأحد المعاجين المنحطة عن درجته22بعد ثلاثين أجود النسخ الذي يعمل عليها الترياق و]أ[فضلها و]أ[كملها أن يؤخذ من:حنين الأقراص المتخذة من العنصل ثمانية وأربعين مثقالا فإن أندروماخس وجميع من كان 15
For Ibn Sina’s version see Qanun III, 313–4; for Galen’s original see note 86 below. 16 Registered by W. Pertsch, Die arabischen Handschriften der Herzoglichen Bibliothek zu Gotha, 1–5 (Gotha 1878–92), IV, 32. 17 MS يتبع 18 MS اسما 19 MS طفلا 20 MS تزيد 21 MS + ٯصىر 22 MS + وستين 482
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
بعده كانوا يلقون من هذه الأقراص هذا المقدار خلا ديمطر]يـ[ـوس المتطبب الذي كان على عهد جالينوس فإنه كان ينقص من هذا المقدار مثقالين ويؤخذ 23من الأقراص المعروفة باندروحورون ومن الفلفل الأسود والأفيون من كل واحد أربعة وعشرون مـ]ـثـ[ـقالا فأما الدارصيني 24فكان جميع القدماء يلقون منه اثنا عشر مثقالا إلا أن مغنيس 25الطبيب الذي كان رئيس الأطباء على عهد جالينوس كان يلقي منه أربعة وعشرون مثقالا ويؤخذ 26من الورد وبزر السلجم البري وأسقورديون وأصل السوسن الآسمانجوني والغاريقون ورب السوس ودهن البلسان الفائق من كل واحد اثنا عشر مثقالا ومن المر والزعفران والزنجبيل 27والريوند والبنطافلون 28والفوتنج الجبلي وفراسيون 29وبزر الكرفس الجبلي وأسطوخودس 30وقسط مر وفلفل أبيض ودار فلفل ومشكطرامشيغ 31من كل واحد ستة مثاقيل ومن اللبنى وبزر الكرفس وسـ]ـسـ[ـاليوس وثلاسفيس 32ونانخواه وكمادريوس وكمافيطوس وعصارة لحية 35 التيس وميبخوشة 33وساذج وجنطيان 34وبزر الرازيانج وطين مختوم وقلقطار محرق وحماما ووج وحب البلسان وهوفاريقون وفوة وصمغ وقردمانا وأنيسون وأقاقيا من كل واحد أربعة مثاقيل ومن الدوقو 36والبار]ز[د والحمر والجاوشير وقنطريون وزراوند من كل واحد مثقالين وفي أمر الزراوند نظر وذلك أن القدماء كلهم يلقون 37الزراوند المعروف 38بحريري إن الزراوند المدحرج ولي بأن يلقى في الترياق و]أ[ما الجندبيدستر فقوم كانوا يلقون مثقالين وقوم آخر يلقون أربعة مثاقيل وأما السكبينج فكان أندروماخس يلقي منه أربعة مثاقيل وكان سانوقراطيس 39وديمقراطيس وماغنس يلقون منه مثقالين ومن عسل النحل الذي يلقطه النحل من نبات الحساء بعد أن يسخن وتؤخذ رغوته عشرة أرطال ومن الشراب العتيق الطيب الرائحة الحاد الطعم ياخد MS الدارالدارصيني MS معتنيل MS ياخد MS الترنحبيل MS البنطاقلون MS قراسىون MS اسطواخودس MS مىىكطرامسير MS بلاسقس MS مىىجيوسه MS خيطان MS علي النصف MS + الدوقور MS يقلقون MS ىالىىكىىلي ]؟[ و MS + ىسابوفراطيس MS 483
23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
قسطين فهذه هي الأدوية التي يتخذ منها الترياق وعددها أربعة وستون سوى العسل والشراب وأما 40صنعته فينبغي أن يذوب وينقع الذي يستنقع وتدق اليابسة وتنخل وتعجن بالعسل ويصير في إناء رصاص قلعي أو في إناء فضة و 41ينبغي أن يترك فيه موضع خال يتنفس فيه الدواء ،قال حنين :ويمكنك إذ هو تنفس أن تستعمله سريعا وذلك أنه يحتاج إلى مدة طويلة ينضج فيها حتى يمكن أن يستعمل لأنه يمكث اثنا عشر سنة حتى ينضج وبعض الناس يحب أن يستعمله وهو طري بعد ويستعمل في مداواة من يلسعه شيء من الهوام بعد أن يعنق لأن سم هذه لا يحتاج إلى دواء على مثال هذه القوة وحد الترياق أن يكون حديثا طريا إلى ثلاثين سنة وأن يكون عتيقا إلى ثمانين سنة فإذا جاوز هذا الحد ضعف وصار إما أن لا يعمل شائبة 42وإما أن يعمل عملا ضعيفا ،ذكر امتحان الترياق :الترياق يمتحن من ثلاثة وجوه ،الوجه الأول أن تعمد 43إلى إنسان قد سقي دواء مقيئا قويا كالخربق الأبيض والسقمونيا أو غيرهما من الأدوية القوية فتسقيه من الترياق مقدار نصف مثقال فإن تعطل به عمل الدواء المسهل فاعلم أنه ترياق جيد 44،والوجه الثاني وهو أن جالينوس يمتحنه به أن يأخذ ديكا بريا لم يترف 45في البيوت يابس البدن فيرسل عليه بعض ذوات السموم لتنهشه ويسقيه من ذلك الترياق إما قبل أن يرسل وإما ساعة ينهش فإن أفلت وعاش فاعلم أن ذلك الترياق فائق وإن عطب فاعلم أن به ترياق سوء لا عمل له، والوجه الثالث وهو الذي امتحن به الترياق قوم ممن قرب عهده من الأطباء بأن يأخذ دواء 46قتالا 47إما أفيون وإما غيره فيسقي 48منه ديكا أو كلبا أو غيرهما من الحيوان ثم يسقيه 49فيعرف بذلك جودته من رداءته 50،ذكر منافع الترياق :أول منافع الترياق وأشرفها أنه ينفع من الأدوية ومن نهش السموم وينبغي أن يكون مقدار ما يسقيه هولا منه كمقدار جلوزة مع مقدار أربع أواقي ونصف شرابا أعني بالأدوية القتالة الأفيون ]و[الشوكران والبنج والذراريح وسائر ما أشبه ذلك وأعني بذوات السموم
ما + ليس + شاىنه يعهد ضعيف ىىرفي دواا قبالا فينقي يسقىيه رداته
484
MS MS MS MS MS MS MS MS MS MS MS
40 41 42 43 44 45 46 47 48 49 50
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
الأفاعي والحيات التي يعتري من تنهشه 51العطش والعقارب والثقابين والحيات التي يقال لها حيات الماء وحيات البر 52والبلوطية والمعفنة 53للدم والرتيلاء ،وللترياق منافع أخر]ى[ يفعلها بعد هذه وهو أن يصلح ما يحدث في البدن من الفساد الشبيه ]بـ[ـسموم ذوات 54السموم وينفع من السعال الحديث والعتيق ومن يوجعه شيء من صدره أو أضلاعه 55ويشربه من كانت به حمى مع عسل وينفع أيضا من به نفخة من معدته أو مغس في أعضائه أو وجع القولنج وينبغي أن يسقي منه 56هولا مقدار مثقال ،ومن منافعه أيضا أن يكسر الشهوة القوية ويقوي الشهوة الضعيفة وينبغي أن يسقي ]منه[ هولا إما بالماء أو بالشراب و 57ينبغي أن الماء والشراب من أربع أو]ا[قي ونصف ولا أقل من ثلاث أواقي ،ومن منافعه أن يدفع النافض الذي يكون بأدوار البرد والقيء العارض في ابتداء نوائب الحمى إذا سقي على ما وصفناه ،وهو مع هذا يدر الطمث ويخرج الأجنة الموتى من الأرحام إذا شرب منه مثقال بعقيد العنب أو بماء العسل الذي طبخ فيه مشكطرامشيغ 58وسذاب ،وينفع أيضا من الهيضة والخفقان فهذا الذي من منافع النوف 59هو أعظم منافعه وأجلها قدرا ،وله مع هذا منافع أخرى كثير]ة[ هي تابعة للمنافع التي ذكرنا]ها[ صورة ،من ذلك أنه 60 إن كان ينفع من الصرع فهو لا محالة ينفع من الصداع القديم ومن الشقيقة والدوار وإن كان يذهب بأورام الكبد والطحال لا محالة يفتح السدد وإن كان ]يحبس[ ما يكون فيها من البول فهو لا محالة يحبس استطلاق البطن ،وجميع ما وصفناه من منافع الترياق هو مما ينتفع به في مداواة 61الأمراض وله مع هذا منافع أخر]ى[ في حفظ الصحة عظيمة وذلك أنه من شرب منه في وقت صحة مر 62به آفات كثيرة تحدث من الأسباب الباردة بمنزلة ما يحدث من نهش ذوات السموم وشرب الأدوية القتالة والوبأ العارض من فساد الهواء وفساد الماء وهو مع هذا إذا شرب يمنع من الأسباب المتقادمة تىنهشه MS البن MS الباعنته MS ادوات MS وينبغي ان يسقي منه هولا مقدار MS + منه MS + لا MS + مسكطراسيرا MS الناف MS الشقبه MS مداوات MS من MS
485
51 52 53 54 55 56 57 58 59 60 61 62
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
1 The preparation of the great theriac known as faruq63 All its warrants are issued by a special property which takes shape and becomes manifest through a (particular) combination of simple drugs; thus it strengthens the innate heat and the vital spirit, thereby opposing corruptive elements (in the body) — be they hot or cold —, and rejects and dispels vicious residues. The theriac is suitable to be used after a year at the earliest — for it is still called a ‘child’ between (the age of) six months and one year; (then) it grows stronger (over a period of) up to ten years in hot countries and up to twenty (years) in cold countries; after that its weakness does not show for (another) ten years in hot countries, whilst in cold countries its antidote ceases to be effective (only) after thirty (years) — this is when it becomes like one of those electuaries which have been degraded from their (high) rank.64 Îunain says: the best, most excellent and most complete copy for the fabrication of the theriac (states) that one should take forty-eight mi†qal from the pastilles which are made with squills — Andromachos65 and all those who came after him put (precisely) this amount of these pastilles (into the theriac), except the doctor Demetrios, a contemporary of Galen,66 who reduced this amount by two mi†qal; (further) take from the pastilles which are known as }ndrwÌwrwn67 and 63
faruq literally means ‘a thing that makes a distinction between two things’ and hence tiryaq faruq is ‘the most approved sort of theriac and the most esteemed of compounds because it makes a distinction between disease and health’, see E.W. Lane, An Arabic-English Lexicon, 1–8 (Beirut 1968 [reprint]), VI, 2386. 64 According to Ibn Sina, Qanun III, 311 ‘the theriac has a childhood (†ufula), time of growth (tara¨ru¨), prime (sabab), old age (saiÌuÌa), and death (maut) — it becomes a child after six months or after one year; then it enters the time of growth, which ends after ten years in hot countries and twenty years in cold countries; then it stays [like that] for either ten or twenty years; then it declines over [a period of] either twenty or forty years; then, [in the course of] either thirty or sixty years, it is stripped of its antidotal property (tiryaqiya) and [eventually] becomes like one of those electuaries which have lost their rank as antidotes’. 65 Andromachos the Elder, physician to the Roman emperor Nero (reg. 54–68 CE), ‘invented’ the so-called Theriaca Andromachi which is preserved in poetic form by Galen (MGO XIV, 32–42) — Andromachos was the one who brought the number of theriac ingredients to sixty-four, including viper flesh; cf. W. Smith, A Dictionary of Greek and Roman Biography and Mythology, 1–3 (Boston 1867), I, 173 [hereafter Smith, Dictionary]. 66 Demetrios Archiater (fl. second century CE), a physician quoted by Galen (MGO XIV); cf. Smith, Dictionary I, 972. 67 The ‘ideograph’ }ndrwÌwrwn is well-attested in Arabic pharmacological literature (variant spellings) and almost always occurs as the name of a particular kind of 486
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
from black pepper and opium twenty-four mi†qal each; as regards cinnamom, all the ancients put twelve mi†qal from it (into the theriac), only the doctor Magnos, who was head physician in the days of Galen,68 used twenty-four mi†qal of it; (then) take roses, the seeds of the wild rape, wood germander, the root of the sky-coloured iris, agaric, liquorice rob, and first-rate balm oil twelve mi†qal of each; from myrrh, saffron, ginger, rhubarb, cinquefoil, mountain mint, white horehound, the seeds of milk parsley, French lavender, bitter alecost, white pepper, long pepper, and Cretan dittany six mi†qal each; from storax, celery seeds, moon carrot, shepherd’s purse, visnaga, wall germander, ground pine, salsify sap, Greek spikenard, laurel, gentian, fennel seeds, sealing bole, burnt yellow vitriol, grape ivy, sweet flag, balm seeds, hypericum, madder, gum-arabic, wild caraway, anise, and gum-senegal four mi†qal each; from carrot, galbanum, Jew’s pitch, opopanax, centaury, and birthwort two mi†qal each — in the case of birthwort it should be noted that all the ancients used the (kind of) birthwort which is known as ‘silky’, (but) rolled birthwort is best suited to go into the theriac;69 as regards castoreum, some people put (only) two mi†qal (from it into the theriac), whilst others use four mi†qal; as regards sagapenum, Andromachos70 put four mi†qal from it (into the theriac), whilst Xenocrates,71 Democrates,72 and Magnos73 pastilles (aqraÒ). In an earlier publication I proposed, on the basis of Galen’s statement oîómenov e¤nai botánjn ≠ ti toioÕto t¬n äpl¬n farmákwn, the emendation idarÌurun < ™duxróon ‘hedychrum’, but that does not actually solve the problem, see O. Kahl, Sabur ibn Sahl. Dispensatorium Parvum (Leiden 1994), 212–13; others considered the term to be a corruption of ˆAndrómaxov, or üdrárgurov ‘quicksilver’, or Persian andaruÌun ‘aspalathus’, but that is not very convincing either, see W. Schmucker, Die pflanzliche und mineralische Materia Medica im Firdaus alÎikma des ™abari (Bonn 1969), 93–4. 68 Magnos Archiater (fl. mid second century CE), a physician who wrote on antidotes and is quoted as such by Galen (MGO XIV) — Serapion calls him ‘Rex Medicorum in tempore Galieni’; cf. Smith, Dictionary II, 902. 69 I cannot establish a connection in Greek between birthwort and ‘silky’ (if this is the correct reading). Ibn Sina, Qanun III, 312 says that ‘according to one account (riwaya) rolled birthwort substitutes long birthwort’. 70 On Andromachos see note 65 above. 71 Xenocrates of Aphrodisias (fl. mid first century CE), a physician who wrote some pharmaceutical works and is quoted a few times by Galen (notably MGO XIV) and others; cf. Smith, Dictionary III, 1294. 72 Democrates Servilius (fl. first half of first century CE), a Greek physician in Rome and author of several pharmaceutical works in verse, of which only the titles remain and some extracts preserved by Galen (notably MGO XIV); cf. Smith, Dictionary I, 935. 73 On Magnos see note 68 above. 487
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
used (only) two mi†qal from it; (finally take) ten ra†l of heated and clarified bee honey, the one which the bees collect from the yielding plants, and two qis† of aged, fragrant, sharp wine — these are the ingredients from which the theriac is made, and their number, not counting the honey and the wine, is sixty-four;74 as regards its preparation, one must melt and macerate that which can be macerated, pulverize and filter the dry (components) and knead them together with the honey, (then) put (everything) into a vessel made of lead (and) Indian tin or into a silver vessel, and make sure to leave (some) empty space in the vessel so that the remedy can breathe. Îunain says: you may use the theriac quickly, even as it is breathing; why, if it needs (such) a long time to mature that it has to remain (in the vessel) for twelve-odd years? Well, some people like to use the theriac whilst it is still (very) young, (for example) when treating someone who has been bitten by a small thing that crawled out of a hole — for the venom of such a crawler does not require a remedy as powerful (as the mature theriac); the term(s) of the theriac are that it is (considered) fresh and young up to (the age of) thirty years and old up to (the age of) eighty years — once it has gone beyond that term it becomes weak and is either (considered) spoiled and not used (at all) or it is used less effectively.75 An account of how to test the theriac: the theriac is tested in three ways. The first method is to approach a person who has been given a strong vomitive drug, like white hellebore, scammony, or another powerful substance, and let him drink a quantity of half a mi†qal of the theriac — if the purging agent is neutralized, you know the theriac is good. The second method is how Galen tested the theriac; he took a free-running cock with a dry constitution, one that had not been spoiled by living in houses, and sent a venomous creature to bite it; before sending (the creature) or straight after the bite, he gave the cock some of the theriac to drink — if (such a cock) gets away and lives, you know the theriac is fine; if (the cock) perishes, you know it has a bad, ineffective theriac inside.76 The third method is one which some modern physicians (employ) to test the theriac; they take a lethal drug, such as 74 The copyist of the manuscript erroneously omitted the following eight ingredients: frankincense (kundur), citronella blades (fuqqaÌ al-i∂Ìir), turpentine (Òamg al-bu†m), malabathrum (saliÌa sauda}), Indian spikenard (sunbul hindi ), germander (ga¨da) [at six mi†qal each], valerian (fu ) [at four mi†qal ], and — most importantly — the viper pastilles (aqraÒ al-afa¨i ) [at twenty-four mi†qal ], see e.g. Ibn Sina, Qanun III, 312 (first prescription). 75 On the ‘terms’ of the theriac cf. note 64 above. 76 Compare MGO XIV, 214–5 (Galen).
488
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
opium and the like, and feed that to a cock or a dog or another animal; then they feed it (the theriac), and from this they learn (how to tell) the good from the bad. An account of the benefits of the theriac: the first and most noble benefit of the theriac is its usefulness against (lethal) drugs and against the bites of venomous (creatures); in the face of such threats, the theriac should be administered in a quantity similar to a hazelnut (and drunk) together with four and a half uqiya of wine — by ‘lethal drugs’ I mean opium, hemlock, henbane, cantharides, and others that run along those lines; by ‘venomous creatures’ I mean vipers, snakes whose bites cause (intense) thirst, scorpions, (other) stingers, the snakes that are referred to as ‘water-snakes’ and ‘land-snakes’, the (one that is called) ballu†iya,77 the one that putrefies the blood, and the tarantula. In addition to that, the theriac has other benefits; thus, it remedies the (kind of) corruption in the body which manifests itself as if it were (caused by) the toxin of some venomous creature; it is useful against recent and inveterate coughing, and for someone who suffers from pain in the chest or (between) the ribs; he who is ill with a fever drinks it together with honey; and it is also useful for someone who is afflicted with gastric flatulence, intestinal gripes, or colical pain — in the face of such threats, (the patient) should be given a quantity of one mi†qal from the theriac. A further benefit is that the theriac suppresses excessive appetite and stimulates poor appetite — when faced with this problem, one should drink (the theriac) with either water or wine, whereby the water and the wine should be (measured out) at about four and a half uqiya and (in any case) no less than three uqiya. Among its benefits is that it fights the shivering which accompanies cold paroxysms and (also) the vomiting which occurs at the outset of fever spells if it is drunk as we prescribed.78 The theriac, moreover, makes the menstrual blood flow and drives dead foetuses out of the wombs if one mi†qal of it is drunk with thickened grape juice or with honey-water in which Cretan dittanies and rue have been cooked. It is also useful against violent dysentery and palpitations — this is its most important and most powerful added benefit. But the theriac still has several other benefits which are secondary to the ones illustrated above. For example, if it is useful against epilepsy then, no doubt, it is (also) useful against long-standing headache, hemicrania, and vertigo; 77 ballu†iya is a calque of ö druñnav ‘serpent living in hollow oaks’, see H.G. Liddell and R. Scott, A Greek-English Lexicon (Oxford 1968), 450 with Ullmann, Medizin 336 (who gives a list of Greek snake names in Arabic garb). 78 That is one mi†qal (Ibn Sina, Qanun III, 311: ‘a lupine’ [turmusa]) of the theriac and between three and four and a half uqiya of water or wine.
489
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
)if it removes tumours of the liver and spleen then, no doubt, it (also )opens obstruction; and if it detains urine then, no doubt, it (also restrains bowel movements. All the benefits we described (so far) are relevant to the treatment of diseases, yet the theriac is also immensely beneficial for the preservation of health; this means that whoever drinks from it in a time of health will be spared a lot of hardship arising from (systemic) coldness, just as it occurs after the bite of some venomous creature, after the intake of some lethal drug, or in — the wake of an epidemic caused by a corruption of air and water the theriac, once drunk, keeps all these troubles at bay. ٢ عمل الأقراص المتخذ]ة[ بلحوم الأفاعي إذا أنت صدت الأفاعي فلا تدعها تلبث أصلا ولكن اقطع رؤسها وأذنابها على المكان ولا تؤخر ذلك وليكن ما تقطعه من كل واحد من الطرفين مقدار أربع أصابع ثم اسلخ جلودها وأخرج ما في أجوافها من الثرب والأمعاء وجميع الأشياء وألقها في قدر فخار جديد أو في قدر نحاس مرصعة ثم صب عليها من ماء العيون ما يغمرها وألق عليها شيـ]ـئـ[ـا من ملح وعيدان الشبت ،وبعض الأطباء القريبين العهد يصب عليها في هذا الوقت زيتا ثم يضع ]في[ القدر لحم بلوط ،ودعها حتى ينضج اللحم حسنا ودليلك على نضجه أن ينتثر اللحم من العظم حتى إذا رأيت ذلك فأنزل القدر عن النار وخلص اللحم ونقه من العظام تنقية جيدة وألقه في قصعة فيها شيء من المرق الذي طبخه فيه كيما لا يجف فإذا اجتمع ذلك اللحم كله ونقي من العظام فاعصره عصرا جيدا من المرق الذي كان منقعا فيه ثم قرنه 79وألقه في مهراس واسحقه ورش عليه في وقت سحقك إياه من المرق الذي طبخته فيه ولا تزال تفعل ذلك حتى يبقى مثل المرهم ]ثم[ اخلط معه كعكا مسحوقا ،وكان أندروماخس وهو الذي تلطف لإ ]لـ[ـقاء لحوم الأفاعي في الترياق أهمل الأمر في مقدار ما يحط من الكعك مع لحوم الأفاعي فأما مغنيس وديمقراطيس وجالينوس فإنهم قدروا أن يكون ما يلقى من اللحم ومن الكعك جز]ئـ[ـين سواء وقد رأى بعض الأطباء ممن كان بعد جالينوس أن يكون ما يلقى من الكعك مقدار الربع أو مقدار الخمس ليكون اللحم أغلب فيكون به الخلط أقوى ،وينبغي أن يكون الكعك من درمك نقي غاية النقاء ويكون قد ألقي في عجينه من الخمير والملح مقدارا معتدلا
ىرنه MS 490
79
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
وخبز في التنور بعناية ويفقد ويجفف في بيت جاف ليس بندى أصلا حتى استحكم وليس ينبغي أن يخلط هذا الكعك بلحوم،جفافه وأمكن أن يسحق في هاون لكن ينبغي أن ينتقع في شيء من مرق ذلك اللحم المسحوق80الأفاعي ساعة تسحق 81 ثم يخلط ويعاد سحقها جميعا حتى يختلطا ثم يتخذ من ذلك أقراص كبار فائقة تلك الأقراص في جام زجاج وتجفف في83 بدهن بلسان وتصير82ويدهن المقرص الظل في بيت دفيء وتقلب في كل يوم ويمسح عنها ما يعلوها من درج إن علا 85 بدهن بلسان ولا يزال يفعل بها ذلك حتى يستحكم جفافها ثم تصير84وتدهن فهذه أقراص الأفاعي،في إناء زجاج وتحفظ في بيت دفيء 2 The fabrication of the pastilles which are made with viper flesh86 Once you have caught the vipers do not under any circumstances let them linger, but rather cut off their heads and their tails on the spot and without delay — a width of about four fingers from either end; later strip off their skin and gut them — lining, intestines, everything; then throw the snakes into a new earthen pot or one made of 80
MS يسحق MS قاق 82 MS + لهايده 83 MS يصير 84 MS يدهن 85 MS يصير 86 See MGO XIV, 265–7 (Galen): kaì d® labónta aûtà t¬ç kair¬ç toútwç pr¬ton mèn aût¬n xr® âpokóptein tàv kefaláv te kaì tàv oûràv, tosoÕton âpokóptontav, Üv e¤nai tò métron t±v âpokop±v tessárwn daktúlwn […] e¤ta metà toÕto âpodérein aût¬n ºlon âkrib¬v tò dérma, êzaírein dè kaì tò stéar Üv ãxrjston kaì tà êntósqia †panta, ∂sti gàr t¬n perittwmátwn doxe⁄a. kaì metà toÕto êmbállein aûtà loipòn e÷v ti keramoÕn âgge⁄on, Üv kállista kataskeuasqèn, Æ eîv lébjta geganwménon kal¬v, kaì to⁄v ãnqrazin ânakekauménoiv êpikeímenon, ÿna âknístwv ™ ∏cjsiv aût¬n génjtai. ëcßsqwsan dè ên Àdati pjgaíwç kaì prosembalésqwsan †lev nearoì, kaì ânßqou m® zjroÕ kl¬nev súmmetroi. e¤q’ ºtan ëcjq¬sin aï sárkev kal¬v, tò métron dè t±v ëcßsewv ∂stw soi, ºtan aï ãkanqai xwrisq¬si t±v sarkòv t¬n qjríwn, tóte ânelómenov tòn lébjta âpò toÕ puròv âkrib¬v xÉrihe tàv sárkav t¬n âkanq¬n, kaì leiÉsav aûtàv, ãrtou Üv málista toÕ kaqarwtátou kaì âpò semidálewv t±v kaqarwtátjv genoménou mísge tò súmmetron pròv t®n ânáplasin, ¿sper kaì ö ˆAndrómaxov boúletai. ö gàr Mágnov, kaì ö Damokrátjv kaì métron tì Ürisménou aûto⁄v mígnusqai qélousin ÷son gàr toÕ ãrtou pròv ÷son t¬n sark¬n âpostßsantev, oÀtwv aûtàv sulleioÕsi t¬ç ãrtwç. e¤ta paraxéav tò aΔtarkev toÕ hwmoÕ, oÀtwv ânáplasse summétrouv troxískouv, paraptómenov ên t±Ç ânaplásei, ôpobalsámou ôlígou, kaì ên ski¢ç âpotiqémenov eîv t®n toÕ ºlou farmákou skeuasían fúlatte. 81
491
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
coated copper, cover them with spring water, and add a bit of salt and dill stalks. At that point some modern physicians pour olive oil over the snakes, (and) then throw (peeled) acorn fruits into the pot. (Now) leave the snakes until (their) flesh is worn soft, which you can tell if it separates from the bone; as soon as you see this (happening), take the pot off the fire, clean the meat and carefully remove the bones, and throw it into a large bowl containing some of the broth wherein it was cooked, so that it does not dry out; when all that meat has been cleared from the bones and thus gathered (in one place), press it thoroughly (and thereby draw off) the broth in which it was soaking; now collect the meat, throw it into a mortar, crush it whilst sprinkling upon it some of the broth wherein you cooked it, and keep doing this until (the mash) looks like an unguent; then admix (some) pounded biscuit. Andromachos87 — though he was the one who promoted the inclusion of viper flesh into the theriac — ignores the issue of how much biscuit (exactly) should be added to the meat;88 as for Magnos,89 Democrates,90 and Galen, they reckon that meat and biscuit should be used in equal parts;91 and some of the physicians who came after Galen think that the amount of biscuit should be one fourth or one fifth, so that the meat dominates and therefore the mixture becomes stronger. The biscuit (itself) should be (made) from the finest flour, and an appropriate amount of leaven and salt should be added to the dough; (then) it is carefully baked in a pit, checked, (taken out), left to dry in a room that is not damp or moist at all, and when the biscuit is completely dry it can be pounded in a mortar. It is (however) not recommended to mix this biscuit into the viper meat straight after pounding, but rather to soak it (first) in some of the broth (obtained from cooking) the crushed meat; (only) then admix (the biscuit), and (now) keep pounding the two (ingredients) until they form an even mixture; from this (mass) you make nice large pastilles, which you roll around in balm oil; place those pastilles into a glass cup, (take them) to a warm shady room to dry, turn them over once a day, wipe off the stuff that slowly rises to settle on their (surface), bedaub them with balm oil, and continue treating them that (way) until they are completely dried out; at last put them into a glass vessel and store them in a warm room. These are the viper pastilles. 87
On Andromachos see note 65 above. Ibn Sina, Qanun III, 314 only says that ‘biscuit is admixed according to the various copies (nusaÌ muÌtalifa), but not so in the copy of Andromachos’. 89 On Magnos see note 68 above. 90 On Democrates see note 72 above. 91 Compare note 86 above. 88
492
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
Glossary92 }ndrwÌwrwn [pastilles] 1, n67 acorn (ballu†) 2 agaric (gariqun) 1 agent, purging (dawa} mushil) 1 air, corruption of (fasad al-hawa}) 1 alecost, bitter (qus† murr) 1 ancients (qudama}) 1 animal (Ìaiwan) 1 anise (anisun) 1 antidote (tiryaq) 1 appetite, excessive (sahwa qawiya) 1 appetite, poor (sahwa ∂a¨ifa) 1 baking-pit (tannur) 2 ballu†iya [snake] 1, n77 balm (balasan) 1, 2 bee honey (¨asal an-naÌl) 1 bees (naÌl) 1 benefits (manafi¨) 1 benefits, added (manafi¨ an-nauf ) 1 benefits, secondary (manafi¨ tabi¨a) 1 birthwort (zarawand) 1 birthwort, long (zarawand †awil) n69 birthwort, rolled (zarawand mudaÌrag) 1, n69 birthwort, ‘silky’ (zarawand Ìariri) 1, n69 biscuit (ka¨k) 2, n88 bite(s) (nahs) 1 blood (dam) 1 blood, menstrual (†am†) 1 body (badan) 1 bone(s) [of snakes] (¨aÂm / ¨iÂam) 2 bowel movements (isti†laq al-ba†n) 1 bowl, large (qaÒ¨a) 2 broth (maraq) 2 cantharides (∂arariÌ) 1 caraway, wild (qardamana) 1 carrot (duqu) 1 castoreum (gundbidastar) 1 celery (karafs) 1 centaury (qan†ariyun) 1 chest (Òadr) 1 ‘child’ (†ifl) 1 childhood (†ufula) n64
cinnamom (darÒini) 1 cinquefoil (ban†afilun) 1 citronella (i∂Ìir) n74 cock (dik) 1 coldness, systemic (asbab barida) 1 combination (imtizag) 1 constitution (badan) 1 corruption (fasad) 1 coughing, inveterate (su¨al ¨atiq) 1 coughing, recent (su¨al Ìadi†) 1 crawler (sai} min al-hawamm) 1 creature(s), venomous (∂awat assumum) 1 Cretan dittany (miski†ramasig) 1 cup made of glass (gam zugag) 2 death (maut) n64 dill (sibitt) 2 diseases (amra∂) 1 doctor (muta†abbib) 1 doctor (†abib) 1 dog (kalb) 1 dough (¨agin) 2 drug, vomitive (dawa} muqaiyi}) 1 drug(s), lethal (dawa} qattal / adwiya qattala) 1 drugs, simple (basa}i†) 1 dysentery, violent (hai∂a) 1 electuaries (ma¨agin / ma¨gunat) 1, n64 elements, corruptive (mufsidat) 1 epidemic (waba}) 1 epilepsy (Òar¨) 1 faruq [theriac] 1, n63 fennel (raziyanag) 1 fever (Ìumma) 1 fever spells (nawa}ib al-Ìumma) 1 flatulence, gastric (nafÌa min al-ma¨ida) 1 flesh [of snakes] (laÌm / luÌum) 2 flour (darmak) 2 foetuses, dead (aginna mauta) 1 frankincense (kundur) n74 French lavender (us†uÌudus) 1 galbanum (barzad) 1 gentian (gan†iyan) 1 germander (ga¨da) n74
92 Plain numbers refer to text one and text two respectively, numbers preceded by ‘n’ refer to footnotes.
493
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
ginger (zangabil) 1 grape ivy (Ìamama) 1 grape juice, thickened (¨aqid al-¨inab) 1 Greek spikenard (maibaÌusa) 1 gripes, intestinal (mags fi l-a¨∂a} ) 1 ground pine (kamafi†us) 1 growth, time of (tara¨ru¨) n64 gum-arabic (Òamg) 1 gum-senegal (aqaqiya) 1 hardship (afat) 1 head physician (ra}is al-a†ibba}) 1 headache, long-standing (Òuda¨ qadim) 1 heads [of snakes] (ru}us) 2 health, preservation of (Ìif aÒ-ÒiÌÌa) 1 health, time of (waqt ÒiÌÌa) 1 heat, innate (Ìarr garizi) 1 hellebore, white (Ìarbaq abya∂ ) 1 hemicrania (saqiqa) 1 hemlock (saukaran) 1 henbane (bang) 1 honey (¨asal) 1 honey-water (ma} al-¨asal ) 1 horehound, white (farasiyun) 1 hypericum (hufariqun) 1 Indian spikenard (sunbul hindi) n74 ingredients (adwiya) 1 intake (surb) 1 intestines [of snakes] (am¨a}) 2 iris, sky-coloured (sausan asmanguni ) 1 Jew’s pitch (Ìumar) 1 ‘land-snakes’ (Ìaiyat al-barr) 1 laurel (sa∂ag) 1 leaven (Ìamir) 2 lining [of snake guts] (†arb) 2 liquorice (sus) 1 liver (kabid) 1 long pepper (dar fulful ) 1 madder (fuwa) 1 malabathrum (saliÌa sauda} ) n74 meat [of snakes] (laÌm / luÌum) 2 method(s) (wagh / wuguh) 1 milk parsley (karafs gabali) 1 mixture (Ìil†) 2 moon carrot (sasaliyus) 1 mortar (hawun) 2 mortar (mihras) 2 mountain mint (futang gabali) 1 myrrh (murr) 1 obstruction (sudad) 1 oil (duhn) 1
old age (saiÌuÌa) n64 olive oil (zait) 2 opium (afyun) 1 opopanax (gawsir) 1 outset (ibtida}) 1 pain, colical (waga¨ al-qaulang) 1 palpitations (Ìafaqan) 1 paroxysms, cold (adwar al-bard ) 1 pastilles (aqraÒ) 1, 2, n67 people (nas) 1 people (qaum) 1 pepper, black (fulful aswad ) 1 pepper, white (fulful abya∂ ) 1 person (insan) 1 physicians (a†ibba}) 2 physicians, modern (a†ibba} qaribun al-¨ahd) 2 physicians, modern (man qaruba ¨ahduhu min al-a†ibba}) 1 plants, yielding (nabat al-Ìisa}) 1 pot (qidr) 2 pot, new earthen (qidr faÌÌar gadid ) 2 pot made of coated copper (qidr nuÌas muraÒÒa¨a) 2 prime (sabab) n64 problem (haul ) 1 property, antidotal (tiryaqiya) n64 property, special (ÌaÒÒiya) 1 rank (daraga) 1, n64 rape, wild (salgam barri) 1 remedy (dawa} ) 1 residues, vicious (fu∂ul radiya) 1 rhubarb (riwand) 1 ribs (a∂la¨ ) 1 rob (rubb) 1 room (bait) 2 roses (ward ) 1 rue (sa∂ab) 1 saffron (za¨faran) 1 sagapenum (sakabinag) 1 salsify (liÌyat at-tais) 1 salt (milÌ) 2 scammony (saqamuniya) 1 scorpions (¨aqarib) 1 sealing bole (†in maÌtum) 1 shepherd’s purse (†alasfis) 1 shivering (nafi∂) 1 skin [of snakes] (gulud ) 2 snake that putrefies the blood (mu¨affina lid-dam) 1 494
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
snakes (Ìaiyat) 1 spirit, vital (ruÌ) 1 spleen (†iÌal) 1 spring water (ma} al-¨uyun) 2 squills (¨unÒul) 1 stingers (†aqqabun) 1 storax (lubna) 1 stuff that rises to the surface (darag) 2 sweet flag (wagg) 1 tails [of snakes] (a∂nab) 2 tarantula (rutaila} ) 1 testing (imtiÌan) 1 theriac (tiryaq) 1, 2 thirst (¨a†as) 1 threats (haul) 1 toxin (sumum) 1 treatment (mudawah) 1 troubles (asbab) 1 tumours (auram) 1 turpentine (Òamg al-bu†m) n74 unguent (marham) 2 urine (baul) 1
valerian (fu) n74 venom (samm) 1 vertigo (dawar) 1 vessel made of glass (ina} zugag) 2 vessel made of lead and Indian tin (ina} raÒaÒ qal¨i) 1 vessel made of silver (ina} fi∂∂a) 1 viper [pastilles] 2, n74 viper flesh (luÌum al-afa¨i) 2 vipers (afa¨i) 1, 2 visnaga (nanÌawah) 1 vitriol, yellow (qalqa†ar) 1 vomiting (qai}) 1 wall germander (kamadariyus) 1 warrants (∂amanat) 1 water (ma}) 1 water, corruption of (fasad al-ma} ) 1 ‘water-snakes’ (Ìaiyat al-ma} ) 1 wine (sarab) 1 wombs (arÌam) 1 wood germander (usqurdiyun) 1
Botanical Names Acacia arabica → gum-arabic Acacia senegal → gum-senegal Acorus calamus → sweet flag Ajuga chamaepitys → ground pine Ammi visnaga → visnaga Anethum graveolens → dill Apium graveolens → celery Aristolochia clematitis → birthwort Aristolochia longa → birthwort, long Aristolochia rotunda → birthwort, rolled Boswellia spp. → frankincense Brassica napus sspp. → rape, wild Capsella bursa-pastoris → shepherd’s purse Centaurium spp. → centaury Cinnamomum tamala → malabathrum Cinnamomum zeylanicum → cinnamom Clinopodium vulgare → mountain mint Commiphora myrrha → myrrh Commiphora opobalsamum → balm Conium maculatum → hemlock Convolvulus scammonia → scammony Crocus sativus → saffron Cymbopogon nardus → citronella Daucus carota → carrot
Ferula galbaniflua → galbanum Ferula persica → sagapenum Foeniculum vulgare → fennel Gentiana lutea → gentian Glycyrrhiza glabra → liquorice Hyoscyamus niger → henbane Hypericum spp. → hypericum Iris persica → iris, sky-coloured Lagoecia cuminoides → caraway, wild Laurus malabathrum → laurel Lavandula stoechas → French lavender Marrubium vulgare → horehound, white Olea europaea sspp. → olive [oil] Opopanax chironium → opopanax Origanum dictamnus → Cretan dittany Papaver somniferum → opium Peucedanum palustre → milk parsley Pimpinella anisum → anise Piper longum → long pepper Piper nigrum → pepper, black/white Pistacia terebinthus → turpentine Polyporus officinalis → agaric Potentilla reptans → cinquefoil Quercus spp. → acorn 495
TWO ANTIDOTES FROM THE ‘EMPIRICALS’ OF IBN AT-TILMIΔ
Teucrium scorodonia → wood germander Tragopogon porrifolius → salsify Urginea maritima → squills Valeriana celtica → Greek spikenard Valeriana jatamansi → Indian spikenard Valeriana officinalis → valerian Veratrum album → hellebore, white Vitis vinifera → grape [juice] Zingiber officinale → ginger
Rheum palmatum → rhubarb Rhoicissus rhomboidea → grape ivy Rosa spp. → roses Rubia tinctorum → madder Ruta graveolens → rue Seseli tortuosum → moon carrot Styrax officinalis → storax Tanacetum balsamita → alecost, bitter Teucrium spp. → germander Teucrium chamaedrys → wall germander
Address for Correspondence: 120 Hunter House Road, Sheffield, S11 8TY
496
Journal of Semitic Studies LV/2 Autumn 2010 doi: 10.1093/jss/fgq010 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
A NEW NABATAEAN FUNERARY INSCRIPTION FROM HUMAYMA GEORGE A. BEVAN AND M. BARBARA REEVES QUEEN’S UNIVERSITY
Abstract This inscription was found in the summer of 2007 at the site of Humayma (ancient Hawara) in southern Jordan and has three lines of Nabataean in a recessed frame (tabula ansata). The text seems not to be associated with a grave site, but with a funerary stele (nefesh) erected by two sons for their deceased father. Though the stele is itself lost, close parallels with Nabataean stelae depicted in relief in Petra suggest what the complete monument may have looked like. Based on its letter forms, the monument dates to the late second or third century and reflects a contemporary re-emphasis on Nabataean culture within the Roman Province of Arabia.
A broken sandstone block with three lines of Nabataean text was found in the summer of 2007 at the site of Humayma (ancient Hawara) by a local Bedouin, Swaylem al Manaja, and photographed and documented by Robbyn Gordon of the University of Victoria while on an ethnographic project in the area (Figures 1 & 2).1 The block is 28.5 cm high and 17.0–26.5 cm thick, with a maximum preserved length of 33.5 cm. Its left side is missing. All extant sides have a pecked appearance but the side bearing the inscription has been carefully smoothed. The inscription is carved within a recessed tabula ansata frame, carefully centred from the top and bottom of the block, and thus probably also from the two sides of the block. The preserved handle of the tabula ansata is 4.0 cm long and flares out to a maximum width of 3.75 cm. The rectangular plaque contain1
Gordon alerted Humayma Excavation Project Directors J.P. Oleson and M.B. Reeves of the presence of the inscription and, with the help of Sausan Fahkri and Manal Basyouni of the Department of Antiquities, had the block transferred to the Aqaba Museum. Brian Seymour drew the block in the Aqaba Museum in June 2008. Ian Babbitt prepared the photograph of the block for publication. The authors thank Professors John Healey and Amir Harrak for their invaluable advice during the preparation of this article. 497
A NEW NABATAEAN FUNERARY INSCRIPTION FROM HUMAYMA
Figure 1. (Photo: R. Gordon)
Figure 2. (Drawing: B. Seymour) 498
A NEW NABATAEAN FUNERARY INSCRIPTION FROM HUMAYMA
ing the inscription is 13.5 cm high, with a preserved width of 17.5. The letters vary in height from 2.25 to 3.30 cm. Two similar recessed frames recording similar funerary monuments were found at Umm el-Jimal (Littmann 1914: no. 40 = CIS II 191, and no. 105). The inscription is a dedication of a funerary monument, a nefesh, by two sons, ‘BDW and SBDY, to their deceased father. In its most basic meaning napsa (Hebrew )נפשׁdenoted the ‘soul’ or ‘self’, but came in Semitic religions to refer to a particular type of tall, pyramidal or conical funerary monument in which the soul of the deceased was said to reside after death.2 Although the dedications on such monuments are highly formulaic (Lidzbarski 1898: 137–48), the present text has important features that are unattested elsewhere in the Nabataean corpus. In other cases of such inscriptions, the verbs bnh and ¨bd are used to indicate that the monument was ‘made’ or ‘built’ for the deceased along with the tomb. In the present text it seems the vox propria of setting up statues (e.g., CIS II 164), the causative hafel form of qwm, was used to denote the monument’s erection. The choice of this verb, if the proposed supplement is correct, strongly suggests that only a monument was erected, divorced from the funerary context, and that the tomb itself was elsewhere (cf. Dalman 1908: 78). This interpretation finds some support in a pair of similar monuments depicted in relief in a tomb in Petra (CIS II 352 and 353), which, according to the most recent discussion (Nehmé 2003: 240), indicated the spiritual presence of deceased family members who were not themselves buried in the same tomb. Another inscription from Petra makes this practice explicit when it records that one Petraios, who died and was buried at Jerash, had a nefesh dedicated to him at Petra (Starky 1965a and 1965b: 44). The Madaba Inscription also indicates that two napsata were set above a tomb (CIS II 196 = Healey 1993: 247–8), although their precise function is unknown. Although the inscription indicates that the block had once formed part of a Nabataean funerary monument erected in the late second or third century CE, the block became displaced from this monument by the fourth century. Instead, the discovery of the block in a heap of stones pushed 25 m to the southwest of the Roman fort, probably during the Department of Antiquities’ clearance of the fort’s walls in 2005, strongly suggests that the Nabataean inscribed block had been incorporated into the fort’s fourth century renovations. As to the original location of the inscribed block, although it is possible that 2 See Healey 2001: 170–1, Patrich 1990: 122–3, Starcky 1966: cols. 951–6, Gawlikowski 1970: 27–9 and 1972: 15. For images, see Starcky 1965a and 1965b.
499
A NEW NABATAEAN FUNERARY INSCRIPTION FROM HUMAYMA
the inscribed block was carried down for the fort’s renovations from the Nabataean and Roman period necropoleis in the hills to the far west side of the settlement (Oleson et al. 1993: 147–9, 1994: 154–5, 1995: 330), Nabataean nefeshata were commonly set up in non-funerary contexts. In particular, nefeshata, both free-standing and in relief, were commonly set up in high traffic areas, such as the twelve at the entrance to Petra’s Siq (Starky 1965a, 1965b) or the one on the staircase leading to Petra’s Great Temple (Joukowsky 2007: 390–1). If the nefesh at Hawara had been set up in a high traffic area, a likely candidate near the stone’s findspot would have been at the side of the major north-south highway through provincia Arabia, the Via Nova Traiana, which archaeological evidence suggests ran alongside the west side of Hawara’s Roman fort (Oleson et al. 2003: 49 and 54). An inscribed funerary monument on the side of this major provincial thoroughfare would have signalled the importance of the deceased man and his family to the numerous soldiers, traders, and travellers passing along this road. The best evidence for the appearance of this nefesh comes from the stone into which the inscription was carved, a horizontally configured ashlar block of a type to be incorporated into a built structure. Thus the stone was intended for a different purpose than the tall, thin inscribed slabs best known from a cluster found in situ in the dromos of a tomb at Umm el-Jimal (Littman 1914: nos. 60–7).3 Moreover, although the monument conceivably resembled the four-to-five course stepped pyramid built over a tomb in the Nabataean necropolis of Mampsis (Negev 1971: 114), the use of the verb qwm suggests that the memorial was not a building but rather a simple, free-standing obelisk. Two types of monuments are known that the Nabataeans explicitly labelled as nefesh.4 The first type, represented by entry number 60 in Littman’s (1914) catalogue, can be quickly rejected as it corresponds to the thin upright slabs mentioned above.5 The second type, in its 3
There are many other examples. At Humayma, three fragmentary slabs 2.4– 4.0 cm thick with traces of Nabataean letters recovered from the fill of a tomb (Oleson et al. 1994: 154–5; Oleson et al. 1995: 330) may correspond to this type. 4 Excluded from this typology are anepigraphic features and artifacts that scholars have suggested might be nefesh-type memorials, but which differ in appearance from the labeled examples; e.g. pyramid and stele on Nabataean tombs at Mampsis (Negev 1971: 114–17), stelae at Khirbet edh-Dharih (Lenoble et al. 2001: 106–7), the obelisk next to tomb Br 813 at Petra (Zayadine 1986: 229 and 232 Abb. 22). 5 Besides, as Littman suggested, the nefesh mentioned in this inscription may refer to the adjacent tomb (Littman 1914: 54), or perhaps to a monument over the tomb. 500
A NEW NABATAEAN FUNERARY INSCRIPTION FROM HUMAYMA
labelled form, survives intact only in bas-reliefs. These labelled basreliefs (along with similar anepigraphic reliefs) can be subdivided into two primary sub-groups, Dalman’s ‘Der einfache Spitzpfeiler’ and ‘Der Spitzpfeiler mit Postament’ (1908: 78).6 The first subgroup, consisting of a triangular obelisk without base, seems least likely to have accommodated our block in a built example; rather the bas-reliefs show this type of nefesh monument as a monolithic structure onto which, in a single case, an inscription has been carved.7 In contrast, the other subtype of inscribed nefesh could easily have incorporated our block in a freestanding example. This subtype takes the form of an obelisk rising from an inscribed base. The base has a width equal to or slightly exceeding the maximum width of the obelisk, and between the obelisk and base is shown a slightly wider but thin horizontal slab; some examples also show a similar slab beneath the base.8 Sometimes (but not always) the obelisk is topped with a flourish. Moreover, what is most relevant about the appearance of the second subtype of monument is that the base is shown as a monolithic block onto which the inscription is carved. It is thus quite conceivable that our block formed the base for a simple freestanding memorial of this type, with a horizontal slab and obelisk placed over top of the base and plastered into place.9 The paleographic features of the inscription strongly suggest it was a product of the late second or third centuries CE, a period which coincides with a documented recrudescence of Nabataean culture after several decades of Roman rule (Bowersock 1990: 33; Patrich 1990: 73–5). At Hawara this resurgence finds dramatic confirmation in the recent excavation of a shrine located in the town centre (Oleson et al. in press; Reeves in press and in preparation). This shrine, which had been dismantled by the Romans following the annexation, was rebuilt in the late second or early third century. At the focal point of Hawara’s third century shrine was a cult statue taking the form of a Nabataean betyl. Next to the betyl, the garrison of the fort set up 6
Dalman (1908: 78) noted one further subtype, ‘Der Spitzpfeiler mit Fuß’, but as there are only two examples of this subtype and neither are inscribed, this subtype will not be discussed further. 7 Inscribed example: Milik and Starcky 1975: 115–16 and Pl. XXXIX, 1; Dalman (1908: 78) noted forty-seven anepigraphic examples. 8 Inscribed examples: Dalman1908: nos. 187, 194c, 198; anepigraphic examples: Dalman 1908: nos. 4b, 49i, 333e, 333f, 335, 342c, 401c, 401e, 755a, 755c, 755d, 755e, 755f, 800; Joukowsky 2007: Fig. 3. 9 Cf. a nefesh composed of small stones plastered together from Bahrain: Daems et al. 2001: 180–1, first century BCE – second century CE. 501
A NEW NABATAEAN FUNERARY INSCRIPTION FROM HUMAYMA
a dedication to their regimental deity but, in an apparent show of solidarity with the townsfolk, left the Nabataean betyl at the shrine’s focal point. Both the betyl, and the garrison’s reaction to it, reflect a new respect for Nabataean culture in the third century. The same values are reflected by the erection of a late second or third century nefesh bearing an inscription in Nabataean (rather than Greek or Latin). Respect for traditional Nabataean culture at Hawara was not, however, destined to last. By the early fourth century, the town shrine and its betyl had been abandoned and forgotten. At the same time the nefesh memorial was apparently dismantled and its carefully carved inscription subsumed into the fort’s late Roman defences.10 Prior to 2007, only a handful of Nabataean inscriptions were known to have come from the Humayma area, of which only two have been published: Graf 1992 and Hayajneh 2001. Several small fragmentary inscriptions remain unpublished. Three headstones were found during a 1993 survey of the tombs (Oleson et al. 1994: 154–5; Oleson et al. 1995: 330). The bottom quarter of a line preserved on a block reused in a Roman cistern was found during excavation in 1987 (Oleson, personal communication). In addition some graffiti, noted by David Graf, were carved into the rocks on the western edge of the settlement (Graf 1992: 69). A dipinto on a potsherd found during the excavations in the Roman fort might also be written in Nabataean (Oleson et al. 2002: 104). Finally, to this meagre corpus can be added the unrecorded and otherwise unknown ‘brief description in Greek and Nabataean’ seen by Musil (1926: 60) while climbing in the hills behind Humayma. Transliteration 1 dnh nps} dy hq[ymw 2 ¨bdw w sbdy lm[… 3 }bwhm bsnt .[… Translation This is the funerary monument that ‘BDW and SBDY erected for M??? their father in year… 10 A similar fate befell blocks bearing Nabataean funerary inscriptions at Umm al-Jimal, which were later reused in Christian buildings (Littman 1914: nos. 39–41 pp. 35–40).
502
A NEW NABATAEAN FUNERARY INSCRIPTION FROM HUMAYMA
Epigraphic Notes The inscription is generally clearly written with little erosion. The mim in line 3, however, seems to extend beyond the prepared surface of the stone and onto the frame. Several features in the script are consistent with a date in the second or third centuries CE, although any firm date cannot be established because of the lack of a full dating formula. First, there is the absence of the X-type aleph, a feature last seen in the Meda’in Saleh inscriptions in 63/64 CE (Healey 1993: 296–7) and in 93 CE in the Hawran (RES I no. 83; Gruendler 1993: 33). Second, a telling feature of the text’s late date are the dots above each of the daleths intended to distinguish them from the similarly formed resh (Healey 1990–1: 45 and Healey 1990: 51). These dots are found in a late Nabataean inscription from Meda’in Saleh, dated to 356 (Altheim and Stiehl 1968: 305–9, pl. 54; image reproduced in Naveh 1987: 159) and a very early Arabic inscription from Raqush in Nabataean script (Jaussen and Savignac 1909: no. 17; republished in Healey and Smith 1989 and discussed further in Healey 2002). Third, the form of the yodh in lines one and two is characteristic of later inscriptions (see the chart in Healey 1993: 297 and the examples cited by Said and al-Hamad 2003: 32, n. 8: Littmann 1914: nos. 40, 83, 98 and 101). Fourth, the squared mim of lines 2 and 3 is also a feature common from the third century on (Greundler 1993: 97). Finally, the inscription also shows the strong tendency of later Nabataean to join letters (see e.g. Littmann 1914: no. 41 = CIS II 192, ca. 250 CE). Below is the inscription with linking between the letters indicated by underlining: 1 dnh nps} dy hq[ymw] 2 ¨bdw w sbdy lm[...] 3 }bwhm bsnt .[…] Comments 1. dnh: Although nps} is feminine, a masculine demonstrative has been used to begin the text. For clear parallels for this relatively common practice, see, CIS II 159, 192 (= RES II 1097.1), 194 (= RES II 1098), 333 and Littman 1914: nos. 39 (= RES IV 2063) and 41. Cf. CIS II 191 and 332, and Littmann 1914: no. 40 for the grammatically correct d} nps}. For dnh as the typical demonstrative in Palmyrene with nps}, see Hillers and Cussini 1996: 390, s.v. nps}, for references. 503
A NEW NABATAEAN FUNERARY INSCRIPTION FROM HUMAYMA
nps}: Properly translated as ‘funerary monument’ (Cantineau 1932: 121–2), this word seems to denote here only a commemorative stele, not a grave, which is typically denominated by qbr} (Cantineau 1932: 141) or kpr} in the Hegra area (Healey 1993: 260). In bilingual Greek and Nabataean inscriptions nps} is translated variously as stßlj (CIS 162 and 192= RES II 1097) or mnjme⁄on (Littmann 1914: no. 105; see also the remarks at Healey 1993: 243). For discussion of the nps} inscriptions with further references see Healey (2001: 170–1). hq[ymw]: The causative hafel form of qwm is usually written with the aleph preformative as }qym in Nabataean (Cantineau 1932: 141–2 and Jobling 1997: 69). The present form seems to be a conscious archaism, as in CIS II 161 col. I line 1 (securely dated to 94 CE), and 349 (Petra). Littmann 1914: no 41 (= CIS II, 192), dating from the third century, contains an even more extreme hypercorrection to what was by then a very archaic language with the noun nps incorrectly given in the construct state as npsw, likely on the model of Arabic names. Cf. Starcky and Strugnell (1966: 42) who, commenting on a text from Petra, suggest the late use of the hafel is indicative of a local dialect. The verb qwm is typically used in Nabataean (and other dialects of Aramaic) for the erection of statues and strengthens the interpretation that a stele, not a grave, is meant (see introduction). The plural form is assumed in supplement from the plural pronominal suffix of }bwhm in line 3. There remains a possibility that ¨bdw of line 2 could be construed as the main verb of the inscription, but this option leaves unintelligible the two characters hq at the end of line 1, which are unattested as the beginning of any Nabataean or Arabian name. 2. ¨bdw w sbdy: The name ¨bdw is a relatively common hypocoristic (‘servant of [divine name]’) found in Nabataean (al-Kraysheh 1986: 133–4, Negev 1991: 46, no. 798 and al-Theeb 1993: 95), as well as in Palmyrene (Stark 1971: 102), pre-Islamic Arabian inscriptions (Harding 1971: 401) and later in Syriac (). The name sbdy presents considerably more problems as it does not appear to be attested, although similar names do exist in Nabataean: sby (al-Kraysheh 1986: 169 and Negev 1991: 61, no. 1093), sbty /}sbtw (al-Kraysheh 1986: 169–71; Negev 1991: 62, no. 1103 and al-Theeb 1993: 242). For reservations about the completeness of al-Kraysheh and Negev, see Macdonald 1999. lm[..]: Several triliteral or quadriliteral male names beginning with mim are possible. For possibilities see al-Kraysheh 1986: 103–17, and Negev 1991: 37–42. 504
A NEW NABATAEAN FUNERARY INSCRIPTION FROM HUMAYMA
3. }bwhm: For a recent instance of this third person plural, masculine pronominal suffix with }b see Said and al-Hammad 2003. The character at the very end of the third line cannot be read clearly. No regnal formula for a Nabataean king can fit in the remaining space, so a number, likely in cipher, must follow bsnt. (e.g. Littmann 1914: no. 101). Given the late forms in the script, a strong possibility exists that the dating is not according to regnal year at all, but by the provincial era. For examples of bsnt or snt followed by a number without a regnal formula, see CIS II 1325 (Bostra Era), CIS II 2666, RES I 128 and RES II 528 (99th year of the provincial era = 205 CE), and Altheim and Stiehl 1969: 306–7. Address for Correspondence: [email protected] and [email protected] Classics Department, Queen’s University, 49 Bader Lane, Kingston, Ontario, K7L 3N6, Canada ABBREVIATIONS Br Brünnow, R.E. and A. von Domaszewski. 1904–9. Die Provincia Arabia. (Strassburg) CIS II Corpus inscriptionum semiticarum. Pars II. Tomus 1–2. (Paris) RES Repertoire d’épigraphie sémitique. 1900–68. (Paris) REFERENCES Altheim F. and R. Stiehl. 1969. Die Araber in der Alten Welt. Vol. V/1 (Berlin) Bowersock, G.W. 1990. ‘The cult and representation of Dusares in Roman Arabia’, in F. Zayadine (ed.), Petra and the Caravan Cities (Amman). 31–5 Cantineau, J. 1930. Le Nabatéen. Vol. I. (Paris) —— 1932. Le Nabatéen, Vol. II. (Paris) Daems, A., E. Haerinck, and K. Rutten. 2001. ‘A burial mound at Shakhoura (Bahrain)’, Arabian Archaeology and Epigraphy 12, 173–82 Dalman, G. 1908. Petra und seine Felsheiligtümer. (Leipzig) Gawlikowski, M. 1970. Monuments Funéraires de Palmyre. (Warsaw) —— 1972. ‘La notion de tombeau en Syrie romaine’, Berytus 21, 5–15 Graf, D.F. 1992. ‘The “God” of Îumayma’, in Z.J. Kapera (ed.), Intertestamental Essays in Honour of Józef Tadeusz Milik (Krakow). 67–76 Gruendler, B. 1993. The Development of the Arabic Scripts: From the Nabataean Era to the First Islamic Century According to Dated Texts. (Harvard Semitic Studies 43. Atlanta) Harding, G.L. 1971. An Index and Concordance of Pre-Islamic Arabian Names and Inscriptions. (Near and Middle East Series 8. Toronto) Hayajneh, H., 2001. ‘Marcus Ulpius Su¨aidu in einem Bruckstück einer nabatäischen Inschrift aus Süd-Jordanien’, ZDPV 117, 117–85 Healey, J.F. 1990. The Early Alphabet. (London) —— 1990–1. ‘Nabataean to Arabic’, Manuscripts of the Middle East 5, 42–52 —— 1993. The Nabataean Tomb Inscriptions from Mada’in Salih. (Oxford) 505
A NEW NABATAEAN FUNERARY INSCRIPTION FROM HUMAYMA
—— 2001. The Religion of the Nabataeans: A Conspectus. (Leiden) —— 2002. ‘New Nabataeo-Arabic: Jaussen-Savignac nab. 17 and 18’, in J.F. Healey and V. Porter (eds), Studies on Arabia in Honour of G. Rex Smith, JSS Supp. 14 (Manchester). 81–9 Healey J.F. and G.R. Smith, 1989. ‘Jaussen-Savignac 17 — The Earliest Dated Arabic Document (A.D. 267)’, Atlal (The Journal of Saudi Arabian Archaeology) 12, 77–84, plate 46 Hillers, D.R. and E. Cussini 1996. Palmyrene Aramaic Texts. (Baltimore) Jaussen, A. and M.R. Savignac. 1909. Mission archéologique en Arabie. Vol. I, (Paris) Jobling, W.J. 1997. Nablex: Nabataean-Aramaic. A Provisional Lexicon. (Kensington Md.) Joukowsky, M.S. 2007. ‘Surprises at the Great Temple from 1993 to 2006’, in T.E. Levi et al. (eds), Crossing Jordan: North American Contributions to the Archaeology of Jordan, (London). 385–92 al-Kraysheh, F. 1986. Die Personennamen in den nabatäischen Inschriften des Corpus Inscriptionum Semiticarum. (Marburg) Lenoble, P., Z. Al-Muheisen, and F. Villenueve. 2001. ‘Fouilles de Khirbet EdhDharih (Jordanie), I: Le cimetière au sud du Wadi Sharheh’, Syria (78) 89–151 Lidzbarski, M. 1898. Handbuch der nordsemitischen Epigraphie I. (Weimar) —— 1902–15. Ephemeris für Semitische Epigraphie I–III. (Giessen) Littmann, E. 1914. Nabataean Inscriptions from the Southern Haurân. Publications of the Princeton University Archaelogical Expedition to Syria, Division IV Section A. (Leiden) Macdonald, M.C.A., 1999. ‘Personal Names in the Nabataean Realm’, JSS 44, 251–89 Milik, J.T. and J. Starcky 1975. ‘Inscriptions récemment découveretes à Pétra’, Annual of the Department of Antiquities of Jordan 20, 111–30, plates XXXVII– XLVII Musil, A. 1926. The Northern Hegaz. (New York) Naveh, J. 1987. Early History of the Alphabet. Second revised edition. (Jerusalem) Negev, A. 1971. ‘The Nabataean Necropolis of Mampsis (Kurnub)’, Israel Exploration Journal 21, 110–29 —— 1991. Personal Names in the Nabataean Realm. Qedem 32. (Jerusalem) Nehmé, L. 2003. ‘Les inscriptions des chambres funéraires nabatéennes et la question de l’anonymat des tombes’, Arabian archaeology and epigraphy 14, 203–58 Oleson, J.P. et al. 1993. ‘The Humeima excavation project, Jordan: Preliminary report of the 1991–1992 seasons’, Echos du Monde Classique/Classical Views 12, 123–58 —— 1994. ‘The Humeima Excavation Project, Jordan: Preliminary Report of the 1993 Season’, Echos du Monde Classique/Classical Views 13, 141–79 —— 1995. ‘Preliminary report of the Humeima excavation project, 1993’, Annual of the Department of Antiquities of Jordan 39, 317–54 Oleson, J.P., Reeves, M.B. and B.J. Fisher. 2002. ‘New Dedicatory Inscriptions from Humayma (Ancient Hawara), Jordan’, Zeitschrift für Papyrologie und Epigraphik 140, 102–21 Oleson, J.P. et al. 2003. ‘Preliminary Report of al-Humayma Excavation Project, 2000, 2002’, Annual of the Department of Antiquities of Jordan 47, 37–64 —— 2008. ‘Preliminary Report on Excavations at Humayma, Ancient Hawara, 2004 and 2005’, Annual of the Department of Antiquities of Jordan 52, 309– 42 506
A NEW NABATAEAN FUNERARY INSCRIPTION FROM HUMAYMA
Patrich, J. 1990. The Formation of Nabataean Art. Prohibition of Graven Images Among the Nabataeans. (Leiden) Reeves, M.B. in preparation. ‘Concordia Hawarenorum: a community shrine in a garrisoned town of the Roman Near East’ —— 2009. ‘Landscapes of Divine Power at Humayma’, Studies in the History and Archaeology of Jordan X. (Amman). 325–38 Said, S. and M. al-Hamad, 2003. ‘A New Nabataean Inscription from Umm elJimal’, JSS 48:1, 29–34 Stark, J.K., 1971. Personal Names in Palmyrene Inscriptions. (Oxford) Starky, J. 1965a. ‘Nouvelle épitaphe nabatéenne donnant le nom sémitique de Pétra’, Revue Biblique 72, 95–7, plates V–VI —— 1965b. ‘Nouvelles stèles funéraires à Pétra’, Annual of the Department of Antiquities of Jordan 10, 434–9, plates XXI–XXII —— 1966. ‘Pétra et la Nabatène’, in Supplément au Dictionnaire de la Bible 7, cols. 886–1017 Starky, J. and J. Strugnell. 1966. ‘Pétra: deux nouvelles inscriptions nabatéennes’, Revue Biblique 73, 236–47 al-Theeb, S. 1993. Aramaic and Nabataean inscription from north-west Saudi Arabia. (Riyadh) Zayadine, F. 1986. ‘Tempel, Gräber, Töpferöfen’ in M. Lindner (ed.) Petra: Neue Ausgrabungen und Entdeckungen, (Munich). 214–69
507
Journal of Semitic Studies LV/2 Autumn 2010 doi: 10.1093/jss/fgq011 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA: AL-SHARTUNI’S POSITION AS A GRAMMARIAN IN SAHM ABDULRAZZAK PATEL UNIVERSITY OF OXFORD
Abstract The Arab Renaissance (nah∂a) in the late nineteenth and early twentieth centuries witnessed a considerable amount of linguistic controversy among scholars. AÌmad Faris al-Shidyaq’s simplified grammar of the Arabic language known as Ghunyat al-†alib provoked Sa¨id al-Shartuni’s refutation, al-Sahm al-Òa}ib. This work subsequently prompted two counter-refutations in al-Shidyaq’s defence by Yusuf al-Asir and Ibrahim al-AÌdab. One of the most significant aspects of this controversy is that scholars have been searching for a solution to the issues it raised up to the present day. Based on an extensive study of original sources, this article sheds light on some of the underlying factors prompting al-Shartuni’s refutation and his position as a grammarian. The article also examines the counter-refutations offered by al-Asir and al-AÌdab in order to provide a fuller picture of the linguistic debate in the nah∂a.
During the Arab Renaissance (nah∂a) of the late nineteenth and early twentieth centuries nearly all intellectuals took an active interest in the Arabic language. As far as the question of language reform is concerned, however, one can isolate the proponents of two main discourses. The conservative reformists which included Bu†rus al-Bustani (1818–83), Ibrahim al-Yaziji (1847–1906), and Sa¨ id al-Shartuni (1849–1912), and the liberal reformists including AÌmad Faris alShidyaq (1804–87), Yusuf al-Asir (1815–89), and Ibrahim al-AÌdab (1827–91).1 These scholars agreed on the need to reform Arabic, but 1 Al-Shartuni was among the prominent group of nah∂a intellectuals who devoted themselves to the preservation of the Arabic language. His principal interests lay in grammar, lexicography and rhetoric though he made important contributions to other fields as well. His close associates include leading nah∂a intellectuals like Bu†rus al-Bustani and MuÌammad ¨Abduh (1849–1905), and later influential figures like Shakib Arslan (1869–1947) and Marun ¨Abbud (1886–1962) who were his students. Al-Asir and al-AÌdab, Lebanese Muslim scholars, were educated at al-Azhar and among leading nah∂a intellectuals who made important contributions
509
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
differed over the scope, method, and especially in their conception of linguistic reform. The conservative concept ought to be understood as a concern for the correct use of Arabic with an intent to purify, while the liberal reformist idea was more about making the language an adaptable means of communication often with an aim to simplify. The later part of the nineteenth and early twentieth centuries witnessed a considerable amount of linguistic controversy between these two groups of scholars. One of the most notorious debates was over al-Shidyaq’s simplified grammar of the Arabic language, Ghunyat al-†alib wa-munyat al-raghib (The Enrichment of the Student and the Object of Desire of the Aspirant, 1872). This work triggered al-Shartuni’s refutation, al-Sahm al-Òa}ib fi takh†i}at Ghunyat al-†alib (The Correct Arrow Concerning the Errors in Ghunyat al-†alib, 1874). In response, al-Asir and al-AÌdab came to al-Shidyaq’s defence in two works entitled Radd al-shahm li}l-sahm (The Response of the Noble to al-Sahm, 1874) and Radd al-sahm ¨an al-taÒwib wa-ib¨adahu ¨an marma alÒawab bi}l-taqrib (Preventing al-Sahm from its Aim and Repelling it from the Right Target, 1874). Based on an extensive study of original sources, this article examines al-Shartuni’s Sahm and attempts to shed light on some of the underlying factors prompting his refutation and his position as a grammarian. The counter-refutations offered by al-Asir and al-AÌdab are also examined in order to get a fuller picture of the linguistic debate in the nah∂a. The article begins by setting out al-Shidyaq’s rationale behind the Ghunya, as a prelude to al-Shartuni’s subsequent response and the counter-response of al-Asir and al-AÌdab. Some interesting features of al-Shartuni’s technique are then outlined, before examining selected linguistic disputes from Sahm, with particular reference to al-Shartuni’s criteria and the counter-arguments of al-Asir and al-AÌdab. Motives and Approaches: al-Shartuni vs. al-Shidyaq, al-Asir, and al-AÌdab One of the most interesting features of al-Shidyaq’s prologue to his Ghunya is the way in which he attempts to articulate the rationale behind the work upon what might be considered four principal bases: pedagogical, theoretical, political, and traditional. Henceforth referred to Arabic language and literature. For more on these scholars, see Abdulrazzak Patel, ‘Sa¨id al-Shartuni: a Humanist of the Arab Renaissance (nah∂ah)’, unpublished Ph.D. thesis (University of Exeter 2007). 510
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
to as (1) pedagogical (2) theoretical (3) political and (4) traditional ‘alibis’ since al-Shidyaq employs them in order to justify his simplified grammar. His employment of a pedagogical alibi is reflected in his statement, ‘I intend to simplify (tashil) the presentation of Arabic grammar [in order] to provide Arabs and non-Arabs in particular with a good command over its grammar.’2 He then prescribes a timeframe of three months, in which he claims the student will have grasped the best part of Arabic grammar by learning one lesson each day with ‘concentration’. For instance, on a theoretical level he argues that Arabic grammar causes considerable difficulties for learners of the language because of the excessive and complex nature of the rules and methods used in its traditional presentation; on a political level he longs for someone to simplify (tashil/taysir) the presentation of Arabic grammar. He was therefore delighted when the Ottoman Chief Minister for Education, Mehmed Saffet Pasha (1814–83), commissioned him to produce a systematic grammar, according to his wishes.3 Al-Shidyaq is keen to authenticate his work in a traditional framework, which he does by stressing that his presentation, as such, does not contravene the conventional presentation of grammar. He therefore lists some of the grammars and commentaries of the medieval period on which he claims to base his Ghunya, including al-Ushmuni’s commentary on Ibn Malik’s Alfiya (The 1,000 Verses) and al-Astarabadhi’s commentary on Ibn al-Îajib’s al-Kafiya fi alnaÌw (The Adequate on Grammar). He states that students should consult these works if they wish to further pursue the ‘more secondary matters’ of Arabic grammar. Al-Shidyaq’s alibis are extremely significant because they are comparable to alibis offered by other proponents of simplification. From the medieval period, Ibn Ma∂a al-Qur†ubi (d. 1194), who presented a detailed proposal for the reform of grammar in his al-Radd ¨ala al-nuÌah (Refutation of the Grammarians) is one scholar who comes to mind. Qur†ubi’s main dispute centred on the theory of governor (¨amil), which he held responsible for the intricate nature of Arabic grammatical presentation. In the post-nah∂a period, Ibrahim MuÒ†afa (1937), Shawqi Δayf (1947) and Fu}ad Tarzi (1973) are just some scholars who come to mind.4 These advocates of simplification share 2
AÌmad Faris al-Shidyaq, Ghunyat al-†alib wa-munyat al-raghib (Istanbul 1872), 3. One of the leading proponents of educational reform within the framework of the Ottoman TanÂimat (1839–78). 4 Kees Versteegh, Landmarks in Linguistic Thought III: The Arabic Linguistic Tradition (London 1997), 150–2. 3
511
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
the alibi that Arabic grammar causes considerable difficulties for learners of the language, or that Arabic grammar needs to be simplified because the linguistic shortcomings of Arabic speakers are attributable to its difficulty.5 In this sense, the significance of al-Shidyaq’s work perhaps lies in being an integral part of the drive towards simplification which notably began with Qur†ubi’s Radd, and was revitalized in the twentieth century by scholars such as Δayf, MuÒ†afa, and Tarzi through their individual calls for the revival of grammar (iÌya} al-naÌw), and for the simplification of the Arabic language (tabsi† al-lugha) and its grammar (tabsi† al-naÌw).6 Al-Shartuni refutes each individual statement in al-Shidyaq’s prologue in order to demolish his pedagogical, theoretical, and traditional alibis. For instance, he writes: ‘It takes a native speaker several years to acquire a reasonable command of Arabic grammar, but with alShidyaq’s work a student will have grasped it in a few days! This can only be possible if someone (i.e. al-Shidyaq) has divorced Arabic grammar from its established roots in the traditional grammars and infringed the sanctity therein. Thus, opening up the way for each and every alien element (dakhil) while leaving no hope for the student.’ Al-Shartuni then scrutinizes al-Shidyaq’s prescribed timeframe of three months within which he claims the student will have achieved a command of grammar by learning one lesson every day: ‘How can this be possible? There are thirty-five lessons on morphology and sixty-six on grammar, bringing the total to 101 lessons. For the student to grasp grammar in three months he would have to learn more than one lesson on some days!’ Al-Shartuni puts al-Shidyaq’s pedagogical alibi under further scrutiny by posing a series of rhetorical questions. For example, if al-Shidyaq’s aim is the simplification (tashil ) of Arabic grammar for Arabs and non-Arabs, is the timeframe of three months 5
Omran notes these scholars attribute the problem of traditional presentations of Arabic grammar to the complex, philosophical and argumentative nature of its presentation. Each states that his goal is to present Arabic grammar in a clearer, less complicated and more integrated manner. See Elsayed Omran, ‘Arabic Grammar: Problems and Reform Efforts’, in K. Dévényi and T. Iványi (eds), Proceedings of the Colloquium on Arabic Grammar (Budapest 1991), 309. 6 This is not to deny the fundamental difference which distinguishes al-Shidyaq’s work. Al-Shidyaq actually attempted to deliver a ‘simplified’ grammar while Qur†ubi and others only presented proposals for its reform. For instance, MuÒ†afa proposed changes to fundamental areas of Arabic traditional grammar e.g. i¨rab, and Tarzi to grammatical categories. For more on the simplification of grammar in the twentieth century, see Yasir Suleiman, ‘The Simplification of Arabic Grammar and the Problematic Nature of the Sources’, JSS 41:1 (1996), 99–119. 512
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
restricted to non-Arabs or does it include Arabs? If so, can an Arab learn it in less than three months? Does the Ghunya require a teacher? If so, are all teachers of the language able to teach it, or is this honour for al-Shidyaq alone? If for al-Shidyaq alone, what would happen after his death!?7 A closer look at al-Shartuni’s rationale behind Sahm clarifies his reasons for launching a refutation against al-Shidyaq. Al-Shartuni warns the unsuspecting reader of the nonsense (khalal ) and faults (kha†a} ) in Ghunya, which he claims ‘embraces all kinds of corrupted beliefs and teachings’. Al-Shartuni’s higher purpose is thus revealed as wishing to highlight the misconceptions (awham) and errors (saqa†at) in al-Shidyaq’s work lest ‘the epidemic of al-Shidyaq’s work spreads among students, I seek refuge with God from this evil!’ Al-Shartuni furthermore announces that his refutation is based on the most authentic traditions and sayings (aqwal ) of the leading authorities of the Arabic language (a}immat al-{arabiya).8 Al-Shartuni’s remarks in the above passages echo the refutational writings of the medieval period, where it was normal to invoke the erroneous nature or ambiguity of an earlier work as a motivation for issuing a corrigendum or writing another ‘better’ work. Further comparisons with these writings can be drawn in al-Shartuni’s choice of language, for instance saqa†at, kha†a} and so forth. Al-Shartuni’s statements and language here not only reflect his prescriptive tendencies, but also inform us that his work is written within the framework of preserving linguistic standards. He is keen to show that he operates within a respected convention. Al-Shartuni’s reference to the leading authorities (imams) of the Arabic language, however, is particularly significant because he announces thereby that his arguments will be based on earlier ‘authoritative’ texts and that he accepts or submits to (taqlid ) their linguistic sayings (aqwal). By aligning his thoughts to the work of the ‘elites’ of Arabic, al-Shartuni not only serves as an embodied vehicle for tradition, but also develops an authoritative alibi. Al-Shartuni exposes the flawed nature of al-Shidyaq’s work by showing that it contradicts the ‘orthodox’ sayings of the imams. AlShartuni, in contrast to al-Shidyaq, also offers a predominantly traditional (medieval) alibi of preserving the Arabic language. In this sense, if al-Shidyaq’s motive is to simplify Arabic grammar, then it is equally plausible that al-Shartuni’s motive is to expose the ‘flawed’ 7
Sa¨id al-Shartuni, al-Sahm al-Òa}ib fi takh†i}at Ghunyat al-†alib (Beirut 1874),
6–8. 8
Ibid., 10. 513
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
nature of al-Shidyaq’s work with the overall aim of preserving Arabic grammar in its traditional framework. The response of al-Asir and al-AÌdab to al-Shartuni’s criticisms reveals their motives and approach. Both portray al-Shidyaq as an imam figure just as al-Shartuni made references to the imams of the language in the medieval period. For instance, al-Asir says: ‘They aim to quell al-Shidyaq’s reputation, and quelling is the say-so of the stupid teacher, so how can one quell him who has become renowned in the lands!?’ Although sycophantic comments similar to these were commonplace in the refutational literature of the medieval period, it is interesting to note the contrasting points of reference between al-Asir and al-Shartuni. Al-Shartuni has recourse to scholars in the medieval period for his inspiration and aspiration, while al-Asir finds his inspiration in al-Shidyaq. Al-Asir’s use of ‘they’ instead of ‘he’ means that he saw al-Shartuni’s refutation as part of a wider campaign to discredit al-Shidyaq. It then follows that one of al-Asir and al-AÌdab’s main techniques in their prologue is to show that al-Shartuni’s refutation is dictated by his personal hostility to al-Shidyaq. One of their main accusations is that al-Shartuni and his supporters (i.e. Ibrahim al-Yaziji and Bu†rus al-Bustani) would stop at nothing to bring al-Shidyaq into disrepute. Therefore, al-Asir asserts, ‘We, his supporters, will not stand in silence because this would lead to much speculation, and people would accuse us of ignorance and dishonesty.’ He adds: ‘Al-Shartuni would regard himself as wise, and the ignorant might believe that al-Shartuni is a scholar and take him seriously!’9 Linguistic Disputes Al-Shartuni’s technique in his overture involved undermining the rationale behind al-Shidyaq’s Ghunya as a prelude to his main refutation. Al-Shartuni’s task in his linguistic disputes is now to prove the flawed nature of al-Shidyaq’s grammar using more meaningful arguments and techniques that bring to mind the refutational literature of the medieval and pre-modern periods. There are two broad aspects of al-Shartuni’s technique that imitate those used in earlier refutational works and commentaries. In the first, the writer presents the assertions of his adversary and then provides a one-sided refutation. For example, al-Shartuni begins his disputes with: min saqa†atihi 9
Yusuf al-Asir, Radd al-shahm li}l-sahm (Istanbul 1874), 4. 514
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
qawluhu fi ÒafÌa… (and among his [al-Shidyaq’s] mistakes is what he said on page…). He then recounts the assertion he intends to criticize, and follows with aqul… (I say…), or wa-huwa kha†a} bayyin… (and that is a clear error…) before presenting his argument. When al-Shartuni wants to bring the readers’ attention to a particular error, he uses categorical phrases such as: qultu hadhihi al-mas}ala aghrab ma warada fi hadha al-kitab (this is the strangest point I have found in this work). Al-Shartuni also employs the question and answer technique, where the critic introduces a series of conditional sentences to torment the adversary. For instance, after presenting al-Shidyaq’s assertions, he speculates over different possible meanings of his words. This technique is also found in medieval and nah∂a refutational works where the writer would torment his adversary with a string of rhetorical questions and answers. For instance, yuqalu lahu (he will be asked) fa-in qala (and if he says) qila lahu (he will be answered), which at times could continue throughout a particular argument.10 While al-Shartuni disputed seventy-six issues in al-Shidyaq’s Ghunya, here I will focus on eight because the nature of al-Shartuni’s arguments in their entirety can be mapped out under the following four headings: (Al-Shartuni’s) personal motives or linguistic scholarship; alShidyaq’s ‘grammatical errors/slips of the pen’; al-Shidyaq’s ‘incorrect’ use of grammatical terminology; and finally his ‘inadequate’ definition of grammatical phenomenon. The eight arguments examined below adequately represent these categories. I must also mention that in the true tradition of medieval works, al-Shartuni, al-Asir, and al-AÌdab cite earlier grammarians by their last names, but rarely mention the particular work let alone page numbers, which has made it difficult to trace their specific sources. However, with some persistence I have located most of these with the specific citations, which should help verify their arguments and illustrate their criteria. Personal Motives or Linguistic Scholarship The following disputes have been selected in view of allegations by al-Asir and al-AÌdab that al-Shartuni misrepresents al-Shidyaq’s words in his Sahm. They use these allegations to support their main 10 al-Shartuni, al-Sahm, 11–84 passim. A good example of these techniques in earlier works is found in Ibn al-Athir, al-Mathal al-sa}ir fi adab al-katib wa }l-sha¨ir, ed. K.M.M. ¨Uwayda (2 vols, Beirut 1998). For examples in the nah∂a, see Mikha}il ¨Abd al-Sayyid, Kitab sulwan al-shaji fi }l-radd ¨ala Ibrahim al-Yaziji (Istanbul 1872). Al-Shidyaq wrote this work under the above pseudonym to refute Ibrahim al-Yaziji. See also al-Asir, Radd al-shahm.
515
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
thesis that personal motives, i.e., hostility and revenge, encouraged al-Shartuni’s refutation. Hence, considering some of these disputes should clarify al-Shartuni’s position and provide a fuller picture of the nature of the debate between these scholars during the nah∂a. The first dispute involves al-Shidyaq’s comments regarding the dual and plural in Arabic, which al-Shartuni presents as follows: wa-ma¨a anna al-jam¨ akthar isti¨malan fi jami¨ al-lughat min al-muthanna faqad uhmila fi al-¨arabiya khilafan li}l-muthanna (Although the plural is used more than the dual in all languages, it has been neglected in Arabic in contrast to the dual).11 Al-Shartuni states: ‘These (al-Shidyaq’s) assertions here are the strangest invented by the writer.’ He adds that if al-Shidyaq had written the work solely for Arabs, it would have been easier to endure, but the fact that the work aims to teach foreigners (a¨ajim) the Arabic language makes these assertions unbearable. Al-Shartuni then alludes to his fear that foreign students would find al-Shidyaq’s comments absolutely absurd and would start wondering how and why Arabs had neglected the plural?! Al-Shartuni asserts this would be considered the biggest defect (¨uyub) of the language, when in fact it represents the incompetence of the perpetrator of these words (i.e. al-Shidyaq). For al-Shartuni, these assertions clearly reflect ‘the disease in al-Shidyaq’s mind and thoughts’. He adds: ‘It is extremely strange that al-Shidyaq should say this in reference to the plural when he himself deals with the plural in detail in his Ghunya! So how can the plural be neglected by the Arabs!? How can the writer allow students to rely on his words!?’ Al-Shartuni concludes: ‘In writing this work, al-Shidyaq has done nothing but dislodge grammar from its original foundations, and expose himself to mockery and laughter!’12 In response, al-Asir claims that al-Shartuni has misrepresented alShidyaq’s words by quoting them out of context, and adds that alShidyaq’s comments are based on the following lines: anna al-¨araba lam taj¨al ∂abi†an jami¨an mani¨an li}l-jumu¨ al-taksiriya kama li}lmuthanna ma¨a annaha fi lughatihim wa-lughat ghayrihim akthar isti¨malan (that the Arabs did not assign a generic standardized paradigm for the broken plural like they did for the dual even though it is used more in their language and in the language of others). Furthermore, al-Asir contends that al-Shidyaq is correct because he did not say: akthar isti¨malan fi sa}ir al-lughat wa-qad uhmila fi al-¨arabiya (it [broken plural] is used more in other languages and has been 11 12
Al-Shartuni, al-Sahm, 33. Ibid., 33–4. 516
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
neglected in Arabic). Rather he said:…fi jami¨ al-lughat wa-min jumlatiha al-¨arabiya (…in all languages including Arabic). Al-Asir adds that al-Shartuni ignores al-Shidyaq’s preceding sentence: wa-fi }l-jumla fa-inna al-jam¨ al-mukassar ghayr mu††arid fi }l-¨arabiya fa-la yumkin ÌaÒruhu wa-la yu¨lam illa bi}l-mumarasa (in sum, the broken plural is irregular in Arabic, so it is not possible to provide a short explanation, and can only be learned with practice). For al-Asir, alShartuni is comparable to the person who hears the following words: ma khalaqa Allahu al-khalqa illa bi}l-Ìaqqi (God did not create the universe, except with truth), and ignores the clause (except with truth). He adds that al-Shidyaq knows that Arabic is the most eloquent and noble of languages. His intention here was only to highlight that the broken plural is irregular, while the dual is regular even though it is used less.13 Like al-Asir, al-AÌdab attempts to clarify al-Shidyaq’s words as follows: ‘What al-Shidyaq meant was that the use of the plural in Arabic differs from the way it is used in other languages, given the existence of the dual.’ Al-AÌdab asks how can al-Shartuni’s accusations be true, given that al-Shidyaq gives exhaustive treatment to the plural in his Ghunya!?14 Al-Asir’s claim that al-Shartuni misrepresents al-Shidyaq’s words by ignoring the relevant sentence (i.e. wa-fi }l-jumla…) is valid because al-Shidyaq’s use of ‘neglected’ (uhmila) referred to the broken plural, not to the plural in general, as al-Shartuni implies. In addition, alShidyaq is referring to the irregular nature of the broken plural in contrast to the dual which is regular. It seems, however, that al-Asir also succumbs to a degree of misrepresentation. For instance, he claims that al-Shidyaq did not say: ‘it [the broken plural] is used more in other languages and has been neglected in Arabic’, but said: ‘…in all the languages including Arabic.’ However, al-Shidyaq does not state: ‘in all the languages including Arabic’, as al-Asir claims.15 The second issue to consider in this category regards al-Shartuni’s sarcastic remarks over al-Shidyaq’s definition of the role of the vowel (a) in the diptote noun, which al-Shartuni presents as follows: wa}lfatÌa takun {alamatan li}l-naÒb fi }l-ism alladhi la yanÒarif (the vowel (a) denotes the accusative/dependent case in the diptote noun).16 13
Al-Asir, Radd al-shahm, 32–3. Ibrahim al-AÌdab, Radd al-sahm ¨an al-taÒwib wa ib¨adahu {an marma alÒawab bi }l-taqrib (Istanbul 1874), 20. 15 Al-Shidyaq, Ghunya, 28. 16 Al-Shartuni, al-Sahm, 73. Earlier Western grammars on Arabic use the traditional terms nominative, accusative and genitive for case endings of nouns and indicative, subjunctive and jussive for verbal moods. More recent grammars use 14
517
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
In response, al-Shartuni merely remarks that this definition calls for laughter rather than a refutation! Al-Asir retorts that al-Shartuni’s remark shows his enmity and hostility towards al-Shidyaq. He explains that al-Shartuni presents alShidyaq’s definition after omitting the clause niyabatan {an al-kasra, which al-Asir claims was present in the original script but was omitted through typesetting errors (ghala† min al-Òaffaf ). He adds that alShidyaq’s full sentence should read as follows: wa}l-fatÌa takun {alamatan li}l-naÒb niyabatan ¨an al-kasra fi }l-ism alladhi la yanÒarif naÌwa marartu bi-Yusufa (the vowel ‘a’ denotes the accusative/dependent case, and substitutes for the vowel ‘i’, in the diptote noun as in ‘I passed by Yusuf’). So al-Asir asks how can an intelligent person question that al-Shidyaq was unaware of this phenomenon? He addresses al-Shartuni directly: ‘Even the shrewd nature of your enmity is better than the thought of being in your mad companionship, your argument here only illustrates your madness!’17 Although it is difficult to substantiate al-Asir’s claim that the clause niyabatan ¨an al-kasra was present in the original script, the clause is present in the version of Ghunya (1872) used here, which suggests al-Shartuni resorts to misrepresentation. Al-Shartuni, however, may have used an earlier published version of Ghunya (1871) with the clause omitted, while the Ghunya used here is the revised version with the clause added and other mistakes rectified.18 I hope to ascertain this in due course, but this would explain why al-Shartuni left out the relevant clause and explain al-Asir’s reference to a typesetting error above. If we assume for now that the clause was absent in al-Shartuni’s version, especially since al-Asir and al-AÌdab do not contest this, then al-Shartuni’s concern is aroused by the ‘inadequate’ nature of al-Shidyaq’s definition, which implies that the vowel (a) in the diptote noun only denotes the accusative/dependent case (naÒb), when in fact it can denote the genitive/oblique case (jarr) also. Al-Asir and al-AÌdab raise additional points to support their theses that al-Shartuni was guided by personal motives. Al-Asir, for instance, claims that al-Shartuni’s use of words like takh†i}at ‘to find fault’, and aÌad ‘someone’ in his refutation show his rivalry with al-Shidyaq. independent, dependent, oblique for nouns, and independent, dependent, apocopated for verbal moods. El-Said Badawi, Michael Carter and Adrian Gully, Modern Written Arabic: A Comprehensive Grammar (London 2004). In order to accommodate both usages, I list the traditional and new terms together in this paper. 17 Al-Asir, Radd al-shahm, 51. 18 Ghunya completed 1870. First ed. published (Istanbul 1871), my copy (Istanbul 1872), later edition (1888). 518
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
For al-Asir, he should have used objections (i¨tira∂at) and Zayd or fulan instead. Al-AÌdab, meanwhile, says that al-Shartuni’s main motive was to exact revenge on behalf of others.19 Was al-Shartuni’s refutation in fact offered on the behalf of others? Personal rivalry was a factor in the linguistic disputes between the conservative and reformist group of scholars. These debates, it appears, could be induced by linguistic issues and then turn into personal battles or vice-versa. Gully explains: ‘The later nineteenth and early twentieth centuries witnessed a considerable degree of controversy amongst scholars of the language. Although linguistic issues were essentially the source of these disputes, the evidence leaves us in no doubt about the ad hominem nature of many of the exchanges that took place in journals and other publications at that time.’20 The rivalry between these two groups of scholars began when al-Shidyaq expressed annoyance in his newspaper al-Jawa}ib (1871) at not being mentioned by name in a poem by NaÒif al-Yaziji. Al-Shidyaq reacted by highlighting NaÒif’s linguistic shortcomings in one of his publications, as Gully notes: ‘He pointed out NaÒif’s incorrect use of the words fĭal (for fi†Ìal), and marabid (for marabi†), in his collection of maqamat entitled Majma¨ al-BaÌrayn.’21 Subsequently, a long and bitter dispute developed between al-Shidyaq and NaÒif’s son Ibrahim in al-Jinan magazine, and often involved their close acquaintances, for instance al-Asir and al-AÌdab in al-Shidyaq’s corner and al-Shartuni and al-Bustani on al-Yaziji’s side. These scholars would often be dragged into the debate between al-Shidyaq and al-Yaziji on charges of jealousy and envy.22 Qasim argues that the disputes, for instance, surrounding fĭal (for fi†Ìal), were responsible for the narrow-mindedness of these scholars. This narrow-mindedness, he says, subsequently filtered into the controversy over grammatical issues. He claims that al-Shartuni’s Sahm was offered on behalf of Ibrahim al-Yaziji while al-Asir and al-AÌdab’s counter-refutations were motivated by similar impulses.23 Shalaq echoes a similar view when he attributes the controversy between these scholars to personal allegiances and rivalry. For instance, he 19
Al-Asir, Radd al-shahm, 4 and al-AÌdab, Radd al-sahm, 7. Adrian Gully, ‘Arabic Linguistic Issues and Controversies of the Late Nineteenth and Early Twentieth Centuries’, JSS 42:1 (1997), 109. 21 Ibid., 111. 22 A. Shalaq, al-Nathr al-¨Arabi: fi namadhijihi wa-ta†awwurihi li ¨asray al-nah∂a (Beirut 1974), 116–17. 23 Riya∂ Qasim, Ittijahat al-baÌth al-lughawi al-Ìadith fi }l-¨alam al-¨arabi (2 vols, Beirut 1982), 1: 263. 20
519
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
states: ‘The criticisms of al-Yaziji and al-Bustani and the defence of al-Asir and others on al-Shidyaq’s behalf was no more than an example of cheap rivalry… it does not even deserve to be mentioned.’24 It would be reductive, however, to put the disputes between these scholars, and the fervour al-Shartuni and others felt for the Arabic language down to personal motives alone. Contrary to these views, here it is argued that their disputes were induced more by linguistic matters and out of their fervour for the language than by personal motives. For example, in Aghla† al-muwalladin, Ibrahim al-Yaziji not only records linguistic errors rife among Arabs since the post-classical period, but also highlights errors in his father NaÒif’s writings, and even his own.25 It is therefore unlikely that he would highlight his father’s errors and his own, if it were not for his fervour for the language. More importantly, the personal animosity between these scholars, for whatever reason, essentially served to remove any reservations they might have had over criticising one another’s incorrect use of language. Al-Shartuni’s Sahm was motivated by factors beyond jealousy. Grammatical Errors/Slips of the Pen Al-Shartuni raises two points. The first concerns al-Shidyaq’s omission of the declensional nun from the verb yatakhalafani (3m. dual) in his description of the substitute (badal). The second involves his use of aw yakuna bi}l-¨aks. Al-Shidyaq states: wa-qad tubadal alma¨rifa min al-ma¨rifa ka-ma fi ¨ihdina al-Òira†a al-mustaqima Òira†a alladhina an¨amta ¨alayhim, wa }l-nakira min al-nakira: ¨inna li}lmutaqqina mafazan Ìada}iqa}.26Wa-qad yatakhalafa imma bi-an yakun al-badal ma¨rifatan wa}l-mubdal minhu nakiratan naÌwa: ¨ila Òira†in mustaqimin Òira†i Allahi}, aw yakuna bi}l-aks naÌwa: ¨la-nasfa{an bi}l-naÒiyati naÒiyatin kadhibatin} (…that they [scholars] disagree whether the substitute becomes definite and the antecedent indefinite e.g.: ‘the straight path, the path of God’, or vice-versa e.g.: ‘we will seize him by the forelock, a lying forelock’).27 24 25
Shalaq, al-Nathr al-¨Arabi, 114–16. Jurji Zaydan. Tarajim mashahir al-sharq fi }l-qarn al-tasi¨ ¨ashar (Beirut 1970),
50. 26 Al-Shidyaq’s definition of badal up to this point is copied directly from al-∑abban’s gloss, which will ensure that al-Shartuni will not refute it. The rest is al-Shidyaq’s own, which al-Shartuni will refute. See MuÌammad ibn ¨Ali al-∑abban, Îashiyat al-∑abban ¨ala sharÌ al-Ushmuni ¨ala ¨Alfiyat Ibn Malik (4 vols, Cairo 1983), 3: 127–8. 27 Al-Shidyaq, Ghunya, 84.
520
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
Al-Shartuni responds that al-Shidyaq has unnecessarily omitted the nun of declension from the verb yatakhalafani, and in so doing, has taken recourse to anomalous material. Al-Shartuni comments that he finds al-Shidyaq’s omission most strange when al-Shidyaq himself had criticized another scholar, Athnaniyus al-Tunutanji, for omitting the declensional nun from la tarawnani in the sentence: innakum min al-an la tarawni ([2m. pl.] will not see me from now on). He argued that this was the la of negation (nafy) and not the la of prohibition (nahy), and therefore al-Tunutanji should have said la tarawnani with two nuns, because the verb after the la of negation is in the indicative/ independent mood. According to al-Shartuni, al-Shidyaq argued this point against al-Tunutanji even though it has been decided among the majority of grammarians (muqarrar ¨inda jumhur al-nuÌah) that the omission of the nun of declension from, for instance, la tarawnani is permissible (sa}igh). He adds that there are clear texts (nuÒus) which attest to this just like there are clear texts which contradict al-Shidyaq’s use of yatakhalafa. For example, he quotes al-∑abban28 as saying: ‘The omission of the nun of declension (nun al-raf ¨ ) is highly permissible (jawazan bi-kathratin) when the verb is connected with the nun of protection (nun al-wiqaya: ni), as in, for instance, ta}muruni, provided that the omitted element is the nun of declension and not the nun of protection (i.e. ni ).’29 Al-Shartuni furthermore contests that al-Shidyaq’s words: aw yakuna bi }l-¨aks (vice-versa), imply that the substitute (badal ) will become indefinite (nakira) and the antecedent (mubdal minhu) definite (mu¨arraf ). Al-Shartuni says: ‘This description is fundamentally deficient because it does not bind the indefinite substitute with the described (mawÒuf ) which is necessary so that the unintended (i.e. definite antecedent) may not be more than the intended (indefinite substitute).’ He quotes Ibn al-Îajib:30 ‘When the substitute is more indefinite than the definite an adjective (na¨t) is required.’31 Al-Asir contends that the error (isqa†) here is out of carelessness (sahw) or a typesetting error (min al-Òaffaf ), and adds: ‘This is permissible 28 MuÌammad al-∑abban (d. 1791), scholar of language and rhetoric. His gloss on al-Ushmuni’s commentary on Alfiya is among al-Shartuni’s main sources. ¨Umar KaÌÌala, Mu¨jam al-mu}allifin (15 vols Damascus 1957),11: 17–18. 29 Al-Shartuni, al-Sahm, 79–80. 30 Ibn al-Îajib (1175–1249), thirteenth-century scholar and grammarian. His works on grammar include al-Kafiya which al-Shartuni, al-Asir, and al-AÌdab cite often. H. Fleisch, ‘Ibn al-Îadjib’, in EI2. 31 Al-Shartuni, Sahm, 80–1. For quote, see Ibn al-Îajib, Kitab al-Kafiya fi }l-naÌw (2 vols, Beirut 197-),1: 268–9.
521
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
(ja}izan) in the speech (kalam) of Arabs!’ Al-Asir contends that the Ghunya was not printed from al-Shidyaq’s original manuscript, but from a manuscript written by al-Shaykh al-Maki, who worked for al-Shidyaq’s journal al-Jawa}ib. Al-Asir claims that the original was proofread for printing errors (ghala† al-†ab¨) and a table of errors and corrections (jadwal) was inserted, but was destroyed with al-Shidyaq’s other books when his house burned down. Al-Asir, therefore, contends that this eliminates the errors in the original and accounts for instances like: al-fatÌah takun ¨alamatan li}l-naÒb, in the new copy. He concludes: ‘Al-Shartuni will not find these in the original, and if he does, they are just out of carelessness.’32 Al-Shartuni’s refutation is aroused by al-Shidyaq’s omission of the nun of declension from yatakhalafani. His point is valid and needs no further elaboration since the imperfect verb in the indicative/ independent mood retains its declensional nun suffixes of the dual. Al-Shartuni’s defence on behalf of al-Tunutanji alerts us to an important aspect of his technique. For al-Shidyaq, al-Tunutanji should have said ‘la tarawnani’ because the la of negation (nafy) retains its nun to give us la tarawna and with ni: la tarawnani. Although al-Shidyaq has a seemingly legitimate point, al-Shartuni is able to overturn him by invoking the level of consensus among the majority of grammarians (muqarrar ¨inda jumhur al-nuÌah). Al-Shartuni then presents clear texts (nuÒus ÒariÌa) from al-∑abban’s work since the sayings (aqwal) therein are to convince us that the omission of the nun of declension is highly permissible when the verb is connected with the nun of protection e.g. in ta{muruni. Similarly, when al-Shidyaq says: aw yakuna bi }l-¨aks, he suggests that the substitute will become indefinite and the antecedent definite. Al-Shartuni therefore presents sayings (aqwal) from earlier works to make his point that, like these scholars, al-Shidyaq should have mentioned that it is obligatory to attach an adjective to the indefinite substitute. When al-Asir excuses al-Shidyaq’s omission of the nun and other errors (i.e. al-fatÌa takun ¨alamatan li}l-naÒb with the omission of niyabatan ¨an al-kasra), he suggests that many of these errors occurred in the process of producing a new script when the original was destroyed in a fire. Of course, it is difficult to verify al-Asir’s claim regarding the correctness of the original. But based on al-Asir’s reference to an original script, a ‘new script’ with errors, and the version of the Ghunya used here, we establish at least three versions: al-Shidyaq’s so-called ‘correct’ original script destroyed in the fire, according to 32
Al-Asir, Radd al-Shahm, 53–4. 522
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
al-Asir; the second version (1871) or ‘new script’ with the clause niyabatan ¨an al-kasra omitted, which is almost certainly the version al-Shartuni used; and finally, a third revised version (1872) with the clause inserted, which is the version used here. This reinforces my earlier hypothesis that al-Shartuni was using the version in which this clause was omitted. Al-Shartuni therefore did not resort to misrepresentation to lend weight to his argument. The second dispute examined in this category involves al-Shidyaq’s description of the Ìal al-mutaradifa (the synonymous circumstantial qualifier) which reads as follows: wa}l-Ìal al-mutaradifa hiya allati takun Ìalan min al-∂amir fi mithal ‘ja}a Zaydun rakiban katiban} fa-inna ¨katiban} Ìalun min al-∂amir fi ¨rakiban} (and the Ìal almutaradifa is the Ìal of the pronoun, as in, for instance, ja}a Zaydun rakiban katiban, katiban is the Ìal of the pronoun in rakiban).33 Al-Shartuni responds that al-Shidyaq has cited Abu }l-Baqa}34 but misrepresented his original definition, which should read as follows: wa}l-mutadakhila wa-hiya allati takun Ìalan min al-∂amir fi mithal: ja}ani Zaydun rakiban katiban. Fa-inna ¨katiban} Ìalun min al-∂amir fi ¨rakiban}. He explains that al-Shidyaq uses al-Baqa’s description of Ìal al-mutadakhila (interposing circumstantial qualifier) to define the Ìal al-mutaradifa (the synonymous circumstantial qualifier). AlShartuni adds: {al-mutaradifa is the Ìal that comes from one subject of state, as in, for instance, ja}a Zaydun rakiban basiman (Zayd came riding and smiling). Besides, al-Shartuni points out that there are two declensional errors (kha†a} i¨rabi) in al-Shidyaq’s description of Ìal which are ‘even more bizarre!’ For instance, al-Shidyaq says: wa}l-mutadakhila ¨ibara ¨an ¨yakun rashidan} Ìal min ∂amir adhhab wa-muhadiyan Ìal. Al-Shartuni refutes that the word Ìal should be in the accusative/dependent case, i.e. yakun rashidan Ìalan… wamuhadiyan Ìalan.35 Al-Asir refutes that an intelligent person (¨aqil) would know that quoting from Abu }l-Baqa} means highlighting only a typesetting error. He adds that if an intelligent reader come across the phrase ya li}l-rijal in a book, they would understand and dismiss it as a typesetting error. 33
Al-Shidyaq. Ghunya, 56. Abu }l-Baqa} Ayyub ibn Musa al-Kaffawi, seventeenth-century linguist and lexicographer and author of the voluminous Arabic language and grammatical dictionary known as al-Kulliyat. 35 Al-Shartuni, Sahm, 64–5. For al-Kaffawi’s description of Ìal, see al-Kulliyat: mu¨jam fi}l-muÒtalaÌat wa}l-furuq al-lughawiya, ed. Adnan Darwish, (5 vols, Damascus 1974–6), 2: 211–12. ‘Declensional errors’ in Ghunya, 56. 34
523
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
Despite this, al-Asir asserts that al-Shidyaq has made no grammatical error (kha†a}) here, rather the mistake is in nomenclature (al-tasmiya).36 Like al-Asir, al-AÌdab concedes that al-Shidyaq’s use of al-mutaradifa, instead of mutadakhila, was carelessness (sahw al-qalam). He furthermore contends: ‘There are many errors in esteemed books which have not been rectified in the table of errors and corrections. This is nothing but stubbornness on al-Shartuni’s part!’37 Al-Shartuni, al-Asir, and al-AÌdab make recurring references to kha†a}, saqa† (saqa†at), ghala†, sahw, Òafaf, laÌn and so forth in the disputes above. These terms deserve closer attention, since they evoke the laÌn literature of the medieval period in which writers used these terms interchangeably.38 While al-Shartuni does not employ the conventional term laÌn ‘grammatical error’ above, his use of kha†a}, saqa†, and ghala† substitutes the former and reflects in al-Shartuni the same prescriptive tendencies of imposing certain linguistic norms to prevent linguistic errors. In this context Gully notes: ‘If the period of ‘Awakening’ began with more than an incidental evocation of the past, then so did the techniques adopted by some of the intellectuals of the time in scrutinizing the legitimacy of linguistic usage. Their attempt to correct, reform and purify the language immediately bring to mind the laÌn literature of the medieval period, in which scholars sought to repudiate unorthodox or patently incorrect use of language in speech and in writing.’39 Two early laÌn works serve to put al-Shartuni’s disputes in a historical context. One of the earliest is {Umar b. Shabba’s ninth century work known as Kitab al-NaÌw wa-man kana yalÌan min al-naÌwiyin (The Book on Grammar and those Grammarians who Committed Errors). In this work, Ibn Shabba highlights errors of individual grammarians whom he names as setting a bad example for others. The other work is al-Îariri’s Durrat al-ghawwas fi awham al-khawaÒÒ (The Pearl of the Diver Concerning the Delusions of the Elite), in which al-Îariri points out errors of the intellectual elites (khaÒÒa).40 Two major works come under laÌn literature in the nah∂a and give meaning to al-Shartuni’s disputes in a contemporary context. Ibrahim 36
Al-Asir, Radd al-Shahm, 47. Al-AÌdab, Radd al-Sahm, 36. 38 Pellat points out that awham, kha†a}, saqa†, and ghalat are often substituted, in earlier treatises, for laÌn, which is the time-honoured term, see his article ‘LaÌn al-¨Amma’, in EI2. 39 Gully, ‘Arabic Linguistic Issues’, 79. 40 Pellat, ‘LaÌn al-¨Amma’, 607–8. 37
524
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
al-Yaziji’s Lughat al-jara}id (The Language of Newspapers, 1901) and As¨ad al-Daghir’s Tadhkirat al-katib (Memoirs of the Writer, 1923) are good examples of this literature in the later part of the nineteenth and early twentieth centuries. While these works prefer different terminology (kha†a} and Òawab instead of laÌn), they nevertheless echo techniques found in similar works of the medieval period and share the same concern for the purity of the Arabic language. Besides, it is significant to note that the concerns over the Arabic language first expressed by conservative scholars like al-Shartuni, alYaziji and al-Daghir, were eventually given a forum through the institutionalisation of language academies in the twentieth century. These academies arbitrated over linguistic issues and published articles on this in journals. For instance, the articles entitled ¨Athrat al-aqlam (Slips of the Pen), in the issues of the Journal of the Arab Academy of Damascus (Arabic) after 1921, were sequels to similar articles published by al-Shartuni and al-Yaziji decades earlier in al-Muqta†af and al-Hilal. Despite this conservative venture, it appears that their calls have largely gone unheeded. In contrast, the efforts initiated by reformist scholars like al-Shidyaq towards the simplification of Arabic have prevailed. This is obvious not only by the development of modern literary Arabic, but by the development of what might be considered a distinct linguistic genre in the Arabic press, mainly concerned with the communicative significance of the Arabic language.41 Grammatical Terminology Before examining al-Shartuni’s arguments over al-Shidyaq’s use of grammatical terminology, it is worth considering some views on the development of Arabic grammatical terminology which may help this discussion. Gully points out that much of the terminology employed by grammarians over the centuries to interpret their sciences remained constant. However, not only did some of these terms assume different meanings later, but their usage also covered a wider sphere as approaches to the study of language diversified.42 A similar view is expressed by Owen who states that the linguistic heterogeneity of the earlier period, where linguists were relatively free to mix ideas and terminology from different sources, gave way to one of greater 41
Clive Holes, Modern Arabic: Structures, Functions and Varieties, rev. edn (Washington D.C. 2004), 314. 42 Adrian Gully, Grammar and Semantics in Medieval Arabic: A Study of IbnHisham’s Mughni (Surrey 1995), 9. 525
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
homogeneity culminating in al-Sarraj’s al-UÒul fi al-naÌw (The Principles of Grammar), a work which effectively standardized Arabic grammar. He explains that the emergence of schools in the ninth and tenth centuries, such as BaÒran and Kufan, can be understood as a reflex of this standardization. This allowed linguists to recognize the diversity of the earlier period by assigning ideas and terminology which did not become standard to a subordinate Kufan position.43 So, as the field of Arabic grammar developed and became increasingly specialized in the ninth and tenth centuries owing to the diversification of language studies and emergence of schools, some terms used by earlier grammarians like Sibawayhi were substituted, some added, and some took on different meanings. However, despite the diversity afforded by these factors, standardization has eventually taken precedence and subsequently ensured the stability of Arabic grammatical terminology. In one sense, it might be argued that this standardization was necessary, if the purpose of terminology was to distinguish the status and function of various grammatical components. In this first dispute al-Shartuni objects to al-Shidyaq’s use of terminology in defining the lam of command (lam al-amr). Al-Shidyaq states: al-amr bi }l-lam an tazid fi awwal al-mu∂ari¨ laman maksuratan wa-tusakkin akhirahu (the imperative with lam adds a genitive lam to the beginning of the imperfect verb and makes quiescent (tusakkin) the final letter).44 Al-Shartuni, however, responds that scholars have unreservedly agreed (qad a†baqat ara} al-¨ulama} ¨ala }l-qa†¨i) that the verb after the lam of command (lam al-amr) becomes jussive (jazm), as the lam al-amr is unanimously (bi }l-ijma{) considered among the jussive particles. He takes issue with al-Shidyaq’s words wa-tusakkin akhirahu (makes the final letter quiescent) which he says is an unequivocal error (kha†a} bayyin) because the quiescence (taskin) belongs to the uninflected/indeclinable words (al-bina}) and not to the inflected/ declined words (al-i¨rab). Al-Shartuni first quotes Ibn Malik’s Alfiya, as follows: bi-la wa-lam in †aliban ∂a¨ jazman fi}l-fi¨l ha-kadha bi-lam wa-lamma (with la, lam, lam and lamma the student makes the verb jussive), and then supports his argument with citations from earlier grammars and commentaries on the Alfiya.45 Al-Shartuni thereby 43 Jonathan Owens, ‘Models for Interpreting the Development of Medieval Grammatical Theory’, Journal of the American Oriental Society 111:2 (1991), 225. 44 Al-Shidyaq, Ghunya, 17. 45 Ibn Malik (1205–74), well-known grammarian of the medieval period. His Alfiya is an abridged linguistic versification on grammar of a 1,000 rhymed couplets
526
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
argues that the imperfect verb with lam of command is inflected in the jussive (mu¨rabun majzum) and adds that inflection (i{rab) is connected to the last part of the inflected/declined (mu¨rab) as is attested by the following definitions. Al-Shartuni cites al-Ushmuni46 who says: ‘The i¨rab change in the last part of a word owing to the governors which interact with them, both in form (lafÂiyan) and in reconstruction (taqdiran).’ He also quotes al-∑abban: ‘Make conventional the use of the vowel ‘u’ (al-∂amma), vowel ‘a’ (al-fatÌa), vowel ‘i’ (alkasra), and quiescence (al-sukun) in al-i¨rab as raf¨, naÒb, jarr or khaf∂, and jazm.47 In al-bina} they should be named ∂amman, fatÌan, kasran and sukunan. So do not mix their usage!’48 Al-Asir accuses al-Shartuni of relying on transmission (naql) without using his brain (¨aql), adding that al-Shartuni has ignored what Ibn Malik, al-Ushmuni, and al∑abban have said on this matter in the same works. For instance, he cites Ibn Malik’s words as follows: Fa-irfa¨ bi-∂ammin wa-anÒiban fatÌan wa-jurra… wa-ajzim bi-taskinin wa-ghayru ma dhukir… 49 For al-Asir, Ibn Malik’s clause (wa-ajzim bi-taskinin) rationalizes alShidyaq’s definition of tusakkin. He explains that al-Shartuni is comparable to someone who reads the Quranic verse la taqrabu al-Òalata (do not approach prayer) and disregards the subordinate clause waantum sukara (while you are drunk).50 Al-AÌdab defends al-Shidyaq’s use of tusakkin by arguing that quiescence (sukun) is a common noun (ism ¨amm) which can function both as an entity of inflection/declension denoting jussiveness (i¨raban bi-ma¨na al-jazm) at the end of a word, and as an entity of indeclinability (bina}). Al-AÌdab notes several instances to support his argument, for example, from Ibn Malik: wa-ajzim bi-taskinin (make jussive with (mathnawi). For al-Shartuni’s quote, see Alfiyat Ibn Malik: sharÌ Alfiyat Ibn Malik fi}l-naÌw wa}l-Òarf, ed. MuÌammad ¨Abd al-¨Aziz (Tanta 1991), verse 695. 46 MuÌammad al-Ushmuni (1435–1513), grammarian of the medieval period. His most enduring work on grammar is his commentary on Alfiya known as Manhaj al-salik ila Alfiyat Ibn Malik. KaÌÌala, Mu{jam, 7: 183. 47 Raf ¨, naÒb, and jarr/khaf∂ denote case of inflected nouns while raf¨, naÒb, and jazm moods of inflected verbs. 48 Al-Shartuni, Sahm,14–15. See also ¨Ali ibn MuÌammad al-Ushmuni, SharÌ al-Ushmuni li-Alfiyat Ibn Malik al-musamma Manhaj al-salik ila Alfiyat Ibn Malik, ed. MuÌammad al-Îamid (4 vols, Cairo 1993), 1: 68–9. Al-Shartuni’s quotes from Alfiya and al-Ushmuni are perhaps taken from al-∑abban’s Îashiyat,1: 66, which incorporates both works. 49 Trans.: ‘Make [nouns and verbs] independent with vowel (u), make dependent with (a), make [nouns] oblique with (i)…Make [verbs] jussive/apocopated (ajzim) with quiescence (taskinin)….’ Al-Asir quotes verses 25–6 of Alfiya. 50 Al-Asir, Radd al-shahm, 15. 527
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
the quiescence), and al-∑abban: al-sukun fi al-i¨rab yusamma jazman (the quiescence used in the i¨rab is known as the jussive). He asks why al-Shidyaq cannot use the sukun to denote jazm, especially when the Kufans made no distinction between the titles of declension and titles of indeclinability.51 The second dispute involves al-Shidyaq’s use of the term raf ¨ in defining the pronoun naÌnu: wa-i¨rabuhu naÌnu ∂amir raf ¨ marfu¨ mubtada} (the inflection of naÌnu is a pronoun in the nominative/ independent case, and topic (of a sentence) in the nominative/independent case).52 In response, al-Shartuni states: ‘It has been agreed (min al-mu†baq ¨alayhi) that the pronoun naÌnu is uninflected/indeclinable (mabniya) like other pronouns. Even primary school students know that the nominative/independent case (raf ¨) is among the case endings of inflected/ declined words (awjah al-i¨rab), and not among the uninflected/ indeclinable words (al-bina} ).’ Al-Shartuni, based on al-∑abban’s sayings (aqwal), argues: ‘All indeclinable words are inflected in the place of another (al-mabniyat kulluha tu¨rab maÌallan). So, one can say that an entity is indeclinable on the vowel ‘u’ (mabni ¨ala al-∂amm), that is, in apposition to the nominative/independent case (raf ¨ ). AlShartuni contends that if al-Shidyaq had said ‘nominative/independent appositively (marfu¨ maÌallan)’, then his words would have escaped criticism. Al-Shartuni explains that his aim here is only to present the sayings (aqwal ) of the imams, and accordingly cites additional sayings to the same effect from al-∑abban’s work.53 Al-Asir responds that the titles of inflection/declension (alqab al-i¨rab) apply to the titles of indeclinability (alqab al-bina} ), and adds that if it were not for elision (al-Ìadhf ) and reconstruction (al-taqdir), there would be no understanding of grammar. Al-Asir comments that the donkey/ass (al-Ìamir) (i.e. al-Shartuni) should know that the taqdir here is in apposition to the nominative/independent case (marfu¨ al-maÌall ), since taqdir is commonplace in the speech (kalam) of the grammarians and al-Shidyaq’s use of it only shows his excellence.54 Before moving on to the next section, some observations on the previous two disputes are in order. It would appear that al-Shartuni’s arguments are aroused by minute variation in grammatical terminology. However, al-Shartuni’s disputes distinguish between the terminology 51 52 53 54
Al-AÌdab, Radd al-sahm, 11–12. Al-Shidyaq, Ghunya, 62. Al-Shartuni, al-Sahm, 70. Al-Asir, Radd al-shahm, 50. 528
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
used for inflected/declined words (al-i¨rab) and uninflected/indeclinable words (al-bina}). In this sense, al-Shartuni clearly represents the traditional view of many grammarians who held that grammar is the study of i¨rab and bina}. Al-∑abban, one of al-Shartuni’s main sources, for instance, defines grammar as: ¨ilm yubÌath fi-hi ¨an aÌwal awakhir al-kalam i¨raban wa bina}an (the science through which case endings of words are studied both in terms of declension and indeclinability).55 From this perspective, al-Shartuni’s disputes are motivated by much more than ‘minute changes in grammatical terminology’, but to appreciate this, we need to go back to the ‘master’, al-Sibawayhi. If we subscribe to the view that al-Sibawayhi’s authority in his Kitab rests on his comprehensive, unified treatment of Arabic, and his application of the theory of governance (¨amal ), governor (¨amil ), and governed on (ma¨mul fih) to all linguistic phenomena in his work, then his most enduring contribution to Arabic grammar lies precisely in making a sharp distinction between those word endings which change as an effect of another word (i.e. ¨amil ) and those words whose endings remain the same. In other words, his systematic, previously unprecedented use, of naÒb, jarr, raf ¨, and jazm to denote inflected/ declined words (al-i¨rab) and fatÌa, kasra, ∂amma, and waqf (sukun) for naming uninflected/indeclinable words (al-bina}). Versteegh’s neat translation from al-Sibawayhi’s Kitab illustrates my point here: …the endings of words follow eight courses: accusative (naÒb), genitive (jarr), nominative (raf¨), apocopate (jazm), a-vowel (fatÌ), i-vowel (kasr), u-vowel (∂amm), zero-vowel (waqf). The eight endings are arranged formally into four pairs: the accusative and the /a/ are one pair formally, likewise the genitive and the /i/, the nominative and the /u/, and the apocopate and the zero-ending. The reason I call them eight endings is to distinguish between … those words that receive one member of these pairs because it is produced by a governor (those are never permanent with the word) and those words whose last consonant has a permanent ending that never disappears as the result of the action of the governor. Each of the governors has its own formal effect on the last consonant of the word, and this consonant is called the consonant of declension.The accusative, the genitive, the nominative, and the apocopate occur on the consonants of the declension.56
Al-∑abban, Îashiyat, 1:15. Al-Shartuni quotes al-∑abban at least 30 times and therefore he is his main source. 56 Versteegh, Linguistic Thought III, 36–7. Here al-Sibawayhi refers to sukun as waqf. 55
529
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
Early grammarians thus primarily used their terminology to describe, categorize, and determine the function of language items in terms of their inflectional/declensional paradigms. Hence, in al-Shidyaq’s sentence: al-amr bi }l-lam an tazid fi awwal al-mu∂ari¨ laman maksuratan wa-tusakkin akhirahu, the lam, as the ¨amil acting on the imperfect verb, modifies the imperfect from the indicative/independent mood to the jussive/apocopate (jazm). So to use the verbal form of an indeclinable entity sukun (i.e. tusakkin), in which the ¨amil is explicitly understood to be ineffective, to denote an entity (imperfect verb) whose inflection/declension is categorically thought to be dependent on the phenomenon of ¨amil (lam), effectively undermines the whole concept of ¨amil as understood by earlier and later grammarians. If this is one of the main factors motivating al-Shartuni’s ‘minute concerns’ over terminology above, then this is essentially the same concern underlying al-Shartuni’s disputes in this section. An interesting phenomenon in al-Shartuni’s disputes is his technique of employing certain terms against al-Shidyaq to implicate a consensus of scholars on grammatical matters.57 Gully points out that various terms are pressed into service in a seemingly hierarchical fashion to deal with the different levels of agreement reached by grammarians on a given linguistic point. He states: ‘At the head of the hierarchy is ijma¨, followed by ittifaq, and to a lesser extent i†baq, which seem to convey the idea of a general agreement, but certainly not consensus. Jumhur and jama¨a ‘group’ are two categories which suggest a majority opinion.’58 Gully indicates that a hierarchy of agreement on linguistic issues would seem to be in operation through these terms, and needs further examination. Before elaborating on this point, Suleiman’s work on Arabic grammatical theory helps shed light on al-Shartuni’s motives. In his treatment of istiÒÌab al-Ìal (presumption of continuity), Suleiman points out that arguments from ijma¨ must always take precedence over arguments based on istiÒÌab. For instance, there is consensus among grammarians to treat jazm in the imperfect verb (jussive) as a declension ending, although it is realized as a zeromorph in speech. Suleiman further cites Astarabadhi as follows: ‘Since the imperfect verb after a jazm particle (e.g. lam) does not appear with an overt vowel declension, and since in this case the verb appears in its emic morphologically unaugmented state, it would be For instance, his reference to qad a†baqat ara} al-¨ulama}…bi}l-ijma¨, and min al-mu†baq ¨alayhi. 58 Gully indicates that ijma¨ is the optimum form of agreement on linguistic issues. Gully, Grammar, 50–5. 57
530
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
possible to argue on the basis of istiÒÌab that zero-ending is not a declension ending’. However, Astarabadhi refrains from following this solution because it would effectively give precedence to an argument based on a putative istiÒÌab over one based on ijma¨. He expresses this view self-critically: ‘Were it not for the fact that it is loathsome to depart from the consensus of the grammarians, it would have been acceptable to regard the imperfect verb in the jussive as indeclinable’.59 It is proposed, then, that a major factor underlying al-Shartuni’s refutation is al-Shidyaq’s contravention of this hierarchical system of agreement among grammarians. His approach has been to evoke this system of agreement at the start of his argument and then present almost homogeneous sayings (aqwal) from earlier works in order to illustrate the level of agreement, lending added authority to his argument, and showing how al-Shidyaq’s grammar falls short. Al-Shartuni is careful to use terms which reflect a higher level of agreement when the sayings of Ibn Malik, al-Ushmuni, and al-∑abban, for instance, have converged on a particular grammatical point, while he uses terms which reflect a lower level of agreement, or no agreement at all, when the sayings of earlier scholars diverge.60 In fact, even when al-Shartuni cites only al-∑abban’s work, the reference is adequate to support a relatively weak argument and reflect a certain level of agreement, particularly given the compilatory nature of al-∑abban’s writing.61 According to this view, it is the convergence of sayings that constitute the jama¨a and jumhur in the hierarchical system, and which becomes the criterion worthy of being imitated. For al-Shartuni these levels of agreement function as a validating instrument, and more importantly signify that a particular issue has become epistemologically certain and thus uncompromising to further interpretation. In other words, he effectively evokes the epistemological value attached to this level of agreement which renders this instrument so powerful in the realm 59 Yasir Suleiman, The Arabic Grammatical Tradition: A Study in Ta}lil (Edinburgh 1999), 36–7. 60 This system of agreement is determined when al-Shartuni refers to these sources and finds that grammarians have agreed on a particular grammatical matter (description, explanation, terminology). The levels of agreement therein are represented through various terms which directly depend on the convergence or divergence of the sayings (aqwal) of imams on that matter. Al-Shartuni adopts this approach throughout e.g.: wa-ajma¨a al-jumhur (the majority have unanimously agreed, 21); and al-rasm ¨inda al-mun†aqin (the practice among scholars, 29). 61 Al-∑abban brings together ‘sayings’ from many commentaries and grammars. Al-Shartuni can thus induce a certain level of agreement from his work alone when he finds ‘sayings’ therein converge on a grammatical matter.
531
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
of criterion, practice, and established norms in the community of grammarians that it can override an unestablished practice as it were. This hierarchical system of agreement is understood as one which knows no time or place, nor is it bound by a certain set of sources. It is instead a self-induced (referential) system in which the level of agreement is generated from the sources at one’s disposal, and then represented in proportion to the convergence of sayings (aqwal). Definition of Grammatical Phenomena In this final category, al-Shartuni takes issue with al-Shidyaq’s simplified descriptions of certain grammatical points. The first dispute regards al-Shidyaq’s definition of the appositives (al-tawabi¨) in Arabic,62 which reads as follows: al-tawabi¨ jam¨ tabi¨ wa-huwa fi ¨urf al-nuÌah kullu thanin tabi¨a ma qablahu fi i¨rabihi (al-tawabi¨ is plural for tabi¨ which according to the grammarians is every second entity which follows the inflection/declension of the first).63 For al-Shartuni, al-Shidyaq’s explanation of the appositive (tabi{) here is clearly corrupt (fasad bayyin), since it would include grammatical phenomena which have no relation to apposition. For instance, the predicate (khabar) in Zaydun qa}imun (Zayd is standing), the accusative of specification (tamyiz al-manÒub) in ishtaraytu ra†lan zaytan (I purchased a ratl of oil), and the circumstantial qualifier (Ìal) in laqaytu Zaydan rakiban (I met Zayd riding). Al-Shartuni cites several examples from earlier works to illustrate his point. For example, he quotes Ibn ¨Aqil64 as saying: huwa al-ism al-musharik ma qablahu fi i¨rabihi mu†liqan (it is the noun which shares the declension of what proceeds it without exception). Ibn ¨Aqil comments: ‘This excludes the subject, predicate, and circumstantial qualifier, as these only follow the declinability of what proceeds them in some circumstances.’ Al-Shartuni also mentions Ibn alÎajib’s definition.65 After recounting al-Shartuni’s side of the argument, al-Asir says: ‘Each entity declines for its completeness (yanÒarif li-kamilihi). So the 62 Five classes of appositives in Arabic: emphasis (tawkid), adjective (Òifa/na¨t), substitute (badal), coordinate (¨a†f), explicative (a†f bayan). These agree with the word followed (al-matbu¨) in number, definiteness, case, and gender. 63 Al-Shidyaq, Ghunya, 74. 64 Ibn ¨Aqil (1294–1367) grammarian of the medieval period. His works on grammar include commentaries on Ibn Malik’s Alfiya and Tashil. J. Schacht, ‘Ibn ¨AÈil’, in EI2. 65 Al-Shartuni}, al-Sahm, 75–6. For his quotes, see SharÌ Ibn ¨Aqil, 2: 151 and Ibn al-Îajib, al-Kafiya, 1: 298.
532
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
reconstruction/implication of a missing syntactical part (taqdir) here is that each and every second entity which follows what comes before it, also follows it in its inflection/declension without exception.’66 Al-Asir claims that al-Shidyaq’s definition is valid because his grammar is abridged (ikhtiÒar). Al-Shartuni’s second objection is to al-Shidyaq’s definition of the position of the circumstantial qualifier (Ìal) in Arabic.67 Al-Shidyaq defines its position in a sentence simply as: yajuz taqdim al-Ìal ¨ala ¨amiliha naÌwa: rakiban ja}a Zaydun (it is permissible to put the Ìal before its governor (¨amil), as in, for instance: riding, came Zayd).68 Al-Shartuni responds that al-Shidyaq’s definition implies that the Ìal can be positioned before its governor (¨amil) under all circumstances while this is certainly not the case, for it is restricted (muqayyad) to certain situations (aÌwal). For example, it can be obligatory (wajiban), permissible (ja}izan) or even forbidden (mumtani¨an), according to the context. He quotes al-Ushmuni that the Ìal in relation to its {amil can occur as follows: ‘It is obligatory (wajib) to place the Ìal before the ¨amil in sentences like: kayfa ja}a ¨Amru (how did Amr come?) while it is obligatory to put the Ìal after, if the amil is invariable, for instance, in ma aÌsana Zaydan kha†iban (how beautiful an orator is Zayd). However, it is permissible (ja}iz) to put the Ìal before or after the ¨amil when none of the above conditions apply, as in, for instance, rakiban ja}a Zaydun.’69 Al-Asir responds that al-Shidyaq has restricted his definition to those instances where it is permitted (jawaz) to place the Ìal before the ¨amil as these are more prevalent. He adds that al-Shidyaq wanted to provide a short explanation (ikhtaÒara) like Ibn Malik, for instance, in: wa-al-Ìalu in yunÒab bi-fi¨lin Òurrifa fa-ja}izun taqdimuhu ka-musri¨an
66
aw Òifatin ushbahat al-muÒraffa dha raÌilun wa-mukhliÒan Zaydun da¨a 70
Al-Asir, Radd al-shahm, 52. The position of the Ìal in a sentence is usually after the governor (¨amil), but can precede a variable ¨amil, e.g. rakiban ja}a Zaydun, but not an invariable like fi¨l al-ta¨ajjub/al-taf∂il. Therefore, one cannot say ∂aÌikan ma aÌsana Zaydan. 68 Al-Shidyaq, Ghunya, 56. 69 Al-Shartuni, Sahm, 63–4. For al-Shartuni’s quote from al-Ushmuni, see SharÌ al-Ushmuni, 2: 306. 70 Paraphrased trans.: ‘The Ìal can precede the governor if it is a variable verb or an adjective resembling the variable verb, e.g. mukhliÒan Zaydun da¨a, and musri¨an dha raÌilun.’ Al-Asir’s quote from verses 343–4 in the Alfiya. 67
533
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
Al-Asir then argues that al-Shidyaq, in line with the aims of his work (i.e. simplification), has excluded rare instances of a phenomena (qalil) in favour of those instances which occur more frequently, because the former would require longer detailed explanations (mu†awwilat).71 Al-AÌdab, however, dismisses al-Shartuni’s arguments as baseless and criticizes him for recounting all the instances of how and where the Ìal should occur, which for al-AÌdab is even known among primary school students. He explains that al-Shidyaq has disregarded these instances as they are well known among Arabs and adds that al-Shidyaq’s aim is to provide an abridged grammar (kitab mukhtaÒir) and is therefore unable to give exhaustive treatment to all grammatical matter.72 In the above disputes al-Shartuni’s objections are aroused by the ‘simplified, abbreviated’ nature of al-Shidyaq’s definitions of grammatical phenomena. For al-Shartuni, al-Shidyaq’s definitions are ‘inadequate’ because they neither take into account the exceptions nor the conditions under which a particular grammatical phenomenon can occur. Using a prescriptive/normative approach al-Shartuni measures al-Shidyaq’s definitions against earlier commentaries. These commentaries serve al-Shartuni’s purpose well against al-Shidyaq’s ‘abridged work’ since they are exhaustive in their descriptions of grammatical phenomena. One may therefore decide to sympathize with al-Asir and al-AÌdab’s reoccurring references to ikhtiÒar, mukhtaÒir in order to justify the ‘inadequate’ nature of al-Shidyaq’s definitions. It would also seem that al-Shidyaq’s ‘abridgement’ does not break away from the Arabic grammatical tradition in any significant way, as al-Shartuni implies, especially since examples of abridged grammars can be found both in the medieval and pre-modern periods. The thirteenth-century grammarian Ibn al-Îajib, for instance, abridged several earlier works. In the pre-modern period, NaÒif al-Yaziji wrote one of the most abridged works of his time, entitled al-Jawhar al-fard (The Unique Gem), in which he claims to have gathered some of the most famous sayings (aqwal) of grammarians in just six pages.73 Al-Shidyaq’s ‘abridgement’, however, differs considerably from these works in one major aspect. The main pedagogical function of abridgements (ikhtiÒarat or takhliÒ), which were often presented as linguistic versifications, was to facilitate the learning (i.e. memorization) of grammar. The Alfiya, for instance, is one work which served this 71 72 73
Al-Asir, Radd al-shahm, 46–7. Al-AÌdab, Radd al-sahm, 35–6. Zaydan, Tarajim, 27. 534
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
purpose. It was the commentary, however, which served the main pedagogical purpose of explaining the grammatical material in these abridgements. So, while the abridgement played a limited pedagogical role as an aid to memorisation in the classroom or reference work for the commentary, the commentary functioned as a pedagogical heavyweight during this time. Al-Shidyaq’s ‘abridged’ grammar, however, was not intended to serve the perennial tradition of learning (memorization), or as a basis for a larger commentary, but for the later pedagogical-cum-commentary purpose of ‘teaching grammar to Arabs and non-Arabs’. Kamal al-Yaziji’s observations serve my point here. He explains that until the nah∂a, commentaries on earlier grammars (e.g. by al-Ushmuni and al-∑abban on Alfiya) were used to teach Arabic grammar, while students were encouraged to memorize linguistic versifications/abridgements because this would develop in them the aptitude to grasp the rules of grammar. The problem, however, arose when some nah∂a scholars began deviating from this practice, and chose to teach the language through abridged/simplified works. He states that they compiled user-friendly pedagogical works which provided brief explanations of grammatical phenomenon, and adds that one of the forerunners in this was al-Shidyaq in his Ghunya.74 Thus, in contrast to the Arabic grammatical tradition where abridgements were employed for learning (memorization) purposes or as cornerstones for commentaries, al-Shidyaq, al-Asir and alAÌdab evoke the ‘abridged’ aspect of the Arabic grammatical tradition as an alibi to simplify the rules of Arabic grammar. Al-Shartuni’s problem, however, is not so much over the abridged nature of Ghunya but with al-Shidyaq’s attempt to break with the perennial tradition of using abridged works for learning and commentaries for teaching. In other words, al-Shidyaq produces an abridged (simplified) work to serve the pedagogical purpose which the commentary tradition had primarily served. Al-Shartuni, in response, compares the Ghunya with these ‘normative’ commentaries and glosses, and highlights the ‘inadequacy’ of al-Shidyaq’s own interpretation and description of grammatical phenomena. In this manner, he not only undermines the pedagogical credibility of his Ghunya, but more significantly warns of the ramifications of breaking from this traditional pedagogical-cum-commentary mode of expression with a simplified/ abridged grammar.
74 Kamal Yaziji, Ruwwad al-nah∂a al-adabiya fi lubnan al-Ìadith 1800-1900 (Beirut 1962), 78–9.
535
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
Conclusion Eight disputes representing four main types of arguments in alShartuni’s Sahm as a whole (i.e. personal motives, grammatical errors/ slips of the pen, grammatical terminology and definition of grammatical phenomena), shed light on some of the underlying factors prompting al-Shartuni’s refutation and his position as a grammarian. The section on ‘grammatical errors/slips of the pen’ not only illustrates the intensity of al-Shartuni’s proofreading skills and prescriptive tendencies, but also the level of scrutiny nah∂a scholars imposed on one another. Al-Shartuni’s concerns in these disputes are genuinely aroused by grammatical errors in al-Shidyaq’s Ghunya, irrespective of the fact that they may have been printing errors. Al-Shartuni’s disputes under grammatical terminology demonstrate that his arguments are motivated by much more than minute changes in grammatical terminology. They are aroused because al-Shidyaq’s unconventional use of terminology undermines the whole concept of {amil as understood by some earlier and later grammarians. One of the main underlying factors motivating al-Shartuni’s disputes is al-Shidyaq’s contravention of the apparent system of agreement among grammarians. For al-Shartuni, these levels of agreement function as a validating instrument, and signify that a particular issue has become epistemologically certain and thus uncompromising to further interpretation. Thus, the role of grammar for al-Shartuni is not so much to communicate or clarify as it is to preserve the norms (sayings) that these levels of agreement reflect in earlier grammars and commentaries. Al-Shartuni’s objections to al-Shidyaq’s simplified definitions of grammatical phenomena reveal that his problem is not so much over the abridged nature of Ghunya but rather with al-Shidyaq’s attempt to break with the pedagogical-cum-commentary tradition. One of the most significant underlying factors motivating al-Shartuni’s refutation therefore lies in the pedagogical ramifications of al-Shidyaq breaking from this tradition with a simplified grammar. Al-Shartuni’s concept of ‘correctness’ is determined by the paradigm of grammatical works and commentaries of the medieval and premodern periods. In setting out al-Shidyaq’s unorthodox position as a grammarian in light of these works, al-Shartuni effectively confirms his own position as an orthodox grammarian. In so doing, it is sufficient for him to show that the Ghunya digresses from the established norms (i.e. aqwal ‘sayings’) laid down in these sources. The strength of al-Shartuni’s argument in this manner rests upon the perennial 536
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
strength of traditionalism in Arabic grammar. For him this derives from the testimony of earlier revered grammars (Alfiya and al-Kafiya) and is preserved through a series of commentaries and glosses which essentially seek to accommodate the norms established in these authorities. The influence of these commentaries and glosses is unequivocal in al-Shartuni’s refutation, a few of which, in particular, constitute the core of his criteria (al-Ushmuni and al-∑abban). Indeed, al-Shartuni’s technique is to echo the ‘sayings’ in these ‘authoritative’ works. This not only serves to authenticate his arguments, but also keeps in check any material that does not meet the established norms laid down therein. It might seem from all this that the Sahm is little more than a record of ‘sayings’ from previous grammars and commentaries, but this would be a gross underestimation of its content and significance. The merit of al-Shartuni resides in the manner in which he brought together these ‘sayings’ and marshalled them in his arguments. AlShartuni adopts a normative approach that involves evaluating the Ghunya against the paradigmatic of earlier commentaries and glosses. In so doing, he employs several terms in an apparently hierarchical pattern to represent the level of agreement, which he felt earlier grammarians had reached on a particular grammatical point. The level of this agreement is represented by terms which are self-induced by al-Shartuni, and whose hierarchical status depends on the extent to which the sayings (aqwal) of these grammarians converge or diverge. He thus not only highlights the inadequateness of al-Shidyaq’s own description of grammatical phenomena, but also warns that al-Shidyaq has contravened the orthodox view among grammarians. From this perspective, al-Shartuni is not so much interested in aspects of organization as he is in the persistent reiteration of a simple formula of grammatical orthodoxy ‘i.e. sayings’, which for him is the only ‘established’ way. Al-Shartuni’s criticisms are therefore limited to those instances where al-Shidyaq has ‘digressed’ in reiterating this formula. He completely bypasses al-Shidyaq’s direct citations from medieval grammars which, in fact, constitute the main bulk of Ghunya, and are by no means easy to distinguish. This not only reflects al-Shartuni’s textualist vision of grammar, but also his in-depth knowledge of these works. More importantly, by employing this normative approach, alShartuni warns of the ramifications of breaking from the pedagogicalcum-commentary traditional mode of expression, and, in effect, reinforces this mode of expression against seemingly innovative attempts (i.e. al-Shidyaq’s simplified Ghunya). Approached from this angle, al-Shartuni’s refutation might be seen as an essential component in 537
LANGUAGE REFORM AND CONTROVERSY IN THE NAHΔA
the tradition of normative grammars, commentaries and glosses, and if one of the aims of these works was to keep grammar and pedagogy alive in its traditional framework, then al-Shartuni’s refutation perhaps contributed to this endeavour in the nah∂a. Al-Shartuni, al-Asir, and al-AÌdab evoke the same sources. Yet, they disagree over the interpretation of the sayings (aqwal) therein. This is because the rivalry between these scholars reflects an irreconcilable methodological polarization whose origins were artificially projected back to earlier grammarians. Al-Shartuni embodies the systemological view of rigour and precision in the Arabic grammatical tradition while al-Shidyaq, al-Asir, and al-AÌdab represent the variationist view of flexibility and adaptability in the same tradition. In this sense, the disputes between these scholars represent the wider power struggle in which the right to interpret the past creates a highly competitive environment of professional scholarship, where the scholar gains prestige as a vehicle for tradition. One is therefore pressed toward demonstrating his grasp of speculative minutiae and skill in deep argumentation. Al-Shidyaq is generally thought to be one of the foremost reformists of his era, and in this sense his taste for the simplification of grammar is firmly rooted in the framework of his wider efforts to make Arabic a workable instrument of modern communication. In this area, the reformist group of scholars probably made substantial progress in introducing a simple and functional style which forms the basis of the newspapers and modern literary Arabic. However, this progress is not matched in the fields of grammar, where conservatism has been particularly fierce. The level of this conservatism is obvious in the nah∂a when it is considered that al-Shidyaq’s work was largely based on medieval grammars and commentaries, but still al-Shartuni managed to single out seventy-six fine points where al-Shidyaq ‘digressed’ out of his drive for simple presentation. In this sense, alShartuni’s conservatism in refuting probably the earliest example of a simplified grammar in the nah∂a, al-Shidyaq’s Ghunya, has perhaps been a significant contributing factor in undermining the reformist calls for simplification. If the function of grammarians is thus to preserve grammar in its traditional framework and hence safeguard the purity of the language whether through treatises, commentaries, abridgements and refutations, then al-Shartuni has a deserved position on the long list of Arab grammarians. Address for Correspondence: [email protected] Oriental Institute, University of Oxford, Pusey Lane, Oxford, OX1 2LE 538
Journal of Semitic Studies LV/2 Autumn 2010 doi: 10.1093/jss/fgq012 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
THE FOLK SONGS OF IRAQI CHILDREN (PART 2) SADOK MASLIYAH DEFENSE LANGUAGE INSTITUTE, MONTEREY, CALIFORNIA
Part D: Teasing and Taunting Songs (Songs 1–32) Certain types of occupations and bodily defects such as baldness, blindness, flat-noses and pug-noses and other bodily defects are a ground for insult in Iraqi culture, because they are considered serious heavenly curses. If several members of the same family are affected with different bodily defects, the people say, ya sattar hadha gha∂ab, ¨O Protector, this is [God’s] wrath’. Professions, like horse-carriage, camel and donkey drivers, are also subjects to insults by some Iraqi children. The following Iraqi proverbs demonstrate the negative attitude towards such deformities and reflect the popular ideas connected with them. a. Agra¨ bi-flaysatha,1 ¨by his money, a bald person [can cover his defects]’, (lit, ‘a hairless person with his money’), said about the power of money. The following proverb is said about a disappointed person who experienced the opposite of what he had expected, jibit il-agra¨ yiwannisni kishaf rasah u-kharra¨ni ,2 ¨I brought the bald man to entertain me, but he revealed his head and scared me’. b. AÌdab u-yitçaqlab,3 ¨he rolls [despite] being a hunchback’, said about someone who misbehaves in a way that reveals his shortcomings. c. To belittle a short person, some Iraqis say, ib†ul il-zibb,4 ¨[he is] as long as the penis’ and †uruh,5 ¨miscarried foetus’, because he was born before due time, hence he is under-developed, and i††uwil yakul tin wal-qaÒir Ìazin,6 ¨the tall person eats figs and the short is sad’. 1 2 3 4 5 6
Al-Îanafi, 1962, I: 151. Ibid., 128. Ibid., 124. Al-Shalji, 1982a, I: 356. Al-Shalji, 1982b, I, 235. Al-Îanafi, 1962, I: 244. 539
THE FOLK SONGS OF IRAQI CHILDREN
Song 1: Agra¨ Mgarga¨
Bald
Baghdad, Îajj 5:90 When a boy cuts his hair with a razor and becomes bald, his playmates tease him. This type of hair cut is known as ‘number zero’ (namrah Òufur) haircut. Agra¨ mgarga¨ 7 Bal bi†-†awah Yibçi ¨ala8 ummah Yrid baqlawah
Bald is worn-out He peed in the frying pan He cries for his mother He wants baklavah
Song 2: Qaççum
The Pug-nosed Boy
Îajj 3:394 The following verses tease a pug-nosed child: Ya-ghrab, ya-ghrab Ma shifit qaççum? Qaççum nayim bish-shamis Wul-khashim mag†um9
O crow! O crow! Haven’t you seen the pug-nosed boy? Qaççum is sleeping under the sun And [his] nose is clipped
Song 3: Îajji GÒayyir
Mr Shorty
Karbala}, Îajj 2:164 Children follow a short man in the alleys and chant behind him: Îajji 10 gÒayyir11 Billayl isayyir Fog il-manarah Yishrab sigarah
Mr. Shorty At night he takes a walk On top of the minaret He smokes a cigarette
7 The word mgarga{ makes no sense, but it creates assonance with agra¨. It means ‘rattling’ (lit., ‘worn out’). 8 In a variant cited in al-Îajjiyyah, 1968, II: 159, the word yibçi, ‘he cries’ is replaced with yiji, ‘he comes’. 9 Mag†um is ‘chewed, cut’. In the Baghdadi dialect /q/ > /g/. 10 Îajji is a person who has performed the duty of pilgrimage in Mecca. 11 Gsayyir, dim. of giÒir, ‘short’. An Iraqi proverb demeaning the short person says: ‘i††uwil yakul tin wul-qaÒir ymut Ìazin’, al-Îanafi 1964, II: 244, ‘the tall person eats figs and the short person dies sad’. If we accept the explanation that figs represent woman’s breasts in Iraqi lore, then the implication is that a short person cannot reach them while standing.
540
THE FOLK SONGS OF IRAQI CHILDREN
Song 4: Qanqun
Qanqun
Karbala}, Q 3:519 Qanqun is the nickname of a nice man who lived in Karbala} in the 1930s. He was a keeper of a mosque and people gave him this nickname because of his nasal-sounding speech. Children sang this song at his funeral and it has become a popular folk song since then. Qanqun is probably a distortion of khankhun (khann ¨to speak from the nose’). Qanqun mat wunkisar ¨aÒ¨uÒah RaÌat ummah lil-gabur datbusah Qanqun mat wunÌiram [min] shamm il-hawa La shirab musahhil Wala ∂ag id-duwa
Qanqun died [after] his coccyx broke His mother went to the grave to kiss it Qanqun died and was deprived From breathing air He neither drank laxatives Nor tasted medicine
Song 5: ¨Arrujah Binti
My Daughter ¨Arrujah
Baghdad, Q 3:84 Trying to soothe the pain of a boy who has injured his leg while playing, the other playmates sing this song, hoping that when he hears the song he will laugh and forget his pain. If the players are girls, they may sing it in order to tease the injured girl. ¨Arrujah binti Kull il-banat itzawjan12 Δallayti inti
My daughter ¨Arrujah All the girls got married [But] you stayed [unmarried]
Song 6: ¨Ishsh il-Laglag
The Stork’s Nest
Baghdad, Q 2:138 One boy secretly puts a small piece of lemon peel or paper on the head of another boy. If the latter does not feel it, all the boys sing this song, hinting that there is something over the boy’s head. ¨Ishsh il-laglag Fog ir-ras
The nest of the stork Is above the head
Likewise, when a big boy hurts a younger boy, the other children rebuke the aggressor (Îajj 2:161), singing: 12 The retaining of the nun al-nuswa in itzawjan as a suffix for f.p. is used mainly by villagers and Bedouins. For a slightly different version, see al-Îajjiyyah, 1968, II: 159.
541
THE FOLK SONGS OF IRAQI CHILDREN
Ibn il ibn il Ummak tiftil wabuk yighzil ib-bab i†-†olah yishrab bolah
Son of a [dog] son [of a dog] Your mother twists [the wool] and your father spins [it] At the gate of the stable he drinks his urine.
Song 7: Ya M ¨aydiyyah!
O the Buffalo Raiser!
Basra, D: 113 Most Iraqis consider breeding buffaloes a demeaning profession and some Iraqi children in villages tease a female buffalo breeder while she is milking her cows and selling the milk in their neighbourhood. The following proverbs demean the M¨aydiyyah (s.f. of m¨aidi): il-m¨aydi ishmidrih ib-akl il-ni¨na¨,13 ¨what does a buffalo-breeder know about eating mint’? and idha indall il-m¨aydi bab Ìoshak Ìawwilha,14 ¨if the water buffalo-breeder finds the door of your house, change it’. One boy pretends he is going to jump into the river with his clothes on, imitating the buffalo-breeder who swims with her clothes on. His friends sing: Ya-m¨aydiyyah tab ¨u-tab! Tab u-†ab! Izbayditiç li-wlaydiç15 U-bintiç li-†-†ullab
O buffalo raiser, how good it is, how good it is! How good it is, how good it is! Your butter is for your little son And your daughter is for those who ask for her hand
Song 8: ¨Arabançi Dhibb il-Qamçi! Horse Carriage Driver, Throw away the Whip! Ramadi, Q 3:72 About the end of the nineteenth century the horse carriage was introduced to Iraq as means of transporting people. The boys used to hang on behind the horse carriage while it was moving and sing: ¨Arabançi ¨Arabançi! Dhibb il-qamçi min idak Hadha sh-shughul mayfidak Labbis martak ¨al-modah WuÒbugh liÌiytak sodah 13 14 15
Horse carriage driver! Horse carriage driver! Throw the whip from your hand This job is useless for you Dress your wife in fashion And dye your beard black
Khayya†, 1986, I: 32 and, McCarthy and Raffouli, 1964, II: 549. Al-Îanafi, 1962, I: 29. Wlaidiç and izbaidatiç are dim. of wuldiç and zibditiç respectively. 542
THE FOLK SONGS OF IRAQI CHILDREN
Song 9: Hadha-il-Qass
This Monk
Mosul, Q 3:358 It is related that a monk lived in a village near Mosul. He used to plant onion seeds (kasaÌ) that produced very small onions. Then he planted a different kind of onion called sawad. A Yazidi man stole them and planted them in his own yard. The monk decided to plant lettuce seeds instead. The same Yazidi stole them and planted them in his yard. Since the Yazidis are forbidden to eat lettuce, this song became popular among the children in Mosul. Hadha-l-qass Ma yinbis Yizra¨ buÒal Yi†la¨ khass
This monk Does not say anything He plants onions Lettuce comes out
Song 10: Hadha-l-Afandi
This Gentleman
Îajj 4:101 These verses mock those who wear western clothes and pretend to be westernized. The afandi is a borrowed Turkish word, effendi. Hadha-l-afandi min çidhib Baskultah16 dhayl-iç-çalib
This gentleman is fake His tassel is a dog’s tail
Song 11: Karradi! karradi!17 Resident of Karradah! Baghdad, Îajj 1:167–8 Girls chant this song and clap hands while sitting in a ring. In the 1930s the residents of Karradah, an outskirt of Baghdad, grew vegetables and greens instead of rice and barley. The neighbouring village, Qal¨at ∑aliÌ, mocked them by employing the insulting expression, ‘abu-l-baytinjanah’, ¨the grower of the eggplants’. Karradi karradi! Yabu baytinjanah Khalli-ç-çalib nayim Wum¨antir adhanah Khashshayt lis-sahaÌ, wulgayt tuffaÌ Wa¬¬ah ma akulha ¨ilma yiji khali Khali yabu baÌriyyah 16 17
O Karradi! resident of Karradah! O eggplants grower Let the dog sleep While he is perking his ears I entered the yard and found apples By God I won’t eat it until my uncle comes My uncle the sailor
Baskulah < bashkultah, where /s/ < /sh/. See a slightly different variant in Samarra} (Q 3:437). 543
THE FOLK SONGS OF IRAQI CHILDREN
Wu-¨mamtah Ìuriyyah Shalha-w-mu†alha bil-ga¨ ™il¨at minha bunayya Bunayya isimha ba††ah ™il¨ab bi-graysh il-Ìun†ah Ya rabbi la-tsa¨idha! Sa¨id banat il-jinjil Wuljinjil bidah rishah Wi†arid il-kidishah Kidishat ¨ammi salim akkalat il-awadim Khara ibruÌ jidditah
[In] his turban is a fair nymph He took it off and tossed it to the ground A little girl came out of it A little girl, her name is Battah Playing in the wheat grains O God, don’t help her! Help the girls wearing anklets And the anklet holds a feather in its hand Chasing the belly goat My uncle Salim’s belly goat that eats humans May manure [be] in the soul of her grandmother!
Song 12: Id ish-Shayib Rikhiyyah
The Old Man’s Hand is Weak
Baghdad, Q 3:45 While dancing, children sing the following song to ridicule an old man who marries young girls and has a second wife. Two boys stand facing each other, holding each other’s wrist. One of the boys shakes both wrists, while both boys sing together. When reaching the last word in the song, one boy tries to slap his partner, but the latter holds tightly the wrist of his friend and does not let him do so, then they change roles. Id ish-shayib tirtikhi Id il-wald gawiyyah
The old man’s hand becomes loose The hand of the boy is strong
Song 13: Shiliç bish-Shayib Dihiddih? What Do You Do with the Old Man? Karbala}, Q 2:314 This play song is similar to the song jatti ish-shamis (part D, song 15). Children hold hands and form a ring. They move around another child sitting crossed-legged in the centre of the ring. This child plays the role of an old man and tries to grab one of the children in the ring. During the game the following dialogue takes place: Group: Gum ya shwayb18 ¨an-it-tannur19 Child: Sirri ¨ayib magdar agum! 18 19
Get up old man, away from the clay oven. My secret is that I have a bodily defect, I can’t get up!
Shwayb, dim. of shayib, ‘old man’ used for demeaning. Tannur, ‘a house of light’, Arm. tannura. 544
THE FOLK SONGS OF IRAQI CHILDREN
Group: Hi wa¬¬ah?20 Child: I¨ta†ni maÌbas21 – Hi wa¬¬ah? – Δa¨ bil-mal¨ab – Hi wa¬¬ah? – Wana al¨ab – Hi wa¬¬ah?
Really? She gave me a ring Really? It got lost in the playground Really? While I was playing Really?
The following variant ridicules an old man by addressing his young wife: Najaf, Q 2:315 – Shiliç bish-shayib – Dihiddih ¨Indah ras çannah fas – Dihiddih! – Shiliç bish-shayib? – Dihiddih! – Kull-in-nas tu†bur bih – Dihiddih! – ¨Indah ∂ahir çannah nahir – Dihiddih! – Kull is-simaç tisbaÌ bih – Shiliç bish-shayib? – Dihiddih!
Song 14: Dahdahit ish-Shayib
What do you do with the old man? Let him go! He has a head like an axe Let him go! What do you do with the old man? Let him go! All people cut with it Let him go! He has a [curved] back like a river Let him go! All the fish swim in it What do you do with the old man? Let him go!
I Rolled the Old Man Down
Misan Province, Q 3: 109 This is another humorous song about an old man who practices polygamy and marries young girls. When girls see an old man walking aided by a stick they tease him. One girl acts like an old man with her back bent wearing worn-out clothes and a grey beard. A group of girls surround her while singing and dancing. Dahdahit ish-shayib Dehiddih Hadha-sh-shayib ghammazi U-¨ala-l-kuÌlah yuwazi Dahdahit ish-shayib La huwwa ¨aÒa-w-tisraÌi bih
I rolled the old man down Let him off This old man leers his eyes And he forces others to [use] kuhl I rolled the old man down He isn’t a stick you [can] graze with
20 Hi wa¬¬ah, When an Iraqi does not believe something said, he may respond saying sarcastically, hi wa¬¬ah!, really! 21 I¨†ini, ‘give me’. Most Iraqis would say in†ini instead.
545
THE FOLK SONGS OF IRAQI CHILDREN
Dahdahit ish-shayib La huwwa hawan wu-nhabbish bih Dahdahit i-sh-shayib Dehiddih! Dahdahit ish-shayib Dehiddih! U-la huwwa magzal u-gazil bih Dahdahit ish-shayib Dehiddih!
I rolled the old man down He isn’t a tray for polishing [grains] I rolled the old man down Let him off! I rolled the old man down Let him off! He isn’t a spindle with threads I rolled the old man Let him off!
Song 15: Ya ¨Ammi Jattak ish-Shamis O Uncle, The Sun Came to You Basra, D: 145–6 Girls stand in a circle in the centre of which sits another girl disguised as an old man. When the ‘old man’ hears her singing that his bride has arrived, he feels full of energy and looks forward to seeing her. The children clap and sing repeatedly, ‘Are you crazy?’ and add the expression ¨al barakah’, ‘congratulations!’ three times at the end of the song, mocking him. A dialogue takes place between the girls and the ‘old man’: Children: Ya ¨ammi jattak ish-shamis Old man: Magdar agum min Ìidbiti Children: Ya ¨ammi jalak iç-çalib Old man: Magdar agum min Ìidbiti. Children: Ya ¨ammi jattak il-faras Old man: Magdar agum mim Ìidbiti Children: Ya ¨ammi jattak il-¨arus
O uncle, the sun came up for you I can’t get up because of my hump O uncle, the dog came for you I can’t get up because of my hump O uncle, the mare came for you I can’t get up because of my hump O uncle, the bride came for you
Hearing this, the old man gathers his strength, gets up and follows the bride, but she walks away. Old man: Na¨am na¨am ¨iduha Yeah yeah. Say it again! Children: Ya ¨ammi laysh itkhabba¬it? O uncle, have you lost your mind?
Then the children clap and dance, singing: Ya ¨ammi laysh injannayt? ¨Al barakah ¨al barakah! ¨Al barakah
O uncle, have you become crazy? Congratulations! Congratulations! Congratulations!
Song 16: Ya Shallal!
O Shallal!
NaÒiriyyah, Q 3:338 A group of girls form a ring, clapping, singing, and dancing. This song is a parody about a man who marries two women. Note the 546
THE FOLK SONGS OF IRAQI CHILDREN
assonances in the proper name Shallal and the verb shallak, ¨may it paralyse you’. Shallal, ya Shallal! 22 Shallak il-bayn! Makhidh ¨ajayiz ¨ithnayn
Shallal, O Shallal! May death paralyze you! You married two old women
Song 17: Abu Jamilah23
Jamilah’s Father
Baghdad Abu Jamilah (d. 1950s) was an old man, who used to go from one celebration to another, especially to those of Jewish families. He earned his living from his performances, singing the same song about a person who had a daughter called Jamilah who caused him a lot of troubles. While singing and dancing he would clap his hands and make humorous motions, and the children would joke with him and touch his head while chanting: Abu Jamilah jimlutu24 ™alla¨ dimlah ib-khiÒwutu
Jamilah’s [daughter] angered him He grew a sore in his testicles
Song 18: Qashmarit ¨Alayk
I Fooled You
Baghdad, Îajj 2:160 When a child believes the lies his friends tell him, they make fun of him: Qashmarit ¨alayk25 Wul [henna] bidak Wul-henna bidayyah Wudh-dhahab bid a¬¬ah
I fooled you That the henna is on your hand But the henna is in my hand And the gold is in the hands of God
Song 19: Bil-¨Arabanah Jaboç In the Horse Carriage They Brought You Karbala}, Q 3:75 During the 1930s there was in Karbala} a ‘crazy’ old woman named Aminah who sold vegetables and fruits. Children used to follow her in the alleys and taunt her: 22
Shallal, admits to two meanings, either a proper name or ‘sloppy’. Kojaman, 2001: 96–7. 24 Jimluto, is a nonsense word to create assonance with jamilah. 25 The quadrilateral verb qashmar is derived from the noun ghashmurah (CA), which means ‘a claim without a proof’, see Îazim al-Bakri, 1972: 388. 23
547
THE FOLK SONGS OF IRAQI CHILDREN
Bil-¨arabanah jaboç Aminah! ¨Ishrin çaylah ∂irboç Aminah!
In the horse carriage they brought you [O] Aminah! Twenty shots they shot you [O] Aminah!
Song 20: Susan
Susan
Mosul, al-¨Ubaydi, 169 Susan is the name of a Christian woman who treated children’s eyes in Mosul applying folk medicine that caused them pain. The children there sang the following song about her: Susan Susani Susani GhaÌit ¨ind ikhtah Susani Îi††itla-l-takhtah Jaghghit il-makh†ah
Susan Susani Susani She went to her sister Susani She put down the armless chair for herself And drew the mucus
Song 21: ¨Al Barakah
Congratulations
Baghdad, Q 3: 97 Several children used to follow a pregnant woman in the street teasing her: ¨Al barakah ¨al barakah! Îiblat u-jabat tanakah
Congratulations! Congratulations! She got pregnant and gave birth to a piece of tin
Song 22: Îawi Ya Îawi
Hawi, O Hawi!
Baghdad, Îajj 2:160 The name Îawi is diminutive of YaÌya. In the 1930s there lived a ¨crazy’ beggar in Baghdad named Îawi. Children followed him in the alleys and teased him by singing a few verses. The name Îawi became an epithet for a madman and for a veteran beggar.26 Îawi ya Îawi! Yabu-tlath khaÒawi! WuÌdah shuguf 27 WuÌdah luguf WuÌdah akalha-l-wawi 26 27
Hawi, O Hawi! O the one who has three testicles! One is up One is down And one the jackal ate
Al-Îanafi, 1982, II: 330, al-Îajjiyyah, 1967, I: 164. It is the name of a children’s game. 548
THE FOLK SONGS OF IRAQI CHILDREN
Song 23: ¨Ali ¨al ¨al
¨Ali Fell Sick
Mosul, al-¨Ubaydi, 209 Playmates tease another player who fights with them and disturbs their games. The quadrilateral verb ¨al¨al, ¨to become ill’ is derived by duplicating the first two radicals of the noun, ¨illah, ¨illness’. ¨Ali ¨al¨al ™ubakh burghul Îa††u-b-¨ubbu ¨Ishta¨al zebbu
¨Ali became sick He cooked burgul He put it in his pouch [And] his penis burned
Song 24: Îawawa Shaqqu Daffu They Tore Hawawa’s Tambourine Mosul, al-¨Ubaydi, 207 A boy who plays with girls is considered a ‘sissy’ and is teased by other boys. Îawawa is a coined word from Ìawa} ¨Eve’, and it refers to a boy who plays with girls and behaves like them. He is also called zananah (Per. zan, ¨woman’). Îawawa shaqqu daffu Kill-il-banat ibÒaffu
Song 25: Zananah
They tore Hawawa’s tambourine All the girls in his class
Woman-like / Sissy
Basra, D: 90 Zananah ghir†afu28 dhanah ¨Ala li¨b il-bnayyat
They clipped the ear of the sissy For playing with the girls
Song 26: Za¨lan
Angry
Samarra}, Q 3:102 Boys sing the following rhymes when a playmate gets mad and leaves the game: Za¨lan ba¬¬ah radduh Bdhayl iç-çalib lakhlikhuh
He is angry, please bring him back Beat him with the tail of the dog
28 Gir†af, ‘to trim, to cut’. This quadrilateral verb is formed by adding /r/ after the first radical of the trilateral verb qa†afa (CA), to ‘pick’ (flower).
549
THE FOLK SONGS OF IRAQI CHILDREN
Song 27: Wargat Fijil
A Leaf of Radish
Musayyab, Q 1:274 Boys strike the ground with their right hands and sing using the cutting word zi¨rayw for teasing donkeys: Wargat fijil Kalha-il-¨ijil 29 Na††-u-mat Zi¨rayw!
A leaf of radish The calf ate He jumped and died Zi¨rayw!
Then one of the boys says †ar il-laglag, ¨the stork flew’ while lifting his hand upward. The rest say †irna wayyah, ¨we flew with it’. The losers are those who reply, ¨we flew with it’ when he says †ar il-ba¨ir, ¨the camel flew’. Song 28: Abu Îalg il-Jayif
The One Who Has a Stinking Mouth
Baghdad, Îajj 2:162 When children see a camel pass by in their neighbourhood they follow it and sing. Abu Îalg il-jayif Ma shayif
[O] the one that has a stinking mouth He does not see [us]
Song 29: Ya Jamal Îukah
O Arrogant Camel
Basra, D: 153 Spotting a camel carrying bags of salt for sale in the alleys, children would become excited and chant these verses several times to tease the camel: Ya jamal Ìukah! Bazzatak shokah Min wara-l-†ofah
O arrogant camel! A thorn pricked you From behind the dividing wall
Song 30: Jamusah ¨Ayniç Îamrah
Buffalo, Your Eye is Red
Baghdad, Q 2:163 Many Iraqi children in the Upper Euphrates and Tigris areas are fond of cats, cows, and camels. When children see buffaloes in the river or when the milk vendor (usually a female) brings her cows to 29
Kalha, the initial hamzah of akal is dropped, akalha > kalha, ‘he ate it’. 550
THE FOLK SONGS OF IRAQI CHILDREN
the customers’ neighbourhood and milks them there, the children would watch her and sing: Jamusah ¨ayniç Ìamrah Bil-layl tij∂a¨ jamrah 30
Cow, your eye is red At night it sparks [like] a piece of coal
Song 31: Abu-sh-Shuwadi
The Owner of the Monkeys
Baghdad, Îajj 2:165–6 Children in Baghdad enjoy the performance of the monkey obeying the commands of its master, which include somersaults, imitating the walking of a drunkard, the way a bride and an old woman sleep, etc. When the performance is over and the owner takes the monkey away, the children follow them, clapping and chanting: Shadi warak il-Ìayyah Urgus wan†ik shahiyyah 31
Monkey, the snake is behind you Dance then I’ll give you shahiyyah
Song 32: Abu Jinnayb Shidd il-Khalalah Hold the Green Date Tightly, O Crab! Basra, D: 57
Iraqis call the crab, abu jinnayb, ¨the one that walks on one side’. When some Iraqi children see a crab holding a small piece of date, they gather around it and sing. They believe that the crab is an arrogant creature; hence they demean it in the song: Shidd il-khalalah bu Jinnayb! Shidd il-khalalah! Jat-il-ghazalah bu Jinnayb! Jat-il-ghazalah!
Hold the green date tightly, O crab! Hold the green date tightly! The gazelle has come, O crab! The gazelle has come!
E. Cumulative and Miscellaneous Songs (Songs 1–65) E. 1. Cumulative songs This part includes a few cumulative verses and many miscellaneous songs about children themselves, people in general, as well as about friends and relatives. Cumulative songs enumerate from the general to the particular and from cause to effect, and give pleasure by noting 30
Tijda¨ < tiqdaÌ, where /q / > /j/ but also /q/ > /g/ > /j/: /qa¨id/ ga¨id > ja¨id, ‘he is sitting’, see ¨Aziz al-Wahhab’s short story, ‘Khiyanat al-¨Uhud’, al-Turath al-Sha¨bi, Baghdad, Vol. 5, Year 6, 1975, 160. 31 Shahiyah, a Persian copper coin of a small value, Per. Shahi. 551
THE FOLK SONGS OF IRAQI CHILDREN
the relativity of all things in nature and their powers to affect each other. There are only a few children’s cumulative songs that are still current in Iraq. The first printed versions of such songs with an English translation are those included in Ethel Stevens’ Folk Tales of Iraq.32 Some variants of these songs continue to exist until the present time. It should be mentioned that such rhymes, and particularly the ones involving water, fire, and sticks, are found around the world. They are so popular in Iraq that portions of these songs are found in other Iraqi folk songs. Iraqi children’s folk songs have two qualities in common: a swinging beat and repetition either of words or sounds. The song Sidi Sidi below reminds us of a late sixteenth-century Aramaic variant, the Îad Gadya (‘the only kid’), which is included in the Jewish Passover liturgy,33 and begins ‘A kid, a kid! My father bought a kid for two zuzim’. This falls into a familiar pattern of the cat that worried the kid; the dog that bit the cat; the staff, the fire and water, and finally, the holy one who destroyed the Angel of Death found in The Old Woman and her Pig, and Mother Goose. It also reminds us of the Iraqi folk tale, Îikayat al-¨anuz wa}l-¨ajuz,34 or ‘the Story of the Goat and the Old woman’ which has a strong likeness to the story in Mother Goose. The variants of the song Sidi Sidi given below differ from each other mainly in the employment of a number of lexical items that are used according to the dialects of the geographical location of the song and the initial verses.35 Such initial verses may be found also in a few other parts of Iraq.36 Song 1: Sidi Sidi
My Grandpa, My Grandpa
Baghdad, Q 2:197–8 Sidi sidi raÌ-il-makkah RaÌ-il-makkah Jabli thob u-ka¨kah Thob u-ka¨kah
My grandpa, my grandpa went to Mecca He went to Mecca He brought me a dress and a ball of thread A dress and a ball of thread
32
Stevens, 1931: 95–7. See Mansour, 1951: 74–6. 34 Stevens, 1931: 7–9. 35 Such initial verses are, ya jdidah nudi nudi, ‘O new nod nod } in Samarra’ (Q 2:293), ya khishbah nudi nudi, ‘O little piece of wood nod nod } in Karbala} and Najaf, (Q 2:295), and ya nakhlah nudi nudi, in the upper Euphrates (Q 2:298) ‘O palm tree’ nod nod, (Q2: 290). The version in Stevens (1931: 95–6) is almost identical to the one here, except that there are also alternative versions of several verses. This song is so popular that it is cited in several other sources. 36 See for example, al-Dulayshi, 1968: 128–9 and al-Samarra}i, 1975: 244. 33
552
THE FOLK SONGS OF IRAQI CHILDREN
Wulka¨kah wayn a∂ummah? Wayn a∂ummah? A∂umma-b-Ònaydgi 37 Ib-Ònaydgi ∑naydgi ma-lah miftaÌ Mah-la miftaÌ Wul-miftaÌ ¨ind-il-Ìaddad 38 ¨Ind il-Ìaddad Wul-Ìaddad yrid flus Yrid flus Wul-flus ¨ind-il-¨arus ¨Ind-il-¨arus Wul-¨arus trid qaqah 39 Trid qaqah Wul-qaqah [ib]ça¨b-ij-jajah 40 [Ib]ça¨b-ij-jajah Wujjajah trid †u¨um41 Trid †u¨um Wul-†u¨um bil-barriyyah Bil-barriyyah Wul-barriyyah trid ma†ar Wul-ma†ar ¨ind a¬¬ah La ¨ilah illa a¬¬ah
The ball of thread, where [shall] I hide it? Where [shall] I hide it? I’ll hide it in my small chest In my small chest My chest has no key Has no key The key is with the blacksmith With the blacksmith And the blacksmith wants money Wants money And the money is with the bride With the bride And the bride wants an egg Wants an egg And the egg is in the bottom of the hen In the bottom of the hen And the hen wants grain Wants grain And the grain is in the desert In the desert And the desert wants water And the rain is with God There is no God but Allah
Song 2 The following version comes from Mosul (Q 2:293–5) Îaddath abuyi jiddi Bis†uÌ bayt il-Ìajji Wul-Ìajji yghid lo bay∂i Wul-bay∂i ibba†in il-jiji
My daddy talked to my grandpa In the house of the Îajji The Îajji wants an egg But the egg is inside the hen
37
Snaydgi, is dim. of sundugi, ‘my chest’. In another version, from Samarra}, this line reads, ‘and the key is with the carpenter’. This may indicate that this version is the oldest version, because making wooden keys for chests and doors in many parts of Iraq was a practice earlier than that of metal keys, see al-Musawi, 1965, VIII–X: 21; al-Samarra}i, 1975: 244 and Îasan, 1966, I: 67–8. Îusayn also says (p. 67) that children in Baghdad end several of their songs with: Ya Fa†mah bint in-nabi O Fatmah the Prophet’s daughter Ukhdhi kitabiç wunzili Take your Book and come down ¨Ala gabur MuÌammad u-¨Ali To the tomb of Muhammad and ¨Ali 39 Qaqa, ‘egg’ in children speech. 40 Only in a few words does /j/ act as a sun letter, and hence we have ij-jajah, ‘the hen’ instead of al-dijajah. 41 ™u¨um, lit. ‘bait’, used here because grains are used as bait to catch birds. 38
553
THE FOLK SONGS OF IRAQI CHILDREN
Wul-jiji tghid lah luqu† Wul-luqu† bil-¨aliyyi Wul-¨aliyyi tghid lah miftaÌ Wul-miftaÌ ¨ind il-Ìiddad Wul-Ìiddad yghid flus Wul-flus ¨ind il-¨ghoÒ Wul-¨ghoÒ bil-Ìimmam Wul-Ìimmam yghid lo qindil Wul-qindil †amis bi-l-bigh Wul-bigh yghid lo Ìabil Wul-Ìabil bÌalq il-jamus Wul-jamus yghid lo Ìshish Wul-Ìshish foq il-jabal Wul-jabal yghid lo ma†agh Wul-ma†agh ¨ind a¬¬ah Tobah wastaghfir a¬¬ah
And the hen wants grains But the grains are in the upper room And the upper room needs a key But the key is with the blacksmith And the blacksmith wants money But the money is with the bride And the bride is in the bath But the bath needs a lamp And the lamp sank in the well And the well needs a rope But the rope is in the mouth of the cow And the cow needs grass But the grass is on the top of the mountain And the mountain needs rain But the rain is with God I ask God’s repentance and forgiveness
Song 3: Dur! Dur! Ya Na ¨ur Go Around! O Water Wheel! ¨Anah, Q 3:346 Children chant this song and clap when strolling or sitting. Dur Dur Ya Na¨ur! Îu†† il-bay∂ah bit-tannur 42 Wut-tannur yrid Ìa†ab Wul-Ìa†ab ¨id il-fallaÌ Wul-fallaÌ yrid il-mayy Wul-mayy ¨id-in-nahar Wun-nahar yrid na¨ur Dur dur ya na¨ur!
Go around! O water wheel! Put the egg in the clay oven And the oven needs firewood And the firewood is with the farmer And the farmer wants water And the water is in the river And the river wants a wheel Turn around, turn around! O wheel!
Song 4: Mgayzli Farraytah43 I Spinned My Spindler Following are two variants, the first one comes from Samarra’ (Q 1:177), and the second is from Baghdad (Q 1:183). Children sing this song on sunny winter days outside their homes. ¨Imgayzli 44 farraytah TaÌt-ish-shijar ∂ammaytah Ja khali }uma-n†aytah In†ayta-l[il]-faÌal wun-nagah
I spinned my spindler Beneath the trees I hid it My uncle came, I didn’t give it to him I gave it to the camel and the she-camel
42
Tannur, Arm., ‘house of light’. See a variant in Stevens, 1931: 195–6 and al-Dulayshi, 1968: 128–9. The variant in Samarra} has several verses not mentioned elsewhere, see al-Samarra}i, 1965: 62. 44 Mgayzil, dim. of magzal, ‘spindle’. 43
554
THE FOLK SONGS OF IRAQI CHILDREN
Ya¨Òayyatna-t-taffagah! Tingatak ¨ala barbinah Barbinah sharji u-gharbi WumÌazmah bil-wardi Wul-wardi ma yug∂ubha Illa-b-jidhayil khali Khali abu baÌriyyah U-¨amamtah shadhriyyah
O, our beating stick! Your jar is on a Barban date tree A date-tree Eastward and Westward It is tied up with flowers And the flowers can’t hold it Except with the hair lock of my uncle My uncle the sailor His turban is bluish
Song 5: Bidi Imghayzil
In My Hand is a Spindle
Bidi imghayzil farraytah45
Ya ÌlayÌil yiÌazzimah Wul-gunbarah tiwaÒwiÒ Wul-b¨ayyir 48 }iy¨anfuÒ
In my hand is a spindle I spinned it Beneath the trees I put it Between the camel and the she-camel O our beating stick! The rifle of my father, ÎlayÌil O ÎlayÌil, O the lion of the desert A quarrel broke out on the other side of the river O Hlayhil he will settle it (lit. tie it) And the lark will chirp And the little camel will kick
Song 6: Rikk il-Markab
Get the Ship Ready!
TaÌt-ish-shijar Ìa††aytah Bayn-il-jamal wun-nagah Ya-¨Òayytna-t-taffagah! Tufgat46 abuyah ÌlayÌil Ya ÌlayÌil 47 ya sab¨il-barr ∑arat ¨arkah ibdhak iÒ-Òob
Mosul, Q 3:527 Two boys sit back-to-back while their hands are intertwined in the shape of a ring. One of the boys raises one hand, bends his back, lifts the other boy and raises the latter’s feet up. The carrier sings the first line and the other boy sings the next line, then they change roles. Rikk il-markab! Ma-rikku Dada49 Rikku! Ma-rikku Dandil buzak! 50 Daqussu 45 46 47 48 49 50
Get the ship ready! I won’t get it ready Brother, get it ready! I won’t get it ready Lower your cheek! So that I cut it
Farraytah admits to two meanings, ‘I spinned it’ and ‘I threw it’. Tufgah, Tur. tufak. ÎlayÌil, dim. of ÌalÌul, ‘weak, ‘loose’. B¨ayyir, dim. b¨ir, ‘camel’. The verb ¨anfaÒ, ‘to kick’ is usually used for a donkey. Dadah (f.), dad (m.), a word used for asking help, Per., dadash, ‘brother’. Buzak, Per. ‘animal mouth’. 555
THE FOLK SONGS OF IRAQI CHILDREN
E. 2. Miscellaneous Songs Song 7: Ana Khalalah
I Am an Unripe Date
Samawah, Q 3:301 Ana Ana Ana Ana Ana Ana
khalalah baydah khalalah Òafrah khalalah Ìamrah khalalah zargah khalalah jozi khalalah wardi
Song 8: WaÌid Rummani
I am I am I am I am I am I am
an an an an an an
unripe unripe unripe unripe unripe unripe
white date yellow date red date blue date walnut-like date pink date
One Pomegranate
¨Amarah, Q 3:370 This song may be considered a number game song. The boys sing only the refrain and the leader sings the rest of the song. Hay Ithnayn rummani Hay Tlathah51 rummani Hay Arba¨ah rummani Hay Khamsah rummani Hay Sittah rummani Hay Sab¨ah rummani Hay Thmanyah rummani Hay Tis¨ah rummani Hay ¨Ashrah Rummani Hay Il-khamis li¨bitna Wul-jum¨ah ¨u†latna
Hay Two pomegrantes Hay Three pomegranates Hay Four pomegranates52 Hay Five pomegranates Hay Six pomegranates Hay Seven pomegranates Hay Eight pomegranates Hay Nine pomegranates Hay Ten pomegranates Hay Thursday is our play time And Friday is our day off
51
Tlathah, a few speakers of IA pronounce /th/ as /t/, especially in numbers 3, 30, 33, etc. The endings /i/ in rummani is to create a rhyme. 52 This verse and the following one are mentioned in the song Sabit Sabbut (part E: 67). For a variant in Basra, see al-Dulayshi, 1968: 123–4. 556
THE FOLK SONGS OF IRAQI CHILDREN
Song 9: Haylah Ya Rummanah!
Come on, O Pomegranate!
Baghdad, Q 1:323 53 A wife is angry at her husband, and her father attempts to reconcile his daughter with his son-in-law. A group of girls stand in a ring holding hands. Another girl, the leader, stands in the centre of the ring and another girl, who plays the role of the angry girl, stands outside the ring facing the backs of the girls in the ring. The leader sings the first verse and every other verse and the group sings the refrain when the girls walk in a ring and dance. The name of every girl in the ring is called one at a time. When this happens she turns around while her back is facing the players. The girl outside the ring replaces the one in the middle after all the names of the girls are called. The rummanah stands here for a beautiful girl. Haylah54 ya rummanah! Haylah yummah! Min hiyya-z-za¨lanah? Haylah yummah! Sa¨diyah za¨lanah Haylah yummah! Minhu-l-yirra∂iha? Haylah yummah! Abuha yirra∂iha Haylah yummah! ∑ayig tiraçiha Haylah yummah!
Come on, O pomegranate! Come on, my dear daughter! Who is the angry girl? Come on, my dear daughter! Sa¨diyah is angry Come on, my dear daughter! Who will reconcile her? Come on, my dear daughter! Her father will reconcile her Come on my dear daughter! He has made her jewellery Come on dear!
Song 10: Ya Baydhinjanah! O Eggplant! Ramadi, Q 3:408 This song is similar to Haylah Ya Rummanah (song E. 9) in form, content, and rhythm. The eggplant stands for a pretty girl in Iraqi lore. Girls form a ring and hold hands. An angry girl stands outside the ring singing the main lines of the song, while the others sing the refrain, Shibak libak, creating a rhythm. The angry girl turns in a way 53 In Basra this game song is played differently than in ¨Anah where the refrain is wayli, ‘Alas to me’ instead of hayla, see al-¨Ani 1975, IV: 50. 54 Haylah < hayya+li- (CA), ‘let’s!’ In an abridged song from Karbala}, the word wayla ‘alas!’ is used instead of hayla. See al-™u¨mah, 1976, 2–3: 202, and al-¨Ani 1975, IV: 47–53. In Iraqi lore, the pomegranate as well as the eggplant are epithets for a beautiful girl.
557
THE FOLK SONGS OF IRAQI CHILDREN
that her back faces the other players. The song alludes to the social phenomenon in which the husband abuses his wife and she leaves home and goes to her parents. The two words shibak libak are derived from the IA verbs libak, ‘to mix’ and shibak, ‘to interjoin’ respectively. Ya Baydhinjanah! 55 Ya baytinjanah! Ho Ho Min hiyya wayyana? Ho ho Fa†am56 wayyanah Ho ho ¨Arusna bil-kallah
Hassah yiji wa-gullah
O eggplant! O eggplant! Ho ho Who is with us? Ho ho Fa†am is with us Ho ho Our bride is [sleeping] under the mosquito net Soon he [the groom] will come and I’ll tell him Our bride is [sleeping] under the mosquito net Soon he [the groom] will come and I’ll tell him Our bride is [sleeping] under the mosquito net Soon he will come and I’ll tell him
Song 11: Bushra Çannatna
Bushra, Our Daughter-in-Law
Hassah yiji wa-gullah ¨Arusna bil-kallah Hassah yiji wa-gullah ¨Arusna bil-kallah
Ramadi, Q 3:310 This song is about an old man who plans to wed a girl by the name of Bushra. Girls stand in a circle and hold hands, singing and dancing clockwise: Bushra çannatna Tugsul brab¨atna Wush-shaykh rayidha ∑ayigh glayidha Digg il-miliÌ57 Wul-miliÌ faÒulah58
Our daughter-in-law Bushra She does her washings in our company The old man wants [to marry] her He has already ordered/made her necklaces [So on so], pound the salt! [but] the salt turned out to be a bean
55
Baydhinjanah: this word is also pronounced betinjanah, bedinjanah, and baydhinjanah from the Persian badinjanah, which means ‘demon eggs’ (testicles). See Shir, 1980: 15. 56 Fa†am, dim. of Fa†imah. 57 The grinding of the salt is an indication of a forthcoming banquet or a wedding celebration. 58 FaÒulah, ‘bean’ (lit.), implying that the celebration did not take place. 558
THE FOLK SONGS OF IRAQI CHILDREN
Song 12: Shushliyyah Rummanah
Shushliyyah is a Pomegranate
Karbala}, Q 3:537 Similar to Haylah Ya Rummanah (song part E: 9). Girls hold hands and stand in a ring, turning clockwise when singing. The song is about a certain beautiful girl by the name Shushliyyah. Group: Shushliyyah Girl: Ya rummanah! – ™abbat li-s-sug – il-kuÌlah tlug – it-†og i†ugg – ¨A†-†irciyyah 59 – Shushliyyah – Ya rummanah!
Shushliyyah, O pomegranate! She entered the market The khul suits [her] The bracelet clinks On the earrings [Of] Shushliyyah O pomegranate!
Song 13: Hdaylah Tgul
Hdaylah Says
¨Amarah, Q 3:299 A group of girls stand in a ring singing. One girl stands in the centre of the ring and sings the refrain Hdaylah, ‘O Hdaylah!’ The other girls sing the rest of the verses. Hdaylah tgul, Hdaylah tgul Hdaylah ¨Abd il-Ìamid }imwaÒÒi Hdaylah Kulman Hdaylah Yumma giraÒni–z-zanbur Hdaylah Aswad brasah Ìanfus Hdaylah Wa¬¬ah laruh il-¨Abbas 60 Hdaylah Wazur yammah wagh¨ud Hdaylah
Hdaylah says, Hdaylah says Hdaylah ¨Abd il-Hamid has ordered [her jewellery] Hdaylah Everyone [marries her fiance] Hdaylah Mom, the wasp stung me Hdaylah Black, in its head there was a sharp needle Hdaylah I swear by God I’ll go to ¨Abbas’s [tomb] Hdaylah And visit with him and sit [by it] Hdaylah
59 ¨al-tirçiyyah >¨attirçiyah, ¨al, is a contraction of ¨ala, ‘on’+il. The /l/ of the definite article is assimilated with the sun letter /t/ of tirçiyyah. 60 Laruh lil-¨abbas, ‘I’ll go to al-¨Abbas, where li+il >lil.
559
THE FOLK SONGS OF IRAQI CHILDREN
Song 14: Bam Bam, Bam!
Bam Bam, Bam!
Basra, Q 1:439 Boys mostly perform this song. One boy hides behind a palm tree. The other boys sing, but they wait to sing the last line when they find him. Bam Bam, Bam! ™ab¨ is-sayyid 61 sharradah WaÌid minhum62 ¨abd il-khan Yirmi-ç-çaylah ¨la-d-dukhkhan ∑ab ÌlayÌil biç-çir¨an Khallah yinkaz bil-midan Jatti-l-ba††ah tsolif 63 lah Ja il-wawi ydaÌlib lah DiÌlib diÌlib diÌlibah!
Bam Bam, Bam! The nature of the sayyid is running away One of them is ¨Abd il-Khan He shoots the bullet at the smoke He hit ÎleÌil in [his] upper and lower parts He caused him to jump in pain in the square The duck came chatting with him The jackal came bending his back to him Milk [it], milk [it], milk [it]!
Song 15: Ya Baba!
O Daddy!
Mosul, Q 3: 396–7 Played by two girls who face each other. One plays the father and the other plays his daughter. When they begin to sing, the right hand of the father claps the left hand of the daughter and the left hand of the daughter claps the right hand of the father until they finish singing the song. D: Ya baba wayÒab rayiÌ?64 WayÒab rayiÌ? F: Ya binti lij-jazirah Lij-jazirah D: Ya baba (i)sh tjibinnah?65 (i)sh tjibinnah? F: Ya binti ward-il-Ìennah Ward il-Ìennah D: Ya baba ashni¨mil biha? Ashni¨mil biha? F: Ya binti nil¨ab biha Nil¨ab biha
O daddy, where are you going? Where are you going? O my daughter, to the island To the island O daddy, what will you bring us? What will you bring us? O daughter, hennah blossoms Henna blossoms O daddy! What will we do with them? What will we do with them? O my daughter, we’ll play with them We’ll play with them
61
Sayyid is a descendant of the Prophet Muhammad’s family. Minhum, refers to those who run away. 63 Tsolif, a quadrilateral verb derived from the noun salufah, ‘tale’. 64 WaiÒob, lit. ‘which direction/side’? 65 Ishtjibinna is a contraction of ish+tjib+ilna, ‘what things’+ ‘you’ll bring’+ ‘to us’, where /l/, ‘to’ is assimilated to the suffix /na/, ‘us’. 62
560
THE FOLK SONGS OF IRAQI CHILDREN
Song 16: ∑ubbu-l-¨Adas! 66
Pour out the Lentil!
Mosul, Q 1:299; D: 134–5 Holding hands in a ring, one girl faces another girl standing in the centre of the ring, dancing and singing. The girls sing the refrain way way and the girl in the centre sings the rest. This song contains details mentioned in the song Haylah Ya Rummanah (part E: 9). ∑ubbu-l-¨adas baÒubbu 67 Way way AÒubbu bil-qawa†i 68 Way way AÒubbu bil-ballu¨ah Way way Yiji ¨ammi wishufah Way way Yuktulni }ib¨aÒatah 69 Way way ¨AÒatah bint il-¨askar Way way Tishrab Ìalib u-tiskar Way way ∑ubbu-l-¨adas baÒubbu
Pour out the lentil! I’m pouring it Way way I’ll pour it into the porcelain jars Way way I’ll pour it into the sewer Way way My uncle will come and see that Way way He’ll beat me with his stick Way way His stick [belongs] to the soldier’s daughter Way way She drinks milk and gets drunk Way way Pour the lentil! I am pouring it
Song 17: Kashk U-¨Adas
Burghul, Wheat and Lentils
Mosul, al-¨Ubaydi: 208–9 Kashk is a mixture of cracked wheat without shells cooked with meat and stems of beet. Sometimes, chickpeas and noodles are added. Girls hold hands while standing in a ring and dancing. One girl stands in the centre and sings the main verses. The rest of the participants sing the refrain. Kashk u¨adas ma-Ìubbu 70 Yoya Yighli ghalway waÒubbu
Burgul, wheat and lentils, I don’t like Yoya It boils once and I empty it
66
See a variant in Q 1:299. Basubbu. The ba in basubbu denoting a progressive action is employed in Eastern Arabic dialects. Its occurence here is probably due to the influence of Syrian and Lebanese teachers who taught in Iraqi schools during the British Mandate in Iraq 1922–32. 68 Qawa†i < bawa†i, ‘jars’. 69 Yuktulni < yuqtulni. In a few words in IA /q/ > /k/. 70 The two words ma and aÌubbu are pronounced as one word, where the hamzah of the first person singular imperfect /a/ of aÌubbu is dropped because it is preceded by a word (ma) which terminates with /a/ too. 67
561
THE FOLK SONGS OF IRAQI CHILDREN
Yoya AÒubbu bil-bawati Yoya Aqiddimu-l-khawati Yoya Khawati ma-ygidunu Yoya An¨al abu abunu Yoya A†ushshu71 bil-billu¨ah Yoya Id jajju72 maq†u¨ah Yoya
Yoya I empty it in the porcelain jars Yoya I serve it to my sisters Yoya My sisters do not want it Yoya Damn It! [lit. ‘may the father of its father’ be cursed] Yoya I’ll pour it in the sewer Yoya The hand of Jurjis is cut Yoya
Song 18: Ya Îajji MÌammad!
O Îajji MÌammad!
¨Anah, Q 3:413 A game song played by girls only and sung while sitting or walking with or without clapping hands. The leader sings the refrain, and the rest of the group sing the verses. Leader: Ya Ìajji MÌammad Group: Yoya ¨I†ini ÌÒanak Yoya Ashiddah warkab Yoya Sakarkab mat Yoya Khallal-banat Yoya Banat usud Yoya Mithlil-qurud Yoya
O Îajji MÌammad Yoya Give me your horse Yoya I’ll saddle it and ride it Yoya The chopped ear horse died Yoya He left [his] daughters behind Yoya Lion-like daughters Yoya Like the monkeys Yoya
Song 19: ¨Ali Ya ¨Ulu
¨Ali Ya ¨Ulu
Hit, Q 3:475 Girls hold hands, sing and dance in a ring. A girl who sings the refrain is either inside or outside the ring. 71 72
A†ushshu, lit. ‘I will scatter it’. Jajju is dim. of the Christian name Jurjis. 562
THE FOLK SONGS OF IRAQI CHILDREN
¨Ali ya ¨ulu! RiÌna tanÌish 73 ¨Ali Lganah darwish 74 ¨Ali
¨Ali, O ¨Ali! We went to gather [grass] ¨Ali A dervish ran into us ¨Ali
Song 20: id-Dishshah
The Dishshah Dance
¨Amarah, Q 3:276 Standing in a ring, the girls hold hands. One girl from the ring stands either in the ring or in its centre and sings all the verses except the refrain ‘ya haza{’ and the other girls sing the rest. The children probably learned this song from adults. Halli ya dam¨at ¨ayni! Ya Haza¨ Kull dimi¨tayn ibghazi 75 Ya Haza¨ Nidhirna intuni shogi 76 Ya Haza¨ Bidi ladug brazi Ya Haza¨
O tear[s] of my eye, flow! Ya Haza¨ Every two drops of tears worth two ghazis Ya Haza¨ I took a vow that [if] you’ll fulfill my desire Ya Haza¨ With my hand I’ll dig a hole Ya Haza¨
Song 21: Sodah Bay∂ah
Black [or] White
Karbala}, Q 3:472 Originally, the mother sang this song to her daughter while playing with her. Girls chant this song while standing in a ring. One girl stands in the middle of the ring and sings the main verses and the other girls repeat the refrain after each verse that the girl in the centre sings. Leader: Sodah bay∂ah Group: Sodah Gha¨dah bil-ma†bakh Sodah Takul u-tu†bukh Sodah
Black White [girl] Black Is sitting in the kitchen Black Eating and cooking Black
73 Tanhish. About the use of ta before the imperfect denoting a progressive action in the present, see part E below, song 40, n.108. 74 Darwish (darpish Per.), lit. ‘threshhold’, an epithet for a beggar. 75 Ghazi, a currency used when Iraq became independent in 1931. 76 ‘We’ refers to ‘I’. intuni shogi, lit. ‘give me my desire!’
563
THE FOLK SONGS OF IRAQI CHILDREN
Tu†bukh ¨aÒidah 77 Sodah Tishrab çigarah Sodah Hiyyah wush-shayib Sodah Hiyyah tridah Sodah
Cooking ¨asidah Black Smoking a cigarette Black She and the old man Black She wants him Black
Song 22: Ya Yummah In†ini Khrayzah! 78 O Mum Give me a Bead! Baghdad, Q 1:245 Played by girls only. Two girls hold hands and sing while dancing around in a circle. One girl sings the refrain, the other the rest. Girl 1: Ya Yummah in†ini Kharzayah! Girl 2: Hay Hay! Ya yummah shiliç biha? Hay Hay! Ya yummah al¨ab biha! Hay Hay! Ya yummah inti waya man? Hay Hay! Ya yummah ani-w-¨ammi Hay Hay! Ya yummah ¨ammiç shismah? Hay Hay! Ya yummah ¨ammi ¨alwan! Hay Hay! Ya yummah rayiÌ sa¨id lil-khan? Hay Hay! Ya yummah dhabli †arbush Hay Hay! Ya yummah aÌmar mangush Hay Hay!
O, Mum Give me a Bead! Hay Hay! O, dear daughter, what will you do with it? Hay Hay! O mum, I’ll play with it Hay Hay! O dear daughter, with whom? Hay Hay! O mum, my uncle and I Hay Hay! O dear daughter, what’s the name of your uncle? Hay Hay! O mum, my uncle ¨Alwan! Hay Hay! O dear daughter, is Sa¨id going to the warehouse? Hay Hay! O mum, he threw a turban to me Hay Hay! O mum, it’s decorated with red Hay Hay!
77
¨AÒidah, a thick porridge made of flour, oil or butter and sugar. In Sa¨d al-Din (1988: 25–6) the song does not have a refrain and in the end of it there are verses from different songs. 78
564
THE FOLK SONGS OF IRAQI CHILDREN
Song 23: Aman Ya Îennah! Alas Hennah! Mosul, Q 3:264 Most Iraqi women prefer the henna from Basra because of its good quality. In the past, when efficient means of transportation was not available, goods took a long time to go from Basra in the south to Mosul in the north. Some Iraqi women used the hennah for treating diarrhoea and dysentery by boiling it in water and drinking some of the boiled water.79 Forming a ring and holding hands, the girls walk around clockwise. One of the girls sings the refrain and the others the rest of the song. Aman80 is a Turkish word for ‘complaint’ and expresses boredom, or a plea for help. Very rarely, it is employed in IA as an expression of admiration as in ‘aman shlon walad hadha!’,81 ‘How wonderful this boy is!’ One characteristic feature of the Iraqi maqam (Arabic classical song) is the inclusion of conventional words and phrases in the song. These have no logical connection with the song itself. The word Aman is one of these words, which is sung in particular moments of the performance, especially in the beginning of the maqam. Maqamat (sing. maqam) are best defined and understood in the context of the rich Arabic music repertoire. The nearest equivalent in Western classical music would be a mode, e.g., Major, Minor, etc. Aman ya Ìennah! Aman! Îennat il-baÒrah Aman! ¨Alayna Ìasrah Aman! Wul-yom ¨idi La, la, la Fistan jdidi La, la, la Fistani kashkash La, la, la Bil-warid mla††ash La, la, la Fistani Mshaqqaq Hi, hi, hi 79 80 81
Alas O henna! Alas! The henna of Basra Alas! We miss it Alas! Today is my feast La, la, la A new dress La, la, la My dress is pleated La, la, la Spotted with flowers La, la, la My dress is torn into pieces Ya, ya, ya
Al-Îajjiyhah, 1986, V: 241. Kojaman, 2001: 161. Woodhead and Beene, 1967: 17. 565
THE FOLK SONGS OF IRAQI CHILDREN
Song 24: Ya Swaydani! 82
O Swaidani!
¨Amarah, Q 3 3:431 A group of girls sit forming a circle. One girl stands in the centre singing the main lines and the rest clap and sing the refrain. The use of Ìayl, Ìayl, Ìayl is for producing a rhythm. Girl: Îayl Ìayl Ìayl ya¬¬ah! Group: Ya swaydani! ¨Idna walad ¨Abdallah Ya Swaydani! Yi¨thar wa-gullah Òmallah 83 Ya Swaydani! Bint il-¨abid basuha Ya Swaydani! Wallah imnaggil labuha Ya Swaydani! Îayl Ìayl Ìayl yallah! Ya Swaydani!
Come on, come on, come on! Let’s go! O, Swaidani! We have a boy, Abdallah O, Swaidani! He stumbles and I say to him ‘may the Name of God protect you’ O, Swaidani! They kissed the daughter of the slave O, Swaidani! By God I’ll tell her father O, Swaidani! Come on, come on, come on, let’s go! O, Swaidani!
Song 25: Bibi Wayn Ummi?
Grandma, Where is Mum?
Basra, Q 1:257 Sung by two girls. One girl acts like a grandmother and another acts like her granddaughter. A reference is made in the song to the custom of visiting the tomb of a saint following a child’s birth. Bibi is ‘grandmother’ or ‘nanny’ used here for endearment. The grandmother may call her grandchild bibi or yummah. Girl 1: Bibi wayn ummi? Girl 2: Bibi bil-Ìsayn 1. Bibi jabat walad? 2. Bibi ¨AbdiÒ-∑amad 1. Bibi khubuz Ìar? 2. Bibi lil-khu††ar
Grandma, where is my mum? Granddaughter, she is [visiting the tomb] of Hussein Grandma, did she give birth to a boy? Granddaughter, [his name] is ¨Abd is∑amad Grandma [is there] hot bread? Granddaughter, [it’s] for the guests
82
See a variant in Samawah in Q 3: 430. Isma¬¬ah, lit. ‘May the name of Allah’ [be upon you’], i.e. ‘May Good guard you!’ This expression is also said when a child falls down or when telling an interlocutor about someone who died, see al-Îanafi, 1978, I: 180. 83
566
THE FOLK SONGS OF IRAQI CHILDREN
1. Bibi }u-çiklet? 2. Bibi lil-khu††ar 1. Bibi khubuz barid? 2. Bibi lil-¨yal 1. Bibi wayn ummi? 2. Bibi bil-Ìsayn 1. Bibi jabat walad? 2. Bibi ¨AbdiÒ-∑amad
Grandma, and chocolate? Granddaughter, it’s for the guests Grandma, [is there] cold bread? Granddaughter, it’s for the children Grandma, where is my mum? Granddaughter, she is [visiting the tomb] of Hussein Grandma, did she give birth to a boy? Granddaughter, [his name] is ¨Abdis∑amad
Song 26: Ya Yumma Δa ¨ MÌaybsi 84
O Mum, My Ring Got Lost
Baghdad, Q 1:265 One girl sits on the ground and digs in it pretending she is looking for her lost ring, but in fact she is a wolf ready to attack the cattle (girls), who stand away from her singing: Wolf: Ya yumma ∂a¨ mÌaybsi 85 Group: Kullah çidhib Wolf: Mal¨un †ash†ash khirzi Group: Kullah çidhib Wolf: WashÌaddah86 yakul wladi Group: Kullah çidhib
O mum, my ring got lost All of this is a lie Mal¨un scattered my beads All of this is a lie I warn anyone who eats my boys All of this is a lie
Song 27: Wardah A∂Ìiyyah
Adhiyyah is Coming
Dhi Qar, Q 3:373 One girl stands inside (or outside) a ring of girls who dance clockwise and sing the main verses of the song while the one in the centre or outside it sings the refrain. Wardah A∂Ìiyyah ¨Ala-l-mayy Shlon nomathah? ¨Ala-l-mayy Hiçi nomithah? ¨Ala-l-mayy Shlon ga¨datha? ¨Ala-l-mayy Hiçhi ga¨datha 84 85 86
Adhiyyah is coming On the water How does she sleep? On the water She sleeps like this On the water How does she sit? On the water She sits like this
About this song in Basra, see al-Dulayshi, 1968: 24–5. MÌaybis, dim. maÌbas, ‘ring’. AshÌaddah is a challenging and warning word. 567
THE FOLK SONGS OF IRAQI CHILDREN
¨Ala-l-mayy Shlon ¨ajnatha? ¨Ala-l-mayy Hiçhi ¨ajnatha ¨Ala-l-mayy Shlon khubzatha? ¨Ala-l-mayy Hiçi Khubzatha ¨Ala-l-mayy
On the water How does she knead [the dough]? On the water She kneads like this On the water How does she bake bread? On the water She bakes bread like this On the water
Song 28: Haylah u-Halli
Come on and Look
NaÒiriyyah, Q 3:362
This song praises the generosity of the villagers in NaÒiriyyah. Girls sing while holding hands and moving slowly in a circle. One girl stands either in the centre or outside the circle. The group sings the refrain and the girl in the centre or outside it sings the rest of the verses. Haylah u-halli! Haylah Tawwiç 87 †ila¨ti Haylah Min [bayt] abuçi Haylah Salem akhuçi Haylah Îakkak jidrah Haylah Yishabbi¨ il-¨aylah Haylah
Come on and look! Hailah You have just left Hailah From your daddy [’s home] Hailah Your brother Salem Hailah He rubs his pot[s] Hailah He feeds the family until they are full Hailah
Song 29: Ana Umm Zluf
I Have Sideburns
Samawa, Q 3:305 Two girls join hands and move around clockwise according to the rhythm of the song. One of the two girls sings the refrain, hi ba¬¬ah (lit., ‘O by God’), which denotes her disbelief in what has been said. The other girl sings the refrain.
87
Tawwiç and tanniç, ‘you (f.) just’, the word ‘tawwa’, ‘just’ or ‘not long
ago’. 568
THE FOLK SONGS OF IRAQI CHILDREN
Girl: Ana umm zluf 88 Group: Hi ba¬¬ah Çannat ir-rwayyis89 Hi ba¬¬ah Abuçin lagra¨ Hi ba¬¬ah Îi††inlah riÌah! Hi ba¬¬ah ¨Aynah wakiÌah90 Hi††inlah Òandal Hi ba¬¬ah LiÌiytah tundil Hi ba¬¬ah Jaytah aguÒÒha Hi ba¬¬ah Ingi∂am nuÒÒha
I have sideburns Yea The chief’s daughter-in-law Yea Your father, the bald Yea He put perfume! Yea He has leering eyes [on women] He put perfume Yea His beard is dangling Yea I came to him to cut it Yea Half of it was cut
Song 30: Ish-shigah Ligah He Who Works Hard Gets What He Wants MaÌawil, Q 3:40 Girls hold hands forming a ring. They sing while going around. One of the girls, the leader, sings the main lines, and the group sings the refrain. Ya luÌ! Bag i¨rani Ya luÌ! Min giddami Ya luÌ! AdhbaÌlak †ili Ya luÌ! Min ghanam hali 91 Ya luÌ! Yabu-l-¨aybah! 92 Ya luÌ! Îa†ni ibjaybah
O LuÌ! He stole my earrings O LuÌ! From in front of me O LuÌ! I’ll slaughter a lamb for you O LuÌ! From my family’s herd O LuÌ! O pure gold [man]! O LuÌ! He put me in his pocket
88 The name of the song admits two meanings, ‘I have sideburns’ and ‘I wear sideburns’. It is unclear to what word the nun of abucin, ‘your (f.) father’ refers. 89 Rawayyis, dim. of ra}is, ‘chief’. 90 ¨Aynah wakÌa, ‘his eye is cheeky (lit.)’. i.e. ‘he is impudent’. 91 Hali < ahli, ‘my relatives’. 92 Abu il-¨aybah, an epithet for ‘gold’, lit., ‘the owner of gold’.
569
THE FOLK SONGS OF IRAQI CHILDREN
Song 31: Ya Bin ¨Ammi Gin†ar
O My Cousin Gin†ar
¨Amarah, Q 3: 402 Girls form a ring and hold hands. One of the girls sings the main lines and the group sings the refrain. Group: Ya Gin†ar! Girl: Simçah ukalha-l-khi††ar 93 – Yabin ¨ammi! – Ya gin†ar! – Wigulli sha¨riç a†wal – Yabin ¨ammi! – Ya Gin†ar! – Wa-gullah gir†afa-l-far 94 – Yabin ¨ammi – Ya gin†ar!
O Gin†ar! The guest ate a fish O my cousin! O Gin†ar! He tells me your hair is [the] longest O my cousin! O Gin†ar! And I tell him the mouse chewed it O my cousin O Gin†ar!
Song 32: IlÌaquni Ya-wlaydati
Follow Me, My Chicks
Mosul, al-¨Ubaydi, 209 Girls stand behind each other in a line. The first girl is the mother hen. Another player undertakes the role of Khajjaw[i], the wife of a criminal in Mosul during World War I, who used to kill children and sell their meat. Khajjaw[i] tries to kidnap one of the chicks, while the mother defends them. IlÌaquni Ya-wlaydati! 95 WayÒ wayÒ La tinzilun lil-wadi WayÒ wayÒ Latimsikkum khajjaw[i] WayÒ wayÒ Latiqlikum bil-†awi WayÒ wayÒ
Follow me my chicks! Wais wais Don’t go down to the valley Wais wais Lest Khajjawi catches you Wais wais Lest she fries you in the frying pan Wais wais
Song 33: Jina Njarris
We Came to Toll the Bell
Karbala}, Q 3:332 This game song is widely known in the middle and south of Iraq. Two girls stand outside a ring of girls. When one of the two girls 93
Kalha < akalha, ‘he ate it’. Gir†af, ‘to gnaw’ (qar†afa, CA), ‘clip’ is a quadrilateral verb formed by adding the infix /r/ after the first consonant of the root q, †, f, ‘to pick up’ (CA). 95 Wlaydati, dim. of wladi, ‘my children’. 94
570
THE FOLK SONGS OF IRAQI CHILDREN
begins to sing, the second one answers her by mentioning the name of one of the girls in the ring. The latter turns her face in the opposite direction to the other girls. The game ends when all of the girls’ names are called and their faces are turned in the direction of the first one. Girl 1: Jina-njarris jina njarris Girl Girl Girl Girl Girl Girl Girl Girl
2: 1: 2: 1: 2: 1: 2: 1:
Ummak tbi¨ il-faras Saççinitha Ìaddah Tiga††i¨ il-labbadah 96 Tannurha Ìami Yishwi-z-zgar wil-kbar il-laylah ¨urus man? ¨Urus sa¨diyyah Tirdin maziqah lo †abul? 97
We came to toll the bell, we came to toll the bell Your mother sells the mare Her knife is sharp It tears the throat Her clay oven is hot It grills the youngsters and the old Whose wedding is tonight? Sa¨diyyah’s wedding Do you want an oboe or a drum?
If she replies †abul, ¨drum’, then the two sing: ™abul… †umba… †umba
Tambourine…tumba…tumba
If he replies maziqah, then the two sing: Maziqaha…maziqah…maziqah
Miziq…miziq…miziq …
The two girls turn their faces from those playing and sing: ∑allu ¨alla-d-dajaja Ò-Òafrah! Group: A¬¬ah wakbar! Girls 1 and 2: ∑allu ¨alla-d-dajaja il-kha∂rah Group: A¬¬ah wakbar!
Pray for the yellow hen! God is great! Pray for the green hen God is great!
All girls sing: Imbarak ¨ursiç ya Nunah! Imbarak ¨ursuk ya Nunah! A¬¬ah wakbar! 98
May your wedding be blessed, O Nunah! May your wedding be blessed, O Nunah! God is great!
Song 34: ¨Indi klayçayah
I Have a Piece of Cake
Q 3:479 ¨Indi klayçayah Îulwah imÌashshayah
I have a piece of cake Pretty [and] stuffed
96
Labbadah, ‘neck’, but here it means ‘throat’. Maziqah is an arabized word from the English word ‘music’, but here it refers to a kind of wood-wind instrument. 98 A¬¬ah wakbar, is a variant of A¬¬ah akbar, ‘God is great’. 97
571
THE FOLK SONGS OF IRAQI CHILDREN
Wa¬¬ah ma ¨akulha Illa yiji khali Khali abu ba¨iyyah Wu¨mamatha Ìarbiyyah Shalha w-riga¨ha bil-ga} Til¨at ibçis il-Ìintah Hiyya isimha ba††ah
By God I won’t eat it Unless my maternal uncle comes My maternal uncle, the owner of the sheep And his turban is a military one He took it off and tossed it on the ground It appeared in the bag of wheat Its name is battah.
Song 35: Qal il-Pasha
The Pasha Said
Mosul, Q 3: 512 Children participate in this song when their mothers put them on their thighs and rock them singing. Mother: Qal il-pasha duqqu-l-ghizz! Children: Dam u-dam u-dam Mother: Qal il-pasha ¨Alqu-n-nag! Children: War u-war war u-war Mother: Qal il-pasha kulu-l-ghizz! Children: Ham, ham, ham, uham
Song 36: ¨Ammi, Yamçari!
The Pasha said, ‘Crush the rice!’ Dum u-dam u-dam The Pasha said, ‘Light the fire!’ War u-war war u-war The pasha said, ‘eat the rice!’ Ham, ham, ham, u-ham99
Sir, the Donkey Driver!
Basra, Q 2:237 When a group of children see the date vendor approaching with his donkey carrying his merchandise, they warn their friend, shouting, ‘ya wahal’ three times, ‘he is coming,’ and sing: ¨Ammi100 yamçari! Tamrak ¨adhari Sug il-faÌÌamah ¨Askar wuzlamah Bidi-l-ba††ushah Îamra-w mangushah Yahu-n-nagishha? Bintak ya mullah? Yaytah101 a∂ughgha Taggatni bil-khashugah
Sir, the donkey driver! Your dates are fresh [In] the blacksmiths’ market An army and young men In my hand is a date Red and decorated Who decorated it? Is it your daughter, O Mullah? I came to ground it [the date] It hit me with the spoon
99 Ham ham, an expression said to a child when feeding him/her with a spoon. In children’s language it means, ‘food’. 100 ¨Ammi, ‘uncle’ and ‘sir’, when used to address a strange man. 101 Jaytah > yaytah, ‘I came to him’. The change /j/ to /y/ is common in some areas of Basra.
572
THE FOLK SONGS OF IRAQI CHILDREN
Song 37: Jasim Jassumi
Jasim, My Dear Jasim
Karbala}, Q 1:79 The song was originally a nursery rhyme, adapted by children as a gaming song. They clap and sing in a ring-shaped formation. Jasim jassumi102 La takul numi Numiytak wuskha Itwassikh hdumi Jatti Zahhuri Kisrat il-quri Jatti Nihayah Kisrat id-dayah
Jasim, my dear Jasim Don’t eat sweet lemons Your sweet lemon is dirty It makes my clothes dirty Zahhuri came She broke the teapot Nihayah came She broke the doll
Song 38: Sunbaylah is-Sunbaylah103
Sunbaylah is-Sunbaylah
Îajj 4:167–8 The name of this game is a distortion of the Turkish words sen böyle, ‘you are like this’. This song is known in Baghdad as sunbaylah issunbaylah and in the southern villages sumbaylah is-sumbaylah. One boy (called nayim, ‘sleeper’) stands and bends his back placing his hands on his knees. Every player jumps over his back to the other side while putting their hands on the back of the nayim. The jumpers’ legs or thighs should not touch any part of the nayim’s body. If this happens, the jumper trades places with the nayim. The players who jump sing a line while jumping. The first jumper sings the first line, the next one the second line, etc. Sunbaylah is-sunbaylah ¨Ala-n-nabi Òallayna ∑allayna ma Bu†inah Bu†inah il-Ìalwani Îalwani-l-çakçakani Çakçakani 104-l-baqarah Ummha wabuha-l-baqara Il-baqara bizwiyyah Wu¨rug galb in-nayim mistuwiyyah
102
Sunbelah is-sunbelah We prayed for the Prophet We prayed without Butinah Butinah, the candy vendor The candy vendor who pricks The one who pricks the cow The mother and father of the cow The cow is in the corner And the veins of the sleeper’s heart are cooked
Jassumi and Zahhuri are dim. of the proper nouns Jasim and Zahrah respec-
tively. 103 104
al-Îanafi, 1962, I: 20. Çakçak, a quadrilateral verb formed by doubling the two root letters /ç/ and
/q/). 573
THE FOLK SONGS OF IRAQI CHILDREN
Dijajah wu-frukhha Itlawwukhni wa-lawwikha Abu-z-za¨ar fog it-tall Bidah ¨aÒayah u-yiftil
A hen and her chicks She hits me and I hit her The bird Abu-z-Za¨ar is on top of the hill He holds a stick in his hand and goes around
Song 39: Yabu Jnayjil!
O Wearer of Anklets!
This song is a song that the children learn from their mothers. Baghdad, Q 1:51 Yabu jnayjil wum¨arrah105
Diç †ufar ¨al-jiran
O wearer of small anklets and the naked boy Beware of going outside [Lest] the envious woman sees you And her insides churn [But] when the loving woman sees you She’ll seek God’s protection [for you] O wearer of the nose ring, O wearer of the nose ring! A rooster jumped on the neighbours
Song 40: Quri Ya Quri!
O Teapot! Teapot!
Balak ti†la¨ li-barrah Wutshufak il-¨aduwwah106 U-MiÒranha yitlawwah Wutshufak il-Ìabibah Tgum t¨awwidh ba¬¬ah U-yabu ¨ran u-yabu ¨ran!
Îadithah, Anbar, Q 3:514 Bedouin girls sing this song at any time of the year. Quri ya quri! Lidaghni zanburi Lidaghni bil-hamah ™ayyar il-khizmah ∑iÌu abu ¨mamah! Tayiksir il-quri! 107 Quri ya quri!
Teapot! O teapot! My wasp stung me In the head it stung me And caused my nose ring to fly away Call for the man with the turban! So that he breaks the teapot Teapot! O teapot!
105
Jnajil is sing. of jinjil, ‘anklet’. ¨Aduwwah, ‘a female enemy’, but the reference here is to the envious woman who is childless. 107 Taykassir, in the Bedouin dialects of Iraq and Jordan, as well as in the southern parts of Basra ta (instead of da) followed by an imperfect verb denotes a progressive action in the present tense. A folk poet in ¨Anah says: ‘Dangarit ana tabusah, danger wayyaya ÌÒani, ‘[while] I bent down to kiss her, my horse bent down [too]’. See al-¨Ani, 1975, V: 199. In IA, the prefix da is usually prefixed to the imperfect verb to denote a progressive action in the present and to indicate a purpose (‘in order 106
574
THE FOLK SONGS OF IRAQI CHILDREN
Song 41: Ya Iblis!
O Devil!
Baghdad, Îajj 5:104 It was a custom in nineteenth-century Iraq that when children lost an item, their mothers would make a knot in a small piece of fabric or a handkerchief, called ¨ugdat dhayl iblis, ‘the Devil’s tail knot’ believing that this will lead to the discovery of the lost item with the Devil. The children sing this song while searching for the lost item. When the children find the lost item, they sing: Ya Iblis! Îillil-çis! Mumalna Malin-nas
O devil! Untie the pouch! [The coin] is not ours [It] belongs to the people
Song 42: WÒalna lo Ba¨ad?
Have We Arrived Yet?
Samarra}, Q 3:375 This song is similar to the blindfolded ‘hot/cold’ game played in the United States. One boy walks ahead and a second boy walks behind him, lowering his head and chanting the question, ‘Have we arrived yet?’ The first boy answers him, it is ‘still a little further to the end’. When the two boys arrive at the designated place, the first boy lies his head on his friend’s back who leads him to the place from which they started. WÒalna lo ba¨ad? Ba¨ad shwayyah liç-çi¨ab108
Have we arrived yet? Still a little further to the end
to’). This prefix has additional meanings in IA. It may have an optative function and can convey a wish, a request, and urging, as in danruÌ, ‘let’s go!’ and in the sentence, ‘…galat-lah dig¨ud maykhalif…,’ ‘she said to him sit down, it does not matter…’ (al-Karmili, 2003: 151). I think it is a remnant of the word ga¨id (q > g) used before the imperfect to denote, among other things, a progressive action. In the Bedouin folk tales da also introduces the English infinitive, as in ‘… u-lizam il-saççinah dayiksir farid wuÌdah’, (al-Karmili, 2003: 206) ‘and he held the knife to open one’ [pomegranate]. The majority of Iraqis use da and very rarely ¨a (See here part A: Song 14, n. 5). However, the Jews of Baghdad and the Christians in Mosul and Baghdad use qa instead of da. In the following example the use of ga occurs with da in the same expression: ‘u-ani hassah ga¨dah adawwir ¨alayh ma da algih’, ‘and I am now looking for him but I am still not finding him’, (al-Karmili, 2003: 152). 108 Lil-çi¨ab, lit. ‘to the bottom’. 575
THE FOLK SONGS OF IRAQI CHILDREN
Song 43: Sayyarah Îamrah
A Red Car
Basra, Q 1: 109 This song is widespread in the south and is sung by girls or boys separately. SiÌgat is-samrah Irja¨ ya ma†or! Jibinna diktor! 109 Dictor hadi Labis taraji110 Taraji ummah Slaymah li†-†ummah! Yil¨ab bi-†-†aÌin Yomiytah ¨ishrin
[The car] ran over the dark-skinned girl Come back, motorcar! Bring us a doctor! Doctor Hadi Who wears earrings His mother’s earrings May the hot south wind bury him! He fiddles with the flour His daily wage is twenty [paras]
Song 44: Aswad Isaywud
Black Blackish
Baghdad, al-Îanafi, Mu¨jam I This children’s song is a riddle. The word isaywid is dim. of aswad, ‘black’ and mraywid is dim. of mrawid, ‘earrings.’ The earrings represent the two holders of the Òaj, the bread baking tin, which is in the shape of a half circle. Aswad isaywid Bidhnah meraywid Yiftil limÌallah ¨Isma ¨abda¬¬ah
It’s black, very black In its ears are earrings It roams in the neighbourhood His name is ¨Abdallah
Song 45: Tanakçi Tanakçi111 Tinsmith! O Tinsmith Karbala}, Q 3:549 Children stand holding hands in a half-circle or in an arc and sing together. The song is a collection of verses taken from different songs. Tanakçi ya tanakçi! Ja ¨ammi u-kitanni 112 Kitanni bil-¨aÒayah
Tinsmith! O tinsmith! My uncle came and beat me He beat me with the stick
109 Jibinna < jibilna. The /l/ of the preposition /lal/ is assimilated with the /n/ of the suffix /na/. 110 Taraji < taraçi, (sing. tirçiyyah and tarchiyyah), ‘earrings’ where /ç/ >/j/. 111 See a variant in Q 3:546. 112 Kitanni < kitalni, ‘he killed/ beat me’, where /l/ of the verb kital is assimilated with the suffix ni, and q/ > /k).
576
THE FOLK SONGS OF IRAQI CHILDREN
¨AÒayah wayn a∂ummah? A∂ummha ¨id 113 Hadiyyah Hadiyyah jabat walad Sammatah ¨abd iÒ-Òamad
The stick, where shall I hide it? I’ll keep it with Haddiyah Hadiyyah gave birth to a boy She named him ¨Abd is-∑amad
Song 46: Futah ¨Ala Futah A Piece of Fabric on a Piece of Fabric Baghdad, Îajj 4:221 Fu†ah ¨ala fu†ah114 Wul-¨ayn majlu†ah YaÌayyat il-laglag Tidbi ¨ala rijli Rijli mÌannayah Wadaytha lil-khan Il-khan mayridha Iyrid ballu†ah Fu†ah ¨ala fu†ah Wul-¨ayn majlu†ah
A piece of fabric on a piece of fabric And the eyes are peeled O, the snake of the stork Crawling on my leg My leg is dyed with hennah I took it to the inn The inn doesn’t want it It wants a piece of oak A piece of fabric on a piece of fabric And the eyes are peeled
Song 47: Bidi Numayah
In My Hand is a Sweet Lemon
Q 1:248 Bidi numayah, numayah Îilwah wumÌashshayah WumÌashshayah Waddaytah lil-khan lil-khan Wul-khan mayridha mayridha Wiyrid ballu†ah Wul-¨ayn magshu†ah Magshu†ah ¨Ammi ya sayyid salman Hizz il-walad khallah ynam! Wugdhayltah ¨ala ¨yunah ¨Ashig uma yin†unah ¨Ashig bnayyat il-sul†an
In my hand is a sweet lemon, a sweet lemon Pretty and stuffed And stuffed I took it to the inn, to the inn But the inn doesn’t want it, it doesn’t want it It wants a piece of oak And the eye[s] are peeled Are peeled My paternal uncle Salman Rock the boy, let him sleep! When his lock of hair is on his eyes He is in love but they do not give her to him He is in love with the sultan’s daughter
113
¨Id < {ind, ‘with’ as in bint>bit, ‘daughter, girl’. Fu†ah, an Indian word, ‘a square or rectangle of fabric used to cover the body from the hip down’ (Tur. futa, al-Çalabi, 1960: 53). In early Islam, the fu†ah covered the middle of the body and the hip only (al-Jadir, 1989: 100). 114
577
THE FOLK SONGS OF IRAQI CHILDREN
Song 48: Ya Layla ¨Ala Waysh Tibçin? Oh Layla Why are You Crying? Basra, Q 1:343 Standing in an arc, girls sing this song. One girl who stands with her back facing the other girls impersonates the childless woman. Girls: Ya layla ¨ala waysh tibçin? ¨Ala waysh tibçin ¨ala waysh tibçin? Layla: Abçi warid bnayyah Bnayyah bnayyah Girls: Shisimha? Shisimha? Layla: Is¨ayyidah, Is¨ayyidah115
O Layla why are you crying? Why are you crying? Why are you crying? I cry [because] I want a daughter A daughter, a daughter What’s her name? What’s her name? Is¨ayyidah, Is¨ayyidah
Song 49: Ani Sihamah
I Am Sihamah
Kufah, Q 1:168 Girls sing this song, sometimes with clapping, while sitting or walking together. Ani Sihamah Mithlil-Ìamamah Amshi wadur Mithl¨aÒfur AruÌ lil-basatin Akul rumman Aji lil-bayt Akul ciklayt
I am Sihamah Like the dove I walk and go around Like the sparrow I go to the orchards I eat pomegranates I come [back] home I eat chocolate
Song 50: Bit il-Rif
The Village Girl
¨Amarah, Q 3:320 Girls sing this song as they stand in a ring. When some European goods were introduced to Iraqi villages, the desire for European furniture and other Western items in the brides’ dowries increased. This song reflects this phenomenon. Children probably adapted this song from adults. Bit ir-rif za¨lanah Wutrid thob iswayki ¨Ammat ¨allaya-l-bufayh Wumnayn ajib il-komidi116 115 116
The villager girl is angry She wants a modern dress The buffet caused me a lot of trouble And from where will I bring the cabinet
Is¨ayyidah, dim. Sa¨diyah. About the words bufayh and komidi, see al-Îajjiyyah, 1986, V: 49 and 103. 578
THE FOLK SONGS OF IRAQI CHILDREN
Song 51: Îajanjali Bajanjali 117
Hajanjali Bajanjali
Mosul, al-¨Ubaydi, 211 Boys and girls form a ring and sit in a room. One girl sits in the centre of the ring while pointing to the legs of the other players. If her hand points to a boy when she sings the last word of the song, that boy folds one leg. If this is repeated, he becomes the winner. A village girl was engaged to a man whom she expected to marry after the harvest. While she was working in the field her anklets were exposed. Awaiting her marriage after the passing of Rama∂an, she asked her uncle how many days were left in Rama∂an. She worked so hard during the harvest that she had a backache. Îajanjali bajanjali 118 ™la¨tu bghas il-fanjali Qiltulu-l-¨ammu Ìsayn Ash baqa r-rama∂an Qalli sab¨ t-iyyam119 wul-tamam Inkasagh Òunduq[i] Ma binu ghayg il-jinajil wul-¨∂am Shil ghijil Ìi†† ghijil ya glam!
Carobs vendor, bajanjali I went to the top of the high land I told Uncle Hussein ‘How many [days] are left of Ramadan’ He said, ‘seven days to the end’ [My] chest broke It has nothing but the anklets and the bones Lift a leg up, put a leg down, boy!
Song 52: Hdayyah Bdayyah
Hdayyah Bdayyah
Qal¨at ∑alih, Q 1:225 This is also a song for determining the leader in a gaming song. Girls and boys sit on the floor and stretch their legs. It is played like the game mini, mini, mo. The last player who remains is the leader. Hdayyah bdayyah120 NaÒir dayyah Yishidd il-kurah ¨Ala-z-zanbur
Hdayyah bdayyah NaÒir is my brother He sets the wasps’ nest On the wasp
117
See a variant from Karbala} in al-Tu¨mah, 1976: 207. Bajanjali, ‘carob’. The meaning of this word is unclear, see Kirkis, 1990, I: 155. 119 Sab¨ t-iyyam, the /t/ of sab¨at, ‘seven’ becomes in the pronounciation a part of the following word iyyam, ‘days’, hence the two words are pronounced as one. 120 The same song is current in Kuwait (see al-Maghribi, 1996: 380). The word hdayyah is derived from the CA root Î, D, Y, ‘to challenge’ and the word bdayyah from (b, d, } ) ‘to begin’. The approximate meaning of these two words is ‘I challenge [s.o.] to start’. For a variant, see Sa¨d al-Din, 1988: 27. 118
579
THE FOLK SONGS OF IRAQI CHILDREN
MsayyiÌha wumfayyiÌha Bab il-Ìilla u-bab ish-sham Shaggif riylak,121 ya Rama∂an
He lets it run and causes it to spread fragrance [At] the Hillah Gate and Damascus Gate Lift your leg, O Ramadan
Song 53: Dir Finjanak
Pour Your Cup
Children play this singing game like the song il-Tha¨lab fat fat (Part C, song 9). They sit on the floor in a circle. The leader goes around the other players while singing one verse, then the group sings the next verse and so on until the end of the song. Dir finjanak Îabiç labiç! 122 Dir finjanak La tiltibik!
Pour your cup Quickly, quickly! Pour your cup Don’t be confused!
Song 54: Ya Samrah!
O Dark-skinned Girl!
Hor SÌayn, Q 3:437 Boys sing this song while sailing in their small boats in the swamps of the wetlands. Ya samrah umm ¨yun is-sud! Hat[i] iÌjayrah123 ¨u-†iggini Halbat124 i††iÌ ib¨ayni Wa-khidhnniç125 ib-taliyyah126 Wahliç yirdun il-miyyah Dhak il-¨amm in-naÒrawi Dhabbani ibdhiç il-çifah Wul-çifa-l-tnaggu† riÌah Ba¨ iÒkhula u-khallani
O dark-skinned girl with black eyes! Throw a stone and hit me Of course it will hit my eye And I’ll marry you in the end Your family will want a hundred That uncle [of yours] from NaÒiriyyah Tossed me in that stink The stink that drips a [foul] smell He sold his billy goats and left me
Song 55: Il-™obah!
O Ball!
Baghdad, Q 3: 269 This song is relatively new. One girl bounces the ball quickly and sings the word çoçah, ‘a chick’, which is an endearing name used by 121
Rijlak > riylak. Îabiç labiç, the meaning of these two words is unclear. It seems that they are like libiç shibik in part C, song 9, n. 2 above. 123 IÌjayrah, dim. of Ìjarah, ‘stone’. 124 Halbat, ‘of course’ (Pers.). 125 Wa-khithniç. The /n/ in wa-khithnniç is for emphasis, used by Bedouins. 126 Ibtaliyyah, more common is ibtaliha, ‘in the end of it’. 122
580
THE FOLK SONGS OF IRAQI CHILDREN
Iraqi children for a girl. In the 1930s, when tennis balls were first introduced to Iraq, girls began to use them in their games. Chabnis is probably a distortion of the English ball brand ‘Champion’. Shifit il-çoçah ¨ala-s-saddah Bidha khubzah u-titgadda Îaçaytah ma Ìaçatni Alif n¨al la††aytha
I saw the pampered girl on the dam Holding bread in her hand, eating her lunch I talked to her, but she didn’t talk to me [Then] I threw one thousand slippers at her
Song 56: ¨Iliç Mayy
Chewing Gum
Saddat il-Hindiyyah, Q 3:495 This game song is played in the villages of the middle Euphrates. When girls play a game and one of the participants gets mad and leaves the game chewing gum (¨iliç Mayy), the other players sing with or without clapping. This kind of gum is extracted from oak trees. Girls chew it after boiling and cleaning it. Some Iraqis use it to help with digestion. Ti¨liç iz-za¨lanah b¨iliç Yom il-mubarak tinmiliç127 Takhudh abu sinn idh-dhab128 Wutnam ¨al-çarpayah129 Wutloli lil-fikhtayah130
The angry girl chews gum Blessed be the day when she gets married She will marry the man with the gold tooth And sleep on the bed And sing a lullaby for the pigeon
Song 57: MuÌammad yal-ghawi!
O MuÌammad, the Rascal!
NaÒiriyyah, Q 2:279 Accompanied by hand clapping, girls chant this song in NaÒiriyyah and Samawah during winter. Earrings represent childhood, hunting represents manhood, and rain is a symbol of bliss and success. At the end of the song the mother takes an oath to slaughter a sheep for the safety of her child who fell down. MuÌammad yal-ghawi Tirçiytah lawi Lawiha lawi RayiÌ lil-ganuÒ 127 128 129 130
O MuÌammad, the rascal His jewellery he bent He bent them a lot He is going to hunt
Timtiliç, lit., ‘she will be owned’. Abu sinn idh-dhab, is an epithet for ‘a rich man’. Çarpayah is Per. çahar, ‘four’ and payah, ‘leg’. Tloli, lit. ‘she will sing luluwwa, ‘lullaby’. 581
THE FOLK SONGS OF IRAQI CHILDREN
GaniÒlah marah Îilwah mgaÒmilah131 Tighsillah sharbah Tighsillah Òaytah132 Min mayy il-mu†ar Halhal wunha†ar Galat ya ¨Ali AdhbaÌlak †ili Min ghanam hali 133 WaÒÒilni-l-hali
He hunted a woman for himself Pretty and tidy She washes his moustache She washes his robe From the rainwater He made sounds of joy and fell to the ground She said, ‘May ¨Ali [protect you] I’ll slaughter one sheep for you From my family’s herd [So] take me to my family’
Song 58: Shandal u-Mandal 134
Shandal and Mandal
al-¨Abbasiyyah-Karbala}, Q 3:533 Children chant this song about a certain girl named Salimah, whose earrings had been stolen. Shandal u-mandal ¨Algaw rayah Salimah bit Jaddu¨ Bagaw Traçiha
Shandal and Mandal They hang a piece of cloth Salimah Jaddu’s daughter They stole her earrings
Song 59: Ummi Tnadini
My Mother is Calling Me
Samarra}, Q 1:359 Girls chant this song before sunset when their mothers call them to come home. They stand in a file. Each girl holds the dress of the one next to her and all sway left and right and sing. Ummi tnadini Gharbil-basatini Madri ashti¨†ini Ti¨†ini tuffaÌah Îamraw-lawwaÌah Îilafit ma ∂ugha
My mother is calling me [From] the west side of the orchards I don’t know what she’ll give me She’ll give me an apple Red and ripe I took an oath that I won’t eat it
131
MgaÒmal, passive participle of qaÒmala (CA), ‘to trim’. ∑aya, (Per. Òayah), a fabric from which the vest is made, worn over the dishdashah. The dishdashah is the standard dress for children and adults not wearing western clothing. It is an ankle-length robe with a buttoned opening halfway down the front. 133 Hali < ahli, ‘my family, relatives’. 134 Two fictitious proper nouns, which create intensification by repeating the first word with its initial consonant changed. 132
582
THE FOLK SONGS OF IRAQI CHILDREN
¨Illa yiji khayyi WÌayyid Khayyi hadha l-luwÌayyid ImÌanni sirwalah135 Sirwalah hal lilo Wush-sha¨ar mÌaylilu
Unless my little brother Whayyid comes This brother, the only brother of mine He dyed his pants with henna His pants [embroided] with pearls And his hair is untied
Song 60: Sha¨ri Tuwil
My Hair is Long
Anbar, Q 3:542 Girls form a ring, hold hands, and sing while dancing clockwise: Sha¨ri †uwil sha¨ri †uwil Mithl il-Ìarir mithl il-Ìarir ¨Iyuni zurug ¨iyuni zurug Tubruq bariq tubruq bariq Ya m¨allimnah u-ya mudir! ∑afguli arba¨ Òafgat WaÌid ithnayn thlatha arba¨ah
My hair is long, my hair is long It is like silk, like silk My eyes are blue, my eyes are blue Shining brightly, shining brightly O our teacher, O principal! Clap four times for me One, two, three, four
The following song is a variant from Hit Q 3: 542–3. The use of the word ‘mama’, for, ‘mum’ attests that this version is relatively recent. Sha¨ri †uwil Ya mama! Sha¨ri †uwil Warid aguÒÒah Ya mama! Warid aguÒÒah Ya mama! Wul-Ìayyah itnadi ya mama Ya mama! Wul-Ìayyah itnadi Ya mama! Wladi il-¨ashrah Ya mama! Wladi il-¨ashrah Ya mama! Wul-¨ashrah gashrah Ya mama! Wul-¨ashrah gashrah
135
My hair is long O mum! My hair is long I want to cut it O mum! I want to cut it O mum! And the snake is calling, O mum! And the snake is calling O mum! My ten sons O mum! My ten children O mum! But ten is unlucky [number] O mum! But ten is unlucky
Sirwal < Shirwal (per.), ‘trousers’. See Dammus, 2001: 56. 583
THE FOLK SONGS OF IRAQI CHILDREN
Song 61: ™ubag ™ubag ÎummuÒ
A Platter, a Platter of Chickpeas
¨Amarah, Q 3:382 Girls stand next to each other, holding hands forming a circle. One girl stands in the centre of the circle and calls the name of a girl in the group. This girl turns her face in the opposite direction of the standing group. When all the names of the girls are called and their faces are turned in the same direction as the first girl, the game ends. ™ubag †ubag ÌummuÒ Ya mindilah! ¨Ala-l-jidir yunbus Ya mindilah! Ija yusuf ¨argan Ya mindilah! Yahu-l-thaffilah? Ya mindilah! Wasan thaffilah Ya mindilah!
A platter, a platter of chickpeas O mindilah! On the pot it is showing O mindilah! Yusuf came sweating O mindilah! Who is the one who’ll fan him? O mindilah! Wasan will fan him O mindilah!
Song 62: Sabit Sabbut
Saturday, Precious Saturday
The Arabs of the pre-Islamic period had sceptical beliefs about the months and the days of the week. One tradition claims that the Prophet MuÌammad advised his followers saying, ‘la tu¨adu al-ayyama fa-tu¨adikum’, ‘do not be hostile to the days [of the week], so they won’t be hostile to you’. They also attributed to the Caliph ¨Ali a poem about the days of the week.136 Some Iraqis are still superstitious about the days of the week. They believe that Saturday is good for hunting and holding weddings, but it is not desirable to eat fish on it, out of fear that this causes poverty. On this day a few Iraqi women avoid going to the public bath and doing hard work. Sunday is a good day to undertake projects like building a room or the like, because on that day God created Heaven. Some women do not weave during this day or sew new clothes or travel. It is desirable to travel on Sunday evening and Monday. Travelling on Monday brings wealth. Al-Maghribi mentions a tradition related by Abu Hurayrah137 that God examines humans’ deeds every Monday and that He forgives everyone who worships Him. On Tues136
For details about this tradition and the Iraqi folk beliefs about the days of the week, see al-∑arraf, 1930, VIII: 467–71 (especially p. 467); al-Îajjiyyah, 1968, II: 109–12 and FraiÌah, 1988: 137. 137 al-Maghribi, 1996: 136. 584
THE FOLK SONGS OF IRAQI CHILDREN
day it is desirable to sew new dresses; the violator will die by fire or drowning. It is advantageous to draw blood (Ìijamah) ‘cupping’ (blood letting) on this day. Wednesday is the best day for taking medications and eating fish, but a woman, who goes to the public bath on this day, will catch cold and have a headache. On Thursday God accepts supplications and He does not like Muslims to travel on Thursday night lest they miss Friday prayers. Friday is a joyful day for holding weddings, enjoying oneself and visiting the graves of relatives. Children in the religious schools used to sing this song about the days of the week (Îajj 2:109–11): Sabit sabbut! AÌÌad ¨ankabut Thnayn babayn Talatha manarah Arbi¨a ziyarah Il-Kamis li¨batna Wul-Jum¨ah ¨I†latna
Song 63: Ba††anna U-¨Azzanna
Saturday, [O] precious Saturday! Sunday is a spider Monday two doors Tuesday is a minaret Wednesday is a visiting day Thursday is our play day And Friday is our holiday
We Stopped [playing] and Finished
NaÒiriyyah, Q 2:248 When called to come home after finishing play, some Iraqi children sing this song on their way home: Ba††anna u-¨azzanna 138 U-ija-l-wawi ukanna 139
We stopped [playing] and finished And the jackal came and ate us
Song 64: ™igg, ™igg, Îajar! Bang, Bang, Stone! Dhi Qar, Q 3:390 Played by either boys or girls alone. A group of boys or girls is divided into two sub-groups. The leader holds two stones and agrees with the other players as to which one is jannah ‘paradise’ and which one is nar ‘fire’. He strikes one stone against the other singing. The leader of the sub-group answers, ‘this one is paradise’ and ‘this one is fire’. When the dialogue is over, the other sub-group begins the game in the same way.
138
Ba††al, ‘to stop+na > ba††anna, where /l/ assimilates with the sun letter /n/. Kanna < u-akalna, /l/ is assimilated with the suffix na and the initial hamzah of the verb akalna is omitted. 139
585
THE FOLK SONGS OF IRAQI CHILDREN
Player 1: ™igg, †igg, Ìajar ™ugg, †ugg Yahu-l-jannah? Yahu-n-nar? Player 2: Hay il-jannah U-hay il-nar Player 1: Yahl il-jannah ishtaklun? Player 2: Khokh u-rumman Khokh u-rumman Player 1: Yahl in-nar ishtaklun? ÎaÒu wu-trab Player 2: ÎaÒu wu-trab
Bang, bang, stones Bang, bang Which one is Paradise? Which one is Hell? This one is Paradise And this one is Hell O inhabitants of Paradise! What do you eat? Peaches and pomegranates Peaches and pomegranates O inhabitants of Hell! What do you eat? Pebbles and dust Pebbles and dust
Song 65: Flanah ™il ¨at li-Barrah So and So [name of a girl] Went Outside Baghdad, Îajj 3:53 In spring when roses are at the peak of blossoming, children pick a petal of a rose and place it on a ring made with the index finger and thumb. After saying the first line, everyone asks, Ìurrah lo gaÌbah, ‘chaste or a whore’, then they strike the petal. If it produces a sound, the other children cry Ìurrah ‘chaste’, if not they cry, gaÌbah, ‘whore’. Flanah †il¨at li-barrah Wu-khdudha muÌammrah Lagoha sittah sab¨ah GaÌbah lo Ìurrah?
So and so [name of a girl] went outside And her cheeks are dyed with red Six-seven [men] run into her Is she a whore or chaste?
Address for Correspondence: 25205 Ward Place, Carmel, CA 93923, USA REFERENCES Abu-Îaidar, Faridah. 1991. Christian Arabic of Baghdad. (Wiesbaden) al-¨Alawçhi, ¨Abd al-Îamid. 1966. Min Turathina al-Sha¨bi. (Baghdad) —— 1975. al-Turath al-Sha¨bi IV (Baghdad), 47–53 ¨Ammar, Hamid. 1954. Growing up in an Egyptian Village. (London) al-¨Ani Walid, ¨Abd al-¨Aziz. 1975. ‘al-Al¨ab al-Mushtarakah lada A†fal ¨Anah’, alTurath al-Sha¨bi IV, 199–204 ¨Athamnah, Khalil. 1987. ‘al-Jinn Jiran la Narahum’, al-Karmil. (Haifa) ¨Awwad, Kirkis. 1990. MaÒadir al-Turath al-{Askari {ind al-{Arab (Beirut) al-Bakri, Îazim. 1972. Dirasat fi }l-Alfa al-¨Ammiyyah al-MawÒiliyyah. (Baghdad) Baneth, D.Z. 1950. Alexander Marx Jubilee Volume. (Newark) Blanc, Haim. 1964. Communal Dialects in Baghdad. (Cambridge) 586
THE FOLK SONGS OF IRAQI CHILDREN
al-Çalabi, Dawud. 1935. al-¨Athar al-Aramiyyah fi Lughat al-MawÒil al-¨Ammiyyah. (Mosul) —— 1960. Kalimat Farisiyyah Musta¨malah fi ¨Ammiyyat al-MawÒil, Baghdad al-Dabbagh, ¨Abd al-Khaliq. 1956. Mu¨jam Amthal al-MawÒil al-¨Ammiyyah I. (Mosul) Dammus, Îalim. 1923. Qamus al-¨Awamm. (Damascus) al-Daywachi, Sa¨id. 1966. ‘Aghani al-¨Id fi }l-MawÒil’, al-Turath al-Sha¨bi (Baghdad), I, 20–2 al-Dulayshi, ¨Abd al-La†if. 1968. al-Al¨ab al-Sha¨biyyah fi }l-BaÒrah. Baghdad Fattal, Salim. 2003. Bai-sim†a}ot Baghdad (Heb., ‘In the Alleys of Baghdad’). (Jerusalem) FraiÌah, Anis. 1988. Asma} al-Ashhur wa}l-A¨dad wa}l-Ayyam. (Lebanon) al-Îajjiyyah, {Aziz Jasim. 1967. Baghdadiyyat I. (Baghdad) —— 1968. Baghdadiyyat II. (Baghdad) —— 1973. Baghdadiyyat III. (Baghdad) —— 1981. Baghdadiyyat IV. (Baghdad) —— 19??. Baghdadiyyat V. (Baghdad) —— 1991. Baghdadiyyat VI. (Baghdad) —— 1999. Baghdadiyyat VII. (Baghdad) al-Îanafi, Jalal al-Din. 1962. al-Ayman al-Baghdadiyyah I. (Baghdad) —— 1964. al-Ayman al-Baghdadiyyah II. (Baghdad) —— 1964. Mu¨jam al-Alfa al-Kuwaytiyyah. (Baghdad) —— 1978. Mu¨jam al-Lughah al-¨Ammiyyah al-Baghdadiyyah, I. (Baghdad) —— 1982. Mu¨jam al-Lughah al-¨Ammiyyah al-Baghdadiyyah, II. (Baghdad) Îasan, Îusayn ¨Abbas. 1966. ‘Îawla Ughniyat Sidi Sidi’, al-Turath al-Sha¨bi V, 66 —— 1988. ‘Lu¨ab wa-Aghani al-A†fal’ , al-Turath al-Sha¨bi (Baghdad) III. al-Jadir, Walid. 1989. al-Azya} al-Sha¨biyyah fi }l-¨Iraq. (Baghdad) al-Karmili, Anastas. 1972. al-Musa¨id I. (Baghdad) —— 2003. Diwan al-Tiftat. (Beirut) Khayya†, Ibrahim. 1986. Majmu¨at al-Amthal al-Baghdadiyyah I. (Baghdad) Kojaman, Y. 2001. The Maqam Music Tradition of Iraq. (London) Leick, Gwendolyn. 1991. A Dictionary of Ancient Near Eastern Mythology. (New York) Lindsay, Jones (ed.). 2004. Encyclopedia of Religion,2 Vols. I, VI, IX, XI. (New York) McCarthy, R.J. and E. Raffouli. 1964. Spoken Arabic of Baghdad II. (Baghdad) al-Ma¨dadi, ¨Abd al-La†if. 1977. ‘al-MilÌ fi }l-Fuluklur al-¨Iraqi’, al-Turath al-Sha¨bi (Baghdad), 67–72 al-Maghribi, Salwa. 1996. Aghani al-Banat al-Sha¨biyyah bayna al-FuÒÌa wa}l¨Ammiyyah. (Kuwait) al-MallaÌ, MaÌmud. 1928. ‘Nukat wa-Ghara}ib Lughawiyyah’, Lughat al-¨Arab, I, year 6. (Baghdad), 349–53. Mansour, Jacob. 1965. Hagaddah shail Pesah (Judeo-Arabic). (Jerusalem) —— 1991. The Jewish Baghdadi Dialect. (Jerusalem) Masliyah, Sadok. 1996. ‘Four Turkish Suffixes in Iraqi Arabic: -Li, -Lik, -Siz and –Ci’, JSS 41/2, 91–8 —— 1998, ‘Abu and Umm in Iraqi Arabic’, JSS 43/1, 113–29 al-Musawi, Shukriyyah. 1965. ‘Min Aghani al-A†fal fi }l-Furat’, al-Turath al-Sha¨bi (Baghdad), 21 Qadduri, Îusayn, Lu¨ab wa-Aghani al-A†fal al-Sha¨biyyah fi }l-Qi†r al-¨Iraqi, Baghdad, I, 1975; II, 1984; III, 1988 Sa¨d al-Din, KaÂim. 1966. ‘al-™uyur fi }l-Fuluklur al-¨Iraqi’, al-Turath al-Sha¨bi (Baghdad), 58–64 587
THE FOLK SONGS OF IRAQI CHILDREN
—— 1979. al-Îikayat al-Sha¨biyyah al-¨Iraqiyyah. (Baghdad) —— 1998 ‘Aghani al-A†fal al-Sha¨biyyah fi }l-¨Iraq’, al-Turath al-Sha¨bi (Baghdad), I, 4–38 al-Samarra}i, Ibrahim. 1968. al-Tawzi¨ al-Lughawi al-Jughrafi fi }l-¨Iraq. (Baghdad) —— 1975. ‘Ahazij al-A†fal fi-Samarra}, al-Turath al-Sha¨bi (Baghdad) VIII/IX al-Samarra}i, Yunus. 1965. al-Al¨ab al-Sha¨biyyah li-∑ibyan Samarra} I (Baghdad), 18–24 al-∑arraf, AÌmad Îamid. 1935. ‘al-Ayyam fi-}l-Mu{taqadat’, Lughat al-{Arab, VIII, year 5 (Baghdad), 466–71 —— Lughat al-¨Arab, year 6, 31–2 al-Shalji, ¨Abbud. 1982. Mawsu¨at al-Kinayat al-¨Ammiyyah al-Baghdadiyyah I. (Beirut) Shawqi, ¨Abd al-Îakim. 1987. Madkhal li-Dirasat al-Fuluklur wal-Asa†ir. (Cairo) Sher, Addi. 1980. Mu{jam al-Alfa al-Farisiyyah al-Mu{arrabah. (Beirut) ShuÌan, AÌmad. 1984. al-Amthal al-Furatiyyah. (Damascus) Stevens, E.S. 1931. Folk-Tales of Iraq. (London) Thinyan, {Abd al-La†if. 2001. SafaÌat min Qamus al-¨Awamm fi Dar al-Salam. (Baghdad) al-Toma, Salih. 1969. The Problem of Diglossia in Arabic. (Cambridge) al-™u¨mah, Salman. 1976. ‘NuÒuÒ Ghina¨iyyah lada A†fal Karbala’, al-Turath al-Sha¨bi, (Baghdad), II–III, 201–8 al-¨Ubaydi, Azhar. 1989. al-MawÒil Ayyam Zaman. (Mosul) al-Wahhab, ¨Aziz. 1975. ‘Khiyanat al-¨Uhud’, al-Turath al-Sha¨bi. (Baghdad) VI, 157–61 Woodhead, D.R. and Wayne Beene. 1967. A Dictionary of Iraqi Arabic, ArabicEnglish. (Washington) al-Zubaydi, Zayn al-Din. 1994. MukhtaÒar ∑aÌiÌ al-Bukhari, MuÌammad MuÌsin Khan (ed.). (Riyadh)
588
Journal of Semitic Studies LV/2 Autumn 2010 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
REVIEWS BILLIE JEAN COLLINS, The Hittites and their World (Society of Biblical Literature Archaeology and Biblical Studies). Society of Biblical Literature, Atlanta 2007. Pp. xvi + 254. Price: $29.95 paperback. ISBN: 978-1-58983-296-1. This is an informative and well documented book, whose publication by the Society of Biblical Literature might lead some readers to suspect an excessive bias, but this would be unjustified. The fifth and final chapter, entitled ‘Hittites in the Bible’, has a length of just 22 pages, but provides a succinct discussion of the Neo-Hittites as well as their neighbours and rivals in the Iron Age and their relations with the Old Testament kingdoms of Israel and Judah. Supplying all the data required for this chapter is a list of all relevant biblical passages (pp. 246–8), though the Assyrian annals are omitted. This reviewer began by looking at the list of ‘Further Reading', described as covering only English-language monographs. The true bibliography is to be found in the numerous full footnotes. Consequently this is not a publication to be dipped into but rather to be read with attention. No criticism! It deals with the surviving written records, in Hittite, Luvian, Akkadian, Hurrian, Egyptian and others, culminating in the Old Testament, relevant only for the Iron Age, the centuries following the collapse of the Hittite empire in the early twelfth century BCE. This book does not aspire to any exhaustive treatment of Hittite material civilization, though the architecture of temples, the majority in Hattusa (Bogazköy), and those artifacts casting light on society and religion (including of course the Yazilikaya reliefs, as well as the Inandik and Bitik vases, depicting festival and marriage scenes respectively) are illustrated and discussed. That the more mundane facets of material culture are left to one side is highlighted by the omissions from the index of modern authors and also from the list of abbreviations such as Anatolian Studies, Anatolica and Belleten. But in a text with a total length of only 223 pages there are certain to be omissions. The first chapter, ‘A Brief History of Hittite Studies’, gives a clear and balanced narrative, centred on the somewhat chequered history of the Bogazköy excavations, fortunate to have had the leadership of Kurt Bittel, Peter Neve and Jurgen Seeher in the decades since the inter-war period. Of course the foundation of Hittite scholarship rests on the contribution of that brilliant but flawed genius, Hugo Winckler. Kurt Bittel’s admirable career included his employment as epigraphist from 1933 and continued support of Hans Güterbock, whose Jewish background might well have consigned him to the academic wilderness. These facts are rightly alluded to in this chapter. Early in the long second chapter, ‘A Political History of the Hittites’, the different theories of Indo-European origins in Anatolia are set out briefly without any strong input from the author, who treats Colin Renfrew’s theory of an origin around Çatal Hüyük with scarcely deserved respect, albeit with some support from evolutionary biology. The major factor of the Early Trans-Caucasian (Kura-Araxes) cultural zone, a great wedge between central Anatolia and the lands north of the Caucasus, is not mentioned, nor the Khirbet Kerak problem. The Proto-Indo-European background is, however, indirectly alluded to in the ‘Afterword', as a footnote reference to Gamkrelidze and Ivanov. 589
REVIEWS
The ongoing story of the defence against the Gasga attacks from the north and the successive campaigns against Arzawa and other enemies in the west is effectively covered, with the increasing international dimension from Suppiluliuma I onwards, long after the daring raid by Mursili I on Babylon. The Hittites were by nature possibly more modest or at least more down to earth than their enemies, notably the Egyptians. Thus, though the great battle of Kadesh can surely be reckoned as a draw in their favour, the Hittites have left no account to counterbalance the boastful inscriptions and reliefs of Ramesses II, giving the impression of a single-handed triumph by the Pharaoh. The tangled Hittite dynastic relationships, at least from the deposition of Urhi-Teshub by his uncle Hattusili III, certainly contributed to the eventual downfall of Hatti, as did the defeat of Tudhaliya IV by the Assyrians at Nihriya, well east of the Euphrates. This undermined the prestige of the monarchy and the loyalty of its vassals, including Ugarit. Economic factors, most significantly widespread famine, were also crucial; and the multi-ethnic composition of society became no longer a strength but a source of discord. The third chapter, ‘Society’, is more wide-ranging than its title implies. The pages on art and architecture (126–41) could perhaps have been somewhat expanded to form a separate chapter. The Yazilikaya reliefs, following modern consensus, are associated with the Hurrian pantheon and specifically with Puduhepa, Hattusi III’s highly influential Hurrian queen. At the head of society stood the king, the most inspirational of all Hittite rulers having been Hattusili I (c. 1650–20 BCE), not only a strong military commander but also one setting standards of justice for himself and his heir. He was moreover probably the originator of the Hittite tradition of historiography, thus having older roots than the Assyrian equivalent. Among illuminating passages is the description of the duties and responsibilities of various officials, from priests and palace staff to the royal bodyguard. Hattusa was controlled by a mayor, with councils of elders in remoter towns (pp. 104–5). One error: the citadel of Hattusa is not Büyükkaya (‘great rock’) but Büyükkale, the former standing north of the citadel (p. 99). District governors were responsible for security, especially along the frontiers, and for supervision of state agricultural lands, upkeep of public buildings and administration of justice. The essential standing army was headed by the elite chariotry, though 90% of personnel were infantry. Kikkuli is mentioned (p. 107), but without reference to his coming from Mitanni. Cavalry, few in number, acted as scouts and dispatch riders. The comparison of the Hittite treaties with the covenant between Yahweh and Israel on Mount Sinai is perhaps plausible, if not taken to indicate any direct derivation from the former. Biblical parallels are discussed in the ‘Afterword'. Hittite society was held together by regulations of every activity and misconduct. A tithe levied on produce and provision for the corvee labour were due from landowners to the state. The king kept an eye on deportees, as when Mursili II brought 66,000 back from Arzawa, excluding those brought by the officers, infantry and horsemen. Slavery arose from debt, crime or captivity in war. As in modern Anatolia, shepherds were of low social standing. The Hittite laws provided for monetary restitution rather than physical punishment: there was no lex talionis. Alien mythological traditions, Hattian or Hurrian, were exploited by the Hittite scribes, with Hurrian songs and Hittite personal prayers. The official religion; places of worship; festivals; the sacrificial cult; divination; necromancy; prayer; deities and demons; sin and pollution; ritual power; sorcery; cosmogony, cosmology and eschatology; death and afterlife — all these are discussed in the detailed fourth chapter, on religion. 590
REVIEWS
The ‘Afterword' outlines etymological parallels between Hittite and Israelite traditions. Among other points is the suggestion of a common Hurrian origin for both Hittite and Israelite traditions of ritual sacrifice. In addition, points of comparison have been suggested between the Apology of Hattusili III and the account of King David’s rise to power (1 Samuel 16 – 2 Samuel 6). These short final pages provide a ‘punch line’, probably appealing especially to the publisher. This reviewer is not competent to assess the plausibility of such conclusions, appealing as they are. The Hittites have come a long way from being regarded as representing a peripheral civilization. Beside their military prowess they displayed a degree of humanity hard to detect elsewhere in the ancient Near East. The author has successfully brought them into the arena appealing to a wider public, at the same time providing food for the serious student. doi: 10.1093/jss/fgq013
CHARLES BURNEY UNIVERSITITY OF MANCHESTER
WILFRED G. LAMBERT, Babylonian Oracle Questions (Mesopotamian Civilisations). Eisenbrauns, Winona Lake 2007. Price: $55.00. Pp. xiii + 156 + 57 plates. ISBN: 978-1-57506-136-8. This book presents the dividends of several decades of research on a corpus of mainly previously unpublished cuneiform tablets of the genre tamitu. In the opening paragraph of his Preface the author defines this genre as follows:1 A Babylonian tamitu text is a question addressed to the Babylonian gods Samas, the sun god, and Adad, the storm god, as a duo. The questioner wanted to know whether something (an eclipse, for example) would occur or not, or whether he should undertake some enterprise. It was customary to commission a professional diviner to put the question because he was expert in the necessary ritual acts and in compiling a comprehensive question, which was essential because ‘Yes' or ‘No' were the only possible answers. The diviners had developed an argot which had become traditional. This could be compared with the language of modern English legal documents: abounding in rare and unusual words and in convoluted syntax, though these ancient diviners had more literary flare than modern English lawyers.
The genre tamitu now takes its rightful place in the history of Mesopotamian divination. Because of the high level of detail in the tamitu texts, the minutiae of their observation and the range of subjects that they cover, these texts provide a series of veritable vignettes on the life and times of ancient Babylonia, while they also have significant lexical and philological interest. This review will first outline some of the cultural material, and then detail some specific items of philological interest. The book’s Introduction collects and discusses evidence for the two gods addressed in the tamitu oracles, Samas and Adad, a discussion of the concept and etymology of tamitu where the author states that the dictionary definition given by the Akkadisches Handwörterbuch, ‘Anfragebeantwortung’, is wrong. A similar definition is given by the T volume of the Chicago Assyrian Dictionary which appeared in 2006 after the completion of Lambert’s manuscript, ‘query for an oracle and the answer to it’. Both definitions are at least misleading since in these texts no answers but only questions survive. BOQ concludes with fifty-seven plates of the author’s autograph copies of the cuneiform tablets and an index of rare words discussed. 1 I am grateful to W.G. Lambert for reading through the text of this review and making some suggestions. He has also asked me to suggest that the book be referred to as BOQ.
591
REVIEWS
Military matters: There is a large body of textual material in tamitu form that deals with military campaigns including some of Abi-esuÌ and Hammurabi, kings of Babylon in the earlier second millennium BCE. Perhaps most remarkable is a section of text detailing the various methods of laying siege to a city including thirty individual terms for the various types of siege engines and tactics. As the author remarks (p. 144) this is ‘probably the most detailed record of the various techniques for besieging a city, from the ancient world’. Reference is made in the book to a similar passage in Ivan Starr’s Queries to the Sun God (State Archives of Assyria, Vol. IV), nr. 102 and attention is drawn here to a further source of relevant information, Israel Eph¨al, ‘Ways and Means to Conquer a City, Based on Assyrian Queries to the Sungod’, in Assyria 1995, Proceedings of the 10th Anniversary Symposium of the Neo-Assyrian Text Corpus Project, S. Parpola and R.M. Whiting (eds), pp. 49–53. Eph¨al has also written a book on the subject in Hebrew, Siege and its Ancient Near Eastern Manifestations (Jerusalem 1996). One tamitu concerns a campaign of a Babylonian king against the Lullubû people (pp. 62–5). In a broken context the reference to the ‘Lullubû enemy’ is followed by the enigmatic phrase ‘whether it be the Ganas or whether it be the Tuktukâ’. Could this phrase, not interpreted by Lambert, be a reference to Lullubian tribes? A variety of circumstances and scenarios involving war are considered in the tamitu texts, including the safety of people and animals inside and outside the city (pp. 184–231) and even the possibility of mutiny and the incarceration of the commanding officer (pp. 126–7). The reliability of messengers is the object of enquiry, including their tact — whether he will find the ‘correct, friendly’ words — and standing (pp. 122–5). Personal ambition: Several tamitu texts enquire about the wisdom and success prospects of individuals aspiring to achieve office, in one case as a eunuch and in another as a charioteer (pp. 110–11): ‘Should he get up on the chariot and hold the reins? Will he be safe with those reins?’ The tamitu of an individual wishing to become a Sangû, temple administrator (pp. 52–3), illustrates the repetitive argot referred to by Lambert. As translated by him it reads: ‘[Samas], lord of the judgement, [Adad, lord of the inspection], / [so-and-so], son of so-and-so, who now in the matter of [… / [has set his face], has made up [his mind], has determined / and is aiming [to be] appointed a temple administrator.’ There then follows a detailed set of questions as to how the man should prepare himself. Ritual: A number of tamitu texts involve river ordeals, successful and unsuccessful (p. 55). These offer unique insight into the superstitions underlying the institution of the river ordeal in Mesopotamia. The tamitu about a river ordeal resulting in acquittal kicks in with the question of whether the man to be tested will be safe or whether he should hire a substitute to undergo the ordeal in his place. The sun god is told to ‘Ignore that the mind of the owner of this (i.e. the man commissioning the tamitu) … may be gloomy, fearful and despondent, and the mind of his accuser may be slow to work and he may be lazy.’ Another ritual questions whether a specific horse is suitable to pull the chariot of the god Marduk (pp. 80–3). This is followed by an incantation to the horse as a constellation among the Pleiades leading to a passage addressing the actual horse that is to pull the god’s chariot and requesting it to take the part of a supplicant. This is accompanied by a ritual in which the incantation is to be whispered three times through a reed into the left ear of the horse. It seems to the present writer that this procedure belongs to personal religion and that the horse is to intercede with the god Marduk on the supplicant’s behalf. 592
REVIEWS
Personal welfare generally: A recurrent theme in the tamitu texts is the personal welfare and circumstances of ordinary people, including women. This includes not only health (pp. 96–7) but also general wellbeing (pp. 34–7) and personal safety (pp. 120–1). Given the strongly male orientation of much of the Mesopotamian textual corpus the frequency of oracle questions concerning women is striking. One tamitu asks whether a woman who has born her husband many girls (expressed with a Gtn of waladu!) will at last produce a male heir (pp. 88–9). Elsewhere the welfare, health and personal safety of particular women are the objects of enquiry (pp. 116– 19, 130–1), while one tamitu asks whether a woman is telling the truth to her husband (pp. 116–17). Matters philological. Lambert (p. 144) convincingly identifies the noun ba}irtu as a feminine collective, ‘a group/many ba}iru’, that is ‘hunters’ and points out that this type of noun is little recognized in Akkadian and refers to GAG 3 §60a where there is indeed scant comment. From a wider vantage point the volume An Introduction to the Comparative Grammar of the Semitic Languages of S. Moscati et al. p. 86 addresses the matter as follows: ‘The feminine morpheme is employed not only to indicate the corresponding natural gender but also nomina unitatis, diminutives and pejoritives, abstract and collective nouns’. No Akkadian examples are given although each of these categories occurs in Akkadian. An example illustrating Lambert’s point is epistu, CAD E 241a, where of the six meanings given five involve handiwork or similar activity while one is ‘working groups, crew (OB and Mari only)’. In fact epistu, ‘working groups, crew’, clearly relates lexically to CAD’s lemma episu (CAD E 245a 3.), ‘workman’. At any rate, as Lambert points out, this class of nouns has been little recognized and would repay further study. A new word is idippu, translated by Lambert (p. 150) as ‘breath of life’ and to be derived from the verb edepu, ‘blow into, inflate’ (CAD E 28a). The purpose of the tamitu in which it occurs is to ascertain that a man who has been condemned by the river ordeal should not have the breath of life restored to him so that he could return and persecute his tormentors. P. 153: Another new word: asmû, ‘favourable acceptance’, from semû, ‘to hear, obey’. Lambert is certainly correct in positing the meaning ‘partition’ for the word Òullupu (p.108): GIM Òer-pi SA5 SAR Òu-ul-lu-pi, ‘written according to a red, fired, tablet, and partitioned’. Produced by Eisenbrauns at an affordable price, this book is a valuable contribution to Assyriology and Mesopotamian cultural studies. doi: 10.1093/jss/fgq014
ALASDAIR LIVINGSTONE UNIVERSITY OF BIRMINGHAM
ISRAEL FINKELSTEIN and AMIHAI MAZAR, edited by BRIAN B. SCHMIDT, The Quest for the Historical Israel: Debating Archaeology and the History of Early Israel (Society of Biblical Literature Archaeology and Biblical Studies 17). Scholars Press, Atlanta 2007. Pp. x + 220. Price: $24.95 paperback. ISBN: 987-58983-227-0. The lectures here transcribed and edited were delivered at the Sixth Biennial Colloquium of the International Institute for Secular Humanistic Judaism, Detroit, in 2005 under the title ‘Digging for Truth’. The topic was very aptly chosen, for recent Israeli archaeology has been at the vanguard of a battle to establish a secular Jewish scientific history of ancient Israel and Judah and to rescue it from on the one hand a version of Zionist archaeology that seeks to colonize the entire history of the land and on the other hand a near-fundamentalist pseudo-science of biblical archaeology 593
REVIEWS
that owes little more to scientific knowledge than creationism does to biology, but manages to capture too much of the popular media with its sensationalist antics. Israel Finkelstein of Tel Aviv and Amihai Mazar of Jerusalem are two of Israel’s most energetic and eloquent archaeologists, and their differences over the interpretation of Israel’s origins in Canaan have been well publicized, if understood only in general terms. These lectures, with the helpful summaries by Brian Schmidt, allow them to explain in some detail their differing reconstructions but rather more impressively to demonstrate their deep agreement over how archaeology can and should be used to reveal something of the true nature and history of ancient Israel and Judah. The lectures begin with the topic of ‘bible and archaeology’ in which both scholars reject biblical archaeology’s aim to confirm the biblical record, but also a ‘minimalist’ position that, as they see it, denies the Bible any historicity. Having tried to secure the necessary claim of a middle-of-the-road position they then consider the biblical pre-settlement traditions. These do not relate strictly to a historical period, but discussion of the stories of patriarchs, exodus, wandering and conquest is important in establishing not only the almost complete lack of archaeological verification but, perhaps more importantly, the need to discover the real meaning of these stories and what they can tell us about ancient Judaeans and Israelites. Both agree that the stories reflect some traditions or memories but Finkelstein ventures further in suggesting specific scenarios for the creation of certain figures and episodes within a critically reconstructed history of Judah. The next topic, the earliest settlement in Canaan, again elicits no fundamental disagreement: both see it as part of a cyclic phenomenon of nomad and sedentary existence; both also see the question of ethnicity among these groups as complex but problematic. Interestingly, neither invokes the ‘urban vs. rural’ duality that has characterized some treatments of the culture of ‘early Israel’, nor assumes any significant interaction between these ‘pre-Israelites’ and the remaining urban centres of Canaan. In the next topic, the tenth century, the differences emerge more sharply. Both recognize David and Solomon as historical figures, though for Finkelstein they are minor chiefs. Mazar accepts the existence of a major political force such as would have invited Shishak’s campaign villages. There are disagreements about the chronology of the Iron I-IIa transition and the dating of the controversial ‘stepped-stone’ structure in Jerusalem. On the next topic, the separate kingdoms, Finkelstein offers his preferred opponent for Shishak’s campaign in the central highlands — the kingdom attributed to Saul, after which, some time later, a more secure kingdom was established on Samaria: Judah only emerged into independent statehood in the eighth century when Israel’s power was weakened by Aram and Assyria. For Mazar, Judah was a state — if weaker and poorer than Israel — already in the ninth century. A final section considers the implications of modern Israeli archaeology for Israelis and other Jews especially. Both Finkelstein and Mazar agree that the biblical stories and their identity-forming power are more important to modern Israel’s identity than any reconstructed history. This book is an excellent read for anyone wishing to see where the modern archaeology of Israel is really going, which challenges are being overcome and which lie in the future. In this reviewer’s opinion one of the major challenges is to equip a generation of scholars in both archaeology and biblical studies, which remain both very different and very technical disciplines, both needed in order to write the a history of ancient Israel and Judah that we need. The way is open for a very fruitful collaboration. doi: 10.1093/jss/fgq015 PHILIP R. DAVIES UNIVERSITY OF SHEFFIELD
594
REVIEWS
LESTER GRABBE (ed.), Israel in Transition: From Late Bronze II to Iron IIa (c. 1250850 B.C.E.) Volume 1. The Archaeology (Library of Hebrew Bible/Old Testament Studies 491; European Seminar in Historical Methodology). Continuum, London 2008. Pp. 239. Price: £65.00 hardback. ISBN: 978-0-567-02726-9. This book draws together contributions presented at an AHRC-funded conference at Hull University in 2006 on the history of Israel in the early Iron Age of the Southern Levant, a period historically associated with the emergence of ancient Israel. This book aims to give archaeological data a prominent place in this muchdebated historical issue. Despite the book’s explicit focus on ‘the archaeology’, the title immediately betrays a strong historical focus within many of the contributions (the volume is published within the ‘Library of Hebrew Bible/Old Testament Studies’ and the ‘European Seminar in Historical Methodology’). As a result, much of the book bypasses any debate over the formation of ancient Israel in the early Iron Age, or even the historical settlement/conquest narrative of the Hebrew Bible/ Old Testament, and instead assumes what it should be demonstrating — a link between the archaeological record and the biblical narrative. There are twelve papers within this book and over half of them are foremost concerned with the chronology of the early Iron Age (Finkelstein et al.; Franklin; Killebrew; Mazar; Sharon et al.; Ussishkin). This is clearly an important issue within Iron Age archaeology of the Southern Levant. However, the majority of these chronological discussions, while arguing over precision in scientific dating techniques and the chronological value of different cultural assemblages, do not address the way by which archaeologists correlate scientifically dated archaeological deposits with historical episodes. Chronology alone appears to be the means by which many scholars aim to establish, or deny, the historicity of the biblical narrative. However, no matter how precise the radiometric dating, radiocarbon dates can contribute little to the historical issue of the emergence of ancient Israel; scholars must first establish a link between ancient Israel and the archaeological record, if indeed this is possible. Scientific dates can only give us accurate temporal information and are unable to coordinate the intersection of historical and archaeological data. Consequently, the old circularity persists: the archaeological record is dated by scientific means, associated with an historical culture, and then used to prove the historical narrative as an accurate reflection of the archaeological record. Amongst the chronologicallyfocused papers is the contribution by Sharon et al., which demonstrates an awareness of how an assumed reliability of the historical narrative has over-determined interpretations of the archaeological record. The result is a discussion of important archaeological, rather than historical, issues in the pursuit of chronological precision for the early Iron Age in the Southern Levant and its implications for the wider Mediterranean region. In the process, these authors make almost no reference to historical events. While the results of their research has not provided the final, definitive word on chronology, it is this type of approach that will inevitably make the greatest contribution to an independent chronological framework for the eastern Mediterranean. Closely connected with the traditional interpretative framework, wherein material culture holds chronological, and therefore, historical implications, is the culturehistorical value supposedly inherent in many aspects of material culture. In other words, certain cultural objects are considered indicative of historical peoples (e.g. Philistines, Israelites). Many contributions within this volume (e.g. Mazar, Ussishkin inter alios) are indicative of this culture-historical framework. And while the contribution by Bunimovitz and Lederman also encompasses a culture-historical perspective, 595
REVIEWS
where Israelites and Philistines are assumed entities, this paper offers important insight into how material culture represents a living, functioning community. More specifically, the authors explore the complex and diverse media by which two different communities in direct contact choose to express individual and communal identity, and how these media may vary greatly to those employed within less confrontational contexts. Unfortunately, key conclusions are undermined by the use of historical/ political borders to define the area of cultural interaction (i.e. the historical border between Philistia and Judaea). In the end, Bunimovitz and Lederman are clearly aware of Barth’s work on ethnic boundaries (Ethnic Groups and Boundaries: The Social Organisation of Culture Difference, Bergen 1969), where the world is rarely divided up into social collectivities that correlate neatly into discrete cultures, yet they persist with the traditional categories of Philistine and Israelite as derived from a literal reading of documentary sources. Nevertheless, this paper is a positive move towards appreciating the complexity of early Iron Age society in the Southern Levant, wherein material culture acts as social agent in the construction of identities. The contribution by Steiner similarly begins with an historical question (was Judah a state at the beginning of the Iron II period?) and consults different aspects of archaeological data to answer this question with little reliance on historical data. Despite the strong historical focus of this volume, a number of contributions demonstrate the growing awareness of the difficulty in correlating history with the increasingly evident complexity of material culture; e.g. Knauf (see below), Nakhai (survey of archaeological data as evidence of social behaviour in Iron I rural communities — minimal historical discussion), Routledge (see below), and Sharon et al (see above). In particular, Knauf begins his paper with a brief discussion on the difference between art-historical ‘archaeology’ and anthropologically-inspired ‘archaeology’; only the latter can reconstruct past human societies. While Knauf’s language might be tongue-in-cheek (enjoyably so), he is clearly aware of how archaeologists frequently make the mistake of applying political time (historical narrative) to archaeological time (physical material culture). He does not advocate for archaeologists to work without texts (and why should we?), but only for them to deny texts any privilege in the reconstruction of the past. Routledge also appreciates the manner by which broad political narratives have over-determined interpretations of archaeological data. He suggests that there are many questions that cannot be resolved by more precise radiometric dating, or the application of a single model (or sequence) for the whole Levant, and instead advocates an approach that begins with ‘local’ analyses in order to appreciate significant complexity in the archaeological record: large regional models tend to homogenize material culture and therefore override small but significant variations in localized behaviour. Only once local behaviour is appreciated will we be able to ask the global question of ‘Levantine-wide interaction and social change’ (p. 173). While this volume does not provide decisive archaeological evidence for the emergence of ancient Israel (it is not clear whether the ‘Israel’ of the book’s title refers to a geographical region or an historical culture already present at the end of the Late Bronze Age!), it does raise some interesting questions regarding current approaches to this important period in the history of the Southern Levant. And after the overdue reconsideration of the archaeological data regarding ‘Israel in Transition’, we are left with one key question, the implications of which are immense: what kind of history will emerge if history is written on the basis of archaeology alone (Knauf, p. 84). doi: 10.1093/jss/fgq016
MATTHEW WHINCOP UNIVERSITY OF DURHAM
596
REVIEWS
DÖRTE BESTER, Körperbilder in den Psalmen (Forschungen zum Alten Testament 2 Reihe 24). Mohr Siebeck, Tübingen 2007. Pp. xii + 304. Price: €49.00 paperback. ISBN: 978-3-16-149361-4. The John Templeton Award for Theological Promise is conferred on the best postdoctoral young scholars globally for their first book on the topic of ‘God and Spirituality’. Dörte Bester was one of the twelve 2008 recipients. We therefore approach Körperbilder in den Psalmen with high expectations. The purpose of this monograph, which results from a doctoral dissertation at the University of Tübingen in 2006, is clear. ‘Body images’ play a central part in the visual discourse in psalmody, and Bester chooses Psalm 22 as one of the best examples of the extensive use of body language in personal prayer, both in the parts of the psalm which are request and complaint (vv. 2–11, vv. 12–22) and also in the final part which is thanksgiving and praise (vv. 23–32). Such a thesis was an ideal candidate for the Templeton Award, in that ‘bodily prayer’ is advocated as an essential articulation of faith and spirituality. The contents of the book are equally clear: the introduction (essentially descriptive and informative) examines the most significant anthropological assessments of ‘body language’ within the Old Testament in general and the Psalms in particular over the last century, referring to works in both German and English. This gives Bester some contextual understanding of the contours of the discussion of the Hebrew language about the body, which she examines both as ‘Körper’ and ‘Leib’, in the psalms. Part One, of some fifty pages, is a textual and literary-critical study of the three parts of Psalm 22. This is entirely exegetical, with lengthy footnotes utilizing other commentaries, examining the psalm as a whole, without special emphasis on its ‘body imagery’. Part Two, the rest of the book, is of some one hundred and sixty pages: this is basically Bester’s own exegesis of the three parts of the Psalm, again heavily annotated, with detailed references to other relevant psalms and texts from, especially, Job and Second Isaiah. Particular attention is given to verses 7–8, 10–11, 15–19 and 27 (EV) where the ‘body imagery’ is the most explicit. Bester concludes with some apposite observations about the silence of God at the beginning of the psalm leading to the silence of the psalmist in the heart of it, and the importance of the community for restoring the ‘körperliches Ich’ back into the presence of God. So were this reviewer’s expectations satisfied? Only in part. The introduction, on the ‘history of the problem’, which assesses the questions raised by other scholars about a Hebraic understanding of ‘self’ and ‘community’, did not seem properly integrated with the commentary on the psalm itself. Issues were raised in the introduction which were not answered within the exegesis. Little was made, for example, of the relationship between the literal and the metaphorical in the use of body language in the psalms. If the literal is so important, as the introduction implied, what do we make of the process of composition of Psalm 22? If the psalmist was so physically debilitated, who would have written the psalm on his behalf, and for what purpose? And what of the editing of the psalm — not only the relationship between the two laments, but also the relationship of the two laments to the final thanksgiving? Does Bester intend us to look primarily at the psalm, or the psalmist(s)? And what of the placing of the psalm within this particular collection — between Psalms 21 and 23? What are we to make of the associations between this psalm and the formulaic language in other psalms? For example, although Bester refers to the several links with Psalms 17 and 18 and 27, the implications of these correspondences are never satisfactorily explained. 597
REVIEWS
The research itself and the resultant exegesis cannot be faulted: it is thorough and comprehensive. But this reviewer came away after each section asking ‘so what?’ or ‘why?’. (Ironically, ‘Warum?’ or ‘Wozu?’ are the very questions which Bester herself spends some twenty-five pages discussing in her exegesis of the first verse of the psalm.) The language of creation and incarnation as an inspiration for the faith of the individual and their place in the community is writ large on every page: but again and again so many opportunities for developing these perceptions further were missed. Nothing was made of the later reception of this psalm: it would be intriguing to have had some discussion as to why it was used so infrequently in later Jewish tradition compared with its prominent place in the passion narratives in early Christian tradition. This said, Bester does offer several important insights. This reviewer found especially helpful her overview of the structure of the psalm as a whole (p. 83) where she illustrates the interplay of various linguistic motifs which recur at significant places within the psalm. One is the use of אליin verses 2 and 11;1 taken together these reveal the suppliant’s continued dependence on God in the bitter experience of his absence. A second, which is found after the most vivid pictures of ‘death-inthe-midst-of-life’ in the second complaint, is the trust that the psalmist will be heard in v. 22 ( )עניתניwhich answers positively the fear of not being heard ( )ולא תענהin v. 3. Another is the repetition of ‘the heart’, which is first described as being ‘like wax’ ( לבי כדונגas in v.15), at perhaps the lowest point of the psalm, whilst later it is ‘the heart’ which will ‘praise God for ever’ ( יחי לבבכם לעדas in v. 27) as the psalm rises to its highest point of praise and trust in God. Yet another is the psalmist’s own final praise in v. 26 ( )תהלתיwhich corresponds with the recall of the ancient praises of Israel ( תהלות ישראלas in v. 4). The overall impression, therefore, is of a connected unity in a psalm which is so full of diverse experiences of intense suffering and intense joy. Only one of the above linguistic connections is actually about ‘body language’, however. But Bester does offer a valuable insight about the way in which the body images do serve to hold the psalm together, whether in the experience of the absence of God or of His presence, and whether in the experience of isolation and being threatened by enemies or of restoration and being upheld by the worshipping community. The most obvious examples of such body images occurring in all three parts of the psalm are ‘from the belly’ ( )מבטןand ‘my mother’s breasts’ ( )על שדי אמיboth in v.10, and ‘from the womb’ ( )מרחםand ‘my mother’s belly’ ( )מבטן אמיin v.11; similarly, ‘all my bones’ ( ) כל עצמותיand ‘my heart’ ( )לביand ‘my bowels’ ()מעי occur in the second lament, in 15, as also does ‘my strength’ ( )כחיand ‘my tongue’ ( )לשוניin v.16 and ‘my hands and my feet’ ( )ידי ורגליin v. 17. Two examples from the final thanksgiving are ‘my soul’ ( )נפשיand ‘my life’ ( )יחידתיin v. 21. Each of these terms Bester assesses in lengthy anatomical excurses to illustrate their significance within the psalm as a whole. A final important point is the way in which Bester shows how the language of the psalm echoes several times the language in Job (interestingly in this particular psalm, more so than in Jeremiah). The context of Job shows us that the language is indeed literal, and so the implications for the interpretation of Psalm 22 suggest physical suffering as well (despite the attendant problems of composition, if so).
1
Hebrew versification is given here and in the following paragraph.
598
REVIEWS
This then would justify Bester’s argument that life in all its extreme physical forms — sickness and health, plenty and want, with or without an experience of God — can be expressed and ‘made good’ through the words of the psalms. And perhaps it was this thesis above all which encouraged the twenty-five judges to nominate Körperbilder in den Psalmen as deserving the Templeton Award for Theological Promise. The promise is undoubtedly there: hopefully Bester’s next publication will explore some of the other profound questions which arise from a work of this nature. doi: 10.1093/jss/fgq017
SUE GILLINGHAM WORCESTER COLLEGE, UNIVERSITY OF OXFORD
STEPHANIE DALLEY, Esther’s Revenge at Susa: From Sennacherib to Ahasuerus. Oxford University Press, Oxford 2007. Pp. xv + 262. Price: £50.00 hardback. ISBN: 978-0-19-921663-5. Stephanie Dalley’s contention in this book is that the biblical story of Esther is descended from a putative Assyrian tale. This is made plausible by indications that the Israelites deported to Assyria by Tiglath-pileser III and Sargon II did not die out but remained a significant minority throughout the life of the Neo-Assyrian empire and beyond. Successive Assyrian kings had wives with West Semitic (possibly Hebrew) names, and similar names occur in other cuneiform documents. An Israelite chariot regiment served in the Assyrian army. Sargon’s allowing a priest to return to Bethel and the Samaritan tradition that he took their sacred books to Assyria suggest that he was at least tolerant of Israelite religion, perhaps even sympathetic. The fact that the stories of Tobit and Judith are given an Assyrian setting, although written long after the Assyrian empire came to an end, suggests a continuing Jewish interest in Assyrian traditions. The legendary Assyrian sage Ahiqar even becomes a Jew in Tobit. The book of Esther opens with the repudiation of queen Vashti and her replacement with the Jewish Esther. This incident may reflect the replacement of Sennacherib’s first wife by the forceful (Hebrew?) Naqia, though whether because the first wife died or was repudiated is not known. It is not only Esther’s name that is reminiscent of the goddess Ishtar, who had major sanctuaries at Nineveh and Arbela. Her grooming recalls the perfuming of Ishtar’s statue. She becomes queen in the tenth month, Tebet. This is the first of a number of dates inserted in the story, a feature that recalls the cultic calendar of Ishtar at Nineveh. Tebet was the month of her akitu festival, which included the sacred marriage ceremony, feasting and the fixing of destinies for the ensuing year, by the casting of lots. This method is used by Haman to determine the most auspicious day for the elimination of the Jewish element in the population. Haman’s plan is foiled by Mordecai and Esther, and the destined day becomes the occasion for the Jews’ revenge on their enemies in Susa and the provinces. Revenge was the motive for the sack of Babylon by Sennacherib in 689 BCE and by Ashurbanipal in 648. The location of the Esther story in Susa is significant, because Ashurbanipal followed up his sack of Babylon with the sack of Susa, then the Elamite capital, and the surrounding countryside. The name of the Jewish festival commemorating these events, Purim, derives from puru, the Assyrian word for ‘lot’, later replaced by the Babylonian iqsu. The ways in which Purim is celebrated also reflect Ishtar’s akitu festival, so that it is likely that it was first celebrated by the descendants of Israelites who had lived in Assyria since 721 BCE. 599
REVIEWS
Dr Dalley’s argument raises two questions: How good is the evidence for an Assyrian story behind the book of Esther? and, How far does it enrich our understanding of the book? One obstacle to accepting the thesis stands out: nowhere does Dr Dalley outline the story she postulates. It is described vaguely as ‘a story based on Assyrian deeds’, a story ‘rewritten to apply to a new situation’ with the king’s name changed and ‘a new group of deportees’ representing ‘the stereotypical foreigners’. ‘An early version of the story … would have been composed in Assyrian Aramaic, in the reign of Ashurbanipal, soon after the sack of Susa’. Since the main characters are Esther and Mordecai it would have been a ‘story of essentially pagan deities and acceptance of the feast of Purim’. Dr Dalley makes much of the way the Assyrian kings attribute success in their campaigns to the aid and intervention of the gods and the adaptation of their myths to reflect historical events in cultic calendars such as the Ordeal of Marduk, but the reverse process, how a myth is changed into ‘a pseudo-historical tale of mortals’, how Haman, Esther and Mordecai shift from ‘deities who controlled fate to mortals who could change it’, is not illuminated. The title of the book suggests that the story would have interpreted Ashurbanipal’s sack of Susa as an act of revenge by Ishtar. This was necessary, according to Dr Dalley, because ‘revenge and retaliation were taboo for mortals, who were polluted by taking their own revenge; such punishments belonged to the gods.’ She does not say what her authority for this idea is: in fact she quotes Ashurbanipal (p. 51) as declaring quite openly that both he and Sennacherib had taken their revenge on Babylon, and produces no clear evidence that Ishtar was associated particularly with the idea of revenge. The suggestion that Esther’s replacement of Vashti reflects Naqia’s replacement of Tashmetu-sharrat is also difficult. To fit Dr Dalley’s scheme the latter would need to have been mythologized as Ishtar’s replacement of another goddess before being retold as a story about human queens. In any case not much is known about Naqia’s character and actions. The picture is built up by inference and by the suggestion that the Semiramis mentioned by Greek historians combines some of her deeds with those of the earlier Sammu-rammat. The demonstration that the dates in Esther are vestiges of the Mesopotamian calendar and coincide with risings of the planet Venus is convincing, but of course only suggests that astronomical phenomena contribute to the background of the Esther story. It does nothing to connect the story to historical events in eighth century Assyria. Dr Dalley shows that a number of words in the Hebrew text have cognates in the Assyrian dialect of Akkadian, and argues that a storyteller writing a new work in the Persian period would be unlikely to choose words from a dialect that had fallen into disuse since the fall of Nineveh in 612 BCE. The present text must therefore be a reworking of the Assyrian story, and any Persian elements must be secondary. However, the Persian elements are not limited to a handful of names and loan words, but provide a whole setting. If this is the result of reworking rather than the original setting of the story, then the case for reworking needs to be spelt out in detail and not simply asserted. The regrettable but inevitable verdict must be that Stephanie Dalley has not produced a convincing case for an Assyrian origin for the book of Esther. Various features in it have reminded her of similar features in Assyrian annals, myths, calendars and language, but she has not succeeded in weaving them into a plausible picture of the precursor of the biblical book. As a result our understanding of the 600
REVIEWS
book has not been significantly enriched. The story is essentially one that embodies what has so often been the experience of the Jews in the Dispersion and gives the reason for the celebration of Purim as a Jewish festival. Dr Dalley seems to lose sight of this almost completely. I suspect that one reason for this is that the book seems to have been compiled from articles she has published that were only marginally, if at all, about Esther. The result is that the argument is held up by discussions of such topics as ‘Who killed Sennacherib?’, the Hanging Gardens, Hezekiah’s relations with Assyria, Aramaic and Egyptian versions of Assyrian Court Narratives, etc. Topics relevant to the argument are addressed piecemeal. The summary given at the beginning of this review took some time to compile; the topic of Naqia, for example, is broached in at least half-a-dozen places. Fortunately there is a reasonably good index. The book also suffers from some uncertainty about the intended readership. The original articles were clearly intended for fellow academics, since the footnotes cite articles in journals and Festschrifts and ancient sources in scholarly editions. There are no references to accessible translations of the myths and inscriptions discussed, not even to her own Myths from Mesopotamia. On the other hand there is an attempt to address a wider readership: there is a glossary, describing Aramaic and Hebrew, Septuagint, Targums and Talmud (rather inaccurately); and occasional excursions into story-book style: ‘[Sennacherib] would have played as a boy with sons and daughters of foreign rulers who were sent to the Assyrian court to be educated.’ doi: 10.1093/jss/fgq018
ROGER TOMES UNIVERSITY OF MANCHESTER
EHUD NETZER, The Architecture of Herod, the Great Builder (Texts and Studies in Ancient Judaism 117). Mohr Siebeck, Tübingen 2006. Pp. xiv + 443. Price: €129.00. ISBN: 3-16-148570-X. This is a useful compendium of the building operations west of the Jordan River undertaken during the rule of Herod the Great (the last third of the first century BCE), useful in having them brought together, but offering few new insights. The book contains 17 chapters, plus epilogue, five appendices, 60 line drawings in text including isometric reconstructions, and 28 pages of half tones. It begins with a short history of Herod’s reign (Herod was born c. 73/2 BCE), and the following chapters, arranged by site, all commence with a summary site history. Netzer attempts to assess Herod’s personal contribution to his building programmes. The result is sometimes rather overweight as detailed description is followed by discussion of the descriptions, and such summaries, however detailed, cannot supersede the fuller, final reports. Inevitably one is left looking for further information — is the important labrum from Herodium, with the head of Silenus, which features twice in the text as the only figurative image known from Herod’s decorative schemes, actually published? There is sadly no reference. And one would like to see drawings, not just photographs, of the ‘flower pots’ from Jericho, the tubuli from the bath houses etc. The photographs of architectural decoration only serve to complement Netzer’s note that Herod’s buildings were not artistically adorned, more provincial, in everything but efficiency and engineering. In the latter respect, the platform and the royal stoa of the Temple in Jerusalem and the harbour at Caesarea, claim pride of place. Netzer concludes that Herod was involved in site selection and function, but had no real interest in decoration. Herod’s funding of these projects may have been drawn from his private wealth, probably including Arabian estates inherited from his mother, 601
REVIEWS
but mainly depended on an efficient system of taxation of his subjects and the customs dues of a facilitated commerce. The photographs of architectural decoration are generally tiny, the use of space and layout of them is poor. Insufficient care is given to scales, orientation and keys to the drawings, although it is very illuminating to have so many of the buildings presented at the same scale for comparison. There are careless typographical errors (‘Cleopatra as a potential thread’; and especially with authors’ names, Barkley instead of Barclay, Dikki instead of Dickie, Condor instead of Conder). The order of site presentation is not particularly relevant, as at some sites, like Jericho, Herod’s projects continued over many years, but prominence is given in the first seven chapters to the many sites at which Netzer has directed excavations. There are briefer summaries of another 14 or so sites, but those east of the Jordan receive only very brief attention, and the location map (p. 9) has the site of Betharamptha/Rama (see also below) badly misplaced to the north. The map also indicates that Hebron was not, as claimed, the most southerly of Herod’s projects, even within his kingdom. There is no coherent statement of the impact of Judaism on the architecture, but it impinges in a number of ways. First is the ruling forbidding the placing of burials in proximity to living quarters, which affects the interpretation of the structures at Herodium and Hebron; another is the observed lack of service facilities, especially of toilets, for which ritual decrees burial of waste outside habitation areas, but also of kitchens, staircases and the like. Netzer also notes the deviations from the Roman style of bath house introduced by Herod, with larger caldaria, and smaller frigidaria usually furnished with stepped pools for ritual cleanliness. The impact of religion is especially noted in Netzer’s reiteration of Herod’s focus on the magnificent royal stoa on the Temple platform, which he regards as linked to Herod’s status as a nonJew, who thus was prohibited entry to the main courts. Elsewhere in Jerusalem, the description of Herod’s work relies heavily on that of Josephus. Occasionally one feels the pressure on an author attempting to deduce and write about the unknown role of Herod in the planning of these building operations: such sentences as ‘probably a street leading to the east gate’ at Caesarea could have been left unwritten. There are other questions and omissions however. Netzer (and others) call the line of fortresses from Alexandrium to Masada ‘desert fortresses’. I would dispute this on two grounds. The first is the location — all except Masada and Herodium (exceptional for other reasons) actually guard the rich estates of the Jordan Valley and the routes to them — they are not ‘desert’ fortresses, but secure control points on dry hill tops above very fertile land, developed for this purpose in the first instance by the Hasmoneans. Secondly, as he shows on the map, Herod controlled much of Peraea, but the only points Netzer mentions across the Jordan are ar-Rama, Heshbon and Machaerus. But other fortresses guard the fertile estates on the east side of the Jordan Valley, and these are ignored (those at Barakat, Habbasa and the older sites of Amathous, Essa and Ragaba further north). On p. 181, n. 15, he denies the existence of the second Herodium across Jordan mentioned by Josephus; but many of the fortresses and sites on the east side of the Jordan also have Hasmonean-style origins and Hellenistic predecessors; have very similar plans to those on the west; and, like Machaerus, are even more relevant not just as controlling fertile estates and the routes and water sources which feed them, but for their role as frontier fortresses in the on-going contemporary politics involving the ‘Arabs’ and the Nabataeans. Netzer is more directly concerned with sites in ‘the Land of Israel’, and the bibliography reflects this inward-looking attitude. 602
REVIEWS
The finds which have emerged since this volume was written are amongst archaeology’s excitements; the main item is Netzer’s claim to have discovered the destroyed tomb of Herod on the flanks of the hill at Herodium, rather than in the previously proposed funerary complex in the lower ‘town’ at that site. While not everyone is convinced that either find is actually the tomb, the discovery would clearly have radically changed much of the text. Another later find is the superb stuccoed ceiling in the reception palace (‘Great Temple’) at Petra which has a direct bearing on Herodian practices, and even more recently, McKenzie’s book on the role of the architecture of Alexandria, which Netzer notes, like Antioch, undoubtedly had an important but yet unknown place in the architectural history of Herod’s kingdom. doi: 10.1093/jss/fgq019
KAY PRAG UNIVERSITY OF MANCHESTER
KAREL JONGELING, Handbook of Neo-Punic Inscriptions. Mohr Siebeck, Tübingen 2008. Pp. xxii + 435. Price: €139.00. ISBN: 978-3-16-149303-4. The Handbook of Neo-Punic Inscriptions is an important tool for further research and study, prepared by a skilled expert in Neo-Punic script, active in this field for almost thirty years. Seven hundred and fifty inscriptions, without counting the Neo-Punic legends on coins, are presented with a translation (if possible), a short commentary, and bibliographical information. Line drawings, made by the author, are added for several texts, while references to photographs and illustrations published elsewhere are given for other inscriptions. These very useful research tools may show immediately that some readings and interpretations are simply impossible and do not deserve to be discussed. The archaeological context of the findings is not indicated, the present whereabouts of the inscriptions is not reported, and no approximate dates are proposed. The texts are presented in a geographical frame: Egypt, Libya, Tunisia, Algeria, Morocco, Greece, Malta and Pantelleria, Sicily, Italy, Sardinia, Spain, Wales. El-Amruni is located in Libya (pp. 9–10), although the site lies in southern Tunisia, north of Remada. The running headline on p. 192 should be 4. Texts from Algeria instead of 3. Texts from Tunisia, and on p. 256 one should read 5. Texts from Morocco instead of 4. Texts from Algeria. In each section, the sites are listed in alphabetical order and the inscriptions are numbered separately for each of the 123 sites. This facilitates the addition of new inscriptions, which will certainly appear ‘in the near future’, as announced in 1963 for the awaited texts from Hr. Maktar, but this system often requires a recourse to the table of contents, as internal references, also in the vocabulary, are given only by naming the site and indicating the number of the inscription, for instance ‘Guelma N 35’. However, Guelaat bou Sba of the onomasticon (pp. 327, 357, 358, 379) is listed in the corpus under Qalat Abi s-Siba (pp. 249–52). The page of the volume and the line in longer inscriptions are never provided. The reviewer would therefore exhort the users of the Handbook to mention also the pages in their publications. The new numbering of the inscriptions, albeit necessary in the frame of the Handbook, will disturb old habits, whereas the useful concordances (pp. XXI–XXIII) do not include either the NP numbers of P. Schröder and Z.S. Harris, or the old Trip. numbers of G. Levi Della Vida. After a historical survey of Neo-Punic epigraphical studies, based on concrete examples (pp. 1–7), the author presents the corpus of inscriptions (pp. 8–289) and of coins with Neo-Punic legends (pp. 290–309). The onomasticon includes readable personal and divine names, as well as place names (pp. 313–80), while the vocabulary in general follows the consonantal form of the radicals (pp. 381–410), apparently 603
REVIEWS
leaving a few unresolved cases aside. The very useful and precise bibliography closes the volume (pp. 411–35). In order to stimulate research, the reviewer will comment on a few points, chosen among the many problems raised by these difficult texts. Some years ago, he had followed J. Hoftijzer’s reading of the Ain Zakkar inscription (N 1, p. 62 = KAI 136; cf. OLA 80, §36.35). It appears at present that some corrections are needed. However, the interpretation proposed in the Handbook on pp. 62–4 is not satisfactory in every point. The first word, read either bmrs or b}rs, must mean ‘at the request’. M}rs occurs in Job 17:11 with this approximate meaning, while }rs is the normal spelling of the root }rs, ‘to desire’. The r¨s¨ at the end of line 5 is a feminine derivative of the root rsy, ‘to have authority’ (DNWSI, pp. 1086–7), attested in Phoenician by the noun rs}t (*rasa}ut) at Karatepe (Phu/A III, 6; PhSt/C III, 20–IV, 1): rs}t n¨mt w¨z }dr, ‘a felicitous rule and a majestic strength’. This would give us the following translation: ‘At the request of Adyat, the illustrious priestess, (who) died at the age of 77 and had been superior for 18 years, Masharat erected (it)’. The central part of the inscription consists of two asyndetic relative clauses (OLA 80, §57.3) or forms a kind of parenthesis. Sp¨r in the inscription of Ras al-Îaddagia (Breviglieri N 1 = Trip. 6), dedicated to Amon, is translated by ‘bull’ (p. 11), although Akkadian sapparu is a word for ‘ram’, from which also derives Hebrew sopar, ‘ram’s horn’. Since Amon, to be distinguished decidedly from Baal Hamon, is the Egyptian god with ram’s horns (cf. Studia Phoenicia IV, pp. 307 ff.), it is evident that m}s }lm sp¨r st means ‘this statue of the Ram god’. The Egyptian roots of the important and influential clan ™bÌpy, ‘Apis is good’, are usually neglected in the study of Leptis Magna texts. Now, the title h-rds of Macer in the building inscription Trip. 2 (Labdah N 2, pp. 14–15) is probably borrowed from Egyptian rw∂w > rwdw (n) s3, ‘supervisor of troops’ or ‘of a gang’ of workmen. Also bry}t, the last word of the inscription, may be an Egyptian loanword, viz. bryt, ‘gang of workwomen’. The preposition b should then mean ‘with’. Such borrowings may be older than the Ptolemaic period. Hebrew b¢riyyot means ‘creatures’ and is unlikely to have any semantic relation with bry}t. The comparison of Trip. 4 with Trip. 5, a trilingual dedication (Labdah N 4 and N 5, pp. 16–17), shows that the stone carver of both inscriptions (or the copyist?) confused ¨ayin with daleth and then provided the lamed of N 5 with a curl, making it similar to ¨l, possibly in order to correct the mistaken d. The photograph provided in G. Levi Della Vida and M.G. Amadasi Guzzo, Iscrizioni puniche della Tripolitania, Roma 1987, Pl. IV, 13, still allows reading [hr]p} instead of hrb} at the end of Neo-Punic inscription N 5, contrary to the drawing (ibid., Fig. 6). The text should thus be understood as follows: B¨l qrt h-Mqrty Qld}y h-rp}, ‘(to the) Baal of the city Meqrati, Clodius, the doctor’. The correct spelling of his name would have been Ql}dy. The Neo-Punic line should have been introduced by the preposition l, which does not need any counterpart in Latin and Greek, which use the dative. The theonym Boncar/BÉnxar (cf. ILPBardo 185) of the Latin and Greek versions precludes the interpretation ‘Citizen of the city Meqrati, Clodius, the doctor’. The town Megérqi is mentioned in the Stadiasmus maris magni §97–8, and the Itinerarium Antoninianum 62:3 refers to Megradi, villa Aniciorum, 25 miles east of Oea / Tripoli (cf. Latomus 64 [2005], pp. 406–7). The Latin and Greek parallel inscriptions call the city Mecrasi / Mexrasi, confirming the dialectal change t > s in spoken Late Punic dialects (OLA 127, p. 351). The place name is missing in the onomasticon. The word mynkd, that appears three times in Trip. 27 (Labdah N 13), once as translation of imperator, is rightly connected with the frequent MNKD(H/}) in 604
REVIEWS
Libyco-Berber texts (p. 22), but the Vocabulary qualifies it as ‘possibly a Libyan loan word’ (p. 396). A doubt would be justified if we only had the Tuareg a-mnukal, ‘king’, to rely on, although the change d > l occurs in several languages and is attested also in Berber dialects (S. Chaker, Linguistique berbère, Paris-Louvain 1995, pp. 194–5). Instead, the proposed etymology (p. 22) is less evident, because the basic meaning of Tuareg nkd is ‘to be careful’, ‘to pay attention’, not ‘to go in front’ as a leader. The larger ‘Afroasiatic' background with Akkadian naqidu and Hebrew nagid seems to be a better choice. The Latin name Roma, used with the article at Leptis Magna (Trip. 28 = N 14, p. 23), can hardly be assimilated to a common noun (p. 373). In early Roman times, the article occurs with several Punic place names (}-slyt, }-sks, }-lpqy, h-mqrty, h-mkt¨r) and this usage is attested also in the Levant (OLA 153, p. 278). The beginning of the second line in the Latin inscription IRT 347, corresponding to line [1] of the Neo-Punic version in Trip. 32 (Labdah N 19), should be translated ‘Tiberius Claudius from the Quirina tribe Sestius’ (p. 28), as Quirina is a qualification of tribus, not of Tiberius Claudius. The name of the chief buried at Bordj Helal N 1 (pp. 68–9, 336) should probably be corrected to Ygw¨kn}, since the deeply engraved St Andrew’s cross of the aleph is recognizable on the photograph thanks to its reduced size (Semitica 27 [1977], Pl. IX). This reading is confirmed by the Libyan inscription and by the latter’s duplicate (Africa 9 [1985], pp. 7–11), since the final sign of the name is H/}. The Neo-Punic plural }bnm }l}, ‘these steles’, refers to the two steles, but the author was apparently unaware of the publication of the duplicate. The inscription from Ghzaizya N 1 (p. 79) is given according to its reading by M. Ghaki, but the photographs published in REPPAL 7–8 (1992–3), p. 174: 1–2, albeit not very clear, show beyond any doubt that this decipherment is not correct. The author does not attempt any independent reading, perhaps awaiting better photographs or a carefully made drawing. The alleged dittography of n in NP 43, 1 (Hr. Maktar N 9, pp. 94–5) might instead confirm the antiquity of the Libyco-Berber genitival preposition n, always followed by the ergative, which is characterized by the u-prefix. The syntagm n-}-qyds would then comprise this preposition, the u-prefix indicated by aleph, as expected, and the noun qides, ‘shrine’. The phrase would mean: ‘to the Baal Hamon of the shrine’ (cf. l}ln }qds in KAI 104, 1). This syntagm is paralleled, for instance, by the Libyan epitaph ZD} / U-DBR / N-KLM} (RIL 642), ‘Zado, son of Dabar, from Calama (Guelma)’. The phrase hnkt ¨bnt, occurring in NP 66, 4 and 67, 4 (Hr. Maktar N 32 and N 33, pp. 103–4), means ‘Here is she bewailed’. The form ¨bnt is a feminine passive participle qal of }bl, ‘to mourn for the death’, with the change l > n, well attested in Phoenician and Punic (PPG3, p. 30, §56), as in other Semitic languages. Weeping for the death of someone was an important element of the Libyco-Berber mourning ritual, often expressed in Libyan funerary inscriptions (K. Naït-Zerrad [ed.], Mémorial Werner Vycichl, Paris 2002, pp. 292–9). The masculine passive participle occurs at Hr. Merah (Suo) N 1 (p. 159), line 2: ws Nymm }bn z m}¨ ‘w’, ‘… and of Nimam. Bewailed is this one, who lived…’. In line 3, s¨nwt ¨ms w ¨ms should not be interpreted as ‘fifty-five years’. The numeral ‘five’ signifies a hand (five fingers) and, as such, is the basic unit of a quinary system, employed occasionally in Libyco-Berber; cf. Ju.N. Zavadovskij, in A. Caquot and D. Cohen (eds), Actes du premier congrès international de linguistique sémitique et chamito-sémitique, The Hague 1974, p. 108. Nimam died thus when he was ten years old. He was probably the son or the grandson of Nabo (line 1), who ‘lived sixty years’ (line 4). The niphal of the same 605
REVIEWS
verb }bl occurs in two funerary inscriptions from Kef Bezioun (N 1 and N 2, p. 245). In one epitaph, the 3rd pers. fem. sing. is used: hnkt n¨bn¨ (line 4), ‘Here is she bewailed’. In the second one, we find the 3rd pers. masc. sing: hnkt n¨bn (lines 4–5), ‘Here is he bewailed’. A new entry }bl or }bn II is needed in the vocabulary instead of ¨bn. The form }bl without the change l > n is attested in Akkadian, Aramaic and Hebrew. The Neo-Punic votive inscription NP 10 of the Leiden Museum, Inv. H. III. S. 1 (OMRO 44 [1963], pp. 93–4, Pl. XXIX, 1), of uncertain provenance, is classified as Hr. Maktar N 40 (p. 107). The author adds [bn] before the patronymic, but the inscription uses the Libyan sign M, which means ‘son’ (mmi) in Berber dialects. For Hr. Maktar N 133 and N 134 (pp. 144–5), the author simply reproduces the transcriptions of M. Ghaki without correcting the errors. The word ‘stele’ is written hbn, as shown by the photograph, not }bn. The patronymic begins with }m, not ¨m, and one can read n/t¨ in line 4. N 134 is illegible on the photograph. A similar approach to the original publication can be observed in the case of Ain Barchouch N 1 (p. 62), of Hr. Oum Guerguer N 1 (p. 160), and of the two Maghraoua steles (p. 167). The photographs of the Maghraoua steles (L’Africa Romana XII, Vol. III, Pl. IV, 11 and 12), not mentioned by the Author, clearly show †¨n} }bn, not †¨n[…]¨bn, for N 1: there is no lacuna between lines 1 and 2. Several words appear in line 2 of N 2, but the author only indicates ¨l¨, following the incomplete deciphering of M. Ghaki. The photograph of Hr. Oum Guerguer N 1 in 30 ans au service du patrimoine, Tunis 1986, p. 131, allows reading †¨n} }b/n z l-Ymrr bn / P¨s ?¨ ?} / ¨ks…, ‘This stele was erected for Yamrur, the son of …’. Iamrur is attested in CIL VIII, 37 = 11050, and Ymrr is found in El-Hofra 132, 3 and in Trip. 38, 2 (Wadi el-Amud N 1, pp. 58–9). Authors dealing with Neo-Punic inscriptions seem to forget sometimes that Massinissa invited a number of Greek artisans to Numidia and that they settled in the country. It is not surprising therefore that the Greek names of their descendants occur in Neo-Punic funerary inscriptions of the first century BCE or CE. This seems to be the case of ¨wrb¨[t.] at Mididi, recorded in RÉS 167 (Hr. Meded N 1, p. 145). The name should possibly be read Eûrúbatov, Eûrúbotov, Eûrúbatav, EûrubÉtav or Eûrubátjv, all attested in Hellenistic times (LGPN I, p. 183c; II, p. 182a; IIIA, p. 174b; IIIB, p. 165a). The trace of the third letter of ¨w¨ in line 2 can be interpreted also as the upper right curl of an aleph, instead of being an ¨ayin. Concerning the funerary inscriptions of Hr. Meded N 27 and N 28 (pp. 154–5), one should know that excellent photographs have been published (upside down) in the Bulletin des Travaux de l’Institut National d’Archéologie et d’Art, Tunis = BTINAAT 3 (1989), pp. 62 and 63. The reading by A. Ferjaoui is correct, except for the patronymic K¨ksn in N 27, 2, that should be read K¨khn, as noticed already by F. Vattioni. This name might correspond to Caggun (acc.) in Corippus, Iohannis VIII, 418, and to KNKN in RIL 627, if we assume an assimilation nk > kk. The patronymic Yms†n in N 28, 3 is spelled Imastan by Corippus, Iohannis VIII, 480, and Ymstn in a monumental Punic inscription from Volubilis, Morocco (L. Galand, J.-G. Février and G. Vajda, Inscriptions antiques du Maroc, Paris 1966, p. 91 and Pl. II, 4, line 3). The last line of N 28 reads s¨nt sysm, ‘sixty years’. Contrary to the author’s statements, Ferjaoui does not propose the reading ysm and he does not explain it as a lapsus for ssm, but there are printing mistakes in BTINAAT. One can manage to correct them by reading Ferjaoui’s comments. The eight fragments of Neo-Punic inscribed votive and funerary steles from Mididi, published by A. Ferjaoui in REPPAL 12 (2002), pp. 65–72, are still missing in the corpus. They should have been classified as N 29–N 36. 606
REVIEWS
The references of El Kef N 1 and N 2 (p. 161) to GHAKI 1998 are incorrect: one should read GHAKI 2002. In N 2, line 2, lmlk} or lmlk is the infin. constr. with or without a proleptic suffix, as in the inscription of Eshmunazor, line 1, and in 2 Kgs 24:12. The author’s transcription of Teboursouk N 4 on p. 177 is not entirely correct, as shown by the excellent photograph published by M.H. Fantar, Carthage, les lettres et les arts, Tunis 1991, p. 66. The reviewer reads there: 1) bym nm wb¨rk ¨bnt †¨n} 2) lnd¨r }nmm slmy pgny 3) rg¨†} w¨brz} bn st 4) y
‘On a pleasant and blessed day a stele was erected as the votive offering of the pleased ones. Performed it, fulfilled it Rogatus and Auruza?, the sons of Sittius.’
The form pgny is the qal perf. 3rd pers. masc. plur. with the suffix of the 3rd pers. sing. and a paragogic nun, attested sometimes in Hebrew (Deut. 8:3, 16; Isa. 26:16) and often in Phoenician after a long vowel with the suffix of the 3rd pers. plur. (-nm). We thus assume a vocalization *paguniu. Since Punic b may indicate an articulated w (PPG3, §38a, pp. 22–3), while a relation ab/al seems to exist in Libyco-Berber names (K. Jongeling, Names in Neo-Punic Inscriptions, Groningen 1983, pp. 90–1), ¨brz} can correspond to Alurusa (CIL VIII, 11308; 14402) and to IURZ (RIL 29). On the other hand, notwithstanding the length of the downward tail, the r might be a d, which would give the Semitic name Abdazus (CIL VIII, 24058). Thibar N 1 (p. 182) refers to BEN YOUNES 1990, but this article is missing in the bibliography. It was published in REPPAL 5 (1990), pp. 27–42. The photograph, not indicated by the author, can be seen there on p. 40. Line 2 of El-Hofra 273 (Constantine N 78, pp. 222, 318) must of course be read Ìmlkt bn }rs †n, as proposed by M.J. Fuentes Estañol, because the inscription would otherwise lack a verb. The ndr of line 1 is thus followed by an asyndetic relative clause: ‘…votive offering, (which) Himilco, son of Arish, erected (*†inne), because…’. The bilingual Latin and Neo-Punic inscription from Sulcis (S. Antioco N 3, p. 280 = KAI 172) contains the active participle qal h-pr†, which must have the same legal meaning as pr† in Talmudic Hebrew, i.e. ‘to specify’, ‘to decide’. As for Hispania 6 (Ibiza N 1 = M.J. Fuentes Estañol, CIFE 07.04), it is not clear what the author means when writing: ‘The earlier reading pl mgn … seems less probable if the drawing presented by Fuentes Estañol is correct’ (p. 285). There is a very clear photograph of the inscription in CIFE, p. 88, which excludes any doubt. The earlier reading, still accepted by W. Röllig in Fs. Ciasca, Rome 2002, p. 451, is certainly incorrect. The entry ¨zrm of the Vocabulary refers to nineteen Neo-Punic inscriptions with this word ‘of uncertain meaning’ (p. 400). The author assumes nevertheless that ‘it is a technical term indicating some sort of human sacrifice’ (p. 184), an opinion expressed recently also by C.R. Krahmalkov, who translates }zrm by ‘infant sacrificial victim’ (OLA 90, pp. 39–40). The most likely explanation relates this word to the root zrm, attested also in Ugaritic (KTU 1. 169:14), and to Hebrew zerem, ‘flow’, ‘shower’ (OLA 55, pp. 277–8), with a prosthetic a required here by the characteristic Late Punic change of the nominal pattern CaCC into CCaC (OLA 80, §24.10). Of course, the sacrificial Neo-Punic term does not refer to rain, but E.W. Lane, Arabic-English Lexicon, London-Edinburgh 1863–93 (reprint, 1980), gives for Arabic zarima such meanings as the following: ‘said of one’s urine … and of his flow of tears …’, and one could add by analogy: ‘and of one’s flow of blood’. *Azrem is thus a particular type of sacrifice, at least initially characterized by drawing of blood from the victim in order to pour it on the altar or stele. The usually 607
REVIEWS
following nouns }s or }st (pp. 384–5) can mean ‘male’ and ‘female’ as in Gen. 7:2, for human sacrifices may be replaced by slaughtered animals, also in Numidia, in the second–first centuries BCE. The simultaneous use of the sacrificial term mlk (p. 396; cf. Fs. Heltzer, Tel Aviv-Jaffa 1999, pp. 243–53) shows nevertheless that animal sacrifices could be substitution offerings in the case of *azrem as well. The indication of the sex of the victim, male or female, was important also in the Hebrew ritual (Exod. 12:5; Lev. 1:3, 10; 3:1, 6; etc.). The same technical phrase, misinterpreted by R.M. Kerr (Latino-Punic and Its Linguistic Environment, Leiden 2007, p. 146), occurs in a Latino-Punic funerary inscription from Roman Tripolitania (IRT 827), as correctly seen by C.R. Krahmalkov (OLA 90, p. 40). The text reads: ly-Mythi csi Nami ce-bal ysrim ysa u-mylthe, ‘Nami took care of Mitti(n), as supplier of a “bloodoffering” of a female and of its full yield’, probably its blood. The inscription contains the syntagm csi ly-Mythi, that parallels k¨s l-p¨l of Trip. 31, 4 (Labdah N 18, pp. 26–7, 393). The corresponding Arabic root is probably kasa, ‘to dress’, and mylthe must stand for ml}t’. We see therefore that the *azrem sacrifice occurred also in a funerary context. We can see it besides in the Neo-Punic inscription from Ksour Abd el-Melek N 5 (pp. 166–7), the interpretation of which is hampered by the separation of ¨zrm from other instances of the same term (p. 400) and by three unexplained words. The first one, s¨bw, is the passive participle qal masc. sing. (*zabu) of the root zbÌ, the spelling of which is paralleled at Guelma. The second concerned word is hn¨, which must be the 3rd pers. fem. sing. of Ìn}, ‘to find rest’, attested in Aramaic (DNWSI, p. 387). This Punic hapax legomenon comes as no surprise, considering our poor knowledge of the Phoenician-Punic vocabulary, while the Neo-Punic inscriptions seem even to reveal a more popular level of speech, shown by phonology, by the use e.g. of }bn in the sense ‘stele’, or by words not attested in earlier inscriptions. The third word, b¨yrst, was explained by F. Vattioni as preposition b + article ¨ + yrst, ‘inheritance’. This interpretation is regarded by the author as ‘far-fetched’, but the lack of any reference to the father, to a husband, or to a son, seems to confirm it and to point to the social status of a woman living in the inherited patrimony. This funerary inscription could then be translated as follows: ‘This stele was erected for Ianuaria. The ‘bloodoffering’ is sacrificed. She found rest. She lived 43 years in the patrimony’. Although the author does not wish to be entangled in onomastic discussions, one should observe that Lbw} in NP 124, 4 (Hr. Medeine N 1, p. 155) cannot mean ‘Libyan’ (p. 344). The w after b points at an archaic form of the word ‘lion’ (Fs. Heltzer, pp. 214–17). The name SÒp, attested by NP 97 at Constantine (N 8, p. 199) and occurring at least twenty-three times in Carthage is no ‘type of sacrifice’ (p. 377): such a word would not have been used as an anthroponym. It is a name borrowed ultimately from Egyptian sspw, ‘sphinx’ (cf. Mediterraneo Antico 8 [2005], pp. 217– 22). N¨pm} in Trip. 47d (Labdah N 57, p. 37; cf. p. 379, s. v. tpm}) is obviously Napamo (CIL VIII, 26944), i.e. N¨m-p¨m-}, ‘Pleasant is his phallus’. This name is attested frequently in Roman times and indirectly attests a popular phallic cult (cf. OLA 64, pp. 215–18). The ¨ayin, omitted by the author, is certain, and the second word of the inscription should probably be read p¨l, ‘has made’, instead of being an unknown proper name pyl (pp. 37, 368) with a very doubtful yod. Concerning the name ¨bdksr (p. 363), one should notice that it was pronounced Auchusor at Guelma (ILAfr I, 233 = CIL VIII, 5306). The name Sdby of Labdah N 17 (p. 25) is attested as Sidbius in Algeria (ILAfr I, 2830 = CIL VIII, 28077). The inscription }mtb¨l h-Òry of Trip. 44a (Labdah N 24, pp. 30, 316) means of course ‘Amot-Baal, the Tyrian’. The feminine ending -t was no longer pronounced, as shown e.g. by sanu, ‘years’, 608
REVIEWS
and it was then easily omitted in writing. There is no parallel for a name such as ‘Servant of the Baal of Tyre’. The proper names of Leptis Magna reveal the wide range of maritime relations maintained by this most flourishing emporium on the coast of the Greater Syrtis. Hksty in Trip. 47b (Labdah N 55, pp. 37, 331) is not a Berber name, but Greek }Akesta⁄ov, }Akestíav, }Akéstiov. As a rule, the aleph is omitted after yod at the end of Greek and Latin names, for instance Srwy, Servius, Slpqy, Sulpicius, Lqy, Lucius. The shin following the Hksty introduces the qualification P¨w}, probably PaoÕv, attested at Cyrene (Prosopographia Ptolemaica 2977). Also Nkspt (Labdah N 60, pp. 38, 358) is not a Berber name, but apparently Greek Nikásippov (LGPN I, p. 331b; II, p. 331c; IIIA, p. 319a; IIIB, p. 301c; IV, p. 251b), since Late Punic t may correspond to Greek or Latin s (OLA 127, p. 351). The ending s of a Greek proper name is indicated by s in NP 104, 1 (Constantine N 14, p. 201), where one should read ¨r†msys, }Artemísiov. The name Q¨tk} (instead of Q¨tkm), written on the lid of the funerary urn Trip. 44aa (Labdah N 50, p. 36), could be identified with Ktßkwn, attested on the island Kos (R. Herzog, Koische Forschungen und Funde, Leipzig 1899, No. 97). A final n is omitted frequently, but the omega is here marked properly. Mk of Trip. 44r (Labdah N 41, p. 34) simply records the name Mica (CIL VIII, 20919; RIL 605, 899: MK}; cf. Latin mica in CIL VIII, 9727; 14319; ICKarth II, 121), spelled Myk} in Trip. 44q (Labdah N 40, pp. 33–4), where it is a feminine name. The title {rks of Annobal in Trip. 44u (Labdah N 44, pp. 34–5) could simply be Greek ârxóv, ‘chief’. An interesting case is provided by the name and the surname engraved on the lid of the funerary urn Trip. 44y (Labdah N 48, p. 35), found in the harbour area. The name P¨wl†} seems to be a derivative of faÕlov in -atos, thus *Faulátov, ‘Worthless’. The surname m}mr looks like the Jewish qualification mumar of an apostate (M. Jastrow, A Dictionary of the Targumim…, p. 744a). If this interpretation is correct, it would attest to the presence of a Jewish community at Leptis Magna with some Judeo-Christian adepts (cf. H. Solin, ANRW II/29, 2, Berlin-New York 1983, pp. 770–9). The interpretation *Faulátov could fit in the frame of the so-called ‘names of humility’ with a pejorative meaning, given in Africa to Christians as a prophylactic means of preventing bad luck caused by the evil eye, etc. This is the explanation proposed by I. Kajanto, On the Problem of ‘Names of Humility’ in Early Christian Epigraphy, in Arctos, n.s. 3 (1962), pp. 45–53. The Handbook of Neo-Punic Inscriptions is a work that should find its place on the desk of every epigraphist dealing with Phoenician and Punic inscriptions, but also in the library of Northwest Semitic scholars and of historians studying the religion and culture of North Africa in Roman times. Let us hope that the author will continue his study of Neo-Punic inscriptions. doi: 10.1093/jss/fgq020
EDWARD LIPINSKI CATHOLIC UNIVERSITY OF LEUVEN
ANDRZEJ S. TURKANIK, Of Kings and Reigns: A Study of Translation Technique in the Gamma/Gamma Section of 3 Reigns (1 Kings) (Forschungen zum Alten Testament. 2 Reihe 30). Mohr Siebeck, Tübingen 2008. Pp. xiv, 231. Price: €55.00 paperback. ISBN: 978-3-16-149541-0. The relationship between the Hebrew books of Samuel–Kings and the Greek versions known as 1–4 Reigns is complex and unclear. Although Turkanik’s neatly titled study does not offer a fully convincing solution, it has the merit of taking us once again through a comparison between Old Greek (OG) and Masoretic Text (MT). 609
REVIEWS
The gamma/gamma section of 3 Reigns (1 Kings; gg in Thackeray’s 1907 schema) comprises the whole book, except for the first and last chapters which have been revised in the distinctive Kaige manner. Chapters 2:12–21:43 constitute a translational unity marked by much reordering of words, verses, and even chapters over against the MT, and also including lengthy additions called ‘Miscellanies’, not covered in this book. Attempts to make sense of the idiosyncratic Greek text have resulted in two opposing positions. According to one, the translator accurately reproduced a Hebrew text widely differing from (and perhaps antedating) the MT; proponents of this position include J. Trebolle Barrera, E. Tov, Z. Talshir and, most recently, A. Schenker and P. Hugo. The opposite view is that the translator used a sourcetext very close to the MT and was himself responsible for the alterations; proponents include J. Wevers, D. Gooding, R. Gordon and, most recently, P. van Keulen and Turkanik himself. For this book, Turkanik has revised his 2002 Cambridge doctoral dissertation. He has taken account of Schenker’s more recent work, Älteste Textgeschichte der Königsbücher (2004), although the bulk of his argumentation concerns the earlier Septante et Texte Massorétique dans l’Histoire la Plus Ancienne du Texte de 1 Rois 2–14 (2000). He has also engaged with van Keulen’s Two Versions of the Solomon Narrative (2005), a substantial work which, while reinforcing his conclusions, inevitably overshadows his own shorter contribution. He does not mention Hugo’s challenging new study (also a doctoral revision), Les deux visages d’Élie: Texte massorétique et Septante dans l’histoire la plus ancienne du texte de 1 Rois 17–18 (2006). Turkanik is convinced that by comparing OG and MT he can ‘discover the translator’s ideological or theological persuasions’ (p. 7). He rightly focuses on translation technique as the means of isolating the translator’s habitual practices, extending the term to include features which could reveal his own, as distinct from his sourcetext’s, point of view. This is standard procedure; Turkanik’s contribution is to investigate the whole of the gamma/gamma section (although he does not treat every verse). In other studies, the focus has been on either the chapters concerning Solomon, or those concerning Jeroboam and the divided monarchy, including the Elijah and Ahab narratives. Turkanik surveys all relevant verses, from the accession of Solomon in 2:12 to Ahab’s Pyrrhic victory in 21:43 (OG reverses MT 20 and 21). Although he presents his argument as inductive (p. 8), he does not proceed in a completely sequential way, as he organizes his material into four thematic chapters (p. 10) covering ‘logical consistency’, ‘piety’, ‘main characters’, and ‘court etiquette’, headings taken from Wevers’ article, ‘Exegetical Principles Underlying the Septuagint Text of 1Kings ii 12–xxi 43’ (1950). The meatiest treatment concerns ‘the Translator’s Quest for Logical Consistency’ (pp. 11–98). There are three sections, on major transpositions, the smoothing out of perceived difficulties, and minor clarifications. Each section discusses relevant verses taken in order, so one keeps starting from the beginning with a different heading. The chapter on ‘Piety’ (pp. 99–126) traces the translator’s concern that people should show proper respect for God and the holy places, while the denigration of idols is enhanced. Oddly, Elijah, the blameless prophet, is treated here instead of in the next chapter which surveys major characters (pp. 127–92). Here, not surprisingly, Solomon has the lion’s share of attention (pp. 127–55); under various headings, he emerges as the one who can do no wrong. Not all the other players really count as ‘major’ (Nadab?). The section on Ahab reveals an interesting ambiguity of outlook (p. 176). In chapter 5, Turkanik develops remarks made by Wevers in his 1950 610
REVIEWS
article to demonstrate the translator’s concern for proper relationships between rulers and subjects (pp. 193–204). The thematic approach makes for a tidy presentation, but it poses problems. Contrary to a truly inductive method, it makes us read the text with the outcome already clear. The frequent return to the beginning of the text under a new heading interrupts a properly sequential reading. Inevitably there are repetitions, as certain verses are relevant to more than one topic. 3 Reigns 4:31 (1 Kgs 3:1, Solomon’s provisions for his Egyptian wife), is discussed four times (pp. 14; 23; 100; 141–2). Other examples include 2:26 (the judgement on Abiathar, pp. 41–2; 128); 3:4 (the high place at Gibeon, pp. 118; 145–6); 3:15 (Solomon’s dream, pp. 43–4; 90); 20:1 (Naboth’s vineyard, pp. 80; 181–2). The different facets needed pulling together, both at the end of each chapter (where ‘conclusions’ are brief in the extreme) and in the final summing up (pp. 205–10), especially as cross-references are sometimes lacking. Overall, however, a picture of the main characteristics of the translation does emerge, and some of Turkanik’s observations and arguments are cogent. Too much argumentation, however, consists in sweeping dismissals of rival views, especially those of Schenker, who is taken to task for being selective, not engaging sufficiently with other scholars, and blurring distinctions between ‘possible’ and ‘probable’ in the translator’s relationship to his source-text. Such critiques may be essential to the genre ‘dissertation’, but sound immature in a published version. I doubt whether ‘possibles’ and ‘probables’ can be sharply distinguished in a debate where what are needed are ‘undeniables’: places where the Greek couldn’t reflect a Hebrew original. There are in fact some hints — not picked up by Turkanik — where minor additions show that the translator had misunderstood a Hebrew idiom (e.g. 3 Reigns 3:20, ‘your servant’, p. 197). A good case can in fact be constructed for both main positions (a radically different source-text, or a creative translator). Individual approaches are often undergirded by presuppositions about whether or not translators had authority to make ‘improvements’. Unfortunately, we have no ancient access to how translators understood their task. If we assume that they did not innovate but stayed close to their source, then radical differences must reflect a text other than the (proto-) MT; in this case it seems surprising that 3 Reigns is often identical to the MT. If we do not make that assumption, the translator can plausibly be credited with innovations; but can we be sure that questions about consistency, cultic propriety, and so on, had not already led to adaptations in Hebrew? Without external evidence such as that provided by Qumran for some controversial texts, the arguments seem evenly balanced. Turkanik would not have lost credibility with this reviewer (who is basically sympathetic to his position) by being more hesitant, especially over the status of pluses and minuses. He does not differentiate between minor and major ones, or engage with Tov’s rule of thumb that only the former are attributable to translators. He might disagree with this distinction, but it needed to be addressed, especially as Tov seems to be vindicated on occasion (cf. the plus in LXX Hos 13:4, attested in Hebrew in 4QXIIc). It does not follow that all pluses necessarily reflect a Hebrew original, only that the possibility can’t be ruled out. Given the impasse, it is a pity that Turkanik did not give more space to the overall character of the Greek text (even without the Miscellanies). Because the translator is credited with almost all the divergences from the MT, some sort of contextualisation seems called for. But Turkanik shies away from this. He says rightly that there is no firm evidence for date or place of translation (p. 1), but 611
REVIEWS
simply to specify ‘Hellenistic times’ and to say that the translator is ‘concerned that the text should be made relevant in the light of present-day geography’ (p. 70), is too vague. A potentially rewarding opportunity is passed up in Chapter 5 where matters of court etiquette might have prompted questions concerning Ptolemaic Egypt (especially as Schenker’s observation about co-regency is emphasized for 3 Reigns 2:19, p. 195). Turkanik merely evokes ‘principles of social behaviour known and accepted in antiquity’, especially ‘in the days of our translator’ (p. 193). His conclusion is that the translation ‘fits well into the religious and cultural milieu of Alexandrian Jewry’ (p. 209). But he does not tell us what this was. There is, however, one good observation, where he relates the treatment of Moses as lawgiver in 3 Reigns 9:6 to that of other Hellenistic Jewish writers (p. 124). More of this might have given Turkanik a firmer anchorage and reinforced the credibility of his pro-active translator. What Turkanik has done well is to have collected a wide range of material in an organized and accessible way. This, together with at least some of his detailed textual discussions, will be useful in the continuing attempts to make sense of distinctive but still ambivalent material. doi: 10.1093/jss/fgq021 J.M. DINES UNIVERSITY OF CAMBRIDGE
EDWARD M. COOK, A Glossary of Targum Onkelos: According to Alexander Sperber’s Edition (Studies in the Interpretation of Aramaic Scripture 6). Brill, Leiden 2008. Pp. xxii + 310. Price: €113.00 hardback. ISBN: 978-90-04-14978-6. This new volume is the first publication devoted exclusively to the vocabulary of Targum Onkelos. Other dictionaries useful for the reading of Targum Onkelos have been far broader in scope, like Marcus Jastrow’s popular Dictionary of the Targumim, Talmud Bavli, Talmud Yerushalmi, and Midrashic Literature. Cook’s glossary, however, is just as the title suggests: a glossary. It is not a detailed dictionary like that of Jastrow, in that the definitions are very brief, and textual passages are not cited. Opening the book is a short introduction (pp. ix–xvii), which discusses previous dictionaries relevant to Targumic Aramaic, the issue of the dialect of Targum Onkelos, and the layout of the present work. The glossary itself (pp. 1– 307) makes up nearly the entire volume. Each page only has about eight entries (by my estimate, there are between 2200 and 2300 total entries), and definitions are usually just one to three words in length. For nouns, the entry is listed under the absolute form, but the emphatic form is normally given also; plural and construct forms are very rarely provided. For adjectives, only the masculine singular absolute form is normally given. Biblical text references (chapter and verse number) are given only if the lexeme occurs less than ten times. References to entries in the other standard Aramaic dictionaries are listed, which is handy, but only very rarely does Cook provide the actual forms from these other dialects. Cook notes when a word is a loan, though only sometimes does he provide the source-form. For example, the entry for naÌtom ‘baker’ (p. 177) cites the Akkadian source nuhatimmu [sic], and the entry for pitgam ‘word’ (p. 231) cites the Persian source patgam, but the entries for †eqas ‘order, regiment’ (p. 105) and n¢Ìasirkan ‘hunter’ note only that the words come from Greek and Persian, respectively, without providing the source lexeme. Only rarely (for about fifty-five entries) does Cook provide a short footnote, with a brief comment or reference to a scholarly article. Following the main set of entries is a short section on Geographical Names (pp. 308–10), which includes just fifteen lemmata. 612
REVIEWS
No doubt this glossary is an extremely useful tool for the student or casual reader of Targum Onkelos. However, this usefulness is unfortunately offset by the price of the book: €113 (or $162, the current retail price in the USA as of April, 2010, even though €113 is currently worth $151). This is a hefty cost for a relatively light volume. In order to really test this glossary, I used it as a companion to reading Targum Onkelos to Genesis 39. I can make the following comments, which are intended not as criticisms, but simply as observations on the contents and scope of the book. Gen. 39:1: MiÒrayim ‘Egypt’ and its corresponding adjective are absent from the glossary, though place names like Y¢ruslam ‘Jerusalem’ appear in the appendix of Geographical Names, and adjectives like y¢huday ‘Jew’ (‘Judean’ is not given as a definition) appear in the main glossary. Gen. 39:2: Though there is an entry for s¢¨ed ‘support, help’ (p. 197), it would be helpful to include there the idiom b-sa¨ad- d- ‘in support of, by the side of’, which occurs in this verse and in a number of other places. Gen. 39:3: The glossary entry for the particle }are (which occurs nearly 1400 times in Targum Onkelos) includes only the definition ‘for, because’ (p. 22). It makes no mention of the use of the particle as a complementizer ‘that’, though it has this function very often, nor does it mention its use as ‘if, when’ (e.g., Gen. 24:41). (Essentially, }are slavishly translates Hebrew kî.) Other common particles also get short treatment in the glossary. For example, kid (p. 125) has only the definitions ‘as, as if’, but in a passage like 43:3, it clearly must mean ‘if’ or ‘when’. Gen. 39:4: The glossary defines the pa¨el verb sames only as ‘to minister’ (p. 287), which is not quite sufficient for the context of 39:4, where the verb is best translated as ‘attend’ or ‘serve’. Gen. 39:6: The emphatic form rewa, which occurs in this verse, is missing from the entry rew (p. 264), though many other entries list the emphatic form, following the absolute; presumably, rewa is not given because its form is predictable. The entry for the adjective ya}e (p. 110) gives only the definition ‘comely’; a definition ‘beautiful, handsome’ would seem more appropriate. The entry for Ìezu (p. 88) gives the emphatic form Ìazwa, yet the form in 39.6 (in Sperber’s Babylonian pointing and in Tiberian pointing) is Ìizawa. Gen. 39:7: The glossary includes the form ha-}illen only under the entry for }illen (p. 14), and there is no comment on the distribution of the two demonstrative forms. On the contrary, the entries for da (p. 55) and den (p. 60) include the abbreviation ‘attrib. adj.’ before the forms with prefixed ha-. Cross-references for all of the ha- forms (under the letter he) would have been ideal. Gen. 39:11: The verb mibdaq must mean something like ‘inspect’ or ‘examine’ in this verse. The glossary lists only the single meaning ‘search’ for this p¢}al verb, which is not sufficient. Gen. 39:14: The adjective ¨ibray (emphatic ¨ibra}a) is missing; see the comments to 39:1, above. Gen. 39:19: The idiom t¢qef rugzeh ‘he grew angry’ occurs (in various forms) at least fifteen times in Targum Onkelos, and at least as many times in Targum Jonathan, but the glossary has only individual entries for t¢qef ‘grow strong, harsh’ (p. 306) and r¢gaz ‘anger’ (p. 259). One could certainly figure out from the glossary, without much trouble, that this means literally ‘his anger grew strong’, and hence ‘he grew angry’, but a note on this idiom would be desirable. Gen. 39:20: The glossary entry for }asir ‘prisoner’ (p. 19) gives the emphatic plural form }asire and the note ‘only in bet }asire prison’. First, the phrase bet }asire 613
REVIEWS
should have a cross-reference under bet, as do several other compounds whose first element is bet. Second, Cook’s note is somewhat misleading. The normal emphatic plural of }asir is }asirayya (e.g., 39:22), which has a construct }asire (39:20). In this entry, if the unexpected plural form that appears in the compound bet }asire is included, then the regular emphatic plural should also have been listed. I will reiterate that the above comments are not all to be taken as criticisms, but rather are meant to show how the glossary performs. The comments show that entries are at times too brief or incomplete, and that they can be inconsistent. But the glossary is certainly helpful in most cases, as a clear definition is given for nearly every word in the Targum. And the large, clear print makes this glossary far easier to read than Jastrow’s dictionary. Targum Onkelos is a wonderful and important text, and it deserves its own dictionary. If the volume were affordable, it would be a very welcome tool, but at the current price, it is hard to see how this will become a popular glossary for students and most scholars. doi: 10.1093/jss/fgq022 AARON D. RUBIN PENNSYLVANIA STATE UNIVERSITY
ROBERT PHENIX, The Sermons on Joseph of Balai of Qenneshrin: Rhetoric and Interpretation in Fifth Century Syriac Literature (Studien und Texte zu Antike und Christentum 50). Mohr Siebeck, Tübingen 2008. Pp. xviii + 333. Price: € 64.00 paperback. ISBN: 978-3-16-149676-9. The interaction of Greek and Syriac in late antiquity has attracted much discussion in recent years, not least in view of its significance within the wider context of the history of classical thought in antique and medieval times. Much of the attention has been focused on philosophical and scientific aspects of this history, but the literary side has also not escaped notice, and given the central position of rhetoric within the educational system of late antiquity, a rounded view of the process is hardly possible without due attention to this element. This work by Robert Phenix on a group of fifth century Syriac sermons by Balai of Qenneshrin constitutes an important new contribution to the discussion. The author notes in his early chapters on the background of these sermons that ‘Balai is the earliest extant author of Syriac who lived west of the Euphrates’ (p. 56). This observation gives considerable importance to the work of Balai within the overall analysis of Greek-Syriac literary interaction. Here we have an author writing in Syriac in an environment which, according to all our other literary and epigraphic evidence, is predominantly Greek. To what extent, therefore, is this early piece of Syriac literature fundamentally ‘Greek’ in inspiration? This is the basic issue with which this study is concerned, and the author’s conclusion is that in the rhetoric of these sermons Balai ‘summons all the sophistication and discipline of his Hellenistic education’ (p. 270). This conclusion is reached after a lengthy (and likewise sophisticated) analysis of the structure and themes of these sermons on Joseph, covering Balai’s employment of rhetorical figures, dramatic irony, invention, arrangement, and techniques of characterisation and persuasion, together with, in the thematic realm, subjects and metaphors such as purity, virtue, honour and battle. The method of rhetorical criticism, to which the author devotes a few brief but useful pages, is not without its difficulties in determining the rhetorical theory which may be supposed to be the basis upon which a speech was constructed, and even after presenting a considerable amount of evidence the author does not overstate his case. Thus after his detailed study of invention and arrangement in the 614
REVIEWS
prefaces of three of the sermons, he considers he ‘has demonstrated the plausibility of the influence of classical rhetoric on the arrangement of some of the speeches’ (p. 198). He recognises, in other words, that a Syriac ecclesiastical poem, albeit consisting mostly of speeches delivered by characters in a narrative, may not fit a classical rhetorical template as well as a prose speech in Greek or Latin, but he has nevertheless made a good case for supposing that a training in classical rhetoric has had a considerable influence on the composition of these sermons. The narrative and speakers are those of the Joseph story of Genesis, but it is unlikely that the biblical text alone was the source for Balai. That text gave rise to a great number of elaborations in both Jewish and Christian circles, and an analytical comparison of Balai’s sermons with this extensive corpus of texts on Joseph is the other main focus of this book. Phenix takes issue with the hypothesis that the section on Joseph in Ephraem’s Commentary on Genesis served as the basis for Balai’s elaboration. In a lengthy chapter and an appendix collating the parallel material, both biblical and extra-biblical, he argues that the author of the Sermons on Joseph could not have been Ephraem (as has sometimes been supposed), and that he probably used Ephraem neither for the outline of the story nor for the extra-biblical material. Phenix therefore turns to the other Joseph material in search of Balai’s sources, Greek and Syriac, Christian and Jewish. After examining the Joseph material for its connections to Balai in Narsai, Pseudo-Basil, Ephraem Graecus, Joseph and Aseneth, Philo, Josephus and Jewish Pseudepigraphic and other sources and concluding, primarily on the basis of the ‘moral’ reading of the story of Joseph as a righteous individual wrongly accused but ultimately vindicated, that one or more popular Jewish works on Joseph (but not the presentations of Philo or Josephus) is closer to Balai than the extant Christian material, he suggests that the Urgeschichte serving Balai was a Jewish Greek narrative transmitted into Syriac in the fourth century. That an original, but no longer extant, Jewish Greek narrative about Joseph (possibly composed in Egypt) was adopted by Christians in Syria is not implausible, but it does not seem necessary to suppose that this narrative was translated into Syriac. If Balai obtained his knowledge of classical rhetoric within the Greek environment of Chalcis, but expressed himself more naturally in Syriac and addressed a Syriac-speaking audience, he could well have made use in his Syriac sermons of a Greek narrative. At a slightly later date, for example, Sergius of Reshaina wrote in Syriac a treatise on Aristotle’s Categories on the basis of the Greek text, without recourse to a Syriac translation (which may not have existed at the time he wrote), and this easy bilingualism (reading Greek and writing Syriac) may be assumed for many of the ‘elite’ in late antique Syria and Mesopotamia. The little that is known of Balai’s life is reviewed in chapter two, as are the reasons for attributing the sermons to him, rather than to Ephraem or another author. Phenix regards the attribution to Balai as a ‘plausible hypothesis’, resting primarily on the manuscript witnesses, which he reviews together with the editions in chapter one. He considers the association of Balai with Chalcis (Qenneshrin) to be consistent with all the evidence, if not provable at the present time, and that Balai could have been born in Edessa or another Syriac settlement. While conscious of the objections that can be raised against the hypothesis that classical rhetoric influenced works written in Syriac, he refers not only to the contact between Greek and Syriac in northern Syria, but also to the fifth century Life of Rabbula (composed c. 436– 49). Surprisingly, he makes no mention of other figures of a slightly later time (late fifth/early sixth century) for whom Syriac and Greek (and indeed Greek rhetoric) were undoubtedly inextricably intertwined, such as John Bar Aphthonia and his 615
REVIEWS
panegyrist, Sergius of Reshaina, and Marutha of Amida. Of course, if it has to be shown that Greek rhetoric influenced Syriac speakers and writers before or during the lifetime of the author under consideration (died c. 437 or 459?), then these figures of a slightly later period will not do, but a plausible case can be made from a slightly later as well as an earlier or contemporary period. The author notes that some of the virtues attributed to Joseph could have been taken from epideictic rhetoric, but might also have come from Jewish tradition, and Balai’s possible indebtedness to epideictic is slight in comparison to Philo’s classical encomiastic presentation of Joseph as the ideal ruler. However the speeches within these sermons which may be categorized as deliberative or judicial are especially interesting, given that the evidence already available for the influence of classical rhetoric in Syriac relates mostly to epideictic. Phenix’s analyses of these speeches with the aid of the rhetorical theory embodied in the corpus of Hermogenes is correspondingly of special value, since that corpus hardly touches on epideictic and has therefore found little application heretofore in rhetorical analysis in Syriac. This richly documented study should therefore prove of great interest not only to students of Syriac literature, but also to those interested in the history of classical rhetoric and the development of biblical traditions in late antiquity. Those who read this book with interest will also look forward to the author’s projected new edition and English translation of these remarkable sermons. doi: 10.1093/jss/fgq023 JOHN W. WATT CARDIFF UNIVERSITY
ROBERT GLEAVE, Scripturalist Islam: The History and Doctrines of the Akhbari Shi¨i School (Islamic Philosophy, Theology and Science, Texts and Studies 72). Brill, Leiden 2007. Pp. xxiv + 344. Price: €125.00. ISBN: 978-90-04-15728-6. In the last two decades plus the Twelver branch of Shi¨ism has attracted growing popular/lay and scholarly attention. Though the field has expanded its output to meet this demand, not all of what has been produced deserves equal attention. Robert Gleave is among those who, equipped with both Arabic and Persian and having accumulated long hours with the relevant textual sources, is well qualified to discuss matters of Twelver jurisprudence. The present volume addresses the roles of the Imams’ sayings (akhbar or aÌadith) and of independent legal reasoning (ijtihad) in the Shi¨i jurispridential process. Those Twelvers identified with opposition to the rise of and increasing reliance on ijtihad in favour of recourse to the akhbar are the Akhbaris. In nine chapters, a conclusion and with three appendices, Gleave discusses Akhbarism and its legacy. Gleave argues that there is no evidence to substantiate suggestions that Akhbari tendencies were visible within Twelver jurisprudence prior to the seventeenth century in favour of the traditional understanding that a ‘rigorous critique’ of rationalist tendencies that had developed to that time was ‘first delineated’ by MuÌammad Amin al-Astarabadi (d. 1036/1626–7) (p. xviii). Gleave also argues that the dispute between the Akhbaris and the proponents of ijtihad (the UÒulis) focussed on technical questions of Shi¨i jurisprudence [and] had ‘limited societal effect’ (xx; cf. p. 296). Akhbarism was not, per se, anti-clerical or opposed to the hierarchical structures of authority (pp. 99, 295–6). Too, Akhbari scholars could hold very different legal opinions (p. 175). As in al-Astarabadi’s most famous work, al-Fawa}id al-Madaniyya, Akhbarism argued that the akhbar constituted scripture and, as such, no other sources of the law could be accepted. Gleave is careful to add that Akharis did disagree as to whether or not theological issues might be informed by additional 616
REVIEWS
sources, including reason or philosophy, for example. Indeed, Gleave argues, alAstarabadi’s own theology was little influenced by his jurisprudence. Gleave sketches the subsequent two hundred year history of the school as its established centres in Iran, Iraq, the Gulf, Lebanon and even in the Subcontinent. He also addresses the several lists of differences between the two schools that appeared over the period, suggests these allow for developing a sense of how Akharis viewed themselves and argues that an Akhbari ‘school’ (madhhab) can be said to have emerged only some 80 years after the death of its founder. The last three chapters examine the development of Akhbarism between the early seventeenth century and the early nineteenth century when Akhbarism began to disappear. Intra-Akhbari disputes were apparent and these, in addition to dialogue between Akhbari and UÒuli scholars, stimulated the further development of Akhbari ideas if not the formation of distinct ‘schools’, per se. Chapter seven addresses the role of the akhbar in the interpretation of the Qur}an and chapter eight discusses what Akhbaris meant by sunna. Chapter nine addresses the question of how the akhbar were more readily understood than the Qur}an that they were supposed to interpret. The appendices, in turn, include a list of Akhbari scholars from the time of al-Astarabadi, a list of the many differences between Akhbaris and UÒulis and, finally, a fatwa of al-Astarabadi on the purity of wine. Gleave offers several answers to the question of why Akhbarism disappeared (pp. 303–4) but, a key factor in its demise was that Akhbarism ‘failed to resolve certain basic questions concerning its role within the Shi¨i scholarly elite’. The historian might make more of Gleave’s mention (p. 300) of the 1818 murder of Mirza MuÌammad al-Akhbari by UÒuli elements in Iraq — his body was fed to dogs — as an incident of note, let alone as a factor in the demise of Akhbarism. The killing of such a prominent figure seems irreconcilable with Gleave’s assertion that these polemics had ‘limited societal effect’. Indeed, in 1985 Cole noted the practical, material implications of being an UÒuli or Akhbari: in the Iraqi shrine cities the former were in receipt of funds from the Shi¨i community in Lucknow and this may have been a factor in the conversion of at least one prominent Akhbari to UÒulism, namely Sayyid Dildar ¨Ali (d. 1820). In his 1988 Roots of North Indian Shi¨ism in Iran and Iraq Cole noted the role of UÒuli clerics in rooting out syncretistic tendencies in Lucknawi Shi¨ism that had developed from the interaction of Indian Shi¨ism with other Indian religious traditions. The ‘societal’ intruded at other points as well. In Safavid Iran popular pressure forced the UÒuli Shaykh Baha}i (d.1621) to abandon his post as Shaykh al-Islam of the Safavid capital of Esfahan. Later the Akhbari MuÌsin Fay∂ al-Kashani (d. 1680) was driven from his post as the capital’s Friday prayer leader. Gleave does note disagreement among members of each ‘school’ over ‘legal’ issues (pp. 174–5). However, he does not explore these disagreements — including, for example, the practical, ‘political’ implications of late-Safavid period disagreements over the permissibility of Friday prayer during the Imam’s absence — and the potential and real very ‘practical’ implications thereof. Gleave’s suggestion (pp. 295–6) that Akhbarism did not fundamentally question the hierarchical manner in which authority within the community had evolved since the onset of the Imam’s absence is a corollary of his view of the dispute as essentially non- or apolitical. In fact, support for hierarchicalisation by the authors of those Safavid-period Akhbari and UÒuli texts interrogated herein is to be expected. Most of these figures are already well-known: some were connected to each other by marriage or, more often, by student/teacher relationships. Too, nearly all enjoyed 617
REVIEWS
favours from the Safavid court and many were also the beneficiaries of court appointments to hierarchical positions within the community. Thus, naturally, they were supportive of such a hierarchy. Al-Kashani, for example, would not question the legitimacy of the post of Friday prayer leader to which he was appointed. Nor would the UÒuli Baha}i. But, among some important (and powerful) sections of society there was clearly sufficient discontent both to force each man to give up a court-appointed post and to forestall any counter-response by either of the two shahs in question. That discontent and its roots in Safavid society are not explored. Gleave also suggests that the Shi¨i †abaqat literature of the period is not a reliable guide to who was/was not an Akhbari as this literature was ‘highly formalised’ (pp. 49ff.). In fact, the authors of the clerical biographies over the whole period, let alone those biographical dictionaries completed in and covering the Safavid period — not all of which were examined — expressed their disapproval of their opponents, contemporary or not, simply by omitting any reference to them. The biographers’ entries on their like-minded co-religionists do certainly emerge as ‘formalised’. Fortunately, a number of those very interesting scholars excised from this literature are otherwise well-known to historians of the period. Finally, to require that any pre-seventeenth century manifestations of Akhbarism meet the criteria said to be characteristic of that seventeenth century Akhbari discourse in order to be called Akhbari (pp. 25f, 177) sets up a self-fulfilling argument. Over the centuries that discourse ought to be expected to have been at least as dynamic as the UÒuli if, owing to the ‘victory’ of the latter, all the harder to track. As Gleave notes at several points (e.g. pp. 179, 215) the story recounted herein involves, in large measure, questions of self-definition and self-perception. As such the story of Akhbari self-identity as revealed by Gleave’s expert, close analysis of the texts selected is fascinating and long overdue. This volume constitutes a major contribution to Shi¨i and Iranian studies and ought to serve to highlight the need for continued examination of the history, meaning and legacy of this discourse. doi: 10.1093/jss/fgq024
ANDREW J NEWMAN UNIVERSITY OF EDINBURGH
GEOFFREY KHAN, Arabic Documents from Early Islamic Khurasan (Studies in the Khalili Collection, Volume V). The Nour Foundation in association with the Azimuth Press, London, 2007. Pp. 183. ISBN: 978-1-874780-71-7. It is a thrill for philologists and linguisticians when new evidence comes to light on a hitherto little-known and little-attested language. This has recently been the case with the Eastern Middle Iranian language Bactrian, once widely used over a large expanse of territories from the eastern Iranian lands through parts of Central Asia to northern India by dynasties like the Kushans. It was previously known to us only from coin legends and a handful of inscriptions from the Eastern Afghanistan-North West Indian region, written in Greek script, one of the long-lasting legacies of Alexander the Great’s Indian campaign of the later fourth century BCE and the subsequent Hellenization of a significant part of northeastern Iran, the Oxus Valley lands and Transoxania. Beginning in the 1990s, there came on to the art and antiques market — probably through the Peshawar bazaar, although the exact provenance is unknown — documents written mainly on parchment, though with some on leather and cloth, in the Bactrian language, also in the Greek script. They apparently came from a cache somewhere in what is now north-central Afghanistan, north of the Paropamisus range and in the classical region of Bactria, and cover a period from the fourth century CE to the eighth century. They have been published, translated 618
REVIEWS
and interpreted by Professor Nicholas Sims-Williams (Bactrian Documents from Northern Afghanistan. I, Legal and Economic Documents, Oxford 2001, IL Letters and Buddhist Texts, London 2007, with a third volume of plates to follow), and have provided a treasure-house of new material for Iranists. Probably emanating from the same source as the Bactrian-language documents, there has also come to light a smaller, but still very significant, number of documents written in Arabic and stemming from the third quarter of the eighth century CE. As well as having an historical value, they are important for early Arabic palaeography and diplomatic history. Until now we have had very little Arabic documentary evidence from the eastern Islamic world beyond an Arabic letter from the year 100/718–19 on parchment and discovered in the ruins of Mount Mugh in the Zarafshan valley in what was ancient Sogdia, now in the northern part of the Tajik Republic (published by I.Y. Krachkovski and V.A. Krachkovskaya, ‘Le plus ancient document arabe de l’Asie Centrale’, in Sogdyskiy Sbornik, Leningrad 1934). The vast majority of early Arabic documents stem, of course, from the papyri preserved in the dry conditions of Egypt, and this has made the study of early Arabic palaeography unavoidably skewed to the west-central part of the Arab world. The early Arabic documentary corpus has now been ‘dramatically increased’, in Professor Khan’s words, by this discovery of 32 legal and administrative documents on parchment coming from what appears to have been a family archive; the names of one Mir b. Bek al-Bamiyani and various members of his family occur frequently. Their subjects include land transactions, financial deals, payment of dowries and the manumission of slaves, but the majority of them, 23, are official quittances for the payment of taxes (here and elsewhere in contemporary Islamic usage, bara}as). From internal evidence, these and parallel documents in Bactrian of the same period come largely from what was in medieval Islamic times the district of Rob, modern Ru}i, some 50 miles south of Samangan on an historic route running from Balkh in the Oxus Valley across the mountains to Bamiyan and Kabul, with a few of the documents coming from the adjacent province of Guzgan to its west. Rob had at this time a local ruler, whom al-™abari calls in his History (II, 1219) the Ru}b Khan; this petty prince assisted the Arabs against the Hephthalite ruler in northern Afghanistan, ™arkhan Nizak, in the year 91/709–10. From the documents, one may deduce that, some fifty years later, Rob was still essentially an ethnically Iranian, Bactrian-speaking region giving a light allegiance to the ¨Abbasid central government in distant Iraq but paying taxes to fiscal representatives of the Arab governors in Khurasan. In such circumstances, the use of Arabic, for administrative purposes at least, was obviously making headway there, but conversion to the Islamic faith must have been only in its very beginnings. Professor Khan has dealt with the documents in exemplary fashion. Deciphering them in the first place must have been a feat in itself. There is a lengthy Introduction (pp. 13–90) on the historical setting of the documents, the information they provide on early Islamic administrative practice, their palaeographical features and the clay bullae attesting authorization attached to the documents. Each document is then illustrated, edited and translated, with a physical description and with notes on features of script, names and official titles and historical significance. In particular, he notes how the names and titles throw light on local divinities of the region, such as the god Zun/Zhun, whose cult in Zabulistan and Zamindawar in eastern Afghanistan was encountered by the incoming Muslim invaders of the eighth and ninth centuries. There seems to have been already an infiltration of Turkish ethnic elements into northern and eastern Afghanistan at this early period, judging by the mention in the documents of ancient Turkish titles like khaqan and †arkhan, used 619
REVIEWS
here, however, as parts of the personal names of persons ostensibly Iranians. The tax-collecting officials, the ¨ummal, have Arabic names and are under an Arab amir, but the tax-payers themselves have Iranian names and do not appear to have been Muslims. Ru}b was only a tiny tax-paying district on the remote eastern fringe of the Islamic oikumene, but the documents do indicate what may have been more general features of financial practice in the lands east of Iraq. In a comparison of the information of these documents on administrative affairs with that of the Egyptian papyri (pp. 42ff.), Khan notes that the first six documents refer to the kharaj tax, its usage here in pre-150 AH years antedating its appearance in the Egyptian papyri some ten or fifteen years later. Likewise of historical significance is the mention of supplementary taxes (qisam ‘portions’) supporting local administrative services, including the barid or government postal system, for whose beginnings we have little evidence except this mention and some references in Egyptian papyri from the early ¨Abbasid period (pp. 34ff. and document no. 6, pp. 102–3, lines 7–8: nafaqat dawabb al-barid wa-qismat nafaqat/al-ar∂ wa-jawa}iz al-burud wa }1-rusul wa-nuzuliha; and cf. Adam Silverstein, Postal Systems in the Pre-Modern Islamic World, Cambridge 2007, 59ff.). In these financial and legal documents from eastern Khurasan (in both the Bactrian and Arabic language ones), clay bullae are attached to the parchment with physical marks of the attesting witnesses, either in the form of a seal or of a thumb-nail, rather than autograph signatures (pp. 57–9). The documents concerning the emancipation of slaves (nos. 29–32) involve both unconditional freeing (¨itq, ¨ataqa) and contractual freeing (mukataba) (pp. 59–64). Matching the quality of its contents, this is a beautifully-produced book, with fine typography, clear illustrations and what seems to be a total absence of typographical errors in both the English and Arabic: a truly splendid achievement! doi: 10.1093/jss/fgq025 C. EDMUND BOSWORTH CASTLE CARY, SOMERSET
620
Journal of Semitic Studies LV/2 Autumn 2010 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
SHORT NOTICES SIMO PARPOLA, ROBERT WHITING et al. (eds), Assyrian-English-Assyrian Dictionary (The Neo-Assyrian Text Corpus Project). State Archives of Assyria, Helsinki 2007. Pp. xvii + 289. Price: $75.00 paperback. ISBN: 978-952-10-1332-4. The Assyrian Dictionary is not intended to replace the traditional Assyriological dictionaries, CAD and AHw, and is avowedly not an Akkadian dictionary. It sets out to provide a glossary for Neo-Assyrian texts. While Old Assyrian and Middle Assyrian are not covered here, some words from these earlier periods are included, alongside words in contemporary use in Standard Babylonian — which functioned as the literary language of Assyria — and words known in modern Assyrian (a variety of Aramaic). The idea behind the inclusion of this additional material is to take account of vocabulary thought likely to have been in use in Neo-Assyrian, and which may become attested as new texts are edited. Symbols are added throughout to allow the reader to identify the source of a form or meaning, if anything other than NeoAssyrian. The dictionary includes about 13,000 Assyrian entries, gathered from letters, administrative and legal texts, royal inscriptions, religious and technical texts. These are matched by about 23,000 English entries. This book is easy to use; readers should not be put off by the somewhat intimidating diagram on p. xvii. The introduction sets out clearly and simply how to navigate the volume. The translations are focussed on the Neo-Assyrian corpus and include new or revised meanings, beyond what is found in CAD or AHw. As such it forms a very useful companion volume to the wonderful State Archives of Assyria series. In common with the other volumes in that series, the translations are not wooden or stilted. While anchored to the literal meanings, the translations convey also the sense of what is written, bringing the texts to life. The dictionary is both Assyrian-English and English-Assyrian (for the first time in this field), and has a variety of users in mind, ranging from experts through to informal learners. Each will obviously use the book in a different way, as appropriate to their own situation. The flexibility to serve such a range of users is a real achievement. Parpola and his international team based at the University of Helsinki have made it easier to study the Neo-Assyrian corpus than any other in the ancient Near East, whatever the reader’s background. doi: 10.1093/jss/fgq026
JONATHAN TAYLOR THE BRITISH MUSEUM
HERMANN GUNKEL (edited and translated by K.C. Hanson), Israel and Babylon: The Babylonian Influence on Israelite Religion. Wipf and Stock, Eugene 2009. Pp. 96. Price: $13.00. ISBN: 978-1-60608-250-8. This is the contribution a reluctant Gunkel made to the Babel-Bible controversy aroused by two lectures the Assyriologist Friedrich Delitzsch gave in 1902 and 1903 to the Deutsche-Orient-Gesellschaft in the presence of Kaiser Wilhelm II. The first lecture was a brilliant piece of popularization, in which Delitzsch outlined the results so far of the excavations in the Middle East. He chose his illustrations well, particularly from the newly discovered Code of Hammurabi, and no doubt enthused the Society to play a full part in further excavations. Unfortunately, he suggested 621
SHORT NOTICES
that much of Israel’s culture and religion was derived from Babylon, and this alarmed the church in Germany. Part of the problem was that the church was unaware of the work of biblical scholars. And so Gunkel comments patiently on the lectures, agreeing that the finds are important, but pointing out that there are significant differences as well as similarities between, for example, the Babylonian and Hebrew flood stories, and reaffirming the belief in a unique revelation through Israel and its prophets. The moderate tone is in places spoilt by disparaging remarks about contemporary Judaism. Hanson has revised an earlier English translation, incorporated some of Gunkel’s endnotes (in which he corrects Delitzsch’s exegesis) into the text, and supplied a somewhat minimal introduction putting the work into its historical context. This would have been done more effectively if it had been issued together with a translation of Delitzsch’s lectures (recently reprinted separately). doi: 10.1093/jss/fgq027
ROGER TOMES UNIVERSITY OF MANCHESTER
G. BOCCACCINI (ed.), Enoch and the Messiah Son of Man: Revisiting the Book of Parables. Wm B. Eerdmans, Grand Rapids 2007. Pp. xv + 539. Price: $50.00/ £27.99 paperback. ISBN: 978-0-8028-0377-1. This book consists of papers and responses from the Third Enoch Seminar, held at Camaldoni in June, 2005. Leading experts on the Similitudes of Enoch were invited, and the reviewer regrets being unable to join them. The result is essential reading for anyone with a serious scholarly interest in ancient Judaism. The importance of these essays lies in providing an up-to-date survey of the present state of scholarship on most aspects of the Similitudes. Very few however advance knowledge, and most of them show little sign of sufficient learning in classical Ethiopic, the only language in which this document has survived, and Aramaic, the language in which most scholars think it was originally written. Perhaps the best is H.S. Kvanvig, ‘The Son of Man in the Parables of Enoch’ (pp. 179–215). He suggests that the three different terms for ‘son of man’ were originally בר אנש, בר אדם, and בר אם כל־חי. The reviewer was unconvinced, but Kvanvig’s learning and ingenuity are essential for progress to be made in any aspect of our understanding of this document. In another fine essay, P. Piovanelli ‘“A Testimony for the Kings and the Mighty Who Possess the Earth”: The Thirst for Justice and Peace in the Parables of Enoch’ (pp. 363–79) proposes a more plausible pattern in the translators’ views of who the terms for ‘son of man’ referred to. Some other essays make interesting points. These include J. Ben-Dov, ‘Exegetical Notes on Cosmology in the Parables of Enoch’ (pp. 143–50), a learned discussion of some cosmological terms: G. Boccaccini, ‘Finding a Place for the Parables of Enoch within Second Temple Jewish Literature’ (pp. 263–89), though, like Henze in his response (pp. 290–8), the reviewer was not altogether happy about Boccacini’s use of ‘paradigms’: and J.H. Charlesworth, ‘Can We Discern the Composition Date of the Parables of Enoch?’ (pp. 450–68), which includes some new reasons for a plausible date under Herod the Great or somewhat later. The most serious gap in this book is the lack of sufficient discussion of the surviving text in the light of recent manuscript discoveries. Only Piovanelli notes the absence of a proper critical edition (p. 368). The conference should have included a discussion of this, and put forward plans to remedy it immediately. 622
SHORT NOTICES
The English text is excellently printed. The Ethiopic is however printed entirely in transliteration, and the Aramaic and Hebrew have words in the wrong order at line-breaks (p. 194 lines 8–9, p. 251 lines 2–1 up). There is a useful bibliography, but no indexes. Despite these criticisms, this book is essential reading for interested scholars, because it provides so much up-to-date information on most aspects of this difficult document. doi: 10.1093/jss/fgq028
MAURICE CASEY UNIVERSITY OF NOTTINGHAM
BRIAN A. CATLOS, The Victors and the Vanquished: Christians and Muslims of Catalonia and Aragon, 1050–1300 (Cambridge Studies in Medieval Life and Thought 59). CUP, Cambridge 2004. Pp. 480. Price. £75.00 hardback. ISBN: 9780521822343. Rarely does one have the privilege of reviewing such a splendid book, combining careful, intelligent and punctilious work in the archive with the constant drive to keep hold of the bigger picture. The geographical focus is the Ebro Valley, where the religious and political merged in the Crown of Aragón’s management of its ancient (and munificent) Muslim communities. These communities themselves were by no means homogeneous; some families acted as tax gatherers for the Crown and as judges facing both inwards to and outwards from their respective communities and who seemed to have been tolerated in positions of externally-nominated authority until their exactions outweighed the protection that they could offer; others — less visible in the court records that have been so ably mined — were merchants and primary contributors to taxation; still others — perhaps the most fortunate — were vassals of the Templars, and could rely upon that formidable organization’s protection. Much legal energy was spent in inter-Muslim dispute, through attempts to force those who claimed exemption from taxes (such as Templar vassals) to cough up and so reduce the common fiscal burden. Although there were examples of violence and kidnapping, Catlos observes that this was by no means any different for Christians, and, during a time of endemic violence, Muslims were not noticeably singled out when it came to the medieval exercise of brutality. Perhaps the most fascinating part of the book is the appendix where the trimmings and tackings of a few well-documented individuals are traced, something which makes the previous analysis come most strongly alive. The one unfortunate element in the book is the author’s tendency to essentialize, both regarding the goodness and rightness of self-governing Muslim society over the border, and also regarding what Aragonese Muslims would have preferred regarding their own government. Although no doubt pious Muslims would have liked a qa∂i appointed by communal consent based on his integrity and knowledge of shar¨ia, not everyone is pious, and these latter would seem to have found the Aragonese system rather congenial. Such an affirmation is supported by continuing Muslim immigration from the Dar al-Islam during this period, despite the repeated fatwas against such a depraved and irreligious practice as to seek a dwelling place within Christian lands. The pious, no doubt and as was their duty, emigrated early on to Islamic lands; the less pious could therefore make their personal and profitable and productive accommodation in a society wholly involved in Aragonese matters, and rather isolated from Islamic affairs. doi: 10.1093/jss/fgq029
ANTHONY JOHN LAPPIN UNIVERSITY OF MANCHESTER
623
Journal of Semitic Studies LV/2 Autumn 2010 © The author. Published by Oxford University Press on behalf of the University of Manchester. All rights reserved.
BOOKS RECEIVED Jeffrey C. Alexander with Martin Jay, Bernhard Giesen, Michael Rothberg, Robert Manne, Nathan Glazer and Elihu and Ruth Katz, Remembering the Holocaust: A Debate. Oxford University Press, Oxford 2009. Pp. xiv + 205. Price: £15.99 hardback. ISBN: 978-0-19-532622-2. Andrew R. Angel, Chaos and the Son of Man: The Hebrew Chaoskampf Tradition in the Period 515 BCE to 200 CE (Library of Second Temple Studies 60). T & T Clark, London 2006. Pp. xiii + 260. Price: £65.00 hardback. ISBN: 0-56703098-9. Adam H. Becker, Fear of God and the Beginning of Wisdom: The School of Nisibis and the Development of Scholastic Culture in Late Antique Mesopotamia (Divinations: Rereading Late Ancient Religion). University of Pennsylvania Press, Philadelphia 2006. Pp. xvi + 295. Price: £45.50 hardback. ISBN: 978-0-8122-3934-8. Jon L. Berquist (ed.), Approaching Yehud: New Approaches to the Study of the Persian Period (Society of Biblical Literature Semeia Studies 50). Society of Biblical Literature, Atlanta 2007. Pp. ix + 249. Price: $29.95 paperback. ISBN: 978-158983-145-2. Ian J. Bickerton, The Arab-Israeli Conflict: A History. Reaktion Books, London 2009. Pp. 244. Price £15.99 paperback. ISBN: 978-1-86189-527-1. David G. Burke (ed.), Translation that Openeth the Window: Reflections on the History and Legacy of the King James Bible (Biblical Scholarship in North America 23). Society of Biblical Literature, Atlanta 2009. Pp. xxiii + 274. Price: $34.95 paperback. ISBN: 978-1-58983-354-2. Richard J. Clifford (ed.), Wisdom Literature in Mesopotamia and Israel (Society of Biblical Studies Symposium Series 36). Society of Biblical Literature, Atlanta 2007. Pp. xiii + 116. Price: $19.95 paperback. ISBN: 978-1-58983-219-0. Philip van der Eijk and R.W. Sharples (translators), Nemesius: On the Nature of Man (Translated Texts for Historians 49). Liverpool University Press, Liverpook 2008. Pp. viii + 273. Price: £16.95 paperback. ISBN: 978-184631-132-1. Walter Dietrich (Translated by Joachim Vette), The Early Monarchy in Israel: The Tenth Century B.C.E. (Society of Biblical Literature Biblical Encyclopedia 3). Society of Biblical Literature, Atlanta 2007. Pp. xvi + 378. Price: $47.95 paperback. ISBN: 978-158983-263-3. E. Earle Ellis, The Sovereignty of God in Salvation: Biblical Essays. T. & T. Clark, London 2009. Pp. xiii + 127. Price: £55.00 hardback. ISBN: 978-0-567-034137. Ziad Elmarsafy, The Enlightenment Qur’an: The Politics of Translation and the Construction of Islam. One World, Oxford 2009. Pp. xiv + 269. Price: £19.99 paperback. ISBN: 978-1-85168-652 O. Gregory Haynes, Tree of Life, Mythical Archetype: Revelations from the Symbols of Ancient Troy. Symbolon Press, San Francisco 2009. Pp. 347. ISBN: 978-09824034-5-7. Abraham Joshua Heschel, God in Search of Man: A Philosophy of Judaism. Souvenir Press, London 2009. Pp. 437. Price: £14.99 paperback. ISBN: 978-0-28563-836-5. 625
BOOKS RECEIVED
Zvi Jonathan Kaplan, Between the Devil and the Deep Blue Sea?: French Jewry and the Problem of Church and State (Brown Judaic Studies 352). Brown University, Providence, Rhode Island 2009. Pp. viii + 140. Price: $19.95 hardback. ISBN: 978-1-930675-61-2. Israel Knohl, Messiahs and Resurrection in ‘The Gabriel Revelation’ (The Kogod Library of Judaic Studies 6). Continuum, London 2009. Pp. xv + 122. Price: £16.99 paperback. ISBN: 978-0-82642507-2. Carleen R. Mandolfo, Daughter Zion Talks Back to the Prophets: A Dialogic Theology of the Book of Lamentations (Society of Biblical Literature Semeia Studies 58). Society of Biblical Literature, Atlanta 2007. Pp. ix + 149. Price: $24.95 paperback. ISBN: 978-1-58983-247-3. Marco Marizza, Lettere ittite di re e dignitary (Testi del Vicino Oriente antico 3). Paideia Editrice, Brescia 2009. Pp. 227. Price: €26.20 paperback. ISBN: 97888-394-0766-5. Michael L. Morgan and Peter Eli Gordon (eds), The Cambridge Companion to Modern Jewish Philosophy (Cambridge Companions to Religion). Cambridge University Press, Cambridge 2007. Pp. xx + 382. Price: £18.99 paperback. ISBN: 978-0-521-01255-3. Simone Mrejen-O’Hana, Le Registre d’Élie Crémieux: Ephémérides de la communauté juive de Carpentras (1736–1769). The Bialik Institute, Centre des langues juives Universite hébraïque de Jérusalem, Institut Ben Zvi, Jerusalem 2009. Pp. lviii + 500 + plates (Hebrew) + xxvii (French). ISBN: 978-965-342-962-8. Jacob Neusner, Theological and Philosophical Premises of Judaism (Judaism and Jewish Life). Academic Studies Press, Boston 2009. Pp. ix + 245. Price: £26.99 paperback. ISBN: 978-1934843-54-3. Old Testament Abstracts. Volume 31:3 (2008), Volume 32:1, 32:2 (2009). ISSN: 0364-8591. Rodney S. Sadler Jr., Can a Cushite Change his Skin?: An Examination of Race, Ethnicity, and Othering in the Hebrew Bible (Library of Hebrew Bible/Old Testament Studies 425). Continuum, London 2009. Pp. xii + 175. Price: £24.99 paperback. ISBN: 978-0-567-02765-8. Jonathan Schorsch, Jews and Blacks in the Early Modern World. Cambridge University Press, Cambridge 2009. Pp. Xiii + 546. Price: £21.99 paperback. ISBN: 978-0521-52723-1. Steven Schweitzer, Reading Utopia in Chronicles (Library of Hebrew Bible/Old Testament Studies 442). Continuum, London 2009. Pp. ix + 205. Price: £24.99 paperback. ISBN: 978-0-567-36317-6. Julian Stern, The Spirit of the School (Continuum Studies in Education). Continuum, London 2009. Pp. xvi + 208. Price: £70.00 hardback. ISBN: 978-1-84706-1096.
626
JOURNAL OF SEMITIC STUDIES VOLUME LV (2010) INDEX ARTICLES
ALEXANDER ANDRASON, ‘The Akkadian Iprus from the Unidirectional Perspective’ 325 GEORGE A. BEVAN and M. BARBARA REEVES, ‘A New Funerary Inscription from Humayma’ 497 GERRIT BOS, ‘Medical Terminology in the Hebrew Tradition: Shem Tov Ben Isaac, Sefer ha-Shimmush, Book 30’
53
FRANÇOISE BRIQUEL CHATONNET and ALAIN DESREUMAUX, ‘A Study and Characterization of the Syro-Malabar Script’ 407 AARON MICHAEL BUTTS and HUMPHREY HILL HARDY II, ‘A Revised Reading of a Nabataean Inscription from Umm al-Jimal’ 385 PATRICIA CRONE and ADAM SIVERSTEIN, ‘The Ancient Near East and Islam: The Case of Lot Casting’ 423 JAMES DICKINS, ‘Basic Sentence Structure in Sudanese Arabic’
237
MIRIAM GOLDSTEIN, ‘“Arabic Composition 101” and the Early Development of Judaeo-Arabic Bible Exegesis’ 451 BAHAA AMER AL-JUBOURI, ‘Nouvelles inscriptions araméennes du temple de Nannay à Hatra’
37
OLIVER KAHL, ‘Two Antidotes from the “Empiricals” of Ibn at-Tilmi∂’
79
ROBERT S. KAWASHIMA, ‘“Orphaned” Converted Tense Forms in Classical Biblical Hebrew Prose’
11
SAM LIEBHABER, ‘Rhythm and Beat: Re-evaluating Arabic Prosody in the Light of Mahri Oral Poetry’ 163 EDWARD LIPINSKI, ‘Le Gérondif en Phénicien’
1
SADOK MASLIYAH, ‘The Folk Songs of Iraqi Children: Part One’
183
SADOK MASLIYAH, ‘The Folk Songs of Iraqi Children: Part Two’
539
CYNTHIA L. MILLER, ‘Vocative Syntax in Biblical Hebrew Prose and Poetry: A Preliminary Analysis’ 347 WILLIAM MORROW, ‘“To Set the Name” in the Deuteronomic Centralization Formula: A Case of Cultural Hybridity’ 365 629
TZVI NOVICK, ‘Naming Normativity: The Early History of the Terms Sûrat ha-dîn and Lifnîm mis-sûrat ha-dîn’ 391 ABDULRAZZAK PATEL, ‘Language Reform and Controversy in the Nah∂a: Al-Shartuni’s Position as a Grammarian in Sahm 509 AARON D. RUBIN, ‘The Development of the Amharic Definite Article and an Indonesian Parallel’ 103 ABDULRAHMAN S. AL SALIMI, ‘Identifying the (Iba∂i/Omani) Siyar’
115
REVIEWS
Mariano Gómez ARANDA, Dos Comentarios de Abraham ibn Ezra al Libro de Ester: Edición crítica, traducción y estudio introductorio 298 (Aaron D. RUBIN) Leora BATNITZKY, Leo Strauss and Emmanual Levinas: Philosophy and 304 the Politics of Revelation (A.H. LESSER) Dörte BESTER, Körperbilder in den Psalmen (Sue GILLINGHAM)
597
Dvora BREGMAN, The Golden Way: The Hebrew Sonnet during the 300 Renaissance and the Baroque (Arie SCHIPPERS) Billie Jean COLLINS, The Hittites and their World (Charles BURNEY)
589
Edward M. COOK, A Glossary of Targum Onkelos: According to Alexander Sperber’s Edition (Aaron D. RUBIN) 612 Stephanie DALLEY, Esther’s Revenge at Susa: From Sennacherib to Ahasuerus (Roger TOMES) 599 Timothy EDWARDS, Exegesis in the Targum of the Psalms: The Old, the 297 New, and the Rewritten (David M. STEC) Israel FINKELSTEIN and Amihai MAZAR, edited by Brian B. SCHMIDT, The Quest for the Historical Israel: Debating Archaeology and the History 593 of Early Israel (Philip R. DAVIES) Elizabeth FROOD, Biographical Texts from Ramesside Egypt (Karl 265 JANSEN-WINKELN) Robert GLEAVE, Scripturalist Islam: The History and Doctrines of the 616 Akhbari Shi’i School (Andrew J. NEWMAN) Lester GRABBE (ed.), Israel in Transition: From Late Bronze II to Iron IIa (c. 1250-850 B.C.E.) Volume 1. The Archaeology (Matthew WHINCOP) 595 Karel JONGELING, Handbook of Neo-Punic Inscriptions (Edward LIPINSKI)
603
P.S.F. VAN KEULEN and W.Th. VAN PEURSEN (eds), Corpus Linguistics and Textual History: A Computer-Assisted Interdisciplinary Approach to 280 the Peshitta (Jerome A. LUND) 630
Geoffrey KHAN, Arabic Documents from Early Islamic Khurasan (C. Edmund BOSWORTH) Wilfred G. LAMBERT, Babylonian Oracle Questions (Alasdair LIVINGSTONE) Jill MIDDLEMAS, The Templeless Age: An Introduction to the History, Literature and Theology of the ‘Exile’ (Ralph W. KLEIN) Khaleel MOHAMMED and Andrew RIPPIN (eds), Coming to terms with the Qur’an. A volume in honor of Professor Issa Boullata (Stefan WILD) John A. MORROW (ed.), Arabic, Islam and the Allah Lexicon: How Language Shapes our Conception of God (Gerhard BOWERING) Christa MÜLLER-KESSLER, Die Zauberschalentexte in der HilprechtSammlung, Jena und weitere Nippur-Texte anderer Sammlungen (Matthew MORGENSTERN) Ehud NETZER, The Architecture of Herod, the Great Builder (Kay PRAG) Christophe NIHAN, From Priestly Torah to Pentateuch (Calum CARMICHAEL) Robert PHENIX, The Sermons on Joseph of Balai of Qenneshrin: Rhetoric and Interpretation in Fifth Century Syriac Literature (John W. WATT) Peter E. PORMANN (ed.), Rufus of Ephesus: On Melancholy (Oliver KAHL) Kay PRAG, Excavations by K. M. Kenyon in Jerusalem 1961 1967: Volume V. Discoveries in Hellenistic to Ottoman Jerusalem (A. Asa EGER) Shmuel SAFRAI lèz, Zeev SAFRAI, Joshua SCHWARTZ, and Peter J. TOMSON (eds), The Literature of the Sages, Second Part: Midrash and Targum, Liturgy, Poetry, Mysticism, Contracts, Inscriptions, Ancient Science and the Languages of Rabbinic Literature (Ephraim NISSAN) Avi SAGI, The Open Canon: On the Meaning of Halakhic Discourse (Azzan YADIN) William M. SCHNIEDEWIND and Joel H. HUNT, A Primer on Ugaritic: Language, Culture and Literature (Wilfred G.E. WATSON) Andrzej S. TURKANIK, Of Kings and Reigns: A Study of Translation Technique in the Gamma/Gamma Section of 3 Reigns (1 Kings) (J.M. DINES) Rainer VOIGT (ed.), From Beyond the Mediterranean: Akten des 7. Internationalen Semitohamitistenkongresses (VII. ISHaK), Berlin 13, bis 15, September 2004 (Aaron D. RUBIN) James W. WATTS, Ritual and Rhetoric in Leviticus: From Sacrifice to Scripture (Walter J. HOUSTON) Archie T. WRIGHT, The Origin of Evil Spirits (Philip R. DAVIES) 631
618 591 271
306 308
282 601 270 614 278 274
293 291 266
609
263 269 273
SHORT NOTES
Fiona C. BLACK, The Recycled Bible: Autobiography, Culture, and the 314 Space Between (Philip R. DAVIES) G. BOCCACCINI (ed.), Enoch and the Messiah Son of Man: Revisiting 622 the Book of Parables (Maurice CASEY) Mark J. BODA, Daniel K. FALK and Rodney A. WERLINE (eds), Seeking the Favor of God. Volume 1: The Origins of Penitential Prayer in Second Temple Judaism; Volume 3: The Impact of Penitential Prayer beyond 316 Second Temple Judaism (George J. BROOKE) Brian A. CATLOS, The Victors and the Vanquished: Christians and Mus623 lims of Catalonia and Aragon, 1050 1300 (Anthony John LAPPIN) Lorenzo DITOMASSO, The Dead Sea New Jerusalem Text: Contents and 318 Contexts (Charlotte HEMPEL) Bart D. EHRMAN, Whose Word is It? The Story Behind Who Changed 321 the New Testament and Why (George J. BROOKE) Frances FLANNERY, Colleen SHANTZ and Rodney A. WERLINE (eds), Experientia, Volume 1: Inquiry into Religious Experience in Early Judaism and Early Christianity (George J. BROOKE) Andrew D. GROSS, Continuity and Innovation in the Aramaic Legal Tradition (Stephen A. KAUFMAN) Hermann GUNKEL (edited and translated by K.C. Hanson), Israel and Babylon: The Babylonian Influence on Israelite Religion (Roger TOMES) Antti LAATO and Jacques VAN RUITEN (eds), Rewritten Bible Reconsidered: Proceedings of the Conference in Karkku, Finland, August 24 26 2006 (George J. BROOKE) Bernard M. LEVINSON, ‘The Right Chorale’: Studies in Biblical Law and Interpretation (Walter J. HOUSTON) Pekka LINDQVIST, Sin at Sinai: Early Judaism Encounters: Exodus 32 (Mila GINSBURSKAYA) Simo PARPOLA, Robert WHITING et al. (eds), Assyrian-English-Assyrian Dictionary (Jonathan TAYLOR) Julius Heinrich PETERMANN, The Great Treasure or Great Book, commonly called “The Book of Adam”, the Mandaeans’ work of Highest Authority (Matthew MORGENSTERN) Melvin K.H. PETERS (ed.), XIII Congress of the International Organization for Septuagint and Cognate Studies: Ljubljana, 2007 (George J. BROOKE) 632
320 315
621
319 312 311 621
321
317
Benjamin RICHLER, Hebrew Manuscripts in the Vatican Library Cata322 logue (Renate SMITHUIS) David T. RUNIA and Gregory E. STERLING (eds), The Studia Philonica Annual: Studies in Hellenistic Judaism Volume XX.2008 (George 320 J. BROOKE) Seth S. SANDERS (ed.), Margins of Writing, Origins of Cultures (Alasdair 323 LIVINGSTONE) James T. SPARKS, The Chronicler’s Genealogies: Towards an Under313 standing of 1 Chronicles 1 9 (Brian E. KELLY) Giancarlo TOLONI, La sofferenza del giusto: Giobbe e Tobia a confronto 315 (George J. BROOKE) Sidnie WHITE CRAWFORD, Rewriting Scripture in Second Temple Times 318 (Ariel FELDMAN) REVIEWERS Bosworth, C.E., Khan Bowering, G., Morrow Brooke, G.J., Boda, Falk and Werline, Ehrman, Flannery, Shantz and Werline, Laato and van Ruiten, Peters, Runia and Sterling, Toloni Burney, C., Collins Carmichael, C., Nihan Casey, M., Boccaccini Davies, P.R., Black, Finkelstein and Mazar, Wright Dines, J.M., Turkanik Eger, A.A., Prag Feldman, A., White Crawford Gillingham, S., Bester Ginsburskaya, M., Lindqvist Hempel, C., DiTomasso Houston, W.J., Levinson, Watts Jansen-Winkeln, K., Frood Kahl, O., Pormann Kaufman, S.A., Gross Kelly, B.E., Sparks Klein, R.W., Middlemas Lappin, A.J., Catlos Lesser, A.H., Batnitzky Lipinski, E., Jongeling Livingstone, A., Lambert, Sanders Lund, J.A., van Keulen and van Peursen 633
Morgenstern, M., Müller-Kessler, Petermann Newman, A.J., Gleave Nissan, E., Safrai lèz, Safrai, Schwartz, and Tomson Prag, K., Netzer Rubin, A.D, Aranda, Cook, Voigt Schippers, A., Bregman Smithuis, R., Richler Stec, D.M., Edwards Taylor, J., Parpola, Whiting et al. Tomes, R., Dalley, Gunkel Watson, W.G.E., Schniedewind and Hunt Watt, J.W., Phenix Whincop, M., Grabbe Wild, S., Mohammed and Rippin Yadin, A., Sagi
634
93397_JOSS_2010/2_Cv
24-06-2010
15:19
Page 1
ISSN 0022-4480 (PRINT) ISSN 1477-8556 (ONLINE)
365 385
391 407 423
451 479 497 509 539 589 621
Semitic Studies
SHORT NOTICES
347
JOURNAL OF
REVIEWS
325
AUTUMN 2010
ALEXANDER ANDRASON ‘The Akkadian Iprus from the Unidirectional Perspective’ CYNTHIA L. MILLER ‘Vocative Syntax in Biblical Hebrew Prose and Poetry: A Preliminary Analysis’ WILLIAM MORROW ‘“To Set the Name” in the Deuteronomic Centralization Formula: A Case of Cultural Hybridity’ AARON MICHAEL BUTTS and HUMPHREY HILL HARDY II ‘A Revised Reading of a Nabataean Inscription from Umm al-Jimal’ TZVI NOVICK ‘Naming Normativity: The Early History of the Terms Sûrat ha-dîn and Lifnîm mis-sûrat ha-dîn’ FRANÇOISE BRIQUEL CHATONNET and ALAIN DESREUMAUX ‘A Study and Characterization of the Syro-Malabar Script’ PATRICIA CRONE and ADAM SIVERSTEIN ‘The Ancient Near East and Islam: The Case of Lot Casting’ MIRIAM GOLDSTEIN ‘“Arabic Composition 101” and the Early Development of JudaeoArabic Bible Exegesis’ OLIVER KAHL ‘Two Antidotes from the “Empiricals” of Ibn at-Tilmi∂’ GEORGE A. BEVAN and M. BARBARA REEVES ‘A New Funerary Inscription from Humayma’ ABDULRAZZAK PATEL, ‘Language Reform and Controversy in the Nah∂a: Al-Shartuni’s Position as a Grammarian in Sahm SADOK MASLIYAH ‘The Folk Songs of Iraqi Children: Part Two’
JOURNAL OF
Semitic Studies VOLUME LV. NO. 2 AUTUMN 2010
VOLUME LV. NO. 2
93397_JOSS_2010/2_Cv
24-06-2010
15:19
Page 2
SUBSCRIPTIONS A subscription to Journal of Semitic Studies comprises 2 issues. Prices include postage by surface mail, or for subscribers in the USA and Canada by airfreight, or in India, Japan, Australia and New Zealand, by Air Speeded Post. Airmail rates are available on request. Annual Subscription Rate (Volume LIII, 2 issues, 2008) Institutional Print edition and site-wide online access: £166//249/US$315 Print edition only: £152//228/US$289 Site-wide online access only: £138//207/US$262 Personal Print edition and individual online access: £54//81/US$103 Please note: £ Sterling rates apply in UK, / rates apply in Europe, US$ elsewhere There may be other subscription rates available, for a complete listing please visit www.jss.oupjournals.org/ subinfo. Full prepayment, in the correct currency, is required for all orders. Orders are regarded as firm and payments are not refundable. Subscriptions are accepted and entered on a complete volume basis. Claims cannot be considered more than FOUR months after publication or date of order, whichever is later. All subscriptions in Canada are subject to GST. Subscriptions in the EU may be subject to European VAT. If registered, please supply details to avoid unnecessary charges. For subscriptions that include online versions, a proportion of the subscription price may be subject to UK VAT. Personal rate subscriptions are only available if payment is made by personal cheque or credit card and delivery is to a private address. The current year and two previous years’ issues are available from Oxford University Press. Previous volumes can be obtained from the Periodicals Service Company, 11 Main Street, Germantown, NY 12526, USA. Email: [email protected]. Tel: +1 (518) 537 4700. Fax: +1 (518) 537 5899. For further information, please contact: Journals Customer Service Department, Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, UK. Email: [email protected]. Tel (and answerphone outside normal working hours): +44 (0)1865 353907. Fax: + 44 (0)1865 353485. In the US, please contact: Journals Customer Service Department, Oxford University Press, 2001 Evans Road, Cary, NC 27513, USA. Email: [email protected]. Tel (and answerphone outside normal working hours): 800 852 7323 (toll-free in USA/Canada). Fax: 919 677 1714. In Japan, please contact: Journals Customer Services, Oxford University Press, 1-1-17-5F, Mukogaoka, Bunkyo-ku, Tokyo, 113-0023, Japan. Email: [email protected]. Tel: (03) 3813 1461. Fax: (03) 3818 1522. Methods of payment. (i) Cheque (payable to Oxford University Press, to Oxford University Press, Cashiers Office, Great Clarendon Street, Oxford OX2 6DP, UK) in GB£ Sterling (drawn on a UK bank), US$ Dollars (drawn on a US bank), or EU/ Euros. (ii) Bank transfer to Barclays Bank Plc, Oxford Group Office, Oxford (bank sort code 20-65-18) (UK), overseas only Swift code BARC GB 22 (GB£ Sterling to account no. 70299332, IBAN GB89BARC20651870299332; US$ Dollars to account no. 66014600, IBAN GB27BARC 20651866014600; EU/ Euros to account no. 78923655, IBAN GB16BARC20651878923655). (iii) Credit card (Mastercard, Visa, Switch or American Express). Journal of Semitic Studies is published two times a year in April and October by Oxford University Press, Oxford, UK. Annual subscription price is £166//249/US$315. Journal of Semitic Studies is distributed by Mercury International, 365 Blair Road, Avenel, NJ 07001, USA. Periodicals postage paid at Rahway, NJ and at additional entry points. US Postmaster: send address changes to Journal of Semitic Studies (ISSN 0022-4480), c/o Mercury International, 365 Blair Road, Avenel, NJ 07001, USA. PERMISSIONS For information on how to request permissions to reproduce articles/information from this journal, please visit www.oxfordjournals.org/permissions. ADVERTISING Inquiries about advertising should be sent to Helen Pearson, Oxford Journals Advertising, PO Box 347, Abingdon OX14 1GJ, UK. Email: [email protected]. Tel: +44 (0)1235 201904; Fax: +44 (0)8704 296864. DISCLAIMER Statements of fact and opinion in the articles in Journal of Semitic Studies are those of the respective authors and contributors and not of Journal of Semitic Studies or Oxford University Press. Neither Oxford University Press nor Journal of Semitic Studies make any representation, express or implied, in respect of the accuracy of the material in this journal and cannot accept any legal responsibility or liability for any errors or omissions that may be made. The reader should make his/her own evaluation as to the appropriateness or otherwise of any experimental technique described. © The University of Manchester 2008 All rights reserved; no part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise without prior written permission of the Publishers, or a licence permitting restricted copying issued in the UK by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1P 9HE, or in the USA by the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923. Typeset by Drukkerij Peeters, Herent, Belgium Printed by Bell & Bain Ltd, Glasgow, UK
INFORMATION FOR AUTHORS Contributions, books for review, and other editorial communications should be addressed to: The Editors Journal of Semitic Studies Middle Eastern Studies School of Languages, Linguistics and Cultures University of Manchester Manchester M13 9PL UK Article Length Editorial policy is normally to favour shorter articles, i.e. no longer than 10,000 words. Typescript layout Articles and reviews should be typed on one side of the paper (A4 or 8.5 x 11 inches) with double spacing and generous margins on one side of the paper only. Authors of articles should submit TWO COPIES, and disk, if available. Non-Roman script is to be written as clearly as possible and used as sparingly as possible. Notes should be numbered consecutively, double spaced and placed at the end. The Editors will not accept for publication any typescripts which do not conform to these standards. For detailed instructions please see the style sheet. Abstracts Authors of articles are required to provide an abstract, which should be a single paragraph, not exceeding 150 words. Submission We will also normally require the final revised version to be supplied in hard copy and electronically. Whilst we can accept most wordprocessor formats the preferred option is Word. The files should be submitted in the wordprocessor's native format and as an RTF (Rich Text Format). Illustrations Electronic submission of illustrations. We can accept 3.5-inch floppy disks, CDROMs and email attachments. High-quality print-outs of all figures are still essential, as they will be scanned for reproduction in the event of problems with the electronic files. Figures must be saved as high-resolution TIF files (at least 300 pixels per inch for photographs, and 1200 pixels per inch for black and white line drawings). Offprints and Proofs Authors of articles will receive one copy of the issue containing their article and twenty-five offprints of their article free of charge. Alteration on the proofs, other than the correction of printer's errors, may be disallowed. Copyright and Permissions Manuscripts submitted will be expected to contain original work and should not have been published in abridged or other form elsewhere. It is a condition of publication in the Journal that authors grant an exclusive licence to The University of Manchester. Any requests from third parties to reproduce articles are handled on behalf of the journal by Oxford University Press. This will also allow the article to be as widely disseminated as possible and will protect the rights of the author and OUP. In granting an exclusive licence, authors may use their own material in other publications provided that the journal is acknowledged as the original authority for publication, and Oxford University Press is notified in writing and in advance.