Journal of the Economic and Social History of the Orient [47/1]

Table of contents :
Article Contents
[unnumbered]
Article Contents
[unnumbered]

Citation preview

JOURNAL OF THE ECONOMIC AND SOCIAL HISTORY OF THE ORIENT Editor NORMAN

YOFFEE

of Michigan,

(University

Ann Arbor)

Editorial Board BARBARA

W.

of Hawaii, Honolulu)?JEAN-PAUL WASHBROOK (CNRS, Aix-en-Provence)?DAVID (University

PASCUAL

ANDAYA

(University

of Oxford)?HARRIET ZURNDORFER (Leiden University) Advisory

Alam

Muzaffar

Board

of Chicago)?Daud Ali (SOAS, London)? Jos Gommans (Leiden University)?Ben Haring Lapidus (Leiden University)?Ira de Pee (Brooklyn College)? of California, (University Berkeley)?Christian Klaas Veenhof (Leiden University) (University

Book Review Editor Gottfried

Hagen

(University

Editorial

Brad

Crowell

of Michigan)

Assistant

of Michigan)

(University

All manuscripts, editorial correspondence, and books for review should be sent to: Norman Yoffee, of Near Eastern of Studies, Frieze Bldg., University Dept. Ann Arbor, MI 48109-1285, USA. JESHO is a peer-refereed Michigan, journal. Correspondence may be sent to [email protected]. Subscriptions

The Journal of the Economic and Social History of the Orient is published quarterly. Subscriptions may be sent to the Subscription Department, Brill, P.O. Box 9000, 2300 PA The

Leiden,

Netherlands.

Subscription price of Volume 47 (2004) (ca. 592 pages in 4 issues): EUR 249.0CVUS$309.00 for institutions, and EUR 69.00/US$ 85.00 Price includes online subscription.

for private subscribers, Claims for missing

and han postage including or damaged issues must be or within within three months of publication five months of pub customers) (European to Brill's lication outside Please address claims and enquiries (customers your Europe). Customer Services Department.

dling. made

Postmaster send address corrections to the JOURNAL OF THE ECONOMIC AND SOCIAL HISTORY OF THE ORIENT, 2323 Randolph Ave., Avenel, NJ, 07001. All prices and postage & handling charges are exclusive of VAT applicable outside the EC). Brill Leiden ? All

Copyright

2004

by Koninklijke

in EC-countries

(VAT not

Boston Brill NV,

Leiden,

The Netherlands

stored in a retrieval translated, may be reproduced, rights reserved. No part of this publication or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the publisher.

system,

Authorization appropriate

to photocopy use is granted by Brill items for internal or personal are paid directly to Copyright Clearance Center, 222 Rosewood Danvers, MA 01923, USA. Fees are subject to change.

fees

This

journal

is printed

PRINTED

on

acid-free

that the provided Suite 910, Drive,

paper.

IN THE NETHERLANDS

This content downloaded from 185.44.78.113 on Tue, 17 Jun 2014 08:09:41 AM All use subject to JSTOR Terms and Conditions

Jorge

Flores

Succession

and

Sanjay

and Imposture

The Shadow Subrahmanyam, . in the Mughal 1628-1640 Empire,

Sultan: 80

Reviews ZAINAB BAHRANI, The Graven . (Andrea Seri) Assyria

Image: 122

Representation

in Babylonia

and

BONGENAAR (editor), Interdependency and Pri of Institutions vate Entrepreneurs. the 2nd MOS Proceedings of Symposium, Leiden, . December 125 11-12, 1998 (Anne Goddeeris) the assistance of Fiorella Horst Klengel with Volkert Imparati,

A.C.V.M.

HASS

and THEO P.J. VAN den

(Richard H. Beal) NAGUIB KANAWATI, Conspiracies . 134 (Petra Vlckova) Reiches

.

Geschichte Hout, 128 in the Egyptian

Palace.

des

hethitischen

Unis

to Pepy

I

in the Middle East, House (editor), Family History . and Gender 137 hold, Property (Nelly Hanna) in Southeast Asia (William H. TONY Day, Fluid Iron: State Formation 139 Frederick).

BESHARA DOUMANI

ISSN 0022-4995

This content downloaded from 62.122.76.54 on Tue, 17 Jun 2014 16:24:19 PM All use subject to JSTOR Terms and Conditions

SHEPHERDS, MERCHANTS, AND CREDIT: SOME OBSERVATIONS ON LENDING PRACTICES IN UR III MESOPOTAMIA BY

STEVEN J. GARFINKLE* Abstract The numerous surviving loan documents from the Ur III period illustrate the vital role that credit played in the early Mesopotamian economy. The availability of credit was critical at every level of society. In this article, the form and function of credit in the Ur III period are described, along with lending practices and the activities of individual creditors. Despite the overwhelming scale of the institutional economies, there was signiŽ cant room for non-institutional households to pursue economic gains through money-lending. This entrepreneurial activity took place in an economy that was familiar with a sophisticated range of possible credit transactions. Les nombreux documents de pr t qui survivent de lՎ poque de la Troisi me Dynastie dÕUr illustrent le r™le essentiel que le crŽ dit joue dans lՎ conomie mŽ sopotamienne archa•que. Ë chaque niveau de la sociŽ tŽ disposer de crŽ dit Ž tait dÕune importance essentielle. Cette contribution dŽ crit les forme et fonction du crŽ dit ˆ la pŽ riode dÕUr III, les pratiques de pr t et les activitŽ s des crŽ anciers. En dŽ pit de la grande taille des Ž conomies institutionnelles, il existait un secteur important o des maisons non-institutionelles pouvaient faire des gains en pr tant ˆ intŽ r t. Cette activitŽ dÕentrepreneur sÕexer ait dans une Ž conomie qui possŽ dait une gamme sophistiquŽe de transactions de crŽ dit. Keywords: Ur III, credit, customary loans, antichretic loans, productive and consumptive loans

* Steven J. GarŽ nkle, History Department/ MS 9061, Western Washington University, Bellingham, WA 98225-9061. steven.garŽ [email protected] This article is a considerable revision and expansion of a paper presented at the 210th Annual meeting of the American Oriental Society. In an article on the subject of debt, it is a particular privilege to recognize some of my own. In the preparation of the original paper, I was fortunate to have the assistance of Seth Richardson (who also read and commented on a draft of this lengthier article) and Jason Freitag, each of whom suggested signiŽ cant improvements. Niek Veldhuis also read a copy of that early paper and provided some very productive criticism. This article has beneŽ ted enormously from the advice and guidance of Marc Van De Mieroop, who has always been an extremely forgiving creditor. I am also grateful to Norman Yoffee and the two reviewers of this article for their comments and assistance. Much of my discussion of credit is adapted from my dissertation, Private Enterprise in Babylonia at the End of the Third Millennium B.C. (Ph.D. Diss. Columbia University, 2000), which includes a more complete analysis of debt and non-institutional archives in the Ur III period. My revised dissertation will appear as a volume of Cornell University Assyriological Studies (CUAS). © Koninklijke Brill NV, Leiden, 2004 Also available online – www.brill.nl

JESHO 47,1

2

STEVEN J. GARFINKLE

INTRODUCTION Few topics are as richly documented in the cuneiform record as the practice of borrowing and lending; and few eras are better documented than the Ur III period at the end of the third millennium BC.1 In a largely agrarian society, the availability of credit is a functional necessity. The Ur III period was no exception, and lending in this era was characterized by both individual and institutional creditors. At the same time, our large collection of texts on credit are among the most laconic in cuneiform literature. Analyzing loan documents brings with it the further problem of assessing economic motivation in antiquity. Only rarely do our texts tell us why two parties entered into a loan agreement; and we cannot even be certain what the survival of the tablets indicates. 2 This article offers a general assessment of credit practices in southern Mesopotamia at the end of the third millennium BC, and in doing so it examines the nature of the surviving documentation. We are assisted in this endeavor both by the size of the corpus, and by the tremendous amount of scholarly work that has been devoted to understanding the Ur III state. Therefore, there are advantages to analyzing lending during this period, and these extend beyond improving our knowledge of that particular era. This study results in a better understanding not only of the operation of the Ur III economy, but also of broader socio-economic developments in early Mesopotamia. Credit transactions played a signiŽ cant role in the institutional and non-institutional economies of southern Mesopotamia at the end of the third millennium BC. The activities of creditors frequently defy easy categorization according to modern economic terms. Therefore, we cannot simply apply the terminology of ÒpublicÓ and ÒprivateÓ to the economic behavior of entrepreneurs in ancient Mesopotamia. Despite the overwhelming scale of the institutional economies of the Ur III period, there was still signiŽ cant room for the economic interests of non-institutional households pursuing their own gain. Moreover, the persistence of local and regional customs and hierarchies presented opportunities for individual entrepreneurs acting as creditors. To begin, we need a working deŽ nition of what constitutes a loan, in order to distinguish loan documents from other types of documents, such as simple

1

2112-2004 BC according to the Middle Chronology. We need to be cautious about making assumptions when we encounter familiar practices, such as lending, in antiquity. As Steinkeller (2001: 48) recently noted, Ò. . . although ancient loaning practices may appear to share formal similarities among themselves, and may seem to be formally identical with or at least analogous with modern loaning practices, one must assume that in each case we are dealing with a different phenomenon, whose precise nature is determined by the economic and social context in which it occurs.Ó 2

SHEPHERDS, MERCHANTS, AND CREDIT

3

receipts and other transfers of property, that often used similar terminology. The fundamental difference between these classes of textual material was the expectation of repayment. A loan document can be readily deŽ ned as any agreement that required the repayment of the principal to the creditor. The following discussion of credit is divided into four parts. First, typologies will be established for both the form and the function of loans. Second, the terminology of the loan documents themselves will be discussed. Third, lending practices will be examined. In particular, I will describe those practices related to interest and to the preservation of loan documents. Finally, the behavior of creditors will be described, focusing on the lending activities of non-institutional creditors.3 A TYPOLOGY OF LOANS IN THE UR III PERIOD The loans that were issued in the Ur III period took three primary forms: customary loans, interest-free loans, and antichretic loans. 4 Within these categories of loans we can also further distinguish between two types of loans on the basis of their function: consumptive loans and productive loans. A Typology of form: Customary, Interest-free, and Antichretic The largest category of loan documents was customary loans, which were issued in a variety of commodities, including silver, barley, and wool, and required the repayment of the principal along with interest. The basic characteristic of customary loans was that they employed a standard terminology (discussed in detail below in the section on terminology). In some cases the time of repayment was expressly indicated in the text, but this was clearly not a requirement.5 3 The reader should be alerted that I have chosen to avoid the language of private and public as far as possible. The use of modern terminology can be misleading in discussions of the ancient economy. Describing a Mesopotamian temple as a participant in the public sector of the economy may promote the idea that that temple behaved in the same manner as a public sector entity in our own economy. Such a view makes it more difŽ cult to assess the operation of an ancient institution in its own historical context. I use the terms institutional for what we traditionally view as the public sector and non-institutional for what we would regard as the private sector. I have discussed this issue at length elsewhere; see, for example, my forthcoming chapter in the Blackwell Companion to the Ancient Near East. 4 There are additional categories of loans attested in ancient Mesopotamia, such as bottomry loans; however, these other types of loans are rare in the Ur III texts. There are also other categories of texts, such as promissory notes and lawsuits that further attest to lending activities. 5 A clause requiring repayment by a speciŽ c date appears most frequently in interest-free

4

STEVEN J. GARFINKLE

In contrast to the Ž rst category of loans, interest-free loans obviously lacked a provision for the payment of interest. In many of the cases where a date of repayment was indicated, the loans were interest-free loans, and interest-free loans in the Ur III period were usually issued for short durations, often not exceeding one month. 6 Some of these loans, issued in barley, may have been attempts by the debtors to alleviate temporary crises within their households. The creditors in such cases may have been a member of the debtorÕs extended family or the debtorÕs superior within a patrimonial or professional organization. Many loans that did not indicate interest were probably Ž ctitious arrangements in which goods were ÒloanedÓ to a craftsman for production. 7 Such work loans were really receipts documenting a work order and the transfer of raw materials. Frequently, these texts did not detail the arrangements of the agreement, but they often provided penalties for a failure to deliver the Ž nished goods. 8 The regular penalty was a payment of twice the value of the original loan. These Òdoubling clausesÓ were very common in the Ur III period, and such penalties may have originated in commercial transactions for the purchase

loans, which were ordinarily of a short duration (for example: MVN 8 165, 166, 167, and 171), and in consumptive, or harvest, loans (for example: MVN 8 153, and 154; ZA 93/2 4; MVN 13 896 + 897), which usually required repayment after the harvest. 6 TMHC NF 1/2 31 is an example of a one month loan that was advanced without interest: 12 1/2 g’n k -babbar m‡á nu-tuku/ ki Lœ-dEN.ZU-ta/ Lœ-áa6-ga/ áu ba-ti/ Witnesses/ iti áu-numun-a u4 1 ba-zal/ iti sig4 gi4-gi4-dam/ mu-lugal-bi “-pad4/ Year/ Seal: Lœ-áa6-ga/ dub-sar Lu-áaga received 12 1/2 g’n of silver without interest from Lu-Suen on the Ž rst day of the 3rd month. He swore in the name of the king to return it in the 4th month. Witnesses. Year. Seal: Lu-áaga, the scribe. There are numerous examples of such short-term loans. See also YOS 4 51, 52. Within the Nippur texts, the 4th month was a frequent month for the repayment of these loans, and this phenomenon requires further study. 7 On the question of work contracts and related legal and economic problems in the noninstitutional economy, see Neumann 1996: 254-64. 8 For example, in the transaction shown below, a silversmith was the debtor, and the customer who advanced him the silver was the creditor. The silversmith was to return all of this silver by a certain month. If he did not return the silver (presumably in the form of a worked object) he had to weigh out barley to the creditor after the harvest. The penalty amounted to a doubling of the original loan. NATN 266: 1/3 (áa) 5 g’n k -babbar/ ki ç-zi-da-ta/ °u-aá-l’ áu ba-ti/ iti apin-du8 NibrukiᏠ/ k -babbar sag-bi gi4-gi4-dam/ tukum-bi nu-gi4/ egir buru14-Ꮰ2.0.0 áe gur-ta/ “-‡g-e/ Witnesses/ Date/ Seal: °u-‡á-li k -simug/ dumu °ar-ru-um-ba-ni °u-aáli received 25 g’n of silver from Azida. He will return the silver in its entirety in month 8 in Nippur. If he does not return it, he will weigh out 2 gur of barley for each shekel of silver after the harvest. Witnesses. Date. Seal: °u-aáli, the silversmith, son of °arrum-bani.

SHEPHERDS, MERCHANTS, AND CREDIT

5

of goods or the commissioning of Ž nished products. Such transactions, which did not originally involve the charging of interest, may have necessitated such a signiŽ cant penalty. 9 Antichretic loans were characterized by arrangements in which borrowers pledged labor as interest on a loan. In antiquity such agreements covered a wide variety of possibilities, including the promise of the labor of dependents of the borrower. The category of antichretic loans can be divided into two broad groups. In all of the antichretic agreements, the creditor sought to acquire labor; however, in some cases the labor involved was that of an individual while in other cases the labor involved was that of a group. The Ž rst set of antichretic loans more frequently included advances in silver. The labor pledged as interest, or as surety, on these loans was that of a skilled dependent laborer, or a slave, of the debtor. 10 The second set of antichretic loans usually involved advances of barley, and the interest was paid in agricultural labor controlled by the debtor. 11 A Typology of function: Productive and Consumptive All interest bearing loans in the Ur III period can be divided into two broad categories on the basis of their function: consumptive loans and productive loans. 12 The impetus behind these loans was very different. The majority of the extant loans from the Ur III period were advances of barley or silver, and we can draw a rough distinction between consumptive and productive loans on the basis of the commodity involved. Productive loans were usually made in silver, and consumptive loans were advanced in grain. Both types of loans could be received on an antichretic basis. Productive loans were sought by debtors because the capital requested would be used to improve the material circumstances of the household. Essentially, then, the interest on such loans was the cost of these improvements. In these arrangements, both the creditors and the debtors were seeking some economic advantage from the transaction, but neither was coerced into the transaction. In consumptive loans the debtor was seeking the loan in order to provide for the immediate subsistence of his own household. These loans were almost always of shorter duration, and were most often harvest loans. That is, loans

9 For an example of a probable purchase agreement involving a doubling penalty, see TMHC NF 1/2 63. For a discussion of these penalties, see Limet 1969. 10 See, for example, MVN 8 168, BE 3/1 19 (which appears to have had a duration of 4 years), and NATN 366. 11 See, for example, JCS 23/4 8, Eames Coll. TT 11, and TIM 3 149. 12 For a similar categorization of loans, see Van De MieroopÕs (2002) summary of credit in Mesopotamia where he differentiates between agricultural and commercial loans.

6

STEVEN J. GARFINKLE

that had to be repaid, with interest, out of the next harvest. However, harvest loans could, on occasion, also be productive. 13 The usual interest rates in the Ur III period were 33% for barley loans and 20% for silver loans. The discrepancy resulted from the fact that most of the barley loans were undertaken during the few months prior to the harvest, when barley was particularly scarce, and these loans were repaid after the harvest, when barley was plentiful and at its lowest value relative to silver.14 The 33% interest rate charged on barley loans was designed to preserve the real value of the loan for the creditor. This does not mean that silver loans could not be consumptive, but only that barley loans were more frequently given for short durations. In a recent article, Piotr Steinkeller (2002) has offered a detailed assessment of lending practices in Ur III Babylonia. I Ž nd his description of the typical money lender especially attractive. His portrait of the creditor indicates that he was a professional, often associated with the royal sector, who held a áuku allotment and had the resources, in barley, silver, and draft animals, to cultivate his allotment and to pursue the control of additional allotments (Steinkeller 2002: 117). What this individual lacked, according to Steinkeller, was the necessary labor to work the land that he could acquire. This raises a crucial issue for observers of the Ur III economy: the shortage of available labor. 15 Our recognition of this problem is central to our understanding of the Ur III economy. In the absence of a free labor market, the scarcity of agricultural labor placed stress on households engaged in the agrarian economy. However, I am convinced that Steinkeller (2002: 118) has gone too far in asserting, regarding the non-institutional money lender, that Òmost of his lending operations were undoubtedly directed towards the acquisition of labor and land.Ó The conclusion that most of the non-institutional loans in the Ur III period were issued in the hope, not of repayment, but of acquiring the labor of the debtor is dependent on four assumptions. First, that there was an acute shortage of agrarian labor available to individual households. Second, that most of the debtors were relatively poor and not in a position to meet their Ž nancial oblig-

13 See ZA 93/2 4. This harvest loan involved an amount of grain too large to have been intended for the subsistence of a single household. (See also below, note 18.) 14 See Van De Mieroop 2002: 84-5 for another perspective on the origins of the different interest rates. 15 The scarcity of labor in the Ur III period has also been discussed by van Driel, see van Driel 1998 and 1999. Van Driel has detailed the way in which labor, which was only extensively required during speciŽ c times of the year, became the limiting factor in the Ur III economy.

SHEPHERDS, MERCHANTS, AND CREDIT

7

ations to their creditor.16 Third, that a signiŽ cant percentage of the loans issued in the Ur III period were not repaid. Fourth, that the lenders were not motivated by the desire to proŽ t from the payment of interest but hoped to gain from the eventual default of the debtor. Only the Ž rst of these assumptions is warranted by the evidence from the Ur III period. The non-institutional archives from the Ur III period indicate that the borrowers were frequently members of the highest echelons of their society.17 These borrowers were not subsistence farmers living perilously close to poverty and the necessity of debt bondage. We can only speculate as to why prominent ofŽ cials and af uent individuals would have required loans in the Ur III period, but they undoubtedly sought out credit. Some of our confusion may arise from the modern assumption that the interest rates charged in the Ur III period were usurious, but we are not sufŽ ciently familiar with Sumerian society to know if this was the case or not. We also know that interest bearing loans were extended to family members.18 There is no reason to suspect that such loans were advanced to bring about the ruin of the creditorÕs relative. The relationship between SI.A-a and his brother NaÀarum may be an isolated example, but I rather suspect that it was close to the norm of creditor and debtor relations in the Ur III period, and was therefore not exceptional. We will never be able to decide what percentage of loans were not repaid in the Ur III period, but the mere preservation of the loan documents in the archaeological record should not be taken as evidence of default.19 Loans survived in considerable numbers for two reasons. First, loans were issued for a

16 This suggestion rests in part on our knowledge of the debt slavery that existed in the Old Babylonian period, and which may have precipitated the royal edicts of debt remission in that era. In this sense, the patterns of the Ur III period are perceived as a precursor to the growth of debt slavery in the succeeding era. At the same time, attestations of debt slavery in the Old Babylonian period are not taken as indications of the absence of commercial or productive loans. 17 SI.A-a, a chief shepherd, was a prominent creditor in the Ur III period, and his debtors included members of the military hierarchy, such as ugula-geátas (MVN 13 890 + 891) and nu-bˆndas (TIM 3 149), as well as prominent ofŽ cials of the temple estates, such as nu-bˆnda gu4s (Eames Coll. UU 16) and áabras (MVN 13 908 + 909). SI.A-aÕs archive, which consists of seventy-six texts, is the largest extant non-institutional archive available for study from the Ur III period. For an overview of the archive, see GarŽ nkle 2003. 18 For example, SI.A-a issued at least two interest bearing loans to his brother NaÀarum (ZA 93/2 4 and MVN 13 902 + 903). ZA 93/2 4 was a massive advance of 23 gur (6900 liters) of barley. I believe that this loan, which was to be repaid after the harvest, was obtained by NaÀarum in order to Ž nance his acquisition of labor for the harvest season, but that does not change the nature of his obligation to SI.A-a. 19 We must also be wary about any quantitative conclusions we attempt to draw from the survival of documents in the cuneiform record (see Civil 1980). Even our analysis of the

8

STEVEN J. GARFINKLE

longer duration than we ordinarily assume; and second, in many instances there was no legal requirement that the loan document be destroyed upon repayment (these points are addressed at length below in the section on lending practices). The suggestion that the creditors of the Ur III period were not primarily motivated by the desire for proŽ t through the payment of interest involves us in the difŽ cult question of evaluating economic motives in antiquity. The familiarity of the modern audience with the idea of money-lending at interest leads to the presumption of the proŽ t motive whenever we encounter this institution in the ancient world. But can we prove that the Ur III creditor was motivated by a desire to recover interest rather than labor (or land) and that this constituted a pursuit of proŽ t? Yes, I am convinced that we can document the concern of the lender for his interest. The promissory notes and court documents that are preserved, for example, in the Ur-Nusku archive from Nippur illustrate the lengths to which creditors went to recover both their principal and the interest that they were owed.20 In these texts, the promised payments were often substantial, as were the penalties for failing to satisfy the creditor.21 Yet in many cases, despite the scarcity of labor, the value of the loans greatly exceeded the potential value of the debtorÕs labor. Finally, the very existence of antichretic loans in the non-institutional archives should cause us to look for a motive other than the acquisition of labor behind the non-antichretic loans. The antichretic loans form a small percentage of the total corpus of loans for the entire period under discussion. It is not warranted to assume that the majority of the customary loans were advanced for the same purpose as the antichretic loans. The motivation behind the customary

overall activity of creditors must be advanced with care. ÒThus some individual who is known to have issued a string of loans, may have been a professional money-lender living off the interest he acquired, a conclusion that is most often drawn by modern scholars. Yet, it is also possible that the loans preserved form only a tiny fraction of his economic activities. He may have been a large land owner enjoying the income of his estate, who on special occasions issued credit.Ó (Van De Mieroop 1997: 16) 20 These categories of texts involve either the promise or the demand of repayment of a debt. For an example of a court document, see NRVN 1 49. The distinguishing characteristic of most of the court documents was the presence of the verb du11, usually in the form “-naan-du11, Òhe said to him.Ó The use of this construction implies that the speaker was making a verbal allegation before witnesses against another individual or group of individuals. For an example of a promissory note, see TMHC NF 1/2 40. 21 In NRVN 1 49, Ur-Nusku demanded repayment of his silver from °a-kuge. °a-kuge promised to return the silver, and its interest. If he failed to do so by the speciŽ ed date, he would pay double the amount owed. The amount involved, 2/3 mana of silver, would have been sufŽ cient to purchase several male slaves. In TMHC NF 1/2 40, the silver that was promised was speciŽ cally identiŽ ed as a payment of interest on silver.

SHEPHERDS, MERCHANTS, AND CREDIT

9

loans, especially on the part of the lender, was not indicated by the texts themselves. In the absence of this crucial information, I see no reason to presume that ordinary loans were anything other than their content indicates: interestbearing obligations, the issuing of which was rewarding to the creditor, and the receipt of which may have been an economic necessity for the debtor. This does not mean that customary loans could not result in debt slavery, but only that this was not the primary intent of the creditor in all such arrangements. TERMINOLOGY In this section, the terminology for customary, interest-free, and antichretic loans is examined in detail. The terms employed in the loan documents provide the clearest means of differentiating between the various forms of loans. Terminology: Customary Loans The customary loans in the Ur III period followed the same basic layout and employed the same terminology: x gur áe or x g’n k -babbar ur5-ra / ki PN1-ta / PN2 áu ba-ti / Witnesses / Date / Seal of PN 2 x amount of barley or silver, as a loan with interest, from PN1, PN2 received. Witnesses. Date. Seal of the recipient.

Further clauses, indicating the precise interest rate or the exact terms of repayment might have been included in the loan documents, but such additional clauses were not necessary. The presence of witnesses or the seal of the debtor, however, was always required. Any information added to the basic structure indicated that the loan was subject to some further conditions that were out of the ordinary. Such conditions might have been an unusual interest rate22 or details concerning the repayment of the loan. 23 Under most circumstances, the only information contained in the loans of this period was the amount and commodity of the principal of the loan, the identities of the borrower and the lender, the interest rate to be charged, the names of the witnesses, and the date, usually only the year, on which the loan was advanced. In the Ur III period, the terms of customary loans were indicated in one of two ways. Either the interest rate was expressed numerically, or it was speciŽ ed with the use of certain stock phrases. In the majority of the loans from

22 23

For example, MVN 8 153 speciŽ es an interest rate of 25%. For example, TIM 3 150 speciŽ es the precise month of repayment.

10

STEVEN J. GARFINKLE

this period, the usual rates of interest were 20% for silver loans (indicated numerically with the phrase: m‡á 5 g’n 1 g’n-ta, Òthe interest is 1 g’n per 5 g’nÓ) and 33% for barley loans (indicated numerically with the phrase: m‡á 1 gur 0.1.4-ta, Òthe interest is 100 s“la per 300 s“laÓ). Therefore, in the customary loans, the payment of interest was always stipulated; however, the exact rate was often determined by custom and was not explicitly stated. For example, in the SI.A-a archive, the rate of interest was expressed numerically in only Ž ve of the thirty-nine surviving customary loans. 24 The reason for this seeming omission was that the interest rates in this period tended to be Ž xed, and they were commonly known. This situation was already addressed by W. F. Leemans (1950) for the Old Babylonian period. Leemans (1950: 14) argued that an explicit numerical indication of the interest rate in Old Babylonian loan documents was not required and could be replaced by certain phrases, such as m‡á gi-na; and he suggested that these phrases were understood to indicate not just the presence of interest but also the actual rate of interest, which was commonly known. The customary interest rates in the Old Babylonian period were the same as those attested for the Ur III period. As Leemans pointed out, his suggestion Ž nds support in the extant Old Babylonian law collections. The Laws of Eánunna and the Laws of Hammurabi both establish that the interest rate for silver was 20%, and the interest rate for barley was 33% (Laws of Eánunna ¶ 18A; Laws of Hammurabi gap ¶ t; see Roth 1997; Van De Mieroop 1995: 357-8). The Laws of Hammurabi underscore the perceived difference between barley and silver loans. Gap ¶ u speciŽ ed that the recipient of a silver loan could repay the loan in barley, if he did not have the necessary silver, but the debtor could not be charged the higher rate of interest, i.e. 33%, on what had originally been a silver loan (Roth 1997: 97-8). In the Ur III period, the same conditions applied to the drawing up of loan documents. In all customary loans, the tablets indicated that interest was being charged but they did not always mention Ž gures, instead relying upon the knowledge of the persons involved, and on the enforcement of customary interest rates. In Ur III loan documents, the phrase ur5-ra, along with its variant ur5Ꮰ, carried the meaning of a debt or obligation with interest.25 The use of the

MVN 8 153, and 158; ZA 93/2 7, and 10; and TIM 3 150. The textual evidence shows that ur5-ra and ur5-Ꮰmust be considered variants of the same phrase. Lutzmann (1976: 40-1) suggested that there were local customs regarding the usage of one of the variants. A survey of the loans from Nippur bears out his conclusion that the phrase ur5-Ꮰwas more common in loan documents at Nippur. Among the dozens of loans from Nippur published in NRVN 1, only three employ the phrase ur5-ra, while forty use the phrase ur5-Ꮰ. 24 25

SHEPHERDS, MERCHANTS, AND CREDIT

11

phrase ur5-ra/Ꮰon its own, or preceded by k , Òsilver,Ó indicated a silver loan carrying an interest rate of 20%, unless a speciŽ c rate was otherwise indicated. 26 The phrase m‡á-g‡-g‡, meaning Òthe interest is established,Ó sometimes accompanied the phrase ur5-ra/Ꮰ. Barley loans were differentiated from silver loans by the addition of áe, Òbarley,Ó to the ordinary phrase, thus áe ur5-ra/Ꮰ. In his exhaustive study of the Ur III loan documents that were available to him, Heiner Lutzmann (1976: 41) arrived at a different conclusion regarding the ur5-ra/Ꮰloans, and determined that these loans did not carry with them an assumption of interest. The original equation of ur5-ra with an interest-bearing obligation came from the lexical lists in which this Sumerian term was paralleled by the Akkadian term for an interest-bearing obligation. Lutzmann, however, noticed that certain texts appeared to provide a notice of the interest-bearing nature of debt twice through the use of the phrase ur5-ra/Ꮰalong with the phrase m‡á-g‡-g‡.27 He felt that such repetition indicated that one of the two phrases must imply something other than interest. What is primarily required in order to accept this suggestion is some evidence that interest was not expected on the loans that simply employed the phrase ur5ra/Ꮰ. (A further requirement would be demonstrating that the nature of such loans had changed over time in order to account for the positive evidence from the later lexical lists and the law codes.) One could just as easily point to the loans that incorporated the customary formula, and at the same time included a phrase indicating that no interest was being charged, m‡á nu-tuku, Òthere is no interest.Ó28 The same principle of unnecessary information would apply in these cases if we follow Lutzmann in assuming that ur5-ra/Ꮰdid not indicate that interest was to be charged. 29 There are a number of possible explanations for this phenomenon. The two clauses may in fact have had slightly different meanings, and, therefore, would imply no repetition. The customary phrase probably signiŽ ed more than simply the interest rate but also the additional terms of the loan agreement, such as the duration of the loan and the form of repayment. The m‡á-g‡-g‡ clause may have been a written indication that the interest rate had been established

26 Wool loans usually employed the same terminology as silver loans. See MVN 13 741, and ZA 93/2 9. 27 See, for example, MVN 8 156. 28 See, for example, MVN 13 746 + 747, and TMHC NF 1/2 91. 29 The same repetition occurs in any loan that bears both the clause ur5-ra/Ꮰand a numerical indication of the interest, as is the case in numerous loans from Nippur, and several in the SI.A-a archive. See, for example, NRVN 1 127, NATN 247, and 316.

12

STEVEN J. GARFINKLE

between the parties, in the presence of the witnesses, at the time of the transaction. 30 In a society that was largely illiterate, the addition of the phrase m‡ág‡-g‡ may have served as a reminder to a scribe reading the loan document that the terms had been agreed upon orally by the participants. Finally, this clause may even have modiŽ ed the customary agreement in a manner that we cannot fully appreciate. For example, the phrase m‡á-g‡-g‡ could have implied some durative force and indicated the operation of compound interest. The problem of repetition is not easy to solve. Aaron Skaist (1994: 126) has suggested, against Leemans, that we should not understand phrases such as m‡á gi-na in the Old Babylonian period as indications of a Ž xed interest rate, but rather as signs of a variable interest rate. This observation could be brought to bear on the Ur III period as well. If we have some ur5-ra/Ꮰloans that provide no additional information, while other ur5-ra/Ꮰloans include a numerical provision for the interest rate, then why should we presume that ur5-ra/Ꮰindicates a Ž xed rate of interest? There is no decisive answer to this question; however, the texts themselves are very suggestive. I have already stated my belief that the use of customary phrases refers to more than simply the rate of interest, and also implies information as to the more general terms of the loan. It is possible that some of these terms were established by oral agreement and are therefore beyond recovery. What we must also keep in mind is the possibility of regional variation in the customs for describing loans. We can see such variation in operation even if we cannot always explain it, as, for example, in a preference for ur5-Ꮰover ur5-ra at Nippur. The centralization of the Ur III state was an extremely rapid and short-lived phenomenon. The law codes of the Ur III and succeeding periods indicate a desire on the part of the central authority to standardize some aspects of lending activity. However, the individual inhabitants of the various districts would have remained familiar with their local customs. 31 Some of the repetition that we cannot readily explain, may be accounted for by the different geographic or social backgrounds of the parties to the loan. Throughout this discussion we must remember that the surviving loan documents did not create legal obligations. The actions taken in the presence of the witnesses when the loan was issued by the creditor established the legal conditions pertaining to the transac30 MVN 8 158 provides support for this understanding of the phrase m‡á-g‡-g‡. This was a barley loan advanced by SI.A-a, and line 2 reads: 1 gur m‡á-g‡-g‡, Òthe interest is established as 1 gur.Ó The clause ur5-ra/Ꮰm‡á-g‡-g‡ should be translated as: Òcustomary interest is established.Ó One of the reviewers of this article suggested that the reduplication of the verb could indicate a plural, thus Òthey agreed on the interest.Ó 31 For example, we know that many local calendars remained in use during the Ur III period, despite the imposition of the so-called Reichskalender.

SHEPHERDS, MERCHANTS, AND CREDIT

13

tion. 32 Under these circumstances, it is hardly surprising that we do not Ž nd the degree of clarity and standardization in the loan documents that we might expect in our own time. In his assessment of loan contracts from the Old Babylonian period, Skaist has reconsidered many of the aspects of the loan documents that had previously been discussed by Leemans, and his conclusions have great signiŽ cance for the discussion of the Ur III loans as well. Most importantly, Skaist (1994: 40) has followed Lutzmann and questioned the assumption that the formula ur5-ra/Ꮰwas a primary indicator that interest was being charged: ÒThe term ur5-ra in the loan contracts of the Ur III period is not to be deŽ ned as Ôinterest bearing loan,Õ nor can interest be said to be a characteristic of the ur5-ra loan.Ó In the Ur III loans from Nippur, which constitute the largest corpus of noninstitutional loans from the period, we are more likely to encounter the numerical indication of the interest rate, rather than the use of the customary phrases. Within the Nippur corpus, there are numerous texts in which both the numerical indication of the interest and the customary phrase appear,33 and there are examples of Nippur loans which use only the customary phrase to indicate the rate of interest.34 Therefore, we encounter examples of what appears to be repetitious information in many of the surviving loan documents; however, we cannot assume that the problem this causes modern scholars in our efforts at arriving at a translation is indicative of any difŽ culty on the part of the ancient Mesopotamian. The parties to the loan agreements understood the customs of their society, and the side by side use of the customary phrase and the numerical indication could just as easily have complemented each other as an indication of different information. There is support for this assertion from the extant law collections of the Ur III period. Martha RothÕs reconstruction of the Laws of X, which may have formed part of the Laws of Ur-Namma, provides a parallel for the repetition of information that characterizes many of the loan documents. According to Roth (1997: 38), the phrases áe ur5-ra-Ꮰand k -babbar ur5-ra-Ꮰappear alongside the numerical indications of the interest rates.35

32 The tablets themselves were always secondary to the public performance of the actions that they record. ÒThere is ample evidence throughout Mesopotamian legal history that legal transactions were concluded in this way, that is, without leaving a trace in the form of a legal document. Symbolic acts performed in the presence of witnesses gave the act its binding force. A written document was not necessary. All this comes as no surprise in a society in which literacy was restricted to a small segment of the population.Ó (Renger 1995: 293) 33 For example: NRVN 1 78, 79, 81, 82, and 84. 34 For example: NRVN 1 105 and 106. 35 For the Laws of Ur-Namma, see also Wilcke 2002: 325. WilckeÕs reconstruction and

14

STEVEN J. GARFINKLE

The phrasing of the laws was precisely the same as that which appeared in the loans themselves. Therefore, the primary indication of a customary loan was the presence of the phrase ur5-ra/Ꮰ. The Sumerian ur5-ra corresponds exactly to the Akkadian Àubullum, meaning interest-bearing debt. The correspondence is made clear in the lexical list ¿AR-ra = Àubullu. The lexical series ana ittiáu provides corroborating evidence for this equation, again giving Àubullum as the counterpart of ur5-ra.36 The manner in which the various provisions in the law collections were written is an indication that the practice of customary lending was identical in the Ur III and Old Babylonian periods. The relevant provision of the Laws of Hammurabi reads: gap ¶ t áumma tamk¨rum áeÕam u kaspam ana Àubullim iddin ana 1 kurum 1 p¨n 4 áâtáeÕam ×ibtam ileqqe áumma kaspam ana Àubullim iddin ana 1 áiqil kaspim IGI.6.GçL u 6 ußßet ×ibtam ileqqe If a merchant gives grain or silver as an interest-bearing loan, he shall take 100 silas of grain per kur as interest (=33%); if he gives silver as an interest-bearing loan, he shall take 36 barleycorns per shekel of silver as interest (=20%). (Roth 1997: 97)

The equivalence of the Akkadian áeÕam/kaspam ana Àubullim with the Sumerian áe/k ur5-ra/Ꮰis therefore obvious. The development of these customary phrases is less obvious. The traditional bias in favor of the great institutions that has pervaded the study of the economy of the Ur III state has led to the assumption that the origin of the áe ur5-ra loans had an institutional background. This assumption has been based upon the idea that the temple estates bore a responsibility to their dependent laborers to provide access to institutional credit in times of need. This was the conclusion drawn by Tom Jones and John Snyder (1961: 249-79) in their assessment of áe ur5-ra loans. Their study also determined that these advances were discounted (which may have been a way of formulating the interest within an institutional setting), and that they may have been a form of Òseparate or private enterpriseÓ within the institutional setting (Jones and Snyder 1961: 271). Steinkeller (2002: 116) has recently reconsidered the institutional áe ur5-ra advances, and he concludes that they constituted a system of advances made by the central administration to its dependents, ordinarily without interest, in order to protect them from economic

translation also support the interpretation of ur5-ra as an interest-bearing loan, carrying the customary interest rates. 36 ana ittiáu II i 57 (MSL 1: 19). Following Skaist (1994: 34-5), we should understand the phrase ur5 -ra to be Òa combination of the noun ur5 plus the locative sufŽ x /a/ . . .Ó Therefore, we should translate the phrase áe/k ur 5-ra/Ꮰas Òbarley/silver as an interest bearing obligation.Ó

SHEPHERDS, MERCHANTS, AND CREDIT

15

hardship. 37 However, the use of this term in relation to lending practices was widespread in the Ur III period and was not conŽ ned to institutional endeavors. Further, the parallel between the use of the terms áe ur5-ra and k ur5-ra, suggests that the key element in these terms is precisely the noun ur5, which can be so closely tied to the Akkadian Àubullum, and which indicated an interestbearing debt. The exclusive use of áe ur5-ra, Òbarley as an interest bearing obligation,Ó in institutional contexts clearly re ects both the availability of vast quantities of barley in the temple estates, and the fact that the needs of the temple dependents were most readily satisŽ ed through advances in barley.38 Based on their terminology, the ur5-ra transactions were interest bearing. Therefore, if Steinkeller is correct in suggesting that the áe ur5-ra advances made by the large institutions, mainly in the far south of Babylonia, were interest-free, then we must assume that the origin of the term ur5-ra lies outside of an institutional context. Indeed, this explanation is made more convincing in light of the probable northern location for the geographic origin of the ur5-ra/Ꮰloans. This is made clear by the prominence of these loans in the north in the Ur III period, 39 and the fact that ur5-ra loans were more commonly found in the north during the Old Babylonian period and, in particular, in the Diyala region (Skaist 1994: 37). The evidence from the SI.A-a archive, which contains numerous ur5-ra loans, supports this conclusion. The origin of this archive is to be found in northern Babylonia, northeast of Babylon and just south of the Diyala. The use of the customary phrase ur5-ra/Ꮰachieved a wide geographic usage during the Ur III period precisely because it expressed the terms that were becoming common practice throughout southern Mesopotamia. In the succeeding Old Babylonian period, a new phrase was introduced from Babylon, m‡á gina, which meant Òof regular interestÓ and obviated the need for the previous formula. Skaist (1994: 126) has concluded that, Ò(t)he ÔsuddenÕ introduction of m‡á gi-na in some of the loan contracts from southern Old Babylonia tends to indicate that m‡á gi-na re ects a concept not in use there.Ó Instead, I believe

37

Temple loans for similar purposes may also have existed in the Old Babylonian period. See Harris 1960, where the °amaá temple at Sippar is described as a beneŽ cent creditor. Recently, Richardson (2002: 171-72) has pointed out that some temple loans in the Old Babylonian period required the payment of interest, and that the temples also engaged in lending for explicitly commercial purposes. 38 As discussed above, in non-institutional settings, we often encounter k ur5-ra, Òsilver as an interest bearing obligation.Ó 39 In his study of the ur5-ra loans, Skaist (1994: 38) provides a map of their distribution that appears to show these loans concentrated in the south, but this map is based on a limited sample of texts.

16

STEVEN J. GARFINKLE

that the concept of a customary interest rate was well-known in all of Babylonia at least since the Ur III period, and the introduction of m‡á gi-na was not the arrival of a new concept but the adoption of a new terminology for an old convention. The value to the ancient scribe of developing a brief terminology for expressing the customary terms of loan agreements should be obvious. Anyone who has worked with Ur III loan documents knows just how small the tablets are. The majority of the loan documents amount to little more than four or Ž ve lines of text along with the witnesses and the date. The most essential characteristic of the customary loans was that the formula allowed for a great deal to be expressed with an economy of writing. Terminology: Interest-free Loans Interest-free loans were usually indicated in one of two ways. Either the phrase m‡á nu-tuku, Òthere is no interest,Ó appeared in the text, or the text included a repayment clause but no indication that interest was due.40 Many of the loan documents that included promissory oaths regarding the repayment of the loan were interest-free loans, and most contained oaths pledging repayment at a speciŽ c time in a given month. 41 The majority of the interest-free loans indicated a date of repayment. Often, the duration of the loan did not exceed one month (some loans were as short as ten days), 42 and none of the loans were likely to exceed one year. We can assume that these loans were due on the month indicated when it next arrived. Interest-free loans may have been common among professional groups and extended families. Steinkeller (1987) has shown that the organization of the foresters at Umma ran along familial lines and that the supervisors were laden with increased responsibility along with increased rewards. Similar arrangements

MVN 8 165: 1 g’n lal igi-6-g‡l/ k -babbar/ ki SI.A-a-ta/ Dingir-ba-ni áu ba-ti/ sag iti a-bu!-/ um/ tu-ru-um/ mu lugal in-pˆd/ Witnesses/ Date 40

Ilum-bani received 1 g’n, less 1/6 of a g’n (7.83 g.) of silver from SI.A-a. He swore in the name of the king to return it at the beginning of the month abum. Witnesses. Date. Of course, a loan document such as this one may have superceded an earlier interest-bearing loan agreement. In essence, this could be a repayment agreement, and as such, the interest on the original loan may be hidden within the total given in line 1. 41 See, for example, from Nippur: NRVN 1 103, 124, 174, 185, and 189, NATN 65, and 442, TMHC NF 1/2 31; from Umma: YOS 4 5, 51, and 52; from the SI.A-a archive: MVN 8 166 and 171. 42 MVN 8 163.

SHEPHERDS, MERCHANTS, AND CREDIT

17

must have been common among other professional organizations, like that of the shepherds, and individuals may have been customarily obligated to advance loans to their subordinates on favorable terms. For example, SI.A-a loaned sheep to two men who may have been shepherds under his supervision. 43 Another possibility is that individuals like SI.A-a may have been temporarily covering a loss among their charges. Robert Englund (1991: 264) has argued that supervisors in the Ur III period were able to derive personal beneŽ t from surpluses created under their care, and he has further suggested that these supervisors were also responsible for deŽ cits that occurred under their watch. There is some evidence as well for the practice of using interest-free loans within extended family groups.44 Terminology: Antichretic Loans Antichretic loans differed from customary loans in several signiŽ cant respects. Similar to customary loans, antichretic loans were ordinarily subject to interest; however, the interest on antichretic loans was not paid in a traditional commodity but was most often rendered in the service, usually labor, of the borrower. Therefore, it is often difŽ cult to calculate the interest rate for a particular loan, and when such calculations are possible, the rates vary widely. Many of the conventions in use for documenting customary loans were also applied to antichretic loans. These loans were often both sealed and witnessed. The duration of antichretic loans is very difŽ cult to determine with assurance. Above, I divided antichretic loans into two categories based on the commodity of the principal advanced. 45 The loans in the Ž rst category were probably similar to customary silver loans in that the terms of the loan could extend beyond one year, and depended upon the ability of the borrower to repay the principal. Those of the second category were probably harvest loans, with the interest payable at the time of the harvest, and the principal returned after the conclusion of the harvest. We do not know when these agreements were contracted, but in all likelihood they were short-term loans. The shortage of labor in the Ur III state that has been identiŽ ed by Steinkeller and others can be readily appreciated in the antichretic loans, which

MVN 8 171. For example, in MVN 8 164 Abaya received an interest-free loan from SI.A-a, his brother. Both brothers were also chief shepherds, and so their familial relationship was paralleled by their professional relationship. 45 See the section above on the typology of loans, and the texts referred to in notes 10 and 11. On the use of antichretic loans in craft production, see Neumann 1993: 154. 43

44

18

STEVEN J. GARFINKLE

place a high value on labor. This pressure affected all households , which explains in part the willingness of the central administration to provide institutional credit on favorable terms for the borrower. The non-institutional household was at a particular disadvantage as it rarely had access to signiŽ cant numbers of dependent laborers. Therefore, access to labor was a more critical concern for the entrepreneur in the Ur III period than access to land. Under these circumstances, the existence of antichretic loans, the purpose of which was the procurement of agricultural labor, is hardly surprising. Steinkeller (2002) has identiŽ ed numerous examples of these loans in both the institutional and non-institutional economies, and he has used these loans as the basis for some of his arguments about the economic motivation of lenders.46 The terminology of the various types of antichretic loans frequently paralleled that of the customary loans, but with the signiŽ cant addition of providing evidence for the purpose of the loan agreement. In the antichretic loans that seek harvest labor, we commonly encounter direct references to the type and duration of the labor.47 In these loans we can often determine that the interest rates were discounted, because the labor being provided had a value that was equal to less than the 33% return that a creditor would ordinarily have received for a harvest loan advanced in barley.48 We also Ž nd antichretic loans that specify the area of land to be harvested as interest on a loan. 49 Therefore, these texts support the theory that labor was in high demand in the Ur III period; however, they cannot be used to support the notion that the majority of lending activity in the Ur III period was geared solely toward the acquisition of land or labor.

46 For non-institutional antichrectic loans of this type, see note 11. For a loan of this type in an institutional setting, see AUCT 3 492. For a more extensive discussion of these loans and their operation, see GarŽ nkle 2000: Chapter II.c.iii. 47 For example, JCS 23/4 8: 2.0.0 gur/ sag-bi gi4-gi4-dam/ 10 guruá u4 1-Ꮰ/ áe-kin-kin-d/ ki SI.A-a-ta/ G“r-ni/ áu ba-ti/ Witnesses/ Date.

Girini received 2 gur (600 liters) (of barley) from SI.A-a, to be returned in its entirety, for 10 man days of labor for harvesting barley. Witnesses. Date. In this transaction, SI.A-a was to recover the principal of the loan (600 liters of barley) and be paid interest in harvest labor. 48 In the text cited above in note 47, the interest rate works out to 10%. The value of the 10 days of labor was equivalent to roughly 60 liters of barley, as the average daily wage for harvesting labor was 6 liters of barley. This issue is discussed at length in GarŽ nkle 2000: Chapter II.c.iii. 49 See, for example, MVN 9 201. On the related topics of the pledge of Ž elds to secure loans, and the rent of Ž elds, see Neumann 1999b: 137-48.

SHEPHERDS, MERCHANTS, AND CREDIT

19

LENDING PRACTICES IN THE UR III PERIOD As noted previously, interest rates in the Ur III period, and in the succeeding Old Babylonian period, were determined by custom. Many of the loan documents did not provide an explicit statement of the interest rate. Instead, certain phrases were employed to indicate the terms of the loan. We should not be surprised that much of the substance of the loan agreement was dependent upon the oral transmission of knowledge and custom in the Ur III period. The loan documents themselves only served as reminders to the participants of what had been transacted. The signiŽ cance of the loan documents was often that they had been executed in the presence of others. The witness lists and the seal impressions were a crucial part of the recorded loan document because they were the locus for the most important legal aspects of the texts.50 The actions taken by the participants before the witnesses at the time that the loan was issued governed the actual obligations of the parties to the loan. 51 It was always necessary for the loan to be either witnessed or sealed. Many of the tablets recording loans also have preserved envelopes. Most loan documents in this period had cases, but it is impossible to prove that this was necessary for the enforcement of the terms of the loan. Indeed, the presence of some sealed loan documents that do not have preserved cases may argue for the possibility that envelopes were not always required when loans were recorded. The fact that the loan documents on their own could not create legal obligations helps to explain their survival in such large numbers. It is common practice in the study of cuneiform texts to assume that the survival of loan documents was the result either of default on the part of the debtor, or the fact that the loan agreement was still in force and survived among the later texts of the archive. The Ž rst assumption rests on the understanding that the loan documents were destroyed at the time of repayment. The information provided, for example, by the SI.A-a archive, in which numerous loans survive from throughout

50 As Greengus (1969: 513) has noted, Òin ancient Mesopotamia written records were basically evidentiary in character rather than dispositive i.e. the documents (as do witnesses) serve only as an additional means of proving that a transaction had taken place but are not, in themselves, instruments of legal change.Ó The loan documents themselves existed to protect the creditor by providing a record of the events and the witnesses to those events. The presence of a loan document was not enough to establish a claim; such a claim could only be demonstrated in conjunction with the witnesses, or, in the case of loans, perhaps with the additional proof of the seal impression of the debtor. 51 It is clear from studying the prosopography of the archives of creditors that both of the active participants in the loan documents furnished witnesses to the transaction. This practice provided obvious protection to both sides in the event that a court case became necessary and the testimony of the witnesses was required.

20

STEVEN J. GARFINKLE

his long career, calls into question these assumptions. Either we must imagine that SI.A-a was plagued by bad debtors or we must understand the survival of these loans in a different way. Steinkeller (2002: 113) has already recognized the problem posed by the presence of so many of these texts: Let me suggest, Ž nally, that the very survival of so many loan documents from ancient Babylonia is yet another potent argument for assuming that most of such transactions were motivated by something other than the gain from interest. As is well known, the principle governing the preparation and handling of loan documents—and here I am talking speciŽ cally of the third and second millennia—had it that only one copy of the documents was prepared, which, after being sealed by the borrower, was handed over to the creditor, in whose possession it would remain until the loan was repaid. At that moment, the loan document would be destroyed, of which procedure numerous testimonies survive.

SteinkellerÕs argument that the impetus for advancing loans was other than the expectation of the receipt of interest has already been addressed, but he has highlighted an intriguing problem. The most immediate answer to the problem of the survival of the loans is that we are making too many assumptions that ancient Babylonian loans were given for a Ž xed duration. Interest rates in the Ur III and Old Babylonian periods were not annual, but were rather calculated on a per loan basis. What this means is that the loans in these eras had no Ž xed term (for more on the duration of loans, see below). I believe that the custom, unless otherwise indicated in the loan document itself, was not to assign a limit to the duration of the loan. The absence of explicit repayment clauses on most of the loans supports the presumption that the loans were repaid, according to custom, based upon the ability of the debtor. Default occurred only when the borrower failed to pay the interest due on the loan, or when the borrower failed to honor conditions that may have been explicitly applied to the individual loan. My second objection to the assumption that the survival of a loan document is an indication of default rests on the legal nature of the documents. We know that the issuing of a loan was attended by witnesses, the testimony of whom constituted legal protection for the creditor. Certainly, the repayment of the loan would have been carried out in the same manner for the protection of the debtor. If the receipt of a loan was a public act, then so was the satisfaction of the creditor. If the repayment was made before the same set of witnesses, then the creditor could hardly use the original loan agreement to establish the basis for a legal claim against a debtor who had already satisŽ ed him. There are obvious reasons why the debtor would have sought the destruction of the loan document at the time that the loan was repaid, but it is not clear that this was a legal requirement. The destruction of the loan agreement, alluded to above by Steinkeller, may in fact have been related directly to the legal obligation cre-

SHEPHERDS, MERCHANTS, AND CREDIT

21

ated by the presence of the borrowerÕs seal on the tablet or envelope. 52 This might explain the greater survival of loan tablets without their cases. Perhaps all that was necessary was the destruction of the sealed envelope and not of the tablet itself. A creditorÕs possession of a defunct loan tablet, without its case, which had been repaid in the presence of witnesses, did not grant him the ability to establish a legal claim against an innocent former debtor. Moreover, the practice of occasionally having copies of the loans drawn up for the archive of the debtor is relevant to the discussion of the survival of the loan documents. 53 These copies, which lacked the seal impression of the debtor and the list of witnesses, could not be used to create a legal obligation and therefore did not have to be destroyed. 54 The absence of an indication of the precise duration of the loan in most of the loan documents can only mean one thing: similar to the customary understanding of the usual rate of interest, the debtor and the creditor must have understood by custom what terms applied to the duration of the loan. 55 The absence of explicit repayment clauses on most of the loans supports the presumption that the loans were repaid based upon the ability of the debtor. Since we know that the scribes of the Ur III state were quite capable of drafting clauses specifying the conditions of repayments, and including speciŽ c dates, they are telling us something signiŽ cant when they omit such clauses. Under these circumstances we can only put forward two suggestions regarding the interest rates. Either the interest was understood to be annual, or the interest was understood to apply to the exact duration of the loan, regardless of its length. It is usually presumed that the latter suggestion would disadvantage the debtor and allow the creditor to collect interest at usurious rates. While I admit that this was often the case, this presumption requires that most loans were of extremely short duration. I think we can suggest precisely the opposite.

52 The fact that the use of a seal could create a legal obligation on the part of its owner is made clear by texts referring to lost seals and their recovery. See Steinkeller 1977: 48-9 and Hallo 1977. 53 This practice is attested in the T¨ram-ili archive, see JCS 38 37. 54 The preparation of copies of legal documents is attested in the cuneiform record, as is a differentiation between copies and sealed texts. See BIN 7 31, an Old Babylonian letter in which the author of the letter threatens to return the copy of a partnership agreement and destroy the sealed documents. 55 ÒThe period of the loan was thus often clear both to the borrower and the lender at the time the contract was drawn up. The exact amount to be repaid was known to both parties, or could have been easily calculated.Ó (Van De Mieroop 1995: 359)

22

STEVEN J. GARFINKLE

According to my understanding of the customary loans, those texts that indicate a precise date for repayment were deviations from the normal practice. We can assume then that short-term loans at high interest were not the normal practice. The custom of applying interest rates to the actual duration of the loan, from its issuance to its repayment, was favorable to the creditor without being onerous to the debtor. The point is that interest was payable for each year of the loan, but was not calculated annually. Any repayment on a loan of less than one year in duration was repaid with the full amount of interest.56 There is evidence that the payment of all of the outstanding interest might accompany the repayment of the principal. 57 Thus, a creditor might not have seen any return on a loan for several years. The obvious implication is that individuals, unless otherwise directed by the terms of the loan, repaid their debts when they were capable, plus the customary interest owed. At the same time, the creditors proŽ ted both from early repayment, and from the opportunities, which arose prior to the harvest, to insist on short-term loans. 58 The evidence from the Ur III period supports the conclusions reached by Marc Van De Mieroop (1995) and Skaist (1994) that Old Babylonian interest rates were not computed annually. Skaist has pointed out that a complete date (day, month, and year) is necessary in order for the amount due to be calculated on a loan with an annual interest rate. The overwhelming majority of the loans from the Ur III period neglect to record the day of the transaction, and many record only the year. This would make perfect sense if the interest rate was to be computed on a per loan basis. 59

56 Therefore, if PN1 borrowed 10 g’n of silver from PN 2 at 20% interest, the following terms would have applied. If he repaid after 9 months, he owed 12 g’n; if he repaid after 12 months, he owed 12 g’n; if he repaid after 18 months, he owed 14 g’n; if he repaid after 24 months, he owed 14 g’n; and so on. 57 There are texts (for example, NRVN 1 61 and NATN 635) which indicate that the interest on a silver loan over a period exceeding one year did not exceed the total of the number of outstanding years of interest on the principal at 20%. Therefore, the payment of interest may have occurred only after several years, or at the time of the Ž nal repayment of the loan, and it did not include any compounded interest. 58 And this was clearly the case also in the Old Babylonian period: ÒMany of the loans we Ž nd attested in the Old Babylonian sources were obtained in times of dire need. In the last month before the harvest a farmer who could not feed his family any longer needed to borrow grain.Ó (Van De Mieroop 1995: 362) 59 The issue of compound interest also arises in any discussion of the duration of loans. It is clear that the inhabitants of the ancient Near East were familiar with the concept of compound interest. Leemans (1950: 22-5; but see also Van De Mieroop 1995: 361) highlighted several mathematical texts that provided for the calculation of compound interest. There is, however, no evidence that compound interest was the norm for loans issued in the Ur III period, and the texts cited in note 57 suggest that it was not.

SHEPHERDS, MERCHANTS, AND CREDIT

23

The following is an example of a loan of this type. Eames Coll. P 1: 1 ma-na 12 g’n k -babbar si-“-tum n“-ka5-ak/ m‡á 5 g’n 1 g’n-ta/ ki Ur-áa5-ga-ta/ ò-tul-Ma-ma/ áu ba-ti/ iti áu-numun-a sag-Ꮰ/ l‡-e-d / tukum-bi nu-l‡ 2-ˆm tab-bŽ -d / mu lugal “-pˆ/ Witnesses/ Date (IS 1, I) Utul-Mama received 1 mana and 12 g’n (576 g.) of silver, the balance carried forward of a balanced account, from Uráaga, with interest of 1 g’n for each 5 g’n (20%). He will weigh it out (repay it) at the beginning of month 4. If he does not weigh it out, then he will pay double the amount. He swore in the name of the king before the witnesses. Date (IS 1, I).

This text, a sizeable loan between two merchants,60 provided a numerical indication of the customary interest rate, and it demonstrated the concern of the creditor for the recovery of his interest. Here we have a short-term loan, from the Ž rst to the fourth month, which included both an indication of the interest rate and the penalty for default. Unfortunately, the laconic nature of most of the texts prevents modern scholars from being wholly conversant with the customs governing lending practices in the Ur III period. However, in the absence of such information, we cannot presume that because the customary rates of interest appear high to a modern audience, and because there is evidence of debt slavery in the ancient Near East, that the practice of money-lending invariably favored the creditor. If the loans of the Ur III period were of a longer duration than we are accustomed to assuming, then we would expect a greater number of loan agreements to survive from throughout the career of a lender. This is exactly the situation that prevails in the SI.A-a archive. Certainly, the survival of some of the loans in this archive may have been the result of default, but there is no evidence to prove that this was the case in the majority of the extant texts. With a larger sampling of loans we might be able to distinguish between the consumptive loans, which were usually issued in barley for a Ž xed duration, and the longerterm loans, which were more frequently issued in silver. Under such circumstances, we might be able to distinguish in an archive such as SI.A-aÕs between the ÒgoodÓ debts and the ÒbadÓ debts. In summary, the loans of the Ur III period were usually subject to a customary interest rate. Compound interest did not generally apply to loans in the

60 It is likely that the participants in this text are two well-known merchants. Utul-Mama was a merchant active in SI.A-aÕs community, and he frequently appeared in the latterÕs texts. Uráaga was a merchant active in Lagaá, where he appears to have been an overseer of a merchant organization that included much of his extended family. This text, based on its probable provenance, also points to the centrality of Nippur in the affairs of the merchants, and this might be related to the bala.

24

STEVEN J. GARFINKLE

Ur III period. The interest rates were not annual, because the entirety of the interest was due at the repayment of the loan, even if the loan was repaid prior to the passage of one year. There was no customary duration for loans in this period. Unless otherwise stated, the interest and the principal were to be repaid when such repayment became possible for the debtor. As a general rule we can state that silver loans tended to be of greater duration than barley loans. CREDITORS The range of possible creditors was extraordinary, including both the large institutional households and the smaller non-institutional households. A debtor might owe an obligation to the temple of one of the chief gods of the pantheon, to a member of the royal family, to an urban entrepreneur, or to a relative or co-worker. This section focuses on the individual entrepreneurs who made some of their living through lending. 61 My examination of the largest non-institutional archives so far available from the Ur III period has led to some side-by-side comparisons of the archival material. Among the most striking aspects of these comparisons is the quantitative difference in the amount of money-lending activity in the different archives. In the largest such archive, that of SI.A-a, a chief shepherd, loans account for 68% of the seventy-six surviving texts. In the archive of Târam-ili, a merchant, loans account for only 17.5% of the Ž fty-seven surviving texts.62 Of course, the surviving texts may not be an accurate re ection of the original volume or importance of any one particular activity within the economies of the households under consideration. At the same time, these are striking numbers, and they challenge some of our assumptions about creditors in the Ur III economy because we have been apt to associate money-lending in Mesopotamia with the activities of merchants.63 The surviving texts of SI.A-aÕs archive have very little to do with the daily business of shepherding. Instead, he appears to have been a creditor on a large scale, even extending loans to members of the provincial administration in his community. 64 In contrast, Târam-iliÕs professional designation corresponded 61 For discussions of entrepreneurs and their activities, see also Neumann 1992 and van Driel 2002: 24-30. 62 For an overview of the SI.A-a archive, see GarŽ nkle 2003; and for an overview of the Târam-ili archive, see GarŽ nkle 2002. 63 Certainly, loan documents are often found in the archives of merchants, but lending was in no way the sole province of merchants. For example, in SteinkellerÕs (2002: 118-24) recent study of money-lending, none of the archives examined belonged to a merchant. 64 See note 17.

SHEPHERDS, MERCHANTS, AND CREDIT

25

much more closely with his business activities. Târam-ili, who was identiŽ ed in one of his texts as an ugula dam-gˆr, Òan overseer of merchants,Ó65 was most often involved in two activities. First, he was active in the management of balanced account operations, similar to the balanced accounts with which we are familiar from Umma and Lagaá.66 Second, he was engaged in the transfer and purchase of various commodities on behalf of others. The role of the merchant in the Ur III period was as a facilitator of exchange. 67 The role of the shepherd was presumably as a minder of sheep. Whereas, from a modern perspective, TâramiliÕs transactions ably Ž t his job description, those of SI.A-a do not. The prominent role of merchants has been highlighted in previous discussions of the early Mesopotamian economy, but we must recognize that within this economy shepherds and merchants could occupy the same space acting as individual entrepreneurs. It is impossible for us to establish clearly all of the economic activity of a household on the basis of the professional designation of one or more of its members. There are, at the same time, some similarities in the conclusions that we can draw about the nature of the Ur III economy from our examination of these two professionals. The Ur III state relied extensively on local, traditional hierarchies in order to manage its economy. This is not a new conclusion, and Steinkeller (1991; see also GarŽ nkle 2002 and 2003) has already made this point in his examination of the administration of the Ur III state; however, the evidence from the non-institutional archives underscores the dependence of the state on the local elites, who organized themselves largely along familial lines. This was certainly the case as well for credit. During the Ur III era, the overwhelming centralization of the state economy and its administration fostered opportunities for entrepreneurs, in part, by creating the acute need for credit that was serviced by non-institutional households, like those of SI.A-a and his family. Advancing loans at interest was an option available to any household that had the means to do so. There were no apparent social or economic disincentives to money-lending. Indeed, the nature of the Ur III state made this activity attractive and, in some cases, necessary. The individual creditor in the Ur III period was usually the head of an urban household with access to surplus wealth, usually in silver or barley, which could be loaned out in both short and long term agreements. These households were then able to invest the proceeds

JCS 38 37. The balanced accounts were studied extensively by Snell 1982. 67 For recent treatments of merchants and their activities, see Englund 1990: 13-55, Neumann 1999a, and van Driel 2002: Part I. 65 66

26

STEVEN J. GARFINKLE

of their lending activities in the acquisition of slaves, draft animals, and access to additional land; and, of course, they could also invest in further lending. CONCLUSIONS Credit played a vital role in the economy throughout ancient Mesopotamian history. The numerous records that survive from the Ur III period not only document this role, but they also provide broader evidence about socio-economic organization in early Mesopotamia. The availability of credit was critical in economic management at every level of society, from peasants and craftsman to the largest landowning institutions of the state. Money-lending in southern Mesopotamia at the end of the third millennium BC was a diverse enterprise that cannot be easily categorized. The precise incentives behind both borrowing and lending are largely hidden from us, and we should not expect the surviving texts to record them. Yet we can establish that there was a wide range of motivations. For example, some of the interest-free loans involving non-institutional households were actually commissions for the production of Ž nished goods. The ur5-ra loans of the temple households may well have been an attempt on the part of the institutions to help maintain their own workforce, and to prevent that workforce from becoming indebted to noninstitutional creditors. The antichretic loans provide evidence of strategies for dealing with the scarcity of agricultural labor in the Ur III state (especially outside of the great institutions). In the end, we must recognize that much of the surviving evidence supports the conclusion that many loans were issued in the hope of achieving Ž nancial gain through the recovery of the principal plus interest. The determination that creditors in the Ur III period issued loans for productive purposes rests on the following Ž ve points. First, the existence of promissory notes and penalty clauses attests to the desire of the creditor to recover the principal and interest, rather than the labor or land of the debtor. Second, even in an era of acute labor shortages, the value of the loans frequently exceeded the potential value of the debtorÕs labor and land by many times. In addition, creditors had access to antichretic loans for the extraction of labor from their debtors. Third, the court documents make it clear that creditors sought redress within their communities over unpaid debts, and that the solutions involved repayment of the principal plus interest. Fourth, many of the surviving loans involved important members of the Ur III hierarchy. The prominence of many of the debtors argues against the assumption that these men were essentially forced into default for their land or labor. Finally, loans at interest within

SHEPHERDS, MERCHANTS, AND CREDIT

27

families appear to defy an explanation based solely on a desire by the creditor to force the debtor into any form of debt slavery. The variety of credit transactions available to the inhabitants of the Ur III state is indicative of the sophistication of its economic enterprises. The agrarian nature of the economy of southern Mesopotamia was the catalyst for the development of many of the lending practices documented in the textual corpora. The consumptive and antichretic loans have an obvious background in the agricultural regime, and they serve as reminders that debt slavery and dependent status were not uncommon. Some of the interest-free loans probably arose for commercial reasons, both because the early Mesopotamian economy was not fully monetized, and because it was in the nature of the cuneiform tradition to generate texts (or receipts) to document ongoing transactions. The development of productive loans, and especially those advanced in silver, went along with the growth of a class of commercial entrepreneurs who sought economic advantages for their households. These households were most often urban households involved in a competition for prestige and advancement within the complex social and professional hierarchies of early Mesopotamia. Finally, the availability of credit Ž t in nicely with the patrimonial nature of the Ur III state and its administration. Within the large institutional households, credit was a means of further binding the dependent households to the great estates and their administrators. The survival of local and regional hierarchies within the patrimonial state was paralleled by the survival of regional credit practices, and by the ability of the creditors to operate across these regional differences. BIBLIOGRAPHY The following abbreviations are used: AUCT 3 = Sigrist 1988; BE 3/1 = Myhrman 1910; BIN 7 = Alexander 1943; Eames Coll. = Oppenheim 1948; JCS 23/4 = Owen and Young 1971; JCS 38 = Van De Mieroop 1986; JCS 54 = GarŽ nkle 2002; MSL 1 = Landsberger 1937; MVN 8 = Pettinato et al 1979; MVN 9 = Snell 1980; MVN 13 = Sigrist et al. 1984; NATN = Owen 1982; NRVN 1: ‚ig and Kizilyay 1965; TIM 3 = Van Dijk 1966; TMHC NF 1/2 = Pohl 1937; YOS 4 = Keiser 1919; ZA 93/2 = GarŽ nkle 2003. Alexander, John B. 1943. Early Babylonian Letters and Economic Texts. Babylonian Inscriptions in the Collection of James B. Nies, Yale University Vol. VII. New Haven: Yale University Press. ‚ig, M., and H. Kizilyay. 1965. Neusumerische Rechts- und Verwaltungsurkunden aus Nippur—I. Ankara: TŸrk Tarih Kurumu Basimevi. Civil, Miguel 1980. Les limites de lÕinformation textuelle. In LÕarchŽ ologie de lÕIraq du dŽ but de lՎ poque nŽ olithique ˆ 333 avant notre  re: Perspectives et limites de lÕinterprŽtation anthropologique des documents, ed. M. T. Barrelet, pp. 225-32. Paris: Editions du Centre National de la Recherche ScientiŽ que.

28

STEVEN J. GARFINKLE

Englund, Robert K. 1990. Organisation und Verwaltung der Ur III-Fischerei. Berlin: Dietrich Reimer Verlag. ——. 1991. Hard Work—Where Will It Get You? Labor Management in Ur III Mesopotamia. Journal of Near Eastern Studies 50/4: 255-80. GarŽ nkle, Steven 2000. Private Enterprise in Babylonia at the End of the Third Millennium BC. PhD Diss. Columbia University. ——. 2002. Turam-ili and the Community of Merchants in the Ur III Period. Journal of Cuneiform Studies 54: 29-48. ——. 2003. SI.A-a and Family: The Archive of a 21st Century (BC) Entrepreneur. Zeitschrift fŸr Assyriologie 93/2: 161-98. Greengus, Samuel. 1969. The Old Babylonian Marriage Contract. Journal of the American Oriental Society 89/3: 505-32. Hallo, William W. 1977. Seals Lost and Found. In Seals and Sealing in the Ancient Near East, eds. McG. Gibson and R. D. Biggs, pp. 55-60. Malibu: Undena Publications. Harris, Rivkah. 1960. Old Babylonian Temple Loans. Journal of Cuneiform Studies 14: 126-37. Jones, Tom B., and John W. Snyder. 1961. Sumerian Economic Texts from the Third Ur Dynasty, A Catalogue and Discussion of Documents from Various Collections. Minneapolis: University of Minnesota Press. Keiser, C. E. 1919. Selected Temple Documents of the Ur Dynasty. Yale Oriental Series, Babylonian Texts Vol. IV. New Haven: Yale University Press. Landsberger, B. 1937. Die Serie ana ittiáu. Materials for the Sumerian Lexicon Vol. I. Roma: Editrice PontiŽ cio Istituto Biblico. Leemans, W. F. 1950. The Rate of Interest in Old-Babylonian Times. Revue Internationale des Droits de lÕAntiquitŽ 5/3: 7-33. Limet, H. 1969. La clause du double en droit nŽ o-sumŽ rien. Orientalia NS 38: 520-32. Lutzmann, Heiner. 1976. Die neusumerischen Schuldurkunden. Teil I: Einleitung und systematische Darstellung. Ph.D. Diss. Friedrich—Alexander—UniversitŠt Erlangen—NŸrnberg. Myhrman, David W. 1910. Sumerian Administrative Documents Dated in the Reigns of the Kings of the Second Dynasty of Ur from the Temple Archives of Nippur Preserved in Philadelphia. Philadelphia: Department of Archaeology, University of Pennsylvania. Neumann, Hans. 1992. Zur privaten GeschŠftstŠtigkeit in Nippur in der Ur III-Zeit. In Nippur at the Centennial: Papers Read at the 35e Rencontre Assyriologique Internationale, Philadelphia, 1988, ed. Maria de Jong Ellis, pp. 161-76. Philadelphia: Occasional Publications of the Samuel Noah Kramer Fund. ——. 1993. Handwerk in Mesopotamien. Berlin: Akademie Verlag. ——. 1996. Zum privaten Werkvertrag im Rahmen der neusumerischen handwerklichen Produktion. Altorientalische Forschungen 23/2: 254-264. ——. 1999a. Ur-Dumuzida and Ur-Dun: Re ections on the Relationship between StateInitiated Foreign Trade and Private Economic Activity in Mesopotamia towards the End of the Third Millennium BC. In Trade and Finance in Ancient Mesopotamia, ed. J. G. Dercksen, pp. 43-53. Istanbul: Nederlands Historisch-Archeologisch Institut. ——. 1999b. Grundpfandbestellung und Feldabgabe unter rechts- und sozialvergleichendem Aspekt (mit Bemerkungen zur Lesung und Interpretation von CST 60, 11 und MVN III 336, 11). In Landwirtschaft im alten Orient, eds. H. Klengel and J. Renger, pp. 137-48. Berlin: Dietrich Reimer Verlag. Oppenheim, A. Leo. 1948. Catalogue of the Cuneiform Tablets of the Wilberforce Eames Collection in The New York Public Library. New Haven: American Oriental Society. Owen, David I. 1982. Neo-Sumerian Archival Texts Primarily from Nippur in the University Museum, the Oriental Institute, and the Iraq Museum. Winona Lake, Indiana: Eisenbrauns. Owen, D. I., and G. D. Young. 1971. Ur III Texts in the Zion Research Library, Boston. Journal of Cuneiform Studies 23/4: 95-115.

SHEPHERDS, MERCHANTS, AND CREDIT

29

Pettinato, G., S. A. Picchioni, and F. Reshid. 1979. Testi Economici dellÕIraq MuseumBaghdad. Materiali per il Vocabulario Neosumerico Vol. VIII. Roma: MultigraŽ ca Editrice. Pohl, Alfred. 1937. Rechts- und Verwaltungsurkunden der III. Dynastie von Ur. Leipzig: J. C. Hinrichs Verlag. Renger, Johannes. 1995. Institutional, Communal, and Individual Ownership or Possession of Arable Land in Ancient Mesopotamia from the End of the Fourth to the End of the First Millennium BC. Chicago Kent Law Review 71/1: 269-319. Richardson, Seth. 2002. The Collapse of a Complex State: A Reappraisal of the End of the First Dynasty of Babylon, 1683-1597 B.C. Ph.D. Diss. Columbia University. Roth, Martha T. 1997. Law Collections from Mesopotamia and Asia Minor. Second Edition. Atlanta: Scholars Press. Sauren, Herbert. 1978. Les Tablettes cunŽ iformes de lՎ poque dÕUr des collections de la New York Public Library. Louvain-La-Neuve: Institut Orientaliste. Sigrist, Marcel. 1988. Neo-Sumerian Account Texts in the Horn Archaeological Museum. Andrews University Cuneiform Texts Vol. III. Andrews University Press. Sigrist, Marcel, David I. Owen, and Gordon D. Young. 1984. The John Frederick Lewis Collection, Part II. Materiali per il Vocabulario Neosumerico Vol. XIII. Roma: MultigraŽ ca Editrice. Skaist, Aaron. 1994. The Old Babylonian Loan Contract. Ramat Gan, Israel: Bar Ilan University Press. Snell, Daniel C. 1980. The E. A. Hoffman Collection and Other American Collections. Materiali per il Vocabulario Neosumerico Vol. IX. Roma: MultigraŽ ca Editrice. ——. 1982. Ledgers and Prices: Early Mesopotamian Balanced Accounts. New Haven: Yale University Press. Steinkeller, Piotr. 1977. Seal Practice in the Ur III Period. In Seals and Sealing in the Ancient Near East, eds. McG. Gibson and R. D. Biggs, pp. 42-54. Malibu: Undena Publications. ——. 1987. The Foresters of Umma: Toward a DeŽ nition of Ur III Labor. In Labor in the Ancient Near East, ed. Marvin A. Powell, pp. 73-115. New Haven: American Oriental Society. ——. 1991. The Administrative and Economic Organization of the Ur III State: The Core and the Periphery. In The Organization of Power: Aspects of Bureaucracy in the Ancient Near East, Second Edition, eds. McG. Gibson and R. D. Biggs, pp. 15-33. Chicago: The Oriental Institute of the University of Chicago. ——. 2001. The Ur III Period. In Security for Debt in Ancient Near Eastern Law, eds. R. Westbrook and R. Jasnow, pp. 47-62. Leiden: Brill. ——. 2002. Money Lending Practices in Ur III Babylonia: The Issue of Economic Motivation. In Debt and Economic Renewal in the Ancient Near East, eds. M. Hudson and M. Van De Mieroop, pp. 109-37. Bethesda, MD: CDL Press. Van De Mieroop, Marc. 1986. Târam-IlÂ: An Ur III Merchant. Journal of Cuneiform Studies 38/1: 1-80. ——. 1995. Old Babylonian Interest Rates: Were They Annual? In Immigration and Emigration within the Ancient Near East, Festschrift E. Lipinski, eds. K. Van Lerberghe and A. Schoors, pp. 357-64. Leuven: Uitgeverij Peeters en Departement Ori‘ntalistiek. ——. 1997. Why Did they Write on Clay? Klio 79: 7-18. ——. 2002. A History of Near Eastern Debt? In Debt and Economic Renewal in the Ancient Near East, eds. M. Hudson and M. Van De Mieroop, pp. 59-94. Bethesda, MD: CDL Press. Van Dijk, J. 1966. Texts in the Iraq Museum, Volume III, Cuneiform Texts: The Archives of Nâr-°amaá and Other Loans. Wiesbaden: Otto Harrassowitz. van Driel, G. 1998. Land in Ancient Mesopotamia: ÒThat what remains undocumented does not exist.Ó In Landless and Hungry? Access to Land in Early and Traditional Societies,

30

STEVEN J. GARFINKLE

eds. R. Haring and R. de Maaijer, pp. 19-49. Leiden: School of Asian, African, and Amerindian Studies. ——. 1999. Capital Formation and Investment in an Institutional Context in Ancient Mesopotamia. In Trade and Finance in Ancient Mesopotamia, ed. J. G. Dercksen, pp. 25-42. Istanbul: Nederlands Historisch-Archeologisch Institut. ——. 2002. Elusive Silver: In Search of a Role for a Market in an Agrarian Environment— Aspects of Mesopotamian Society. Istanbul: Nederlands Historisch-Archeologisch Institut. Wilcke, Claus. 2002. Der Kodex Urnamma (CU): Versuch einer Rekonstruktion. In Riches Hidden in Secret Places: Ancient Near Eastern Studies in Memory of Thorkild Jacobsen, ed. T. Abusch, pp. 291-333. Winona Lake, Indiana: Eisenbrauns.

LE GRAND JASAQ DE GENGIS-KHAN, LÕEMPIRE, LA CULTURE MONGOLE ET LA SHARÁ®A BY

DENISE AIGLE* Abstract Mongol law, the jasaq, has provided the basis for a long tradition of studies which were inaugurated by Petis de la Croix in 1710. He was the Ž rst to deŽ ne a list of precepts of the jasaq, but without taking into consideration either the chronology or their origins. Most subsequent scholars dealing with the question revived this same vision of the jasaq. Debate was especially focused on whether or not the Mongols possessed a written code of laws. But, until now, little discussion has taken place concerning what the jasaq represented for the Mongols themselves and how this Mongol law was perceived by Mediaeval authors who, on the whole, confused the imperial edicts ( jasaq) with customs (yosun). The present article is an attempt to clarify these issues. The author examines the jasaq in its politico-cultural context and, in particular, the analysis of the precepts takes into consideration shamanism, the Mongol system of representations. Reasons for the lack of understanding by Muslims of certain customs in disharmony with Islam are thereby highlighted, reasons which led them to see, in the jasaq, an equivalent of the shar®a: a Mongol order imposed on populations which had fallen under their domination. La loi mongole, le jasaq, a donnŽ lieu ˆ une longue tradition dՎ tudes qui fut inaugurŽe par Petis de la Croix au XVIIe si cle. Il fut le premier ˆ dresser une liste des prŽ ceptes du jasaq, sans tenir compte de la chronologie des sources et de leur provenance. Les chercheurs qui se sont penchŽ s par la suite sur cette question ont repris, pour la plupart dÕentre eux, cette vision du jasaq. Le dŽ bat sÕest surtout focalisŽ sur lÕexistence ou non dÕun code de loi Ž crit chez les Mongols. Mais, jusquÕˆ prŽ sent, il a Ž tŽ peu discutŽ de ce que le jasaq reprŽ sentait pour les Mongols eux-m mes et comment cette loi mongole a Ž tŽ per ue par les auteurs mŽ diŽ vaux qui confondaient, la plupart du temps, les Ž ditŽ s impŽ riaux ( jasaq) et les coutumes ( yosun). Cet article fait le point sur ces questions. LÕauteur examine le jasaq dans son contexte politico-culturel et, en particulier, il prend en compte, dans lÕanalyse des prŽ ceptes, le syst me de reprŽ sentations des Mongols, le chamanisme. Il met ainsi en lumi re les raisons de lÕincomprŽhension, de la part des musulmans, de certaines coutumes en dŽ saccord avec lÕislam, ce qui les a conduits ˆ voir dans le jasaq lՎ quivalent de la shar®a : un ordre mongol imposŽ aux populations tombŽ es sous leur domination. Keywords: jasaq, shar®a, yosun, Mongols, Mamelouks, Gengis-khan, Tenggeri

* Denise Aigle, Institut Fran ais du Proche-Orient, BP 344, Damascus, Syria, [email protected]. Merci ˆ Marie-Dominique Even qui a bien voulu relire une premi re version de cet article qui a pu ainsi sÕenrichir de ses commentaires. © Koninklijke Brill NV, Leiden, 2004 Also available online – www.brill.nl

JESHO 47,1

32

DENISE AIGLE

Au XIIIe si cle, les Mongols, et, ˆ leur t te, les qan de la lignŽ e de Gengis-khan (Cinggis-qan), sont passŽ s de lÕobscuritŽ au rang de grands conquŽ rants. La formation de lÕempire mongol provoqua une cŽ sure dans lÕhistoire de lÕEurasie : pendant un si cle et demi environ, des pays de vieilles traditions sŽ dentaires, comme la Chine et lÕIran, furent soumis ˆ la domination dÕun m me peuple des steppes 1. LÕirruption dans lÕempire musulman, le d¨r al-isl¨m, de ces tribus nomades, dont le mode de vie et la religion Ž taient Ž trangers aux musulmans, a dÕautant plus frappŽ les esprits que, pour la premi re fois, une grande partie des territoires islamiques situŽ s ˆ lÕest de lÕƒgypte se trouva assujettie ˆ un pouvoir non musulman. Ce bouleversement gŽ opolitique a rapidement suscitŽ nombre de rŽ actions, de rŽ cits et de mythes qui ont Ž voluŽ au fur et ˆ mesure que les Mongols sÕinstallaient au cœur de lÕempire. Cette domination Ž trang re sur les pays dÕIslam a, notamment, Ž tŽ symbolisŽ e par un « grand y¨s¨2 », une lex mongolica qui sÕopposait ˆ la loi islamique, la shar®a. D s lÕorigine, les mentions du y¨s¨ dans les sources mŽ diŽ vales tŽ moignent de lÕambigu•tŽ du terme dans lÕesprit des auteurs. Les Mongols Ž taient un peuple de tradition orale jusquÕˆ Gengis-khan, qui aurait ordonnŽ dÕadapter lՎ criture des Ou•ghours ˆ leur langue. MalgrŽ cela, aucun ouvrage historique nÕa subsistŽ sous sa forme originale en mongol, mais seulement ˆ travers des versions chinoises et tibŽ taines3. CÕest le cas de lÕHistoire secr te des Mongols, texte fondateur dans lequel sont mentionnŽ s un certain nombre de « jasaq », au sens de r gle Ž dictŽ e par un souverain. Les seules sources qui parlent dÕun « grand y¨s¨ », ou m me dÕun ensemble de y¨s¨, sont extŽ rieures ˆ la culture mongole. Il sÕagit, en particulier, des sources islamiques mŽ diŽ vales. La tradition Ž rudite sÕest appuyŽ e sur ces derni res pour reprendre lÕidŽ e dÕun « grand code (yeke jasaq) » mongol.

1 Une vue dÕensemble sur les Ž tudes mongoles est prŽ sentŽ e par David Morgan, « The Mongol Empire : A Review Article », BSOAS (1981) 14 : 121-125 ; idem, The Mongols (Oxford : Blackwall, 1986) : 5-31 et Peter Jackson, « The Mongol Empire, 1986-1999 », Journal of Medieval History (2002) 26/2 : 189-210. Voir la bibliographie et les abrŽ viations en Ž n dÕarticle. 2 Le terme appara”t, dans les sources islamiques, sous des graphies diffŽ rentes : yasaq, j¨s¨q, y¨s¨q, y¨s¨. Voir lÕimportante notice « Y¨s¨q » dans Doerfer, vol. IV, n¡ 1789. Nous avons adoptŽ la forme jasaq lorsquÕil sÕagit de sources en mongol classique (mo. cl.) et y¨s¨ lorsquÕil sÕagit de sources en persan et en arabe, car cÕest sous cette forme que le terme appara”t le plus souvent. Pour la translittŽ ration des termes arabes et persans, nous suivons le syst me anglo-saxon, proche de celui de lÕEncyclopŽ die de lÕIslam ; pour les termes chinois, nous avons adoptŽ le syst me Wade. Les noms propres mongols sont translittŽrŽ s selon le syst me de la langue mongole, sans voyelles longues. 3 Sur les rapports entre les historiographies mongole, chinoise, tibŽ taine et persane, voir Shira Bira, « Historiography among the Mongols », in History of Civilisations of Central Asia, vol. IV, part 2, The age of the achievement : A.D. 750 to the end of the Ž fteenth century (Paris : Unesco Publishing, 2000) : 156-160 ; Thomas Allsen, Culture and Conquest in Mongol Eurasia (Cambridge : Cambridge University Press, 2001) : 83-102.

LE GRAND JASAQ DE GENGIS-KHAN

33

Ce paradoxe est le point de dŽ part du prŽ sent article. Nous verrons que les dŽ veloppements autour du y¨s¨, d s le XIIIe si cle, sÕinscrivent dans un contexte historique prŽ cis : lÕimposition dÕun ordre mongol sur une partie du d¨r al-isl¨m et la guerre idŽ ologique ˆ laquelle se livr rent les Mamelouks et les Mongols dÕIran, ˆ la suite des prŽ tentions de ces derniers sur les pays du Levant. La prŽ sence mongole dans le d¨r al-isl¨m incita les musulmans ˆ ampliŽ er lÕopposition entre la shar®a et un « grand y¨s¨ » qui devint, au Ž l du temps, de plus en plus mythique. LÕexamen des sources mŽ diŽ vales montre, en effet, que la plupart des auteurs regroupaient sous le terme y¨s¨ des dŽ crets Ž dictŽ s par les qan ainsi quÕun certain nombre de r gles coutumi res mongoles. Nous proposons, dans cet article, de clariŽ er la notion de y¨s¨ en distinguant son sens strict en mongol et son acception Ž tendue chez les auteurs musulmans. Nous discuterons de la signiŽ cation des dŽ crets impŽ riaux, puis des coutumes, selon une perspective historique et anthropologique qui tienne compte du milieu dÕorigine et du syst me de reprŽ sentations des Mongols. Nous mettrons parall lement en lumi re les raisons de lÕincomprŽ hension, de la part des musulmans, envers certaines coutumes et interdits. Pour cela, il est nŽ cessaire de poser le cadre historique dans lequel sÕest dŽ veloppŽ e, dans lÕempire musulman, lÕidŽ e dÕun grand y¨s¨ mongol. LA PERCEPTION DU Y¤S¤ DANS LÕORIENT MUSULMAN LÕirruption des Mongols dans le d¨r al-isl¨m Ë la veille de leur entrŽ e dans lÕhistoire, ˆ lÕaube du XIIIe si cle, les clans instables des nomades de Mongolie furent rassemblŽ s par un chef Ž nergique, TemŸjin, le futur Gengis-khan, qui pratiqua une politique dՎ pouvante aŽ n dÕeffrayer les populations quÕil souhaitait soumettre, sur le mode de la vengeance Ž pique 4. Ceux qui acceptaient la soumission avaient la vie sauve tandis que la rŽ sistance Ž tait assimilŽ e ˆ une offense et punie par un massacre gŽ nŽ ral. Cette fŽ dŽ ration marqua un tournant dŽ cisif dans lÕhistoire de la sociŽ tŽ mongole. En effet, en quelques dŽ cennies, ces nomades de la steppe form rent un immense « ƒtat » qui, bien que dirigŽ ˆ la mani re patrimoniale traditionnelle, fut rapidement dotŽ des structures administratives et judiciaires nŽ cessaires pour contr™ler lÕensemble des territoires conquis. Soumettre le monde au Tenggeri – LÕhistoriographie contemporaine a souvent mis lÕaccent sur la croyance en un « mandat cŽ leste » dont aurait Ž tŽ gratiŽ Ž Gengis-khan Sur les conqu tes et lÕexpansion de lÕempire, voir Fran oise Aubin, « Mongolie (histoire) », in Encyclopaedia Universalis, 1974, vol. 11 : 656-684 ; David Morgan, The Mongols (Oxford : Blackwall, 1986). 4

34

DENISE AIGLE

par le « Ciel Ž ternel (mšngke tenggeri) » pour rendre compte de la rapiditŽ des conqu tes. Dans un article paru en 1993, Marie-Lise Beffa, sÕappuyant sur lÕanalyse linguistique du concept « Ciel (tenggeri) » dans lÕHistoire secr te des Mongols a fortement nuancŽ cette th se 5. La notion de « mandat du Ciel Ž ternel » ou « dŽ cret du Ciel Ž ternel (mšngke tenggeri-yin jarliq) » nÕappara”t en effet quÕapr s lÕintronisation du qan en 1206 et, de surcro”t, dans la bouche de Teb-tenggeri6 pour mettre en doute la lŽ gitimitŽ du nouveau chef : « Le dŽ cret du Ciel Ž ternel dit le prŽ sage de qan, une fois il dit que TemŸjin prenne lÕulus 7, une autre fois il dit que ce soit Qasar8. » Aussit™t, Gengis-khan Ž limina Teb-tenggeri, le « porte-parole du Ciel9 ». Marie-Lise Beffa estimait, quÕˆ cette Ž poque, il nՎ tait plus possible de se dŽ barrasser de Gengiskhan par le seul jeu des rapports de force et que, pour cette raison, il Ž tait fait appel au surnaturel. Mais elle constatait que, dans les derniers chapitres de lÕHistoire secr te des Mongols, Gengis-khan, lui aussi, faisait clairement appel au Ciel comme, par exemple, au paragraphe 240, o il est question de donner ˆ lÕarmŽ e lÕordre de « prier le Ciel Ž ternel » pour soumettre les TŸmet10. Au dŽ but de sa carri re, Gengiskhan nՎ tait pas encore « investi dÕun mandat », mais il Ž tait « protŽ gŽ par le Ciel », qui lui envoyait des signes de faveur, manifestŽ s par les victoires quÕil remportait sur les ennemis.

5 Marie-Lise Beffa, « Le concept de tŠnggŠri, “ciel” dans lÕHistoire secr te des Mongols », ƒtudes mongoles et sibŽ riennes, (1993) 24 : 215-236. La notion de mandat Ž ternel est toujours sujette ˆ discussion. David Morgan dans The Mongols avait suggŽ rŽ que cette notion Ž tait postŽ rieure ˆ Gengis-khan. John M. Smith a prŽ sentŽ une opinion contraire dans « The Mongols and World-Conquest », Mongolica (1994) 5 : 206-214. Cependant, il semble confondre deux notions : celle de lÕempire et celle du ciel protecteur Ž ternel. 6 Teb-Tenggeri Ž tait le Ž ls de Menglig, de la tribu des Qongqotan. 7 Le patrimoine dÕun qan se composait de lÕulus, « gens, peuple », cÕest-ˆ-dire dÕun certain nombre de feux nomades (ayil) et du territoire (nuntuq) sur lequel ces gens nomadisaient, voir Boris Vladimirtsov, Le rŽ gime social des Mongols. Le fŽ odalisme nomade (Paris : Adrien Maisonneuve, 1948) : 131. 8 Histoire secr te des Mongols (HS) § 244. Qasar Ž tait le fr re de Gengis-khan, de deux ans son cadet. 9 Bien que Teb-Tenggeri ne soit pas explicitement dŽ signŽ comme un chamane, Gengis-khan en lՎ liminant cherchait ˆ marginaliser le chamanisme. Il fut mis ˆ mort pour avoir semŽ le trouble entre lui et ses fr res. Al-Juvayn donne une version totalement diffŽ rente du r™le de Teb-tenggeri : « Dieu (khâd¨) a parlŽ avec moi », dit Teb-tenggeri, « et il a dit jÕai donnŽ toute la face de la terre ˆ TemŸjin et ˆ ses enfants ( farzand¨n-i hâ), et je lÕai nommŽ Gengis-khan. Dis-lui dÕagir avec justice (®adl) », al-JuvaynÂ, Ž d., vol. I : 29 ; trad., vol. I : 39. LÕhistorien persan Ž crit ˆ lՎ poque de Mšngke ; on trouve, dans ce rŽ cit, lÕidŽ ologie du « mandat cŽ leste » mise en œuvre par les successeurs de Gengis-khan. HŸlegŸ prŽ senta encore Teb-tenggeri (en latin : Temptemgri/Teptemgri) comme garant de la mission de Gengis-khan dans une lettre quÕil adressa ˆ Louis IX en 1262, voir Paul Meyvaert, « An unknown Letter of Hulagu, Il-Khan of Persia, to King Louis IX of France », Viator (1980) 11 : 252. 10 Marie-Lise Beffa, « Le concept de tŠnggŠri, “ciel” » : 222-223. Les TŸmet faisaient partie des peuples turco-mongols de la for t sibŽ rienne.

LE GRAND JASAQ DE GENGIS-KHAN

35

La conqu te de lÕIran oriental – Le « mandat cŽ leste » ne fut pas la cause de la premi re invasion des pays iraniens, en 1219, mais le rŽ sultat de la politique dŽ sastreuse du Khw¨razmsh¨h, ®Al¨¾ al-DÂn Mu½ammad, qui dominait alors la majeure partie de lÕIran oriental. En 1215, peu apr s la chute de PŽ kin, Gengis-khan chercha ˆ Ž tablir des liens commerciaux avec la cour du souverain iranien ; il envoya une dŽ lŽ gation porteuse dÕun message lui demandant de laisser circuler librement dans leurs territoires respectifs les marchands et leurs caravanes. Le qan traitait alors ®Al¨¾ al-DÂn Mu½ammad sur un pied dՎ galitŽ : « Je suis le souverain [des territoires] o le soleil se l ve, vous  tes le souverain [des territoires] o il se couche 11. » En 1218, une caravane conduite par quatre marchands arriva ˆ Ußr¨r ; tous furent massacrŽ s par le gouverneur du Khw¨razmsh¨h, lequel refusa de livrer le coupable. Ces Ž vŽ nements dŽ clench rent lÕinvasion de lÕIran oriental, marquŽ e par le massacre gŽ nŽ ral (qatl-i ®¨mm) de la population des grandes villes de Transoxiane et du Khorassan. Outre ses consŽ quences politiques immŽ diates, lÕaffaire dÕ Ußr¨r tŽ moignait de lÕimportance que Gengis-khan accordait au commerce. On verra que certains prŽ ceptes du y¨s¨ visaient ˆ faciliter la circulation des grands nŽ gociants dans lÕempire et ˆ rŽ glementer leurs activitŽ s. La marche vers les pays du Levant – La conqu te des pays musulmans, commencŽ e par Gengis-khan, fut achevŽ e par Mšngke (r. 1251-1259), son troisi me successeur ˆ la t te de lÕempire. Il dŽ cida que, tout en restant sous son autoritŽ , les pays iraniens seraient constituŽ s en un ulus au proŽ t de son fr re HŸlegŸ, aŽ n que celui-ci contr™le les richesses du pays et mette en œuvre les coutumes (rusâm va yâsân12) et la r gle (y¨s¨) de Gengis-khan dans les territoires situŽ s « de lÕOxus ˆ lÕƒgypte 13 ». La conqu te achevŽ e, HŸlegŸ se retira en Azerba•jan, o les troupes mongoles pouvaient trouver les zones de p‰turages nŽ cessaires pour nourrir troupeaux et chevaux. Les prŽ tentions des Mongols sur les pays dÕIslam ne sÕarr t rent pas lˆ. En dŽ cembre 1259, les armŽ es de HŸlegŸ sÕinstall rent dans le nord de la Syrie. Elles ne purent pŽ nŽ trer plus avant, car leur progression fut arr tŽ e, le 3 septembre 1260, par le sultan mamelouk al-Malik al-Muúaffar Qußuz et par Baybars, qui remport rent une victoire mŽ morable ˆ ®Ayn J¨lât 14. LՎ tat de guerre entre Mamelouks et Ilkhans dura plus de soixante ans, jusquÕˆ la signature dÕun traitŽ de paix en 1323 15. Al-Jâzj¨nÂ, Ž d. : 336 ; trad., vol. II : 966. Le terme yâsân vient du mongol yosun. 13 RashÂd al-DÂn, Ž d. Alizade, vol. III : 23. 14 Voir Reuven Amitai-Preiss, « In the Aftermath of ®Ayn J¨lât : The Beginnings of the MamlâkÁlkh¨nid Cold War », al-Mas¨q (1990) 10 : 1-21 ; idem, « ®Ayn J¨lât Revisited », Tarih (1992) 2 : 119-150. 15 Sur lՎ tat de guerre entre Ilkhans et Mamelouks, voir Reuven Amitai-Preiss, Mongols and Mamluks. The Mamluk-Álkh¨nid War, 1260-1281(Cambridge : Cambridge University Press, 1995). 11 12

36

DENISE AIGLE

Entre y¨s¨ et shar®a, une guerre idŽ ologique LՎ chec des troupes mongoles ˆ ®Ayn J¨lât avait Ž xŽ des zones dÕin uences : les Mamelouks dominaient les pays du Levant tandis que, de lÕautre c™tŽ du dŽ sert syrien, les Ilkhans tenaient la MŽ sopotamie, le cœur du d¨r al-isl¨m abbasside. Les prŽ tentions des Ilkhans sur la Syrie sont attestŽ es par les nombreuses missions envoyŽ es en Occident latin aŽ n de rechercher lÕalliance des princes chrŽ tiens et de la papautŽ . Ils adress rent, parall lement, une sŽ rie de lettres et dÕambassades aux sultans mamelouks les invitant ˆ la soumission. Ces Ž changes diplomatiques donn rent lieu ˆ une vŽ ritable guerre idŽ ologique entre les deux ƒtats. Les Mamelouks Ž taient arrivŽ s au pouvoir au moment o Ž tait perceptible, dans le d¨r al-isl¨m, une nouvelle idŽ ologie : les successeurs de Gengis-khan revendiquaient la « bonne fortune impŽ riale (suu) » octroyŽ e ˆ la famille royale par le « Ciel 16 ». Dans ce contexte, tout souverain qui refusait de se soumettre au pouvoir mongol Ž tait considŽ rŽ comme rebelle, non seulement contre les Gengiskhanides, mais aussi contre « Tenggeri ». On trouve ce th me de mani re rŽ currente chez les successeurs de Gengis-khan : toutes les nations de la terre sont invitŽ es ˆ «  tre en paix et en harmonie avec les Mongols », en dÕautres termes ˆ une soumission totale et inconditionnelle 17. La plupart des lettres, en mongol et en latin, envoyŽ es en Occident par les Ilkhans mettent lÕaccent sur le mandat cŽ leste ; elles commencent par la formule : « Dans la force du ciel Ž ternel, dans la fortune du qan [. . .] parole de nous (mšngke tenggeri-yin kŸcŸn-dŸr, qan-u suu-dur [. . .] Ÿge manu)18. Les Mamelouks, dÕorigine servile, nÕavaient rien ˆ produire dՎ quivalent ˆ ce nasab « mongol »: ils Ž taient marquŽ s par les stigmates du manque dÕascendance, 16 Voir Igor de Rachewiltz, « Some remarks on the ideological foundation of Chinggis khanÕs empire », Paper on Far Eastern History (1973) 7 : 21-36. Ce concept Ž tait connu dans la doctrine politique des Chinois, avec lesquels les Mongols Ž taient en contact depuis longtemps. Selon Peter B. Golden, les parall les avec les Turcs sont plus pertinents, voir « Imperial ideology and the sources of political unity amongst the pre-cinggisid nomads of the western Eurasia », Archivum Eurasiae Medii Aevi (1981) 2 : 37-76. 17 Reuven Amitai-Preiss, « Mongol Imperial Ideologie and the Ilkhanid War against the Mamluks », in The Mongol Empire and its Legacy, Ž d. Reuven Amitai-Press et David Morgan (Leyde : Brill, 1999) : 62. 18 Lettre envoyŽ e par Arghun ˆ Philippe le Bel en 1289, Ž d. Antoine Mostaert et Francis Woodman Cleaves, Les lettres de 1289 et 1305 des Ilkhan Argun et …ljeitŸ ˆ Philippe le Bel, (Cambridge, Mass. : Harvard University Press, 1962) : 17-18. Nous disposons dÕune lettre en latin adressŽ e Ž galement par Arghun au pape HonorŽ IV en 1285 qui commence par la m me formule : « Potentia dei vivi et gratia magni cam », Ž d. Karl-Ernst Lupprian, Die Beziehungen der PŠpste zu islamischen und mongolischen Herrschern im 13. Jahrhundert anhand ihres Briefwechsels, (Cittˆ del Vaticano : Biblioteca Apostolica Vaticana, 1981) : 245. Remarquons que les lettres envoyŽ es par les Ilkhans convertis ˆ lÕislam ne dŽ butent pas par la rŽ fŽ rence au ciel, mais on la trouve dans le corps des lettres. …ljeitŸ, dans sa lettre envoyŽ e ˆ Philippe le Bel en 1305, Ž crit : « Maintenant que dans la force du ciel (tenggeri-yin kŸcŸndŸr) nous sommes assis sur le grand tr™ne », Les lettres de 1289 et 1305 des Ilkhan Argun et …ljeitŸ : 55.

LE GRAND JASAQ DE GENGIS-KHAN

37

dans un monde obsŽ dŽ par la gŽ nŽ alogie. En 1269, lÕIlkhanide Abaqa fait dire ˆ Baybars par un messager : « La meilleure chose que tu puisses faire est la paix (×ul½) avec nous [. . .]. Tu es un esclave achetŽ ˆ SÂw¨s, comment peux-tu tÕopposer aux rois de la terre19 ? » Abaqa, souverain de sang impŽ rial et mandatŽ par le ciel, ne pouvait quÕexprimer son mŽ pris pour des rebelles illŽ gitimes et sans anc tres. Baybars nÕavait dÕautre choix, pour se forger une lŽ gitimitŽ , que de se prŽ senter comme le protecteur de la vraie foi face ˆ une dynastie per ue comme pa•enne et tyrannique. Le y¨s¨ fâst ainsi un des Ž lŽ ments de la guerre idŽ ologique ˆ laquelle se livr rent Mamelouks et Ilkhans. Le message envoyŽ par Abaqa ˆ Baybars en 1269 contenait quatre rŽ fŽ rences au y¨s¨20. Dans ce texte, les mentions du y¨s¨ Ž taient associŽ es aux termes farm¨n et yarlÂgh et au titre qan, dŽ signant Qubilai, dont Abaqa dŽ pendait. Le contexte invite ˆ traduire farm¨n et yarlÂgh par « commandements et dŽ crets » de lÕactuel qan, ce qui leur donne un statut Ž quivalent ˆ ceux de Gengis-khan. Dans sa rŽ ponse ˆ Abaqa, Baybars proclama : « Notre y¨s¨ est supŽ rieur ˆ celui de Gengiskhan. Dieu nous a donnŽ de rŽ gner sur quarante rois21. » Nul doute que Baybars utilisait ici le terme y¨s¨ dans le sens de shar®a22. On ne sera pas Ž tonnŽ que les circonstances historiques aient conduit les musulmans ˆ percevoir le y¨s¨ comme la shar®a et la sunna des Mongols, un ensemble de lois en dŽ saccord avec lÕislam, qui resta en vigueur m me apr s la conversion ofŽ cielle de lÕIlkhan Ghazan-khan ˆ lÕislam, en 1295. LÕhistorien persan RashÂd al-DÂn tŽ moigne en effet ˆ plusieurs reprises de lÕattachement de Ghazan-khan au y¨s¨. Il rapporte que ce dernier avait lÕhabitude de rassembler les compagnons de son ‰ge (atr¨b, pl. de tirb) et de leur enseigner les coutumes (yâsân) et le y¨s¨. Il avait assignŽ un rang ˆ chacun (martabat) : les anciens (¨q¨), les jeunes (ÂnÂ), les alliŽ s jurŽ s (anda 23) et les familles par alliance (qud¨y24) ; 19 Reuven Amitai-Preiss, Mongols and Mamluks : 121 ; Anne F. Broadbridge, « Mamluk Legitimacy and Mongols : The Reigns of Baybars and Qal¨wân Ó, Mamluk Studies Review (2001) 5 : 107. 20 Reuven Amitai-Preiss, « An Exchange of Letters in Arabic between Abaga Álkh¨n and Sultan Baybars (A.H. 667/A.D. 1268-69) », Central Asiatic Journal (1994) 38/1 : 28-30. 21 Reuven Amitai-Preiss, « An Exchange » : 30. 22 Voir la discussion sur le sens, dans cette lettre, donnŽ au terme y¨s¨ dans Reuven AmitaiPreiss, « An Exchange »: 31 ; Anne F. Broadbridge, « Mamluk Legitimacy and Mongols » : 108109, et dÕune mani re plus gŽ nŽ rale sur le y¨s¨ voir David Ayalon, « The Great y¨s¨ », Part. A. 23 Un anda est un alliŽ de chasse ou de relation matrimoniale en position de rŽ ciprocitŽ, voir Roberte Hamayon, La chasse ˆ lÕ‰me. Esquisse dÕune thŽ orie du chamanisme sibŽ rien (Nanterre : SociŽ tŽ dÕethnologie, 1990) : 768, note 9. 24 Qud¨y est la forme persanisŽe du mongol quda, terme qui signiŽ e « preneur de femme », « comp re en mariage », il sÕagit dÕun alliŽ matrimonial en position non rŽ ciproque. Mais le sens exact du terme est discutŽ. Doerfer (vol. I, n¡ 296), par exemple, voit dans le terme quda la rŽ ciprocitŽ entre les deux parties. Sur cette question, voir Ž galement Rodica Pop, Le mariage chez les Mongols. Rites et textes (Th se de doctorat EPHE, sous la direction de R. N. Hamayon, Sorbonne, Paris, 2001).

38

DENISE AIGLE

si lÕun dÕentre eux franchissait cette limite, il Ž tait rappelŽ sur le chemin du y¨s¨25. Ghazan-khan respectait ainsi la hiŽ rarchie classique de la sociŽ tŽ mongole prŽ gengiskhanide, conservŽ e dans le cadre du nouvel ordre impŽ rial. Quelle que soit lÕhistoricitŽ de ce passage, la dŽ votion de Ghazan-khan au y¨s¨ est prŽ sentŽ e de mani re positive par RashÂd al-DÂn26. Il est lÕun des rares historiens persans ˆ transmettre les dŽ crets impŽ riaux (yarlÂgh) de Ghazan-khan. Selon lÕun de ces yarlÂgh, il aurait ordonnŽ la distribution dÕiqߨ® aux soldats mongols, cÕest-ˆ-dire de terres dont ils avaient lÕusufruit en guise de salaire27. Cette pratique Ž tait Ž trang re au mode de rŽ munŽ ration des armŽ es gengiskhanides, dont la base Ž tait constituŽ e par la distribution du butin de guerre. Le dŽ cret dŽ butait par la louange du y¨s¨ de Gengis-khan, ˆ qui Ghazan-khan attribuait les conqu tes victorieuses de ses prŽ dŽ cesseurs. Mais David Morgan consid re que cette louange du y¨s¨ avait pour but de dŽ tourner lÕattention des Ž carts que le nouvel Ilkhan commen ait ˆ se permettre28. Les sources mameloukes mettent, elles aussi, lÕaccent sur lÕattachement de Ghazan-khan au y¨s¨. Selon al-ÑafadÂ, lorsque Ghazan-khan accŽ da au pouvoir, il suivit la mani re de gouverner (al-siy¨sa) de Gengis-khan et il Ž tablit le y¨s¨ des Mongols (al-y¨s¨ al-mughâliyya)29. AŽ n de remettre en cause la sincŽ ritŽ de la conversion de Ghazan-khan, al-Ñafad dŽ non ait son respect des coutumes mongoles. En effet, peu apr s sa conversion, le nouvel Ilkhan prit pour Ž pouse Bulughan Khatun, la veuve de son p re Arghun, qui lui plaisait. Cela lui Ž tait possible, les Mongols pratiquaient le lŽ virat du cadet envers les veuves de son p re, en dehors de sa propre m re, ou dÕun fr re a”nŽ . Mais cette pratique Ž tait en dŽ saccord total avec la shar®a. Ghazan-khan aurait abandonnŽ lÕislam si un savant en sciences religieuses nÕavait trouvŽ une solution ˆ cette impasse en dŽ clarant le prŽ cŽ dent mariage de Bulughan Khatun avec Arghun non valide du point de vue de la shar®a, puisque ce dernier nՎ tait pas musulman30. En rapportant, dans le dŽ tail, le mariage de Ghazan-khan avec Bulughan Khatun, al-Ñafad accentuait, de mani re implicite, la disparitŽ entre le y¨s¨ et la shar®a.

25

RashÂd al-DÂn, Ž d. Alizade, vol. III : 251. Reuven Amitai-Preiss, « Ghazan, Islam and Mongol Tradition : A View from the Mamlâk Sultanate », Bulletin of the School of Oriental and African Studies (1996) LIX/1 : 4. 27 Voir la discussion sur le sens ˆ donner au terme iqt¨® dans le contexte de ce yarlÂgh dans Reuven Amitai-Preiss, « Turko-Mongolian Nomads and the Iqt¨® System in the Islamic Middle East (ca 1000-1400 AD) », in Nomads in the Sedentary World, Ž d. Anatoly M. Kazanov et AndrŽ Wink (Curzon : Curzon-IIAS Asian Studies Series, 2001) : 159-160. 28 David O. Morgan, « The ÔGreat Y¨s¨ of Chingiz Kh¨nÕ and Mongol Law in the Álkh¨nate », Bulletin of the School of Oriental and African Studies (1986) 49/1 : 172. 29 Reuven Amitai-Preiss, « Ghazan, Islam and Mongol Tradition » : 3-4. 30 Sur ce mariage, voir Reuven Amitai-Preiss, « Ghazan, Islam and Mongol Tradition » : 1-3. 26

LE GRAND JASAQ DE GENGIS-KHAN

39

MENTIONS TEXTUELLES SUR LE Y¤S¤ ET TRADITION ERUDITE Le y¨s¨ dans les sources mŽ diŽ vales Nous disposons sur le y¨s¨ mongol dÕun ensemble de sources de nature et dÕorigine multiples : les historiographies mongole, persane, arabe, syriaque et armŽ nienne, ainsi que les rŽ cits en latin des missionnaires occidentaux qui furent envoyŽ s dans lÕempire. Nous nÕavons retenu ˆ lÕappui des analyses que les sources qui contiennent des Ž lŽ ments substantiels sur le y¨s¨, soit sous la forme dÕune liste des infractions aux lois et aux coutumes mongoles passibles, le plus souvent, de la peine capitale,soit de fragments de chapitres consacrŽ s aux prŽ ceptes et coutumes mentionnŽ s dans les listes31. La littŽ rature historique mongole – La source mongole la plus ancienne est lÕHistoire secr te des Mongols 32, rŽ digŽ e peu de temps apr s la mort de Gengis-khan 33 ; elle avait pour vocation de cŽ lŽ brer ses hauts faits. LÕHistoire secr te est fondatrice de lÕidentitŽ mongole ; elle est dÕune importance capitale pour conna”tre lÕorganisation sociale, les valeurs et lÕunivers culturel des tribus mongoles. Ce texte, qui se trouve ˆ la croisŽ e de plusieurs genres, peut laisser lÕhistorien dŽ semparŽ . Il est, en premier lieu, une « histoire clanique » que les membres de la lignŽ e doivent avoir en mŽ moire. Il est nŽ cessaire, en effet, de conna”tre les noms de ses anc tres, des alliŽ s et des ennemis, aŽ n de pouvoir se situer dans le syst me vindicatoire 34. LÕHistoire secr te tient aussi de lՎ popŽ e, dont la fonction est de vŽ hiculer les valeurs de la sociŽ tŽ . Ce texte, enŽ n, est un rŽ cit historique qui relate « la saga gengiskhanide », depuis les origines de TemŸjin jusquÕˆ lÕav nement dÕ…gšdei, en 1229 35.

31 Nous ne prŽ tendons pas analyser ici de mani re exhaustive les mentions isolŽ es du y¨s¨ dans toutes les sources mŽ diŽ vales. 32 Par la suite : Histoire secr te. 33 Les spŽ cialistes sÕaccordent pour dater le texte de 1240, voir Igor de Rachewiltz, « Some Remarks on the Dating of The Secret History of the Mongols », Monumenta Serica (1965) XXIV : 185-205. Pour les besoins de cet article, nous avons utilisŽ la translittŽration en caract res latins du texte mongol par Igor de Rachewiltz (Index to the Secret History of the Mongols) et la traduction fran aise de Marie-Dominique Even et Rodica Pop (Histoire secr te des Mongols. Chronique mongole du XIIIe si cle), dans laquelle nous avons introduit la translittŽration de certains termes mongols et des noms propres. 34 Sur le syst me vindicatoire, voir infra. Chez les Bouriates de lՎ poque prŽ -soviŽ tique, les enfants devaient  tre en mesure de rŽ citer leur gŽ nŽ alogie aŽ n de se souvenir des atteintes portŽ es ˆ la famille, voir Roberte Hamayon, « MŽ rite de lÕoffensŽ vengeur, plaisir du rival vainqueur. Le mouvement ascendant des Ž changes hostiles dans deux sociŽ tŽ s mongoles », in La Vengeance. ƒtudes dÕethnologie, dÕhistoire et de philosophie, vol. II, Vengeance et pouvoir dans quelques sociŽ tŽ s extra-occidentales, Ž d. Raymond Verdier (Paris : Ed. Cujas, 1980) : 128. 35 Marie-Dominique Even, Introduction ˆ lÕHistoire secr te : 11.

40

DENISE AIGLE

Notre connaissance de lÕHistoire secr te est tributaire des sources chinoises 36. En effet, le texte original ne nous est pas parvenu en Ž criture mongole – le mongol, langue alta•que, utilisait un alphabet dÕorigine ou•ghoure –, mais dans une transcription syllabique en caract res chinois. Chaque syllabe mongole est remplacŽ e, de fa on systŽ matique, par un caract re chinois Ž quivalent, phonŽ tiquement proche 37. Dans lÕHistoire secr te, le terme jasaq est employŽ neuf fois38. Il est glosŽ en chinois par fa-tu, mot qui signiŽ e r gles, r glements, statuts 39 ; toutes les r gles nՎ manent pas de Gengis-khan. Les sources persanes – Les informations sur le y¨s¨ proviennent essentiellement de lÕhistoriographie ofŽ cielle ilkhanide, et en particulier de ®Al¨¾ al-DÂn ®Aߨ¾ Malik alJuvayn (m. 1283). Il Ž tait issu dÕune famille de notables du Khorassan, dont plusieurs membres se mirent au service des Mongols apr s quÕils eurent conquis lÕIran oriental. Al-Juvayn est le seul historien persan ˆ avoir sŽ journŽ en Mongolie : une premi re fois de 1249 ˆ 1251, puis, apr s un bref retour dans ses terres, de 1251 ˆ 1253, pendant le r gne de Mšngke. Al-Juvayn a consacrŽ au y¨s¨ le deuxi me chapitre de son T¨rÂkh-i jah¨ngush¨ intitulŽ : « Des r gles (qaw¨®id) mises en œuvre par Gengis-khan et des y¨s¨ promulguŽ s apr s son accession au pouvoir 40 ». Les qaw¨®id et les y¨s¨ dont parle alJuvayn concernent lÕƒtat : la chasse comme entra”nement ˆ la guerre, lÕorganisation de lÕarmŽ e, le rŽ seau postal pour les communications ofŽ cielles ( yam), la levŽ e des imp™ts dans les territoires conquis. Dans le T¨rÂkh-i jah¨ngush¨, al- Juvayn associe souvent le terme y¨s¨ ˆ un autre mot du m me champ lexical : qaw¨®id (r gles) wa y¨s¨-h¨ ; ahk¨m (dŽ crets) wa y¨s¨-h¨, par exemple. Il emploie la plupart du temps le terme y¨s¨ dans le sens de « r gle, rŽ gulation » ou « punition 41 ». On ne trouve pas, dans ce chapitre, une liste des coutumes et des interdits mentionnŽ s par les missionnaires occidentaux et par les auteurs arabes. En revanche, al-Juvayn dŽ crit la plupart de ces tabous dans le chapitre consacrŽ aux : « Actes et actions du qan […gšdei]42 ». 36 Une partie du texte se trouve dans des chroniques mongoles plus tardives, comme lÕAltan Tobci, mais la version la plus compl te de lÕHistoire secr te se trouve bien dans sa version chinoise. 37 Sur lÕhistoire de la Ž liation du texte en Chine, voir William Hung, « The Transmission of the Book Known as The Secret History of the Mongols », Harvard Journal of Asiatic Studies (1951) 14 : 433-492. 38 HS § 153 : 2 fois ; § 189 ; § 193 ; § 197 ; § 199 ; § 227 ; § 257 ; § 278 ainsi quÕune autre occurrence au § 74 dans le sens de « ceux qui donnent des lois ». 39 Dictionnaire Ricci, n¡ 1494, sous le caract re fa. 40 Al-JuvaynÂ, Ž d., vol. I : 16-25 ; trad., vol. I : 23-34. 41 LÕexpression bi-y¨s¨ ras¨nÂdan est le verbe le plus couramment utilisŽ dans les textes persans mŽ diŽ vaux pour signiŽ er « exŽ cuter, mettre ˆ mort ». Voir Ma®sâme Ma®dankan, Bi-y¨s¨ rasÂdag¨n (Tehran : Iran University Press, 1996) : 1-18. 42 Al-JuvaynÂ, Ž d., vol. I : 158-191 ; trad., vol. I : 201-236.

LE GRAND JASAQ DE GENGIS-KHAN

41

Il prŽ sente ces interdits en les associant, le plus souvent, au terme y¨s¨ et sous la forme dÕanecdotes Ž diŽ antes aŽ n de montrer la mansuŽ tude dÕ…gšdei envers les musulmans qui ne les respectaient pas. Le caract re Ž diŽ ant de ces rŽ cits invite lÕhistorien ˆ se montrer prudent ; cependant, les tabous dŽ crits par al-Juvayn recoupent les autres sources43. Les sources arabes – Les sources arabes sont paradoxalement plus disertes sur le y¨s¨ que les sources persanes. Ibn Fa´l All¨h al-®Umar (m. 1349) appartenait aux services de la chancellerie mamelouke44. Dans son grand ouvrage encyclopŽ dique, le Kit¨b mas¨lik al-abרr wa mam¨lik al-amרr, se trouve la premi re liste organisŽ e des r gles (qaw¨®id) qui auraient Ž tŽ Ž dictŽ es par Gengis-khan et des coutumes (¨d¨b) mongoles 45. Il ne prŽ sente jamais Gengis-khan comme un lŽ gislateur ; il souligne simplement que beaucoup dՎ lŽ ments du y¨s¨ ne sont pas en accord avec la shar®a. Il est le premier historien ˆ mentionner de nombreuses coutumes (¨d¨b) qui ne Ž gurent pas dans la chronique dÕal-JuvaynÂ, mais dans les sources latines. Al-®Umar prŽ tend sÕappuyer sur al-JuvaynÂ, dont il cite le nom, alors quÕaucune source biographique ne lui attribue des connaissances en persan. Selon son propre tŽ moignage, il pourrait donc avoir eu acc s au texte, difŽ cile, dÕal-Juvayn par lÕintermŽ diaire de divers fonctionnaires et marchands en provenance de lÕilkhanat et qui lui auraient transmis le texte oralement46. Taq al-DÂn ®Abd l-®Abb¨s A½mad al-MaqrÂz (m. 1442) reprend dans ses Khiß¨ß 47 la liste Ž tablie par al-®UmarÂ, mais son exposŽ sur le y¨s¨ est polŽ mique48. AŽ n de dŽ montrer le caract re anti-islamique du y¨s¨, al-MaqrÂz alt re le texte dÕal-®Umar en attribuant ˆ Gengis-khan la paternitŽ de ce quÕil consid re comme des « lois ». Il

43 Par exemple, lÕhistorien persan al-Jâzj¨nÂ, qui rŽ sidait ˆ la cour des sultans de Delhi et qui est reprŽ sentatif de lÕhistoriographie persane non dŽ pendante des Mongols, rapporte des anecdotes similaires. Dans son ouvrage, Þabaq¨t-i N¨×irÂ, rŽ digŽ en 1259-1260, il insiste sur la mansuŽ tude du qan …gšdei envers les musulmans : « Des personnes dignes de conŽ ance ont dit que les faveurs et les Ž gards dÕ…gšdei envers les musulmans Ž taient la preuve quÕil Ž tait devenu musulman en secret. » Ž d. : 387 ; trad., vol. II : 1115. On trouve les m mes anecdotes chez RashÂd al-DÂn, historien ofŽ ciel des Ilkhans sous Ghazan-khan. 44 Grande famille de juristes, les Banâ Fa´l All¨h ont occupŽ des postes importants dans lÕadministration mamelouke. 45 Al-®UmarÂ, Ž d. : 7-10. Cette liste fut intŽ gralement reprise au XVe si cle par al-MaqrÂzÂ. 46 Sur les sources orales dÕal-®UmarÂ, voir Klaus Lech, Introduction ˆ lՎ dition du texte : 29-41. Al-Kutub le crŽ dite de connaissances inŽ galŽ es sur lÕhistoire des Mongols, des souverains de lÕInde et des Turcs, voir Faw¨t al-wafay¨t, vol. I : 159. 47 Al-MaqrÂzÂ, Ž d. Bulaq : 357-359 ; Ž d. Beyrouth : 383-385. Silvestre de Sacy a traduit et annotŽ le passage dÕal-MaqrÂz sur le y¨s¨, voir Chrestomathie arabe : 157-190. 48 Sur al-MaqrÂz et le y¨s¨, voir lՎ tude de David Ayalon, « The great Y¨s¨ of Chingiz Kh¨n. A Re-examination », Studia Islamica 33 (1971) : 97-140 (part A, A Basic Data in the Islamic Sources on the y¨s¨ and on its Contents).

42

DENISE AIGLE

utilise pour les introduire des termes du vocabulaire juridique islamique : « Gengiskhan Ž t une loi (shara®a) » ; « il Ž dicta comme prŽ cepte (sharaßa) » ; « il ordonna (alzama) » ; « il interdit (mana®a) ». Les interdits, qui sous la plume dÕal-®Umar nՎ taient que des coutumes mongoles, sont ainsi prŽ sentŽ s comme des lois instaurŽ es par Gengis-khan 49. Le tŽ moignage dÕIbn ®Arabsh¨h (m. 1450) sur les r gles de Gengis-khan et sur les coutumes mongoles 50 apporte des complŽ ments intŽ ressants par rapport au rŽ cit dÕal®UmarÂ. Ibn ®Arabsh¨h avait Ž tŽ emmenŽ en captivitŽ par Tamerlan en 1401, apr s le si ge de Damas. Il sŽ journa dans le monde turco-mongol jusquÕen 1422, date de son retour dans les terres mameloukes. Ibn ®Arabsh¨h produisit, en 1448, un Miroir des princes, le F¨kihat al-khulaf¨¾ wa muf¨kah¨t al-úuraf¨¾, adaptation en arabe dÕune collection de fables animales persanes du XIIIe si cle51. Sans aucun doute, Ibn ®Arabsh¨h, qui connaissait le persan, tire une partie de ses informations dÕal-JuvaynÂ, mais il semble disposer Ž galement de rŽ cits de premi re main, oraux, collectŽ s durant son long sŽ jour en Eurasie. En tout cas, sa prŽ sentation du y¨s¨ vise essentiellement ˆ montrer en quoi les « lois » de Gengis-khan sont en dŽ saccord avec la loi islamique. Les rŽ cits dÕal-MaqrÂz et dÕIbn ®Arabsh¨h peuvent  tre qualiŽ Ž s de sources antimongoles. Ces auteurs prŽ sentent le y¨s¨ en prenant comme mod le de rŽ fŽ rence la shar®a ; ils en font une sorte de lŽ gislation mongole en mettant lÕaccent sur ce qui, aux yeux des musulmans, peut disqualiŽ er les coutumes ainsi que la mani re de gouverner des Mongols. Les sources syriaques et armŽ niennes – Nous disposons dÕune source syriaque de premi re importance, rŽ digŽ e par le plus haut dignitaire de lÕƒglise jacobite de lÕempire ilkhanide, Bar Hebraeus (m. 1286), qui, d s 1268, Ž t de longs sŽ jours ˆ M¨ghar¨, la capitale des Ilkhans en Azerba•djan. Par ses fonctions religieuses, Bar Hebraeus fut amenŽ ˆ frŽ quenter les autoritŽ s politiques. Il eut ainsi acc s ˆ la biblioth que royale des Mongols, dont il semble quÕil connaissait la langue. Il Ž tait bien placŽ pour disposer dÕinformations de premi re main. CÕest en particulier le cas pour les Ž vŽ nements qui eurent lieu en Iran et dans le Bil¨d al-Sh¨m. En 1276, il commen a la rŽ daction dÕune grande chronique, connue sous le nom de Maktebonž th Zabn . Un chapitre de cette chronique est consacrŽ aux « lois Ž dictŽ es par Gengis-khan 52 ». Bar Voir infra. Il nÕutilise pas le terme y¨s¨, mais târ¨ (mo. cl., tšre, r gle). Sur ce terme, voir Doerfer, vol. I, n¡ 134 ; Lessing : 835-836. 51 Ibn ®Arabsh¨h : 227-250. Sur cet auteur, voir Robert G. Irwin, « What the Partidge Told the Eagle : A Neglected Arabic source on Chinggis Khan and the Early History of the Mongols », in The Mongol Empire & its Legacy, Ž d. Reuven Amitai-Press et David Morgan (Leyde : Brill, 1999) : 5-11. 52 Bar Hebraeus, Ž d. : 411-412 ; trad. : 354-355. 49 50

LE GRAND JASAQ DE GENGIS-KHAN

43

Hebraeus mentionne neuf lois (n‰m‰s , sing. n‰m‰s‰53) ; elles concernent la rŽ glementation de lÕarmŽ e, la mani re de se comporter avec lÕennemi et certaines r gles sociales ; une infraction ˆ une seule de ces lois, en rapport avec la discipline militaire, est passible de la peine de mort. David Ayalon a suggŽ rŽ que Bar Hebraeus avait tirŽ ses informations sur le y¨s¨ dÕal-Juvayn 54. Il est vrai que les tŽ moignages de ces deux auteurs sont tr s proches, mais lÕon trouve chez Bar Hebraeus des dŽ tails supplŽ mentaires et la vision quÕun auteur chrŽ tien pouvait avoir de ces r glementations. LÕhistorien armŽ nien Kirakos Ganjakec®i (m. 1272) rŽ digea une histoire de lÕArmŽ nie, Patmut®iwnk® Hayoc®, vers 1241, ˆ partir de ses impressions personnelles. Il avait Ž tŽ capturŽ par les Mongols et obligŽ de les servir en tant que secrŽ taire. John A. Boyle a traduit la partie de ce texte la plus intŽ ressante pour notre propos, car lÕauteur y fait une « description de lÕapparence et des coutumes des Tatars » ; il cite nombre dÕinterdits et de croyances mongoles 55. Il est question des r gles du y¨s¨ dans une source armŽ nienne plus tardive, intitulŽ e T®at®arac® Patmut®iwnk® (Histoire des Tatars), de Grigor Akanc®i (m. 1335) 56. Ce dernier nÕest pas, comme Kirakos Ganjakec®i, un tŽ moin oculaire des Ž vŽ nements quÕil dŽ crit ; il sÕappuie vraisemblablement sur des sources orales57. Dans le deuxi me chapitre : « De la reviviŽ cation de la foi [des Mongols], de leurs prŽ ceptes et de leur chef », il dresse une liste de six r gles qui, Ž crit-il, « leur ont Ž tŽ imposŽ es par Dieu et quÕils nomment eux-m mes yasax 58 » ; la transgression de lÕune de ces r gles Ž tait suivie dÕune mise ˆ mort sans pitiŽ . Les sources latines – Les rŽ cits des envoyŽ s des papes et des souverains occidentaux aupr s des qan sont tr s prŽ cieux pour conna”tre le syst me de reprŽ sentations de la sociŽ tŽ mongole mŽ diŽ vale ; elles Ž clairent utilement les donnŽ es des auteurs arabes, qui prŽ sentent, hors contexte, de s ches listes de r gles et dÕinterdits. Franciscain italien, Jean de Plan Carpin fut envoyŽ par le pape Innocent IV aupr s de Batu sur la Volga et du grand qan GŸyŸk ˆ Qaraqorum. Il sŽ journa dans lÕempire mongol de 1245 ˆ 1247. Son rŽ cit est la premi re source occidentale sur les Mongols 59 : toutes les connaissances quÕil rassemble sur les mœurs, les croyances, le 53 Sur le terme n‰m‰s‰, voir lÕimportante notice dans le Thesaurus Syriacus : 2383-2385. Le Thesaurus Syriacus cite comme Ž quivalents arabes au terme : far´a, shar®a, sunna. 54 Voir David Ayalon, « The great y¨s¨ », Part A : 127. 55 John Andrew Boyle sÕest appuyŽ pour traduire cette partie de la chronique de Kirakos Ganjakec®i sur lՎ dition du texte rŽ alisŽ e en 1961 par K.A. Melik®-Ohanjanyan. 56 Grigor Akanc®i, Ž d. et trad. Robert P. Blake et Richard N. Frye, Harvard Journal of Asiatic Studies (1949) 12/3-4 : 269-443. 57 Voir la discussion sur les sources utilisŽes par Akanc®i par Blake et Frye, ibid. : 274-276. 58 Akanc®i : 291. 59 Iohannes de Plano Carpini, Ystoria Mongalorum, in Sinica Franciscana, vol. I, Ž d. P. Anastasius Van den Wyngaert, 1929 : 27-143. Traduction fran aise par Dom Jean Becquet et Louis Hambis (Paris : Adrien Maisonneuve, 1965).

44

DENISE AIGLE

gouvernement et lÕart militaire sont destinŽ es ˆ livrer des informations pratiques. Jean de Plan Carpin parle des lois et des ordonnances (leges et statuta) promulguŽ es par Gengis-khan, quÕil distingue des coutumes et des pratiques ancestrales (traditiones) des Mongols 60. Guillaume de Rubrouck 61, lui aussi franciscain, fut envoyŽ en 1253 par Louis IX aupr s de Sartaq, le qan des Mongols de Russie mŽ ridionale, dont les rumeurs sur sa conversion au christianisme avaient atteint les armŽ es croisŽ es de Palestine ; il se rendit ensuite aupr s de Mšngke ˆ Qaraqorum, puis de Batu sur la Volga62. La motivation premi re de Guillaume de Rubrouck Ž tait dÕordre pastoral 63. Il fait alterner des sc nes descriptives et vŽ cues, des informations historiques et gŽ ographiques et ses propres observations. Ses matŽ riaux sur les coutumes des Mongols et sur le chamanisme sont dÕune valeur inestimable pour une aire gŽ ographique tr s peu documentŽ e par les sources Ž crites de lՎ poque 64. La tradition Ž rudite sur le y¨s¨ Au dŽ but du XVIIIe si cle, PŽ tis de la Croix a inaugurŽ s, dans son Histoire du grand Genghizcan 65 les recherches sur « le code de loi mongol ». Il a reconstituŽ le y¨s¨ ˆ partir de fragments relevŽ s dans diffŽ rentes sources, souvent tardives66. Ce code aurait renfermŽ les prescriptions de Gengis-khan en mati re de gestion de lÕƒtat, de r gles militaires, dՎ changes diplomatiques, de justice et de rapports entre tribus ; le y¨s¨ aurait Ž tŽ promulguŽ au quriltai 67 de 1206. La faiblesse de lՎ tude de PŽ tis de la Croix Ž tait due ˆ son manque de recul critique par rapport aux textes. NŽ anmoins, sa reconstruction du y¨s¨ a marquŽ les Ž tudes postŽ rieures68, et en particulier Valentin

Carpin, III, 7, Ž d. : 40. Village des Flandres fran aises. 62 Guillemus de Rubruc, Itinerarium, in Sinica Franciscana, vol. I, Ž d. P. Anastasius Van den Wyngaert, 1929 : 164-332. Traduction fran aise par Claude et RenŽ Kappler (Paris : Payot, 1985) ; traduction anglaise par Peter Jackson et David Morgan (Londres : Hakluyt Society, 1990). 63 Dans une large mesure, Guillaume de Rubrouck aurait fait ce voyage pour apporter son soutien ˆ un groupe de Germains qui avaient Ž tŽ capturŽ s pendant lÕinvasion de lÕEurope orientale, en 1241-1242, et qui Ž taient employŽ s dans les mines de Tal¨s. Sur les motivations de Rubrouck, voir Peter Jackson, « William of Rubruck : A Review Article », Journal of the Royal Asiatic Society (1987) : 92-97. 64 Rubrouck, chapitres XXV, 9-10 et XXXV. 65 Histoire du grand Genghizcan, par feu PŽ tis de la Croix, le p re, (publiŽ par PŽ tis de la Croix, le Ž ls), Paris, Vve Jombert, 1710 ; traduction anglaise en 1722. 66 Il utilise un grand nombre de sources en arabe, en persan, en turc et en latin, mais il les met toutes sur le m me plan et ne tient pas compte de leur chronologie. Le passage sur le y¨s¨ se trouve aux pages 99-110. 67 Le quriltai est une assemblŽ e de tous les chefs mongols. 68 Voir, en particulier, G. Vernadsky, « JuwainiÕs Version of Chingis KhanÕs yasa », Annales de lÕInstitut Kondakov (1940) XI : 33-45 ; A.N. Poliak, « The in uence of Chingiz-Kh¨nÕs Y¨sa upon 60 61

LE GRAND JASAQ DE GENGIS-KHAN

45

A. Riasanovsky qui, dans un ouvrage publiŽ en 1937 et aujourdÕhui dŽ passŽ , prŽ sentait ce quÕil appellait « des fragments du grand yassa ». Il en rŽ pertoriait trente-six, dont la plupart Ž taient tirŽ s des Khiß¨ß dÕal-MaqrÂzÂ69. La vision du y¨s¨ ne sՎ carta de cette ligne quÕen 1971, avec lÕimportante Ž tude de David Ayalon 70. Il constatait que les chercheurs nÕavaient jamais tentŽ de retrouver la source dÕal-MaqrÂzÂ, ni les raisons qui lÕavaient incitŽ ˆ Ž crire sur le y¨s¨. David Ayalon rŽ futait la prŽ tention dÕal-MaqrÂz dÕavoir eu connaissance du contenu du y¨s¨ par lÕintermŽ diaire dÕun certain Ibn Burh¨n qui aurait « vu » une copie du y¨s¨ ˆ Bagdad. Il sugg rait, puisque la source dÕal- MaqrÂz Ž tait manifestement al-®Umar – quÕil copiait sans mentionner son nom –, que toutes les sources arabes sur le y¨s¨ dŽ rivaient, directement ou non, dÕal-Juvayn 71. LÕapport majeur de David Ayalon a Ž tŽ dÕavoir montrŽ que lÕobjectif dÕal-MaqrÂz Ž tait de dŽ montrer que les sultans du Caire avaient incorporŽ des pratiques mongoles dans les instances administratives et judiciaires mameloukes72. Dans un article iconoclaste publiŽ en 1986, David Morgan a remis en question une grande partie de ce qui avait Ž tŽ Ž crit sur le y¨s¨73. Il constatait que, dans les sources persanes, les plus anciennes les mentions du y¨s¨ Ž taient rares et imprŽ cises ; le concept semblait plut™t avoir existŽ dans lÕesprit des historiens ˆ partir du milieu du XIVe si cle74. RŽ cemment, Robert Irwin, ˆ partir de lÕanalyse du F¨kihat al-khulaf¨¾ dÕIbn ®Arabsh¨h, a repris la th se dÕun code de loi Ž crit. Puisque les Mongols sՎ taient dotŽ s dÕune Ž criture du vivant de Gengis-khan, il supposait que les lois du fondateur de lÕempire avaient Ž tŽ mises en prioritŽ par Ž crit75. Mais nous verrons que lÕobjectif essentiel dÕIbn ®Arabsh¨h Ž tait de souligner les diffŽ rences entre les pratiques judiciaires des Mongols et la justice islamique rendue par le cadi. Les recherches des spŽ cialistes de lÕhistoire des YŸan, la dynastie gengiskhanide qui rŽ gna en Chine (1272-1368), apportent des informations complŽ mentaires sur le y¨s¨76 ; les sources chinoises, bien quÕelles aient Ž tŽ rŽ digŽ es apr s la chute des the General Organization of the Mamlâk State », Bulletin of the School of Oriental and African Studies (1942) 10/4 : 862-876 ; Mansura Haider, « The Mongolian Traditions and their Survival in Central Asia (XIV-XV Centuries) », Central Asiatic Journal (1984) 28/1-2 : 57-79. 69 Valentin A. Riasanovsky, Fondamental Principles of Mongol Law (Bloomington : Indiana University Publications : Uralic and altaic Series, vol. 43, 1937) : 83-86. Voir la critique de cet ouvrage par David O. Morgan, « The Great Y¨s¨ » : 165-166. 70 David Ayalon, « The Great Y¨sa of Chingiz Kh¨n, A Re-examination », Studia Islamica (1971) 33 : 97-140 (part A, The Basic Data in the Islamic Sources on the Y¨sa and on its Contents). 71 David Ayalon, « The Great Y¨sa » : 101-104. 72 Voir lÕargumentation dÕal-MaqrÂz dans David Ayalon, « The Great Y¨sa of Chingiz Kh¨n, A Re-examination », Studia Islamica (1973) 38 : 107-156 (part C2, al-MaqrÂzÂÕs Passage on the Y¨sa under the Mamlâks). 73 David O. Morgan, « The Great Y¨s¨ » : 163-176. 74 David O. Morgan, « The Great Y¨s¨ » : 172-173. 75 Robert G. Irwin, « What the Partidge Told the Eagle » : 5-11. 76 Voir une bonne prŽ sentation de ces Ž tudes par Igor de Rachewiltz, « Some Re ections on Cinggis QanÕs jasag », East Asian History (1993) : 91-104.

46

DENISE AIGLE

YŸan, sont en effet plus prŽ cises que lÕhistoriographie islamique. Les importants travaux de Paul Ratchnevsky ont abouti ˆ la conclusion que le y¨s¨ nՎ tait ni un code de loi rŽ digŽ ˆ une pŽ riode particuli re, ni un document homog ne et systŽ matique, mais un ensemble de dŽ crets promulguŽ s par Gengis-khan, aŽ n de rŽ pondre aux besoins crŽ Ž s par des circonstances particuli res77. Par la suite, sous lÕin uence de lÕislam et du bouddhisme, lÕimportance du y¨s¨ diminua progressivement dans la mesure o lÕempire entrait en symbiose avec les cultures sŽ dentaires. Alors que la plupart des chercheurs considŽ raient que Gengis-khan avait codiŽ Ž la loi coutumi re mongole, Paul Ratchnevsky contestait ce point en arguant quÕune loi coutumi re nՎ tait jamais consignŽ e par Ž crit, mais que, en revanche, il aurait Ž tŽ nŽ cessaire dÕenregistrer les nouvelles lois de Gengis-khan 78. Cet argument est contestable, car la formation de lÕempire et la sujŽ tion de peuples non mongols auraient pu conduire ˆ mettre par Ž crit un certain nombre de coutumes en vue de les imposer aux populations soumises. Paul Heng-chao ChÕen a discutŽ le probl me du y¨s¨ (chinois, ta-cha-sa, « grand jasaq ») dans le cadre du syst me lŽ gislatif des YŸan79. Selon le YŸan-shih, la chronique historique ofŽ cielle de la dynastie des YŸan80, au moment de son intronisation au quriltai de septembre 1229, …gšdei Ž t promulguer le « yeke jasaq », glosŽ en chinois « la grande [loi] (ta-fa-ling)81 ». Paul Heng-chao ChÕen a dŽ montrŽ que le y¨s¨ nՎ tait pas un traitŽ juridique, organisŽ de mani re systŽ matique ; il ne sÕappliquait pas ˆ lÕensemble des peuples qui se trouvaient sous la domination des Mongols, mais, par la suite, il constitua, en Chine, la principale source en mati re juridique 82. Le dŽ bat, comme on peut le constater, sÕest focalisŽ sur lÕexistence ou non dÕun « code de lois Ž crit » dans lÕempire gengiskhanide. Ces Ž tudes nÕont ni examinŽ de mani re approfondie ce que reprŽ sentait effectivement le y¨s¨ pour les souverains mongols et comment il fut compris par les auteurs mŽ diŽ vaux, ni tentŽ de dŽ terminer quelles Ž taient les populations soumises au y¨s¨, deux questions en rapport direct. 77 Paul Ratchnevsky, « Die Yasa (Jasaq) Cinggis-khans und ihre Problematik », Schriften zur Geschichte und Kultur des alten Orients 5 (1974), Sprache, Geschichte und Kultur der altaischen Všlker, Berlin, Akademie-Verlag ; Cinggis-khan sein Leben und Wirken (Wiesbaden : Franz Steiner Verlag, 1983) : 164-172 ; trad. anglaise par Thomas N. Haining, Gengis-khan. His Life and Legacy (Oxford & Cambridge Mass. : Blackwell, 1991) : 187-196. 78 Paul Ratchnevsky, Cinggis-khan : 165-166 ; trad. : 188-189. 79 Paul Heng-chaÕo ChÕen, Chinese legal tradition under the Mongols. The code of 1291 as reconstructed (Princeton : Princeton University Press, 1979). 80 Le YŸan-shih, selon la tradition historiographique en Chine, fut compilŽ par les savants de la dynastie suivante, les Ming (1368-1644), voir Francis W. Cleaves, « The Memorial for Presenting the Yuan shih », Asia Major (1988) 1 : 59-69. 81 En chinois, fa-ling signiŽ e : lois, dŽ crets, voir Dictionnaire Ricci, n¡ 1494, sous le caract re fa. 82 Paul Heng-chaÕo ChÕen consid re la promulgation du jasaq par …gšdei comme le premier stade (1229-1270) du dŽ veloppement des codes chinois, Chinese legal tradition under the Mongols : 4-10.

LE GRAND JASAQ DE GENGIS-KHAN

47

Le double sens du y¨s¨ dans les sources islamiques Il est nŽ cessaire de partir de lÕHistoire secr te pour prŽ ciser la distinction qui existe en mongol entre les dŽ crets impŽ riaux, le jasaq, et la coutume, le yosun 83. Dans lÕHistoire secr te, le terme jasaq est invariablement employŽ dans le sens de loi dÕun souverain dans lÕexercice de son autoritŽ : « AcŽ rŽ e fut la loi de GŸrbesŸ notre reine84 », de prŽ cŽ dent lŽ gal, de r gle dont la transgression expose ˆ une peine sŽ v re. En mongol, le mot « jasaqla- » signiŽ e agir selon la loi, administrer, gouverner 85 ; jasaq est donc clairement un terme en rapport avec les r gles qui permettent dÕadministrer lÕƒtat 86. Le terme yosun, quant ˆ lui, appara”t vingt-deux fois dans lÕHistoire secr te87 pour exprimer une mani re de faire ou un usage : « Telle fut la mani re dont YesŸgei emmena Hš¾elŸn dans sa tente88. » Yosun est Ž galement utilisŽ pour dŽ crire la mani re de fonder une relation dÕalliŽ s jurŽ s (anda). Il est associŽ deux fois ˆ r gle (tšre) ; au paragraphe 216 dans le sens dÕusage : « Dans les r gles (tšre) existe la coutume ( yosun) de devenir noyan 89, une position de seigneur », et au paragraphe 263 au moment o Gengis-khan nomme des gouverneurs en Chine du Nord et en Transoxiane : « parce quÕils Ž taient bons connaisseurs des r gles (tšre) et des principes ( yosun) des villes ». Dans lÕHistoire secr te, yosun renvoie toujours ˆ un usage, une pratique, une mani re traditionnelle de procŽ der. La source de validitŽ du jasaq et celle du yosun ne sont donc pas les m mes. Dans le premier cas, il sÕagit dÕun pouvoir, celui du chef qui Ž dicte la loi, dans le second, la lŽ gitimitŽ Ž mane de la tradition. Enfreindre le jasaq comme le yosun peut conduire ˆ une punition, temporelle dans le cas du jasaq, ˆ laquelle on peut tenter dՎ chapper, surnaturelle, dans le cas du yosun, et lÕon Ž vite dÕy contrevenir. La terminologie islamique manque de prŽ cision par rapport ˆ la langue mongole. Al-Juvayn utilise un large champ lexical pour exposer la mani re dont Gengis-khan a lŽ gifŽ rŽ . Il Ž crit : « En accord avec sa propre opinion (r¨y), [Gengis-khan] a Ž tabli une loi (q¨nânÂ) pour chaque occasion (k¨r), une r gle (dastâr) pour chaque circonstance [. . .], une peine lŽ gale (½addÂ) pour chaque crime ( gun¨hÂ) [. . .], ces y¨s¨ et ces a½k¨m (dŽ crets) Ž taient enregistrŽ s sur des rouleaux quÕon appelait y¨s¨n¨ma-yi buzârg 90 [Grand Livre du y¨s¨]. » Dans lÕesprit dÕal-JuvaynÂ, le y¨s¨ concerne les affaires de lÕƒtat (q¨nân), mais il lÕassocie Ž galement ˆ une loi ˆ 83

Sur ce terme, voir Doerfer, vol. I, n¡ 408. HS § 189. GŸrbesŸ Ž tait la m re de Tayan, le roi des Naiman. 85 Sur ce terme, voir Lessing, « zasagla- » : 1040. 86 En mongol moderne, zasag dŽ signe le gouvernement. 87 HS § 9 ; 56 ; 96 ; 110 ; 116 ; 117 ; 139 ; 147 ; 150 ; 164 ; 177 ; 180 ; 216 ; 241 ; 244 ; 263 ; 270 ; 272. 88 YesŸgei Ž tait le p re de Gengis-khan ; HS § 56. 89 Le terme noyan (plur. noyad) dŽ signe ˆ la pŽ riode impŽ riale les commandants de lÕarmŽ e. 90 Al-JuvaynÂ, Ž d., vol. I : 17 ; trad., vol. I : 25. 84

48

DENISE AIGLE

connotation religieuse, puisquÕune infraction est sanctionnŽ e par une peine lŽ gale, ½add, terme empruntŽ au vocabulaire du droit musulman. Al-Juvayn utilise le terme y¨s¨ pour dŽ signer les coutumes mongoles et il lui associe souvent un autre mot, par exemple ½ukm (dŽ cret), ¨dhÂn ou yâsân (coutume) : « Selon le y¨s¨ et le yâsân des Mongols, la place du p re passe au plus jeune Ž ls par sa femme principale 91. » Mais il lui arrive de mentionner un interdit coutumier sans utiliser le terme y¨s¨. AlJâzj¨n semble donner une description du y¨s¨ plus en rapport avec le sens prŽ cis du terme en mongol : « Ces prŽ ceptes (a½k¨m), auxquels ils [les Mongols] ont donnŽ le nom de yasa dans la langue mongole, signiŽ ent dŽ cret et ordre (½ukm wa farm¨n)92. » Mais al- Jâzj¨n mentionne, lui aussi, des coutumes qui, non respectŽ es, sont passibles de la peine capitale 93. Dans les sources arabes, on retrouve la m me ambigu•tŽ dans lÕemploi du terme y¨s¨ puisquÕil est utilisŽ pour dŽ signer les dŽ crets impŽ riaux et les coutumes mongoles. Al-®Umar attribue ˆ Gengis-kahn les y¨s¨ dont il dresse la liste : « Gengiskahn a Ž tabli une r gle (waqa®a q¨®ida) pour toutes les choses importantes et une punition (®uq¨ba) pour tous les crimes (li-kulli dhanbi), [. . .] ces r gles furent enregistrŽ es dans un livre appelŽ al-y¨s¨ al-kabÂra et il a ordonnŽ que [le livre] soit dŽ posŽ dans son trŽ sor94. » Il dresse une liste des dŽ lits passibles de la peine capitale et une liste des coutume (¨d¨b), mais, dans la liste des dŽ lits, il inclut des coutumes, tels les interdits sur lÕeau et lÕabattage des animaux. Quant ˆ al-MaqrÂz et ˆ Ibn ®Arabsh¨h, ils regroupent sous le m me terme les dŽ crets impŽ riaux et les coutumes. Les auteurs mamelouks consid rent sans aucun doute le y¨s¨ mongol comme un code de loi ayant une fonction identique ˆ la shar®a, mais en dŽ saccord avec elle. LÕanalyse du y¨s¨, pris au sens large des sources islamiques, nous informe, comme nous le verrons, sur la mani re dont Gengis-khan et ses successeurs ont mis en place les structures nŽ cessaires pour faire fonctionner lÕƒtat mongol ; cette analyse permet Ž galement, en rattachant les interdits du y¨s¨ au syst me de reprŽ sentations du chamanisme, dÕexpliciter les raisons de lÕincomprŽ hension de certains tabous par les musulmans 95. Il est nŽ cessaire de dŽ crire au prŽ alable le milieu socio-culturel dans lequel vivaient les tribus mongoles au moment o sÕinstaura dans la steppe le nouvel ordre gengiskhanide.

91

Al-JuvaynÂ, Ž d., vol. III : 3 ; trad., vol. II : 549. Al-Jâzj¨nÂ, Ž d. : 381 ; trad., vol. II : 1108. 93 Il fait rŽ fŽ rence ˆ la condamnation, par Gengis-khan, de lÕadult re (zin¨), du vol (duzdÂ), du mensonge (durâgh), de la trahison (khiy¨nat kardan), dŽ lits mentionnŽ s dans la plupart des sources, et ˆ deux coutumes : prendre un morceau dans la bouche dÕun autre et entrer dans une pi ce dÕeau, quÕelle soit grande ou petite, Ž d. : 381 ; trad., vol. II : 1107. 94 Al-®UmarÂ, Ž d. : 8. 95 Voir en annexe le tableau rŽ capitulatif des principaux y¨s¨ et yâsân, sanctionnŽs par la peine capitale, et qui sont mentionnŽ s dans les sources utilisŽes pour cette Ž tude. 92

LE GRAND JASAQ DE GENGIS-KHAN

49

LE Y¤S¤ ET LÕEMPIRE La culture mongole et lÕempire gengiskhanide Milieu Ž cologique, syst me de reprŽ sentations et organisation sociale – Les Mongols pratiquaient le chamanisme, un syst me symbolique englobant la vie matŽ rielle, lÕorganisation sociale et les reprŽ sentations religieuses de la sociŽ tŽ . Au XIIe si cle, les tribus Ž taient rŽ parties en deux groupes, selon leur mode de vie et le caract re de leur Ž conomie : les chasseurs « forestiers » et les pasteurs de la steppe96. Le chamanisme de chasse conditionnait le genre de vie des petits groupes qui parcouraient la for t. Dans lÕHistoire secr te, le territoire dÕorigine des anc tres de Gengis-khan, situŽ ˆ la source de lÕOnon, est riche « en zibelines, en Ž cureuils et en gros gibier97 ». Sous cette forme ancienne, le chamanisme reposait sur le principe dÕun Ž change entre deux mondes : les hommes se nourrissaient de gibier, tandis que les esprits animaux sauvages consommaient la force vitale des humains. LՎ change entre ces deux mondes Ž tait con u dans le cadre dÕune « alliance » avec les esprits animaux, sur le mod le de lՎ change matrimonial. PŽ riodiquement, au nom de la communautŽ , le chamane rŽ actualisait cette alliance par le rituel qui reprŽ sentait son mariage avec la Ž lle de lÕesprit de la for t, pourvoyeur de gibier 98. LÕimportance Ž conomique de lՎ levage, qui ne cessa de sÕaccro”tre avec lÕempire, Ž t Ž voluer le mode de vie et donc le syst me de reprŽ sentations des Mongols 99. Dans le chamanisme dՎ levage, en effet, ce qui importait, cՎ tait la Ž liation : les anc tres Ž taient au premier plan et les vivants se trouvaient sous leur dŽ pendance. Les morts devinrent alors « lՎ lŽ ment actif de la Surnature100 ». La fonction du chamane se transforma, elle aussi, car la lŽ gitimitŽ de sa charge venait alors des anc tres. Il assumait, lors des rituels claniques, le r™le de mŽ diateur entre les esprits des morts et le clan ; il assurait la pŽ rennitŽ du groupe et en rŽ parait les dŽ sordres101. Boris Vladimirtsov, Le rŽ gime social des Mongols : 39. HS § 9. 98 Roberte Hamayon, « Pragmatisme et ritualisation dans le chamanisme », in Essais sur le rituel, II, dir. Anne-Marie Blondeau et Kristofer Shipper (Peters : Louvain-Paris, 1988) : 156. En dressant une synth se des syst mes de reprŽ sentation de lÕensemble turco-mongol, ˆ travers les si cles et comme un tout homog ne, Devin DeWeese prŽ sente une vue diffŽ rente du chamanisme. Chez ces populations, la vision religieuse du monde serait ainsi marquŽ e par des structures spirituelles et cosmologiques complexes, en relation avec des expŽ riences spirituelles individuelles orientŽ es vers des pratiques extatiques, voir Islamization and Native Religion in the Golden Horde (The Pennsylvania State University Press : University Park, 1994) : 32-50. 99 Roberte Hamayon, « MŽ rite de lÕoffensŽ vengeur » : 113 ; voir Ž galement Boris Vladimirtsov, Le rŽ gime social des Mongols : 39-44. 100 Roberte Hamayon (La chasse ˆ lÕ‰me : 613) emploie le terme « Surnature » pour dŽ signer lÕensemble des instances animant les  tres naturels ou, plus gŽ nŽ ralement, les instances symboliques associŽ es ˆ la nature. 101 Le caract re oral de la pratique chamanique, qui prend pour base le rapport personnel que 96 97

50

DENISE AIGLE

Les tribus Ž taient construites selon une variŽ tŽ de clans et de lignages qui remontaient, ˆ travers un syst me patrilinŽ aire, ˆ un fondateur Ž ponyme 102. Ces clans pratiquaient une stricte exogamie ; le mariage Ž tait interdit entre descendants dÕun m me « os (yasun) » ou lignŽ e103. AŽ n de respecter la r gle exogamique, les Mongols devaient pouvoir citer les noms de leurs anc tres sur neuf gŽ nŽ rations104. La sociŽ tŽ mongole Ž tait, par ailleurs, fondŽ e sur la segmentaritŽ : les lignages patrilinŽ aires (oboq) et les clans formaient des unitŽ s structurellement Ž quivalentes, tandis quÕune hiŽ rarchie, fondŽ e sur la sŽ nioritŽ , ordonnait les rapports entre consanguins au sein de ces groupes. Deux instances se rŽ partissaient ainsi le pouvoir : les a”nŽ s du clan, qui en assuraient la gestion, et le chef, sur lequel reposaient les capacitŽ s du groupe en tant que « communautŽ de chasse et de guerre105 ». Les Mongols, comme les autres nomades de la steppe, inscrivaient leurs rapports avec lÕextŽ rieur dans le cadre dÕun syst me vindicatoire, basŽ sur la vengeance et lՎ change matrimonial. LÕHistoire secr te « se lit comme un chapelet dÕactes de vengeance » mais, dans ce syst me ritualisŽ , se combinaient lՎ change, la rŽ gulation et la dynamique sociale106. Le nouvel ordre gengiskhanide – LÕuniŽ cation de la steppe par Gengis-khan marqua donc une rupture par rapport ˆ la situation antŽ rieure107. En quelques dŽ cennies, « tous les peuples aux tentes de feutre », de la Mandchourie ˆ la Hongrie, Ž taient devenus, de grŽ ou de force, membres dÕun vaste empire dont la classe dirigeante Ž tait nomade. Les expŽ riences de TemŸjin enfant dŽ termin rent de mani re dŽ cisive sa politique envers les tribus108 : il dŽ tourna ˆ son proŽ t le syst me vindicatoire en vigueur. le chamane est censŽ Ž tablir avec les esprits, sÕoppose a priori ˆ toute Ž xation Ž crite. CÕest ce qui explique quÕon trouve assez peu de mentions des rituels et des pratiques dans les sources mŽ diŽ vales.Sur les pratiques chamaniques chez les Turcs, voir Peter B. Golden, « Religion among the Qipcaqs of Medieval Eurasia », Central Asiatic Journal (1998) 42/2 : 180-237. 102 Boris Vladimirtsov, Le rŽ gime social des Mongols : 56. Sur la sociŽ tŽ mongole, voir Sechin Jagchid et Paul Heyer, MongoliaÕs Society and Culture (Boulder and Folkestone, 1979) ; Joseph F. Fletcher, « The Mongols : Ecological and Social Perspectives », Harvard Journal of Asiatic Studies (1986) 46/1 : 1-35. 103 LÕos symbolise les descendants dÕun anc tre paternel commun dÕo le sens de lignage, clan, ethnie. Il existe une hiŽ rarchie sociale entre les « os blancs », lignages des chefs, et les « os noirs », lignages du bas peuple. 104 Marie-Dominique Even, Histoire secr te, note 59 : 257. 105 Roberte Hamayon, « MŽ rite de lÕoffensŽ vengeur » : 112. 106 Roberte Hamayon, « MŽ rite de lÕoffensŽ vengeur » : 107-108. 107 Voir Thomas J. BarŽ eld, The Perilous Frontier. Nomadic Empires and China, 221 B.C. to A.D. 1757 (Cambridge Mass. & Oxford UK : Blackwell, 1989) ; Thomas T. Allsen, « The Rise of the Mongolian Empire and Mongolian Rule in North China », in The Cambridge History of China, vol. 6, Alien regimes and border states, 907-1368, Ž d. Herbert Franke et Denis Twitchett (Cambridge : Cambridge University Press, 1994) : 321-413; Isenbike Togan, Flexibility and Limitation in Steppe Formations : the Kerait Khanat and Chinggis Khan (Leyde : Brill, 1998). 108 YesŸgei, le p re de Gengis-khan, fut empoisonnŽ par les Tatars ; apr s sa mort, la famille fut exclue du sacriŽ ce clanique et abandonnŽe.

LE GRAND JASAQ DE GENGIS-KHAN

51

Isenbike Togan qualiŽ e de « dŽ tribalisation » la politique mise en œuvre par Gengiskhan. Cette dŽ tribalisation se manifesta dÕabord par une adhŽ sion volontaire et individuelle au futur qan109. D s 1187, Gengis-khan Ž tait entourŽ dÕune suite considŽ rable de « compagnons (nškšd, sing. nškšr110) » qui avaient Ž tŽ attirŽ s vers lui par sa force de caract re et son habiletŽ . La qualitŽ de nškšr reposait sur la loyautŽ et le talent. Le nouvel ordre crŽ Ž par Gengis-khan Ž tait un syst me artiŽ ciel dans lequel les anciennes loyautŽ s tribales, incertaines, Ž taient supplantŽ es par la loyautŽ des individus envers le chef, contr™lable. LÕingŽ niositŽ du qan fut dÕavoir su maintenir la hiŽ rarchie entre les a”nŽ s et les cadets, le principe de sŽ nioritŽ qui, traditionnellement, conditionnait lÕavancement dans le syst me lignager, et de lÕavoir fusionnŽ avec un principe de relations individuelles, qui constituait la base du rapport entre les nškšd et le chef, mais dans un cadre hiŽ rarchique. Cette politique visait ˆ crŽ er une sociŽ tŽ dans laquelle les individus, quelle que soit leur origine tribale, ethnique ou religieuse, pouvaient œuvrer vers des buts communs. Du syst me tribal ˆ lÕempire. RŽ glementations politico-militaires Un juge supr me pour un « ƒtat » patrimonial – La Ž gure de Gengis-khan lŽ gislateur est prŽ sente dans le rŽ cit de Jean de Plan Carpin : « Il [Gengis-khan] revint dans son pays et y Ž t toutes sortes de lois et dÕordonnances (leges et statuta) que les Tartares observent exactement111. » Gengis-khan Ž tait en effet conscient quÕil fallait, pour assurer la stabilitŽ de cet empire naissant, crŽ er une autoritŽ centrale face aux anciennes structures claniques et retirer aux instances traditionnelles lÕautoritŽ dont elles Ž taient investies ; il pouvait chercher ainsi ˆ Ž carter « toute forme de solidaritŽ qui ne serait pas au service du pouvoir central112 ». Au quriltai de 1206, au moment o Gengis-khan fonda le « grand ƒtat mongol » (yeke monggol ulus 113), il le dota dÕune instance judiciaire supŽ rieure. Sigi-Quduqu 114, dŽ signŽ comme juge supr me, fut chargŽ dÕexercer un contr™le sur lÕensemble des populations de lÕempire : « Sois pour moi des yeux qui voient, des oreilles qui Ž coutent [. . .]. Que nul nÕaille ˆ lÕencontre

Isenbike Togan, Flexibility : 131. Le terme nškšr signiŽ e compagnon dÕarmes, fŽ al ; il dŽ signait aux XIe et XIIe si cles celui qui dŽ cidait de quitter sa famille pour se mettre au service dÕun autre chef, Marie-Dominique Even, Histoire secr te, note 31 : 266. 111 Carpin, V, 18, Ž d. : 64 : « Et inde in terram propriam est reversus et ibidem leges et statua mutiplicia fecit, que Tartari inviolabiliter observant ». 112 Roberte Hamayon, « MŽ rite de lÕoffensŽ vengeur » : 121. 113 Le terme ulus dŽ signe ici des peuples ; le sens « ƒtat dotŽ dÕun territoire dŽ limitŽ » est postŽ rieur. 114 Sigi-Quduqu, un enfant trouvŽ dans un camp de Tatars, avait Ž tŽ adoptŽ et Ž levŽ comme un fr re de Gengis-khan. Sur son r™le, voir Paul Ratchnevsky, « Sigi-Qutuqu, ein mongolische Gefolgsmann im 12.-13. Jahrhundert », Central Asiatic Journal (1965) X/2 : 87-120. 109 110

52

DENISE AIGLE

de tes dŽ cisions ! ordonna-t-il. Dans toute la nation punis le vol, fais taire le mensonge ; mets ˆ mort ceux qui doivent  tre mis ˆ mort, ch‰tie ceux qui doivent  tre ch‰tiŽ s115. » La fonction de Sigi-Quduqu, chargŽ de juger les crimes commis dans lÕempire, reprŽ sente le prototype de la machine judiciaire mongole, le y¨rghâ. Dans les sources persanes, le terme y¨rghâ est utilisŽ pour dŽ signer la cour dÕinvestigation chargŽ e – souvent avec usage de la torture – dÕenqu ter sur le cas des ministres dŽ chus, des fonctionnaires corrompus et autres « ennemis » de lÕƒtat116. Le terme est dŽ rivŽ du mongol jargu qui signiŽ e : affaire, plainte, justice 117. Les procŽ dures du y¨rghâ furent per ues par les musulmans comme Ž tant contraires ˆ la shar®a et relevant de lÕarbitraire le plus total 118. Ibn ®Arabsh¨h mentionne parmi les r gles du y¨s¨ la pratique de « rŽ duire les hommes libres ˆ lÕesclavage et de se transmettre les paysans par hŽ ritage119 ». Il est vrai que les informations rapportŽ es par Ibn ®Arabsh¨h sont tardives par rapport ˆ lՎ poque impŽ riale mongole, mais, comme on lÕa vu, cet auteur a vŽ cu tr s longtemps en milieu turco-mongol et a disposŽ de nombreuses sources orales. On peut formuler lÕhypoth se quÕil y a dans ce tŽ moignage une allusion ˆ une institution impŽ riale : le partage des terres et des populations entre les Gengiskhanides 120. Dans la sociŽ tŽ mongole prŽ -impŽ riale, nuntuq (turc yurt) dŽ signait un territoire sufŽ sant ˆ lÕentretien dÕun certain nombre de feux nomades ; ces populations appartenaient ˆ un « seigneur (noyan) » qui contr™lait Ž galement le territoire sur lequel elles vivaient 121. Gengis-khan conserva la base de cette organisation des terres et des hommes, mais il la transforma ˆ lÕavantage de la famille impŽ riale en considŽ rant lÕempire, tout entier, comme un « bien patrimonial » appartenant ˆ lui-m me, aux quatre Ž ls de son Ž pouse principale et ˆ cette derni re. Cette nouvelle disposition est rapportŽ e dans lÕHistoire secr te au paragraphe 203. Gengis-khan dit ˆ SigiQuduqu : « Et, tandis quÕavec la protection du Ciel Ž ternel je donne une direction ˆ la nation (ulus) [. . .]. Divise ceux qui ont des parois de feutre [les nomades], sŽ pare ceux qui ont des portes (egŸde) de bois [les sŽ dentaires], partage lÕenti re nation

HS § 203. Sur le y¨rghâ, voir David O. Morgan, « The Great Y¨s¨ » : 173-175 ; Ann K. S. Lambton, Continuity and Change in Medieval Persia (Londres : Tauris, 1988) : 95-96 ; 274-275. 117 Lessing, « zargu » : 1037. 118 Voir infra la critique des pratiques judiciaires des Mongols par Ibn ®Arabsh¨h. 119 Ibn ®Arabsh¨h : 233. 120 Pour analyser les prŽ ceptes du y¨s¨, relevŽ s dans les diffŽ rentes sources, nous partons toujours, lorsque cela est possible, de sa signiŽ cation dans lÕHistoire secr te. 121 Boris Vladimirtsov, Le rŽ gime social des Mongols : 144. 115 116

LE GRAND JASAQ DE GENGIS-KHAN

53

selon le nom des peuples et donne-les en apanage (qubi) ˆ notre m re, ˆ mes cadets et ˆ mes Ž ls122. » Ce syst me tirait son origine dÕune coutume nomade : le partage, en guise de butin, des populations vaincues et des terres de p‰ture. Apr s avoir Ž tabli leur domination sur les populations sŽ dentaires, en Chine et en Iran, les Mongols Ž tendirent cette pratique de partage aux terres agricoles, qui furent allouŽ es aux princes de sang et ˆ certains fonctionnaires123. Les sources chinoises dŽ signent les terres donnŽ es en apanage par le terme fen-ti ; on en trouve la premi re mention sous …gšdei124. La distribution des terres avec les populations agricoles et sŽ dentaires qui y vivaient fut pratiquŽ e ˆ grande Ž chelle par Mšngke, notamment apr s le recensement gŽ nŽ ral de la population de lÕempire quÕil avait ordonnŽ aŽ n de se rŽ server lÕacc s aux ressources humaines et matŽ rielles de lÕensemble des territoires et dÕen redistribuer une partie entre les membres de la famille impŽ riale125. Le partage du territoire, et des revenus affŽ rents, Ž tait une forme de redistribution qui, lorsque les lignŽ es princi res entr rent en luttes intestines apr s 1259, fut utilisŽ e dans la recherche perpŽ tuelle dÕalliances. Les YŸan, par exemple, ajout rent, en 1281, soixante mille familles aux possessions des Jšchides en Chine aŽ n de les inciter ˆ revoir leur alliance avec les …gšdeides et

122 HS § 203. Il sÕagit de la plus ancienne mention textuelle de donation de terres aux membres de la famille gengiskhanide. Al-Juvayn mentionne aussi ce partage, Ž d., vol. I : 31-32 ; trad., vol. I : 42-43. Dans sa traduction de ce passage, F.W. Cleaves ne voit pas dÕopposition entre nomades et sŽ dentaires : « Dividing Those who have skirts of fet [which cover their tents as outward walls], Making to separate themselves Those who have doors of boards [on their tents] », The Secret History of the Mongols (Cambridge, Massachusetts : Harvard University Press, 1982) : 143-144. Il faut remarquer que egŸde, le terme utilisŽ ici pour dŽ signer la porte de bois dŽ signe, dans la terminologie mongole de la tente, la porti re de feutre. La porte de bois, dont les tentes peuvent Ž galement  tre pourvues, est dŽ signŽ e par le mot qaghalgh-a. En contexte nomade, parler dÕune « egŸde en bois » est donc contradictoire. Nous trouvons bien dans ce passage lÕopposition entre les nomades et les sŽ dentaires. Peter Jackson consid re que cette rŽ fŽ rence ˆ des peuples sŽ dentaires est anachronique, car Gengis-khan nÕavait pas, en 1206, soumis de pays sŽ dentaires et quÕelle re  te lÕidŽ ologie des successeurs de Gengis-khan. Il ne sÕagissait pas, ˆ cette Ž poque, de donations territoriales au sens strict du terme, mais de terres de p‰turages, voir Peter Jackson, « From Ulus to Khanate : The Making of the Mongol States, c. 1220-c. 1290 », in The Mongol Empire & its Legacy, Ž d. Reuven Amitai-Press et David Morgan (Leyde : Brill, 1999) : 18. 123 Thomas T. Allsen, « Sharing out the Empire : Apportioned Lands under the Mongols », in Nomads in the Sedentary World, Ž d. Anatoly M. Kazanov et AndrŽ Wink (Curzon : Curzon-IIAS Asian Studies Series, 2001) : 183-184. 124 On trouve Ž galement les termes : ai-ma (mongol aimaq, tribu) et tÕou-hsia (apanage), Thomas T. Allsen, « Sharing out the Empire » : 176 ; Elizabeth Endicott-West, Mongolian Rule in China. Local Administration in the Yuan Dynasty (Cambridge Mass. : Harvard University Press, 1989) : 89. 125 Sur le recensement et la politique centralisatrice de Mšngke, voir Thomas T. Allsen, « Guard and Government in the Reign of the Grand Qan Mšngke 1251-59 », Harvard Journal of Asiatic Studies (1986) 46/2 : 495-521 ; Mongol Imperialism: the Policies of the Grand Qan Mšngke in China, Russia and the Islamics Lands (Berkeley : University of California, 1987) : 116-143.

54

DENISE AIGLE

les Chaghataides 126. Al-Jâzj¨n et al-JuvaynÂ, contemporains des Ž vŽ nements quÕils dŽ crivent, font clairement allusion aux terres donnŽ es en apanage en Iran par Mšngke ˆ sa parentŽ 127. Ces dons eurent des consŽ quences importantes, puisque la population aussi Ž tait donnŽ e en apanage ˆ la famille impŽ riale : des centaines de milliers de foyers agricoles devinrent ainsi la possession des Mongols 128. Une armŽ e disciplinŽ e – Toutes les sources tŽ moignent de lÕimportance que Gengiskhan accordait ˆ lÕarmŽ e. Dans lÕHistoire secr te, en particulier, la plupart des jasaq sont des directives militaires, quÕil sÕagisse de discipline ou de tactique. Au paragraphe 153 sont mentionnŽ s deux jasaq de Gengis-khan : « Au moment dՎ craser lÕennemi, nous ne nous arr terons pas pour le butin. Une fois que nous les aurons compl tement Ž crasŽ s, ce butin sera bel et bien n™tre [. . .]. Si nous sommes contraints par lÕadversaire ˆ reculer, nous retournerons [. . .]. Ceux qui nÕy retourneront pas, nous leur trancherons le cou ! » Ici, lÕidŽ e sous-jacente qui entra”ne cette sanction peut se comprendre non pas comme une atteinte ˆ lÕautoritŽ du chef, mais comme une trahison envers ce dernier et les compagnons dÕarmes129. Cet acte mŽ rite la dŽ capitation, la peine la plus dŽ gradante, car elle comporte lՎ coulement du sang, « incompatible [dans le cas dÕun humain] avec une survie dÕanc tre et donc la transmission de lÕidentitŽ clanique 130 ». En effet, selon le syst me de reprŽ sentations du chamanisme, faire couler le sang interdit toute survie sociale. LÕinterdiction du pillage, lÕun des aspects de la vie dans la steppe qui engendrait non seulement des actions de vengeance, mais perturbait fortement le commerce, fait Ž galement lÕobjet dÕun jasaq dans le cadre dÕopŽ rations militaires au paragraphe 257 de lÕHistoire secr te. En 1219, pendant la campagne contre le Khw¨razmsh¨h, Toqucar, un des gŽ nŽ raux de Gengis-khan, avait sans autorisation du qan attaquŽ le gouverneur de HŽ rat et asservi les paysans. Toqucar Ž chappa ˆ la mort, mais il fut dŽ mis de son commandement militaire. Cette sanction est le re et de la discipline de fer imposŽ e par Gengis-khan ˆ ses troupes. LÕHistoire secr te mentionne une directive, donnŽ e par Gengis-khan ˆ son gŽ nŽ ral SŸbeÕetei BaÕatur sous la forme dÕun jasaq, et qui atteste ˆ cette Ž poque (en 1205) de lÕimportance symbolique du chamanisme de chasse : « Sur votre route, le gibier sera sans doute nombreux [. . .], ne laisse pas les soldats [. . .], sՎ lancer sur le gibier. Ne Voir Thomas T. Allsen, « Sharing out the Empire » : 184-185. Thomas T. Allsen, « Sharing out the Empire » : 177. 128 Sur cette question, voir Thomas T. Allsen, « Sharing out the Empire » : 172-190. Au moment des conqutes, des groupes de populations furent Ž galement rŽ duits ˆ lÕesclavage. Par exemple, lorsque Samarcande tomba, en 1220, Gengis-khan partagea trente mille artisans entre ses Ž ls, voir Peter Jackson, « From Ulus to Khanate » : 20. 129 Akanc®i : 290 mentionne la m me peine en cas de trahison. 130 Roberte Hamayon, « MŽ rite de lÕoffensŽ vengeur » : 124. 126 127

LE GRAND JASAQ DE GENGIS-KHAN

55

chasse pas en battue avec dŽ mesure. Et, si tu chasses en battue dans le dŽ sir dÕaugmenter les provisions pour les soldats, chasse avec mesure131. » Ce jasaq sÕexplique, comme nous lÕavons vu ci-dessus dans le cadre du cycle de lՎ change qui sous-tend le chamanisme, dans lequel le chasseur Ž tait tenu de faire preuve de modŽ ration aŽ n de prŽ server lՎ quilibre entre le monde des hommes et celui des animaux, ceci pour ne pas  tre obligŽ de rendre trop en retour. Les sources non mongoles mentionnent Ž galement de nombreuses rŽ glementations en mati re de discipline militaire132, toutes passibles de la peine capitale sans que soit prŽ cisŽ e la mani re dont le coupable Ž tait exŽ cutŽ , les auteurs nÕayant pas saisi la portŽ e symbolique du mode de mise ˆ mort. Jean de Plan Carpin et al-Juvayn rapportent quÕil Ž tait interdit, sous peine de mort, de changer dÕunitŽ militaire. Bar Hebraeus corrobore cette information en prŽ cisant que le chef qui acceptait de recevoir le contrevenant ˆ la r gle subissait le m me sort133. Comme le refus de revenir au combat, sanctionnŽ dans lÕHistoire secr te, ce dŽ lit Ž tait considŽ rŽ comme un acte de trahison : celui qui sÕen rendait complice Ž tait lui-m me passible de la peine capitale. On trouve une version un peu diffŽ rente de cette r gle militaire chez al-®Umar : celui qui trouvait un fuyard et qui ne le restituait pas ˆ son chef Ž tait exŽ cutŽ 134. Al-®Umar mentionne dÕautres r gles en mati re de discipline militaire et de responsabilisation des troupes : « Au cours de lÕattaque ou de la fuite, lorsquÕun soldat perd son paquetage, son arc ou son bagage [carquois], celui qui le suit doit descendre [de cheval] pour lÕaider, sinon il est tuŽ 135. » Il est possible que cette r gle se fonde sur la notion dÕomoq 136 qui est sous-jacente dans lÕHistoire secr te au paragraphe 190 : « Si, alors que nous sommes vivants, nos carquois nous ont Ž tŽ ™tŽ s par des ennemis, ˆ quoi bon vivre ! » En mongol, omoq (Ž  rtŽ ) et oboq (famille, clan, tribu) pourraient  tre liŽ s Ž tymologiquement, mais ils ont Ž voluŽ sŽ parŽ ment137. Le lien sŽ mantique entre les deux termes pourrait expliquer la signiŽ cation de ce jasaq. En effet, la Ž ertŽ du clan, et sa survie, rŽ sidait dans sa capacitŽ ˆ se dŽ fendre ou ˆ  tre vengŽ par les membres du lignage : celui qui perdait son carquois ne pouvait plus sÕacquitter de ce devoir fondamental, il perdait alors son honneur 138. HS § 199. Comme on lÕa vu, al-Juvayn consacre une partie de son chapitre sur les y¨s¨ Ž dictŽ s par Gengis-khan ˆ dŽ crire comment il organisa lÕarmŽ e. 133 Bar Hebraeus, Ž d. : 412. 134 Al-®UmarÂ, Ž d. : 8. Al-MaqrÂz reprend lÕensemble de la liste Ž tablie par al-®Umar ; par consŽ quent, pour ne pas alourdir lÕexposŽ , nous ne le mentionnerons que lorsque son point de vue diff re de celui dÕal-®UmarÂ. 135 Al-®UmarÂ, Ž d. : 9. 136 Sur cette notion, voir Roberte Hamayon, « MŽ rite de lÕoffensŽ vengeur » : 122. 137 Voir Lessing, « obog » et « omog » : 611. Sur lՎ volution Ž tymologique du terme, voir Doerfer vol. IV, n¡ 572. 138 Sur la notion dÕomog, voir Roberte Hamayon, « MŽ rite de lÕoffensŽ vengeur » : 122-123. 131 132

56

DENISE AIGLE

Le chamane, la famille impŽ riale et lÕƒtat – Le chamane, comme on lÕa vu, Ž tait un personnage central dans le syst me de reprŽ sentations du chamanisme. Le terme bš¾e, qui en mongol dŽ signe le chamane, appara”t cinq fois dans lÕHistoire secr te139. Dans ce texte, les chamanes remplissent leurs fonctions traditionnelles : ils tirent des augures et dialoguent avec les esprits de la dynastie des Chin 140, qui in igent la maladie ˆ …gšdei pour se venger des incursions mongoles dans leurs territoires141. Sous le r gne de Mšngke, comme Guillaume de Rubrouck en tŽ moigne, les chamanes Ž taient nombreux ˆ Qaraqorum 142. Il les dŽ crit comme les « pr tres des Mongols 143 » et les nomme divini, devins 144, ce qui rend bien compte de la nature de leurs activitŽ s145. Les chamanes prŽ disaient les jours fastes et nŽ fastes : lÕarmŽ e nÕentrait jamais en guerre sans leur avis. Ils Ž taient chargŽ s de puriŽ er par des fumigations tout ce qui Ž tait apportŽ ˆ la cour146. Ils Ž taient consultŽ s pour rendre un avis dans les affaires de « sorcellerie ». Une accusation de sorcellerie pouvait en effet motiver les ch‰timents les plus graves et servir de prŽ texte pour Ž liminer des opposants politiques. Guillaume de Rubrouck rapporte : « Mšngke mÕa dit de sa bouche que Chamus [Oghul Qaimish] Ž tait la pire des sorci res et que, par ses sortil ges, elle avait dŽ truit toute sa parentŽ 147. » Ce tŽ moignage est lՎ cho des r glements de compte sanglants qui eurent lieu entre les descendants dÕ…gšdei et les partisans de Mšngke apr s lÕaccession de ce dernier au pouvoir. Oghul Qaimish, la veuve de GŸyŸk, qui lui avait succŽ dŽ en 139 HS § 181, une fois et § 272, quatre fois. Sur le r™le des chamanes, voir Elizabeth EndicottWest, « Notes on Shamans, Fortune-Tellers and Yin-Yang Practitioners and Civil Administration in YŸan China », in The Mongol Empire & its Legacy, Ž d. Reuven Amitai-Press & David Morgan, (Leyde : Brill, 1999). 140 La dynastie des Chin (1115-1234) est dÕorigine non chinoise. Elle est issue de la victoire des nomades sur la dynastie Liao (i.e. Kitan) et des Song du Nord par des nomades de Mandchourie, les JŸrcen. Sur les dynasties nomades en Chine, voir Thomas J. BarŽ eld, The Perilous Frontier : 164-186 ; Herbert Franke, « The forest peoples of Manchuria : Kitans and Jurchens », in The Cambridge History of the Early Inner Asia, Ž d. D. Sinor (Cambridge : Cambridge University Press, 1990) : 400-423. 141 HS § 272. 142 Rubrouck Ž crit quÕils ont toujours un chef dont la maison est placŽ e ˆ la distance dÕun jet de pierre de la plus grande maison de Mšngke, XXXV, Ž d. : 300 ; trad. Kappler : 216 ; Jackson/Morgan : 240. 143 Rubrouck, XXXV, Ž d. : 300 : « Divini ergo, sicut ipse confessus est, sunt sacerdotes eorum ». 144 Rubrouck et Ganjakec®i consid rent les femmes chamanes comme des sorci res. 145 Al-Juvayn conna”t le nom turc q¨m ; il le dŽ crit comme un « expert dans les sciences de la magie et le traitement des maladies », John Andrew Boyle, « Turkish and Mongol Shamanism in the Middle Ages », Folklore (1972) 83 : 179. Voir le tŽ moignage dÕIbn Shadd¨d sur les q¨m en MŽ sopotamie ca. 1259 dans Reuven Amitai-Preiss, « Evidence for the Early Use of the Title Âlkh¨n among the Mongols », Journal of the Royal Asiatic Society (1991) I/3 : 355. 146 Rubrouck, XXXV, 3, Ž d. : 301 ; trad. Kappler : 216 ; Jackson/Morgan : 241. 147 Rubrouck, XXXVI, 9, Ž d. : 308 : « Quod Chamus fuit pessima sortilega, et quod per sortilegia sua destruxerat totam parentelam suam ».

LE GRAND JASAQ DE GENGIS-KHAN

57

tant que rŽ gente, fut exŽ cutŽ e, sans aucun doute, ˆ cause de son opposition ˆ Mšngke 148. Le tŽ moignage de Guillaume de Rubrouck sur la sanction des actes de sorcellerie est conŽ rmŽ par al-®Umar qui Ž crit : « Celui qui pratique la magie (si½r) est tuŽ 149. » Les sources persanes mentionnent aussi la condamnation de plusieurs personnages qui furent exŽ cutŽ s pour avoir exercŽ des sortil ges ou sՐ tre livrŽ s ˆ des pratiques magiques 150. Tughanjuq Khatun, lÕune des femmes dÕArghun, fut accusŽ e par les chamanes (q¨m¨n) dՐ tre responsable « par magie (si½r va j¨dâ) » de la maladie de son Ž poux. JugŽ e au y¨rghâ et condamnŽ e, elle fut noyŽ e ainsi que dÕautres femmes de son entourage 151. Nous pouvons Ž galement mentionner le cas dÕun fonctionnaire persan, Majd al-Mulk al-YazdÂ, qui, inquiŽ tŽ pour son inimitiŽ avec les JuvaynÂ, fut accusŽ de sorcellerie sous le prŽ texte quÕil Ž tait en possession dÕun « morceau de peau de lion sur lequel quelque chose Ž tait rŽ digŽ avec une Ž criture inconnue 152 ». EnvoyŽ au y¨rghâ, Majd al-Mulk al-Yazd fut condamnŽ ˆ mort. AŽ n de lui interdire toute possibilitŽ de survie symbolique, son corps, coupŽ en morceaux, fut « mis sur le feu », puis mangŽ , tandis que son sang Ž tait bu153. La famille impŽ riale et les marchands – Gengis-khan accorda, comme on lÕa vu avec lÕaffaire dÕUßr¨r, une grande importance au commerce d s le dŽ but de ses conqu tes. Les recherches sur lՎ mergence des empires nomades ont montrŽ que lÕalliance entre un « clan charismatique » et les grands nŽ gociants a favorisŽ lÕexpansion des ƒtats nomades, qui ont ainsi Ž tŽ en mesure dÕexercer un contr™le politique sur les rŽ seaux commerciaux154. Paradoxalement, seuls deux auteurs du corpus font allusion ˆ une rŽ glementation des activitŽ s des marchands dans le cadre du y¨s¨. Al-JuvaynÂ

148 Les sources persanes conŽ rment les purges qui eurent lieu dans la famille impŽ riale apr s lÕintronisation de Mšngke. 149 Al-®UmarÂ, Ž d. : 8. 150 Voir Ma®sâme Ma®dankan, « Y¨s¨ dar ®ahd-i Álkh¨ni¨n », Ma®¨ref (1371sh) 9/1: 3-14. 151 RashÂd al-DÂn, Ž d. KarÂm : 823-824. 152 RashÂd al-DÂn, Ž d. KarÂm : 785-787. 153 Al-Va×רf : 108. Il est possible que la description de cette forme dÕanthropophagie rel ve de lÕimagination de lÕauteur. Il semble cependant que les Mongols ont pratiquŽ une certaine forme de crŽ mation, si lÕon en croit un curieux passage dÕal-Nasav : « Ils [les Mongols] avaient lÕhabitude de brž ler les os de tout sultan enterrŽ dans quelque pays que ce fž t, croyant que tous les sultans Ž taient descendants dÕun anc tre commun », citation dÕapr s Boyle, World Conqueror, vol. II, note 77 : 387. Al-Nasav cite en exemple le cas du sultan Ma½mâd de Ghazna, dont le corps fut exhumŽ et brž lŽ , et celui du Khw¨razmsh¨h, ®Al¨¾ al-DÂn Mu½ammad, dont les os, dŽ terrŽ s, furent envoyŽ s ˆ …gšdei, ˆ Qaraqorum, pour quÕils soient brž lŽ s. Cette pratique trouve son origine dans la croyance, selon le syst me de reprŽ sentations du chamanisme, que la survie dÕanc tre a lieu ˆ travers les os ou lÕappareil respiratoire. 154 Voir la bibliographie dans Thomas T. Allsen, « Mongolian Princes and Their Merchant Partners, 1200-1260 », Asia Major (1989) 2/2 : 83-84.

58

DENISE AIGLE

rapporte une anecdote o il est question dÕun marchand musulman qui, ayant perdu le capital qui lui avait Ž tŽ conŽ Ž par …gšdei, est pardonnŽ par ce dernier155, tandis quÕal-®Umar Ž crit que le commer ant qui perd, trois fois de suite, le capital qui lui a Ž tŽ remis est passible de la peine de mort156. Le r™le et lÕimportance des marchands dans lÕempire mongol ont Ž tŽ Ž tudiŽ s par Thomas Allsen et Elizabeth Endicott-West dans deux articles parus en 1989 157. Le terme utilisŽ pour dŽ signer les grands nŽ gociants qui circulaient dans lÕempire Ž tait ortoq 158, un mot dÕorigine turque qui signiŽ e « partenaire » ; un ortoq Ž tait un marchand qui effectuait des transactions avec un capital qui lui avait Ž tŽ conŽ Ž par un prince gengiskhanide ou par un haut fonctionnaire de la cour159. Pendant le r gne dÕ…gšdei (1229-1241), lÕin uence des marchands sÕaccrut, comme en tŽ moignent les observateurs envoyŽ s en Mongolie par les Sung ; lÕun dÕentre eux, qui arriva ˆ Qaraqorum en 1234, rapporte que lÕempereur, les princes et les princesses du sang, leurs familles, tous conŽ aient leur argent aux musulmans pour le faire fructiŽ er160. Ces marchands avaient pris lÕhabitude de dŽ clarer faussement que les marchandises quÕils transportaient avaient Ž tŽ dŽ robŽ es et ils en demandaient la restitution ˆ la population locale. Dans une sociŽ tŽ nomade, o les troupeaux sՎ garaient souvent, et en conformitŽ avec les r gles dÕaide mutuelle, la coutume obligeait en effet les voisins ˆ restituer les animaux perdus. Cette pratique, dŽ crite par de nombreux voyageurs, fut institutionnalisŽ e dans le cadre de lÕempire161 et les marchands musulmans rŽ alis rent quÕils pouvaient en tirer avantage. Pour mettre Ž n aux abus, Mšngke sÕeffor a de placer les marchands sous le contr™le de lÕadministration centrale : il leur fut interdit dÕutiliser les chevaux du rŽ seau postal ofŽ ciel, le yam (mo. cl. jam)162, leurs revenus furent contr™lŽ s et, comme les autres sujets de lÕempire, ils furent soumis ˆ lÕimp™t163. La r gle citŽ e par al-®UmarÂ, qui comme on lÕa vu a recueilli nombres dÕinformations aupr s des marchands, atteste lÕimportance du commerce dans lÕempire, et elle pourrait  tre un Ž cho lointain et dŽ formŽ des rŽ glementations imposŽ es aux marchands164.

Al-JuvaynÂ, Ž d., vol. 1 : 167 ; trad., vol. I : 209. Al-®UmarÂ, Ž d. : 8. 157 Voir Thomas T. Allsen, « Mongolian Princes »; Elizabeth Endicott-West, « Merchant Associations in YŸan China : The Ortoy », Asia Major (1989b) 2/2 : 127-154. 158 Voir Doerfer, vol. II, n¡ 446. 159 Thomas T. Allsen, « Mongolian Princes » : 85. 160 Thomas T. Allsen, « Mongolian Princes » : 94. 161 Thomas T. Allsen, « Mongolian Princes » : 98. 162 Sur le yam, voir Doerfer, vol. IV, n¡ 1812. 163 Thomas T. Allsen, « Mongolian Princes » : 105. 164 Qubilai (r. 1260-1294), fondateur de la dynastie des YŸan, mit en place des mesures pour rŽ duire les privilges dont jouissaient les marchands, voir Elizabeth Endicott-West, « Merchant Associations » : 133-149. 155 156

LE GRAND JASAQ DE GENGIS-KHAN

59

Les prŽ ceptes du y¨s¨ et la rŽ glementation des rapports sociaux Au moment de la fondation de lÕ« ƒtat mongol », Gengis-khan lui avait donnŽ un cadre juridique. Tous les actes qui pouvaient  tre une source dÕanimositŽ entre clans ou entre individus et qui, avant la pŽ riode impŽ riale, Ž taient sanctionnŽ s dans le cadre du syst me vindicatoire, furent rŽ glementŽ s. En effet, dans lÕancien syst me tribal, les infractions envers un rival nՎ taient pas considŽ rŽ es comme telles, mais comme une revanche lŽ gitime ; elles Ž taient externes au clan et, m me, valorisantes, comme en tŽ moigne lÕHistoire secr te. Seuls les actes qui mettaient en danger lÕordre clanique interne Ž taient considŽ rŽ s comme des dŽ lits. En revanche, dans le syst me Ž tatique instaurŽ par Gengis-khan, toute infraction devint interne et fut donc considŽ rŽ e comme une atteinte ˆ lÕordre impŽ rial. Condamnation du vol et du mensonge – Au quriltai de 1206, comme on lÕa vu, Gengis-khan nomma un juge supr me. Il conŽ a ˆ Sigi-Quduqu le soin de punir le vol et dÕinterdire le mensonge. Le vol occasionnait en effet des campagnes de revanche et lÕhostilitŽ entre les tribus. Toutes les sources mentionnent la condamnation du vol 165, mais les tŽ moignages des auteurs diff rent quant ˆ la sanction quÕencourait le voleur. Kirakos GanjakecÕi Ž crit : « Ils dŽ testent ˆ tel point le vol quÕils le punissent de la mort la plus terrible166 », alors que Guillaume de Rubrouck prŽ cise que la peine de mort nՎ tait appliquŽ e que pour un « vol considŽ rable167 ». Le vol dÕun mouton, par exemple, nՎ tait sanctionnŽ que par la bastonnade 168. Marco Polo, dans le Devisement du monde, dŽ crit la mani re dont les Mongols rendaient la justice. Il Ž crit que les vols de petite importance Ž taient sanctionnŽ s par un certain nombre de coups de b‰ton (bastonnees), en fonction de la gravitŽ du dŽ lit, mais que le vol des chevaux et autres choses importantes (et si emble un cheval ou autre grant chose) Ž tait passible de la peine de mort (le font trenchier parmi dÕune espee)169. Les animaux reprŽ sentaient en effet une part importante du patrimoine de chaque lignage. Ibn ®Arabsh¨h mentionne aussi la condamnation du vol, mais, ayant pour cadre de rŽ fŽ rence la shar®a, il sÕintŽ resse surtout ˆ la peine lŽ gale (½add ) appliquŽ e au voleur. Selon son tŽ moignage, la sanction dŽ pendait de la mani re dont Ž tait commis le dŽ lit. On coupait la main du voleur si le vol Ž tait commis « en per ant (bi l-naqb)

165 HS ; Carpin ; Rubrouck ; al-Juvayn ; al-Jâzj¨n ; Ganjakec®i ; Akanc®i ; al-®Umar ; alMaqrÂz ; Ibn ®Arabsh¨h. 166 Ganjakec®i : 202 ; Akanc®i : 291. 167 Rubrouck, VIII, 2, Ž d. : 186 : « Item enorme furtum puniunt morte ». 168 Rubrouck, VIII, 2, Ž d. : 186 ; trad. Kappler : 103 ; Jackson/Morgan : 93-94. 169 Marco Polo, Le Devisement du monde, tome II, Ž d. Jeanne-Marie Boivin, Laurence HarfLancner et Laurence Mathey-Maille (Paris : Droz, 2003) : 35.

60

DENISE AIGLE

[les tentes] », en dÕautres termes par effraction ; le voleur Ž tait cruciŽ Ž si le vol avait lieu « dans les tentes170 », sans autre prŽ cision, cÕest-ˆ-dire sans effraction. Dans lÕislam, la peine lŽ gale (½add ) dŽ signe la sanction de certains actes et interdits mentionnŽ s dans le Coran171. Le vol nÕest sanctionnŽ par lÕamputation de la main que si plusieurs conditions sont rŽ unies. LÕobjet volŽ doit avoir une valeur au moins Ž gale ˆ dix dirhams, le vol doit  tre commis par effraction et la stipulation dans le droit islamique « en garde (½irz)172 » exclut le vol ˆ lՎ gard dÕun proche parent ou dans une maison o le voleur peut  tre autorisŽ ˆ entrer. LÕobjectif dÕIbn ®Arabsh¨h Ž tant de montrer que la loi mongole est en contradiction avec la shar®a, il Ž crit que le vol commis sans effraction est sanctionnŽ par une peine plus forte, la mort par cruciŽ xion, que le vol par effraction173. Il est peu probable que ces peines, la cruciŽ xion ou lÕamputation de la main, aient Ž tŽ appliquŽ es par les Mongols dans les cas de vol, la sanction la plus courante pour ce dŽ lit Ž tant, comme on lÕa vu, la bastonnade. On peut cependant supposer, dÕapr s le tŽ moignage dÕIbn ®Arabsh¨h, que les vols commis dans les tentes dont on trouve des mentions dans lÕHistoire secr te, furent sanctionnŽ s par lÕordre gengiskhanide au m me titre que les vols de bŽ tail. La condamnation du mensonge sÕexplique Ž galement dans le cadre du nouvel ordre impŽ rial174. Dans lÕancien syst me pluri-clanique, comme en tŽ moigne lÕHistoire secr te, il nՎ tait pas interdit dÕutiliser la ruse et le mensonge pour vaincre les ennemis extŽ rieurs, alors que, dans le cadre de lÕempire, tout mensonge pouvait porter atteinte ˆ lÕordre Ž tabli par le « super-clan » gengiskhanide, et donc  tre considŽ rŽ comme un dŽ lit envers le qan. La croyance, ancrŽ e chez les Mongols mŽ diŽ vaux, que la sanction du mensonge Ž tait surnaturelle : maladie, mort ou autres catastrophes pour soi-m me ou sa descendance, a pu  tre utilisŽ e comme un moyen pour imposer cette r gle. Homicide et droit de compensation – Dans la sociŽ tŽ mongole traditionnelle, le meurtre Ž tait expiŽ dans le cadre du syst me vindicatoire et, notamment, par le droit de compensation (haci ) dont bŽ nŽ Ž cierait la famille de la victime. Le terme haci nÕexprime pas lÕidŽ e dÕune vengeance, mais plut™t celle dÕun jugement. Il est teintŽ

Ibn ®Arabsh¨h : 233. Il existe quelques lŽ g res diffŽ rences entre les diverses Ž coles juridiques, voir Joseph Schacht, Introduction au droit musulman (Paris : Maisonneuve & Larose, 1983). 172 Le ½irz signiŽ e que lÕobjet doit  tre conservŽ, soit dans un lieu sž r, soit en prŽ sence dÕun gardien. 173 Ibn ®Arabsh¨h dŽ nonce aussi la pratique de conŽ squer « les biens hŽ ritŽ s (til¨d ) » du voleur alors que, selon le droit musulman, il est interdit de sÕen emparer Ž tant donnŽ que ces biens, dont le voleur a hŽ ritŽ , nÕont pu  tre acquis de mani re illicite. 174 HS ; al-Juvayn ; al-Jâzj¨n ; al-®UmarÂ. Akanc®i mentionnent le faux tŽ moignage. 170 171

LE GRAND JASAQ DE GENGIS-KHAN

61

dÕune valeur positive : le « retour vers soi dÕun dž 175 ». La sanction de lÕhomicide est rarement mentionnŽ e dans les sources non chinoises176 ; seul Guillaume de Rubrouck rapporte quÕun attentat ˆ la vie Ž tait passible de la peine de mort177. Il nÕest pas certain que les auteurs mŽ diŽ vaux aient saisi le fonctionnement du droit de compensation. En effet, al-Juvayn semble y faire allusion lorsquÕil Ž crit : « Par la bontŽ du crŽ ateur qui conna”t le rang de chaque peuple et en conformitŽ avec le y¨s¨ ancien (y¨s¨-yi qadÂm) de Gengis-khan, le prix du sang (qiר×) versŽ pour le meurtre dÕun musulman est de cinquante lingots dÕor (balish) alors quÕil est dÕun ‰ne pour un Khitai [Chinois] 178. » Si nous faisons abstraction du c™tŽ Ž diŽ ant de ce rŽ cit179, il semble que lÕhistorien persan, tout en utilisant le terme qiר×, correspondant islamique de haci, mentionne le droit de compensation appliquŽ dans la sociŽ tŽ de la steppe. Il est possible en effet de rapprocher le tŽ moignage dÕal-Juvayn de la lŽ gislation mise en place par les YŸan en Chine180. Dans les cas dÕattentat ˆ la vie, il existait un type de dŽ dommagement, le shao-mai-yin, littŽ ralement « argent [pour] brž ler et enterrer », indemnitŽ Ž xŽ e ˆ cinquante onces dÕargent (liang), convertibles en billets de banque, qui Ž tait versŽ e par le responsable de la mort aux parents de la victime. Dans cette rŽ paration de lÕhomicide, fusionnaient la peine compensatoire mongole et une coutume issue du culte des anc tres, tr s important dans la culture chinoise. Condamnation de lÕadult re – Le mariage reprŽ sentait pour les Mongols la r gle (tšre) par excellence, car il conditionnait lÕensemble des rapports sociaux. En effet, tŸre, la forme bouriate du mongol classique tšre, dŽ signe aujourdÕhui, dÕune part, « lÕinstitution et le rituel du mariage » et, dÕautre part, « le [mode de] gouvernement ». La notion de base de cette racine est celle « dÕengendrement (tŸreke) » : elle articule ainsi les principes dÕalliance et de Ž liation et consid re donc la reproduction de la sociŽ tŽ dans lÕabsolu181. 175 Sur le terme haci et sur cette pratique, voir Roberte Hamayon, « MŽ rite de lÕoffensŽ vengeur » : 110-111. 176 Le YŸan-shih mentionne 73 cas dÕhomicides, voir Fran oise Aubin, « Les sanctions et les peines » : 252-255. 177 Rubrouck, VIII, 2, Ž d. : 185-186 : « homicidium puniunt capitali sententie ». 178 Al-JuvaynÂ, Ž d., vol. I : 164 ; trad., vol. I : 207. 179 En valorisant les musulmans par rapport aux Chinois, al-Juvayn satisfait ˆ la fois les populations musulmanes tombŽ es sous la domination ilkhanide et ses ma”tres mongols qui avaient subjuguŽ une grande partie de la Chine. 180 Sur le droit sous les YŸan, voir Paul Heng-chaÕo ChÕen, Chinese legal tradition under the Mongols ; Fran oise Aubin, « Les sanctions et les peines » : 253-254 et Paul Ratchnevsky, « Jurisdiction, penal Code, and cultural Confrontation under Mongol-YŸan Law », Asia Major (1993) 6 : 161-179. 181 Roberte Hamayon, La chasse ˆ lÕ‰me : 224. Voir lՎ tude par Marie-Dominique Even et Rodica Pop sur les diffŽ rentes associations linguistiques entre tšre, dŽ signant le mariage, la coutume, la r gle, et lÕƒtat chez les populations mongoles bouriates et khalkhas dans Rodica Pop, Le mariage chez les Mongols : 9-10.

62

DENISE AIGLE

Toutes les sources, quÕelles soient chrŽ tiennes ou islamiques, mentionnent que lÕadult re Ž tait sanctionnŽ par la peine de mort182. Il faut donner au terme « adult re » le sens quÕil avait dans la sociŽ tŽ mongole qui pratiquait une stricte exogamie183. En effet, on devrait plut™t parler ici dÕun « inceste au sein du clan » qui rompt le principe de lÕexogamie et non dÕun « adult re » dans le sens quÕil rev t dans le christianisme et dans lÕislam. La sanction de ce dŽ lit visait ˆ prŽ server le principe de lÕexogamie, comme en tŽ moignent les sens du terme tšre et de son dŽ rivŽ , tšrel, qui dŽ signe lÕunitŽ de parentŽ exogame184. LÕadult re pouvait  tre source de dŽ sordres au sein du clan, mais, commis avec une femme dÕun lignage extŽ rieur, il Ž tait sans aucun doute tolŽ rŽ , puisquÕil ne mettait pas en danger lÕharmonie interne au clan. LE Y¤S¤ ET LA SHAR®A Y¨rghâ mongol et droit musulman Ë lÕexception dÕal-JuvaynÂ, tous les auteurs musulmans ont relevŽ les incompatibilitŽ s entre certaines coutumes mongoles et la shar®a. Il est dÕautant plus surprenant quÕIbn ®Arabsh¨h, qui situe son discours dans le cadre du droit musulman, nÕait pas dŽ noncŽ les interdits sur lÕeau et lÕabattage des animaux selon le rite islamique, les r gles coutumi res mongoles qui ont le plus frappŽ les musulmans. Il est vrai que sa prŽ occupation majeure Ž tait ailleurs : critiquer les procŽ dures judiciaires du y¨rghâ mongol. La juridiction religieuse exercŽ e par le juge dans le cadre de la shar®a constitue, sans aucun doute, le mod le de rŽ fŽ rence dÕIbn ®Arabsh¨h quand il dŽ crit les procŽ dures des cours dÕinvestigation chargŽ es dÕenqu ter sur les « malversations » des fonctionnaires au service des pouvoirs mongols. On peut, en effet, comparer le y¨rghâ mongol aux « cours des plaintes (maú¨lim) », qui Ž taient destinŽ es ˆ recevoir les rŽ clamations au sujet des erreurs judiciaires, des dŽ nis de justice et autres torts causŽ s par les fonctionnaires ; cette justice Ž tait rendue par les autoritŽ s politiques 185. Dans la procŽ dure juridique islamique, les preuves reposent sur la dŽ position (shah¨da) des tŽ moins et, de ce fait, la qualitŽ de tŽ moin est strictement rŽ glementŽ e. Ibn ®Arabsh¨h dŽ crit le syst me de tŽ moignage dans le y¨s¨ comme totalement contraire ˆ la shar®a. Il cite en exemple la condamnation pour la fornication (zin¨) qui peut  tre prononcŽ e sur le tŽ moignage dÕun seul tŽ moin alors que, selon le droit musulman, quatre tŽ moins hommes sont nŽ cessaires. Il Ž crit que le tŽ moignage

182 Carpin ; Rubrouck ; al-Juvayn ; al-Jâzj¨n ; Ganjakec®i ; Akanc®i ; al-®Umar ; al-Maqriz ; Ibn ®Arabsh¨h. 183 Voir infra. 184 Roberte Hamayon, La chasse ˆ lÕ‰me, note 2 : 760. 185 Voir J.E. Nielsen, « al-Maú¨lim », EI2 VI : 925-927.

LE GRAND JASAQ DE GENGIS-KHAN

63

dŽ posŽ sur « les hommes adultes » par certaines catŽ gories de personnes ( jeunes hommes, jeunes Ž lles, femmes et esclaves), strictement rŽ glementŽ dans lÕislam, est Ž galement dŽ clarŽ recevable dans le syst me lŽ gislatif mongol. En effet, selon le droit musulman, le tŽ moignage de deux femmes nÕest recevable que sÕil se rapporte ˆ des sujets pour lesquels celles-ci sont compŽ tentes, la naissance par exemple ; le tŽ moignage des jeunes hommes et des esclaves nÕest valable quÕˆ lÕintŽ rieur de leur groupe social ; quant au tŽ moignage des jeunes Ž lles, il nÕest m me pas mentionnŽ . En disqualiŽ ant les tŽ moins, Ibn ®Arabsh¨h rŽ cuse les procŽ dures juridiques mongoles. La sanction de lÕadult re est Ž galement critiquŽ e par Ibn ®Arabsh¨h. Dans lÕislam, il est passible de lÕune des peines les plus sŽ v res, la lapidation 186, alors que, selon la loi mongole, il est sanctionnŽ par « lՎ tranglement ». Ibn ®Arabsh¨h fait sans doute rŽ fŽ rence ˆ lՎ touffement qui, avec la brisure des os, Ž tait la peine la plus pratiquŽ e par les Mongols lorsque les condamnŽ s appartenaient au clan gengiskhanide, dont il ne fallait pas verser le sang aŽ n de leur permettre une survie dÕanc tres. Il est vrai que lÕon peut Ž galement supposer que la dŽ nonciation, par Ibn ®Arabsh¨h, de la pratique judiciaire mongole vise, ˆ travers elle, ˆ critiquer la rŽ alitŽ timouride. En effet, les sources tŽ moignent de lÕadhŽ sion de Tamerlan aux lois de Gengis-khan pour administrer son empire et elles attestent de la persistance, sous Tamerlan et ses successeurs, dÕun syst me pŽ nal sŽ parŽ (y¨rghâ), parall le ˆ la juridiction islamique. Au y¨rghâ, la justice Ž tait rendue par un personnel spŽ cialisŽ chargŽ de mettre la justice administrative en conformitŽ avec la loi mongole 187. Quoi quÕil en soit, m me si Ibn ®Arabsh¨h dŽ crit les procŽ dures timourides, le y¨rghâ ˆ lՎ poque de Tamerlan Ž tant hŽ ritier des pratiques mongoles, cÕest bien la justice mongole qui est visŽ e par lÕauteur. R gles coutumi res contraires ˆ la shar®a La distinction entre le pur et lÕimpur – Les musulmans ont Ž tŽ frappŽ s par les interdits sur lÕeau. Al-®Umar rapporte que, dÕapr s « le y¨s¨ ancien des Mongols », il Ž tait interdit de se tremper dans lÕeau, sous peine de mort188. Al-MaqrÂzÂ, pour mieux montrer que les r gles coutumi res mongoles sont en dŽ saccord total avec la loi islamique,

186 LÕislam condamne toute forme de relations sexuelles en dehors des liens du mariage, mais la sanction est diffŽ rente sÕil sÕagit dÕune personne cŽ libataire ou dÕun esclave : dans ce cas, la sanction dÕun tel dŽ lit est la  agellation. 187 Niú¨m al-DÂn Sh¨mÂ, Ž d. F. Tauer, Histoire des conqutes de Tamerlan intitulŽ e, øafarn¨ma, par Niú¨m uddÂn °¨mÂ, vol. II : 88-89 ; John E. Woods, « TimurÕs Genealogy », in M. Mazzaoui et V. Morreen (Ž d.), Intellectual Studies on Islam. Essays Written in Honor of Martin Dickson (University of Utah Press, 1990) : 101. 188 Al-®UmarÂ, Ž d. : 9.

64

DENISE AIGLE

dŽ forme le rŽ cit dÕal-®Umar : « Il [Gengis-khan] leur enjoignit de porter leurs v tements sans les laver […], il leur dŽ fendit de dire quÕune chose est impure (najis), car il disait que toute chose est pure (ߨhir) et il ne faisait aucune distinction entre le pur et lÕimpur189. » LÕimportance des esprits dans le syst me de reprŽ sentations du chamanisme, o il y a homologie entre « ‰me » et « esprit », permet dÕexpliquer ces tabous sur lÕeau. Apr s la mort, lÕ‰me Ž tait supposŽ e avoir une autre modalitŽ dÕexistence, sous la forme dÕun esprit. CՎ tait en vertu de cette croyance dans les esprits des morts, et dans les esprits de mani re plus gŽ nŽ rale, quÕil Ž tait interdit, par respect pour eux, de se plonger dans lÕeau des rivi res, dÕy laver les v tements et la vaisselle190, aŽ n de ne pas polluer leurs lieux de rŽ sidence. LÕinterdit sur lÕeau, en rapport avec les esprits qui y rŽ sidaient, se conjuguait chez les Mongols mŽ diŽ vaux ˆ un autre facteur : la crainte dÕattirer lÕorage, per u comme la sanction surnaturelle dÕun manquement aux obligations envers les esprits. Al-Juvayn et les sources latines prŽ cisent que ces interdits sÕappliquaient au printemps et en Ž tŽ , Ž poque o , tr s souvent, hommes ou b tes tombaient foudroyŽ s : ces morts tragiques Ž taient interprŽ tŽ es comme une vengeance des esprits. Le r™le central des ablutions rituelles dans lÕislam a conduit les musulmans ˆ dŽ nier aux Mongols toute idŽ e de « puretŽ ». Cette notion existe cependant dans la culture mongole, mais sous dÕautres formes, comme en tŽ moignent les mesures prises envers les malades, les morts et leurs familles. La cause de la maladie nՎ tait jamais per ue comme naturelle ; les Mongols considŽ raient quÕelle rŽ sultait soit dÕun manquement envers les esprits, soit dÕune vengeance. Par consŽ quent, sans rem de naturel, la maladie Ž tait traitŽ e par une action symbolique sur des Ž gurations dÕesprit, les ongon 191. Les sources mŽ diŽ vales occidentales dŽ crivent les « idoles (ydola)192 » ou les « images de feutre des morts193 » qui Ž taient vŽ nŽ rŽ es et nourries. Elles Ž taient placŽ es 189

Al-MaqrÂzÂ, Ž d. Bulaq : 358. Les sources mentionnent dÕautres interdits en rapport avec le respect dž aux esprits : ne pas uriner dans les cendres ou dans lÕeau, ne pas heurter le seuil de la tente, ne pas enjamber un plat de nourriture, ne pas jeter de la nourriture dans les cendres mais les dŽ poser avec la main, voir Carpin, Rubrouck et al-®UmarÂ. 191 LÕongon est un «  tre sacrŽ , envisagŽ dans son support matŽ riel » ; le terme sÕapplique ˆ la fois ˆ lÕesprit et ˆ lÕobjet dans lequel il rŽ side, le nourrir a pour objectif de le maintenir dans son support, Roberte Hamayon, La chasse ˆ lÕ‰me : 404. Certaines populations tŸrkes dÕAsie centrale, apr s leur conversion ˆ lÕislam, ont continuŽ ˆ croire dans les pouvoirs des ongon, voir Vladimir N. Basilov, Samanstvo u narodov Srednej Azii i Kazaxtana (Moscou : Nauka, 1992) : 232-234 ; Patrick Garrone, Chamanisme et islam en Asie centrale (Paris : Jean Maisonneuve, 2000) : 149-150. Les Tatars Baraba (XVIIIe-XXe s.) fabriquaient des Ž gurines (qong•rchaq) quÕils nourrissaient, voir Frank J. Allen, « Varieties of Islamization in Inner Asia. The case of the Baraba Tatars, 17401917 », in En islam sibŽ rien, Ž d. StŽ phane Dudoignon, Cahiers du Monde russe, (2000) 41/2-3 : 256-260. 192 Carpin, III, 2, Ž d. : 36 ; trad. : 36. 193 Rubrouck, XXV, 9, Ž d. : 232 ; trad. Kappler : 146 ; Jackson/Morgan : 156. 190

LE GRAND JASAQ DE GENGIS-KHAN

65

des deux c™tŽ s de la tente ou sur des chariots qui Ž taient gardŽ s par le chef des chamanes194. Le monde des esprits Ž tant per u comme un monde dÕaffamŽ s en qu te de gibier humain, il fallait, par consŽ quent, nourrir les ongon aŽ n de limiter leurs nuisances : maladie ou mort195. Les Mongols considŽ raient, jusquÕˆ une Ž poque rŽ cente, que la maladie survenait soit dans le cadre de lՎ change entre les deux mondes – Roberte Hamayon la qualiŽ e de « maladie-contrepartie » –, soit ˆ la suite de la transgression dÕun interdit, et cՎ tait la « maladie-sanction », situŽ e, elle, dans le champ du dŽ sordre196. Atteint par cette seconde forme de maladie et devenu la proie des esprits, le malade Ž tait Ž cartŽ de son groupe. On mettait un signe sur sa demeure aŽ n que personne ne sÕen approche, laissant au chamane le soin de dŽ couvrir lÕesprit qui, en dŽ vorant le malade, se vengeait de lui. La famille des morts, elle aussi, Ž tait considŽ rŽ e comme « impure » du fait de la croyance que la mort, comme la maladie, Ž tait une punition ou une vengeance des esprits. Tous les membres de la famille du dŽ funt Ž taient Ž cartŽ s du clan pendant une pŽ riode variable, car ils pouvaient  tre porteurs des esprits responsables de la mort de leur parent. Al-Juvayn Ž crit que, selon la loi et la coutume mongole (dar y¨s¨ va adhÂn-i mughâl), si un homme tombait foudroyŽ , sa famille Ž tait Ž cartŽ e de sa tribu pendant trois ans ; sÕil sÕagissait dÕun animal, la mise ˆ lՎ cart durait plusieurs mois197. Cette mesure touchait Ž galement ceux qui avaient assistŽ ˆ la mort de quelquÕun198. Il revenait au chamane de puriŽ er par des fumigations tout ce qui pouvait  tre porteur de malŽ Ž ces : malades, familles des morts ainsi que les animaux et les objets qui leur appartenaient 199. Les Mongols puriŽ aient Ž galement par le feu tout ce qui pŽ nŽ trait ˆ la cour du grand qan : hommes, b tes et cadeaux, au cas o « ils auraient fait un sortil ge et apportŽ du poison ou quelque malŽ Ž ce200 ». Bien entendu, lÕidŽ e de puretŽ chez les Mongols nÕest pas comparable avec cette notion dans lÕislam, puisquÕelle nÕest pas dÕordre rituel. En effet, les puriŽ cations par fumigation quÕeffectuait le chamane, tout comme lՎ loignement des personnes susceptibles dՐ tre sous lÕemprise dÕesprits malins, Ž taient des mesures de protection. En tout cas, le terrain culturel Ž tait favorable pour le dŽ veloppement rapide dÕune notion

194 Jean de Plan Carpin dit quÕon attachait ces idoles sur le lit dÕun enfant malade : III, 2, Ž d. : 36 ; trad. : 36. Akanc®i : 289 dŽ crit Ž galement le culte rendu ˆ des idoles. 195 Roberte Hamayon, La chasse ˆ lÕ‰me : 415. 196 Roberte Hamayon, La chasse ˆ lÕ‰me : 408. 197 Al-JuvaynÂ, Ž d., vol. I : 162 ; trad., vol. I : 205. 198 Rubrouck, VIII, 3, Ž d. : 186 ; trad. Kappler : 103-104 ; Jackson/Morgan : 94. SÕil sÕagissait dÕun adulte, la mise ˆ lՎ cart Ž tait dÕune annŽ e, dans le cas dÕun enfant, elle durait une lunaison. 199 Carpin, III, 10, Ž d. : 42 ; trad. : 40-41 Ž crit : « Si le feu tombe du ciel sur les troupeaux ou sur les hommes [. . .], ou sÕil leur arrive quelque accident [. . .] qui les fait se croire sous le coup dÕune impuretŽ ou dÕun mauvais sort, il leur faut se faire puriŽ er par les sorciers. » 200 Carpin, III, 10, Ž d. : 41-42 ; trad. : 40.

66

DENISE AIGLE

de puretŽ rituelle : le bouddhisme, notamment sous la forme du lama•sme tibŽ tain rŽ introduit par Altan-khan au XVIe si cle, nÕeut aucun mal ˆ imposer ses r gles et ses rites puriŽ catoires lorsquÕil devint la religion dominante en Mongolie 201. Les interdits sur le versement du sang – La mani re dÕabattre les animaux, sans faire couler le sang, a Ž galement suscitŽ la critique des musulmans, car cette pratique Ž tait contraire ˆ lÕabattage rituel dans lÕislam. Seules les sources islamiques mentionnent cette mani re de faire. Al-Juvayn est le premier ˆ Ž crire : « Au moment o les Mongols sont arrivŽ s au pouvoir, ils ont fait un y¨s¨ de ne pas abattre les animaux selon le rite islamique 202. » Al-®Umar prŽ cise : « LorsquÕils Ž gorgent un animal, ils lui lient les pattes, ils percent son ventre, lÕun dÕentre eux introduit sa main jusquÕau cœur et il le serre jusquÕˆ ce que lÕanimal succombe ou bien il lui arrache le cœur203. » Dans le syst me de reprŽ sentations du chamanisme, aŽ n de permettre la survie symbolique dÕun animal, il ne faut pas verser son sang. Il est nŽ cessaire de sauvegarder lÕappareil respiratoire ( jŸlde 204) constituŽ de la t te, de la trachŽ e, du cœur et des poumons, porteurs du souf e vital (amin). LÕarrachage lib re lÕ‰me de mani re que la viande de lÕanimal soit consommŽ e sans risque que lÕ‰me le soit aussi. LÕappareil respiratoire de lÕanimal abattu, prŽ servŽ , est considŽ rŽ comme un support de « potentialitŽ de vie [dÕ‰me] », destinŽ ˆ permettre symboliquement lÕapparition dÕun nouvel animal205. La description donnŽ e par al-®UmarÂ, qui sÕappuie vraisemblablement sur des informations orales, correspond parfaitement ˆ la mani re dÕabattre lÕanimal selon le rite chamanique. Ces rŽ cits ne permettent pas dÕafŽ rmer que cette r gle Ž tait imposŽ e aux musulmans, comme on le voit encore Ž crit dans la production scientiŽ que 206. On peut en effet supposer que le y¨s¨ mentionnŽ par alJuvayn est destinŽ aux Mongols aŽ n de leur interdire dÕimiter les musulmans, hypoth se quÕal-®Umar conŽ rme de mani re implicite puisquÕil Ž crit : « Celui qui abat un animal comme les musulmans est Ž gorgŽ 207. »

201 Sur les religions en Mongolie, voir W. Heissig, The Religions of Mongolia (Londres, 1980) ; bonne introduction ˆ lÕhistoire de la Mongolie dans : C.R. Bawden, The Modern History of Mongolia (Londres et NewYork : Kegan Paul International, 1968). 202 Al-JuvaynÂ, Ž d., vol. I : 163 ; trad., vol. I : 206. 203 Al-®UmarÂ, Ž d. : 8-9. 204 JŸlde signiŽ e « essentiel, principal, central », Marie-Dominique Even, Histoire secr te, note 34 : 255. 205 Chez les Bouriates, la force vitale de lÕappareil respiratoire est conservŽe lors des sacriŽ ces des animaux domestiques aŽ n que le troupeau puisse  tre renouvelŽ, voir Roberte Hamayon, La chasse ˆ lÕ‰me : 397. 206 Voir par exemple, Richard Foltz, « Ecumenical Mischief Under the Mongols », Central Asiatic Journal (1999) 43/1 : 44-45. 207 Al-®UmarÂ, Ž d. : 9.

LE GRAND JASAQ DE GENGIS-KHAN

67

La r gle de lÕexogamie – Les pratiques matrimoniales des Mongols ont Ž tŽ mal per ues par les chrŽ tiens et par les musulmans, qui nՎ taient pas en mesure de les situer dans leur contexte social. Jean de Plan Carpin constate quÕun homme peut Ž pouser sa sœur par une autre m re et la veuve dÕun fr re a”nŽ ou dÕun parent proche 208,tandis que Guillaume de Rubrouck Ž crit : « Au sujet de leurs mariages, vous saurez quÕici nul ne prend femme sÕil ne lÕach te […], ils peuvent Ž pouser en m me temps ou successivement deux sœurs […], il arrive parfois que le Ž ls Ž pouse toutes les femmes de son p re, ˆ lÕexception de sa m re209. » Ces tŽ moignages sont corroborŽ s par Ibn ®Arabsh¨h : « Les proches parents du mari [dŽ funt] peuvent Ž pouser la femme et ils se lՎ changent ˆ tour de r™le210. » Ces coutumes matrimoniales sÕexpliquent par lÕexogamie, pratiquŽ e soit par alliance avec la famille de la femme, soit par achat ou rapt, le cas le plus frŽ quent dans lÕHistoire secr te211. Selon la r gle de rŽ sidence patrilocale en vigueur chez les Mongols, les Ž lles Ž taient destinŽ es ˆ vivre, au loin, dans le clan du mari212. Apr s le mariage, la femme Ž tait donc en rupture avec son milieu dÕorigine. La coutume, qui voulait quÕapr s la mort du mari, lÕun des Ž ls, gŽ nŽ ralement le cadet, Ž pouse les veuves de son p re, autres que sa m re, avait pour fonction de restaurer la prise de possession originale de la femme aŽ n de maintenir lÕintŽ gritŽ du clan et de son patrimoine. Le mariage avait en effet un triple intŽ r t : accro”tre la descendance, fournir de la main-dÕœuvre, contracter, dans certains cas, une alliance ; il fallait donc maintenir les femmes du dŽ funt mari dans le clan. Cette pratique sociale, bien entendu, Ž tait en dŽ saccord complet avec la loi islamique 213. CONCLUSION Notre vision des prescriptions du y¨s¨ est, nous lÕavons vu, obscurcie par la distance qui sŽ pare de lՎ poque de Gengis-khan la plupart des sources. Chez les auteurs musulmans, confrontŽ s ˆ un ordre politique et ˆ une culture Ž trangers ˆ lÕislam, le y¨s¨ a pris tr s vite une dimension mythique. Une source cependant, lÕHistoire secr te, donne un Ž clairage central, quoique tŽ nu, sur la notion de y¨s¨ et sur son contenu. Les jasaq dont on trouve la mention dans ce texte, Ž dictŽ s par Gengis-khan, tŽ moignent de la volontŽ du qan de transformer le syst me vindicatoire prŽ -impŽ rial en un

Carpin, II, 3, Ž d. : 33 ; trad. : 32-33. Rubrouck, VII, 4, Ž d. : 185 ; trad. Kappler : 102 ; Jackson/Morgan : 91-92. 210 Ibn ®Arabsh¨h : 233. 211 Chez les Bouriates prŽ -soviŽ tiques, un simulacre de rapt permettait au pauvre de ne pas verser la compensation matrimoniale, Roberte Hamayon, « MŽ rite de lÕoffensŽ vengeur » : 114. 212 Rodica Pop, Le mariage chez les Mongols : 9-10. 213 Voir supra la discussion sur le mariage de Ghazan-khan avec la veuve de son p re. 208 209

68

DENISE AIGLE

moyen dÕexercer son pouvoir sur toute la sociŽ tŽ . Cette volontŽ se manifeste, entre autres, par lÕimposition dÕune rigoureuse discipline militaire ˆ ses troupes et par lÕinstauration dÕune instance juridique supŽ rieure, chargŽ e de punir les dŽ lits dans tout lÕempire. Mais cÕest Mšngke, le troisi me successeur de Gengis-khan, qui a le mieux ma”trisŽ la direction de lÕƒtat mongol 214. Les sources islamiques sur le y¨s¨ (mo. cl. jasaq) et plus particuli rement les sources mameloukes du XVe si cle, Ž tablissent rarement une distinction entre les r gles qui rel vent de lÕorganisation et du fonctionnement de lÕƒtat (cÕest le sens premier du terme jasaq en mongol), et les r gles coutumi res, ou yosun. Or, nous avons montrŽ que cette distinction est importante puisque la source de validitŽ nÕest pas la m me pour le jasaq et le yosun : dans le premier cas, cÕest le pouvoir, dans le second la tradition. Cette confusion entre jasaq et yosun, qui nÕest pas faite par les auteurs chrŽ tiens, est ˆ la source de la mythiŽ cation de la loi mongole dans les pays dÕIslam. La vision du y¨s¨ par les historiens mamelouks a variŽ entre le XIIIe et le XVe si cle. Nous avons vu le r™le actif quÕa jouŽ le y¨s¨ dans la guerre idŽ ologique ˆ laquelle se livr rent les Mamelouks bahrites et les Ilkhans de Perse. En revanche, pour les auteurs du XVe si cle, le y¨s¨ est per u comme un ordre social mongol opposŽ ˆ la shar®a islamique : ils incorporent dans le y¨s¨ nombre de coutumes quÕal-®Umar avait distinguŽ es des r gles Ž dictŽ es par Gengis-khan. Dans son importante Ž tude sur le y¨s¨ dans le sultanat mamelouk, David Ayalon avait montrŽ quÕal-MaqrÂz avait eu recours au concept de y¨s¨ pour condamner le r™le, de plus en plus prŽ dominant ˆ son Ž poque, du ½¨jib qui Ž tait chargŽ de la justice administrative 215. Al-MaqrÂz prŽ tendait, pour fonder cette th se, que la siy¨s¨ suivie par les sultans nՎ tait quÕune forme du y¨s¨ dont le nom avait Ž tŽ dŽ formŽ ˆ dessein : dans le sultanat mamelouk, la justice administrative avait donc pris le pas sur la justice religieuse rendue par le cadi. Comme on lÕa vu, la juridiction mongole, le yarghâ, est, elle aussi, implicitement dŽ noncŽ e par Ibn ®Arabsh¨h, lorsquÕil dŽ crit les lois de Gengis-khan. Ainsi, alMaqrÂz et Ibn ®Arabsh¨h accentuent fortement la tension entre le y¨s¨ et la shar®a, ce qui nՎ tait pas le cas dÕal-®Umar qui disait simplement que beaucoup des prŽ ceptes de Gengis-khan nՎ taient pas en conformitŽ avec la loi islamique. Au XVe si cle, la position de ces historiens mamelouks sÕexplique vraisemblablement par le contexte historique auquel les Pays du Levant venaient dՐ tre confrontŽ s : un nouveau « pŽ ril tatar », placŽ cette fois sous la houlette dÕun Turc, Tamerlan, qui, bien quÕil soit musulman, se prŽ sentait comme le restaurateur de lÕordre gengiskhanide. Au XVe si cle, le y¨s¨ semble intŽ riorisŽ dans la conscience de ces historiens, il est devenu,

Voir Thomas T. Allsen, « Guard and Government » ; Mongol Imperialism ; « Mongolian Princes and Their Merchant Partners ». 215 Voir David Ayalon, « The Great Y¨sa (part C2) » : 107-156. 214

LE GRAND JASAQ DE GENGIS-KHAN

69

en quelque sorte, un argument cherchant ˆ justiŽ er la crainte de lÕautre, le conquŽ rant tatar venu dÕOrient. Il y a peut- tre aussi, chez ces historiens liŽ s au sultanat mamelouk circassien, le dŽ sir de disqualiŽ er lÕoeuvre du sultan Baybars, le premier vainqueur des Mongols. La mani re dont Ibn TaghrÂbrid prŽ sente ce dernier dans la notice biographique quÕil lui consacre dans ses Nujâm pourrait aller dans ce sens : « Toute la conduite dÕal-Malik al-ø¨hir [Baybars] nÕavait pour base que les usages des rois des Tatars, et les r glements (al-a½k¨m) de Gengis-khan, connus sous les noms de yasaq et târ¨216. » Mais le y¨s¨ a-t-il Ž tŽ effectivement imposŽ avec rigueur aux populations musulmanes placŽ es sous la domination mongole ? La mani re dont il fut appliquŽ en Chine sous les YŸan (r. 1271-1368) pourrait Ž clairer la situation dans les pays dÕIslam et permettre de formuler des hypoth ses sur son r™le rŽ el et la durŽ e de son in uence. Dans les sources chinoises, nous trouvons des documents qui font rŽ fŽ rence ˆ lÕapplication du y¨s¨. Ces citations ont Ž tŽ examinŽ es par Paul Heng-chaÕo ChÕen217. Il constate que la majoritŽ des rŽ fŽ rences au y¨s¨ concernent les intŽ r ts de lÕƒtat : jugements dÕaffaires militaires et gestion du syst me des relais postaux (yam), ainsi que des dŽ lits moins graves comme, par exemple, la pratique de la magie, la vente de poison, la contrefa on. LÕanalyse de ces citations montre que les peines ne sÕappliquaient aux Chinois que lorsque les infractions concernaient les affaires de lÕƒtat. Paul ChÕen note quÕˆ partir de 1280 les rŽ fŽ rences au y¨s¨ sont en dŽ clin, preuve quÕil est manifestement inadaptŽ ˆ la sociŽ tŽ sŽ dentaire chinoise : les YŸan se tournent alors vers la culture juridique chinoise. Faute de disposer pour les pays musulmans dÕune documentation aussi prŽ cise, il est cependant possible de proposer quelques hypoth ses sur lÕimpact rŽ el du y¨s¨ dans les pays dÕIslam. On peut supposer que les prescriptions touchant aux intŽ r ts de lÕƒtat, ˆ la discipline militaire et ˆ la gestion du syst me de contr™le dans lÕempire sÕimposaient ˆ tous les sujets, m me dans les territoires les plus Ž loignŽ s de la capitale, Qaraqorum. Il est peu probable, en revanche, que les r gles coutumi res aient Ž tŽ imposŽ es ˆ lÕensemble des populations soumises. Elles sont liŽ es, on lÕa vu, au syst me de reprŽ sentations du chamanisme. Al-Juvayn rapporte un certain nombre dÕanecdotes Ž clairantes sur lÕapplication des r gles coutumi res dans les pays dÕIslam. Il parle ainsi dÕun musulman qui, ayant achetŽ un mouton au marchŽ , revient chez lui, ferme bien toutes les portes de sa maison avant de lՎ gorger selon le rite islamique. Un Turc, qui lÕavait suivi, sÕintroduit brutalement chez ce musulman et le conduit devant …gšdei, sous le prŽ texte que lÕhomme nÕa pas respectŽ le y¨s¨. Apr s avoir rŽ  Ž chi, le qan dit : « Ce pauvre homme a respectŽ les commandements de notre y¨s¨ tandis que ce Turc

216 217

Nujâm, vol. VII : 182. Voir Paul Heng-chaÕo ChÕen, Chinese legal tradition under the Mongols : p. 4-8.

70

DENISE AIGLE

les a transgressŽ s218 », et cÕest le Turc qui est mis ˆ mort ˆ « cause de sa mauvaise nature (bad sÂrat) ». Le rŽ cit dÕal-Juvayn est marquŽ par son caract re Ž diŽ ant, mais il tŽ moigne de lÕindiffŽ rence des Mongols en mati re religieuse, soulignŽ e par toutes les sources, m me celles qui leur Ž taient hostiles. Un principe du y¨s¨ donnait dÕailleurs un statut Ž quivalent ˆ toutes les religions et visait ˆ protŽ ger tous les personnels religieux. En dÕautres termes, il Ž tait permis de faire ce que lÕon voulait chez soi du moment que cela ne portait pas atteinte ˆ lÕordre de lÕƒtat. Ë la cour des qan, en revanche, il fallait respecter ces interdits, notamment ceux qui Ž taient en rapport avec les esprits, redoutŽ s des Mongols. Le compagnon de Guillaume de Rubrouck manqua dՐ tre exŽ cutŽ parce quÕil avait heurtŽ le seuil de la tente de Mšngke. Il eut la vie sauve, mais il ne fut plus autorisŽ ˆ pŽ nŽ trer dans une demeure du qan219. En mati re de loi privŽ e, les peuples soumis Ž taient vraisemblablement rŽ gis par les principes locaux traditionnels, en territoire islamique par la shar®a : elle nÕa jamais Ž tŽ abolie par les pouvoirs mongols. Si les r gles coutumi res des Mongols avaient Ž tŽ imposŽ es, ˆ grande Ž chelle, dans lÕensemble des territoires musulmans tombŽ s sous leur domination, on en trouverait bien plus de traces dans les sources islamiques. Le y¨s¨ Ž tait bien, pour les Mongols, une affaire dÕƒtat. QuÕest devenu le y¨s¨ apr s la chute de lÕempire mongol ? Il est restŽ une rŽ fŽ rence, dans son acception large, mythique, pour les peuples turco-mongols dŽ sirant afŽ rmer leur rattachement ˆ Gengis-khan 220. Il fut manifestement revendiquŽ ˆ des Ž ns politiques par Tamerlan et par nombre de dynasties qui fond rent, en Asie centrale, leur lŽ gitimitŽ ˆ la fois sur lÕislam et la culture de la steppe 221 .

Al-JuvaynÂ, Ž d., vol. I : 163 ; trad., vol. I : 206. Rubrouck, XXIX, 28, Ž d. : 262 ; 37, Ž d. : 265 ; trad. Kappler : 177, 179 ; Jackson/Morgan : 193-194, 196. 220 Sur les dynasties qui en Asie centrale ont recherchŽ une lŽ gitimitŽ gengiskhanide, voir Denise Aigle, « Le mythe crŽ ateur dÕhistoire », in Figures mythiques des mondes musulmans, D. Aigle (dir.), REMMM, n¡ 89-90 (2000) : 7-38 ; « Figures mythiques et histoire. RŽ interprŽ tations et contrastes entre Orient et Occident », in Figures mythiques des mondes musulmans, D. Aigle (dir.), REMMM, n¡ 89-90 (2000) : 39-71, notamment la bibliographie prŽ sentŽ e dans ces articles. 221 Voir BŽ atrice Manz, « Tamerlane and the Symbolism of Sovereignty », Iranian Studies (1988) 21/1-2 : 105-122. Sur les enjeux du y¨s¨ dans les probl mes de successions de Tamerlan, voir BŽ atrice Manz, « TemŸr and the Problem of a ConquerorÕs Legacy », Journal of the Royal Asiatic Society (1998) 3/8 : 21-41 et John E. Woods, « TimurÕs Genealogy » : 85-125. 218 219

Rubrouck

Interdiction de sՎ tablir en dehors de lÕendroit assignŽ

R gle sur la prise du butin

Organisation Interdiction de de la garde sÕautoproclamer qan sous peine de mort

StratŽ gie militaire

Fuyards (ou tout lÕescadron) mis ˆ mort

Celui qui ne retourne pas au combat est dŽ capitŽ

CommencŽ en Mongolie en 1252 (persan)

Al-JuvaynÂ

Ne pas changer dÕunitŽ militaire sous peine de mort

Espionnage Faux puni de tŽ moignage mort (faux dÕun ambassadeurs) « informateur (aiqaq) » puni de mort

Milieu du XIIIe Mission entre Mission entre s. (mongol) 1245-1247 1253-1255 (latin) (latin)

Histoire secr te Plan Carpin Auteur mort en 1272 (armŽ nien)

Ganjakec¾i

Ne pas changer dÕunitŽ militaire sous peine de mort

CommencŽ ˆ Magh¨r¨ en 1276 (syriaque)

Bar Hebraeus

Faux tŽ moignage puni de mort

Auteur mort en 1335 (armŽ nien)

Akanc®i

Celui qui ne porte pas secours au compagnon dÕarmes est mis ˆ mort

Celui qui ne restitue pas un fuyard est mis ˆ mort

Interdiction de venir en aide au prisonnier sous peine de mort

Espionnage puni de mort

Auteur mort en 1349 (arabe)

Al-®UmarÂ

Liste des principaux y¨s¨ et yâsân rŽ pertoriŽ s dans les sources mŽ diŽ vales, et qui, non respectŽ s, sont le plus souvent sanctionnŽ s par la peine capitale

Celui qui ne porte pas secours au compagnon dÕarmes est mis ˆ mort

Celui qui ne restitue pas un fuyard est mis ˆ mort

Interdiction de venir en aide au prisonnier sous peine de mort

Espionnage puni de mort

Auteur mort en 1441-1442 (arabe)

Al-MaqrÂzÂ

RŽ digŽ en 1448 (arabe)

Ibn ®Arabsh¨h

LE GRAND JASAQ DE GENGIS-KHAN

71

Crime de l se majestŽ

R gle sur la prise du gibier ˆ la guerre

Interdiction du pillage

Le voleur est exŽ cutŽ

Trahison ˆ la guerre punie de 100 coups de b‰ton

Histoire secr te Plan Carpin

Table (cont.)

Homicide puni de mort

Le voleur est exŽ cutŽ pour un vol considŽ rable

Celui qui pratique la sorcellerie est exŽ cutŽ

Rubrouck

Mention du qi×¨× (peine compensatoire)

Il est interdit de mentir sous peine de mort

Mention de la banqueroute du commer ant, pas de sanction

Al-JuvaynÂ

Vol puni de la mort la plus terrible

Les femmes sÕoccupent de magie

Ganjakec¾i

Bar Hebraeus

Le voleur est exŽ cutŽ

Trahison punie de mort

Akanc®i

Celui qui pratique la magie est exŽ cutŽ

Al-MaqrÂzÂ

Il est interdit de mentir sous peine de mort

Le voleur est exŽ cutŽ

Il est interdit de mentir sous peine de mort

Le voleur est exŽ cutŽ

Le commer ant Le commer ant qui perd trois qui perd trois fois le capital fois le capital conŽ Ž est conŽ Ž est exŽ cutŽ exŽ cutŽ

Celui qui pratique la magie est exŽ cutŽ

Al-®UmarÂ

Le voleur doit  tre cruciŽ Ž

RŽ duire les hommes libres ˆ lÕesclavage

Ibn ®Arabsh¨h

72 DENISE AIGLE

Adult re puni de mort

Mention des interdits sur lÕeau (sans peine de mort)

Interdiction de fouler le seuil de la tente sous peine de mort

Adult re puni de mort

Mention des interdits sur lÕeau (sans peine de mort)

Interdiction de fouler le seuil de la tente sous peine de mort Interdiction dÕabattre les animaux selon le rite islamique

Mention des interdits sur lÕeau (sans peine de mort)

Adult re puni de mort

Ils ne commettent pas lÕadult re car il est vu comme une offense

Adult re puni de mort

Interdiction dÕabattre les animaux selon le rite islamique sous peine dՐ tre Ž gorgŽ

Interdiction dÕuriner dans lÕeau ou dans les cendres (peine capitale)

Interdiction dÕuriner dans lÕeau (peine capitale)

Interdiction dÕabattre les animaux selon le rite islamique sous peine dՐ tre Ž gorgŽ

Mention des interdits sur lÕeau (peine capitale)

Adult re puni de mort

Mention des interdits sur lÕeau (peine capitale)

Adult re puni de mort

Adult re puni de mort (strangulation)

LE GRAND JASAQ DE GENGIS-KHAN

73

Coutume dՎ pouser les veuves du p re, sauf sa propre m re

Histoire secr te Plan Carpin

Table (cont.)

Interdiction de sÕinterposer entre deux personnes sous peine de mort

Coutume honteuse dՎ pouser les veuves du p re, sauf sa propre m re

Rubrouck Le Ž ls Ž pouse les veuves de son p re, sauf sa propre m re

Al-JuvaynÂ

Ganjakec¾i

Bar Hebraeus

Akanc®i

Al-®UmarÂ

Al-MaqrÂzÂ

Les proches parents du mari peuvent Ž pouser ses femmes (pas de pŽ riode de ®idda)

Ibn ®Arabsh¨h

74 DENISE AIGLE

LE GRAND JASAQ DE GENGIS-KHAN

75

BIBLIOGRAPHIE SOURCES PRIMAIRES Sources mongoles Histoire secr te des Mongols. Chronique mongole du XIII e si cle, traduit du mongol, prŽ sentŽ et annotŽ par Marie-Dominique Even et Rodica Pop, prŽ face de Roberte N. Hamayon (Paris : Gallimard, 1994). The Secret History of the Mongols, trad. Francis W. Cleaves (Harvard University Press : Cambridge, Massachusetts, 1982). Les lettres de 1289 et 1305 des Ilkhan Argun et …ljeitŸ ˆ Philippe le Bel, Ž d. Antoine Mostaert et Francis Woodman Cleaves (Cambridge, Mass. : Harvard University Press, 1962). Rachewiltz, Igor de, Index to the Secret History of the Mongols (Bloomington : Indiana University, Uralic and Altaic Series, vol. 121, 1972). Sources persanes Al-Juvayn [= Al-JuvaynÂ], T¨rÂkh-i Jah¨ngush¨, Ž d. Mu½ammad QazvÂn (Leyde-Londres, 2 vol., 1912). [RŽ impr. de lՎ dition de M. QazvÂn ˆ TŽ hŽ ran en1379sh/ 2000.] Al-JuvaynÂ, T¨rÂkh-i Jah¨ngush¨, trad. John Andrew Boyle, The History of the World Conqueror (Manchester, 2 vol. 1958). Al-Jâzj¨n [= Al-Jâzj¨nÂ], Þabaq¨t-i N¨×irÂ, Ž d. W. Nassau Lees (Calcutta : Bibliotheca Indica, vol. XLIV, 1864). Al-Jâzj¨nÂ, Þabaq¨t-i N¨×irÂ, trad. H. G. Raverty, 2 vol. (Londres, 1881). RashÂd al-DÂn [= RashÂd al-DÂn], J¨mi® al-taw¨rÂkh, Ž d. A.A. Alizade (Baku, vol. III, 1957). RashÂd al-DÂn, J¨mi® al-taw¨rÂkh, Ž d. B. KarÂm (TehŽ ran, 2 vol., 1959). Al-Va×רf [= Al-Va×רf], Tajziyat al-amרr wa tazjiyat al-a®×¨r (Lith. Bombay, 1852). Sources arabes Ibn ®Arabsh¨h [= Ibn ®Arabsh¨h], Liber Arabicus sive Fructus imperatorum et Jocatio ingeniosorum [F¨kihat al-khulaf¨¾ wa muf¨kah¨t al-úuraf¨¾], Ž d. G. Freytag (Bonn, 1832). Ibn Taghr Bird [= al-Nujâm], Abâ l-Ma½¨sin, al-Nujâm al-z¨hira f muluk Mi×r wa l-Q¨hira, Le Caire, 1930-1956, vol. VII. Al-KutubÂ, Waf¨t al-wafay¨t, Ž d. I½s¨n ®Abb¨s (Beyrouth, 1973, vol. I). Al-MaqrÂz [= Al-MaqrÂzÂ], al-Maw¨®iú wa l-i®tib¨r f” dhikr al-khi×¨× wa l-¨th¨r, Ž d. Boulaq, 1270h. [RŽ impr. Beyrouth ca. 1970.] Al-®Umar [= Al-®UmarÂ], Das Mongolische Weltreich : al-®UmarÂÕs Darstellung der mongolischen Reiche in seinem Werk Mas¨lik al-abרr wa mam¨lik al-amרr, Ž d. Klaus Lech (Wiesbaden : Asiatischen Forschungen, vol. 14, 1968). Sources syriaques Bar Hebraeus [= Bar Hebraeus], Maktebonž th Zabhn , Ž d. Paul Bedjan (Paris : Maisonneuve, 1890). Bar Hebraeus, Maktebonž th Zabhn , trad. Ernest A. Wallis Budge, The Chronography of Gregory Abž Õl-Faraj, 1225-1286, 2 vol., I : traduction anglaise ; vol. II : reproduction texte syriaque, facsimilŽ (Londres, 1932). [RŽ impr., Amsterdam, 1976.] Sources armŽ niennes Akanc®i [= Akanc®i], History of the Nation of the Archers [T®at®arac® Patmut®iwnk®], Ž d. et trad. Robert P. Blake et Richard N. Frye, Harvard Journal of Asiatic Studies, vol. 12/3-4 (1949) : 269443.

76

DENISE AIGLE

Ganjakec®i [= Ganjakec®i], [Patmut®iwnk® Hayoc®], trad. partielle John A. Boyle, « Kirakos of Ganjak on the Mongols », Central Asiatic Journal 8, (1963) : 200-211. [RŽ impr. dans John Andrew Boyle, The Mongol World Empire 1206-1370, Londres, Variorum Reprint, 1977.] Sources latines Guillaume de Rubrouck, Voyage dans lÕempire mongol, trad. Claude et RenŽ Kappler (Paris : Payot, 1985). Guillemus de Rubruc [= Rubrouck], Itinerarium, in Sinica Franciscana, vol. I, Ž d. P. Anastasius Van den Wyngaert (Quarrachi-Firenze : Apud Collegium S. Bonaventurae, 1929), 164-332. Iohannes de Plano Carpini [= Carpin], Ystoria Mongalorum, in Sinica Franciscana, vol. I, Ž d. P. Anastasius Van den Wyngaert (Quarrachi-Firenze : Apud Collegium S. Bonaventurae, 1929), 27-143. Jean de Plan Carpin, Histoire des Mongols, trad. Dom Jean Becquet et Louis Hambis (Paris : Adrien Maisonneuve, 1965). Lupprian, K.E., Die Beziehungen der PŠpste zu islamischen und mongolischen Herrschen im 13. Jahrhundert (Rome : Studi et Testi 291, 1981). Meyvaert, Paul, « An Unknown Letter of Hulagu, Il-Khan of Persia, to King Louis IX of France », Viator 11 (1980) : 245-259. The Mission of Friar William of Rubruck. His Journey to the Court of the Great Khan Mšngke 12531255, trad. Peter Jackson et David Morgan (Londres : Hakluyt Society, 1990). Sources en ancien fran ais Marco Polo, Le Devisement du monde, Ž d. sous la direction de Philippe MŽ nard, t. II TraversŽ e de lÕAfghanistan et entrŽ e en Chine, Ž d. Jeanne-Marie Boivin, Laurenced Harf-Lancner et Laurence Mathey-Maille (Paris : Droz, 2003). ƒtudes Aigle, Denise, « Le mythe crŽ ateur dÕhistoire », in Figures mythiques des mondes musulmans, dir. D. Aigle, REMMM, 89-90 (2000) : 7-38. — — , « Figures mythiques et histoire. RŽ interprŽtations et contrastes entre Orient et Occident », in Figures mythiques des mondes musulmans, dir. D. Aigle, REMMM, 89-90 (2000) : 39-71. Allen, Frank J., « Varieties of Islamization in Inner Asia. The case of the Baraba Tatars, 17401917 », in En islam sibŽ rien, Ž d. StŽ phane Dudoignon, Cahiers du Monde russe, 41/2-3 (2000) : 245-262. — — , « Historical Legends of the Volga-Ural Muslims concerning Alexander the Great, the City of Yelabuga and B‰chm‰n Kh‰n », in Figures mythiques des mondes musulmans, dir. D. Aigle, REMMM, 89-90 (2000) : 89-107. Allsen, Thomas T., « Guard and Government in the Reign of the Grand Qan Mšngke 1251-59 », Harvard Journal of Asiatic Studies, 46/2 (1986) : 495-521. — — , Mongol Imperialism: the Policies of the Grand Qan Mšngke in China, Russia and the Islamics Lands (Berkley : University of California, 1987). — — , « Mongolian Princes and Their Merchant Partners, 1200-1260 », Asia Major, 2/2 (1989) : 83-126. — — , « The Rise of the Mongolian Empire and Mongolian Rule in North China », in The Cambridge History of China. vol. 6, Alien regimes and border states, 907-1368, Ž d. Herbert Franke et Denis Twitchett, p. 321-413. (Cambridge : Cambridge University Press, 1994). — — , « Sharing out the Empire : Apportioned Lands under the Mongols », in Nomads in the Sedentary World, Ž d. Anatoly M. Kazanov et AndrŽ Wink, p. 172-190. (Curzon : Curzon-IIAS Asian Studies Series, 2001).

LE GRAND JASAQ DE GENGIS-KHAN

77

— , Culture and Conquest in Mongol Eurasia (Cambridge : Cambridge University Press, 2001). Amitai-Preiss, Reuven, « In the Aftermath of ®Ayn J¨lât : The Beginnings of the Mamlâk-Álkh¨nid Cold War », al-Mas¨q 10 (1990) : 1-21. — — , « An Exchange of Letters in Arabic between Abaga Álkh¨n and Sultan Baybars (A.H. 667/A.D. 1268-69) », Central Asiatic Journal 38/1 (1994) : 11-33. — — , « Evidence for the Early Use of the Title Âlkh¨n among the Mongols », Journal of the Royal Asiatic Society (1991) I/3 : 353-361. — — , Mongols and Mamluks. The Mamluk-Álkh¨nid War, 1260-1281 (Cambridge : Cambridge University Press, 1995). — — , « Ghazan, Islam and Mongol Tradition : A View From the Mamlâk Sultanate », Bulletin of the School of Oriental and African Studies LIX/1 (1996) : 1-10. — — , « Mongol Imperial Ideology and the Ilkhanid War against the Mamluks », in The Mongol Empire & its Legacy, Ž d. Amitai-Press, Reuven & David Morgan, p. 57-72. (Leyde : Brill, 1999). — — , « Turko-Mongolian Nomads and the Iqߨ® System in the Islamic Middle East (ca 1000-1400 A.D.) », in Nomads in the Sedentary World, Ž d. Anatoly M. Kazanov et AndrŽ Wink, p. 152-171. (Curzon : Curzon-IIAS Asian Studies Series, 2001). — — , « ®Ayn J¨lât Revisited », Tarih (1992) 2 : 119-150. Anderson, J.N.D., « Homicide in Islamic Law », Bulletin of the School of Oriental and African Studies XIII/4 (1951) : 811-828. Aubin, Fran oise, « Mongolie (histoire) », Encyclopaedia Universalis, vol. 11, p. 656-684, Paris, 1974 [3¡ Ž d. en 1989]. — — , « Les sanctions et les peines chez les Mongols », in La peine. Punishment, p. 242-293. (Bruxelles : De Boeck UniversitŽ , 1991). Ayalon, David, « The Great Y¨sa of Chingiz Kh¨n, A Re-examination. Preface », Studia Islamica 33 (1971) : 97-140 (part A, The Basic Data in the Islamic Sources on the Y¨sa and on its Contents). [RŽ impr. in Outsiders in the Lands of Islam : Mamluks, Mongols and Eunuchs, Variorum Reprints, Londres, 1988.] — — , « The Great Y¨sa of Chingiz Kh¨n, A Re-examination. Preface », Studia Islamica 38 (1973) : 107-156 (part C2, al-MaqrÂzÂÕPassage on the Y¨sa under the Mamlâks). [RŽ impr. in Outsiders in the Lands of Islam : Mamluks, Mongols and Eunuchs, Variorum Reprints, Londres, 1988.] BarŽ eld, Thomas J., The Perilous Frontier. Nomadic Empires and China, 221 BC to AD 1757 (Cambridge Mass. & Oxford UK : Blackwell, 1989). — — , « Steppe Empires, China and the Silk Route : Nomads as a Force in International Trade and Policies », in Nomads in the Sedentary World, Ž d. Anatoly M. Kazanov et AndrŽ Wink, p. 234248. (Curzon : Curzon-IIAS Asian Studies Series, 2001). Basilov, Vladimir N., Samanstvo u narodov Srednej Azii i Kazaxtana (Moscou : Nauka, 1992). Bawden, C. R., The Modern History of Mongolia (Londres et NewYork : Kegan Paul International, 1968). Beffa, Marie-Lise, « Le concept de tŠnggŠri, « ciel », dans lÕHistoire secr te des Mongols », ƒtudes mongoles et sibŽ riennes 24 (1993) : 215-236. Bira, Shira, « Historiography among the Mongols », in History of Civilisations of Central Asia, vol. IV, part 2, The age of the achievement : A.D. 750 to the end of the Ž fteenth century, p. 156-160. (Paris : Unesco Publishing, 2000). Boyle, John Andrew, « Turkish and Mongol Shamanism in the Middle Ages », Folklore 83 (1972) : 177-193. [RŽ impr. in The Mongol World Empire 1206-1370, Variorum Reprints, Londres, 1977.] Broadbridge, Anne F., « Mamluk Legitimacy and Mongols : The Reigns of Baybars and Qal¨wân », Mamluk Studies Review 5 (2001) : 91-118. ChÕen Heng-chao, Paul, Chinese legal tradition under the Mongols. The code of 1291 as reconstructed (Princeton : Princeton University Press, 1979). Cleaves, Francis W., « The Memorial for Presenting the Yuan shih », Asia Major 1 (1988) : 59-69. —

78

DENISE AIGLE

Chrestomathie arabe = Sacy, Silvestre de, Chrestomathie arabe ou Extrait de divers Ž crivains arabes, tant en prose quÕen vers, tome II, p. 157-190. (Paris : Imprimerie royale, 1826). [voir Sacy, Silvestre]. DeWeese, Devin, Islamization and Native Religion in the Golden Horde (The Pennsylvania State University Press : University Park, 1994). Doerfer, Gerhard [= Doerfer], TŸrkische und mongolische Elemente in Neupersischen, 4 vol. (Wiesbaden : Franz Steiner Verlag, 1963-1975). Endicott-West, Elizabeth, Mongolian Rule in China. Local Administration in the Yuan Dynasty (Cambridge Mass. : Harvard University Press, 1989). — — , « Merchant Associations in YŸan China : The Ortoq », Asia Major 2/2 (1989) : 127-154. — — , « Notes on Shamans, Fortune-Tellers and Yin-Yang Practitioners and Civil Administration in YŸan China », in The Mongol Empire & its Legacy, Ž d. Reuven Amitai-Press & David Morgan, p. 224-237. (Leyde : Brill, 1999). Fletcher, Joseph F., « The Mongols : Ecological and Social Perspectives », Harvard Journal of Asiatic Studies 46/1 (1986) : 1-35. Foltz, Richard, « Ecumenical Mischief Under the Mongols », Central Asiatic Journal 43/1 (1999) : 42-69. Franke, Herbert, « The forest peoples of Manchuria : Kitans and Jurchens », in The Cambridge History of the Early Inner Asia, Ž d. Denis Sinor, p. 400-423. (Cambridge : Cambridge University Press, 1990). Garrone, Patrick, Chamanisme et islam en Asie centrale (Paris : Jean Maisonneuve, 2000). Golden, Peter B., « Imperial ideology and the sources of political unity amongst the pre-cinggisid nomads of the western Eurasia », Archivum Eurasiae Medii Aevi 2 (1981) : 37-76. — — , « Religion among the Qipcaqs of Medieval Eurasia », Central Asiatic Journal 42/2 (1998) : 180-237. Hamayon, Roberte, « MŽ rite de lÕoffensŽ vengeur, plaisir du rival vainqueur. Le mouvement ascendant des Ž changes hostiles dans deux sociŽ tŽ s mongoles », in La Vengeance. ƒtudes dÕethnologie, dÕhistoire et de philosophie, vol. II, Vengeance et pouvoir dans quelques sociŽ tŽ s extraoccidentales, Ž d. Raymond Verdier, p. 107-140. (Paris : Ed. Cujas, 1980). — — , « Pragmatisme et ritualisation dans le chamanisme », in Essais sur le rituel, II, dir. AnneMarie Blondeau et Kristofer Shipper, p. 149-169. (Peters, Louvain-Paris, 1988). — — , La chasse ˆ lÕ‰me. Esquisse dÕune thŽ orie du chamanisme sibŽ rien (Nanterre : SociŽ tŽ dÕethnologie, 1990). — — , « Chamanisme, bouddhisme, hŽ ro•sme Ž pique : quel support dÕidentitŽ pour les Bouriates post-soviŽtiques ? », ƒtudes mongoles et sibŽ riennes 27 (1996) : 327-355. Haider, Mansura, « The Mongolian Traditions and their Survival in Central Asia (XIV-XV Centuries) », Central Asiatic Journal 28/1-2 (1984) : 57-79. Heissig, W., The Religions of Mongolia (Londres, 1980). Hung, William, « The Transmission of the Book Known as The Secret History of the Mongols », Harvard Journal of Asiatic Studies 14 (1951) : 433-492. Irwin, Robert G., « What the Partidge Told the Eagle : A Neglected Arabic source on Chinggis Khan and the Early History of the Mongols », in The Mongol Empire & its Legacy, Ž d. Reuven Amitai-Press & David Morgan, p. 5-11. (Leyde : Brill, 1999). Jackson, Peter, « William of Rubruck : A Review Article », Journal of the Royal Asiatic Society (1987) : 92-97. — — , « From Ulus to Khanate : The Making of the Mongol States, c. 1220-c. 1290 », in The Mongol Empire & its Legacy, Ž d. Reuven Amitai-Press & David Morgan, p. 12-38. (Leyde : Brill, 1999). — — , « The Mongol Empire, 1986-1999 », Journal of Medieval History (2002) 26/2 : 189-210. Jagchid, Sechin et Heyer, Paul, MongoliaÕs Society and Culture (Boulder : Folkestone, 1979). Lambton, Ann K.S., Continuity and Change in Medieval Persia (Londres : Tauris, 1988). Lessing, Ferdinand D. [= Lessing], Mongolian-English Dictionary (Berkeley/Los Angeles : University of California Press, 1960).

LE GRAND JASAQ DE GENGIS-KHAN

79

Ma®dankan, Ma®×âme, Bi-y¨s¨ rasÂdag¨n (TŽ hŽ ran : Iran University Press, 1996). — — , « Y¨s¨ dar ®ahd-i Álkh¨ni¨n », Ma®¨ref 9/1 (1371sh) : 3-14. Manz, BŽ atrice, « Tamerlane and the Symbolism of Sovereignty », Iranian Studies 21/1-2 (1988) : 105-122. — — , « TemŸr and the Problem of a ConquerorÕs Legacy », Journal of the Royal Asiatic Society 3/8 (1998) : 21-41. Morgan, David O., « The Mongol Empire : A Review Article », Bulletin of the School of Oriental and African Studies 44 (1981) : 121-125. — — , The Mongols (Oxford : Blackwall, 1986). — — , « The ®Great Y¨s¨ of Chingiz Kh¨nÕ and Mongol Law in the Álkh¨nate », Bulletin of the School of Oriental and African Studies 49/1 (1986) : 163-176. Poliak, A. N., « The in uence of Chingiz-Kh¨nÕs Y¨sa upon the General Organization of the Mamlâk State », Bulletin of the School of Oriental and African Studies 10/4 (1942) : 862-876. Pop, Rodica, Le mariage chez les Mongols. Rites et textes. (Th se de doctorat EPHE, sous la direction de R. N. Hamayon, Sorbonne, Paris, 2001). Rachewiltz, Igor de, « Some Remarks on the Dating of The Secret History of the Mongols », Monumenta Serica XXIV (1965) : 185-205. — — , « Some Remarks on the Ideological Foundation of Chinggis khanÕs empire », Paper on Far Eastern History, vol. 7 (1973) : 21-36. — — , « Some Re ections on Cinggis QanÕs jasag », East Asian History 6 (1993) : 91-104. Ratchnevsky, Paul, « Sigi-Qutuqu, ein mongolische Gefolgsmann im 12.-13. Jahrhundert », Central Asiatic Journal X/2 (1965) : 87-120. — — , « Die Yasa (Jasaq) Cinggis-khans und ihre Problematik », Schriften zur Geschichte und Kultur des alten Orients 5 : Sprache, Geschichte und Kultur der altaischen Všlker (Berlin : Akademie-Verlag, 1974), 471-487. — — , Cinggis-khan sein Leben und Wirken (Wiesbaden : Franz Steiner Verlag, 1983). [Trad. anglaise par Thomas N. Haining, Gengis Khan. His Life and Legacy (Oxford & Cambridge Mass. : Blackwell, 1991]. — — , « Die RechtsverhŠltnisse bei den Mongolen im 12.-13. Jahrhundert », Central Asiatic Journal 31 (1987) : 64-111. — — , « Jurisdiction, penal Code, and cultural Confrontation under Mongol-YŸan Law », Asia Major 6 (1993) : 161-179. Riasanovsky, Valentin A., Fundamental Principles of Mongol Law (Bloomington : Indiana University Publications : Uralic and altaic Series, vol. 43, 1937). [2e Ž d. 1965.] Ricci = Dictionnaire fran ais de la langue chinoise (Institut Ricci : Kuangchi Press, 1994). Sacy, Silvestre de, Chrestomathie arabe ou Extrait de divers Ž crivains arabes, tant en prose quÕen vers, tome II, p. 157-190. (Paris : Imprimerie royale, 1826). [voir Chrestomathie arabe]. Schacht, Joseph, Introduction au droit musulman (Paris : Maisonneuve & Larose, 1983). [Premi re Ž dition : An Introduction to the Islamic Law, (Oxford : Oxford University Press, 1964)]. Smith, John M., « The Mongols and World-Conquest », Mongolica (1994) 5 : 206-214. Thesaurus Syriacus [= Thesaurus Syriacus], Ž d. R. Payne Smith (Hidelsheim/New York : Georg Olms Verlag, 1981) [1re Ž d., 1879-1901.] Togan, Isenbike, Flexibility and Limitation in Steppe Formations : the Kerait Khanat and Chinggis Khan (Leyde : Brill, 1998). Vladimirtsov, Boris, Le rŽ gime social des Mongols. Le fŽ odalisme nomade (Paris : Adrien Maisonneuve, 1948). Vernadsky, G., « JuwainiÕs Version of Chingis KhanÕs yasa », Annales de lÕInstitut Kondakov, XI (1940) : 33-45. Woods, John E., « TimurÕs Genealogy », in Intellectual Studies on Islam. Essays Written in Honor of Martin Dickson, Ž d. M. Mazzaoui et V. Moreen (University of Utah Press, 1990) : 85-125.

THE SHADOW SULTAN: SUCCESSION AND IMPOSTURE IN THE MUGHAL EMPIRE, 1628-1640 BY

JORGE FLORES AND SANJAY SUBRAHMANYAM* Abstract This essay explores the problem of imposture in the Mughal empire, through the case of Sultan Dawar Bakhsh, or Bulaqi, who ruled brie y in the late 1620s. Though ofŽ cial Mughal histories had it that he was executed in January 1628 along with several other princes, various persons claiming his identity surfaced, Ž rst in India and then in Iran. We examine the views of Mughal, Portuguese, Iranian and other sources on these claimants, and also explore what forms of proof were sought by different early modern agents in order to satisfy themselves of the identity of a returning prince. Cette contribution examine le problme de lÕimposture dans lÕEmpire moghol en Ž tudiant le cas du Sultan Dawar Bakhsh ou Bulaqi, qui a rŽ gnŽ pendant quelques mois en 1627-28. Selon les chroniques mogholes de lՎ poque, Bulaqi aurait Ž tŽ exŽ cutŽ en janvier 1628 avec plusieurs autres princes. Mais lÕon sait que pendant la dŽ cennie suivante, plusieurs personnages se sont manifestŽ s, tout dÕabord en Inde et ensuite en Iran, prŽ tendant  tre le sultan disparu. En croisant les informations fournies par les textes et des documents dÕarchives assez variŽ s, en provenance de lÕEtat portugais des Indes, de lÕEmpire moghol et de lÕEtat safavide, lÕanalyse suit pas ˆ pas le parcours de ce Martin Guerre moghol pour apprŽ cier les preuves apportŽ es sur son identitŽ. Keywords: Mughal empire, imposture, succession, Portuguese, Safavids One of these men is Genius to the other; And so of these: which is the natural man, And which the spirit? Who deciphers them?

—Shakespeare, The Comedy of Errors, Act V, Scene 1.

* Jorge Flores, Centro de Estudos Asi‡ticos, Universidade de Aveiro, Campus de Santiago, 3810-193 Aveiro, Portugal, jm [email protected]; Sanjay Subrahmanyam, Oriental Institute, University of Oxford, Pusey Lane, Oxford OX1 2LE, UK, [email protected] We are grateful to Muzaffar Alam, Saqib Baburi, Shailendra Bhandare and Caroline Ford, to the referees of JESHO, and to a seminar audience at the SOAS, London, for their valuable suggestions and comments while writing and revising this essay. © Koninklijke Brill NV, Leiden, 2004 Also available online – www.brill.nl

JESHO 47,1

THE SHADOW SULTAN

81

Early modern historiography has in the past decades devoted considerable attention to emerging notions of personal identity, which in turn are linked to theories of the historical trajectory of the idea of the ÔindividualÕ in various cultures. To be sure, some elements of this problem can be traced back to medieval (or even older) re ections on self-knowledge, and it has even been argued from this perspective that Òthe discovery of the individual was one of the most important cultural developments in the years between 1050 and 1200Ó in western Europe.1 The issue has also been posed from a rather different angle, namely that of the speciŽ city of the individual, which in turn takes us to the very widespread theme of the ÔimpostorÕ or Ôdouble,Õ of which an outstanding example may be found in Akira KurosawaÕs 1980 Ž lm Kagemusha (The Shadow Warrior) (cf. Desser 1983). The Ž lm is situated in JapanÕs sixteenth-century warring states period, and narrates how a condemned thief is saved from execution by agreeing to serve as a double for the powerful warlord Takeda Shingen (1521-73) of Kai. In this case, the key elements are somewhat particular: the ÔdoubleÕ is in a sense authorised by the original to substitute for him, and the tensions revolve largely around the difference in the social status of the two. But a series of other questions also emerge from such materials. Can one person somehow ÔbecomeÕ another? Can one think of historical cultures in which personal identity was effectively subordinated to social identity, and, by an extension of the same logic, to social function? Can a person simply be produced as a social actor, in the sense that some have argued that authors are produced? 2 What notions of proof of identity do different historical cultures demand, and can we assume that we are moving historically from more credulous cultures to others where ever more rigorous ideas of proof are being set down by an intrusive state apparatus? India has had its fair share of Tichborne Claimants and Martin Guerres who lend themselves to this type of re ection, of which one of the most celebrated modern cases is that of the so-called Kumar of Bhawal in Bengal, that has recently been studied in considerable detail by Partha Chatterjee (2002). Though the Bhawal case, which was brought to trial in the 1920s and 1930s, is unusual for the fact that a considerable amount of forensic material was brought to bear on the matter, Chatterjee and others have noted that it also followed on a longer Indian trend, of which a notable earlier Ž gure is that of the so-called ÔFalse Pratapchand,Õ an instance referring to a claimant to one of the zam”nd‰r” estates in Bengal in the Ž rst half of the nineteenth century (McLane 1993: 316-22; cf.

1

Morris 1972: 158; also, more recently, the related re ection in Gurevich 1995. Foucault 1987: 124-42; and the further discussion in Grafton 1990, and Chartier 1994: 25-60. 2

82

JORGE FLORES AND SANJAY SUBRAHMANYAM

Bhadra 2003). However, if we cast our glance back as far as the sixteenth or seventeenth centuries, we can already Ž nd such instances, of which one wellknown case concerns Shah Shuja®, the son of the Mughal emperor Shahjahan who was apparently killed in Arakan in the 1660s, but who turned up periodically in quite diverse parts of the Mughal domains. ChatterjeeÕs purpose in his work is complex, but one of his intentions is to juxtapose Òmodern Western philosophical discussion on identity,Ó with what he terms ÒIndian philosophical discussions on identity and recognition,Ó arguing that these produced quite different notions of both personhood and proof (cf. Chatterjee 2002: 115-37). An examination of the situation in the seventeenth century may demonstrate that this opposition was by no means quite so radical, in particular if one moves from the realm of philosophical re ection to the methods actually used by states to separate the sheep from the goats, as it were. The question of the ÔimpostorÕ may of course exist at various levels in a society: the long-absent merchant, the renouncer who returns after a long interval, and so on. Yet the issue takes on a particular charge when the personage in question, like the Kumar of Bhawal, has a function in guaranteeing social order by virtue of his or her association with royal power. Now the theme of the royal ÔdoubleÕ is a powerful one that potentially possesses the capacity to transcend cultural divisions, but nevertheless seems more anchored in some cultures than in others. The central idea is quite simple. A royal Ž gure, who could be a reigning or former monarch or is at any rate close to the throne, possesses an exact double, who resembles him or her perfectly—and is either a perfect stranger or an identical twin. In a variant, the ÔdoubleÕ is in fact an impostor, a false claimant who will either be unmasked for what he or she is, or again will manage successfully to impose this false claim on a credulous population. The Ž rst of these broad lines of development gives us popular literary themes such as that of Louis XIVÕs identical twin, developed by Alexandre Dumas (or his own ghostwriter-doubles) in The Man in the Iron Mask. The most recent literary version, albeit not concerned with a royal double, is the work entitled O homem duplicado by the Nobel Prize winning novelist JosŽ Saramago, who adds a further twist by making his ÔheroÕ a professor of history (Saramago 2002). In an extensive post-Dumas sub-literature, we already have examples such as Anthony HopeÕs late nineteenth-century work, The Prisoner of Zenda, to which Saramago incidentally does refer. The second line is far more fecund from a historical point of view. For instance, we know that in Tsarist Russia, a vast number of claimants purporting to be this or that prince existed from the sixteenth century on, and that this tradition even continued after the Russian Revolution, with several persons claiming to be Aleksei or Anastasia, or other members of the TsarÕs family who had survived the mass execution by the Bolsheviks

THE SHADOW SULTAN

83

(Szv‡k 2000). Again, in England in the late Ž fteenth century, the nascent Tudor dynasty had to face a challenge in the form of various rival claimants, who stated that they descended from the Yorkist kings. One of these claimants, Lambert Simnel, was dealt with relatively easily by Henry VII, but the other, Perkin Warbeck, posed a rather more major threat and was hence eventually hanged for treason in 1499 after mounting several unsuccessful attacks and rebellions with the aid of a number of European monarchs who recognised his claims.3 What makes the case of Warbeck particularly curious is his Portuguese ÔconnectionÕ; for though he was probably born in the Flanders, he had spent time in Portugal, and one of his major supporters was a Jewish adventurer of Portuguese origin, Duarte Brand‹o or Edward Brampton (Aubin 2000). Closer geographically to the case we shall examine presently, we have an important incident in late sixteenth-century Iran, namely of a qalandar called Muzawwar who appeared in 1580, claiming to be the recently deceased monarch Shah Isma®il II. This man had managed at a certain stage to gather together as many as twenty thousand supporters, and to defy the Safavid regional governors, until his movement was brutally repressed (Savory 1971). A less well-known instance that is relevant to the subject of the present essay is narrated by the Portuguese ofŽ cial chronicler Diogo do Couto, in his account of the Sultanate of Gujarat in the sixteenth century, and draws upon a number of other sources. In this view, the life of Sultan Bahadur of Gujarat (r. 1526-1537) followed a certain familiar pattern. Couto writes: [The Sultan] Modafar (. . .) had many sons, and the Ž rst-born was called Bador, and not Badur, as the chronicles of India call him. When he was still a boy, either because the astrologers told his father that that kingdom would be lost in the hands of his oldest son, or (as seems more likely) because he wanted to give it over to a younger son for whom he had more affection, it seems that he [Modafar] became ill-disposed towards Bador, who—either because he knew of this or because he feared that his father would have him killed—escaped bodily and went around that Indust‹o as a pilgrim, or what is called a calandar, and in this fashion wandered many years, learning different languages, observing and noting new rites and customs, and many new and fugitive things. (Couto 1999, i: 55).

So we have the theme of the royal exile, who becomes a renouncer (Persian qalandar), and wanders about incognito. Couto then develops the logically necessary second part of this theme as follows. While he was going about in this manner, his father died, and the brother to whom the father wanted to give the kingdom succeeded, but this lasted only a short while, and he

3

Arthurson 1994; also, more recently, the popular work by Wroe 2003.

84

JORGE FLORES AND SANJAY SUBRAHMANYAM

left the kingdom to another younger brother, but we do not know their names. When news of the death of Modafar travelled through that Indust‹o, it reached Bador, who at once returned in order to come and demand his kingdom, and—so they say—still in the clothes of a calandar entered the court of Amadaba, where his mother was still alive, as well as her son the king, who was still a boy, and without identifying himself to anyone, he [Bahadur] went in to see his mother and revealed himself to her, asking her that she should give the necessary orders for him to have his kingdom. The queen (and it may well be that she was not his mother but his stepmother, and that the Guzarates who recount this got it mixed up) told him that he should leave, for if his brother were to Ž nd out about him, he would have him killed, from which we may infer that she was the mother of the king and not of Bador. He, on being disillusioned, went to the king of Mandou, and revealed himself to him (se lhe descobrio), asking him for help and favour in order to recover his kingdom. And since, with the great people of the world, there is nothing that moves them so much as the miseries of an exiled prince, he promised him every favour and even that he would go in person to help him: and to this end, he asked some of the neighbouring kings that they should join him with great might.

Sultan Bahadur thus recovers his throne either with the help of the ruler of Mandu, or (in a variant that Couto also recounts) through the help of the Queen of Chitor. However, once he is on the throne, Òas he was evil, cruel and weakÓ he loses no time in attacking those who have helped him. The origins of a part of this narrative, in which Bahadur wanders in exile, may already be found in an anonymous Portuguese text of the mid-1530s, that describes how Òthe said Bador became a calamdar, going about barefoot and dressed in skinsÓ (Subrahmanyam 1998). Couto, who clearly had a fondness for this theme, subsequently uses it somewhat indiscriminately in his text, in relation for example to an ancestor of the Mughals whom he seems to have largely invented. In his view, then, one of the sons of Timur called ®Umar Shaikh (ÔHaomarxacÕ) also had a similar career. Couto thus writes of how in the course of a political sibling rivalry in Ž fteenth-century Central Asia, ÒHaomarxac, who is called Balobo by some, was a boy (. . .) and was left with nothing, because the two other brothers took hold of everything they could.Ó Faced with these circumstances, the inevitable took place: ÒThis Haomarxac, or Balobo, on seeing himself disinherited, decided to serve Mafamede and leaving those states that had belonged to his father in the clothes of a calandar (which is to say a pilgrim) went off in the direction of India, and traversed all of Industan, and eventually halted in the kingdom of Deli, where he decided to stayÓ (Couto 1999, i: 544-45). Couto then goes on to describe how he became popular amongst the other calandares there on account of the Òfame of his life and religion, which was wonderful;Ó he thus built up this following into a sizeable army, and eventually attacked and killed the king of Delhi, seizing power there himself and creating a sizeable domain through a series of conquests. Couto even tells us that this Balobo Òwho entitled himself king of the MagoresÓ was the founder of Mughal rule in India, and

THE SHADOW SULTAN

85

the grandfather of Babur, who eventually came to rule in the sixteenth century. The manifest inaccuracy of CoutoÕs account, which confuses Mughal history with that of the Afghan ruler Bahlul Lodi (almost certainly the ÔBaloboÕ of the account) need not concern us here; what is of interest is that the theme was clearly a common and widespread one and that Couto made use of it while drawing upon a vein of ÔfolkloreÕ in his chronicle.4 It is clear then that by the sixteenth century, a tradition already existed in northern India of wandering princes in exile who might make claims to a throne while appearing ostensibly to be mere calandares. Whether these claimants were true or false is not a theme that concerns Couto much, for he assumes that in general they were authentic—whether with Sultan Bahadur, or with his account of Balobo. But in the case that we shall consider below, matters appear far more complex, and lead us in the direction of a different variant, namely one where ÔfalseÕ and ÔtrueÕ claimants appear in the same political panorama. MUGHAL SUCCESSION AND THE C ASE OF B ULAQI Mughal succession had been much contested in the early phase of the dynasty, which had established itself in India in the 1520s. At the death of the Ž rst emperor, Zahir-ud-Din Muhammad Babur, his sons fought for years over the succession, with the designated heir Humayun being forced to contend time and again with his brothers Hindal, ®Askari, and especially Kamran, who operated from a base in Kabul. Eventually, Humayun was able to put an end to this threat only by blinding Kamran. However, if we return to the moment of BaburÕs death, we note that some sources suggest that this fact was initially kept secret for a certain time in order to avoid the social turmoil on account of which such moments were particularly feared. The account of BaburÕs own daughter, Gulbadan Begam, in her Hum‰yž n N‰ma, makes this clear enough; she sugges ts that only after the elapse of some time was it eventually announced that the emperor had Òbecome a dervishÓ (Hazrat B‰bur B‰dsh‰h darw”sh shudand), and that he had left his domains to his oldest son, Humayun (Gul-Badan Begam 1974: 109; text, p. 25). On HumayunÕs own death a quartercentury later, in 1556, matters proved somewhat complicated as well. The designated heir was Jalal-ud-Din Muhammad Akbar, but he too had a half-brother, Mirza Muhammad Hakim, who though an infant at the time also had his own

4

Compare this with the account of ÔBallu,Õ as Bahlul Lodi is termed, in the sixteenth-century text of Rizqullah Mushtaqi 1993: 1-15; for the Persian text, see Rizqullah Mushtaqi 2002: 3-16.

86

JORGE FLORES AND SANJAY SUBRAHMANYAM

partisans. The Ottoman admiral, Seyyidi ®Ali Reis, who was present at the time of HumayunÕs death in Delhi (in January 1556), claims that a ruse had to be used in order to conceal the emperorÕs death for a certain time, while Akbar returned to the capital. He writes: Amongst those close to the emperor [Humayun], there was a person called Munla Bikesi who had a certain resemblance to him, but who was somewhat smaller. Finally, on Tuesday, he was installed on the throne that had been placed under an arch that overlooked the river, he was dressed up in the emperorÕs clothes, his face and eye were covered in bandages, Khushhal Beg stayed close to him and the Mir Munshi directly in front. All the Sultans, Mirzas, the subjects and the common people arrived, saw the emperor from the bank of the river and made prayers for him. Joyful music was played, the physician was given a robe of honour, and it was asserted that the emperor was in good health.5

On AkbarÕs eventual return, this imposture could Ž nally be dispensed with, and the death of the emperor was at last announced to the public. AkbarÕs own death, a half-century later, was attended by somewhat less subterfuge. His successor Nur-ud-Din Muhammad Jahangir had seen his brothers and major potential rivals, Daniyal and Murad, die in quick succession, and his only serious rival was his own son, Khusrau, whose ambitions were initially suppressed by deft political alliance-building in 1605. It turned out however that the enthronement of Jahangir had by no means resolved his problems with Khusrau. The prince rebelled in 1606, and besieged Lahore; imperial armies had to be sent into the Punjab to quell the movement, which apparently enjoyed quite extensive popular support. Khusrau was captured, and others around him were meted out exemplary punishments. A year later, a fresh ÔconspiracyÕ was unearthed, with the alleged intent of assassinating Jahangir and placing his son on the throne; the conspirators were again severely castigated. This time, Khusrau himself was partially blinded and placed under strict guard. But then, in late April 1610, a further strange incident occurred. A man—later identiŽ ed as a certain Qutb from Uchh—appeared in the eastern Indian province of Bihar in the guise of a faq”r, and managed to gather together a fair degree of armed support in the region of Ujjainiya by claiming to be Khusrau who had escaped from prison. 6 The provincial governor, Afzal Khan, was absent from his post at the time, and when the rebel army, which was supported by some zam”nd‰rs, approached

5

Seyyidi ®Ali Re¾is 1999: 90-91. For the original text, see Seyyidi ®Ali Re¾is 1999a: 120-

21. For a detailed account, see Mirza Nathan 1936, i: 89-91; also Khwaja Kamgar Husaini 1978: 129-31, section entitled: ÔThe rising of the dust of sedition in the city of Patna, and its quelling by the irrigation of imperial fortune.Ó 6

THE SHADOW SULTAN

87

the provincial capital of Patna, the other ofŽ cials took fright at their numbers and  ed without offering too much resistance. The Ôfalse KhusrauÕ and his supporters were able to help themselves liberally to the treasury, and thus gain even greater popular support. Eventually, Ž ve days later, Afzal Khan returned and a pitched battle ensued, in which the rebel forces were defeated, and the impostor himself was executed by the governor. The imperial court for its part was greatly displeased, not least of all by the behaviour of its ofŽ cials who, it seemed, were all-too-ready to believe rumours of rebel princes and draw unwarranted conclusions therefrom. Some of these ofŽ cials were hence publicly humiliated, with Òtheir heads and beards shaved . . ., dressed in womenÕs head coverings, mounted on donkeys, and paraded around the city and marketplace to serve as a warning to others.Ó7 Afzal Khan himself barely escaped punishment on account of the intervention of his friends. Yet, even if the rest of his life was spent in obscurity, the Ž gure of Khusrau had not quite lost its aura. We see this some two decades later, in 1627, when Jahangir lay on his deathbed. On 28 October that year, when he died in Rajauri, in the outskirts of the city of Lahore, several rival claimants to the throne existed. Even if the robustness of the political institutions of the Mughal empire was sufŽ cient to ensure no truly deep crisis at a moment such as the death of a monarch (Alam and Subrahmanyam 2002), there is little doubt that such periods of transition were always somewhat disturbed. The succession to the throne of Shahjahan (r. 1628-1658) was no exception to this rule. The Dutch East India Company employee, Dirck van der Lee reported from Surat that the CompanyÕs factors in Agra had reported con icting rumours concerning JahangirÕs state of health. This had led Òsome of the more important persons in the city to leave the town with their costly belongings; others employed guards in their houses.Ó 8 Besides, the roads had become unsafe, and the price of provisions too had risen unusually. Now, the attention of historians has usually focused on a brief period of three months, dominated by a marked degree of political instability caused by the succession. Prince Parwiz, the eldest son of Jahangir and principal claimant to the throne in the eyes of some, had already died a year earlier, in October 1626. Khusrau himself had died in prison in early 1622, under obscure circumstances. The contest thus seemed to centre essentially on the rivalry between Shahryar, the youngest son of Jahangir, who was married to a daughter of Nur Jahan, and Prince Khurram, who had rebelled against his father in 1622, and who some

7 8

Jahangir 1999: 112-13; for the Persian text, see Jahangir 1864: 83-84. Van der LeeÕs report is summarised and paraphrased in Prakash 1998: 144-45.

88

JORGE FLORES AND SANJAY SUBRAHMANYAM

suspected in the convenient deaths of his two brothers. In the succession struggle, a central role was to be played by Asaf Khan, brother of Nur Jahan and waz”r of the empire. While Nur Jahan seems to have preferred Shahryar, Asaf KhanÕs own preferences clearly lay with Khurram. Yet, given the fact that the latter found himself in the distant Deccan, and was some daysÕ march from Lahore, and the need was felt to neutralize Shahryar, Asaf Khan seems to have found a temporary solution in order to gain some time. With the support of a sizeable party of Mughal nobles, he organised the crowning of Dawar Bakhsh, the son of Khusrau, who had meanwhile been kept in prison. According to the chronicler Muhammad Hadi: Asaf Khan, a loyal supporter of Shahjahan, conspired with Iradat Khan to release KhusrawÕs son Dawar Bakhsh from prison and give him the good news of a phantom rulership. However, he would not believe it until they swore solemn oaths that it was so. Then Asaf Khan and Iradat Khan put Dawar Bakhsh on a horse, raised the royal parasol, and set out for the next station. No matter how many messengers Nurjahan Begam sent to summon her brother, Asaf Khan made excuses and did not go to her.9

In this way, Shahryar, who for his part had also declared himself emperor and taken over the imperial treasury at Lahore, was obliged to react and to combat this unexpected rival. The battle took place near Lahore itself, and Asaf Khan managed to capture Shahryar alive, obliging him to submit formally to Dawar Bakhsh. This latter prince, also known in popular parlance as Sultan Bulaqi,10 seems to have beneŽ ted from the re ected glory of his fatherÕs charisma, and had also been somewhat favoured by Jahangir in the latter years of his reign.11 Thus, in 1623, Dawar Bakhsh had been named governor of Gujarat, and in the same year had also received the high mansab of 8000/3000 (Jahangir 1999: 397; also Jahangir 1864: 361). But, in the strategic conception of the waz”r, Bulaqi was a Ž gure whose prime signiŽ cance was decorative, and he is even described by the Mughal chroniclers as no more than a ÒsacriŽ cial goatÓ (gosfand-i qurb‰n”).12 Roughly twenty days after the death of Jahangir, Prince Khurram in the

Jahangir 1999: 456-57; for the Persian text, see Jahangir 1864: 421. Bulaqi, which is also written at times in Safavid texts as ÔBulaghi,Õ seems to be a name of Turkish rather than Persian origin. It does not seem to have any relationship with Bulaq, a small town near Cairo. 11 Cf. Richards 1993: 103, where the author notes that ÒKhusrau universally was seen as a tragic and popular Ž gure both by the populace at large and by the royal harem women who resented Nur JahanÕs power.Ó 12 Jahangir 1864: 422 (addendum by Muhammad Hadi); Khwaja Kamgar Husaini 1978: 484; also the later reference in Shahnawaz Khan 1911-52, i: 290, for the Persian text of which see Shahnawaz Khan 1888-90, i: 154. 9

10

THE SHADOW SULTAN

89

Deccan received news of his fatherÕs demise. He left immediately for Agra, but Ž rst sent a farm‰n to Asaf Khan, ordering him to eliminate Shahryar, Dawar Bakhsh, as well as other possible candidates to the throne who might represent a potential threat. Thus, on 19 January 1628, Bulaqi was made prisoner, at the same time that the Friday prayers were read in the name of Khurram, who was now proclaimed ruler with the title of Shahjahan. Some days later, we hear of the execution of Shahryar, Dawar Bahksh and his brother Gurshasp, as well of two sons of the deceased prince Mirza Daniyal (Tahmurs and Hoshang), brother of Jahangir (Jahangir 1999: 460; 1864: 425). Yet, even the ofŽ cial Mughal tradition attributes a certain sang-froid and gallows humour to the unfortunate Bulaqi. Khurram, so it is reported, had sent a certain Riza Bahadur to Lahore, with orders that Òhe should cleanse the world from the contamination of the existence of sundry princes who were the source of disturbance.Ó The account then continues: They say that Sultan Dawar Bakhsh known as Sultan Bulaqi—whom Asaf Khan had placed upon the throne for some days out of certain considerations of policy—was playing chess (shatranj m” b‰$t) with his brother Sultan Gurshasp. When he heard the noise of Riza BahadurÕs arrival, he divined his object and said to his brother: ÒVirtue (Riz‰) has not come, it is your and our Fate (Qaz‰) that has arrived.Ó Yamin-ud-Daula [Asaf Khan] in accordance with the order made over the blinded Sultan Shahryar, Sultan Bulaqi and Tahmuras and Hushang, the sons of Sultan Daniyal [to Riza Bahadur]. He on 25 Jumada I 1037 H. [22 January 1628], put them all to death in one day. (Shahnawaz Khan 1911-52, i: 811-12; Shahnawaz Khan 1888-90, i: 713-14).

Shahjahan waited outside Agra for the appropriate moment. Finally, on 3 February (7 Jumada II), scrupulously following the instructions of astrologers as to the date and hour that were best for entering the city, he was crowned in Agra.13 Once on the throne, the new ruler was soon to face serious challenges on the southern fringes of his domains. In October 1629, Khan Jahan Lodi, an Afghan noble of some importance who had been close to Jahangir (but who had chosen to support the candidacy of Dawar Bakhsh), decided to  ee Agra and take refuge in the Deccan. Since he had already served as Mughal governor in the region, he was immediately welcomed by Murtaza Nizam Shah II, the much-weakened Sultan of Ahmadanagar. Shahjahan, who greatly feared a generalized rebellion amongst the Afghans who were in his service, decided to react with the utmost harshness. He at once mounted an expedition to the south, and installed his camp at Burhanpur, in the region of Khandesh, where he arrived in March 1630. He was to remain there until March 1632, launching a series of attacks on the forces of Khan Jahan Lodi during the rest of 1630. The

13

For a summary of the major events, see Saksena 1968: 56-65; Richards 1993: 116-18.

90

JORGE FLORES AND SANJAY SUBRAHMANYAM

rebellious Afghan noble was eventually captured and executed in January 1632.14 The presence of the Mughal court in the Deccan seemed to be a clear enough sign that the emperor himself had every intention of closely supervising the governor of that provinc e, Mahabat Khan, in the process of forcing the autonomous Sultanates of the area to submit to Mughal power. An initial campaign against Ahmadnagar, which had chosen to welcome Khan Jahan, went on until 1633, culminating in the siege of the major fort of Daulatabad. In 163536, besides formally terminating the existence of the Nizam Shahi dynasty of Ahmadnagar, Shahjahan also managed to ÔdomesticateÕ the Sultanates of Bijapur and Golconda. Both these rulers were obliged from that time on to recognise Mughal hegemony and suzerainty, by way of paying a tribute and in the fact that the $utba was henceforth read in the name of Shahjahan, while excluding mention of the Shi®i Imams. Nevertheless, the Ž rst years of ShahjahanÕs reign were far from being as simple and linear in character as the rapid summary we have presented above might suggest. Besides many other severe problems, including the major famine that ravaged much of western India in 1630-31, the Mughal ruler also had to come to terms with a sort of ghost. For, soon after the execution of the Ž ve princes in January 1628, a rumour came to be heard that Sultan Bulaqi had in fact been spared, and that as the legitimate heir to Jahangir, he was in the process of getting together a powerful rebel army in order to challenge the ÔusurperÕ Shahjahan. His escape, according to this view, had been managed with the complicity of Asaf Khan, who had sacriŽ ced another person in the princeÕs place. The collaboration in the matter of Mahabat Khan, an important noble of Persian origin who had been particularly close to the deceased Prince Khusrau, was also portrayed as being decisive. At much the same time, rumours also came to be heard that Mirza Baisunghar—another son of Prince Daniyal, who had decided to support Shahryar in the succession struggle after the death of Jahangir—had also survived after the defeat that he and Shahryar had suffered at the hands of Asaf Khan near Lahore. It would in fact seem that this cousin of Bulaqi did manage to escape to Badakhshan, where he died. However, someone else managed to pass himself off as the prince, and traveled Ž rst to Balkh, and then to Iran, before eventually settling for a time in Istanbul. Here, he seems to have made grandiose plans to depose Shahjahan from the Mughal throne with the help of the

14

On Khan Jahan Lodi, see Husain 1999: 149-50 and Rahim 1961.

THE SHADOW SULTAN

91

Ottoman Sultan Murad IV (r. 1634-1640), and even claimed to have the support of the Shaibanids, as well as supporters in Kabul who would revolt at a signal from him. However, the poor Ž gure that he cut at the Ottoman court led to his being identiŽ ed as an impostor. He was apparently sent to India on the request of Shahjahan, and executed in 1636. 15 The incident is reported in a variant version as follows in an eighteenth-century Mughal text. In the 9th year [1636], he [Daulat Khan Mayi, governor of Sind] arrested the false Baisunqar and brought him to court. That trickster was an obscure person who gave himself out as the Baisunqar who in ShahryarÕs battle was the leader of the latterÕs army (sard‰r-i lashkar), and who after  eeing to the fort of Kaulas in Telingana died a natural death. The pretended Baisunqar went to Balkh. Nazr Muhammad Khan, the ruler (w‰l”) there wanted to make him a relation by marriage, but as his claims did not prove to be true the connection did not take place. Then he went to Iran, and though Shah SaŽ did not admit him to his presence, he showed him some favour. From thence he went to Baghdad and Rum. After a long time, willingly and unwillingly, the hand of death seized his collar and brought him to Thatta. Daulat Khan arrested him and sent him to court and he was put to death (ba qatl ras”d).16

Contemporary European observers, with the exception of the Portuguese (for reasons that will presently become clear) do not pay much attention to the episode of Baisunghar. However, they were far more interested in the legend of Bulaqi, which appears in a wide variety of writings from the seventeenth century. Besides the brief references to Bulaqi in the writings of Johannes de Laet (Hoyland and Banerjee 1975: 238) and Peter Mundy (1914: 105, 107, 206), a handful of other interesting versions of this incident can quite easily be cited. The German traveller Mandelslo, after noting that Shahjahan had usurped the throne from his own nephew, tells us that he actually met the prince ÔPolagoÕ in the Iranian city of Qazwin (Olearius 1679, ii: 179). For his part, Thomas Herbert is rather more prolix. He thus explains that Jahangir had decided to name Bulaqi as his successor, but that the latter—since he was very young and inexperienced—was quickly outmanoeuvred by Nur Jahan and Asaf Khan, who ordered a certain Radgy Bandor (that is, Riza Bahadur) to kill him while in Delhi. He then closes the history of the Ôpoor princeÕ who apparently only managed to reign for three months, but does mention the fact that two other false Bulaqis then appeared, each of them with sufŽ cient charisma to gather together

15 Farooqi 1989: 6, 26, 46-48. Farooqi bases his account on the chronicle of Mustafa NaÔima (1655-1716), T‰r”$-i Na®ima, or Rauzat al-Husain f” §ul‰sat A$b‰r al-§‰Ž qain. For a partial translation of this text, see Fraser 1832; and also the modern edition, Mustafa NaÔima 1967-69: iii: 1260-64, passim. 16 Shahnawaz Khan 1911-52, i: 468; Persian text, Shahnawaz Khan 1888-90, ii: 26. For another version, with some minor additional details, see Begley and Desai 1990: 191.

92

JORGE FLORES AND SANJAY SUBRAHMANYAM

a band of followers.17 Other writers present minor variants. In the account written in 1638 by an anonymous Italian, a native of Piedmont, Asaf Khan once more is given a central role. He writes: ÒThe dead King made him [Asaf Khan] swear that he would make a certain Sultan Bolaschi the king, since the one who now lives [Shahjahan] had made war on his father [Khusrau] some twenty-Ž ve years ago, but he did not keep the promise, and put him in prison and under guard, and when this King was crowned, he ordered that the other [princes] should be killed, amongst whom two were Christians, and Bolachi  ed to Persia where he is now.Ó 18 The excursus devoted by the French jeweller Jean-Baptiste Tavernier to Bulaqi, whose  ight to Iran is recounted by him as an absolute fact, is also of some interest for us. The French traveller refers to the unfortunate prince in the context of the later war of succession (1657-58), and draws his own moral connections between crime and punishment. For Tavernier, the imprisonment of Shahjahan by his own son, Aurangzeb, was simply the return for the injustice that he himself had committed three decades earlier, when he had ordered the assassination of his own nephew Bulaqi (Tavernier 1995, i: 267ff.). Precisely the same idea may be found in the text of the Italian traveller Gemelli Careri, who recounts the story in the following manner. ÒIf we now look back into the life of Sciah-Gehan, we shall Ž nd that he was punishÕd by the Hand of God as he had deserved, for the wrong he had done his Nephew Bulaki, usurping the Crown from him.Ó He then narrates the succession struggle after the death of Jahangir, concluding with the  ight of Bulaqi with the aid of Asaf Khan, who we are assured had promised Jahangir on his deathbed, Òthat they would never consent that Bulaki should be put to Death; which he afterwards faithfully performÕd, but not to settle him on the throne.Ó Careri then concludes his version of the story as follows: Bulaki receiving this dismal News [of the succession of Shahjahan] by the way, being in consternation had no hopes in safety but in  ying; which was easie to be done, because his Enemies thought not proper to pursue him. He wandered about India a long time, becoming a Fachir; but at last tirÕd with that painful Employment he retirÕd into Persia, where he was nobly receivÕd and entertained by Scia SoŽ . (Sen 1949: 232-33).

We return here to the idea of a prince who has been transformed into a qalandar or faq”r, a theme for which Couto already shows a weakness. Later,

17 Herbert 1663: 175-79. The edition by William Foster (cf. Herbert 1972) does not include the part relating to the Mughal empire. 18 BNF, Collection Dupuy 587, ÔPartenza del Re Gran Mogor della citta dÕAgra per la citta di LaorÕ (1638),  s. 300-312, citation on  s. 310v-11.

THE SHADOW SULTAN

93

at the end of the seventeenth century, Niccol˜ Manuzzi provides us with a version full of fantastic details, and having totally confused the relationships between the principal protagonists, recounts the triumphal entry into Agra of Shahjahan, and the fall of Bulaqi who, so it would appear, had no other option from that moment on but to  ee (Manucci 1990, i: 172-75). ManuzziÕs text is accompanied by a portrait in the Mughal style of Bulaqi, with the following legend: This is the portrait of the unhappy prince Sultan Bolaqui, oldest son of Djehanguir; as he wished to live a life of pleasure like his father, this permitted his brother to usurp the throne; on account of which in the chronicles he is no longer permitted the title of king; to save himself he  ed in haste to the king of Persia, and remained there until his death.19

As we shall see below, the later existence of a person in the Safavid domains who was believed to be the real Bulaqi is usually also accepted by the greater part of the texts written by European travellers, as also in some of the Portuguese sources of the epoch. We do not know, however, to what extent this recurrent rumour of the ÔresurrectionÕ of Bulaqi, taken together with the opposition centred around the false Baisunghar, actually affected Shahjahan. We need to know the manner in which such episodes might have conditioned the emperorÕs actions, both at the level of the assertion of his internal authority and in terms of the external relations of the Mughals, particularly the rather delicate dealings with Safavid Iran. The present state of information does not permit us to pursue the comparison with the well-documented reaction of the ChÕing emperor Yongzheng (r. 1723-1735) to a rumour concerning a possible revolt both against him personally, and against the Manchu domination of China in general, that came to light in 1728. 20 THE PERSPECTIVE FROM GOA The legend of Bulaqi equally attained some proportions in the Portuguese documentation of the period. In contrast to the other Europeans, in the case of Manucci 1986, i: 98. Few other known portraits of Bulaqi exist; but see Beach 1978: 61-63, Ž gure 14, where he appears clearly identiŽ ed in a group portrait. We are grateful to Saqib Baburi for this and other references. 20 A certain Zhang Xi, a messenger from a teacher from Hunan called Zeng Jing, attempted to bring around to this cause the Governor-General of the provinces of Shaanxi and Sichuan, General Yue Zhongqi. However, the latter immediately informed Beijing of what had happened, leading to an extremely complex process of inquiry promoted by the emperor himself, as has been studied recently by Spence (2002). In this case too, there is the problem of the authenticity of one personage, a certain Wang Shu, and issues of the proof of identity that require resolution. 19

94

JORGE FLORES AND SANJAY SUBRAHMANYAM

the Portuguese, it is neither through the accounts of travellers nor even through the missionary sources that we can follow the trajectory of the rumours. Rather, in the Portuguese case, it is necessary to look to the materials preserved by the state, the Estado da êndia, both in terms of the correspondence exchanged between Goa and Lisbon (or Valladolid), and in terms of the re ections of the Portuguese viceroy of that time, and in the minutes of the formal meetings that he held at Goa to discuss the ÔBulaqi affair.Õ It was recognised by the early seventeenth century that the huge dimensions of the Mughal empire, and its growing ascendancy over the politics of the Deccan, could have an immediate and rather devastating effect on the Portuguese presence in maritime western India. Through the detailed information that they regularly received from the Jesuits at the Mughal court, taken together with intelligence that was gathered through other channels, the viceroys at Goa from the 1570s on came to acquire the re ex of closely monitoring and evaluating the activities of the Mughals, in particular with regard to their Ôexternal relations.Õ But they also paid great attention to the internal revolts and the massive movements of Mughal armies, even those that occurred on the somewhat remote frontier with Central Asia. It is therefore scarcely surprising that the Portuguese should also have paid particular attention to processes of transition and imperial succession. In critical situations, when they were faced with a Mughal central power that seemed to them to be particularly adverse to the interests of Goa, the fratricidal struggles for power in Agra, Delhi or Lahore were important processes for the Portuguese, simply from the viewpoint of safeguarding their own Estado da êndia. Two examples will bring this out clearly enough. In view of the periodic illness of Akbar, taken together with the revolts of Prince Salim, the question of imperial succession in India was amply discussed in the correspondenc e between Goa and the Habsburgs in the early years of the seventeenth century. In March 1604, Philip III believed that things could turn out very well indeed for the Portuguese. He wrote to Goa: ÒThe state in which you tell me that Equebar Ž nds himself with his oldest son is what is best for the Estado, in view of the intentions of this king. May it please God that as long as he lives this division between them continues, for on his death we may well expect that it will spread throughout his domains.Ó 21 A year and a half later, the death of Akbar and the succession of Jahangir were followed very closely by the Jesuits

21

207.

King to viceroy, Valladolid, 23.III.1604; AHU, Cod. 282,  s. 206-207v, citation on  .

THE SHADOW SULTAN

95

who were at that time resident in the Mughal court (cf. Alam and Subrahmanyam 2002: 108-12). We may also brie y consider the situation in the 1650s, and the Portuguese reaction at that time to the succession of Aurangzeb (r. 1658-1707) . This emperor, the son and successor of Shahjahan, awakened the deepest fears of the Portuguese who thus produced extremely critical images of him. But, once again, the possibility of long-lasting internal wars and of a sort of political ÔimplosionÕ gave hope to the administration of the Estado da êndia. Thus, the viceroy wrote to Lisbon in 1666: ÒThe brothers, sons of the Mogor, are at (. . .) war, and the oldest has left for the parts of Persia, and nothing is known with certainty about him as he was defeated; the other brothers have had various confrontations, the one of who calls himself king is Auranzeb, and he is in possession of the kingdom. They say he is a tyrant and cruel, and not particularly well-disposed towards the Christians.Ó 22 The proŽ le of Aurangzeb that is set out here naturally meant that the Portuguese were open to an alternative solution. It is in this context that we must understand the instructions given to Greg—rio Pereira Fidalgo when he was sent on his embassy to Iran in 1696-97. In view of the advanced age of Aurangzeb at that time, the ambassador was asked to advance his contacts with the Mughal prince, Muhammad Akbar, who was in those years an exile in the Safavid court: ÒTo motivate him in this sense, you will demonstrate to him the great beneŽ ts that will follow from our friendship for the pretensions that he may have in Indust‹o as the son of the Mogor, since the latter is so old, it is evident that his life must be of short durationÓ (Aubin 1971: 99-128, esp. 115; cf. Pissurlencar 1928). The Portuguese reaction to the rise to the throne of Khurram in the late 1620s, and the movements around Bulaqi, can be seen within the ambit of a very similar set of tensions and motivations. As we shall see, the emergence of the Ž gure of Shahjahan was followed in Goa, as also in Lisbon (and Madrid) with a particular attention. The Ž rst actions of the Mughal emperor on coming to the throne already carried a premonition of the difŽ cult relations that he would have with the Portuguese, and with Christianity in general. It would seem that the religious eclecticism of Akbar and Jahangir, which had been at the origin of both dreams and disappointments concerning the possible conversion of the Mughal court, was now giving way to the assertion of a regime with a

22 Viceroy to king, Goa, 22.III.1666; HAG, Mon ›es do Reino, no 36,  . 535. The Ž rst part of this sentence seems to relate to rumours that Dara Shikoh was still alive, as we know that in 1663-64, someone claiming to be this deceased Mughal prince had appeared in Gujarat and led a brief revolt; cf. Pearson 1976: 226.

96

JORGE FLORES AND SANJAY SUBRAHMANYAM

somewhat more orthodox Islamic character, which was then pushed forward and intensiŽ ed by Aurangzeb. The period thus sees some actions against the places of worship of other religions, and the hardening of the attitude towards Christian missionaries seems to be a part of this more general movement.23 The Jesuits did maintain their ÔMogor mission,Õ but the ambience had changed from what had obtained in the late sixteenth century. This reverse, which was both religious and political, was not entirely autonomous of another jolt, this one of a politico-economic nature. We refer here to the destruction of the Portuguese establishment of Hughli in 1632, which, no matter what the precise reasons for it were, does reveal quite clearly the mercantile aspect of the imperial strategy followed by Shahjahan (cf. Subrahmanyam 1993: 396-401; Flores 2002). On account of all this, it is hardly surprising that the image of Shahjahan in the Portuguese texts of the epoch is strongly negative, much worse than those of his father or his grandfather. According to the Jesuits who lived in Agra, this emperor Òhas a natural aversion for the Padres and all the frangis.Ó24 ÔTyrannousÕ and ÔcruelÕ are two of the most common adjectives that the Portuguese use to characterise the monarch whom they often continued to designate not by his chosen title, but in a somewhat pejorative fashion as ÔSult‹o Corromo.Õ The enthronement of Shahjahan and the Ž rst phase of his reign as emperor happen to coincide, grosso modo, with the start of the Ž rst mandate of the Count of Linhares as viceroy of the Portuguese Indies.25 D. Miguel de Noronha began to govern in October 1629, more or less at the moment when Khan Jahan Lodi decided to abandon Agra and  ee to the Deccan. The march of Shahjahan towards the Deccan, the desertion of the great Afghan noble, as well as the reports of the death of Bulaqi are all subjects that are referred to in some detail in the Portuguese documentation of the epoch. Besides, the administration of the Count of Linhares is particularly well-documented in view of the fact that D. Miguel de Noronha—in keeping with royal instructions—maintained a public diary of the events during his government.26 It is this valuable text, taken See the interesting discussion in Bernier 1934: 286-88. ÔRela ‹o dÕalgumas cousas que passar‹o na Miss‹o do Mogol, des do Ž m do anno 627 te ao dia presente 13 de Junho de anno 1628Õ; BL, Addn. Ms. 9854,  s. 116-130, published in Rego 1963: 188. 25 On this Ž gure, see Disney 1992; 2002. 26 However, the diary of Linhares is incomplete, and three distinct parts exist to it. The Ž rst is to be found in the Biblioteca da Ajuda, Lisbon (BA), Codex 51-VII-12, and deals with the period from 3 March 1630 to 6 February 1631. The second, also unpublished, is to be found in the Biblioteca Nacional, Lisbon (BNL), Codex 939, and covers practically the whole of the year 1631 (9 February to 20 December). The third section has been published as Linhares 1937-43, and runs from February 1634 to January 1635. 23 24

THE SHADOW SULTAN

97

together with the letters that one Ž nds in the so-called Livros das Mon ›es (Monsoon Codices) and the minutes of the Council of State (Conselho do Estado), which permit us to reconstruct the manner in which the Portuguese struggled with the legend of ÔSult‹o Bolaquim.Õ The Ž rst time Bulaqi appears in the materials of the Estado da êndia may be dated to a document on the ÔNeighbouring KingsÕ (Reis Vizinhos) from the end of 1629. Here we Ž nd a reference to the disturbances caused by the death of Jahangir, and the fact that the throne belonged by rights to his Ҏ rst-born grandson (neto morgado) by name Sult‹o Bulaqui, somewhat above twenty-three years of age.Ó In this version of events, Jahangir had handed over his grandson to the charge of Asaf Khan but the latter, as soon as the emperor was declared dead, quickly summoned Khurram to the court telling him that he had Bulaqi Òsecure in his hands.Ó Khurram then ascended the throne, Òand Bulaqui has gone away to the borders of the kingdom of the Xaa, and until now they said he was dead, or had been killed, and now in the last two months they say he is alive, and that the Xaa has promised him one of his legitimate daughters, and has given him great succour, and they say that he has returned to the Kingdom of Dely.Ó27 The following reference to Bulaqi that appears in the Portuguese ofŽ cial sources dates to 15 March 1630. On the basis of information that had been sent to him from one of his spies in the ÔMogorÕs camp,Õ the viceroy made a note in his diary of the advance of Shahjahan in the direction of the Deccan, and further noted that the emperor had left Agra, Òwhich is an open and defenceless cityÓ (sic) out of fear of the reaction of Bulaqi, rather than to really conquer the Sultanate of Ahmadnagar. He then goes on to describe the bloody succession of Shahjahan, and makes it a point to argue that it was KhusrauÕs son, Bulaqi, who was in fact the legitimate heir. The version that had reached Goa had it that it was the wife of Mahabat Khan who had saved Bulaqi, Òby placing another dead man in his place.Ó The real Bulaqi for his part had  ed to Delhi, and still under the protection of Mahabat Khan, had even married one of the latterÕs daughters. Linhares appears excited at the prospect of a possible alliance between Khan Jahan and the Sultan of Ahmadnagar, and concludes by noting that if Bulaqi seizes power once more, the Dutch would be expelled summarily from Surat.28 HAG, Mon ›es do Reino, liv. 13 (AN 15), published Pissurlencar 1930: 52, 54. BA, Cod. 51-VII-12,  s. 13v-14. For a Dutch perspective on the events of the period, see the diary of Commissioner Dirck van der Lee, summarised in Prakash 1998: 149-50. The Dutch factor notes that the death of Bulaqi had caused Òa certain amount of resentment among both the big and the small.Ó 27 28

98

JORGE FLORES AND SANJAY SUBRAHMANYAM

Two months later, there is further news of Bulaqi. On 12 May, Linhares notes that letters have arrived in Goa from Bengal, either written by or sent through a certain Gaspar Pacheco de Mesquita, who was a particularly important informant who kept the viceroy abreast of developments in that region. These letters claim, he writes, that ÒBolaquim is alive, and that he is much loved and adored by all, and that many people follow him, and that he has put together a very large army, and that the Nababo of Bengal and his son-in-law have taken four large ships and  ed out of fear to Massulapat‹o and that he [Bulaqi] is very soundly situated in those partsÓ (BA, Cod. 51-VII-12,  . 36v). Then, on 28 August, further letters arrived for the viceroy from a certain Diogo Saraiva, his informant in the court of Ahmadnagar. This time it was reported from Daulatabad Òthat Bulaquim, king of the Mogors, is alive, and I have mentioned that it was believed that he was dead, and he is bringing together an army of twenty thousand men to attack the Idalc‹o [®Adil Khan]Ó (BA, Cod. 51-VII-12,  . 87). After trying to render coherent these rather confusing and contradictory rumours, Linhares then spent the rest of the monsoon season in Goa, without receiving any further details concerning the strange peregrinations of Bulaqi. But then, in October, the viceroy received a veritable avalanche of intelligence, which eventually led him to act. We may follow the chronology of events. On 11 October, a fresh and very happy rumour reached the viceroy. It was said that Shahjahan had died, and this was not surprising since the banias had held that according to a prophecy, Òhis time of dominance was overÓ (BA, Cod. 51-VII-12,  . 97).29 Linhares reacted to this news with enthusiasm mixed with caution, and three days later, further news arrived from Bijapur: ÒI received letters from Vizapor [written on] the Ž rst of this month, in which they say that Bulaquim, the Mogor King, who had disappeared and who was believed to be dead, was now in the company of the Mellique [Nizam Shah], even though he scarcely had four thousand horsemen, and that his ambassadors were now forty leagues from Vizapor, and that the baneanes claimed that his [ShahjahanÕs] time of dominance was over, for this other one [Bulaqi] had come to take possession of his kingdom, though they say that the peace between the Mogor and the Mellique is being negotiated with great urgency, but for the Estado it would be of very great effect if Bulaquim were to seize hold of the kingdomÓ (BA, Cod. 51-VII-12,  . 98).

Shahjahan, born in the year 1000 of the Hegiran calendar (1591-92), seems to have possessed a somewhat millenarian aura about him. See, in this context Begley 1979. However, this particular prophecy appears to be of Vaishnava or Jain origin. 29

THE SHADOW SULTAN

99

Meanwhile, Diogo Saraiva too had written Ž ve times to the viceroy from Daulatabad during the month of October, giving him a close account of various affairs, including the adventures of Bulaqi. Taking all these pieces of information together, the viceroy decided to send a Ž rst summary re ection to Portugal on the whole delicate question. This was done in the form of two letters, the Ž rst dated 4 November 1630, 30 and the second dated 6 December of the same year (BA, Cod. 51-VII-12,  s. 178-178v; see Pissurlencar 1953: 523-25). It is really at this moment that Linhares makes the transition from simply collecting  oating rumours and scarce information, to seeking out systematic intelligence on the matter in order to formulate a coherent policy in consultation with the Habsburg monarchy. This appears clearly enough in his letter to Diogo Saraiva, written on 6 November: I am not fully satisŽ ed with what you tell me about Sultam Bulaquim, and it is important that you should dilate on it by writing a longer relation, from the time that he escaped from death, and how and where he went and was hidden, and how he reappeared and at what time, and for what reason, and who he depends on and who helps him, and where he is and with which people, and what he intends to do, and if he is coming there to seek the aid of the King Melique and Canaj‹o [Khan Jahan], and what it is believed the captains of the Mogor Xaajahan will do on his arrival, and whether they want him and will acclaim him as their true King which he is, and also what his ambassador told you of the desire that he had to deal with me concerning various matters, setting them out in order to see if they could be attained; and send me this relation [which should be] very well written and with all the necessary declarations [documentation] that are necessary, and send it to me so that I can send it on to His Majesty. (BA, Cod. 51-VII-12,  . 107).

The viceroy thus clearly felt the need to evaluate the rigour of the intelligence that was coming in to him, in order to deŽ ne a coherent line of action. For, already in less than a year, Bulaqi had been spotted on the border with Iran, in Delhi, Bengal and in various parts of the Deccan, and the most diverse projects and alliances had been attributed to him. To make matters even more confusing, Linhares received news from Bijapur on November 16, which he considered to be extremely doubtful, and which claimed that Bulaqi was preparing still another alliance, this one with ®Abdullah Qutb Shah, Sultan of Golconda (r. 1626-1672) (BA, Cod. 51-VII-12,  .110v.). Now, the Portuguese viceroyÕs own diplomatic project came to centre on the

30 HAG, Mon ›es do Reino, liv. 14 (AN 16),  . 175. It is probably this letter, or the one that is cited in the note below, that arrived in Lisbon the next year with the title ÔSobre o netto del Rey Jahamguir, que escapou, e estado em que Ž caÕ; see BNL, Cod. 11410,  . 150, ÔLista de todas as cartas que vier‹o na via da nao Almiranta deste governo, anno de 1631.Õ

100

JORGE FLORES AND SANJAY SUBRAHMANYAM

idea of an alliance between the Sultan of Ahmadnagar, Khan Jahan Lodi, and Bulaqi, against Shahjahan. His plan was that the Estado da êndia would provide various sorts of discreet support to this alliance, allowing the allies for instance to use the ports in western India that were under Portuguese control. Based on this view of things, Linhares Ž nally wrote a letter to Bulaqi himself on 2 December, Òtelling him that he should choose the direction in which he wanted to proceed, either to the north or to the south, and that I would have him fetched from there in  eets so that he could be secure, and that he should let me know as quickly as possibleÓ (BA, Cod. 51-VII-12,  . 121v.). 31 The idea was that the Mughal prince should enter Ahmadnagar territories as rapidly as he could as a measure of security (BA, Cod. 51-VII-12,  s. 120v, 121). 32 The advantages of such an alliance were obvious from the point of view of Goa, both from a political and from an economic perspective. On the one hand, the Estado da êndia would thus be able to limit the power of an emperor who was clearly hostile to the Portuguese, and instead have in his place the far more consensual Bulaqi. Throughout this period, Linhares continues to harp Ž rmly on his legitimacy. He calls him the Ôtrue Mughal KingÕ (verdadeiro Rei Mogor), insists on the empathy of the populace for this prince, and as importantly (or perhaps even more so), refers to how a great proportion of the most in uential Mughal am”rs are in fact ready to adhere to his cause. On the other hand, he also suggests that in case of a victory for Bulaqi, it is certain that Òthe lands and customs-houses of Surrate, Cambaja, and BarocheÓ will be handed over to the control of Goa. Apparently, the envoys from Khan Jahan had already reached an agreement with Bulaqi in this respect (BA, Cod. 51-VII-12,  . 119). The month of December 1630 proved absolutely decisive in the matter. On the 6th of the month, the viceroy received letters from the port of Chaul, which would turn out to be crucial for the dŽ nouement of the Bulaqi affair in its Portuguese incarnation. Besides the captain of the Portuguese fort, Linhares also received letters from a certain Manuel de Azevedo, stating that Bulaqi was now to be found resident in Upper Chaul (Rewadanda). He had apparently made his way there discreetly from Gogha, in a small vessel and accompanied by a

HAG, Mon ›es do Reino, no 16,  s. 178-178v. The letter from the viceroy was apparently carried by an Ôintelligent BrahminÕ (bramane inteligente), whom Pissurlencar identiŽ ed with a certain Ramoji Sinai Kothari; cf. Pissurlencar 1952: 30. 32 Here too, there is a parallel to be drawn with the projected alliance between the Estado da êndia and Prince Akbar at the end of the seventeenth century. In the latter case, the viceroy instructed Greg—rio Pereira Fidalgo that the exiled son of Aurangzeb should be incited to seize the throne by making use of one of the Portuguese-controlled ports, Òfor we have so many doors for him, all between Dio and GoaÓ (Aubin 1971: 115). 31

THE SHADOW SULTAN

101

handful of servants. Azevedo sent him a few trusted bania intermediaries, with money and some gifts, and these men stated that they knew the prince well and that they could guarantee that it was in fact the real Bulaqi. The captain of the Portuguese fortress of Chaul for his part had also offered the Mughal prince his help, so that he could leave Upper Chaul in a secure fashion (BA, Cod. 51-VII12,  . 122v). It was claimed however that the local governor of Chaul (sarsamat, or in Indo-Portuguese, ÔsarssamataÕ) was isolating the faq”r Bulaqi from external contacts, and subjecting him to a propaganda campaign against the Portuguese. He had hence placed him in a fortress that was some leagues from the city proper, and was reportedly trying to convince him that the Portuguese were his enemies and that they planned to kill him (BA, Cod. 51-VII-12,  s. 122v, 123v, 126). The pressing importance of this news from Chaul led the viceroy to summon a meeting of the State Council on the following day, in order to decide the next step to take (BA, Cod. 51-VII-12,  . 122v). 33 The meeting concluded that a certain Jesuit called Francisco Le‹o should be asked to leave Goa immediately for Chaul in order to identify Bulaqi. This Padre Le‹o had been superior of the mission at the Mughal court between 1627 and 1628, and in the context of the court had followed the transition between Jahangir and Shahjahan. He also claimed to know Bulaqi rather well.34 But the dream of an alliance between the Estado da êndia and the supposed heir to the Mughal throne went up in smoke even before the start of the year 1631. For, on 27 December, viceroy Linhares received a letter from Padre Francisco Le‹o, announcing his arrival in Chaul. The Jesuit had quickly found out that Bulaqi was exchanging letters with Khan Jahan Lodi, and declared that he intended to leave the next day for the fortress where the Mughal prince was thought to be (BA, Cod. 51-VII-12,  . 129). Two days later, Linhares received a fresh letter. The man whom the Jesuit had visited was not the one whom he had known at the Mughal court. The missionary stated bluntly that the ÔBulaqiÕ he had seen in Upper Chaul was no more than a blatant impostor (BA, Cod. 51-VII-12,  . 129v; HAG, Mon ›es do Reino, no 16,  . 202v). The idea of proof of identity that was used here was thus simple and direct: the testimony of one reliable witness, in this case the Jesuit, was seen as enough to discredit

33 Also see ÔConselho sobre Solt‹o Bolaquim que veyo a Chaul de CimaÕ; Goa, 7.XII.1630; in Pissurlencar 1953: 331-33. 34 Padre Le‹o was removed from his post in 1628, largely on account of an imprudent con ict that he had with three Venetians who lived at the Mughal court; he also fell out at much the same time with the powerful Asaf Khan.

102

JORGE FLORES AND SANJAY SUBRAHMANYAM

the claims of the man who called himself Bulaqi. Things were not so simple in most other cases of imposture in the sixteenth- and seventeenth-century world (cf. Eliav-Feldon 1999). A few days later, on 12 January 1631, the viceroy ruefully related what had happened to Philip IV in a letter: ÒOn the twentieth of November of the last year, a veiled man (hum homem embussado) arrived in Chaul by sea, with a voice that suggested that he was the said Bolaquim, and he persuaded all the Moors and even the Christians who were there that he was the true Mogor King, and he went to the Morro de Chaul on disembarking with much adoration, and veneration, and he was then secured in a fortress that is in the mountains which they call Drugo. I was advised with all speed so that I could have him stay over in the fortress of Chaul, and as I am not given to believing in such matters so quickly, it seemed to me better to Ž rst send a Padre of the Company, called Francisco de Le‹o who had spent many years in educating this prince. Francisco de Le‹o carried out this task very well indeed, even though it was at risk to his own person. He entered the fort of Drugo, and in a letter of 24 December, wrote to me that this was not Bolaquim.Ó 35 The Count of Linhares immediately lost interest in the subject. In the following years, the viceroy preferred to gamble on a delicate strategy of an alliance between Bijapur, Ahmadnagar and Golconda with the aim of checking the advance of Shahjahan into the Deccan, and thus attempted to guarantee the safety of Goa and the Portuguese fortresses of the Prov’ncia do Norte. During the rest of his government, it is rather difŽ cult to see any coherent line of development in the shifting logic of the alliances and con icts between the Estado da êndia, Shahjahan and the Sultanates of the Deccan. An anonymous Italian account written in Goa in 1632 resumes the Bulaqi affair to that point, and already in January of that year, takes note of the new in ection in Portuguese ofŽ cial attitudes on the question: ÒThe legitimate Bolachino, the true king of the Empire of the Mogor, persecuted by Corumbre [Khurram] who reigns now, came as far as Carapatan, 18 leagues from Goa, and they said that he was coming here, but he went off instead to Persia, which was good for him and a favour to us, because besides the fact that they could not swear it was him, it would have given cause [for con ict] to the Mogor who has just become our neighbour.Ó 36

35 Viceroy to king, Goa, 12.I.1631; HAG, Mon ›es do Reino, no 16,  . 202v. By a scribal error, the letter is dated 12 December 1630. 36 Biblioteca Casanatense, Rome, Ms. 2681, ÔRelatione di alcune cose che succederono in India nellÕanno 1632,Õ  s. 142-59, citation on  . 142v.

THE SHADOW SULTAN

103

A somewhat similar logic as that which appears in the case of the ÔBulaqi affairÕ also appears to work for a time in the case of the ÔdoubleÕ of Baisunghar, in whom D. Miguel de Noronha also seems to have believed for a time. This time however, the mirage of an alliance between the Estado da êndia and an alternative to the detested Shahjahan moved from the southern frontier of the Mughal empire to its northernmost conŽ nes. What lent the affair a particular piquancy was that Baisunghar was one of the three sons of Mirza Daniyal who had been baptised by the Jesuits in Agra in 1610, an episode that had a great place in the missionary sources of the period. As might have been predicted, the conversion of D. Carlos (Baisunghar), D. Filipe and D. Henrique, was seen as a preŽ guration of the conversion of Jahangir himself, as well as of his whole court and kingdom. 37 In a meeting of the State Council dated 6 November 1632, there was a discussion on whether Portuguese support should be lent to this Mughal prince, even if he was now an apostate from Christianity. It was noted that Òin this interim period which was one in which the army of the Mogor was destroying the kingdom of the Idalc‹o [Bijapur], in the kingdom of Cabul which belongs to the Mogor, there was a rebellion by one of his cousins who had been baptised by the Padres of the Company, and who was called Dom Carlos, and later he gave up the faith, and he took control of the said Kingdom, with the help of the people of the Usbeques, who are the bravest and most feared of all these lands, because Dom Carlos married a daughter of that King. Because of this, the Mogor found it necessary to go to the rescue of Cabul, since it is from those parts that all the horses that make up his army come, besides which he also feared that Dom Carlos could easily seize hold of the kingdom of Laor, which borders that of Cabul, and it is understood that he can do it quite easily because the Mogor is a tyrant and unpopular (. . .) and Dom Carlos is valiant and liberal.Ó 38 At the same time, the references to Bulaqi in the Portuguese sources become less frequent. It is of course true that the years 1632-33 are not as well documented as the preceding ones, since the diaries of viceroy Linhares from the period have not come down to us. But there is still little doubt that the subject of Bulaqi ceased for the most part to interest Goa. For Linhares, the project of an alliance with Bulaqi vanished at much the same time that the idea of

For an account of these events, see Camps 1957: 8-10; as also Hern‡ndez 1958: 226-32. ÔSobre a guerra do Mogor suas detremina ›es e retirada entre aqui a carta sobre as cousas do Melique e tambem a retirada do Turco de Babilonia,Õ Goa, 6.XII.1632; in Pissurlencar 1953: 556-57. 37 38

104

JORGE FLORES AND SANJAY SUBRAHMANYAM

proŽ ting from the revolt of Khan Jahan Lodi against Shahjahan began to appear less and less realistic. Finally, we are aware that the Afghan noble was captured in January 1632, and his head taken to the emperor who was at the time in Burhanpur, an episode that the artists who illustrated the celebrated Windsor P‰dsh‰hn‰ma brought out in rather gruesome detail.39 Besides, the destruction of the Portuguese settlement of Hughli in 1632 was to pose a challenge to the viceroy that was almost as important as those that he faced in the Deccan, forcing him to strike a balance between the neighbourhood of Goa and Portuguese interests at the eastern extremity of the Mughal empire. It was only in a letter of 28 February 1632 that Philip IV eventually reacted to the information that had been sent to him from Goa in November and December of 1630. This meant of course that the response from the Habsburg centre was already somewhat out-of-date, since the later developments in Chaul were not taken into account. Even so, there is a curious detail that catches oneÕs eye. The king counsels prudence, noting that Òthe Estado should always maintain peace with the Mogor,Ó in view of the latterÕs enormous power. We Ž nd the same preoccupation a half-century later, when the Count of Vila Verde was to urge Greg—rio Pereira Fidalgo to seek an alliance with Prince Akbar, the rebel son of the emperor Aurangzeb. 40 Hence, it is suggested that any negotiation with Bulaqi should be as discreet as possible, so that Shahjahan should not grow suspicious and hence take it upon himself to make war on the Portuguese. The pragmatic posture of the Habsburg monarch with regard to a possible alliance with Bulaqi is also worth noting. For it would seem that Philip IV was even willing eventually to consider an alliance with a false Bulaqi, that is with Òa man who under his name can lay claim to the kingdom of Mogor.Ó A ÔgoodÕ impostor was thus as useful as the real thing. 41 A similar move had been proposed a century earlier by the governor Nuno da Cunha (1529-1538), in relation to Sultan Bahadur of Gujarat and his rivals. A contemporary observer— the celebrated Goan Brahmin Krishna—claimed that the governor had tried to

Painted by Abid, ca. 1633; for this painting see Beach and Koch 1997: 50-51. ÒIn all these negotiations that concern Prince Acabar, you will be cautious in two matters (. . .), the second in trying to Ž nd out if his father, the Great Mogor, has some of his agents in the court, who serve him as spies to observe the dealings of his son, be very careful with these men, and once you get to know them, show them how faithful we are to the Mogor, for if not you will scandalise him, and since he is our very close neighbour and his power embraces our territories, if he is scandalised he may break off with us to the great damage of the Estado, which is not in a condition to resist so great an enemyÓ; Aubin 1971: 116. 41 IAN/TT, Livros das Mon ›es, liv. 30,  . 7. Also see the version in HAG, Mon ›es do Reino, liv. 16 A (AN 18),  . 3. 39 40

THE SHADOW SULTAN

105

promote the case of one of BahadurÕs brothers by Òbringing from Dabul a black (huum negro) saying he was the brother of the King of Canbaya, and he did him much honour and lavished great expenses [on him].Ó42 In his own reply to Philip IV, written in the closing days of 1632, D. Miguel de Noronha informed the king of his decision not to proceed further on the question of ÔBulaqi,Õ who Padre Francisco Le‹o believed had now  ed to Iran. In point of fact, after this Portuguese moment, the spectre of Bulaqi travelled north, and we will follow it presently into the Safavid domains. According to some Portuguese sources, Shah SaŽ (r. 1629-1642) covered the man who they saw as an impostor with grants and beneŽ ces, giving him an army and even entire charge of Qandahar, which the Safavids had conquered from the Mughals in 1622. Even though he did not wish to involve the Estado further in the matter, Linhares could not help gloating over this: ÒIf this is so, nothing could be more convenient for us.Ó43 IN THE COURT AT ISFAHAN It was indeed the case that a certain ÔBulaqiÕ had sought exile in Iran, and was attempting to persude Shah SaŽ to aid him in a project to regain power at the expense of Shahjahan. The English factor, William Pitt, claimed at much the same time to have lent money to the ÒMogulls brother which  ed into Persia,Ó and stated that he feared a violent reaction from Shahjahan on this account. 44 The Persian sources of the Safavids are in consonance with the Europeans in some measure on this question, but there are also signiŽ cant points of divergence. The ofŽ cial chronicler of Shah ®Abbas, Iskandar Beg Munshi, provides us a Ž rst version of events at the Mughal court which is as follows. After the death of Jahangir, he writes, Òthe principal ofŽ cers of the Mogul state, apparently in accordance with a testamentory disposition made by Jahangir at the

42 Summary of a letter from Krishna to D. Jo‹o III, (ÔSumario das cartas que vieram este anno de 1534 na armada da India . . .Õ), in Rego 1974: 193-94. On Sultan Bahadur and his brothers, see the note by Sanjay Subrahmanyam, ÔIrm‹o de Bah‰dur (Badur) que viveu em Goa,Õ in Couto 1999, ii: 88-89. On Krishna, also see Pissurlencar 1952: 1-16. 43 IAN/TT, Livros das Mon ›es, liv. 30,  . 7. Philip IV appears to have understood the new situation in November of the following year, and broadly agreed with the new strategy proposed by Linhares. The king however did not see it as if Bulaqi had been handed over charge of Qandahar, and instead interpreted matters as if the Mughal prince had been put in charge of the same army as had been used by Shah ®Abbas I to conquer Qandahar in 1622; letter dated Lisbon, 12.XI.1633, IAN/TT, Livros das Mon ›es, liv. 31,  . 65. In any event, the version in the Persian sources is quite different, as we shall see below. 44 Letter from Dabhol to William Methwold in Surat, 19.V.1636; in Foster 1911: 259.

106

JORGE FLORES AND SANJAY SUBRAHMANYAM

time of his death, nominated as his successor Dawar Bakhsh the son of the blind Sultan Khusrau, who was known as Sultan Bolaghi, and minted coins and had the $utba recited in his name.Ó The other princes are said to have agreed to this, except Shahryar in Lahore. The con ict between Bulaqi and Shahryar is then described, ending with the defeat and blinding of Shahryar. Iskandar Beg now passes to a discussion of the fortunes of Sultan Khurram, and his changing relations with Jahangir. He even suggests that at one point in the 1620s, wearied by his miserable condition, Khurram Òtried to go to Iran and seek assistance from Shah ®Abbas.Ó At the death of his father, his position is thus portrayed as rather weak, save for the fact that he was supported by Mahabat Khan, who it is stated did not get along with Asaf Khan, the chief supporter of Bulaqi. It was with the aid of this powerful noble and the Òprinces of the DeccanÓ that Shahjahan is said to have marched northwards, gathering up support on the way from ÒChaghatays, Mughals, Afghans and Rajputs.Ó This unexpected swell of support eventually led to a sharp weakening in the situation of Bulaqi. So, writes Iskandar Beg: Sultan BolaghiÕs fortunes began correspondingly to decline, and his principal ofŽ cers of state and the wisest heads among his military leaders saw no future for a grandson now that an ambitious natural son of the late emperor was in the contest for the throne. Their loyalty to Sultan Bolaghi wavered, and they began to leave Lahore and go to Agra to join Shahjahan, striving to outdo one another in the race to pledge their fealty and services.45

Abandoned by his nobles and troops, Bulaqi was also soon betrayed by his own chief supporter Asaf Khan, who is hence portrayed here not as the initial mastermind behind the operation to place Shahjahan on the throne, but rather as an unwilling accomplice.46 It is only to ingratiate themselves to Shahjahan that Asaf Khan and his faction kill the Ž ve princes: and one of those killed is termed by Iskandar Beg ÒDawar Bakhsh ibn Sultan Khusrau, known as Sultan Bolaghi and entitled Sher Shah.Ó This last piece of information, concerning the title that Bulaqi allegedly assumed on mounting the throne, is not to be found on the coins of the epoch, and appears only on a single farm‰n from 1627. 47

45 Munshi 1978-86, ii: 1290-93; for the Persian text, see Munshi 1956-57, ii: 1066-69. For a general re ection on the Safavid chronicling tradition in the period, see Quinn 2000. 46 For still another version of the succession, in which Asaf Khan is portrayed as having acted in the best interests of the empire, and Dawar Bakhsh is portrayed as somewhat less of a victim than usual, see Kamboh 1967-72, i: 169-82. 47 Compare the titulature used on the surviving coins of Bulaqi, for which see the accompanying illustration of a silver rupee of Dawar Bakhsh: Ashmolean Museum, Oxford, Heberden Coin Room (ex-Thorburn Collection, acquired 1965-66). The coin is minted in Lahore in 1037 H. and carries the legend on the obverse: ÒAbuÕl Muzaffar D‰war Ba$sh

THE SHADOW SULTAN

107

However, the information presented by Iskandar Beg in this chronicle is contradicted by another Persian source, paradoxically itself the continuation of this very chronicle by the very same Iskandar Beg and a certain Muhammad Yusuf, and entitled Zail-i T‰r”$-i ®ålam‰r‰-yi ®Abb‰s” (Munshi and Yusuf 1938: 12026).48 The Zail, in its account of events of the year 1041-42 H. (1632), contains a detailed account of the affair of ÔSultan Bulaghi,Õ which it may be worth paraphrasing here in some detail, for the light it sheds on a number of questions. The relevant section of the chronicle begins by noting that among the Tarin Afghans (on the border between the Safavid and Mughal domains), there was a Ž fteenyear old boy, who had been brought clandestinely (duzd”da wa posh”da) to their leader Sher Khan Tarin, who had decided to employ him as his cup-bearer and lamp-lighter. This boy was said to be the son of the Mughal prince, the late Sultan Khusrau, himself the son of Shah Salim, the son of Jalal-ud-Din Muhammad Akbar, the erstwhile ruler of Hindustan. The boyÕs name was apparently Nabdi Mirza, and it is reported that Sher Khan kept him very secretly, so much so that few people knew about him until he was captured by ®Ali Mardan Khan the Safavid governor (beglerbeg”) of Qandahar during a skirmish with the Tarins. ®Ali Mardan Khan then sent him to the Safavid Shah, with details of the story about him and his royal Mughal origins. The Shah for his part placed him in the custody of the waz”r of Isfahan, asked him to keep him in the fort of Tabrak, with the full honours and dignity appropriate for such a prince, until the full truth about him was ascertained. In the meanwhile, however, another young man, this one about twenty years old, arrived via the port of Surat in the southern Iranian region of Fars. He claimed for his part to be none other than Dawar Bakhsh Mirza, the son of Khusrau, and said that he had been crowned with the help of some nobles as the successor of Sultan Salim after the latterÕs death, on the way to Kashmir (dar r‰h-i Kashm”r). This prince, it is noted, was popularly known not by his formal name, but rather as Sultan Bulaghi (note: not ÔBulaqiÕ). The authors of the Zail remark that Dawar Bakhsh had already been mentioned in the T‰r”$-i ®ålam‰r‰-yi ®Abb‰s” in the context of the account of the last years of Jahangir. They sum up the story as follows: at the time of the death of Jahangir (or Salim), Bulaghi (taking advantage of the fact that Prince Khurram was not in the good books of the late emperor, and that he was far away in the Deccan)

B‰dsh‰h.Ó We are grateful to Shailendra Bhandare of the Ashmolean Museum for his help in this matter. For other discussions, see Wright 1905; and Delmerick 1884. For the farm‰n from Dawar Bakhsh to Raja Jai Singh, conserved in Bikaner, see Ali 1978: 39. 48 The authors are grateful to Muzaffar Alam for his unstinting help with this text.

108

JORGE FLORES AND SANJAY SUBRAHMANYAM

had approached Lahore and crowned himself as emperor; but when his supporters heard of the approach of a huge army under Khurram, they had all deserted him and run away. Khurram had thus managed to take possession of the Mughal capital, Agra, and had Ž rmly established as the new emperor with the name Shahjahan. Asaf Khan, who is now presented as the main prop of Khurram (in contrast to the earlier version in the Safavid chronicle), and other nobles thought it better to make peace with Bulaghi and other rival princes. He hence persuaded Bulaghi to submit to the new emperor, explained his position to Shahjahan, and tried to have him pardoned. But the emperor would have none of it and ordered BulaghiÕs life to be terminated together with that of several other princes. Subsequently, it is said, he was executed (badarja-i shah‰dat) with Ž ve other ill-starred princes. Since this was what was believed in Iran in 1630, when the man claiming to be Dawar Bakhsh had entered the land of Fars, the governor there, Imam Quli Khan, had sent a report about him to the Shah. The Shah had prudently directed that he be received with the appropriate royal dignity, and sent to the court if, after investigation, his story was found to be true. The chronicle continues: The story that he related about himself at the Bandar [®Abbas] and which I have heard is as follows: ÒWhen Shahjahan ascended the throne, he ordered my execution. I was told by some of my friends about the plot, and also about the night when I was to be arrested. I had a slave of almost the same age as myself. I told him to sleep in my bed in my place. He hesitated initially, but I assured him that no harm would come to him, as it was I and not he who was the target of the emperor. The same night I  ed in the darkness towards the jungle, lived on mere grass for fourteen days while wandering in the wilderness, and hiding from people. Then, in the guise of a dervish, I joined a group of the faq”rs, and Ž nally arrived in Surat. There, I met the governor (h‰kim) of the port, and told him the tale of my plight. He was very helpful and friendly, guarded the secret, assisted me with people and money, and advised me to leave for Iran, in order to avoid any harm coming both to me and also to him. With the help of God I set out towards this royal court [Isfahan]. When I arrived at the port of Fars [Bandar ®Abbas], many thought I was a cheat and a liar. Some however noticed the signs of nobility in me, and believed in what I said. I remained suspended between truth and falsehood, fear and hope (sidq-o-kizb, $auf-oraj‰) for quite some time after I entered the province of Fars, until I encountered some people there who had seen me as Sultan on the way to Kashmir. They recognized me, and spread the news about me to others. I was then surrounded by numerous servants ($idmatg‰r‰n). People from India and Hindu traders came to meet me with whatever help they could extend to me in cash and kind (naqd-o-jins). Imam Quli Khan, in the Ž rst instance, believed in my statement, and showed me appropriate respect, but later due to the denial (ink‰r) of my identity by some traders, he grew suspicious, and accordingly, I declined in his esteem. He dispatched a report about me to the Shah stating his own doubts. In consequence, I was summoned to Isfahan, and there again, I remained suspended between hope and fear. The Shah however was very kind, saw to it that I was shown due respect, that my comfort was ensured and that my requirements were attended to, in keeping with the rules of hospitality.Ó

THE SHADOW SULTAN

109

This extended Ž rst-person account, in the voice of the man claiming to be Sultan Bulaghi, is a particularly valuable addition to our understanding of what transpired in Iran. It contains many familiar elements, including that of the time spent as a dervish, but also introduces an additional plot-theme, namely that of the substitution (of the slave for the prince).49 We thus see that even as Linhares sent his Jesuit representative to verify the identity of the claimant, in Iran too, the claims of this ÔBulaghiÕ were met with skepticism and put to the test. The Zail resumes its account by noting that Bulaghi now lived for some time as a guest of an important ofŽ cial, the n‰zir-i buyž t‰t. In due course, in the presence of the latter, a couple of Indians visited Bulaghi and veriŽ ed his claim. Hearing about him, many Multani merchants (tabaqa-i mult‰niy‰n) who then lived in large numbers in Isfahan, also collected at the princeÕs residence to pay him their respects (kž rnish). After about two months, the Shah at last expressed a desire to meet him. When Bulaghi had his audience with the Shah, he was honoured with the award of a special horse, and the Shah is said to have remarked: ÒVerily, high birth and greatness are evident on his countenance. From his utterances too, he seems very sedate and wise; he uses no words without a purpose (bi-mah‰sal).Ó The Shah then ordered that Bulaghi and Nabdi Mirza, the young boy who had arrived via Qandahar (and who was also allegedly a son of Sultan Khusrau), should meet. Perhaps this too was a test, though in this instance, the claims of not one but both were in question. As it happened, when the two met, they did not recognize each other. However, they then spent a night together, and mentioned the names of several women from the harems of Khusrau, Shahjahan and other princes, hinted at their good and bad qualities in order to test each otherÕs claims, and to assess how much each of them was familiar with the people who were mentioned. It then presently became clear to Sultan Bulaghi that this Nabdi was indeed his brother, and he embraced him with affection (‰#osh-i mihrb‰n”). The Shah ordered that both brothers should henceforth live together. Later, it is stated, whosoever came from India conŽ rmed that in Hindustan too the talk about the escape of Bulaghi was widespread, even though—write the authors of the Zail—BulaghiÕs story itself sounded rather incredible. It was really a wonder of the time that he escaped his fate, despite the fact that there

49 It is possible to trace the idea of substitution, and the smuggling out of a prince from a palace in a moment of siege, to a series of texts from the Vijayanagara period onwards in south India. We cannot explore this theme here, but see Narayana Rao et al. 2001, as also Sewell 1900: 222-31.

110

JORGE FLORES AND SANJAY SUBRAHMANYAM

was so vigilant a watch over and guarding of the princes opposed to Shahjahan by that emperorÕs well-wishers. What made the story stranger still was that Bulaghi had nurtured the ambition to be emperor himself, and a royal order had been issued to terminate his life. So, write the Iranian chroniclers, it is possible that some of the guards let their vigilance grow slack, or that they themselves helped him  ee and then killed the slave in order to escape the wrath of the king. Yet, God alone knew the truth of all this. But the chronicle has not Ž nished with its account of fugitive Mughal princes and claimants, for it now passes on to the case of the other prince whom we have encountered above, namely Mirza Baisunghar, the son of the Mughal prince Daniyal, who had passed away in the time of his father, Akbar. It is reported in the Zail that after Sultan SalimÕs death, Baisunghar had joined the supporters of Shahryar, when the latter had fought against Sultan Bulaghi in the neighbourhood of Lahore. But Shahryar was defeated, and his army was shattered, so that Baisunghar managed to  ee towards Kabul with the help of some of ShahryarÕs soldiers. On his way there, he heard about the plight of Bulaghi and the establishment of ShahjahanÕs power. He then thought it unwise to go to Kabul, turned towards Balkh instead, and lived there a while among the Uzbeks and their ruler, Muhammad Khan. But since he was devoid of wisdom, and was also a mere fugitive, he could be of little real use to the Uzbeks. He was rather ill tempered as well and the Uzbeks, for their part, grew indifferent and cold to him. So, with the intention of leaving for the Safavid ShahÕs court, he is reported to have left for Qandahar, where he met ®Ali Mardan Khan, and stayed there as his guest for some time before he set out for the ShahÕs court with some of the governorÕs men. Shah SaŽ , it is reported, sent a large number of people to receive him in style. All this had taken place before the arrival in Iran of the man claiming to be ÔBulaghi.Õ Indeed, when the latter arrived, Baisunghar refused to accept his claim, and since he had a surer position in the esteem of the Shah, he even misbehaved with Bulaghi. Unseemly exchanges took place between the two who claimed to be Mughal princes, and once, in an assembly organized for them to meet and in which the Shah was especially to be present, Baisunghar is reported to have used impolite and unjust words for the other prince, to have shouted at him in anger, and called him a fraud. But Bulaghi maintained his calm, ignored his invectives, and attributing all that he said to his illiteracy and madness ( jahl wa junž n), then left the assembly. Hearing all about this, the Shah also declined to come there. Still later, at the time of the ShahÕs march against the Georgian Tahmurs, who had been instigated to rebel against the Safavid sovereign by the brother of Imam Quli Khan and others, Baisunghar sought the ShahÕs permission to leave his domains. The Shah had no hesitation in allowing him to go

THE SHADOW SULTAN

111

back. At this time, Baisunghar submitted that he wanted to return to India in the hope of support to him from the Sultans of the Deccan. He received nothing else besides the hundred tom‰ns that the Shah had earlier sent for him for his expenses in Iran. The Zail notes that he was an able young man, but much too proud of his bravery and royal position. He lived during his stay in Iran like an arrogant upstart, with an unbecoming comportment, had little wisdom or sagacity, and was also intolerant and narrow-minded. The extended and valuable account in the Zail does leave a number of crucial questions open. To begin with, it does not by any means resolve our doubts about the genuineness of the man who claimed to be Bulaqi or Bulaghi, as the Zail itself admits. There is notably the problem of the relationship between Bulaqi and Baisunghar, the fact that the latter called the former a hoax or fraud, while other evidence (as we have noted above, from his later career with the Ottomans) suggests that it was in fact Baisunghar who was a fraud. If so, why did the Bulaqi of the Safavid court (if he was genuine) not unmask him? A further set of problems concerns the relationship between the man encountered by Padre Le‹o, and the man who arrived in Bandar ®Abbas from Surat. Were they the same or not? In the case of the claimant in Iran, how do we come to terms with the fact that large numbers of people who had known Bulaqi in the Mughal court came forward to identify him? In any event, it appears that after initial problems, the claimant was accepted by the Safavids as genuine, and we encounter him periodically in the chronicles of the epoch, both in later sections of the Zail, and elsewhere.50 Besides the narratives in the chronicles, it may also be useful to look brie y at the correspondence that was exchanged between the Safavid monarchs Shah SaŽ and Shah ®Abbas II, and Sultan Bulaqi. The Ž rst set of such letters come to us from around 1632, thus shortly after the arrival of the claimant in Iran. Here, Shah SaŽ assures the man claiming to be the Mughal prince that he will look after his welfare, and that he is sending a certain Zu¾lŽ qar Beg Qurchi Baiburdlu to him in this connection. The letter addresses ÔBulaghiÕ as a genuine Mughal prince, with titles such as sh‰hz‰da and far‰zinda-i sar”r-i saltanat (Islam 1979: 243-44). We also possess a letter from the same period, written by Bulaqi/Bulaghi himself to Shah SaŽ , though it is unclear whether it precedes the above letter, or is in response to it. In this, the letter-writer terms himself Bulaghi p‰dsh‰hz‰da-i Hindust‰n (Òdescendant of the emperor of HindustanÓ), and refers to his own illustrious ancestors from Timur to Humayun (who had

50

See, for example, Munshi and Yusuf 1938: 230-40, passim.

112

JORGE FLORES AND SANJAY SUBRAHMANYAM

sought help from Shah Tahmasp in the sixteenth century). An interesting feature of this letter is its reference to Shi®i holy sites in Iran, whether Karbala, Ardabil or Najaf, to which the letter-writer claims to have a special attachment. Humble in tone, the purpose of the letter is clearly to  atter Safavid pretensions and suggest that the author is not far from becoming a Shi®i himself. The writer also states that he hopes to put an end to the oppression of the people by Shahjahan with the help of Shah SaŽ (Islam 1979: 245-46). Portuguese sources, as we have seen above, suggest that in the next few years, the Safavids did give this Bulaghi some resources, allowing him to make a minor nuisance of himself on the Mughal-Safavid frontier. This seems to have lasted only into the mid-1630s however, after which the prince (whether real or false) would appear to have retired to enjoy whatever revenues the Safavids had given him. The Safavid support for this putative prince was clearly to continue for a time. Still in the time of Shah SaŽ , we have a further exchange of letters, such as one in November 1640, concerning gifts sent by Sultan Bulaghi to the Shah, including a shield and a sword. The letters of this period suggest a certain familiarity and even intimacy between the Mughal ÔprinceÕ and the Safavid ruler, and we know that at much the same time, this Bulaghi accompanied Shah SaŽ on a visit to Ashraf in the region of Mazandaran (Islam 1979: 276-77). 51 In the same year, 1640, Augustinian sources from Iran also inform us of a con ict between members of their order and Òthe secretary of Bolaquin, son or nephew of the Great Mogor, successor to that kingdom,Ó over some matters of conversion in Isfahan. Here, the Mughal prince and his entourage are presented as staunch Muslims; Òthe Mogors,Ó the Augustinian writer declares, Òprofess that they are great observers of their sect.Ó 52 Two years later, when Shah SaŽ died and was succeeded by Shah ®Abbas II (r. 1642-66), Bulaghi was amongst those who wrote a letter of condolence and congratulations to the new ruler, leading to a further friendly correspondence between the two as well as an exchange of gifts (Islam 1979: 281-84). In one of these letters, a ruq®a drafted by Mirza Muhammad Riza, the recipient Bulaghi is given particularly high titles such as Niz‰m al-saltanat waÕl-$il‰fat, suggesting that by this time any residual doubts concerning his identity had been resolved. These letters must have been written at the time that Bulaghi was resident at Qazwin, and met his various European visitors—as has been noted above. We are also aware that Shah ®Abbas II probably took this ÔBulaghiÕ along when he mounted his expedition to recover Qandahar from the Mughals in 1648. And when the false rumours of the death

51 52

Also see the account of the visit to Ashraf, in Munshi and Yusuf 1938: 237-38. See the report from Pietro dei Santi reproduced in Alonso Va–ez 1962: 319.

THE SHADOW SULTAN

113

of Shahjahan reached Iran in 1658, the same Bulaghi apparently grew keen to ask for the aid of the Safavid monarch in order to regain the Mughal throne, and rushed to Isfahan from Qazwin.53 This would seem to be the last trace we possess of ÔBulaghiÕ in the Safavid documents, though we do Ž nd some correspondence relating to revenue-grants to his son, Sultan Khusrau (Islam 1979: 424-25). THE CONTINUITY OF TRADITION As for the Portuguese, their deep interest in Bulaqi did not really survive the disappointment of the meeting with Padre Le‹o in December 1630. Even so, Bulaqi was to have one last encounter with the Count of Linhares. On 18 July 1634, the Portuguese viceroy received news from Surat, through the Jesuit Paulo Reim‹o. He was informed by his correspondent Òthat the Mogor king was in Laor to prepare a response to the Persian regarding the affairs of Bulaquim, to whom many of the MogorÕs Ž dalgos had gone over, as was the case with Mirzamadafar, who by pretending to depart for Mecca, embarked with a great quantity of riches in a ship in Surrate, and as soon as he reached the high seas, he expelled the pilgrims to Meca in some shipÕs boats, and those who resisted were thrown into the sea, and he then went off to Persia to join BulaquimÓ (Linhares 1937-43, ii: 146-47). A week later, on 25 July, the viceroy once more had news from Surat through the same Padre Reim‹o. This time, it was stated that Shahjahan was in Lahore en route to Kashmir, where he hoped to confront Bulaqi. It was claimed that the Mughal prince had decided openly to challenge his uncle, sending him an embassy, Òand as a present, a sword and a bed made of gold, and the embassy told him to choose the one that was more to his taste: the bed on which he could take rest once he had handed over the kingdom, or the sword with what he could expect from it. The Mogor felt this deeply, and wished to kill the envoys, but [instead] had contempt for them, and without deigning to reply prepared himself for war, since the Persian has given much power and help to BolaquiÓ (Linhares 1937-43, ii: 149-50). Then, on 19 January 1635, we have the last echo of the legend of Sultan Bulaqi in the Portuguese sources. That evening, the viceroy Linhares read with some attention a letter that had just arrived from Bijapur, between two unknown

53

Qazwini 1951: 234-35, section entitled ÔSharh-i ‰madan Sult‰n Bul‰gh” ba D‰r alSaltanat Isfah‰n.Õ These moves seem to have been related to a Safavid project to support the prince Murad Bakhsh in 1657-58; cf. Quamruddin 1974: 117.

114

JORGE FLORES AND SANJAY SUBRAHMANYAM

correspondents (the recipient clearly is a high-placed Portuguese ofŽ cial, perhaps the Secretary of the Estado).54 Amongst other questions, there was an outline of a proposed alliance between Bulaqi and the Estado da êndia: Sult‹o Bolaquim asks His Excellency [Linhares] for a royal safe-conduct, [issued] in the name of His Majesty, so that he can stay securely in Goa and from there consult with the captains of the kingdom of the Mogor, and that another such [safe-conduct] should be given to his brother who also escaped,55 and is with the King of Persia, and once this has been agreed upon, that the lord viceroy should give him passage through one of the ports that he could designate, and to this end he would meet all the accords and conditions that the lord viceroy might desire, carrying this matter out at once. Your Grace should send this messenger for I have agreed to this with the Solt‹o Bolaquim, and I do not write to His Excellency [directly] for I do not know if he will wish to enter into this affair, and I greatly desire that through the mediation of Your Grace it may come to the knowledge of the lord viceroy, and if he wishes to do this, that I may be informed immediately of the response of the lord viceroy, so that it may be written out in [proper form]. Your Grace should carry out this affair with rapidity. (Linhares 1937-43, ii: 265.)

But the circumstances had changed drastically from those in late 1630. At that time, the Count of Linhares believed in a possible alliance and, in view of the varied origins of the different pieces of intelligence that he had before him, his main preoccupation was with the authenticity of the person, and then, with the feasibility of a political manoeuvre to unseat Shahjahan. Now, in early 1635, Linhares was totally unbelieving. His own viceroyalty was in its Ž nal phase, and the viceroy contents himself by noting a last  eeting thought on the subject, two days before concluding his diary for that year and sending it back to Europe: ÒI hold this to be a fable, and this Bolaquim seems to be for the Mogores, another king Sebasti‹o for the Portuguese.Ó It was the second time that the history of the Ôhidden oneÕ (o encoberto) was to cross that of the Mughal emperors. If one is to follow the Jesuits who lived in the court of Akbar, that ruler (who was ShahjahanÕs grandfather) was much impressed by the Ž gure of Dom Sebasti‹o: Òand he feels deeply for the case of the king Dom Sebasti‹o, and when he talks of that affair, he praises the forceful spirit of the king Dom Sebasti‹o.Ó 56 Half a century later, we have this The letter could have been written by one of a number of Portuguese agents at the Bijapur court, such Vicente Ribeiro, Fern‹o Lopes, Fr. Jo‹o da Rocha, or a certain Jorge da Costa, who had just arrived in Bijapur as LinharesÕs envoy in late December 1634. 55 Is this a reference to the mysterious Mirza Nabdi, whom we have encountered in the Zail? According to Mughal chronicles, Bulaqi had only two brothers, Gurshasp (who was killed in January 1628), and Buland Akhtar, who died young; the latter name seems closer to Nabdi, but the association is still far-fetched. 56 Excerpt from a letter written by R. Acquaviva, A. Monserrate and F. Henriques to the Portuguese captain of Daman, Fatehpur Sikri, March-April 1580; published in Wicki 1972: 23. 54

THE SHADOW SULTAN

115

interesting comparison from the pen of D. Miguel de Noronha. There are of course various parallels between the different Bulaqis and the different (and false) claimants to being D. Sebasti‹o in the sixteenth century, such as the King of Penamacor, the King of Ericeira, the pastrycook of Madrigal, and the socalled Calabrian. 57 The omnipresence of the legend and the central place of rumour in both cases are fed by the wandering presence of a veiled personage. The tragic component of both stories and of both personages can even reputedly move foreign observers, as the Portuguese were moved by the misfortunes of Bulaqi, and Akbar by those of D. Sebasti‹o. In both cases, we also see the same problems of legitimacy and illegitimacy, the same problem of a popular movement, and the same lack of stability in an early phase of a ruler who had seized power in a disorderly succession. The case of Sultan Bulaqi was certainly not the last of its kind in Mughal India. We Ž nd other instances around the brother of Aurangzeb, Shah Shuja® in the 1660s and 1670s, and in the early eighteenth century, such cases of persons pretending to be Mughal princes continue to proliferate, perhaps owing to the fact that very large numbers of princes of royal blood now in fact existed. But certain Ž gures seem to attract particular attention. Thus, as late as 1717, we Ž nd an instance of a man (apparently really called ®Aqibat Mahmud), who appeared in the Deccan claiming to be the Mughal prince Muhammad Akbar, and who was eventually imprisoned by the Mughal governor of Arcot (see Subrahmanyam 2001: 128-29). Mughal chroniclers equally mention other incidents concerning the same prince; in the early eighteenth century, the Maratha Raja Shahu is reported to have raised ÔcommotionÕ and sought strength by picking up a certain Mu®in-ud-Din who claimed to be a son of Prince Akbar; while a few years later, during the con icts between the emperor Farrukhsiyar and the Sayyid brothers, Husain ®Ali Khan deŽ ed the emperor by returning to court from the Deccan, while claiming to have a son of Muhammad Akbar with him. On this occasion, we are told that Òthe emperorÕs gall-bladder melted with fearÓ on account of the rival claimant.58 And later still, after the fall of the Safavid dynasty, men claiming to be princes of that house would appear periodically in India, where on some occasions at least their claims were upheld, as we see with AbuÕl Fath Sultan Muhammad Mirza Safavi in the late eighteenth century (Rota 1996; 1999). Again, in the 1790s, the wandering Mughal prince Mirza ®Ali Bakht ÔAzfariÕ was disconcerted to arrive in the eastern Indian town of

57

On this subject, see Pires 1982: 59-64; Hermann 1998, and the classic work by DÕAntas 1866. 58 These incidents may be found in Kamwar Khan 1980: 228, 254.

116

JORGE FLORES AND SANJAY SUBRAHMANYAM

Murshidabad, for he found there to his chagrin an impostor who was already in place, claiming to be none other than Mirza Azfari himself (Azfari 1957: 102103). 59 Did such men know the story of Sultan Bulaqi, from almost two centuries before? Or had it passed into folk memory as a more generic theme, even as the legend of Bulaqi itself drew from the same corpus of beliefs and legends pairing kings and renouncers, faq”rs who might reveal themselves to be rulers? The episodes that we have discussed in the preceding pages can, as we have suggested at the outset, be read as part of a larger history of royal ÔdoublingÕ in the early modern world, or—as one recent author has put it—as part of the question of how ÔidentityÕ and ÔcredulityÕ came together in the period, which is termed Òa golden age for impostors and pretendersÓ (Eliav-Feldon 1999: 203). Yet, as we have seen, credulity was always tempered with certain down-to-earth notions of the empirical testing of claims, even if the criteria used differed from one context to another. If the Portuguese viceroyÕs method was simple enough, and involved sending an accredited eyewitness, the Safavids judged their own claimant by a more complex means, having to do with his civility and mastery of the appropriate etiquette for a prince, including the notion that Òhigh birth and greatness [were] evident on his countenance.Ó The problem could also Ž nd resolution of another sort if one were cynical enough: we observe this in Philip IVÕs view that a plausible enough impostor was about as good as the genuine article. As regards the Safavids, we may have some doubts as to whether they thought the man who claimed to be Sultan Bulaghi really was what he declared, for it is interesting that they chose to keep him not at the court in Isfahan (where he would have to meet visiting Mughal dignitaries), but rather in Qazwin. The difference in the treatment of ÔBulaghiÕ in this period, and Prince Muhammad Akbar later in the seventeenth century is thus marked. Still, true or false, so long as an element of doubt subsisted, such a claimant could always play a role on the chessboard of inter-state politics in the period. To this extent, the Bulaghi who spent the 1630s, 1640s and 1650s on a comfortable Safavid pension may not have eventually mounted the Mughal throne; but if he was indeed an impostor, he must be deemed a successful one. This success was not predicated though on simple credulity, or on the absence of an empirical spirit, but because ambiguity, bluff and doubt always had a place in the political systems of the time.

59

For an Urdu translation, Azfari 1937.

THE SHADOW SULTAN

APPENDIX Obverse:

Reverse:

Silver rupee coin of Sultan Dawar Bakhsh, 1037 H., Lahore mint, Ashmolean Museum, Oxford. ARCHIVAL REFERENCES AHU BA BL BNF BNL HAG IAN/TT

Arquivo Hist—rico Ultramarino, Lisbon. Biblioteca da Ajuda, Lisbon. British Library, London. Biblioth que Nationale de France, Paris. Biblioteca Nacional de Lisboa, Lisbon. Historical Archives of Goa, Panaji. Instituto dos Arquivos Nacionais/Torre do Tombo, Lisbon.

117

118

JORGE FLORES AND SANJAY SUBRAHMANYAM

BIBLIOGRAPHY Alam, Muzaffar, and Sanjay Subrahmanyam. 2002. Witnessing transition: Views on the end of the Akbari dispensation. In The Making of History: Essays presented to Irfan Habib, eds. K.N. Panikkar et al., pp. 104-40. New Delhi: Tulika Publishers. Ali, M. Athar. 1978. Towards an interpretation of the Mughal Empire. Journal of the Royal Asiatic Society of Great Britain and Ireland 1: 38-49. Alonso Va–ez, Carlos, OSA. 1962. Stato delle missioni agostiniane nelle Indie orientali secondo una relazione inedita del 1640. Analecta Augustiniana 25: 291-325. Arthurson, Ian. 1994. The Perkin Warbeck conspiracy, 1491-1499. Stroud: Alan Sutton. Aubin, Jean. 2000. D. Jo‹o II et Henry VII. In Le Latin et lÕAstrolabe, Vol. II: Recherches sur le Portugal de la Renaissance, son expansion en Asie et les relations internationals, by J. Aubin, 83-92. Paris: Fondation Calouste Gulbenkian. ——. 1971. LÕAmbassade de Greg—rio Pereira Fidalgo ˆ la cour de Ch‰h Solt‰n-Hosseyn, 1696-1697. Lisbon: ComitŽ National Portugais pour la celebration du 2.500e anniversaire de la fondation de la monarchie en Iran. Azfari, Mirza ®Ali Bakht. 1957. W‰qi®‰t-i Azfar”, eds. T. Chandrasekharan and Syed Hamza Hussain Omari. Madras: Government Oriental Manuscripts Library. ——. 1937. W‰qi®‰t-i-Azfar”, Urdu trans. Muhammad Husain Mahvi Siddiqi. Madras: University of Madras. Beach, Milo C. 1978. The Grand Mogul: Imperial Painting in India, 1600-1660. Williamstown, MA: Sterling and Francine Clark Art Institute. Beach, Milo C., and Ebba Koch. 1997. King of the World, The Padshahnama: An Imperial Mughal Manuscript from the Royal Library, Windsor Castle. London: Thames and Hudson. Begley, Wayne E. 1979. The myth of the Taj-Mahal and a new theory of its symbolic meaning. Art Bulletin 3: 7-37. Begley, W.E., and Z.A. Desai, eds. 1990. The Shah Jahan Nama of ®Inayat Khan: An abridged history of the Mughal Emperor Shah Jahan, compiled by his royal librarian, trans. A.R. Fuller. Delhi: Oxford University Press. Bernier, Fran ois. 1934. Travels in the Mogul Empire, AD. 1656-1668, trans. Irving Brock and Archibald Constable, ed. Vincent A. Smith. London: Oxford University Press. Bhadra, Gautam. 2003. J‰l r‰j‰r galpa. Kolkata: Ananda Publishers. Camps, Arnulf. 1957. Jerome Xavier S.J. and the Muslims of the Mogul Empire: Controversial and missionary activity. Schšneck-Beckenried: Nouvelle Revue de Science Missionaire Suisse. Chartier, Roger. 1994. The Order of Books: Readers, authors, and libraries in Europe between the fourteenth and eighteenth centuries, trans. Lydia Cochrane. Stanford: Stanford University Press. Chatterjee, Partha. 2002. A Princely Impostor?: The Kumar of Bhawal and the Secret History of Indian Nationalism. New Delhi: Permanent Black. Couto, Diogo do. 1999. DŽ cada Quarta da çsia, ed. M. Augusta Lima Cruz, 2 Vols. Lisbon: Imprensa Nacional. DÕAntas, Miguel. 1866. Les faux Don SŽ bastien: ƒtude sur lÕhistoire de Portugal. Paris: Auguste Durand. Delmerick, J.G. 1884. On a silver coin of Dawar Bakhsh. Proceedings of the Asiatic Society of Bengal 20: 60. Desser, David. 1983. The Samurai Films of Akira Kurosawa. Ann Arbor: UMI Research Press. Disney, Anthony. 2002. The Fiscal Reforms of Viceroy Linhares at Goa. Anais de Hist—ria de AlŽ m-Mar 3: 259-75. ——. 1992. On attempting to write an early modern biography: My encounter with the life of Dom Miguel de Noronha, Fourth Count of Linhares (1588-1656). Indica 29/2: 89-106.

THE SHADOW SULTAN

119

Eliav-Feldon, Miriam. 1999. Invented Identities: Credulity in the Age of Prophecy and Exploration. Journal of Early Modern History 3/3: 203-32. Farooqi, Naimur Rahman. 1989. Mughal-Ottoman Relations (A study of political and diplomatic relations between Mughal India and the Ottoman Empire, 1556-1748). Delhi: Idarah-i Adabiyat-i Dilli. Flores, Jorge Manuel. 2002. Relic or Springboard?: A Note on the ÔRebirthÕ of Portuguese Hughli, ca. 1632-1820. The Indian Economic and Social History Review 39/4: 381-95. Foster, William, ed. 1911. The English Factories in India, 1634-1636. Oxford: Clarendon Press. Foucault, Michel. 1987. What Is an Author? In Twentieth-Century Literary Theory, eds. Vassilis Lambropoulos and David Neal Miller, 124-42. Albany: SUNY Press. Fraser, Charles. 1832. Annals of the Turkish Empire, from 1591 to 1659. London: Oriental Translation Fund. Grafton, Anthony. 1990. Forgers and Critics: Creativity and Duplicity in Western Scholarship. Princeton: Princeton University Press. Gul-Badan Begam. 1974. Hum‰yž n-N‰ma, ed. and trans. Annette S. Beveridge. Reprint, Lahore: Sange-Meel Publications. Gurevich, Aaron. 1995. The origins of European individualism, trans. Katherine Judelson. Oxford: Basil Blackwell. Herbert, Thomas. 1972. Travels in Persia, 1627-29, ed. William Foster. Reprint Freeport, N.Y.: Books for Libraries Press. ——. 1663. Relation du voyage de Perse et des Indes Orientales. Paris: Jean Dupuis. Hermann, Jacqueline. 1998. No reino do Desejado: A constru ‹o do sebastianismo em Portugal, sŽ culos XVI e XVII. S‹o Paulo: Companhia de Letras. Hern‡ndez, çngel Santos. 1958. Jeronimo Javier S.J. Ap—stol del Gran Mogol y Arzobispo electo de Cranganor, en la India, 1549-1617. Pamplona: Instituci—n Pr’ncipe de Viana. Hoyland, J.S., and S.N. Banerjee. 1975. The Empire of the Great Mogol: De LaetÕs Description of India and Fragment of Indian History. Reprint, Delhi: Idarah-i Adabiyat-i Dilli. Husain, Afzal. 1999. The Nobility under Akbar and Jahangir: A study of family groups. New Delhi: Manohar. Islam, Riazul. 1979. A Calendar of Documents on Indo-Persian Relations (1500-1750), Vol. I. Tehran: Iran Culture Foundation. Jahangir, Nur al-Din Muhammad. 1999. The Jahangirnama: Memoirs of Jahangir, Emperor of India, trans. Wheeler M. Thackston. New York: Oxford University Press. ——. 1864. Toozuk-i Jahangeeree, ed. Syud Ahmud. Aligarh: The editor. Kamboh, Muhammad Salih. 1967-72. Am‰l-i S‰lih, eds. Ghulam Yazdani and Wahid Quraishi, 3 Vols. Lahore: Majlis-i Taraqqi-yi Adab. Kamwar Khan, Muhammad Hadi. 1980. Tazkirat-us-Sal‰t”n Cha#t‰, ed. Muzaffar Alam. Bombay: Asia Publishing House. Khwaja Kamgar Husaini. 1978. Ma¾asir-i Jahangiri: A contemporary account of Jahangir, ed. Azra Alavi. Bombay: Asia Publishing House. Linhares, Conde de. 1937-43. Di‡rio do 3o conde de Linhares, vice-rei da êndia, 2 Vols. Lisbon: Biblioteca Nacional. Manucci, Niccolao. 1990. Storia do Mogor or Mogul India, trans. William Irvine, 4 Vols. Reprint, Delhi: Low Price Publications. ——. 1986. Storia del Mogol di Nicol˜ Manuzzi veneziano, ed. Piero Falchetta, Vol. I. Milan: Franco Maria Ricci. McLane, John R. 1993. Land and Local Kingship in Eighteenth-Century Bengal. Cambridge: Cambridge University Press. Mirza Nathan. 1936. Baharistan-i-Ghaybi: A history of the Mughal wars in Assam, Cooch Behar, Bengal, Bihar and Orissa during the reigns of Jahangir and Shahjahan, trans. M.I. Borah, 2 Vols. Gauhati: Narayani Handiqui Historical Institute.

120

JORGE FLORES AND SANJAY SUBRAHMANYAM

Morris, Colin. 1972. The Discovery of the Individual, 1050-1200. Toronto: University of Toronto Press. Mundy, Peter. 1914. The Travels of Peter Mundy, in Europe and Asia, 1608-1667, ed. Richard C. Temple, vol. II (Travels in Asia, 1628-1634). London: The Hakluyt Society. Munshi, Iskandar Beg. 1978-86. History of Shah ®Abbas the Great (Tarikh-i ®Alamara-ye ®Abbasi), 3 Vols., trans. Roger M. Savory. Boulder: Westview Press. ——. 1956-57. T‰r”$-i ®ålam‰r‰-yi ®Abb‰s”, 2 Vols., ed. Iraj Afshar. Tehran: Ma¾assasa-yi Matbu¾at-i ®Amir Kabir. Munshi, Iskandar Beg, and Muhammad Yusuf. 1938. Zail-i T‰r”$-i ®ålam‰r‰-yi ®Abb‰s”, ed. Suhaili Khwansari. Tehran: Chapkhana-yi Islamiya. Mustafa Na®ima. 1967-69. Na”m‰ T‰r”h”, 6 Vols. Istanbul: Zuhuri Danishman. Narayana Rao, Velcheru, David Shulman and Sanjay Subrahmanyam. 2001. Textures of Time: Writing History in South India, 1600-1800. New Delhi: Permanent Black. Olearius, Adam. 1679. Relation du voyage DÕAdam Olearius en Moscovie, Tartarie, et Perse, augmentŽe en cette nouvelle edition de plus dÕun tiers, particulierement dÕune seconde voyage de Mandelslo aux Indes Orientales, 2nd edition, 2 Vols. Paris: Jean Dupuis. Pearson, M.N. 1976. Shivaji and the Decline of the Mughal Empire. Journal of Asian Studies 35: 221-35. Pires, Ant—nio Machado. 1982. D. Sebasti‹o e o Encoberto. 2nd edition, Lisbon: Funda ‹o Calouste Gulbenkian. Pissurlencar, P.S.S. ed. 1953. Assentos do Conselho do Estado, 1618-1633, Vol. I. GoaBastor‡: TipograŽ a Rangel. ——. 1952. Agentes da diplomacia portuguesa na êndia (hindus, mu ulmanos, judeus e parses). Bastor‡-Goa: TipograŽ a Rangel. ——. 1930. A êndia em 1629: Rela ‹o dos Reis Visinhos do que por ora passa e cont‹o. Boletim do Instituto Vasco da Gama 7: 52-61. ——. 1928. Prince Akbar and the Portuguese. Bengal Past and Present 35: 163-69. Prakash, Om. 1998. Archival source material in the Netherlands on the history of Gujarat in the early modern period. In Sources europŽ ennes sur le Gujarat, ed. Ernestine Carreira, 141-51. Paris: LÕHarmattan. Qazwini, Muhammad Tahir Wahid. 1951. ®Abb‰s N‰ma y‰ Sharh-i Zindag‰n”-yi 22-s‰la-yi Sh‰h ®Abb‰s-i s‰n” (1052-1073), ed. Ibrahim Dihgan. Arak: Dawudi. Quamruddin, Mohammad. 1974. Life and Times of Prince Murad Bakhsh (1624-1661). Calcutta: The author. Quinn, Sholeh A. 2000. Historical writing during the reign of Shah ®Abbas: Ideology, imitation, and legitimacy in Safavid chronicles. Salt Lake City: University of Utah Press. Rahim, M.A. 1961. History of the Afghans in India, A.D. 1545-1631, with especial reference to their relations with the Mughals. Karachi: Pakistan Publishing House. Rego, Ant—nio da Silva, ed. 1974. As Gavetas da Torre do Tombo, Vol. 10, Lisbon: Centro de Estudos Hist—ricos Ultramarinos. ——. ed. 1963. Documenta ‹o Ultramarina Portuguesa, Vol. 3. Lisbon: Centro de Estudos Hist—ricos Ultramarinos. Richards, John F. 1993. The Mughal Empire. The New Cambridge History of India, Vol. I. 5. Cambridge: Cambridge University Press. Rizqullah Mushtaqi, Shaikh. 2002. W‰qi®‰t-i Musht‰q”, eds. Iqtidar Husain Siddiqui and Waqarul Hasan Siddiqi. Rampur: Raza Library. ——. 1993. Waqi®at-e-Mushtaqui of Shaikh Rizq Ullah Mushtaqui: A Source of Information on the Life and Conditions in Pre-Mughal India, trans. Iqtidar Husain Siddiqui. New Delhi: Northern Book Centre. Rota, Giorgio. 1999. The Man who would not be King: AbuÕl Fath Sultan Muhammad Mirza Safavi in India. Iranian Studies 32/4: 513-35. ——. 1996. Un SoŽ tra i Nababi: LÕultimo Safavide a Lucknow. In Ex libris Franco Coslovi, eds. Daniela Bredi and Gianroberto Scarcia, 337-80. Venice: Poligrafo.

THE SHADOW SULTAN

121

Saksena, Banarsi Prasad. 1968. History of Shahjahan of Dihli. Reprint Allahabad: Central Book Depot. Saramago, JosŽ . 2002. O homem duplicado. Lisbon: Caminho. Savory, Roger M. 1971. A Curious Episode in Safavid history. In Iran and Islam: In Memory of the late Vladimir Minorsky, ed. C.E. Bosworth, 461-73. Edinburgh: Edinburgh University Press. Sen, Surendranath. 1949. Indian travels of Thevenot and Careri. New Delhi: National Archives of India. Sewell, Robert. 1900. A Forgotten Empire (Vijayanagar): A contribution to the history of India. London: S. Sonnenschein & Co. Seyyidi ®Ali Re¾is. 1999. Le miroir des pays: Une anabase ottomane ˆ travers lÕInde et lÕAsie centrale, trans. Jean-Louis BacquŽ -Grammont. Paris: Actes Sud. ——. 1999a. Mir¾‰tŸ¾l-Mem‰lik, ed. Mehmet Kiremit. Ankara: TŸrk Dil Kurumu. Shahnawaz Khan, Nawwab Samsam al-Daula. 1911-52. Ma¾asir-ul-Umara, being biographies of the Muhammadan and Hindu ofŽ cers of the Timurid sovereigns of India from 1500 to about 1780 A.D., 3 Vols, trans. H. Beveridge and Baini Prashad. Calcutta: Asiatic Society of Bengal. ——. 1888-90. Ma¾asir al-Umara, eds. Maulavi ®Abdur Rahim and Maulavi Mirza Ashraf ®Ali, Vols. 1 and 2. Calcutta: Asiatic Society of Bengal. Spence, Jonathan D. 2002. Treason by the Book. New York: Viking. Subrahmanyam, Sanjay. 2001. Penumbral Visions: Making Polities in Early Modern South India. Delhi: Oxford University Press. ——. 1998. A Cr—nica dos Reis de Bisnaga e a Cr—nica do Guzerate: Dois Textos IndoPortugueses do sŽ culo XVI. In Os Construtores do Oriente Portugus, ed. Jorge Manuel Flores, 131-54. Lisbon-Oporto: CNCDP. ——. 1993. Through the looking glass: Some comments on Asian views of the Portuguese in Asia, 1500-1700. In As rela ›es entre a êndia portuguesa, a çsia do Sueste e o Extremo Oriente. Actas do VI Semin‡rio Internacional de Hist—ria Indo-Portuguesa, eds. A. Teodoro de Matos and L.F. Thomaz, 377-403. Macau-Lisbon: The editors. Szv‡k, Gyula. 2000. False tsars, trans. Peter Daniel. Boulder: Social Science Monographs. Tavernier, Jean-Baptiste. 1995. Travels in India by Jean-Baptiste Tavernier, trans. V. Ball and William Crooke, Vol. 1. Reprint Delhi: Oriental Books Reprint Corporation. Wicki, Josef, ed. 1972. Documenta Indica, Vol. 12. Rome: Monumenta Historica Soc. Iesu. Wright, H. Nelson. 1905. Dawar Bakhsh. Journal and Proceedings of the Asiatic Society of Bengal, 2nd Series, 1: 127-28. Wroe, Ann. 2003. Perkin: A story of deception. London: Chatto.

REVIEWS Zainab Bahrani, The Graven Image: Representation in Babylonia and Assyria. Philadelphia: University of Pennsylvania Press, 2003. Pp. x + 242, 26 ills. ISBN 0-8122-3648-3. Ancient Near Eastern studies originated as a byproduct of the colonial rule that European governments imposed upon the Middle East during the nineteenth and twentieth centuries, and interest in ancient monuments can already be traced in the writings of early European travelers who visited the Middle East. Monumental remains and uncanny inscriptions enhanced the curiosity to learn more about those peoples mentioned in the Bible and in Classical works. Scholars of ancient Mesopotamia are familiar with the romantic and adventurous origins of their discipline. Stories and anecdotes of early travelers and diggers usually appear in works dealing with the foundations of Assyriology.1 The history of the Ž eld, however, is generally divorced from an evaluation of the in uence that Western political, economic, intellectual, and religious history has had on the recovery of an ancient Mesopotamian ethos. Similarly, approaches to ancient artistic remains usually dissociate writing from visual representations. In The Graven Image, Zainab Bahrani studies the ways in which scholarly tradition rooted in Western intellectual practices has modeled a particular perception of Near Eastern art. She further provides an alternative way to understand ancient and modern representations of Assyrian and Babylonian art, and she challenges standard interpretations of art historians. The Graven Image focuses on the concept of representation in the ancient Near East and on the Assyro-Babylonian practice of combining writing and visual representation for the production of images as a form of essential presence. The main thesis of the book maintains that the image takes the place of the real because the image carries an essential and conjuring presence. The study is also concerned with the practice of image making in academic discourse. This is an interdisciplinary approach encompassing theoretical knowledge from art history, anthropology, ethnography, historiography, semiology, cultural studies, post-structural literary theory, political practices, psychology, and philosophy. Bahrani incorporates Jacques DerridaÕs deconstruction and anti-Platonism, as well as the contributions of thinkers such as Edward Said, Timothy Mitchell, Michael Taussig, John and Jean Comaroff, Michel de Certeau, Hayden White, and Jacques Lacan, among others. Following the theoretical stance of post-processual archaeology, post-structuralist, deconstructivist, and postcolonial scholarship, the author asserts that ancient contexts do not simply emerge from the available data. Rather, they are forged by means of interpretative decisions and assumptions of modern scholars. Since the reconstruction of contexts makes culture an enunciative site, Bahrani regards the art historian as a translator of culture or ethnographer. The introduction explains that the Graven Image was not written as a seamless narrative but as a series of essays confronting unexamined theoretical assumptions and proposing new readings of Assyro-Babylonian art. The Ž rst three chapters deal with the ways in which colonial and post-colonial discourses have in uenced the study of ancient Near Eastern art, and have applied categories alien to Mesopotamian ontology. Chapters four through seven analyze speciŽ c examples of Assyro-Babylonian representation: cuneiform writing, the ×almu, image abduction, and the altar of Tukulti-Ninurta. The Ž rst chapter refers to aesthetics, epistemics, and the notions of race, culture and antiquity, all of them related to the civilizing mission of imperialism. The discussion also includes Orientalism, Hellenism and barbarism, the idea of a Òtranscategorical primitive,Ó and the natural history of art, all of which contributed to reinforce an alleged European superiority. The second chapter shows that Western characterizations of the Orient made of it an ÒextraterrestrialÓ entity, and a land governed by despots. Bahrani challenges the idea of Mesopotamia as a discursive formation by questioning the ontological concept of Mesopotamia. Chapter three examines two sets of oppositions: © Koninklijke Brill NV, Leiden, 2004 Also available online – www.brill.nl

JESHO 47,1

REVIEWS

123

the Ž rst is between the subject and object of study. The second is the division between the perceptual (sign) and the conceptual (symbol) representation, as equivalent to West/other. The analysis lays the theoretical bases to demonstrate that the divide between representation and the real does not apply to ancient Mesopotamia, which is the leitmotiv of the next four chapters. The second part of the book aims to demonstrate that the Assyro-Babylonian tradition does not aspire to mimesis but Òis conceived of as being part of the realÓ (p. 5). This section starts with a reexamination of certain interpretations concerning the cuneiform script to demonstrate that for the Assyrians and Babylonians the visual and the verbal were not clearly separated, but that they constituted one interdependent symbolic system. Chapter four characterizes cuneiform writing as a pluridimensional system narrowly connected with magic and divination. Bahrani deŽ nes cuneiform writing as an Òimage-text,Ó an interpretation based on the evolution from the Uruk IV pictograms to the later shape of the signs. Thus the author states that, although in later periods it is no longer possible to see the original pictogram, Òthe script always retained its pictographic origin within its logic as a systemÓ (p. 106). Due to this characteristic, cuneiform had unlimited possibilities for signiŽ cation because each sign could be used for Òpictographic or phonetic valueÓ (p. 114). Here the author seems to equate pictogram with logogram. This equation is difŽ cult to sustain because a pictogram is a symbol intended to be a recognizable picture of what it means, while a logogram is a symbol representing a complete word. The inscriptions mentioned in the book, however, used a combination of syllabic and logographic writing. It seems highly unlikely to me that a Ž rst millennium scribe could easily have recognized the pictogram behind a given logogram, because by then cuneiform writing, already a couple of millennia old, bore little resemblance to hypothetically original pictograms. The characteristics of Ž rst millennium lexical lists and commentaries show the need to record (for pedagogic and utilitarian reasons) the shape, readings, and meanings of those signs used to write in Sumerian and Akkadian.2 None of these lists includes, as far as I know, the evolution of signs as a modern scholar can trace it from looking at LabatÕs handbook.3 Bahrani further illustrates the multiple and hermeneutic readings of cuneiform signs by referring to BottŽ roÕs study of the Ž fty names of the god Marduk listed in the last tablets of Enâma elish. This example also supports the idea that writing was considered an act of creation. It should be noted, however, that the section listing the Ž fty names of Marduk in the last part of the sixth and in the seventh tablet of Enâma elish is an unusual text, and it is hardly representative of the logic of other texts such as royal inscriptions or legal documents. As Michalowski suggested, the exaltation of Marduk in Enâma Elish belongs to what Machinist called Kulturkampf, that is, a political and literary contest between Assyrians and Babylonians in the Late Bronze Age.4 The Ž fty names worked as a lock with a complex linguistic mechanism. The names were written in such a way so that the Assyrians could not appropriate Enâma elish. In principle it was impossible to unlock the names and change Marduk to the god Assur, a replacement that the Assyrians unsuccessfully tried. The potential hermeneutic reading of MardukÕs Ž fty names is not a suitable analogy to interpret the whole writing system because it is a late composition, and it does not account for a variety of texts beyond literature, magic, and divination. The Ž fth chapter focuses on the concept of ×almu, an Akkadian word usually translated as Òstatue, relief, or monument.Ó Bahrani challenges the traditional rendering of the term, arguing that visual representation cannot be separated from the verbal system of the script. In that sense, then, Assyro-Babylonian manner of representation is better deŽ ned as a Òpluridimensional chain of possible appearances,Ó similar to the writing system. Ñalmu, therefore, should be considered Òas a form of image that circulates within the realÓ (p. 127). The author proposes that rather than being an aesthetic concept, ×almu is an ontological category because through representation it substitutes for the real thing. This interpretation implies that the image of the king is not simply an image of the ruler, but that after the performance of rituals the statue becomes a substitute for the monarch himself. This conclusion is further

124

REVIEWS

expanded in the next chapter, aimed at studying the abduction and mutilation of royal images in the Near East. Chapter six deals with two sets of images that were assailed and robbed in antiquity. The Ž rst group comes from Nineveh and is now in the British Museum, and the second group comes from Elam and is in the Louvre. The analysis includes reliefs from the palace of Sennacherib, the Laws of Hammurabi, and the stele of Naram-Sin, among others. The author compares the abduction of royal statues with the capture of the statues of Marduk and his consort Ñarpanitum by the Hittites in 1594 B.C., and by the Elamites in the twelfth century. Since the loss of the image of the god implied the loss of divine favor, and since the image of the king represented the king himself, the author concludes that the abduction of images was an act of political control. Thus, Òhaving control of a personÕs image was one more way of having control of that personÓ (p. 183). This interpretation, according to Bahrani, challenges the view of traditional scholarship that regards mutilation and looting of statues as barbaric acts undertaken for the material value of the object. Instead, she proposes that deportation of royal monuments should be considered as an act of magical and psychological warfare. This statement also questions the traditional interpretation of royal monuments as mere royal propaganda. Although I agree with Bahrani that royal propaganda is not necessarily a manifestation of Oriental despotism, I would like to express my reserve with respect to her conclusions regarding image abduction. An important thing to consider is what happened to the presence embedded in the image once the king died. Did the statue become an empty shell? This question is relevant because some of the images that Elamite kings looted were of kings who had died many centuries before, as was the case with Naram-Sin and Hammurabi. Shutruk-NahhunteÕs scribes seem to have even mistaken the name of Hammurabi for that of Manishtushu, according to an Elamite inscription on HammurabiÕs stele. If indeed Shutruk-Nahhunte thought that by capturing images he was taking control of the king himself, then one would have to assume that the Elamite king did not know the monarch against whom he was Ž ghting. The abduction of divine images is a different matter, because gods are immortal. But the stories about these statues are also problematic, for all the examples that have reached us are accounts from those kings who returned the statue and had texts written to praise their actions. As an example I shall mention the so-called ÒMarduk prophecy,Ó where the god himself predicts his own return to Babylonia thanks to Nebuchadnezzar I.5 The seventh chapter examines the altar of Tukulti-Ninurta, a monument well known to Near Eastern art historians. Bahrani offers a novel interpretation based on her theoretical premises. The author questions generic identiŽ cations of the altar as a political portrait of the king for public display, and her rereading of the piece emphasizes the integral visual and verbal character of the monument. Thus, the altar Òis a representation about an episteme and ontology, about repetition and doubling, about representation itselfÓ (p. 201). In chapter eight, Bahrani integrates her questionings of colonial discourse and traditional academic practices with her conclusions regarding Assyro-Babylonian ontology. She stresses the importance of being aware of the cultural project of imperialism in order to write a post-Orientalist history of Mesopotamia. An evaluation of the in uence that colonial and Western discourses have had on the writing of Mesopotamian art history is undoubtedly needed. Post-colonial scholars deal with discursive practices that shape otherness, the struggles between imperial powers and the oppressed, and resistance. By considering the conclusions of post-colonial and post-structuralist studies, the author has successfully questioned traditional assumptions of Near Eastern art historians. However, it seems to me that focusing almost exclusively on AssyroBabylonian ontology when analyzing art monuments leaves aside questions of politics and power. For instance, the author seems to imply that the propagandistic effect of royal image was only tangential. That is the case when she afŽ rms, ÒThe system functioned through the ontological notion of the word-image-being entity. If this system served the ideology of

REVIEWS

125

kingship, so much better, but it was not simple propagandistic assertion of absolute powerÓ (p. 145). Even if we do not agree with the simplistic characterization of Oriental despotism, we should not forget that the surviving artistic works of the period originated in the ruling class: Assyrian and Babylonian kings were indeed the masters of people and land in the Ž rst millennium. Zainab BahraniÕs book is a thought-provoking work that undermines antiquarian approaches to ancient Near Eastern art and breaks the closed circle of the specialist by means of an interdisciplinary perspective. Andrea Seri University of Michigan [email protected]

1 See, for instance, Mogens T. Larsen, The Conquest of Assyria. Excavations in an Antique Land 1840-1860 (London: Routledge, 1996). 2 For the demise of cuneiform script see S. Huston, J. Baines, and J. Cooper, ÒLast Writing in Egypt, Mesopotamia, and Mesoamerica.Ó Comparative Studies in Society and History 45, no. 3 (2003): 430-479. 3 R. Labat and F. Malbran-Labat, Manuel dÕƒpigraphie Akkadienne (Signes, Syllabaire, IdŽ ogrammes), 6th ed. (Paris: Librairie Orientaliste Paul Geuthner, 1995). 4 P. Michalowski, ÒPresence at Creation,Ó in Lingering over Words. Studies in Ancient Near Eastern Literature in Honor of William Moran, eds. T. Abusch, et al. (Atlanta, Georgia: Scholars Press, 1990), 381-396. See also P. Machinist, ÒLiterature as Politics: the TukultiNinurta Epic and the Bible.Ó The Catholic Biblical Quarterly 35 (1976): 455-482. 5 For the ÒMarduk Prophecy,Ó see B. Foster, Before the Muses. An Anthology of Akkadian Literature, vol. 1 (Bethesda: CDL Press, 1996), 302-305.

A.C.V.M. Bongenaar (editor), Interdependency of Institutions and Private Entrepreneurs. Proceedings of the 2nd MOS Symposium, Leiden, December 11-12, 1998. (PIHANS, 87; MOS Studies, 2). Leiden: Nederlands Historisch-Archaeologisch Instituut te Istanbul, 2000. vii, 284 p. EUR 57.—. ISBN 90-6258-088-2 The book contains the proceedings of the second international symposium of the research project ÒThe Economy of Ancient Mesopotamia.Ó In this second symposium, the issue of the institutional production and management was addressed. As Renger points out (p. 153), Mesopotamian institutions managed their assets through a combination of two different systems: the self-sufŽ cient oikos system and the agency of entrepreneurs (whether private or not). These two systems may co-exist in one institution. In the symposium, the position of the entrepreneur was studied. Apparently, the original goal of the symposium was to address only agricultural entrepreneurship (Van Driel p. 5), but this scheme could not be maintained if Ur III, Old Assyrian, Nuzi and Ugarit material were to be included. The point of departure for the symposium is the situation in the Neo-Babylonian period, because both institutional and private archives from different cities have been recovered (p. 1). This is also the case for the Old Babylonian period, when the edicts and the Code of Hammurabi provide more general information about the entrepreneurs. However, the NeoBabylonian texts are much more informative concerning institutional-human relations (van Driel, p. 7). The purpose of the symposium was threefold (p. 1): © Koninklijke Brill NV, Leiden, 2004 Also available online – www.brill.nl

JESHO 47,1

126

REVIEWS

1. to verify the thesis of the relation between the private entrepreneur (an outsider leasing institutional property or an insider providing commodities such as labor force) and the institutions (palace or temple) in the Neo-Babylonian period and to study this relationship in detail, 2. to investigate the importance of this connection for earlier periods (evident for OB, but other periods as well), 3. and to answer the question whether there are indications in the written sources for a sector of the economy which had no connections with the institutions. Unfortunately, the book contains no conclusion in which these purposes are resumed. Thus, this book will be used mainly for the information contained by individual articles. It could have offered more, since the system of entrepreneurship has been studied for the individual archives and periods but not yet from a diachronic perspective. Not all of the articles actually address one of the three objectives formulated in the introduction. Thus, Charpin studies an aspect of the m”áarum edicts which is not related to the connection between the private entrepreneur and the institutions, namely the re ection of the periodicity of a m”áarum promulgation in the archives. Just as the private archives from Ugarit, the Nuzi archives do not allow a study of the entrepreneurial activities. Therefore, Jas and van Soldt examine the nature of the archives in order to determine who kept an archive in Nuzi and in Ugarit respectively. Finally, the aim of JongmanÕs article is to give some new perspectives on the economic history of Mesopotamia and he does not touch upon entrepreneurial activities. The contribution of the much-regretted van Driel, ÒInstitutional and Non-institutional Economy in Ancient Mesopotamia,Ó gives a diachronic perspective and touches upon the third purpose of the symposium, the question of a sector of the economy not connected to the institutions. Before addressing the central issue of his paper, the degree of institutionalization of Babylonian agriculture, van Driel emphasizes the bias of the documentary evidence. Since practically all Mesopotamian archives are connected to an institution, the size of the non-institutional segment of agriculture is very difŽ cult to assess. An important non-institutionalized part of society was the pastoral world. These tribal groups could be integrated in the Mesopotamian social fabric by incorporating them in the military structures. Only the Ur III period provides enough data to quantify the institutional agriculture, and allows Van Driel to conclude that enough agricultural land is left for non-institutional management. The size of the personnel of the Neo-Babylonian temple of Sippar points to the same proportions. The scarce documentary evidence concerning Babylonian villages, the actual loci of Mesopotamian agriculture, shows that they could be managed by institutions, by high ranking persons and by the military as maintenance land. Villages operating on subsistence level are of course not archivally documented. In the Ž rst Neo-Babylonian contribution, Joann s (ÒRelations entre intŽ r ts privŽ s et biens des sanctuaires ˆ lՎ poque nŽ o-babylonienneÓ) tries to situate the prebendary system and the Fermiers GŽ nŽ raux in a larger social context. After an overview of the possible transfer of goods between temples and private individuals, he examines the background of some wellknown prebendary holders and the Fermiers GŽ nŽ raux of the Eanna. Prebends generally remained in the family, but the opposite is documented as well. Most of the prebendary holders were members of the local nobility. The Ž rst Fermier GŽ nŽ ral was sent and supported by the king. The later ones belonged to the nobility of Uruk. The other articles concerning Neo-Babylonian material all illustrate the entrepreneurial activities documented in a(n) (group of) archive(s). Thus, Beaulieu (ÒA Finger in Every Pie: The Institutional Connections of a Family of Entrepreneurs in Neo-Babylonian LarsaÓ) discusses the activities described in a small archive from Larsa. The texts illustrate diverse activities in the agricultural and commercial spheres. Apparently, the family, which was not part of the hereditary urban elite, proŽ ted from the rebuilding of Larsa by Nebukadnesar II to expand its economic assets and to gain a foothold in the temple economy. Eventually, one of the family members marries into a family with patronymic.

REVIEWS

127

Bongenaar (ÒPrivate Archives in Neo-Babylonian Sippar and their Institutional ConnectionsÓ) gives an overview of the private archives from Sippar. However, these were recovered in the temple complex, and thus, the families are related to the temple administration. Temple ofŽ cers and prebend holders kept the archives. The documents concerning their private activities such as crediting were kept in the temple as well. Finally, Wunsch (ÒNeubabylonische GeschŠftsleute und ihre Beziehungen zu Palast- und TempelverwaltungenÓ) investigates the institutional relations of the Egibi family. She gives an overview of the references to institutions in the texts from the different generations. Contrary to the families discussed in BeaulieuÕs and BongenaarÕs articles, the EgibiÕs were not dependent on the institutions. They took care to maintain excellent relations with the royal ofŽ cials both on a private (through leases and loans) and on an ofŽ cial level. These contacts were very important for the business of the family. The military and the palace were the best market for products acquired through harr¨nu businesses. They also seem to have taken over some tax-farming activities from the Esagila. The chronological overview of the connection between institution and entrepreneur starts with the Ur III period. Neumann (ÒStaatliche Verwaltung und Privates Handwerk in der Ur III Zeit: Die AuftragstŠtigkeit der Schmiede von GirsuÓ) argues that, in spite of the impression one gets from the Ur III documentary evidence, craftsmen operated on an independent basis. Since the Ur III texts originated from the institutional bureaucracy, we are informed only rarely about their non-institutional orders. However, the state was by far the most important commissioner of the craftsmen. Dercksen (ÒInstitutional and Private in the Old Assyrian PeriodÓ) tries to locate the institutional connections of the Old Assyrian entrepreneurs. Since they were discovered in commercial outposts, the texts from this period describe the commercial activities from the point of view of the entrepreneurs, and not that of the institutions. The king of Assur appeared as an entrepreneur between the others. The city of Assur provided the conditions for the entrepreneurial activities (it regulated, monopolized and established the political network), but it did not seem to play an active role in the trade. The Assyrian traders functioned as intermediaries between the different Anatolian palaces. Since the symposium departs from the Neo-Babylonian period, RengerÕs contribution (ÒDas PalastgeschŠft in der altbabylonische ZeitÓ) about Old Babylonian PalastgeschŠfte is included in the chronological overview. However, it deserves a more prominent place since it examines the nature of the relations between the entrepreneur and the Old Babylonian palace. As Bongenaar observes (p. 3), the Neo-Babylonian entrepreneurship was organized along the same principles, only the temples seem to have taken over a large part of the role of the Old Babylonian palace. Renger reviews all of the Old Babylonian evidence concerning the theme and places it in its context. He arranges the evidence (institutional archives, letters, administrative documents, paragraphs of the edicts and of the Codex Hammurabi) thematically (relating to agrarian production and herding, to the management of natural resources and to the service sector). Thus, the article provides us with an excellent framework in which to situate the PalastgeschŠfte. The private entrepreneurs studied in CharpinÕs contribution (ÒLes pr teurs et le palais: Les Ž dits de m”áarum des rois de Babylone et leurs traces dans les archives privŽ esÓ) are creditors issuing loans to private individuals. Since the loans could be subjected to a m”áarum edict, these creditors depended on the palace institution. Charpin illustrates how the promulgation of a m”áarum was preceded by an increase in the number of loan contracts found in the archives of the creditor. The creditor apparently did not throw away the canceled contracts, but kept them in a separate Ž le. The archives from Nuzi are not informative about entrepreneurial businesses. On the basis of scarce evidence, it can be concluded that the merchants were employed both by the palace and by private entrepreneurs. Jas (ÒOld and New Archives from NuziÓ) reconstructs some unstudied archives and situates them in a city quarter where military ofŽ cers (several of them from Hanigalbat) lived.

128

REVIEWS

Only a few private houses in Ugarit contained documentary evidence. In his contribution (ÒPrivate Archives at UgaritÓ), van Soldt tries to discover the raison dՐ tre of these archives. All of the administrative texts and letters can be related to the palace administration. Some juridical texts can be placed in a private context. The lexical and cultic texts allow us to determine whether we are dealing with a (teaching) scribe or a priest. All of the archive holders in Ugarit occupied a high position in the palace bureaucracy. Since some of the archive holders were deŽ nitely priests, the cultic organization(s) in Ugarit seem(s) to have been dependent on the palace as well. As far as the textual evidence allows any conclusions, there was not much room for entrepreneurial activities in New Kingdom Egypt. High ofŽ cials nominally supervised temples and managed royal lands, but agents, deputies or scribes represented them. Haring (ÒOutsiders in Charge of Institutional Property in New Kingdom EgyptÓ) thus situates the entrepreuneurship on an ofŽ cial level. The contribution of Jongman (ÒHunger and Power: Theories, Models and Methods in Roman Economic HistoryÓ) is not directly connected to entrepreneurship. It is included in the volume because the recent developments in the study of Roman economy provide new perspectives on the economic history of Mesopotamia. A correct use of economic theory, a thorough study of factors like demography, food supply, agricultural and transportation technology and the use of parametric data such as model life tables, may rejuvenate the debate concerning ancient economy. Resuming the purpose of the symposium, we can state that entrepreneurship is a widespread phenomenon in Neo-Babylonian archives. Families owning an archive were tied to institutions through ofŽ ces, prebends, because they managed institutional land or bought taxfarming rights. The management of the institutions depended for a large part on these entrepreneurs. Already in the documentary material from Ur III institutions, entrepreneurial activities can be detected. The system is well documented in the Old Babylonian period—though not as prevalent as during the Neo-Babylonian period. The documentary evidence from other periods (Old Assyrian and Middle Babylonian) does not give speciŽ c information on the nature of the relation between entrepreneurs and institutions, but archive holders from those periods often were involved in business relations with institutions. According to Van Driel, the pastoral world was the only sector of the economy which had no connections with the institutions mentioned in the written sources. Some villages may have operated independently on subsistence level but there are no traces in the documents of those segments of the economy. The book contains some important contributions for students of Babylonian society, in particular for the Neo-Babylonianists. A concluding summary would have offered the opportunity to compare the situation of the entrepreneur in the different periods, especially between the Old and the Neo-Babylonian period. Anne Goddeeris [email protected]

Horst Klengel with the assistance of Fiorella Imparati, Volkert Hass and Theo P.J. van den HOUT , Geschichte des hethitischen Reiches. Handbuch der Orientalistik 1/34, Leiden: Brill, 1999. xxvi + 430 + 63 illustrations. The book under review here is written by one of the leading historians of the Hittite kingdom, and its neighbors, Horst Klengel. The introduction has sections discussing the geography where the history played out, the story of the rediscovery of the Hittites, and notes on the further development of Hittitology. The next Ž ve sections proceed chronologically. First there is a short discussion of Anatolia in the early Bronze Age. This is followed by a chapter © Koninklijke Brill NV, Leiden, 2004 Also available online – www.brill.nl

JESHO 47,1

REVIEWS

129

on Anatolia in the age of the Assyrian merchant colonies, during which period Hittites are Ž rst mentioned in written sources. Sections II, III and IV proceed chronologically with each king given a separate chapter. Unlike its English language rival, Trevor BryceÕs The Kingdom of the Hittites, KlengelÕs work is not straightforward narrative history. Rather the section on each king begins with an introductory paragraph. Then every text datable to that kingÕs reign is listed and given a number [A1], [A2], etc. This is followed by a list of every text from later reigns that refer back to this kingÕs reign. These are given numbers [B1], [B2], etc. The entry for each text gives its text number, the name the text is known by today (ÒAnnals of Muráili IIÓ) or its type (Òfragment of a prayerÓ), its number in LarocheÕs catalogue, a sentence or two describing what of its content is relevant to the reign in question and where the text has been edited or translated. After this there is a narrative of the reign of the king. In the case of Muráili II, about whom the most is known, and for whom events can be easily placed chronologically, the chapter is divided into theaters of operations: north and northeast, west, and southeast, which are then discussed chronologically. The Ž nal section of political history deals with the fall of the Hittite empire and the general crisis at the end of the Bronze Age. It is followed by a section discussing the two portions of the empire that survived or may have survived the general catastrophe, Kargamiá and TarÀuntaááa. There are in addition copious footnotes, many of which contain important details. Some are sufŽ ciently important that one wonders why they are not up in the text, e.g. p. 22 n. 18 on the geographical suitability of Kaniá as a node point for the Old Assyrian trade, n. 19 on the archaeological chronology of KŸltepe and p. 23 n. 28 that PitÀana, previously only known from AnittaÕs inscription is now attested in contemporary Old Assyrian texts. This political history written by Klengel is followed by a section of seven chapters written by Imparati on the Organization of the Hittite State. Chapters discuss the texts, the royal family, the exercise of power, the administration of the kingdom, die kollegialen Gremien (Òthe collegial boardsÓ?), the organization of work, the government of the empire and international relations. Toward the end of the book are charts of synchronisms, a detailed bibliography divided up into twelve parts by subject covered, indices of personal names, place names and texts (according to Laroche catalogue numbers). Rounding out the book are 63 illustrations, including plans of Hittite cities and buildings so far excavated, photographs of Hittite tablets, drawings of Hittite seals, drawings and photographs of the Hittite kings who portrayed themselves on rock reliefs, and Ž nally excerpts from Egyptian reliefs showing Hittites. As we have come to expect from Klengel, the book relies on facts and sound historical judgments and eschews trendy theory. Low probability suggestions and out and out errors are remarkably few. p. 17, 3 lines from bottom, read STT 51, 78 + 166 (Sultantepe). p. 18. The Old Assyrian tablet concerning the deeds of Sargon of Akkad has since been published in photo, copy, transliteration, Turkish translation and English summary by G€ nbatt in Archivum Anatolicum 3 (1997) [GsBilgi ] 131-155. pp. 33, 35, 43. The long established view that king ¿attuáili I moved the Hittite capital to newly refounded ¿attuáa will be challenged by R. BealÕs contribution to the Festschrift for Harry A. Hoffner, Jr. KlengelÕs understanding (p. 37) that Labarna I was ¿attuáilliÕs auntÕs husband and heir of ¿attuáiliÕs grandfather by adoption is an excellent one. p. 40 § [A 9]. This important text, which should be known as ÒAnecdotesÓ since it is not a ÒChronicle,Ó is edited with German translation by O. Soysal, in his dissertation Muráili I Eine historische Studie (WŸrzburg, 1989) and more recently with Italian translation by Paola Dardano, LÕaneddoto e il racconto in etaÕ antico-hittita” (Rome, 1997). p. 53. I wonder why the Siege of Uráu text is declared a legend Òund wohl nicht historische Ereignisse re ektiert.Ó The text is certainly not ÒhistoryÓ but appears rather to be ÒsatireÓ and although differing in style it appears to Ž t in with the moralizing on the failings of royal ofŽ cials seen in writings from the time of ¿attuáili I and Muráili I such as the Òanecdotes.Ó As the latter seem to have historical bases, so probably the Uráu siege.

130

REVIEWS

p. 60 n. 123, read AoF 18 (1991). p. 63. I do not see why ÒVS NF XII 2Ó (a.k.a. VS 28.2) is mentioned under Muráili I, since in this text ÒMuráiliÓ occurs in a position after Ò°uppiluliumaÓ is seems clear that Muráili II is meant. pp. 64-65. Although ¿antili I apparently tried to paint MuráiliÕs I raid on Babylon as sacrilegious to justify his own usurpation, it seems unlikely that it is due to this judgment that he is omitted from later genealogies. The fact that Telipinu paints Muráili as one of the good kings the sack of Babylon as a Hittite success, and the fact that there are later kings named Muráili points to the fact that Muráili was not seen by later generations as thoroughly disreputable. It seems most likely that he was not mentioned in later genealogies since he was in fact not a direct ancestor of any kings, as he was murdered before he managed to reproduce (Telipinu doesnÕt mention the murder of any of MuráiliÕs children, unlike in the cases of ¿antili IÕs and AmmunaÕs children). p. 72, last paragraph, for Taurus read Amanus, since according to Telipinu Adana (on the far side of the Taurus, but near side of the Amanus) was lost by Zidanta IÕs successor Ammuna. p. 76 [B3]. ¿uzziya in company with ¿attuáili, Labarna and Pimpirit is more likely to be ¿attuáiliÕs son ¿uzziya of ¿akpiá or even ¿uzziya 0, distant ancestor of ¿attuáili I and dynastic founder, rather than king ¿uzziya I. Similarly the ¿uzziya of [B4] in the company of PapaÀdilmaÀ probably dates to the time of ¿attuáili I. p. 76. For a suggestion that Zuru the GAL ME°EDI was AmmunaÕs brother see Beal, THeth 20:329 w. n. 1257. p. 80. I do not think that the plot that resulted in the murder of deposed king ¿uzziya I and his brothers is properly understood. Those who instigated the plot are called Ògreat onesÓ (meggaeá) by Telipinu, but saying they were Òdie hšchstenÓ is overdoing it. They are UGULAs ÒsupervisorsÓ not GALs ÒchiefsÓ of departments. Even if we understand the military ofŽ cer as a Òoverseer of 1000Ó this is still colonel level, subordinate to the GAL GE°TIN and GAL ME°EDI. What appears to have happened is that on AmmunaÕs death his brother Zuru and ZuruÕs sons Tanuwa, TaruÀáu and TaÀurwaili removed AmmunaÕs legitimate sons and installed the illegitimate ¿uzziya as their puppet. Telipinu, husband of a legitimate daughter, overthrew ¿uzziya and took over and banished ¿uzziya, his brothers and his supporters. Later while Telipinu was on a distant campaign with his way home conveniently blocked by a revolt in Lawazantiya, a group of magnates back home plotted a palace coup to depose Telipinu, should he manage to get home, and to replace him, not with the weakling ¿uzziya, but with one of ZuruÕs sons (who must Ž rst eliminate ¿uzziya and his brothers). It should be noted that it is Tanuwa, TaruÀáu and TaÀurwaili who are singled out for particular opprobrium, so they were probably to be the beneŽ ciaries of the plot. And, of course, eventually ZuruÕs son TaÀurwaili did manage to seize the throne, temporarily interrupting the reigns of TelipinuÕs descendants. (See p. 88.) p. 95. For a suggestion that Zidanta II was the son of ¿aááuili, the GAL ME°EDI and brother of his predecessor ¿antili II see Beal, THeth 20:330. p. 102f. For a suggestion that Kantuzili was son of ¿uzziya II and father of TudÀaliya II and perhaps king after the overthrow of the usurper Muwattalli I see Beal, FsHoffner (forthcoming). pp. 103ff. Klengel, correctly in my opinion, posits only two kings named TudÀaliya between Telipinu and °uppiluliuma I. These have since virtually the beginning of Hittitology been known as TudÀaliya II and III. When TudÀaliya II was split into two kings called TudÀaliya I and II and then subsequently reintegrated with himself, unfortunately it became the fashion, followed by Klengel, to call the king TudÀaliya I. This means that his grandson who was previously always known as TudÀaliya III is now called TudÀaliya II (III) on p. 127. The next TudÀaliya to certainly become king remains with his traditional number TudÀaliya IV. TudÀaliya the younger, who probably never was king, may or may not be

REVIEWS

131

called TudÀaliya III (p. 148 n. 27). Thus we have massive confusion. Originally the number I was given to a pre-¿attuáili I prince named TudÀaliya. While it is true that this ancestor may or may not have been king (see my discussion in FsHoffner [forthcoming]), it is better to reserve the number I for this obscure Ž gure as 3/4 of a century of scholarship has done and return to the traditional number of all the later TudÀaliyas, getting rid of parenthetical numbers and restoring missing numbers in what is a relatively clear sequence. I will continue to use the traditional numbering in this review and elsewhere. p. 109f. Klengel is surely correct to point out that TudÀaliya II did not found a new dynasty. p. 113 l. 19 Ò†berfŸhrungÓ and p. 174 [A21] Ò†berstellungÓ and Ò†bersiedlung.Ó I once based a small part of an argument about political control on the evidence of the ÒtransferÓ of the Goddess of the Night from Kizzuwatna to °amuÀa, an argument IÕm happy to see Klengel follows. However, more recently a closer look at the verb áarra- which does not otherwise mean ÒtransferÓ has convinced me that the goddessÕs divinity was ÒdividedÓ (the usual meaning of the verb) so that she would be resident in both Kizzuwatna and °amuÀa. (See R. Beal, in Magic and Ritual in the Ancient World, eds. Paul Mirecki and Marvin Meyer [Brill, 2002] 197-208.) The political implications are unchanged. p. 116 [A1]. Kuwattalla is not a ÒHieroduleÓ (Òa temple slave, especially the temple courtesans at Corinth and elsewhereÓ Liddell and Scott 821, OED s.v.). I thought that this traditional, but baseless, translation of SU¿UR.LAL had Ž nally been discarded. There are no sacred prostitutes in ¿atti (or Babylonia for that matter), and slaves of gods are not called SU°UR.LALs. The original Sumerian meaning of the term appears to have been hairdresser. As GŸterbock and others have pointed out (JAOS 103 [1983] 159 [ÒladyÕs servant, attendant womanÓ], Beckman, BiOr 40 [1983] 113 [ÒmaidÓ], Neu and RŸster, HZL [1989] no. 349 [Ò(Kammer-)Zofe, DienerinÓ]) perhaps the best translation for the term at ¿attuáa is ÒladyÕs maid.Ó pp. 125-126. Concerning ¿attuáili II, the offering texts VS 28.2 i 10-13, KBo 39.86 ii 1113, KBo 39.88 ii 7-9, KBo 39.89 iv 6-8, and KBo 39.91 ii 1-3 should have been brought up. In VS 28.2 offerings are made to the statues of ¿attuáili, then TudÀaliya, then °uppiluliuma, then Muráili. While these could be offerings to HattiÕs greatest kings ¿attuáili I, TudÀaliya II, °uppiluliuma I and Muráili II, they could just as well be offerings to Muwattalli IIÕs immediate predecessors, ¿attuáili II, TudÀaliya III, °uppiluliuma I and Muráili II (with the ephemeral Arnuwanda II skipped). In favor of the latter, note van den HoutÕs comment p. 252 n. 486. p. 148. It has always bothered me that when Muráili mentions the incident of the murder of the heir TudÀaliya the younger to the beneŽ t of MuráiliÕs father °uppiluliuma in the context of the gods being angry at him for his fatherÕs crimes that he does not say ÒMy father killed his own brother.Ó p. 158. Press reports indicate that newly found letters from the chancellery at Qatna will further illuminate °uppiluliumaÕs conquests of Mittanni and Syria. p. 163 n. 103. I do not see how Liverani can claim that the name mZannanza, borne by °uppilulumaÕs son sent to become king of Egypt, was an Egyptian epithet for ÒprinceÓ and not a Hittite name. While a Hittite might mistake an Egyptian title of an Egyptian for that personÕs personal name, it is hard to see how Muráili (the author of the text) would not have known his own brotherÕs name from an Egyptian title. p. 180. It is good to see emphasized Muráili IIÕs constant need to keep the lid on the closein Kaáka, before he could lead campaigns further aŽ eld, against places that we (and probably the Hittites) consider more interesting. p. 193 l. 15, for ÒOnkelÓ read ÒVetter.Ó ¿udupianza is the son of Ziti the GAL ME°EDI, °uppiluliuma IÕs brother, and Muráili IIÕs uncle and thus ¿udupianza is Muráili IIÕs cousin. p. 208 l. 26, for GAL GE°TIN read GAL ME°EDI. This is correct on p. 255. p. 211 l. 24, for SU-an read °U-an.

132

REVIEWS

p. 240 n. 451. It is distressing to see from this footnote total agreement on the location of Ura at the mouth of the Kalycadnus (Gšk Irmak), agreement created by ignoring my article arguing against this location and in favor of Kelenderis (Gilindere), AnSt 42 (1992) 65-73. p. 249 [A36], to the text KUB 21.20 (+) KBo 16.36 add + HHT 82 + KUB 48.87, ed. Alp, Belleten 41/164 (1977) 644-646 repeated in HBM 32-35. p. 274 n. 561, add here R. Beal, THeth 20 (1992) 387 n. 1466. p. 286. For a newer suggestion concerning what had been know as ÒThronnamenÓ and ÒPrinzennameÓ see Beal, FsImparati (Eothen 11) 55-71. p. 294 n. 633 l. 9, for StBoT 18 read StBoT 38. p. 300. Merneptah??? why not Siptah/Tawasret? p. 314. ÒDer Weg zu Mittelmeer via TarÀuntaááa war(en) offenbar fŸr das †berleben ¿attis immer wichtiger geworden,Ó implies that unrest in TarÀuntaááa would cut off grain shipments from Syria to relieve famine in ¿atti. This ignores the fact that the best road from the Mediterranean coast to the Anatolia plateau is the Cilician Gates which runs through Kizzuwatna not TarÀuntaááa. p. 314. Modern historical periodization here forces disparate events into a straightjacket. Although Greece, Anatolia, Syria and Palestine all suffer destruction of civilization and Egypt goes into steep decline after about 1180 B.C., the scholarly export of a simultaneous ÒEnd of the Bronze AgeÓ to Mesopotamia does not work. The Kassite Dynasty lasts another 25 years and its end comes not from the west but from the Elamites to the east. It is replaced by the 2nd Isin dynasty, including the powerful Nebuchadrezzar I (1125-1104), a dynasty unaccountably ignored in GHR. It is not until 1026 that this dynasty dissolves into the obscurity of the 2nd Sealands Dynasty. In Assyria, too, this periodization gets in the way of understanding. It is true that the assassination of Tukulti-Ninurta I (some twenty to thirty years before the fall of ¿atti!) ends a period of Assyrian expansionism, and AssyriaÕs brief rule of Babylonia is ended, but there was no ÒMachtrŸckgangÓ due to massive Aramean invasions for another hundred years, until the Middle Assyrian kingdomÕs arms had reached their farthest extent yet and then suddenly collapsed due to over expansion and Aramean incursions. It is quite clear from excavations at Dur-Katlimmu that Assyrian rule over ¿anigalbat (the former Mittanni) continued for another century after Tukulti-NinurtaÕs death. p. 323. Imparati follows the excellent suggestion of Beckman that the title ÒMy Sun” for the king is more likely to have come from North Syria than Egypt. However, ImparatiÕs further suggestion that the link was with the Sungoddess of Arinna, wife of the Stormgod, chief deities of the pantheon, seems unlikely. It was the Hittite queen that was linked to the Sungoddess according to the iconography: at Fraktin, Queen PuduÀepa is shown dressed the same as the Sungoddess. However, as is well known, at Yazõlõkaya the Sungod, dresses differently from all of the other gods and wears the same style clothing as the Hittite king, portrayed there and elsewhere, often does. This Ž ts perfectly with Mesopotamian concepts where the Sungod is also god of justice and so also of kingship (but is not the chief god of the country). p. 335 w. n. 55. It is disheartening to see an old translation (LòI°/KU°7 = ÒWagenlenker/ KnappeÓ) that one has spent considerable effort to correct (actually Lò°ô°/KU°7 = Òchariot Ž ghterÓ) not only continue to be used unchanged but to see oneself cited favoring this old translation. p. 337. That following Starke ÒWie zuvor die Großen, hŠtten sie (the Lò.ME°SAG) nun mehr eine politische Kraft dargestellt, die gemeinsam mit dem Kšnig regierteÓ seems off base. The Hittite king appears to have been the active and unchallenged ruler of the kingdom, obviously needing the help of grandees to rule, but Hatti certainly does not appear to be a limited monarchy. The opinion of many that the Lò.ME°SAG were eunuchs should have been taken into consideration. p. 342 n. 83. How do we know that °aÀurunuwa bore the three titles he is given in the edict dividing his estate at the same time. In my study of military titles, I suggested they were borne consecutively (THeth 20.383-385, 387 n. 1466).

REVIEWS

133

p. 344. It is hard to believe that with communications what they were that ÒDennoch kontrolierte und bestimmte die Zentralverwaltung auch die administrative Strukturen von Dorf gemeinden und ließ ihnen nur wenig Autonomie.Ó It is certainly clear that the defenses and water supplies of small walled towns were strictly and minutely watched over by the governor during his periodic visits, and that the government clearly took an interest in getting vacant land back into taxpaying production. However, there are no instructions concerning villages, nor are there building codes for houses. Statements such as Òwherever they execute let them execute him, wherever they exile, let them exile himÓ argue for differences of law and lack of the centralizing standardization that only really becomes government theory with the European ÒenlightenedÓ despots and subsequent revolutionaries of the 18th and following centuries A.D. It seems clear from the instructions for the governor that even in provincial towns legal cases would be expected to be settled by the local elders (who Imparati discusses earlier) and notables and only a difŽ cult case would be passed on to governor or to the king. He is not instructed to examine settled cases to see that the settlements correspond to some central government policy. p. 344. ÒEine ganze Reihe von Dokumenten re ektieren eine kž nigliche Politik, die auf eine gleichmŠßige Verteilung des Grundbesitzes abzielte.Ó I donÕt know what those documents are. Sure, land was granted, land was conŽ scated from rebels or fell to the crown in default of direct heirs, and regranted, and the king might conŽ rm a complicated division of an estate among heirs, but I do not see anything that aims at an equal division of land. pp. 349f. Imparati follows Diakonoff in seeing beside temple and palace agriculture, Òcommunity agricultureÓ (Gemeinde-Wirtschaft). I would agree it was the case that much of ¿atti was farmed outside the temple and palace sectors at a more or less subsistence level. It is also true that there was a certain amount of village corporate behavior (village elders [§ 71], the men of the village shall [temporarily] work the empty land [§ 40], collective village responsibility if no individual could be penalized for a murder [§ IV]). However, the texts do not refer to a village short a person, but a plot of land that is empty. They envision land sales (§§ 39, 169), an individual taking the initiative in farming an empty Ž eld (§ 40, 41), division of land as part of marriage settlements (§ 46), individual ownership of animals (§ 71, XXXV), wage labor at harvest time (§ 158). It thus appears that most land was held as private property and that the villages had leadership, who could act for the village vis-ˆ-vis the central government if necessary. There is no evidence here for primitive Communism. It is clearly Orientalist slander (in the sense of the word invented by Edward Said) to state that ÒDie Nahrungsproduzenten waren gehalten, einen großen Teil ihrer Erzeugnisse der zentralen AutoritŠt zu Ÿberliefern.Ó The statement would probably hold if the word ÒErzeugnisseÓ (ÒyieldsÓ) were changed to Ò†berschusseÓ (ÒsurplusesÓ) or the word ÒgroßenÓ were removed. The NeoAssyrian government took 10% of the harvest (N. Postgate, Taxation & Conscription 176) (which is considerably less than the US government takes from my income). The Islamic land tax (kharaj) was according to the jurists also 10%, but the Imam could raise or lower this percentage based on the quality of the individual cultivatorÕs land and the individualÕs ability to pay (A. Lambton, Encyclopaedia of Islam2 4 [1978] 1037). There is no evidence for the Hittite rate, but it was probably not that different. p. 351 w. n. 115. Imparati is to be commended for recognizing (unfortunately only in a footnote) that in addition to the palace and temple sectors of the artisanal economy there was a private sector that is largely absent from our palace and temple centered documents, few of which in any case are administrative. p. 353. ÒWie in den anderen Kšnigreichen des alten Nahen Ostens war auch in ¿atti der Arzt dem Palastbereich verbunden und widmete seine BemŸhungen vor allem der EliteÓ makes no sense and cannot be supported by any facts. It is quite clear that the aáipu, who was both the doctor and the exorcist of Mesopotamia (see J. Scurlock, in Mesopotamian Magic, eds. K. van der Toorn and T. Abusch [Styx, 1999] 69-79) was attached to temples and was thus in a position to provide services for free (except for drugs and equipment) even to the lower echelons of society (note comments Òif he is poor, substitute the followingÓ).

134

REVIEWS

Nor is there any evidence that all pharmacists (asž ) worked for palaces. Kings may have sent their favorite practitioners to other kings, but since there are no private records for anything in ¿atti it is pure cynical speculation that there was no medicine outside the palace. p. 357. Concerning trade, the fact noted by Imparati that we have almost no private records of trade or anything else for that matter should not lead to her conclusion that trade was dependent on the palace and merchants were royal functionaries, and that private trade was but a tri ing. There are, after all virtually no documents concerning trade from the palace, which we at least know was producing documents on other subjects. That the king of Ugarit calls the merchants of Ura Òmerchants of Your MajestyÓ may mean they worked for the Hittite king (see also pp. 261f. w. n. 508), but it seems more likely that the term simply means Òmerchants (who are subjects) of Your Majesty.Ó The whole problem between the merchants of Ura and the king of Ugarit concerns the merchants buying up or foreclosing on property in Ugarit. This sounds like something merchants buying and selling to make a proŽ t would do with their proŽ ts and not something government buying agents would do. In any case, Ugarit already belonged to the Hittite king as overlord, so why would his government buying agents be buying up its real estate? If the Hittite king had wanted to build something for himself in Ugarit, eminent domain would have been so much easier. Beside the king of Ugarit does not complain to the Hittite king that the Hittite king is buying up all of the land, but that the merchants are. p. 365. What, I wonder, leads Imparati to say that the tributary king could possess no fortiŽ ed cities? One of his main jobs was to repel his and thus the HittitesÕ enemies. How ever was he to do that without fortiŽ ed cities on his borders? This is an important book that generally provides a thorough and well founded survey of Hittite history and society. Richard H. Beal Hittite Dictionary Project The Oriental Institute of the University of Chicago

Naguib Kanawati, Conspiracies in the Egyptian Palace. Unis to Pepy I. London: Routledge, 2003. IX, 208 pp. £55.00 (cloth). Was king Teti really assassinated? And did his conŽ dants and dignitaries take part in this plot? Professor Naguib Kanawati chose for his latest book an extremely interesting and thought-provoking subject from the history of the late Old Kingdom and tried to Ž nd the answer to a question that is crucial for our understanding of the culture and history of the ancient Egyptians—were they able to murder their king, the only living god on earth? Professor Kanawati is mainly connected with the excavations, clearance works, and publications of the tombs located in the Teti Pyramid Cemetery at Saqqara. The archaeological material originating from this restricted necropolis forms the basis for the book under review. The author conceived the whole book as a sort of Òhistorical investigation.Ó The core of the book concerns the suspicious circumstances surrounding the death of the founder of Dynasty 6, King Teti, who according to Manetho was murdered by his own bodyguards. With this in mind, Kanawati named the three main chapters of the book ÒAssassination claim,Ó ÒThe Suspects: Case Studies,Ó and ÒThe Investigation.Ó An introductory chapter, conclusions, notes and a detailed bibliography complement the main part of the study. The author pays particular attention to the archaeological and historic evidence of the three Òpalace conspiraciesÓ that occurred during the reigns of Teti (his assassination) and his son Pepy I (the queen trial and the plot of vizier Rawer). His examination is based on material from the period between the reigns of Unas and Pepy I. He analyzes the reasons behind TetiÕs enthronement, the course of his reign, and the occasion of his death. In his overview © Koninklijke Brill NV, Leiden, 2004 Also available online – www.brill.nl

JESHO 47,1

REVIEWS

135

of the reign of Pepy I, Kanawati focuses mainly on the punishments handed out to the conspirators. In the introduction, the author claims that it was not his intention to write a history of Dynasty 6, but rather to focus on some speciŽ c events that occurred within this unstable period. He then brie y describes the history of the Old Kingdom in general, with special attention to the royal succession. However, his emphasis on the secrecy of the royal palace and the description of the royal court as a Òfertile land for intriguesÓ is somewhat exaggerated—perhaps done to suit the taste of the general public. Although the genealogical connections among the Old Kingdom kings are sometimes not easily determined, according to the preserved written evidence, the mother of the king Snofru was queen Nimaathap and not Meresankh, the queen Khentkaus I from Giza was rather the daughter of Menkaure and not of Djedefhor, and king Neferirkare was the husband (and not the son) of queen Khentkaus II from Abusir, who bore him two sons—the succeeding kings Neferefre and Niuserre. The Ž rst chapter deals in detail with two subjects—the authenticity of ManethoÕs statement about the assassination of Teti by his bodyguards, and the related problem of deŽ ning those ÒbodyguardsÓ in the preserved written material from the Old Kingdom. The author argues that the Egyptian equivalent of the term ÒguardÓ is the title ½ntjw-á, predominantly translated as Ò(palace) attendantÓ and he describes this term in detail. The holders of this title were qualiŽ ed by the reference to a palace (pr-®3) or to a mortuary temple of a king. There is no doubt that the bearers of this title performed personal services for the king in the palace for they hold other titles referring to feeding, bathing and clothing the king. This personal aspect of the title contrasts with the tasks they performed in the royal mortuary cult, such as transporting food, and dressing and feeding the cult images of the deceased king. At the dawn of Dynasty 6, the number of ÒguardsÓ rapidly increased and the nature of this title underwent distinct changes. During the previous dynasty, the tomb owners formed a rather independent social class of middle rank. In Dynasty 6 this title occurs among titles held by ofŽ cials with two different positions within the social structure of the Egyptian society. For one, it was incorporated into the extensive title strings of the most powerful dignitaries at the court, perhaps marking the initial stage of their careers. On the other hand, among the titleholders were also still middle class men, sometimes with no other responsibilities. It remains hard to say to which extent these ofŽ cials guarded the king and his security. The second chapter represents the most important part of the book. Here the author deals with the available archaeological evidence that originates mainly from two parts of the Saqqara necropolis—the cemeteries lying to the north of the Unas causeway and his mortuary temple and those to the north of TetiÕs pyramid. Altogether, evidence of 47 tombs, whose owners bear the title ÒguardÓ and which date to the period between the reigns of Unas and Pepy I, is taken into account. The description of each tomb is structured following a similar pattern. First, Kanawati gives a brief overview of the career of the tomb owner and his titles, then he discuses the architecture and position of the tomb within the cemetery, and Ž nally he suggests a date for the construction of the tomb. The dating is of special importance because it sometimes differs from the dates indicated in the archaeological publications of the separate tombs. This is not without importance when considering the time/spatial development of the cemetery as a whole. Each entry ends with an overview of later treatments done to the tomb architecture, and the epigraphic and decorative program. Although the author gathered a great amount of reliable information, there seem to be some omissions. Not all recorded tomb owners bear the title Òguard.Ó For at least eighteen of them the title is not recorded on the preserved epigraphic material—Akhethetep/Hemi (No. 1), Ihy (No. 3), Geref (No. 13), HeŽ (No. 14), Inumin (No. 17), Iries (No. 20), Kaaper (No. 22), Kagemni (No. 23), Memi (No. 27), Mereruka (No. 30), Nedjetempet (No. 32), Nikauisesi (No. 35), Rawer (No. 36), Shepsipuptah (No. 38), Tetiankh and Hesy (No. 40), Tjetji (No. 42), name lost (No. 45), and Sabu (No. 47). There is no doubt that further archaeological excavations will reveal more material about this unstable and lesser known period of Egyptian history. Tombs dated to this period were

136

REVIEWS

also discovered in other parts of the Memphite necropolis, e.g. by the team of the Czech Institute of Egyptology in the area of Abusir South—the tombs of the vizier Qar and his family members. One of the advantages of this book lies in the possibility to read chapters 2 and 3 independently from one another. The last chapter was written as a self-contained part of the book for those readers who are only interested in the historic analysis, free from the archaeological descriptions contained in chapter 2. The investigation develops chronologically and the chapter is further subdivided into smaller entities dealing with different aspects of the period. A study of the development of the Teti Pyramid Cemetery, with special attention to the decision of Teti to allocate this speciŽ c place at Saqqara for the cemetery, opens the chapter. Based on an analysis of the titles of the dignitaries and the archaeological evidence, the author reconstructs the events that may have led to the enthronement and death of Teti, the turmoil concerning the ephemeral reign of Userkare, the seizing of power by Pepy I, and the punishment of disloyal ofŽ cials by the king. Kanawati also mentions the difŽ cult years of the reign of Pepy I, in particular the trial of the anonymous queen and the conspiracy led by the vizier Rawer around the twenty-Ž rst occasion of account of Pepy I. The results of the analyses of the archaeological evidence are gathered in two sets of charts—one for the reign of Teti and the other for Pepy IÕs—according to the extent of the damage executed on the tombs. In the concluding chapter, the author has gathered a group of kings with several aspects in common (the location of their mortuary complexes, exclusion of Re from the royal name and perhaps a similar attitude towards the priesthood of the god Re), namely Unas, Djoser, Userkaf, Teti and perhaps Menkauhor (if his pyramid is also located in this area). Sometimes the author tends to read more in the evidence than there actually is. For example, the usage of the name of Djoser is attested for the Ž rst time only in texts dated to the Middle Kingdom. Moreover, nothing can be guessed from the absence of the name of Re in his royal name (Netjerykhet) because it was formed according to other rules than later Old Kingdom royal names. There is no doubt that the mortuary complex of Djoser, who the ancient Egyptians considered to be the founder of the Old Kingdom, played a central role in the development of the royal necropolis at Saqqara. Perhaps the weakest point of the book is the stress the author puts on the rather spurious role of the priesthood of Re and the ambiguous attitude of kings under discussion towards it. The author considers the priesthood of Re to have been a mighty power, which sometimes forced the kings to make compromises and even appeasements, e.g. Userkaf (the enlargements of his sun temple) and Pepy I (the change of the throne name from Nefersahor to Meryre). However, there is no written evidence that could support the theory of a struggle between the royal and divine power. The reviewed book is a very important contribution to a better understanding of the highly problematic history of the later part of the Old Kingdom. More importantly, the study reveals to the eyes of a non-professional reader the method, fascinating in itself, of extracting historical information from a variety of archaeological material. As for the professional reader, the main asset of the book lies in the extensive collection of relevant archaeological and textual material—sometimes not yet published—, the historical analyses of this evidence, and the ingenious personal interpretation of the sources by the author. The author is known for his great scholarly erudition, yet his style of writing is vivid and the language elegant. We are presented here with a study Ž lled with material sources, arranged in a concise and accessible manner to the beneŽ t of our reading pleasure. Petra Vl ková Czech National Centre for Egyptology, Prague Czech Institute for Egyptology, Prague [email protected]

REVIEWS

137

Beshara DOUMANI (editor), Family History in the Middle East, Household, Property and Gender, Albany, SUNY Press, 2003. 340 pp. $86.50 (cloth), $29.95 (paper). ISBN 0-79145680-3. This book is one of a growing number of works devoted to family in the Middle East. It contains twelve articles that cover a regional focus on Egypt and Syria in addition to an article on Iran. The last article in the book, written by Akram F. Khater, is the only one which studies a region outside of the Middle East; notably it explores Lebanese migrants in the United States, tracing some of the tensions they faced as they settled in their new homes and the painful conditions of their interaction with middle class America. The articles focus on the period from the mid- to late nineteenth century until the contemporary period, with the exception of two articles, by Beshara Doumani and Heather Ferguson, which deal respectively with Greater Syria in the period 1700-1860 and with Tarablus al-Sham in the seventeenth century. More importantly, they explore some little known regions and localities, rural, provincial and tribal families, material that is not often available to the historian. For instance, Erika FriedlÕs study of the villagers of Deh Koh in Iran between 1880-1990 explores a subject that few people know about, notably the inhabitants of a village in the Zagros Mountains, and tracing the long-term development of marital customs and relations as well as the expectations of the young spouses in this crucial period. The book, moreover, brings together research from several disciplines. The contributors are roughly evenly divided between historians on the one hand, and social scientists and demographers, on the other. The first two articles, by Phillipe Fargue and Tomoki Okawara, based on population census, identify demographic trends in nineteenth century Cairo and Damascus. The Cairo census of 1848 that Philippe Fargue analyses shows some surprizing results. It provides some important data on little known subjects like children, showing that many of them were separated from their parents at an early age, either by the death of their parents or by living outside their family residences. Fargue also indicates that the majority of households, some 70%, were structured along nuclear lines. Unlike Cairo, the dominant pattern in Damascus was a large household with a mean size of over eleven persons. Taken together with DubenÕs work on Istanbul households, FargueÕs and OkawaraÕs articles clearly show the great diversity in family structure in different regions of the Middle East and consequently the inadequacies of grouping together ÔArab familiesÕ and ÔMuslim familiesÕ as clearly identifiable entities. The issue of gender, the methods that women used to maintain some level of control over wealth and property, are explored by Annelies Moors, Martha Mundy and Richard Saumarez Smith. They show the spaces within which women negotiated their way within predominantly male-oriented societies. By exploring the ways in which wives tried to control possessions in gold, MoorsÕ study of a village near Nablus shows the tensions between Islamic law which privileged male descendents and inheritance practice in relation to gold. Likewise, Mundy and Smith explore the strategies used by women in the village of Kufr Awan in northern Jordan, in relation to the wifeÕs bridal payment or mahr which sometimes consisted of property rather than money. The larger issue of writing family history on a non-linear non-Eurocentric approach is somewhat more complex. This has long been an objective for historians of non-European societies whose histories still bear the stamp of the trends born of colonialism. The difficulties and complexities of this issue is that many decades after it was formulated, it still remains with us and the solutions offered remain partial. This book is not an exception. The transition to the modern family is a major concern in a number of articles. Fargue links this modernity to the fact that the focus of the 1848 population census was based on the individual as the statistical unit rather than a collectivity. For instance, it focussed on a personÕs economic activity, his employment and whether or not he was a student. Most of © Koninklijke Brill NV, Leiden, 2004 Also available online – www.brill.nl

JESHO 47,1

138

REVIEWS

the other articles dealing with the issue of modernity link it to the emergence of the nuclear family and as the move from a polygamous to a monogamous family. There is certainly much truth in this view, but the issue nevertheless remains problematic on a number of levels. One of these relates to the absence of references to the period prior to the nineteenth century. The presumption that multiple households or extended families were the only pattern or the dominant pattern in the seventeenth or eighteenth centuries has yet to be proven. As a matter of fact, studies on Cairene housing seem to indicate that the nuclear household existed and was perhaps even fairly common. One need only look at the living quarters of the rab¾, made up of small apartments which sometimes consisted of a single room, to suggest that the issue needs further consideration. My own book Habiter au Caire: les maisons moyennes et leurs habitants aux XVIIe et XVIIIe siecles (Cairo, 1991) explored various housing types used by the middle class very likely to have been inhabited by a nuclear family. In more ways than one, it would have been interesting to see how this phenomenon can be linked to the developments of the nineteenth century. Such links could also constitute one of the ways in which historians can move away from a non-Eurocentric approach. Likewise, the presumption that ÔnuclearÕ was evidence of ÔwesternizationÕ echoed in various articles in the book, can occasionally be misleading if it presupposes a single model of nuclear families. In reality, families (or couples) could be nuclear in their residential patterns while maintaining very close relations, financial or other, with their ÔextendedÕ families or with their parents. They could be bearing financial responsibility for their parents, or for their respective siblings, or on the contrary be financially dependent on them. Likewise, ÔnuclearÕ families might live in an independent lodging while at the same time they resided in the same locality or district as their extended families, presuming frequent interaction. Finally, the expansion of a nuclear family structure, in the sense of the monogamous marriage replacing the polygamous one and the disappearance of the tradition of having concubines also needs to be rethought. Both Kenneth Cuno and Mary Ann Fay, who studied royal and elite families in late nineteenth century Cairo, were interested in an exploration of the move from polygamy to monogamy. Both have insisted on the internal forces that could have brought this change about, rather than simply attributing it to western influence. One nevertheless wonders if any connection could be made between this trend towards the monogamous marriage and the radical expansion of prostitution which we know to have occurred at about that same time. Did the expansion of prostitution impact the nuclear household, was it one of the factors that allowed it to flourish? These are as yet unanswered questions, and the connection between the two trends may be impossible to establish. Yet to avoid a positivist approach, the darker sides of the nineteenth century should be evoked. As announced in the introduction, the book combines different methodologies. Historical demography based on population census was already mentioned. A number of articles by anthropologists are essentially based on field work and interviews. With their emphasis on material conditions, they develop an entirely different approach than analysis of the ÔIslamicÕ family based on an exploration of religious texts. The emphasis on practice rather than on legal provisions also permeates the articles based on court records. These three articles, by Beshara Doumani, Iris Agmon and Heather Ferguson, approach court records by an in-depth study of one or two cases. We thus get a close view of family disputes and of the way that family members, on the one hand, and the legal system, on the other hand, dealt with such disputes. We also see that the bulk of such disputes revolved around financial matters, like the wifeÕs claim to maintenance from her husband; disputes between various family members about an inheritance, about illegally withholding waqf revenues from the legal beneficiaires, or about misuse of funds, the allegations and counter-allegations. Admittedly, the balance between the individual case and the broader society, between the little known village or locality and the region as a whole, is not an easy one. The reader may occasionally find himself lost in the minutie of a family dispute narrated in great length, and ask what relevance the wealth of details of a single case recorded in a court register or of

REVIEWS

139

an interview with a family member can have for the larger picture and in what way it advances research towards a particular issue or debate of interest or controversy. The book contains some sloppy footnoting. In spite of its shortcomings, this collection of articles is an important addition to a growing literature on the family which scholars in the field hope with time will help provide an understanding not only of the family but of the broader society. Nelly Hanna Department of Arabic Studies American University in Cairo [email protected]

Tony Day, Fluid Iron: State Formation in Southeast Asia. Honolulu: University of Hawaii Press, 2002. Pp. xii, 339. References, Index. $56.00 (cloth); $21.95 (paper). This is a fascinating book, full of lively ideas synthesized from far-ranging and systematic reading, rich in comparative suggestion, and offering critiques of some more-or-less received ideas about the subject at hand. It is hard to imagine a Southeast Asianist with more than a passing interest in culture, politics, or history who wonÕt Ž nd this thought-provoking, and many scholars outside that Ž eld should also see in it a great deal that is worthwhile noting. At the same time, however, the book possesses a dark side: it is full of conceptual contradiction and hesitation, is deeply con icted about the nature of history as a discipline, and in the end it careens into precisely the postmodern cul-de-sac it appears to have hoped to avoid. More than most, this is a study both to learn from and argue about. Fluid Iron is primarily a sophisticated exercise in reconsidering how scholars have thought about the (more-or-less) Weberian ÒstateÓ in Southeast Asia, and in discussing certain elements which the author believes may be important to its formation (or, it might be argued, non-formation) and persistence. Despite the implication of its subtitle, the book offers neither a history of state formation in the region nor an argument explaining how states came into being and developed there. Questions about when a state may or may not be said to exist, or when a state may be described as ÒlocalizedÓ or not, are raised but not answered. Tony DayÕs chief aims are, instead, to put culture (or elements of culture such as kinship and violence) and cultural analysis Òback intoÓ the study of the Southeast Asian state, past and present; to suggest Ò ways of thinking across and beyond the reigning dichotomies that separate ÔtraditionalÕ from ÔmodernÕ Southeast Asia. . . . [and] one part of Southeast from anotherÓ (p. 37); to Òraise doubts about ÔchangeÕ . . . that great fetish-concept of the historianÓ (p. 290). These goals are pursued sensitively and with politesse, but what Day seems to be after is not merely an adjustment in what and how we consider history, but a tectonic shift. The book consists of an introductory chapter reviewing the (mostly theoretical) literature on Southeast Asian states (again, not so much their historical formation as their nature), followed by four chapters discussing the topics of kinship, knowledge, bureaucracy, and violence, and their relationship to state structure and behavior. A brief conclusion does more than summarize the whole. The opening chapter will be of widest appeal, and has already found its way into reading lists for advanced students. It offers deft commentary on a number of ideas about states and state formation from Weber, Marx, Foucault, and Tilly, as well as from Furnivall, Wolters, Anderson fr res, Geertz, Aung-Thwin, Leiberman, and Robinson, among others. We are reminded, for example, of the heavy legacy of dichotomous social science thinking about states—modern/traditional, Western/non-Western, rational/cultural, state/ society, and so on—which Day characterizes as Òaccount[ing] for virtually everything written about the state in Southeast AsiaÓ (p. 7). An important concomitant of this binarism, Day

140

REVIEWS

believes, is that ÒcultureÓ—a term which he does not deŽ ne—has been identiŽ ed, where it is considered at all, with whatever is not modern, and is neglected as a factor in studies of the state, particularly comparative ones. This is what he strives to escape. The argument driving the topical chapters is that the historical disciplineÕs propensity to view the trajectory between past and present in linear and essentialized or reductivist terms prevents us from appreciating both the ambiguities of culture and continuities across time. Day uses literary, social science, and historical studies to attempt to alter the balance, and many of the insights arising from this will resonate with the contemporary Southeast Asian scholar. He suggests, for example, that Southeast Asian societies are best viewed as the product of constant tension between opposites; that ÒWestern causes alone are not responsible for Ôre-feudalizingÕ and ÔcentralizingÕ indigenous Southeast Asia social and political practicesÓ (p. 89); that Southeast Asian societies have a greater tolerance for ambiguity than many others (p. 260ff); that coercion and resistance are interrelated, as are security and freedom; the West and Southeast Asia are (in many respects) not so different after all (p. 224). These and other ideas are skillfully drawn or constructed from parts provided by other authors, a welcome process of synthesis all too few Southeast Asian specialists have been willing to undertake. There are a few lapses, such as when an already rather rococo conceptualization of Thongchai WinichakulÕs is further ornamented to conclude nothing more startling than that the great nineteenth century Thai ruler Mongkut (Rama IV) was a transitional Ž gure (p. 99), but on the whole the result is both helpful and plausible. What Indonesia watcher is not grateful for, even seduced by, the proposal that the New Order may be seen as Majapahit, or that Taman Mini is akin to Angkor Thom? But how useful are such plausible ÒinsightsÓ as history? Problems arise when Day insists upon advancing what are not—or not yet, anyway—more than what the French call correspondances (roughly, connections or harmonies) to the level of (soft?) fact, and using them as principles with which to deconstruct fundamentals of the historianÕs craft. He seems determined, for example, to argue that there is no dichotomy—and therefore, one must infer, no appreciable or analytically useful distinction—between the traditional and the modern. Such distinctions, he says, are mere artifacts of Òhistoricism,Ó a troublesome term which he uses (not entirely defensibly, I think) to mean the practice of essentializing by theorizing back from the present to portray (and to measure and evaluate) the past. Day wants to Òdehistoricize our understanding of the pastÓ (p. 290). Here Day wavers. In several instances he takes a moderate approach, suggesting for example that IndonesiaÕs New Order was the Òpredictable outcomeÓ of centuries of Òdeveloping repertoiresÓ (p. 222), or that nineteenth century bureaucracy on Java owed something to Majapahit and also to seventeenth century Holland (p. 217). We are provided with no details, but we do get a sense of a gradual, time-sensitive or evolutionary process at work. In other cases, however, Day seeks to convince us that there is in fact no such process, that culture is culture then as now; it  oats above or outside of Time. A Ž lm about Flor Contemplacion is thus fundamentally the same as an ancient ceremony mobilizing departed kin (p. 53); contemporary states are fundamentally nothing more than the expanded family networks of ancient Òstates.Ó In order to make the argument work, Day emphasizes a postmodern insubstantiality of things: bureaucracies are simply Òrepertoires,Ó and states mere ÒeffectsÓ; they donÕt possess any real content that we can compare or analyze further, certainly not linearly, and they are what they are. He also leans heavily on a concept of hybridization, by which he appears to mean a kind of change which involves no long-term, evolutionary alteration, at least not of essences, but only dead-end, sui generis forms. In short, Day wants to say that change takes place, but at the same time to see it as non-temporal in nature and to insist that it cannot be real change. This is hardly a satisfactory way of solving the riddle posed by the tension between change and continuity in our representation of past and present, a riddle particularly important to Southeast Asian studies between the 1950s and 1970s. Most historians have long since come to terms with this tension and found ways of dealing with it. They understand that if history is about anything, it is about charting, or precisely describing, or ÒmeasuringÓ (in any num-

REVIEWS

141

ber of ways that may be chosen) change over time. They know as well that hybridization is not only about mules and ligers, which are sterile; in other (probably more frequent) cases hybridization is part of a continuing chain of evolutionary change, which sometimes produces new species. And they also know that Òculture,Ó which they recognize as a vague formulation, both changes and stays the same over time, like DNA strung over generations. Yet they expect (hope?) that both what changes and what does not change can be identiŽ ed, and that this identiŽ cation may result in a better understanding of both the process and the outcome. The concept of a linear—not predetermined and not at all necessarily Òstraight,Ó but linear in the sense of Ò traceableÓ—trail of causation, is key to making things understandable. Without it, past and present collapse into one another, reduced to nothing more than an indispersible fog; this too is a form of essentializing, and a particularly obscurantist one at that. To argue, as Fluid Iron does, that historians often underestimate and fail to appreciate the power of human agency (a.k.a. ÒcultureÓ), is fair enough, a sensible warning. But to argue as well, as Fluid Iron frequently appears to do, that the solution is to abandon linearity altogether and see change as a mere Ž ction, goes much too far. A notable curiosity in Fluid Iron is the authorÕs con icted approach to Southeast Asia as a region. This is a hoary issue over which generations of scholars have labored and argued, for often rather obscure reasons. Once again, Day wants to have it both ways: on the one hand Southeast Asia is not a region, certainly not in the supposedly self-evident sense of, say, Europe or East Asia; nothing there is Òauthentic,Ó and whatever civilization or civilizations may be said to exist there are ÒpolyvalentÓ and Òcon ictualÓ rather than ÒpureÓ (p. 293). At the same time it is a region; borrowing from de Certeau, its cultural interactions deŽ ne it and distinguish it from other regions. It is even unique (implicitly, among regions): its states, for example, are Òlike no other historical formations on earthÓ (p. 291). At the very end of the book, Day struggles to describe Southeast Asia, rejecting several possibilities raised in recent works, ending by declaring that Southeast Asia is Òincongruous,Ó though it is not entirely clear whether this means it is unique or not, a region or not. (Are other world regions or, for that matter, nations not incongruous? Where in this case does the incongruity shade off into congruity? South Asia? China? How? Surely there are strong arguments against placing much stock in the characterization.) Very early in the book, Day seeks to head off the problem by pleading that he is merely using the concept of Southeast Asia as a Òheuristic frameÓ (p. 37). (Presumably in a postmodern ÒborderlessÓ world it is politically incorrect to speak seriously about regions just as it is about nations.) On the one hand, this is stating the obvious: of course the idea of regions is a device, a tool of fairly high-level generalization well-recognized as such, no more no less. On the other hand, using such a device to shape a bookÕs worth of argument only to attempt to disavow its value in the end seems to me perverse. Throughout Fluid Iron Day in fact provides us with ample and vivid evidence that across the region comparisons and connections and harmonies can indeed be found that give us more reason than ever to see Southeast Asia as a region whose parts have something to do with each other and may be usefully and responsibly generalized about. Waf ing over or throwing up oneÕs theoretical hands at inauthenticity and incongruity is unwarranted, selfdefeating and—given the material presented here—quite unnecessary. William H. Frederick Ohio University, USA [email protected]