Al-ʿUṣūr al-Wusṭā 26 (2018): 258-269 Book Review The book focuses on a central topic of medieval Islam: Baghdad and its elites under the Saljuqs, between 447/1055 and 575/1180. This Baghdad was the caliphal capital where al-Ghazālī (d. 505/1111) taught and where Ibn al-Jawzī (d. 597/1200) preached—a city trans- formed by the foundation of madrasas and ribāṭs, and fought over by the Saljuq Turks, their Iranian viziers, their emirs, the ʿayyārs, and the Abbasids Caliphs. The city was indeed the heart of “Traditional Islam,” to quote the late George Makdisi (d. 2002), the leading figure in “Baghdad studies” during this key period. For all these reasons, the potential readership of Van Renterghem’s book is far greater than what is normally expected for monograph on a medieval city. T h e a u t h o r a i m s t o o f f e r a s o c i a l history of Baghdad by focusing on the best documented section of the population: its elites (p. 21-22). Her book is based upon a PhD dissertation submitted in 2004 at the University of Paris, Sorbonne. The result is two volumes, 3.4 kg, a thousand pages in-quarto; 30 maps and graphics, 53 tables and 34 genealogical trees; a lexicon including about 550 entries; and a bibliography of more than 650 titles. Imposing in its size, the work is also unique in its statistical basis. At the core of the book is a database of 2,639 persons having lived in Baghdad long enough to be considered as Baghdadi. To generate t his p ros op og rap hic al d at abas e, V an Renterghem has dug through in a vast corpus of 23 biographical sources (listed in Table 1.1, and discussed in Vol. 1: 25-39). Unexpectedly, Ibn al-Jawzī’s Muntaẓam a n d I b n a l - ʿ I m ā d ’ s ( d . 1 0 8 9 / 1 6 7 9 ) Shadharāt contain the highest number of relevant notices. But other sources, often disregarded, have proven invaluable: for example, 70% of the notices drawn from al-Bundārī’s (d. after 639/1241–2) Dhayl Taʾrīkh Baghdād (still in manuscript at the Bibliothèque Nationale de France) are not found anywhere else, and the figure is 74% for Ibn al-Najjār (d. 643/1246) (see Vanessa Van Renterghem, Les élites bagdadiennes au temps des Seldjoukides. Étude d’histoire sociale, 2 vols. (Beirut-Damascus: Presses de l’IFPO, 2015), Vol. 1 xx + 530 pp., Vol 2. 493 pp. ISBN: 978-2-35-159704-0. Price: €150 (Paper). Available in OpenEdition Books (https://books.openedition.org/ifpo/9172). David Durand-Guédy (david_durandguedy@yahoo.com) mailto:david_durandguedy%40yahoo.com?subject= Al-ʿUṣūr al-Wusṭā 26 (2018) 259 • DaviD DuranD-GuéDy Table 1-2). Van Renterghem explains in detail how the database was designed, with its 97 fields (see Vol. 2: 5-21). This huge dataset fuels the maps, graphics and tables that make up most of the second volume. These figures are, in turn, commented upon in great detail in the first volume of text. As such the work honors the French tradition of statistical method applied to history. After all, it was Pierre Chaunu who founded quantitative history in the 1950s.1 For Medieval Islam, it follows Dominique Urvoy’s milestone work on the ʿulamāʾ of al-Andalus, as well as the Onomasticon Arabicum project.2 Outside of France, this statistical approach has been developed and theorized by Richard Bulliet.3 These studies are mentioned in the introduction, along with a discussion on the concept of elites (its inception in the seminal work of the Italian sociologist Pareto as a reaction to Marxist theory; the new approach brought by prosopography; the major contributions of Elias and Bourdieu) and an analysis of the Arabic t e r m s u s e d i n t h e s o u r c e s ( k h ā ṣ ṣ a / ʿāmma, aʿyān, bayt, etc. ). Eventually, Van Renterghem defines the Baghdadian elites as follows: they enjoyed a superior social status recognized by their peers and by the rest of the population; they developed strategies of legitimation and distinction; and they lived in Baghdad long enough to implement these strategies (p. 21). Van Renterghem thus differentiates herself from the classical Weberian trilogy of power, money, and prestige as she 1. Pierre Chaunu, “Histoire quantitative ou histoire sérielle,” Cahiers Vilfredo Pareto 2/3 (1964): 165-176. 2. Dominique Urvoy, Le monde des ulémas andalous du ve/xie au viie/xiiie siècle. Étude sociologique (Geneva: Droz, 1978). The Onomasticon Arabicum is available online: onomasticon.irht.cnrs.fr (it now aggregates over 25,000 records). 3. Richard Bulliet, Conversion to Islam in the Medieval Period: An Essay in Quantitative History (Cambridge: Harvard Univ. Press, 1979). considers that economic factors are not paramount. T h e a u t h o r s u c c e s s i v e l y t a k e s t h r e e d i f f e r e n t a p p r o a c h e s : f u n c - tional, sociological and spatial. After a presentation of the sources, the first section of the first volume contains a functional analysis of the various fields in which elites can be identified: religious authorities (Chap. 2), sufism (Chap. 3), traditional sciences (i.e. transmitters and ʿulamāʾ) (Chap. 4), the judiciary (Chap. 5), the military (Chap. 6), the divans (Chap. 7), the Abbasid court (Chap. 8), and the bazars (Chap. 9). I n t h e s e c o n d s e c t i o n , t h e a u t h o r switches to a socio-historical approach. It focuses on the “practices of distinction” in a set of activities: eating, moving, and getting dressed (Chap. 10). Van Renterghem then turns to the “signs of reverence” d u r i n g t h e p r o c e s s i o n s ( m a w ā k i b ) , funerals, or investiture ceremonies (Chap. 11). Elaborating upon this Bourdieusian m o d e l , t h e n e x t c h a p t e r h i g h l i g h t s practices of “social reproduction” through the example of fifteen great families, such as the Dāmghānis (who controlled the office of the qāḍī al-quḍāt throughout the period), the family of Ibn al-Jawzī, the Zaynabīs (the great Hanafi family of Baghdad), and the Banū Muslimas. This part, which highlights the links and the unity within the elites, counterbalances the analytical presentation of the first section (although we had already been told http://onomasticon.irht.cnrs.fr Al-ʿUṣūr al-Wusṭā 26 (2018) Vanessa Van Renterghem’s Les élites bagdadiennes au temps des Seldjoukides • 260 that the Sufis and the faqīhs had followed the same curriculum, p. 103). The third and final section of the book takes a spatial turn. After a presentation of the topographical framework, Van Renterghem analyzes in great d et ail where these men (and women) lived and died, and also how their presence was felt in the urban space (through mawākib, funeral processions, or festivals) (Chap. 13). Next, she scrutinizes the competition between various actors for urban control: the Saljuqs (through the ʿamīd, and the shiḥna), the caliph (through the ḥājib and the muḥtasib), and the Hashemites (through the naqīb) (Chap. 14). Finally, C h a p t e r 1 5 e x a m i n e s t h e p o l i t i c s o f architectural patronage in Baghdad. This last chapter is typical of how Van Renterghem proceeds. The text comments on two detailed tables and five maps. Table 15-1 (12 pages long) is a list of the 100 building projects launched between 447/1055 and 573/1177 in Baghdad, not counting the madrasas and ribāṭs, which are dealt with elsewhere. Table 15-2 lists the 66 known patrons of building projects, this time including madrasas and ribāṭs (112 in total because powerful and wealthy patrons could launch several projects). Map 17 synthetizes all the data, and maps 18 to 21 contain a more precise representation of the areas favored by v a r i o u s a c t o r s ( r e s p e c t i v e l y S a l j u q s , Abbasids, Abbasid officials, and civilians). The tables are extremely detailed, and conveniently include mention of the primary sources. The maps are well drawn, in color, and very readable. Many of the diagrams and figures which aggregate the results of the database are truly enlightening. For example, a series of four charts elucidates the evolution of hadith transmission in Saljuq Baghdad. Elaborating upon a method designed by Dominique Urvoy, Van Renterghem is thus able to identify four generations of hadith transmitters. With the exception of the first generation, they include an equivalent number of masters (about 17) and disciples (about 550). The diagrams (4-2 to 4-5) show how many disciples the various transmitters had in common. In the first period (before the arrival of the Saljuqs), we see a clear hierarchy within two groups of transmitters which do not communicate; conversely, in the second period (transmitters who died between 439/1047 and 479/1086), the network is remarkably integrated (i.e. a student has many professors, and no professor stands out); in the third period, the network has lost density and five transmitters prevail; finally, in the last period (transmitters who died between 510/1116 and 564/1168), the network is strongly polarized around t h e f i g u r e o f o n e t r a n s m i t t e r ( t h e pro-Ḥanbalī Ibn al-Ḥusayn, d. 525/1131). Van Renterghem concludes that “we can witness a tightening of the circle of hadith transmitters around ever less numerous figures and ever more dominant ones” ( p . 1 3 9 ) . S h e o f f e r s n o d e f i n i t i v e explanation for such a trend—though she suggests that the distinction between the Ḥanbalīs and the other Sunni groups may have played a role—but the visual representation of this phenomenon is striking. Another valuable series of illustrations are the maps of residential areas. Several filters are successively applied and the results are telling. While hardly any social zoning can be highlighted (see maps 2 and 3, which show that simple transmitters and high-profile elites live Al-ʿUṣūr al-Wusṭā 26 (2018) 261 • DaviD DuranD-GuéDy in the same districts), we clearly see that the center of gravity of Baghdad moved to the East during the century of Saljuq rule (compare maps 4 and 6). Also compelling is the evidence of residential segregation according to madhhab (map 10) and occupation (map 7-9). We now have a clear idea of where the Ḥanbalīs were living: South-West of Dār al-Khilāfa on the East side, and South of the Jāmiʿ al-Manṣūr on the West side. I could mention dozens of other such valuable “hard facts” offered throughout the book. I cannot resist mentioning two further examples: of the 31 Sufis who are known to have studied fiqh, 29 are Shāfiʿite (p. 105)! This is quite striking. Similarly, I found interesting the low proportion of mystics among the persons who enjoyed divine baraka (4 out of 32): unlike later periods, the Sufis did not have the monopoly on baraka. The endeavors of the author to provide her readers with a robust documentary basis on which the analysis is built is admirable. History, however, is not a science and Van Renterghem does not pretend that she has found the Grail (she is aware of the unbalanced character of her corpus), but as long as new sources are not discovered (which is very hypothetical), it seems unlikely that the general picture painted in her tables can be challenged. That said, the whole project suffers from two problems. The first is the decision to hermetically isolate the documentary evidence (maps, tables, trees, graphics) from the text; all this valuable material is 4. The running head at the top of the page only refers to the number of any given table, not to its subject. Since the volume lacks a detailed list of figures, precious time is wasted in searching for them. 5. Strikingly, one previous reviewer of the book starts by stating that he will discuss only the first volume, as if the “annexes” (appendices) contained only marginal information. See M.H. Benkheira, Studia Islamica 112/2 (2017), 303-314, here p. 303. relegated to the second volume (alongside the bibliography and indices). This choice is questionable for a volume of this scale, because following the argumentation requires constantly navigating between two in-quarto volumes, an operation which is hampered by the type of binding (the volume often closes by itself) and the absence of clear running heads.4 I am aware that one figure can be used in multiple passages, and appendices work just fine with significantly smaller books. In this case, however, this sort of reasoning has proved counterproductive, and many figures (all the genealogical trees, all the diagrams and maps; many tables) could easily have been inserted into the text section.5 Another problem is the content and scope of the book itself. It says too much and not enough at the same time. Too much because Van Renterghem seems to refuse to choose what story she wants to tell with her abundant material. As a consequence, for each category or issue, she adds up quantitative data, a n t h r o p o l o g i c a l a n a l y s i s , a n d c a s e - studies. Here again there is a scale effect: what is possible for other cities less well documented does not work well here. Take one example: in the corpus of sources on Saljuq Isfahan, I have not been able to find a single mention of a muḥtasib, though such figures naturally existed in that place and time as in any other city; but for Baghdad, 23 muḥtasibs can be identified, often with many details (see Table 14-3). Multiply this by the number of “catégories élitaires” Al-ʿUṣūr al-Wusṭā 26 (2018) Vanessa Van Renterghem’s Les élites bagdadiennes au temps des Seldjoukides • 262 defined by Van Renterghem and it is easily understandable how we end up with this behemoth of a book. This is aggravated by the overreaching approach: thus, in the first section of the book, Van Renterghem deals with the “milieux élitaires,” but i n s t e a d o f r e s t r i c t i n g h e r s e l f t o t h e relevant positions (preachers, etc.), she also adds a spatial dimension and speaks about the institutions (she deals with the ribāṭ in the chapter on the Sufis, with the madrasa in the chapter on the traditional sciences). This leads to repetition in the last section of her discussion, which is space-oriented. There are, indeed, a very high number of repetitions throughout the text. These are sometimes small anecdotes,6 but also include much longer developments.7 At some point we would have expected the author to choose an angle and stick to it, even if it meant leaving aside a vast amount of hard-won data (scholars’ computers are full of such treasures waiting to be dealt with!). At the same time, much of the story that the book contains has already been told. Van Renterghem’s work is truly admirable, and every scholar has a natural tendency to overvalue his or her contribution to the field. But the case of Saljuq Baghdad 6. 50 pages apart, we read exactly the same anecdote about the faqīh Abū Isḥāq Shīrāzī, who refused to pray in the madrasa in which he teaches; cf. pp. 423-424 and p. 486, with full references in both cases. 7. The lexicon of social preeminence is dealt with in the introduction (pp. 14-20), then again in Chapter 11 (pp. 326-329); the function of the Abbasid ḥājib is presented in the functional analysis (Chap. 7, pp. 220-222) and also in Chapter 14, which focuses on social order (p. 468). The mawkib is dealt with in Chapters 11 (pp. 329-330) and 13 (p. 436). Even the concept of elites is explained as far into the book as page 416. Other examples abound: e.g., the issue of clothing imposed on dhimmis (p. 302 = p. 209), and the role played by women in urban development (pp. 248-251 and pp. 495-500). Other examples abound: e.g., the issue of clothing imposed on dhimmis (p. 302 = p. 209), and the role played by women in urban development (pp. 248-251 and pp. 495-500). 8. George Makdisi, “The Topography of Eleventh Century Baġdād: Materials and Notes,” Arabica 6/2 (1959): 178-97 and 6/3 (1959): 281-309. can hardly be compared to the state of scholarship on pre-Mongol Nishapur when Bulliet started investigating it, or on Saljuq Isfahan when I embarked on its study. “Very few works have been carried out on Iraqi cities, including Baghdad, despite its status as the seat of the caliphate,” Van Renterghem writes in the introduction (p. 9). The presentation of the scholarship which follows aims to substantiate that claim. Le Strange’s topographic study is mentioned briefly in a footnote. George Makdisi’s seminal work is alluded to in only a few lines, in a rather curious way: can we really say that he limits himself to “political and intellectual history” and “does not aim to analyze urban society” (p. 10)? While he never engaged in quantitative analyses, Makdisi insisted on the fact that the life of a public person such as Ibn ʿAqīl ( d . 5 1 3 / 1 1 1 9 ) c o u l d o n l y b e t r u l y understood through a global analysis of the period, space, and milieu in which he lived.8 Also, his substantial article on the madrasas not only includes a list of all the “institutions of learning” in the Saljuq period, but also proposes a new understanding of their function (a tool for the elites “to control the masses”) based on a reflection on the power relationship Al-ʿUṣūr al-Wusṭā 26 (2018) 263 • DaviD DuranD-GuéDy in that very time and space.9 Likewise, his article on the topography of Baghdad10 is much more than an arid list of toponyms: Makdisi shows how the urban space was fought over between the Abbasid caliph and the Saljuq sultan, which is precisely the subject of Van Renterghem’s last chapter (15). And Makdisi’s annotated translation of the Journal of Ibn al-Bannāʾ ( d . 4 7 1 / 1 0 7 9 ) , w h i c h i s n a t u r a l l y abundantly used by Van Renterghem, provides a unique insider perspective on an urban (Ḥanbalī) community.11 In the same section, one would have expected at least one reference to Simha Sabari’s study on popular movements in Abbasid Baghdad, insomuch as Claude Cahen— Sabari’s supervisor—was the first to connect the fitnas with power struggles among the elite (Sabari is dismissed much later in a footnote as too “descriptive”).12 B r u s h i n g a w a y M a k d i s i , V a n Renterghem dedicates a little more to E p h r a t ’ s w o r k o n t h e S u n n i ʿ u l a m ā ʾ of eleventh-century Baghdad. It was published in 2000 and was not particularly well received. Indeed, reviewers pointed out that her study was marred by too many factual errors, by several essential sources 9. George Makdisi, “Muslim Institutions of Learning in Eleventh-Century Baghdad,” Bulletin of the School of Oriental and African Studies 24/1 (1961): 1–56. 10. George Makdisi, “The Topography of Eleventh Century Baġdād: Materials and Notes,” Arabica 6/2 (1959): 178-97 and 6/3 (1959): 281-309. 11. George Makdisi, “Autograph Diary of an Eleventh-Century Historian of Baghdād,” BSOAS 18/1 (1956): 9-33, 18/2 (1956): 239-260 and 19 (1957): 13-48. 12. Simha Sabari, Mouvements populaires à Bagdad à l’époque ʿabbasside, ixe-xie siècles (Paris: Maisonneuve, 1981). Claude Cahen, “Mouvements populaires et autonomisme urbain dans l’Asie musulmane du Moyen-Age,” Arabica 5 (1958) and 6 (1959) (with separate pagination). 13. Daphna Ephrat, A Learned Society in a Period of Transition: The Sunni ʿUlamaʾ of Eleventh-Century Baghdad (Albany: State University of New York Press, 2000). See, notably, the reviews by T. El-Hibri in the International Journal of Middle East Studies 34 (2002): 736-738; V. Van Renterghem in the Bulletin Critique des Annales Islamologiques 18 (2002): 65-67; and, most importantly, Shahab Ahmed in the Journal of the American Oriental Society 123/1 (2003): 179-182. and studies simply overlooked and, above all, by a thesis which proved untenable.13 That being said, and despite what Van Renterghem implies, her book and Ephrat’s have much in common: the same main biographical sources, the same focus on the ʿulamāʾ (not the only elites dealt with by Van Renterghem, but certainly the most documented), and the same themes (e.g. assessing Baghdad’s attraction or the family background of the ʿulamāʾ). M o r e o v e r , i n a n u t s h e l l b o t h b o o k s share the same approach: statistics cum historical anthropology. Van Renterghem has not referenced modern authors in her 88-page index section, but an online search reveals that she refers only exceptionally to Ephrat’s study outside the introduction. These few examples lead to a broader concern: Van Renterghem too seldom refers to the existing scholarship in the course of her argumentation. She justifies herself in the introduction by the wish to stay immune from any “foreign problematics,” be it that of Medieval Europe (e.g., the issue of autonomy, which largely defined the history of Western cities) or of “Islamic cities,” especially in Syria and Egypt. The fact is that the Al-ʿUṣūr al-Wusṭā 26 (2018) Vanessa Van Renterghem’s Les élites bagdadiennes au temps des Seldjoukides • 264 questions at the heart of Ephrat’s study were derived from Michael Chamberlain and Joan Gilbert’s studies on Damascus.14 I can understand this line of reasoning (i.e., to let the sources speak for themselves), but this works best when no source stands above the others, like in Bulliet’s Patricians, Garcin’s Qūṣ, or the urban studies based upon Ibn ʿAsākir (d. 571/1176) or al-Khatīb al-Baghdādī (d. 463/1071).15 This is not the case with Saljuq Baghdad. In some cases, Van Renterghem adds a note to say that her results confirm other studies (e.g., on the precarious situation of the viziers, p. 243, or on the versatility of the emirs, p. 218). On rare occasions, recent studies are discussed in the text (Aloha’s u n p u b l i s h e d s t a t i s t i c a l i n q u i r y i n t o al-Khaṭīb al-Baghdādī’s dictionary is used as a comparendum, p. 131; the same is true for Tor’s book on the ʿayyār, p. 457).16 But the level of critical engagement with the existing scholarship remains insufficient. For example, at the outset of the second section, the author introduces examples o f “ s o c i a l d e a t h s ” ( d i s g r a c e , i n f a m y parade) without referring to Christian Lange’s essential book on the subject.17 14. Joan Gilbert, The Ulama of Medieval Damascus and the International World of Islamic Scholarship (Ph.D. dissertation, University of California Berkeley, 1977); Michael Chamberlain, Knowledge and Social Practice in Medieval Damascus, 1190-1350 (Cambridge: Cambridge University Press, 1994). 15. Richard Bulliet, The Patricians of Nishapur: A Study in Medieval Islamic Social History (Cambridge, Mass.: Harvard University Press, 1972); Jean-Claude Garcin, Un centre musulman de la Haute Egypte médiévale: Qūṣ (Cairo: IFAO, 1976 ; 2nd ed. 2005). 16. Deborah Tor, Violent Order: Religious Warfare, Chivalry, and the “ʿAyyār” Phenomenon in the Medieval Islamic World (Würzburg: Ergon, 2007); Judith Ahola, The Community of Scholars: An Analysis of the Biographical Data from Ta’rīkh Baghdād (PhD St Andrews, 2004). 17. Christian Lange, Justice, Punishment and the Medieval Muslim Imagination (Cambridge: Cambridge Univ. Press, 2008). 18. See Muḥammad Badrī Fahd, Taʾrīkh al-ʿIrāq fī al-ʿaṣr al-ʿabbāsī al-akhīr, Baghdad: Irshād, 1973. Fahd has also investigated the amīr al-ḥajj before Van Renterghem (p. 206-10); see his article “Taʾrīkh umarāʾ al-ḥajj,” al-Mawrid 4 (1981). 19. See Kazuo Morimoto (ed.), Sayyids and Sharifs in Muslim Societies: The Living Links to the Prophet (London and New York: Routledge, 2012) with extended reference to previous scholarship. Van Renterghem rightly considers the niqāba as a key institution, and the naqīb appears in several passages of the book as one of the most important public persons in Baghdad. But can we say with the author that “the niqāba is an institution not well known” (p. 84, repeated p.474, with a reference to a single article in Italian), when the naqībs of Baghdad are dealt with by Badrī Muḥammad Fahd, a scholar who, incidentally, has published extensively on Baghdad during the Saljuq period, but is never mentioned?18 This is to say nothing of a new stream of research, best exemplified by the studies of Kazuo Morimoto, who has profoundly renewed our understanding of the issue.19 Can we speak of the guilds (“corps de métiers,” p. 440) without hinting at the substantial bibliography on the subject, in particular in relation to the Abbasid capital studied long ago by Massignon. A quick look at the footnotes reveals that Van Renterghem hardly refers to the relevant scholarship, even when she tackles issues as hotly debated as the attitude of the ʿulamāʾ toward rulers, the relationship between Al-ʿUṣūr al-Wusṭā 26 (2018) 265 • DaviD DuranD-GuéDy the Saljuqs and the Shīʿites, and the place of the horse in medieval societies. This is also true for individuals or specific families. The Ibn Jahīr family and Abū Shujāʿ are well known to specialists of the period, but a reference to Hanne’s study on the Abbasid vizierate during the Saljuq period would be expected.20 With regard to the shiḥna (i.e., the military governor representing Saljuq power), Van Renterghem’s lengthy discussion offers a case study on the career of Jawhar Āʾīn. But she does not mention the key thing for which Jawhar Āʾīn was famous, namely his role in the capture of the Byzantine emperor at Manzikert, nor an article encapsulating the career of the same emir.21 For the many individuals connected to Isfahan and discussed by Van Renterghem throughout her book, much more relevant material would have been found in my own work.22 And that is not even to mention scholarship in Turkish and Persian. Another consequence of her lack of engagement with the existing scholarship is that Van Renterghem misses several key problematics. I will limit myself to three examples that illustrate the problem. Van Renterghem claims that ʿAmīd al-Mulk 20. Eric Hanne, Putting the Caliph in His Place: Power, Authority, and the Late Abbasid Caliphate (Madison, N.J.: Fairleigh Dickinson Univ. Press, 2007). 21. Kosuke Shimizu, “Amīr Gawhar Ā’īn,” Orient 32 (1997): 26-36. 22. David Durand-Guédy, Iranian Elites and Turkish Rulers: A History of Iṣfahān in the Saljūq Period (London & New York: Routledge, 2010), e.g., p. 185 about the two qāḍīs al-Khāṭibi, who played a critical political role at the Saljuq court; or p. 466, n. 146, regarding an emir who occupied the post of shiḥna in Isfahan and in Baghdad (for whom Van Renterghem erroneously writes that “no notice is available”). 23. Heinz Halm, “Der Wesir al-Kundurī und die Fitna von Nīšāpūr,” Die Welt des Orients 6/2 (1971): 205-233. Generally speaking, Van Renterghem lists German scholarship in the bibliography but never in the text. For example, about the Shiʿis during the Buyid period (p. 81), Van Renterghem refers to Donohue’s book (which she criticizes elsewhere), but not Busse’s authoritative work. John J. Donohue, The Buwayhid Dynasty in Iraq 334 H./945 to 403 H./1012: Shaping Institutions for the Future (Leiden: Brill, 2003). See Heribert Busse, Chalif und Grosskönig. Die Buyiden im Iraq (945-1055) (Beirut: Orient-Institut der Deutschen Morgenländischen Gesellschaft, 1969), 405-431. al-Kundurī, Toghrïl Beg’s vizier, was “well-known” for his “Hanafi leanings” (p. 184). This view is that of Ibn al-Athīr and al-Subkī, two of the main sources used by Van Renterghem. But one of the most famous—and brilliant—scholarly article about Saljuq rule (Halm’s on the fitna in Nishapur) precisely shows that al-Kundurī’s madhhab was not clear, and that he remained close to the great Shāfiʿī families of Nishapur to whom he owed his rise.23 As a second example, in the section on clothing, Van Renterghem refers to an Ashʿarī preacher walking through the city surrounded by armed bodyguards ( p . 3 0 4 ) . T h i s a n e c d o t e c o u l d h a v e provided the occasion to discuss the issue of the militarization of society. Following Cahen’s pioneering analysis, important research (in particular by Jürgen Paul) has been dedicated to this important issue of “civilian” elites engaging in warfare in the Iranian world. Interestingly enough, the preacher mentioned by Van Renterghem was an Iranian (his nisba is al-Ṭūsī). As a final example, also concerning military matters, the term khādim is recurrent in the sources, but beyond its generic meaning of “servant,” it could mean more specifically “eunuch,” a key figure Al-ʿUṣūr al-Wusṭā 26 (2018) Vanessa Van Renterghem’s Les élites bagdadiennes au temps des Seldjoukides • 266 in pre-modern Muslim polities as Ayalon has shown. (Van Renterghem exclusively translates khādim as “serviteur,” hence s h e c o n s i d e r s t h a t a k h ā d i m c a n n o t be shiḥna, p. 467). This is all the more frustrating given that Ayalon’s relevant studies on the subject are duly listed in the bibliography.24 The lack of a broader perspective is particularly problematic when dealing with the Saljuqs themselves. In the very first pages of the book, we read that “the sultans did not stay much [résident peu] in Bagdad, which was a secondary basis of their power” (p. 4); later, in the chapter on the military elites, the “instability of the Saljuq regime” (p. 197) is noted, but merely as one of the various threats to Baghdad’s security. But nowhere do we read the obvious: any Saljuq ruler with imperial ambition had to control Baghdad, the seat of the caliphate. At the same time, however, he could not cut himself off from his main source of power, which remained on the Iranian plateau. This is the reason why we see an itinerant pattern emerging at the end of the reign of Malik-Shāh that would last until 547/1152, when the Saljuq state imploded. According to this practice, the sultan and his court spent their winters in Mesopotamia, usually in Baghdad, and moved back to the plateau during the spring. (Isfahan, the first dār al-mulk, was abandoned for Hamadan when the “instability of the Saljuq regime” imposed not only shorter routes, but also 24. See especially David Ayalon, “The Mamlūks of the Seljuks: Islam’s Military Might at the Crossroads,” Journal of the Royal Asiatic Society, Series 3, 6/3 (1996): 305-333, here p. 306. Reprinted in Eunuchs, Caliphs and Sultans: A Study of Power Relationships (Jerusalem: The Magnes Press, 1999). 25. This is an issue I have tackled in my book (Iranian Elites, pp. 319-323 with a table listing the stays of the sultans in Isfahan or elsewhere). I have further investigated the topic in subsequent articles (quoted by Van Renterghem in her bibliography), e.g. “Where Did the Saljūqs Live? A Case Study Based on the Reign of Sultan Masʿūd b. Muḥammad (1134-1152),” Studia Iranica, 42 (2011): 211-58. greater proximity to Azerbaijan, the new mainstay of power).25 After the end of the dynastic crisis (485/1092–498/1105) and the victory over the Ismailis of Isfahan in 500/1107, Sultan Muḥammad travelled to Baghdad every winter, not just three times as posited by Van Renterghem (p. 484, n. 3). Although his son Masʿūd spent 12 of the 18 winters of his long reign in Baghdad, it was clearly not he who “made Baghdad one of his residences” (p. 225). Assessing the Saljuq presence in the city correctly is not a small issue for understanding the consequences of the Saljuq domination of local society. Beyond some of the specifics, it is the global logic of this pattern that is missed: Alp Arslan never bothered to go to Baghdad because the presence of the most powerful Saljuq Sultan was not needed there to have his authority respected. But his great grand-son Masʿūd kept going to Baghdad precisely because he was weak: he had to be physically present to keep the centrifugal forces in check (since he could not physically visit all his territories, he focused on Baghdad and Hamadan, and let Fārs and Azerbaijan slip away). Similarly, Van Renterghem is, I believe, mistaken about the relationship of the Saljuqs to city life. She contends that the sultans preferred the way they were received in Iranian cities (p. 485) but, in Bagdad as in Iran, the sultans actually did not live inside the city. At best they spent time by the walls, in a military camp. In Baghdad, this camp dated back to the Buyid Al-ʿUṣūr al-Wusṭā 26 (2018) 267 • DaviD DuranD-GuéDy period and is called in the sources the Dār al-mamlaka; it had several buildings and its own Friday mosque. But elsewhere the Saljuq court was set in a tented encampment, sometimes with a pavilion called kūshk in the Persian sources (see Turk.: köşk, but not kishk as the author writes erroneously on p. 493).26 Does this mean that the Saljuqs were “Türkmens”? Van Renterghem constantly r e f e r s t o t h e m a s s u c h : “ s o u v e r a i n s t u r k m è n e s , ” “ d y n a s t i e t u r k m è n e , ” “sultanat turkmène,” “empire turkmène.”27 They were, indeed, the leaders of Türkmens at the very beginning, but they quickly became much more than that and soon presented themselves as legitimate Iranian rulers. It is true that, contrary to what has been argued in the past, the Saljuqs did not sever their links with the Türkmens after the establishment of the sultanate.28 Likewise, Saljuq court poets celebrated the Turkish identity of their patrons. But it is inaccurate to systematically speak 26. See David Durand-Guédy, “The Tents of the Saljuqs,” in Turko-Mongol Rulers, Cities and City Life ed. David Durand-Guédy (Leiden-Boston: Brill, 2013), 149-189. 27. Ibid., 4, 48, 506, 507, 516. 28. For a critic of “diversion theory” (i.e. Iranized Saljuq sultans diverting turbulent Turkmen tribes to frontier regions in the Caucasus and Anatolia), see, following Cahen’s insight, A.C.S. Peacock, “Nomadic Society and the Seljūq campaign in Caucasia,” Iran and the Caucasus, 9/2 (2005): 205-230; A.C.S Peacock, Early Seljūq History. A New Interpretation (London and New York: Routledge, 2010), 139-151. 29. When Van Renterghem says that the “Saljūqs kept their nomadic habits and… were accompanied by their families, mounts, and herds” (p. 314), the first part of the sentence can be supported, but the last half is at best misleading (Saljuq armies could be followed by some cattle, just like any other army). 30. See pp. 393-394, about “the presence of the Turks and the Turkmens” in Baghdad; see p. 459 about the departure of the “Turkmen troops from Baghdad.” 31. Qimāj (cf. index) for Qumāj. Qumāj of Balkh was the key actor of the fall of Sanjar’s rule in the East. 32. Actually all the names in Table 14-2 are Turkish. The “Salār Kurd” of the year 542/1147 is not a name but a function. of the Saljuqs as Turkmens. It gives the false impression that they were nothing more than pastoral nomads, which was certainly not the case, at least from the time they occupied Baghdad in 447/1055.29 Powerful groups of Türkmens occupied parts of the mountainous regions East of Baghdad (especially in the Liḥf region, or around Shahrazūr), but Van Renterghem’s text give the impression that they were systematically inside Baghdad.30 The overall lack of familiarity with the Saljuqs and with Saljuq rule is all too visible throughout the text. Thus, V a n R e n t e r g h e m f o r g e t s t o m e n t i o n N i s h a p u r o r M a r w a m o n g t h e m o s t important Saljuq cities (p. 4); misses the panj nawbāt (privilege to have a fanfare played five times a day) when dealing with a text obviously referring to it (p. 437); improperly vocalizes a well-known Turkish name;31 assumes that the position of shiḥna could be given to non-Turkish emirs (p. 464);32 or confuses the functions Al-ʿUṣūr al-Wusṭā 26 (2018) Vanessa Van Renterghem’s Les élites bagdadiennes au temps des Seldjoukides • 268 of shiḥna and ʿamīd.33 I also spotted a fairly significant number of factual errors.34 But enough with the minutia. A broader issue is that the general conclusions are frustrating. Vague statements are quite typical: “les collections de notables de l a B a g d a d s e l d j o u k i d e f o r m a i e n t d e s ensembles mouvants, traversés par des lignes de fracture et composés d’individus hétérogènes en terme de statut social, de richesses matérielle ou de prestige” (p. 505). This generality may be due to the fact that Van Renterghem never clearly articulates the main questions guiding her investigation. By contrast, it was easy to get the thesis of Ephrat’s work (“madrasas were not that important after all”), and Ahmed was right to note that despite all its shortcomings, it was a book that “makes people think, and think hard.” One problem with the present book is that, after reading over 500 pages, I have found a lot of data and am convinced that these data are as exhaustive as possible—even if not entirely new—but I have not found a clear thesis. One option could have been to clearly focus on what happened in Baghdad during the Saljuq period. Makdisi and even Ephrat tried to answer this vexing 33. On p. 460, Van Renterghem apparently does not see that the shiḥna commanded the garrison, while the ʿamīd was in charge of what can be called non-military affairs. This is evidenced by the latter’s actions: seizing the iqṭāʿ of the caliph, abolishing the unlawful taxes (mukūs), presiding over the maẓālim court, initiating building projects. 34. For example, on p. 4, Malik-Shāh’s death (492/1085) did not mean the end of the Great Saljuq period (which occurred 60 years later, with the death of Sanjar in 552/1157; some scholars even speak of the Great Saljuqs until the demise of the dynasty in Iran in 590/1194). On p. 30, it should be noted that Bundārī did not write a history of the Saljuqs, he merely abridged one. On p. 440, there is confusion between Muḥammad b. Malik-Shāh and his son Maḥmūd b. Muḥammad. On p. 455, it should be clarified that the “Ibn Qāwurd” mentioned by Ibn al-Jawzī is not “the son of” Qāwurd, but his descendant. On p. 485, Toghrïl Beg did not build the masjid-i jāmʿi al-sulṭān in Baghdad—it was actually Malik-Shāh. On p. 488, the vizier of Malik-Shāh in 480/1087 was not Tāj al-Mulk but Niẓām al-Mulk (Tāj al-Mulk was not promoted to the vizierate before Niẓām al-Mulk’s murder in 485/1092). 35. Jean-Michel Mouton, Damas et sa principauté sous les Saldjoukides et les Bourides, 468–549/1076–1154 (Cairo: IFAO, 1994). issue. This is obviously not a question for me to answer, but what struck me while I was reading the book was the importance of the Iranian presence in Baghdad. The database is filled with Iranian nisbas. The function of qāḍī al-quḍāt was entrusted t o t h e D ā m g h ā n i s ; t h e p r e e m i n e n t figures of Baghdadian Sufism were all Iranian (Zawzanī, Mayhanī, Suhravardī and Nīshāpūrī); a great proportion of mudarrisūn (23 out 70) were connected to Iran; most of the ʿamīds (Nihāvandī, Iṣfahānī, Dihistānī) were Iranian (and naturally all the Saljuq secretaries as well). This was to be expected: Jean-Michel Mouton has shown a similar pattern for Saljuq and Burid Damascus,35 but it is naturally more conspicuous here. Indeed, the importance gained by Iranians among the local elite would have deserved study for its own sake. It is only at the very end of the book that we read: L’intégration de Bagdad dans un ensemble oriental de tradition turco- iranienne consolida les liens de la ville et de la société locale avec une sphère culturelle non arabophone porteuse d’héritage propres et de penchants idéologiques se distinguant, par Al-ʿUṣūr al-Wusṭā 26 (2018) 269 • DaviD DuranD-GuéDy certains aspects de ceux prévalant dans les régions centrales de l’ancien empire abbaside (p. 515). In other words, Saljuq Baghdad was connected first and foremost to Western Iran. Here, it seems to me that the most important sources on Saljuq Baghdad encapsulate a Ḥanbalī-Abbasid point of view that, to some extent, was embraced by Van Renterghem. Following Bulliet, I would argue that adopting the view from the edge is always fruitful, and so looking at Baghdad from Iran would have much to offer. In her introduction, Van Renterghem notes that the scholarship on urban studies deals primarily with the Egyptian and Syrian cases (pp. 6-7); and yet in her conclusion she does not refer to Cairo (Lapidus) or Damascus (Chamberlain, Gilbert), but focuses on Nishapur (Bulliet) and Isfahan (myself) (while Jean Aubin is notably absent from her discussion). The above criticisms should not obscure the fact that Van Renterghem has done a tremendous service to the scholarly community. Her monumental book is a product that perhaps only publishing houses of state-funded French research institutes can publish. The author should be thanked for having provided fellow scholars with an indispensable tool to n a v i g a t e t h e c o m p l e x w a t e r s o f t h e Abbasid capital in Saljuq times. Given the importance of Bagdad in the Muslim world, any specialist of the pre-Mongol period will benefit from her work, notably thanks to the valuable indices and tables, covering a wide array of subjects. The formidable quantity of data carefully amassed throughout the book will foster future research on a variety of topics. For these reasons, Van Renterghem’s volume should be kept close at hand, and will likely become a standard reference work. This is all the easier now that the book is freely available online (https://books. o p e n e d i t i o n . o r g / i f p o / 9 1 7 2 ? l a n g = e n , accessed 23 Sept 2018). https://books.openedition.org/ifpo/9172?lang=en https://books.openedition.org/ifpo/9172?lang=en