The State and the Military : The Samanid Case

Papers on Inner Asia No. 26 Jiirgen Paul Hamburg University THE STATE AND THE MILITARY: , THE SAMANID CASE INDIANA UNIVERSITY RESEARCH INSTITUTE FOR INNER ASIAN STUDIES Bloomington, Indiana 1994 Indiana University Research Institute for Inner Asian Studies PAPERS ON INNER ASIA Yuri Bregel, Editor Robert Dankoff, Devin DeWeese, Peter Golden, Gyorgy Kara, Nazif Shah rani , Elliot Sperling Consulting Editors No. 26 (Subseries: Central Asia) Papers on Inner Asia No. 26 Jiirgeo Paul Hamburg University THE STATE AND THE MILITARY: TIlE SAMANID CASE -- Bloomington, Indiana 1994 The series ｐ｡ｰｾｲｳ＠ on ｉｮｾｲ＠ Asia consists of scholarly papers of variable size (from 20 to 100 pages), each published and bound separately. loner Asia is defined as the region that includes Islamic Central Asia (the areas sometimes called Western, Eastern, and Afghan Turkestan), Mongolia, Manchuria, and Tibet. The papers deal with various topics related to this vast region: history, philology, linguistics, anthropology, archeology, economics, contemporary problems, and so on. Works on certain subjects that transoeDd the boundaries of Inner Asia in its strict sense, but are relevant for the study of its languages, history, and culture, are also included, for instance, the ancient and medieval history etc. of the peoples of the East European steppes, or the modem history of Afghanistan. The series is divided into the following subseries: (1) General; (2) Ancient Inner Asia; (3) Central Asia; (4) Mongolia, Manchuria, and Tibet; (5) Altaic linguistics. The series ｐ｡ｰｾｲｳ＠ on ｬｾｲ＠ Asia is designed to ensure prompt publication of scholarly papers and to facilitate the pUblication of longer papers, which are large enough not to be accepted by most scholarly journals. As a rule, no stylistic editing is done, with possible exceptions for papers written in a language other than the author's native tongue. After the manuscript submitted for publication is reviewed by the editors, it is returned to the author whose responsibility is to prepare a camera-ready copy of the text in accordance with technical requirements set up by the Research Institute for Inner Asian Studies, which undertakes to publish the paper within one month after the receipt of the camera-ready copy. All correspondence related to the ｐ｡ｰｾｲｳ＠ on ｉｮｾｲ＠ Asia (requests for the copies of the technical requirements, proposals of papers for publication, submission of manuscripts, orders for single on ｬＢＬｾｲ＠ Asia, papers and standing orders) should be addressed to TIle Editor, ｐ｡ｰｾｲｳ＠ Research Institute for Inner Asian Studies, Indiana University, Goodbody Hall 343, Bloomington, Indiana 47405, U.S.A. The price of individual copies is $ 3.50 for papers up to 50 pages in length and $ 4.50 for papers of more than 50 pages. For overseas customers the rates are: surface mail - $ 3.95 for papers up to 50 pages and $ 5.50 for papers over 50 pages; air mail - $ 4.95 and $ 7.55, respectively. (C) Research Institute for Inner Asian Studies 1994 ISSN 0893-1860 States are, Weber's defmition runs, whuman associations that succ:essfully claim the monopoly of legitimate use of physical force within a given territoryw. This seems to mean that where military activities, let alone military institutions, are not entirely controlled by the state, this latter institution is either very weak or entirely absent. It can be argued that Weber's is too narrow a definition. I Above all, it has been stated that in this definItion -every single key word begs the historical question of when, at what particular date, the 'state' can be said to have emerged. The degree of success achieved by a government in claiming the legitimate use of physical force; the completeness of its monopoly of this, the assurance of its legitimacy, and the extent of its territorial jurisdiction - every one of these fluctuated during the last millenium, and every one admitted and still admits of degrees- .2 If this is true even for Western Europe, the area where modern states first originated, the questions asked by Finer are all the more urgent in a Middle Eastern context. Middle Eastern history has more often than not been written as a history of states (or, more precisely, of sue«;eeiing dynasties), and it is only recently that historians are beginning to find this approach unsatisfactory. Nevertheless, it seems that until now, the state has more or less been taken for granted: questions of the kind cited above have not yet been asked in a Middle Eastern context. It is furthennore all too frequently repeated that Islam is not only a religion, but also a political system; Mul)ammad founded a political community as well as a religious one. Even if this is true (we cannot be altogether sure), it does not mean that the political community of early Islam was what we would call a state or else had to evolve into such an institution; moreover, it does not mean that such states as existed I Other definitions of the State are provided in CJaesseo, 'the &rly ｓｴ｡ｊｾＬ＠ Claes&eD, Early StIlU Dyn.amia. 2 Finer, 85. Italics are Finer's. and further diacu. . . in 2 throughout Muslim history have conformed to the traditions of the early community; and last but not least, we do not have to suppose that statehood, once emerged, will continue to dominate historical processes. Not only may individual states decay and eventually disappear, and even sink into oblivion, but also the institution of the state, statehood in itself, may lapse. 3 Along with this, we have to be aware that we tend to take the state for granted because we are so much affected by it in our own lives. "Things being so we are hardly able to envisage a state entirely different from what we are accustomed to. The Muslim state, like most others in the past, was concerned mainly with supplying two precious commodities: security and justice. Not more. The welfare of people - providing them with food, clothing, shelter, medical care, education, and other needs of life - was the concern of God [ .•. ]".4 Thus, even in this reduced set of tasks, "security", internal and external, the police and the military, rests with the state, and Goitein offers numerous examples and cases as proof for his assertion. It will be argued in this paper, however, that Egypt (as analyzed by Goitein and others) was not a paradigm for the rest of the Muslim world, and that at least some degree of qualification has to be added to the statement quoted above. S The question is, to what degree did the state in a given region of the Muslim world supply ｳ･｣ｵｲｩｴｹｾ＠ Historians and sociologists concerned mainly with the history of Western Europe have developed a distinction between direct and indirect modes of state government.7 Such concepts have been shown to yield important insights into the , The breakup of tho state in itself will DOt be dealt with in this paper. 4 Goitein. ll:404. j Egypt may have beea exceptional in that it almost Devet departed from its reL.tively ceatnIizI:d modo of 8dministration. This in tum is at least partly explained by naturaJ conditiOllll. • The paper will not toucIa upon tho question to what degree the state was CODCOI'IIed willa IIdjudicatioa. In this respect, we OUJht to keep in mind that le,islatioa was not amonc the tub of the rulet, wbetber'in EI)'Pt or elaewbeIe. Qlldjs, even when appointed by tho rulet, may have felt obIipcl to the law and its religious foundatiOllll rather tbu to the state. 1 For a pnenI discussioa of the development of statehood in Western Europe cf. Tilly. For direct and indirect modes of covernment: Chapter 4 ·States and their Citizeas-. AccordinC to Tilly, direct rule was achieved only after the French Revolution. 3 , structure of societies like China under Mao Zedong.' Indirect rule means that the state, even if its power is paramount in a given society, is not able to exercise this power directly, through a closely controlled bureaucratic apparatus, but bas to rely on a stratum of intermediaries. 9 Indirect rule concerned most (or even all) sectors of governmental activity, most notably taxation and warfare. Examples for indirect levy and command of armies are the feudal host and early modern mercenaries. Knights bringing their own arms, horses and equipment as well as their own followers can only with some difficulty be compared to modem officers who, after all, do not own the tanks and airplanes they use. Men:eDaries were brought to the field by military entrepreneurs who were quite willing to serve the highest bidder. Direct rule, in a military context, implies standing armies composed of natives (volunteers or conscripts). These appeared in Western Europe on a large scale only in the 18th century.tO This should remind us that in a Muslim context as well, we should expect indirect forms of military organization instead of direct ones. Rulers, army commanders, warlords and state-builders alike have to consider three questions in choosing, more or less consciously, an army: They have to decide whether the army they are going to have will serve its purpose (effectiveness), whether the state, with the means of extraction it commands, will be able to meet the costs occasioned by the army (expense), and whether the army will continue to obey its commanders and the commanders the ruler (loyalty).lt These questions make themselves felt even if the responsible leaders do not take them into account. Military disaster, bankruptcy of the state or undesired military intervention into the affairs of the state are possible outcomes if they are not properly considered, so much so that rulers may be expected to accept considerable efforts in trying to have the army most suited to their situation, that is, the most effective and most loyal one they can possibly afford. This paper proposes a study of the following questions: To what extent were military institutions controlled by the state in a given Muslim society? The examples will be chosen from the Eastern Muslim world, Khurasan and Transoxiana in the 9th and 10th century A.D. If there was indirect control, who, then, were the • Tilly ; Sbue. 9 HOUI"8Di, in his ...,,;ul article OIl the -Politics of Notables-. can be cited UDODI the lint and foremost to have used coacepta of intermediaries in the study of Muslim IOCiaI bistoty. 10 For this, II Cf. FiBer. 94. BCe FiBer's study. 4 intermediaries? If there was direct control, how was it ensured'? What consequences did considerations of effectiveness, loyalty, and expense have on the evolution of military institutions over time? And finally, what can we say about the way military institutions and the state were connected? How did the role of the state in the societies in question evolve? ••• Before plunging into a study of the available source material, a note on slave soldiery seems in order. It has been stated that slave soldiers are the solution to problems of effectiveness, loyalty and expense in military matters that has prevailed over most of Muslim-governed territory from the early 9th century to modern times, the institution of slave soldiery having existed up to its abolition in Egypt by Napoleon and Mul;lammad . Ali. 11 Developing a view most conspicuously taken by Ayalon, Pipes has explained the importance of slave soldiery in Muslim societies by a feature characterizing, in his opinion, the way Muslims behave in public affairs. According to Pipes, Islam, by setting unattainable standards as an ideal of public life, has discounlged and even deterred participation from a very early date on.13 Armies were staffed by slaves since Muslims had retired from military activities to a very high degree. Crone, the other author who has written extensively on slave soldiery, shares Pipes' views as far as the involvement of Muslims in public affairs is concerned. 14 12 AyalOD, -L'esclavqe du Mamelouk-, i. I' Pipes. The mo&t fi ......mental critique of this book seems 10 be the article by Beckwith wbida formally is DOt a review. There is something of an Ihistoric:allouch about the writings of Pipes - . 10 a les&cr degree, Ayaloo. in that they teod 10 live the impressiOD that the iDstitutiOD of slave soIdiory. afta- baving omeraod in tho be£inning of the 9th century A.D., remained substantially unchanged for oearly a miUeD.ium or so. - The specificity of the Mamliiks of Egypt is stressed by Garcin. 14 Crone's book bas met with substantial criticism. out of which the review by Dooaec is maybe ooIy the mo&t exteosive, but also most 10 the point. 00 Pipes and Crone. see also Garcia', saudy OD the Mamliiks of EI)'Pt. DOt 5 Thus, scholarship since the writings of Ayalon has tended more and more to look on military sJavery as not only very common indeed throughout Muslim history, but also as induced by certain features characterizing Islam as a rdigion. 'lbe question emerging from this strain of thought has been placed by Shoshan in a context including military affairs. The question is not, however, restricted to urbanites: 15 -[W]hy is [it] that despite the uninterrupted existence of urban life in the House of Islam, town dwellers were not entitled nor were they able to claim the right to handle their own finances and taxation, to supervise public works, to decide about matters such as fortifications and food provisions, to control weights and measures in the markets and, above all, to make war and conclude peace. _16 The argument in this paper is that it is far too early to ask such a question. It will be contended that it has not been proved that Muslims, town dwellers and even rural people, were not, at least at times, in some places and to a certain extent, able, and sometimes even entitled, to look after their affairs (including problems of internal and external security). The stress laid on military slavery tends to obfuscate the degree to which free Muslims wielded weapons. The question therefore is not if there is a causal tie between slave soldiery as a military institution and Islam as a religion (Pipes' question), but: To which considerations of efficiency, loyalty and expense did slave soldiery respond in a given historical setting? ••• State institutions are characterized in a way by the nature of their military base. The development of the military institutions in a given society may serve as an indicator for the development of the relationship between the state and the society it is trying to control. .., Cook. in bia comment on Garcin's paper, puts it like this: -[I]t is remarkably bard to fiDd in laIamic history instaDces of what misht be called citizen armies - armies locally recruited. by • lUte ideotified with the area in question, from a settled population that wu DOt tribal-. He aoes on to mention the $affirids as one possible exception. (133). - The point is that Cook, even here, aumot thiok of armies recnUted by 8Omebody else as a state, however' ideotified. Tbua, be ....ects aU forma of -private armies- (e.,., tboae reliaiously motivated) and local efforts to repel agresaon. Even IIpUt from this. exceptions miebt DOt be as rare as Cook suggests. II Shosbao, 210. 6 If the state, in order to build military might, has to rely upon active participation of non-statal groups, it will most probably have to look for a legitimizing rationale: it has to give reasons for participation in military activities that are liable to convince a satisfactory number of volunteers and to ensure sufficient motivation. The question is, which forms of legitimation are available in a given context, and who are their social carriers? Is it a state apparatus which is in charge of mobilizing the volunteers, or are groups intervening which the state does not control? If the legitimation offered is strong enough, problems of loyalty may tend to lessen; loyalty, however, is not to the state as such and not even to the ruler, but to the legitimizing purpose and eventually to the persons embodying this purpose (leaders of volunteer troops or specialists for legitimation as e.g. religious leaden, see below). Efficiency may be another problem (volunteers may lack serious training not always compensated for by high motivation), though if military skills are well diffused throughout society and not centered in one or just a few of its segments while lacking in others, the efficiency of volunteer troops may well be high enough. Expenses may be comparatively low if volunteers provide their own equipment. If the state bases itself on the military support of a given sector of the society, e.g., nomads, the remaining sectors, e.g., the sedentary-agrarian and the urban sectors, will have to serve as a resource from which the necessary funds are extnlcted in order to meet military expenses. The need for legitimation is still there; loyalty is once again not to the state as an institution, but to an individual ruler, in some cases to a dynasty. There tends to be a tiered process of transmission of loyalty: rank-andfile warriors are loyal to their immediate leaders and so forth up to the ruler. 17 But . the need for legitimation does not seem to be as important as in the first case. . Efficiency is ensured again not by systematic training, but by diffusion of military skills; as is known, nomads bring up their children (boys in the first place, but sometimes girls also) as riders, hunters and fighters. Expenses may vary due to the warriors' claims on the ruler. Possibilities of extraction include not only the nonmilitary sector of society, but also external sources of revenue (booty, tribute). If the state tries to build up a military basis entirely of its own, that is, to avoid the problems of loyalty arising in the first two cases, society as a whole is regarded as a source for extraction of resources. In this case, expenses tend to be significantly higher. By the same token, the demands on state apparatuses rise. Apart from extraction, the state has to ensure regular and full payment of stipends; there has to be a military bureaucracy responsible for recruitment, training and control of 11 It seems that tho loyalty of nomadic warriors to a ruler CIIIl take 00 two basic f0nD8: trit.! loyalty which implies tho tiered system sugested in tho text, and loyalty to a charismatic leader' wbicb may tnosccad previously oxistiq tribal bonds. Amon, the best studies of these pbeaomeaa in a Central AsimlIranian context are Manz and Woods. 7 the army. Depending on the degree of monetization of the economy at large, state apparatuses charged with catering for the needs of a professional standing army (this is what we are taUdng about) take on a buIlder or leaner fOrmal. II Professional standing armies in a Muslim context have typically been staffed and led by -slaves-. In these cases, loyalty is again not to the state as an abstract principle, but to an individual, mostly a ruler. 19 The question is: What happens if the state apparatuses cannot meet the army's requirements? What are the options open to military leaders in this case? Thus, the degree to which the state makes itself felt in a given society depends, among other variables, on the format of the military institutions. Statehood can be expected to penetrate society most deeply in the case of professional standing armies (-slave armies- in the cultural context of this study). With nomadic warriors, intermediaries may be prominent, as would be the case with tribal chiefs. Non-statal forms of organization ought to be most evident in the case of volunteer troops; intermediaries may be those ensuring legitimation as well as small-scale military leaders. On the other hand, communities and segments of a given society will react according to demands as engendered by varying military format of the statecontrolled army. In the case of volunteer troops, forms of cooperation between local leaders (legitimizing and/or military groups) and the ruler and his governmental apparatuses will emerge. This cooperation is limited by the purposes given for military action. In the case of troops made up of nomadic warriors, the non-military segments of the society, the sedentary-agrarian and the urban sectors, are likely to look for ways of reducing the tax burden or to escape taxation altogether. In the case of ·slave· armies, nomads as well will strive to avoid being used as resources.20 All three forms of military organization were present in Eastern Iran and Transoxania in the period from ca. 820 to 1220 A.D. But even the most cursory perusal of the sources is enough to show that change ｯ｣ｵｾ＠ not in a steady ·upward- evolution to ever more developed forms of statehood, but that the overall II Cf. the results of Finer, Tilly, Rokbn and others. In a European context, the pantdiamatic cue of a bulky state apparatus of extraction designed for taxation of a largely non-mooetized agriculture is Russia, the other extreme beiDg the Italian city-state or city-empire (Venice and Genoa). (Rokkan). l' Cf. Mottabedeh. lD Taxpayers may. apart from avoiding taxes, also try to wiD control over the state and. by the same over the military iD order to easure that it is used acconliDg to their pen:eived iDten8ta. A classical case of such a strategy is coadea&ed iD the slogan -No taxation without represeotMioa· • toIe_. 8 trend to nomadbation is visible in the military sector as well. 21 Apart from that, all three forms coexisted most of the time; predominance of a given form never reached exclusiveness. *** The Samanids, the dynasty which ruled at first over Transoxiana and parts of Khurasan, later over most of the area in question in the 3rd/9th and 4th/10th century, are sometimes said to have made particularly intensive use of slave soldien. 22 MOROver, slave soldiery has clearly a Central Asian component and it may have originated out of pre-Islamic Central Asian institutions, whether Turkic or not. Military activities in Eastern Iran and Transoxania under Samanid rule are therefore the focus of this section. The Samanids began their career in 204/819 as governors over different parts of Khurasan and Transoxiana for the at least semi-independent rahirids who had their military and administrative base in Nishapur. rahirid overlordship weakened in the course of time and was insignificant long before the dynasty's downfall in 873.23 lust a little later, in 2611874-5, the best-known Samanid, Isma 'n b. Al}mad, was sent to Bukhara by his brother ｎｾｲＬ＠ who as the head of the family ruled in Samarqand. Bukharan ulama had requested a Samanid governor. With Ismi'n, Samanid rule began to spread over Khurasan and other parts of the Eastern Iranian world. Clannish principles, however, were still prominent in the ways of government. . In the decades following Isma'n (r. as Samanid overlord 279 - 295/892 - 907), the dynasty is said to have reached the peak of its power. ｾ＠ Following the death of N3¥ b. AI}mad b. Isma 'n in 3311943, a long period of -decay- lead to the dynasty's overthrow by the Qarakhanid Turks at the end of the 10th century A.D. Military power at the beginning of the Samanid period was in the hands of local people led by their -gentry-, the dahlJqrn. This stratum was precisely the background the Samanid family itself came from. The dahilqrn's power can be dated 21 Cf. Bregel. 22 AyalOll, -Preliminary Remarks-, 53; and idem, -Aspects-, 207. 23 Spuler, 77. :at In EI I s. v. (BOdmer); see also: Frye. BuIcIuJra. 17w Medieval ａｾｮｉＮ＠ his traDsIatioa of Narsbakbi and his chapter in CHI IV. Frye's writings are at the time being the best available studies 011 the Samanids. 9 back to pre-Islamic times, and the leading families of that period continued to play an important role during most of the first two centuries of Islam in Transoxiana. Z5 After their conversion to the new religion, the descendants of local rulen made at times brilliant careers in the Abbasid army.» The Samanid army, at least down to the times of Ismi 'n, and to a lesser degree all along, was in part a volunteer army led by the dIlhIJqt'n gentry. It was such an army that won the victory over 'Amr b. al-Laith the ｾ､＠ (at Balkh in 287/9001'. An echo of this state of affairs is found in the works of the Arab geographers of the 10th century. MuqaddaS tells us that there are 16000 settlements in the region of Cbagbiniyin, and numerous warriors take to the field with their own equipment and animals if there is an attack states the same procedure for the whole of Transoxiana: from the outside. 21 ｾｉ｡ｫｨｲｩ＠ from every village, at least one horseman and one soldier on foot will go to war.29 The Samanid army, the geographers say, is made up of Turkic slaves and free men and dIlhIJqt'n ·whose region aJ\d place, whose family and neighbours are known·. 30 It is only reasonable to suggest that it was not a military apparatus, but the dIlhIlqtn themselves who recruited the volunteers they led afield. In order to draft ·one man per village- via a governmental bureaucracy, the state has to have a command of the 25 This is best shown in the case of Bukbara. since we ue lucky enougb to have NarsbUbr. local history. A delegation to the JOvernor in Marw asking to have the Outer Wall arouod the Bukbann oasis rebuilt wu led by YaDd b. Obiirak, descendant of • prominent family from pre-Islamic timee (edition R.i4awi. 40, traosIation Frye, 33). Similar initiatives were fakco by local people coocemin, the IDnea' Wall (41:34). the mintin, of coins (edition Scbefec, 35), uteGsiOD . . . rebuildin, of the Friday moeque (edition Scbefec, 48. mel edition Ri4awi. 59. tnnslation Frye, SO). The overaU pictwe coocemiDa Bukbara is OM of inteaso cooperation between the Arab aovemor bued ill Maw, the ladiDa families from pre-Islamic times mel the DeW Muslim 6lite. the qadi ill the tint place (the qadi is shown supervisilla the buildin, of the Outer Wall. fiahtin, the -false prophet- Muqaaaa'. fixiDa rules for the distributioa of water in the irrigation system and 80 on - edition Schefer. 4. 33.67). » The ruler of Usriisbaua, called the Afsbin, is only the most DOled of these. Other eump_ III'e the BukbirIdwdit who equally served in the Abbasid army. The backlfOUDd of the IIcbsh'id, fouDder' of. (shortlived) dyuasty in BJYPl, is that he claimed descent from the priDcely house of Ferp.na. Cf. ｾ＠ 292: The com....1MIenJ of the ca1ipbaI army ue people from FeraJuma mel Turb. He sepuately aameI (amooa others) the A&bin mel the dl Abr '1-SlJj family from Usn-.... the Ikbshid from Samarqand, various Sosdian J1lUldees mel the Bukbirkhudit. The same auIbor . . . . little euliet Ibat the Tutb make up the calipbal army, mel that dDhIJqliJ from TnIIISOxiaoa III'e their leaden: 291. See also Ayaloo: -Preliminary Remarks-, 51 - 54. %7 Narsbakhi, ed. Ric;lawi lOS, translation Frye 89. The popular character of this army is strused by Tahm: m:2194, mel equally in the TtJrtkh-i Sfstan. 2S5. » Muqadd,si, 283. ﾻｾＬ＠ 291. 3D ｾＬ＠ 291; b. I;Iawqal, 471. 10 society the Samanids certainly did not have; otherwise, conscription of this Jdnd can be ensured through a stratum of intermediaries who control their constituencies closely enough to know who might be willing to go or else whom they might coax or coerce into going. Thus, as far as military affairs are concerned, there was no direct rule, not even in the central lands controlled by the Samanids. The first stratum of intermediaries to emerge clearly out of the source material are the gentry, the dah/Iqrn.3' There is a second ODe, religious dignitaries and leaden of religiously legitimated (and at least sometimes, motivated) fighten. But we will stay a little while with the gentry and the Samanid rulers' dealings with them. -- *** Since volunteers are so prominent in the early Samanid army, we will have to look for a legitimizing rationale with sufficient mobilizing power. Two possibilities come to mind, one material, the other religious. To deal with the material aspect first: The first Samanids (largely speaking, up to Isma 'n), being little more than dah/Iqrn themselves, shared their power with this stratum. The second aspect, which evidendy is the complex of jihiJd, fighting the infidels, will be dealt with later on. The pre-Islamic regional structure of power showed a more or less nominal . overlord (the Ikhshid of Samarkand) and a great many local leaders (in Sogdian: ). n This situation continued with the addition of some more levels of (more or . less nominal) overlordship, the caliph and the Arab governors in Marw, later the Nishapur-based rahirids. But regional dynasties still held considerable power; the tides of a number of such princelings have been transmitted in various sources." These petty rulers were prominent above all in peripheral areas. But they (or descendants of such families) were important in the central regions as well. This is best shown by the stories relating how Ismi 'n got rid of them. After initial triumphs (against outside enemies and so-called bandits, that is, armed groups of popular origin) Isma 'n thinks about the situation and comes to ｾ＠ conclusion that -the nobles of Bukhara did not show him sufficient honor. There was no respect in their '1 Archeological surveys of Transoxiana show • landscape literally strewn with castles. D See e.g •• Frye: -Tarxun - TiitxUn- which offers refereoces to the older Soviet litaaluro. More receut archeological evideaco is discussed in Raspopova. A discussion of rulers in SoJdiua cf. Livshits. 11 eyes and their gathering together did not portend good for him-.34 The nobles did not think that Ismi 'n was in any practical way their superior (he was not even the head of the Samanid family at that time), so he did not have hoybat for them; this expression means the -authority- coupled with kingship. The stratagem used by Isma 'n is designed to take their power (and some of their wealth) while allowing them to keep - at least in apparence - their elevated status: he wants them to become a nobksse de coW'. Before that and even for a while after Ismi 'n, though, the gentry seem to have cooperated, serving as officers in the Samanid army, above all the -volunteer- part of it. The Samanid army seems to have comprised an element of mass mobilization down to the fall of the dynasty, even if these troops were reduced to an auxiliary status.'5 The contemporary $afflrids (in Sistln) also used a kind of peasant levy.36 Even if they were not always of great military value, they may in some cases have contributed in attrition warfare. Sometimes, volunteers are seen to be bard to control, since they are all too keen on booty.n That this should have been so does not come as a surprise. The geographers as well as other sources frequently mention that people in Khurasan, Transoxiana and Sistin are well trained in the use of arms." In the story of an walleged embassy from the Emperor of China to the Amir Na,r b. ,.. In Frye's traDsIatioa (82). in the oriJiDal: dar 1J4l-; I'IUdJc fa 'amnud Iuud wa ""' 'lilm Iuud lei li- nt ba mlhtariJn-1 BIlkJa4rd cIumtl4n IJMrnuztf VyiId4J n£st wa ba-clta.rhm·i rsluJn IuzybaIt n£st wa til jam' shudan-i rshlm malffa 'atf bel-way r4ji' na-ldawtlhad shud. Schefer's edition 80, Ri4awi 96. The reodering of h4ybat by -respect- is maybe not precise enough. Perhaps -In their view. __ 'i1 did DOt have the aura of kinpbip- is another possibility. J5 Terms are IJashar aDd rajjtJla, for iDstaoce: 'UtbiIM8Dini, 1:307f, cf. 'UtbilJurfadbeqinl 117; 'Utbi/Maniai 1:211 aDd 1:120, ct. 'Utbi/Jurfidbaqini, 120. • ｾ＠ rtIsUI'f. 7'iJrIIdH SfSI4n. 291. See also ｍｾＮ＠ -Simcuriler-, 1:74, and Bosworth. "The Armies-, SS4. n 000 term for this kind of troops is a/JdIIJh. 1be&e seem to have bad their own commanclen <ru 'u), but the ·official- military tried to integrate them. The Samanid geoeral Abii 'Ali S"aagiir was ....... colllm&Dder-iD-chf of the auxiliary troops (siplJhsdliJr( WW'I 'antal-i ma '1InllI MId al}dlJth-i NtshilpIIr wa HanJI MIG QuhisI4n); Gardizi, 16S, ace also ｍｾｩＱＮ＠ ·Simcuriler-, m: 122 and IV:SSO. Other IDNItiorur of a/JdIIJh in this meaning: 'UtbllJurfidbaqini. 106. 120. 175; 'UtbilM8DiDi, 1:303. ] I Many meatioas in lfudIkJ al- 'alam; ace also 'nJrtkh-i SlSt"", 11, and Yiqiit, m:l9Ob. DeWlod referaacea in my -Nacluicbtea - • 12 A1Jmad W , the author mentions infantry troops that obviously are local levies (rajjiJkJ).39 -- At the end of the Samanid period, however, we learn that a group of t/QhIJqIII who felt that this dynasty had outlived itself called in the Qarakhanid Turks. 4O This was by no means unusual: calling in the Turks had a long tradition in Transoxiana and was used as a strategy in winner conflictsW ; even internal problems in the Samanid family were sometimes solved this way.4. A last effort of the Samanid ruler at Bukhara to win popular support among the urban population met with no success. He tried to bypass the dah/lqrn as military leaders and ordered religious figures, the Ichu!aba' (who deliver the Friday sermons) to organize the BukharanS. 42 At the end, thus, the Samanid rulers seem to have lost contact with the intermediaries who previously were active in mobilizing and leading volunteer troops. In the first period, the Qarakhanid Turks apparently made use of the mobilizing power of the dah/lqrn in a similar fashion, for offensive as well as defensive purposes. The nile Khan called not only the nomadic Turks, but also the dahIlqrn for a campaign directed at Kburasan;43 enemies from further East were repelled, as in the best Samanid days, by a levee en 11IIlSSe..... Thus, they seem to ai-Rabid b. aI-Zubair. 142. See also Bosworth's trans1ation in: -An allepd embaay-. serious teIIeI'Ve8 as to the reliability of this soun:e. Bosworth aeems ript eoouP . ia ""mina a depee of tnIatworthiness as far as Samanid military affain are coacemed. The IICCOUDt . fits ia with what is kDowa from other soun:e&. If Althouab there are • 'Utb'ilMaDini, 1:163. In the Persian translation (which dates from the 12th ceotuty A.D.), tho expressiOD (4 'ifa mill al-JohlJqtn is reudered by jam 'r at ma 'iJrif (92); it is well known that tho word diJIqIJIt bas como to meaD -peasant-, -tiller of the soil-. The change will have occurred during tho lUb ceotuty A.D. - See abo BartoI'd, 318; Spuler, 99. •• Examples from Nanbakhi (eel. Ri4awi): 45,55,97; also b. aI-Atb1r. vm:41S. - The teadeocy a decisive step towards cultural decay would DOt have met with approval by tho dahiIqIft of this period. Central Asia bad looa before been an uea of mixeel culture, and Turtic overlordsbip wu nothing new; OD the coatnuy, it seems that Turkic rule wu __ .. _ aIleviMioa wbea compared to the last decades of tho Samanids. to Iameot the coming of tho Turks as 42 BartoI'd, ｾＬ＠ 328f, according to HiliI aI--$&bi. The preachers said that since the Turks were just as aood Muslims as the Bukbaraas. it would be better for them not to parUcipMe in tho fighting. 4) oM 291): ·UtbilManiDi.II:82. In the year 403: 'Utb'ilMaaioi, 11:220. Only one example from Samanid times (in the yOIII' VII:S33. Tabari. ID:2249, shorter in b. aI-Athir. 13 have attained a certain social basis in the gentry. What reasons did the gentry have for defecting from the Samanid dynasty? Before turning back to this question, however, we will have a look at how the religious legitimizing rationale worked and how the Samanids were able to make use of it. ••• Transoxiana, the centtal region of Samanid rule, has been one of the most important frontier regions of Islam for a very long period. Waging war against the ·infidels· was the concern not only of rulers and their military apparatus, but also an endeavour which Muslims could voluntarily choose to participate in, particularly in order to gain the other-worldly rewards connected with it. This alone should suggest that activity in this field was not an affair controlled by the state to the exclusion of others, was not a monopoly of the ruler; on the contrary, conflicts over control of volunteer armies and legitimizing purposes seem much more likely. Relations between rulers and volunteer fighters may range from close cooperation to open . conflict. Once again, loyalty of the armed personnel to the state, the dynasty or the individual ruler is the issue at stake. At first, participating in military activities against non-Muslims was part of an ascetic lifestyle (ZMhd. t 5 There are several cases on record in biographical dictionaries (a certain type of local history) where this link is mentioned.4t Later on, however, this kind of practical ascetism was rejected by teachers of mysticism: -The true exercise of the spirit is not that you go fighting the infidels in order to get .., Tbia kind of activity is in ODe c::ase even cbaracterized as au ascetic exercise of miooI' difliculty, easier ...... speading the oiJht stanclina up (qiyIIm 1Il-lllil): BaIkhi. 84. In Sam'iai. we .... of lIOIDebody who wu DOled for' his ucetism (which eamed him his oisba ＱｉｬＭｾ［＠ be weal fOr die I]oJJ oace ia tea yean and to gIulzw oace in three yeas (Sam 'w, VI:239) • .. Soun:es of this typo iDcJude the Arabic Qandlyya; see my -Histories of Samarqaad-, e.peciaUy p. 82; see abo (one of the two extant fragments ooIy) Dodkhudoeva. See aIao the euIi. study by J. Weinbeqer. I ovedooked this uticle during my work on the Arabic Qandiyya. Weinbeqer's reaulta are very clOIIO to my own, as far as the history of the Arabic work itself is coocemed. - Furtber 1IOUI'Ce8: Sabmi. 394. - FIJIj4 'II-I Balkh: 129. lSI. - For the HLrtorla of Nishapur, . . Cbabbi. -Iba Funduq (p. 220) meotioas that a family descending from the Companioa of the Prophet, Abu Obarr, who was DOled for his &SCdism. was regularly involved in ghllzt fightina down to a mao who died ill 401 A.H. 14 ldIled-•Q Some fighters do indeed seem to have sought death during their ghllzw activities. <II But at least down to the 4th century A.H., we have to presume that for many men, participating in such activities was only natural once you took your religious and other duties seriously enough. These men did not necessarily fight out of political considerations; they did not want victory so much as the other-worldly rewards mentioned above. This further suggests that their cooperation with professional soldiers and officers cannot always have been a smooth affair. From the 3rd century A.H. on, fighting the infidels took on a more organized and professional cbaract.er. (The Samanids acted as governors in Transoxiana and parts of Khurasan from 204 A.H. on.) Comparatively large groups of fighten set out.'" The question in our context is: To what degree was there cooperation between the Samanid rulers and the fighting groups? Did they have their own leaders and who were they? What do the sources tell us about military organization? There are some people who appear to be leaders of ghllzltroops. One of them is shown in an official mission, although not a military one; he is styled -commander of the Bukharan ghIJzB-, and is sent as an envoy to the Simjurid amir in Khurasan. 50 This might mean that his position is an ·official· one, that is, the ruler apparently is able to give him orders. But with other men in a similar position, this cannot be surmised. In the extant fragments of the biographical dictionary for Samarqand, the Arabic Qandiyya, several names of men are handed down who are called -leader of ghllz.r troops·. 51 At least their training, whether practical or ｾ＠ ｾｳ＠ COIlUDeOWy 00 the statement of an ascetic (d. 303 A.H.) who thouJht that COIIIpUOCl to the gMzrs way, aU the mystical eodeavours of the sufis were nothing but idle talk. Tabaqill alｾｦｖｩｹ｡Ｌ＠ 219. bodhl ｡ｬＭｾ＠ IJtJ /III biJshmJ lei btI-ghazA bi-rawc t4 '"ora b"-IaISIumd. • Paul, "The Histories of Samarqaod-, 82. • Some 200 meo were tilled durina. raid to Sbivdir (Arabic Qandiyyd. IDS Turbm Valide 70, toI. S4b). This is quite Dar Samarqand. Christians settled there in early Islamic times: ｾ＠ 321. b. J.IawqaI 498. Fnatz GreDel (Paris) told me (penoaal QOIDDUJDication) that durina arcbeoloaical surveymar. Christian eraffiUi have been found 00 some rocks there. If the ghIJzI raid in questioo was directed apinst these Cluistiaoa. legitimation is questiooable if they were UDder protectioo (dJUmma). » GualZi, 165. dated 365. The man is called Abii •Abdallib b. 1;Iaf" stJl4r-i gh4:Vyilll-l Bllkh4ra. 51 Ms Paris. BN arabe 6284. fal. 13b. S9a; DIS Turban Valide 70, fal. 16b, 23.. 16S•• 187•• 18Sa. Detailed information: Paul. "The histories of Samarqaod-, 83. 15 -theoretical-, was organized in the medieval Muslim way, as private schooling.52 This would suggest that the rulers did not control the training available to volunteers and their leaders. A late mention of a commander of ghilzlS is in b. Funduq's History of Bay/rQq: this man is in charge of the defense of the region during an attack in 378 A.H. Another person is said to have led volunteers to Tarsus (on the Byzantine frontier) regularly. No date is provided in this latter case, but no goverDmeat controlling Bayhaq/Sabzavir in the period from the 3rd to the 5th century A.H. can have been interested in such activities. 53 Thus, leaders of ghiJzI troops enjoyed varying, but mosdy significant degrees of independence; they felt responsible to the groups of fighters rather than to the dynasty. Their loyalty was to their community and to the purpose of fighting the infidels, not to the state, not even to an individual ruler. 54 Prominent cases of conflict demonstrate this state of affairs very clearly. In the middle of the 4th century A.H., the Byzantines regained the military initiative in the frontier region in Eastern Anatolia. Quite in keeping with the gh4d traditions of Kburasan, large groups of fighters set out from there. No mention is made of any·. Samanid sponsoring. In the second case (where indeed Samanid instigation, but no direct material support is suspected), the Khurasani volunteers are said to have numbered 20 000. Even if this number evidently is subject to serious reservations, they must have been very numerous indeed. From the start, they Wel'C unruly, and their leaders were either unable or unwilling to prevent disorders. They came to Rayy Sam 'w Vl:49. quoting a (otherwise lost) passage of the Arabic Qandlyya (01' Ode of its Idrisrj: Somebody caUed MlIa,.1IUDIId b. al-' Abbis aI-Gbizi ai-Rami (-the bowman-) was an authority in -the science of shooting- (al-mutabba' ft 'ulIIm al-ramy), Idrlsi himself studied aln.IIII)' with him for loa, yeara, aad it was he who gave corresponding diplomas to the gla4:r leaden in Samuq.Dd (wa-bihll6kh4rrrtJa ,." 'tU4 ' al..,hu:IJt bi-8omarqaltd). - I am aware that tJu. _ may have tauaht tnditioDs aad the like ｾＬ＠ the bow as well as shooting arrows in • more pnICticaI way. But aiDce he is described as a friead of teamed people. be may not have beeIl a 't2lim himaelf. $l 1!IOUI'I:e8. S) b. FUDduq 51, 124. It seems that Khurasani volunteers kept flocking to the Tanua m&r'IIbes from eaty AbbMid times 0DWaI'ds; a high point was reached in the middle of the 4th ceatury A.H.; cf Bosworth, C.B.: "The city of Tarsus-. 54 It seems that under the Ghamavids, serious attempts were made to gain control of the reliJiously motivated volunteers. When Sultan Mas 'WI wanted to go to war in India in per1IOIl iDstad of &ciDa the Seljuq meoace in Khurasao, his adviset'l told him that the task the sultan b8d daoaeo c:ouW very well be fuI-filled by the sill4r-i gM.:.iytJn of Labore: BU'l-Fa4I, 531. It seems that the ghdtlleadw is -' the COIDID8Dd of the military leaders of the replar troops. Commandin, the gh/JzI troopI is regarded as a govemmeot office under the Gbaznavids: Bii'I-Fa4I, 254. - There certaiDly was cIoae cooperatioo between the volunteer troops and the regular Ghazoavid army in India. Mer91 ｾ＠ 17 campaipa of ｍｾ＠ alooe: Gaz.neliler. 16 - 28. 16 which was under Biiyid rule, and their leaders demanded that the entire land tax (kharilJ) of that rqion be given to them, since it was meant exactly for the purpose they were serving, fighting the infidels and defending the d4r al-isllJm. Not surprisingly, the governor flatly refused, and they started plundering and treating the local Daylamis (Biiyid soldiers who were shiites) as if they were infidels.55 There are clearly two political principles in conflict here: The state (in this case, the Biiyid governor) insists on its right to decide on matters of peace and war, and above all, of taxation, whereas the volunteers brandish the banner of their religious legitimation. In this particular case, no cooperation and no agreement is reached. But we have to keep in mind that this is a re1atively late case, although it is quite interesting to encounter religious leaders in military functions . ••• Ribill fighting is another form in which cooperation between rulers and volunteers could be organized. In distinction to ghilzw, ribil! fighting is carried out in smaller groups from fortified strongholds, often in a givUl season of the year." The necessary buildings were sometimes provided by the ruler or other prominent persons, mostly military leaders. This took on the form of pious endowments made from the privy purse, not the state treasury (since it is uncertain whether the state treasury contains legally obtained monies only). The fighters, on the other hand, brought their own equipment and provisions. Towns and even villages all over . Transoxiana and Khurasan had their ·own· ribiJ!s in frontier regions. Thus, in the . vicinity of Paykand near Bukhara, there were ·more than a thousand ribll1;s corresponding to the number of villages of Bukhara [ ..• ]. The people of every village built a ribiJ! there and settled 55 b. 8l-Atblr. VID:SS2 (year: 3S3). S69f. (year:3SS). See also Frye in CHI IV: ISS; Bosworth, -TU'MIS-. - ODe of the Khuruaoi leaders was the noted sbifi 'jte jurist AbU Bakr a1-Qaftil a1-sbiabi who later came to be IIC)IDetbjn. lite the patron saint of Tasbkeat: Castap, SU. His numJr in Tubkeat is still standing. There are - in later aources - stories showing him as a gMtlfigbter: Jamil Qarsbi. fol. 7tb. ·Cf. Noth. 17 a group. They sent them their living expenses from the village. In the winter, when the attacks of the infidels occurred, many people from every village gathered there to attack (the infidels). Every group went to its own . ' 57 n"batOther ribiJIS linked to individual towns include: Dihistan which was manned by people from Iurjin;5I Nur, for Bukharans;59 in Usriishana, for people from Samarqand;60 on the eastern frontier t there were several strongholds meant for fighters from individual regions or cities. 61 It is quite evident from these sources that at least until the middle of the 3rd century A.H., but in certain Rgions, particularly on the eastern frontier, obviously longer, ribiJl fighting was carried out by volunteers organized on a local basis, with their own leadership and at their own expense, at most partly sponsored by endowments from leading governmental and military personS.62 Thus, the role of the ruler in ribaf fighting was rather small. ••• Volunteers in ghazw, in ribiJ{ or other forms of warfare are often called JnUIttIWWi 'a. It is not always clear whether they come together on private or governmental initiative. Sometimes, though, they clearly have their own leadership $7 Nar'IIbIIkhi. in Fryo" traosIation p. 18. ed. RicJawi. 22. Paykand is suited as a wiDta' camp for DOIIIIIda. Tbe quoted aouroo mentioas that this form of figbliDa fell into disuse in 240. The aeopapben (as quoted by Frye) aaree that there were many rib4{s in that J'elion. Sam 'w teUs .bout rum. he .... 800D thoro. only a few Turkmeos were Iivinl there: 11:404 (BaiJcandt). Some of the ruins are atiU above JIOUDCI. but of c:oune DOt 1000 of them: Gorodishdu! Paykntd. 10 and 116f. • saboi. 139. 149. , Today Nurata, to tho ｡ｯｲｴｨｾ＠ of Bukhara. Narsbakhi. edition RicJaw'i. 13, trmslatioa Frye, 12; Arabic Qandiyya. ms Paris, fol. 24b (ef my -Histories of Samarqand-. 861). • b. III J:lawqal. S04f. Muqaddasi. 273. 291): a .... C Private initiative in military matters is further mentioned b. 1;IawcpI. 468 ｾＮ＠ from Tasbk_t equipped many animals, horses and beasts of bwdeo. and the soun:es ...... that he was DO official fipre. Yiqiit tolls that a wall of defease against Turkic IIOIDIIds was built by a private penoo (I:SOOb with refereoQe to ｾＮ＠ Private initiative in IIUCh matters is dillC''SIIed by MiwardI. 236. He states that in a town where the w.l1 is in bad repair, the affluent peopIo have to see to ita I'OItontioa if the bail td-m4l (the state treasury) fails to do so. 18 -- and are at nobody's command. They base their activities on the well-known prescription to enforce the good and hinder the evil which is a guideline for the good Muslim life. 63 Examples (to cite but a few) are: The u1ama and ·volunteers· in Nishapur sided with 'Amr b. Ya'qiib who had a caliphal diploma of investiture;" one 'Abd al-RaI}min al-Muttawwi'r had gathered a group of fighters in Nisbapur without the governor's consent;65 somebody called Badil al-Kisbsbi, an acquaintance of Tabari's, was active in Isfara'in.M But in other cases, IIUI(!QWWi 'Q just means those men defending their village against an attack. 67 All these volunteers did not think they needed an authorization by the ruler to bear arms and to use them. They felt that Islam offered a legitimation for that; in this case, it was not the injunction to fight the infidels so much as the prescription mentioned above. Thus, even in a context confronting Muslims and Muslims, ·private- initative leading to some form of military activity is not at all excluded. ., al....",. hi 'l-ma 'rqf WQ 'l-nohy 'an al-mllllkDr. The coasequeocea of this pnacriptioa can budly be overestimeted. It is discuaed at Ieosth by al-Miwardi, 231. See allO Garcfa-AJeDal. - Tho meDtioned prescription was (and is) used to call for Muslim participation in public affairs. Pi.,. who tbiDb Ibat Islam as a reliJioo bas led Muslims to massive IIDd loag-Iasting absteation from tbiDp political doe&: not offer a diSCWllioa of this principle• .. GanIizi, 142. liS In tho year lOS A.H. laban, Ill:I043; b. al-Athir, VI:361: inna 'Abd al-RtU,m"" alM"If4WWi 'r janul '4 jlllll/j 'cue IcaIhInIItut bi-Naisllbllr U-yuqtJIila bihim al-l!artIrlyya b,-,hair ."". walr The J:lariiriyyab are a Kbirijite sect according to Sam'inl: IV: 134 s. v. I...ootiDa at tho ...... Sam'iDi quotes for representatives of this sect, they were also active in tho East (Rayy aod Nisbapur) • KluIrils"". - .. Taben, tho ｾｦｴｩｲ､＠ Ill: ISS3, year. 260 A.H. [Badil] yu;hiru al-tlJ!4WWII' wa'l-amr bl'l.",. 'rtJf. Ya 'qiib tbeo used flattery and other tricks to get hold of him. WI Thus a village in the reaioo of Nasal was attacked at niaht by eaemiea called tbI' 'IJT, ."...Imehly Turkic marauders. Tho mll/{llWWi '4 fought back: QantIiyyG. ms Paris, BN foI. 31b. ma. Turtwl Valide 70. fol. 91.. The eveot is dated 392 A.H. At that time, SaowUd rule bad c:oUapsed for aU practical purposes, so that people bad to see themselves to their security. """'y 19 There were, however, attempts at integrating these volunteers into the wider' army controlled by the ruler. TIle qadi b. al-Zubair mentions volunteers alongside regular troops. Like the infantry levies, they were organized on a regional basis. He even names their commander who does not seem to be known otherwise. 1bese troops receive mounts and equipment from the treasury, if only, as in this case, for parading purposes." ••• The 'ayyar groups of fighters have been dealt with in Iraqi and Syrian contexts. 69 But they were present in Eastem Iran also, in Sistin most prominendy, but also farther north, in Kburasan and Transoxiana.1O In Samarqand, 'ayylJr groups are mentioned even before the Samanids' rise to power, and they are again (or still?) in place in the years following the dynasty's downfall. 71 In the latter case the fighters may well have had an organization of their own; at least, they had their own commander. In the account of the qadi b. al-Zubair, the 'ayylJr of Bukhara are seen parading with their flags and banners; they are organized according to . neighbourhoods. 72 In Samarqand and in Bukhara, there were towers in the city wall called wthe 'ayyar tower w.73 But in sum, the source material does not give sufficient - Voluoteen who joined • ai-Rashid. 141f. The CXHJlIIYlOkr is called AI;uaad b. 'Abd ｡ｉＭｗｾ､Ｎ＠ aD army OIl aD iodividual bula <aDd without the leptimation provided by al-amr hi 'I.",. .,.., were perbape DOt called mlllflJWWl'". 800 e.g. the report of how Ya'qIib the ｾｦＦｲｩ､＠ recruited aucIa voIuoteen: Maa'iidi. IV:200 ff. See &lao Bosworth. "Tbe umies-. 541. T'beae were men:eaarie8 rather dum volunteers in that they were DOt part of a Ievk ell masse. • See. above all. CI. Cabea's study. and Havemann. Out of recent studies on the city history of Bqbdad. I'U quote ODly Mikhailova• .. For S-Ultin cf. Bosworth, -The umies-. 538 ff. For Khuruan: Bosworth. ｾ＠ - 171. ｾＮ＠ 167 11 At the beJUming of the revolt of Rifi' b. aI-Laitb: Bal'ami • 1204; tnasl.tioa 2deabera. IV: 472. - At tho ead of the Samanid period: GarOIZi. 176; 'UtbiIMaaioi, 1:341. 'Udi/Jur&dbaqini. 196. Based OIl 'Utbi: b. aI-Atbir. IX:158. '7l ai-Rashid. 145. From fNery neighbourhood. between 200 and 1000 't.l)lJldrlln participated in the .,.,... The ghIJzI fighters bad their own buIners. The two categories of mea . . . have been separated at least OIl 80IDe occ....... - ID this particular cue. the author of the aoun:e _y have been inftueoced by the situation ia _laqi homeland and the period he wrote in (11th ceotUIy A.D.). 7J blRj al- 't.l)lJldr. ID Samarqaud. this tower was OIl one occasion defended by a group UDder' the COIIIIIIaDd of an A1awi who was obviously DOt a professioaal military mao: b. aI-Atbir. X:113 (year: 482). The Bukhanm 'ayydr tower is mentioned in MU'in aI-fuqari, 71. It is of c:oune ooIy a 20 evidence to state the existence of town militias as distinct bodies of fighters haviDa their own commanders, present and active not only on certain occasions, but also in times of peace, and forming their own organization separate from the ablebodied male population at large. 74 It is, however, evident that there were numerous forms in which ·civilian· persons could participate in military activities, and that only a few of these forms were controlled by the ruler. What control there was had to be exercised indiRctly, by intermediaries: the gentry in the case of levh en 11I4fSe, specializ.ed leaders in the case of gh/lz( warfare, local leaders of unspecified background in the case of rlbil{ fighting and local volunteers (rajjlJla or mugawwi '4), more or less popular (and at times unruly) elements with the 'ayyilT groups. The state could (and did) not ·successfully claim the monopoly of legitimate use of physical force-. Military skills, too, were widely diffused throughout society, and the state could not (and did not) control legitimation. It tried to integrate volunteers of all brands, and it tried to organize some kind of cooperation with various groups of fighters. But it is easy to imagine that state builders and rulers would feel uncomfortable as long as this situation lasted, and that they would endeavour to build a military force of their own. -- * •• Thus, volunteers play a considerable part in military activities throughout the Samanid period. At fmt, the Samanid rulers and the social groups who were active . in mobilizing the volunteers and providing the necessary legitimation may have . cooperated quite closely. The first Samanids based their power on the gentry (dahlJqrn) on the one hand and on the ulama on the other hand. That relations between the ulama and the early Samanids were good indeed is shown by a number of reports in the Arabic Qandiyya (and to a lesser degree in other sources as well) relating that Samanid amirs read the funeral prayers for prominent religious dignitaries, above all the qadis of Samarqand. But there also is a leader of ghlJ:zf SUuesbOD. that it just mi&ht have beeo the local 'ayyllrs' task to care for these towers in war and peace. A number' of refereac:es for the terms aIJd4Ih and fltylJn arc givea in my Herncher. Vnminkr. chapter m. section 3, ·Stadtmiljzen-. - In-depth studies of individual towDa may however yield more positive results. as is the case with Hent. See Les hlstoiru • HIrtIt. peper preaeoted by the preseal writ« to the Table Roode meeting on local history. PariaIIvry. Decembec 1993. 14 ｾＱｊｗｦＡＶｍＮ＠ 21 troops among the persons thus honoured. 75 The cases on record encompass the period from 210 - 260 (ca. 825 - 873) with only one exception, that is, roughly the period from the first nomination of Samanids as govemon in Kburasan and Transoxiana (204) to Ismi'U's arrival at Bukhara (261). Two more cases of the same kind are recorded for Balkh.'76 It also is significant that Samanid amin wae accepted as transmitters of traditions from the Prophet, but only down to Ismi 'n; later members of the family either did not care about Islamic knowledge or did not bother to ensure their acceptability.TI Thus, some change in the relations between religious circles and the ruling family must have occurred in the times of IsmI'n or a liUle later. In some cases, religious figures went to war themselves. But their importance for warfare lay more in providing legitimation and legal regulations for it than in personal bravery. An analysis of traditions (l}adIlh) connected with the names of some religious figures in the biographical sources shows that they aimed at making it easier for people to depart for war. Other-worldly rewards were crucial in the fighters' motivation, and sometimes. ulama offered, in the guise of traditions, egregiously high remunerations for participation in ghazw.71 It was indeed essential . for the Samanid rulers to gain and keep the religious dignitaries' support: when the last ruling Samanid asked the Bukharan 1chu!abt2 ' for their assistance in mobilizing the townspeople against the Qarakhanid Turks, they refused and thus sounded the dynasty's death knell. 79 The Samanid rulers, although they clearly had the gentry 11 cues are OIl record in the Arabic Qandiyya, see my -Histories of SamarqIIDd-. $am'iai ODe more cue: Abii Ibribim JsI;aiq b. Ismi'U aI-Bibkissi. d. 259. For him, lsmi'U b. A\lmad rad the funeral prayen (ll:7). This man was a noted preacher (1IIIItlhaIrJdr). he al80 led the building of the ribiI! al-llUll'tlbba • in Samarqaad• 7S offen .. BaIkbi. 214 (year: 239) aad 246 (year: 244). Balkh was DOt under Samaaid rule at that time. It was the pernor of the city. Da'Ud b. 'Abbis. who said the funeral prayers. This man is probIbly a member of tho local rulin& family. the Baoi Di 'w. cf. Narsbakhi, traoslatioo Frye. DOle 297 (p. ISO). ADOtbet cue in tho 88IIIe source uaderlines tho difference from later periods: ID 471. a Seljuqicl prince helped CUI)' tho deceued cpdi to tho cemetery. But this was DOt plaDned. . . . there was DO political desip behiad it; he just bappeoed to meet the funeral processioo oublide the city walla. and lie probIIbly c:ouId not help stoppio& and acting as he did (p. 326). T1 Examples in "The Histories of Samarqand-. Sam'iDi equally reports the teadaen ... tnDsmitten of several early Samanids (s, v. StJnuJnl); he based himself on the 88IDe sources as the staat frapwmts of the Arabic Qandlyya• .. ct. -The Histories of Samarqand-• ." See above. note 42. Bartol'd's presumption that the Bukbaran populace followed -unofficial· preachers in this matter, is hardly supported by the sources. There is nothin& to iDdiCl'te Ibat the ·official- preKben t.d lilly Rl&SOG to strive for the prolongation of Samanid rule. 22 and the religious dignitaries on their side in the beginning, eventually lost not only the sympathy, but above all the active support of both these groups. It can be argued that this was indeed instrumental in their downfall. There is an episode which illustrates how much the relationship between the Samanid amir and his circle and the religious dignitaries had deteriorated. The amir in question is Nqr b. Abmad (r. JOI - 331) in whose time the Samanids possibly rached the peak of their power.1O At the end of his ragn, N3¥ is said to have converted to the Ismi 'wyya.ll He was followed by an unknown number of leading memben of the -civilian- administration.a This has been explained by the Samanids' drive for independence from the caliphs in Baghdad.13 We know in fact too little about the Samanid conception of rule and kingship to provide decisive answen in this case. M The Isma'ms were eventually massacred after N3¥'S death by the -Turkish military-.15 As a consequence, the civilian administration lost much of its influence. But also such support as the Samanids might still have had in religious cireles was definitely lost. Thus, the second (or third) important group of intermediaries linking the Samanid rulers to the volunteer troops were religious dignitaries. They occupied • Frye in eHllV; Bosworth. ·An alleged embassy". I. See EI2 s.v. Nasajt (I. Poooawa1&). I l In the Arabie Qandiyya DOt formerly used in this cootext. we reIId: ·Their fon:e JI'OW. mel tbeir missiooary IICtivities radled the horizoas. ud 10caI nalcrs. powerful mea aDd dahiJqDl joined their cause as well as prominent mea in the administration". ma. Twbau Valido 70, fol. 23b. See ..., my "Histories of Samarqand". 90 ud note 49 for the Arabie text. IS al-Nadim, truLslation Dodge, 468 note 71. .. It would 80CIIl plausible to explain the Ismi iii flirtation of the civilian 8dmiDistratioa by • desUe to have their own ntJigious orientation. lID Qite faith whiell would ad them apart from uIama aDd dtlh4qDl alike. What is kDown about the speeial brand of Ismi'i1ism brought to TJ'IID8OxiIIIla (it bad • distinct Neoplatonie taste about it) is certainly in keeping with the .ppeaI it ..,.,....,..dy bad ro.Qitist circles. It would make sease, too. if rulers who are trying DOt to depend on forc:es in the midat of society whom they caanot cootrol (the religious dignitaries and the tIahaqtn) for military pIItpOa. were looking for • religion whiell would further set them apart from these forces. The failure to coovert the military leaden. however. would meao that opposition between the civilillll admiai8Inaion aDd the military bad grown too strong beforeband. Bureaucrats were ever suspect of boIdiq beck resources the military felt they deserved. - For Ismi'ili doctrine. see MadoIuog. IS The Arabie Qandiyya says that the "military" (1Jasham) killed one of the most importIIIlt Ismi 11i figures. the vazir AbU rayyib al-M... 'abi (paris ma, BN 6284. fol. 43a-b; Turban Valido 70. fol. IOSa). On this mao. see EI2 S.v. MlI.1'abtand -The Histories of Samarqaod-, DOte 49. 23 themselves more with providing legitimation and offering legal and theological rulings for the tighten than with warfare itself, although in some cases, they led ghIJzI groups in person. As happened with the gentry, the Samanids lost contact with this group, or maybe more or less consciously put an end to the tacit compact of cooperation the first representatives of the dynasty had concluded with the Is1amic Bite. Both the dDhilqtn and the ulama cases show that there was no monopoly of the state on the legitimate use of physical violence. The rulers neither controlled the fighting groups directly, nor did they have significant influence on the legitimizing rationale. The state did not reach very far into society, but had to make use of intermediaries to a very large extent. ••• Up to now, the focus of this paper has been volunteer troops. The gentry were important in that they recruited and commanded them; religious dignitaries, apart . from personally participating in jihlJd or ribtl{-type warfare, provided legitimation. Neither group, however, was loyal to the state or the dynasty as such. The gentry were loyal to the dynasty insofar as they held a share of power; the ulama and the leaders of ghIJzI troops were loyal to the ruler insofar as he acted as a ghlJzr himself. Thus, problems of loyalty were bound to arise. The volunteer armies could only be used in such campaigns as the legitimizing rationale permitted: above all, against non-Muslims, that is, against the Turks to the north and east of Transoxiana. It has often been noted that with the Turks' conversion to Islam,jihlJd warfare lost some of its edge." Conversion on a massive scale, however, is reported only from the middle of the 4th century on.17 Accounts of earlier contacts do not as a rule include conversions. U • Cf. Spuler. Frye. BartoI·d• ., Cf. b. aI-AdUr. VIU:532 (year: 349). S6S (year: 354); IX:474 (about the early Soljuqid biItory); 520 (year: 435). 'I1U. is supported by the somcwbat blurred IICCOUDt of Jamil QanrIai: The tint COUDtry of the Turks to IICCept bIam was aI-Shish (Tashkeat); the cIatin, is iocoDclusive: foI 43-. sec also Bartol·d. Enalisb tr8DsIation 2S5f. • Cenpipa of Iami 11: Tabari. m:2138. year: 280 (cf. b. aI-Adir. Vll:464f); W:2248. yea': 291 (b. aI-Athir. Vll:533). See also Bartol'd. 'Turkestan. 317. - The accouot of tbe 280 eampeip ia NanrbKhi bas the defeated ruler of Tariz -.:cept IsIam-: ed. Ri4awi. 102. tn...l.tioa Frye. 87. tIIfIIt'-i rarllZ blHbl iImIId wa i.r1lJm tJwartl b4 bi.ry4r-i diJrqibtiIn. The aettiDa is dIanderizod __ by political surreodet &:baa by penoaaI conversion. - AcceptiD, blame may be cotermmou. with the rule of. Muslim overlord-. -.:c:eptin, 24 In some of the writings of the geographers, it is likewise taken for granted that the Turks are not Muslims.19 Even if the geographers' accounts cannot be taken at face value for the state of affairs in their own time since they often reflect earlier periods, it seems that mass conversion of Turks had taken place only in a limited number of regions until the middle of the 4th century A.H. Cooperation between groups of fighters and the Samanid rulers in ghazw had considerably weakened by then. No important campaigns against the Turks are mentioned for the two rulers following Isma 'n who is in fact the last Samanid to be credited with such activities. 90 Thus, the conversion of the Turks and a subsequent decline in ghazw activities can hardly be the reason why the Samanids no longer relied on volunteer forces to the same extent as they had done before Ismi 'n. From the time of Isma'D onwards, the Samanids began a career of conquest. After his victory over the ｾ｡ｲｩ､＠ 'Amr b. a1-Layth (achieved with an army composed mostly of volunteers led by the gentry),91 Isma'D's military endeavors were directed more towards Iran than to the east. For his conquests, he could not use the -traditional- volunteer armies any longer. There was no legitimizing purpose for them in the eyes of the ulama, and even the dahlJqtn may have been reluctant to embark on this enterprise. Garrisons in distant parts of Iran could not possibly be manned in the same way as ribilfs. Empire-building was the issue now, not defending one's own country or fighting the infidels. Out of considerations of effectiveness and loyalty, Isma'n had to have a professional standing army. What he created, was a • B.g., Y. 'qiibi. 295: They are surrounding Khurasan. and there is continual warfare betweea the Turb themselves OIl the ooe band. and betweea the Turks and the Muslims OIl the other bud. In the same vein b. u-Faqih, 316; Ibis author thinks that the Turks are mostly 1.IlIIiIdiqa which Day or Day DOt mean MaoicbeaDs: 329. Muslim Twb are livina in the Isfijib repOll: b. I;Iawqal. 510; l!udtId al- '4lam, 117. Muslims are livina in the regiOil of al-qarya ｡ｬｾ＠ (Yiaaikiot.1ower S)'I' Daya). although it beloap to the Gbuzz (who evidently are not Muslims). b. J:lawqal. 512. III In the Arabic QandiyyG. in Sam'iDi and in Narsbakhi. 1be ｾ｡ｦｲｩ､＠ could be seen as a menace to Transoxiana. so this was. defeasive war. Besides. were upIIWts without ally Iiob to the ｰｾｉｳｊ｡ｭｩ｣＠ past. ｾｦｴｩｲ､＠ overlordship would have been uloJetber UIIIICCeptabIe to the Tranaoxaniao dohIJqDl. This is clearly visible in the account of CIIlvay. and IIIe88IIp8 excbaqcd betweea Ismi'iI and 'Amr: Narshakhi. traasIatioo Frye, 87f. Whereas the Samanids could with some de&ree of plausibility claim de&ceat from Bahdm CbUbia &ad certainly were of dihqIln. that is. noble origin. the ｾｦｴｩｲ､ｳＧ＠ claim to Sago;_o desceat was very much opeo to quesbOll. See Bosworth. -The bcritap-, 59. 9\ the ｾｦｴｩｲ､ｳ＠ 25 • slave army· - he had the Abbasid model for that.92 Volunteers were still used in many campaigns, but the core of the army was now made up of troops led by the ruler's personal retainers. 93 When Isma 'n became governor of Bukbara and tried in earnest to gain control of the region, he bad more than one problem to tackle. As a wily ruler, he did not deal with them simultaneously. His behaviour indicates how much personal power he bad (or did not have). He first cleared the region of a band of ·robbers· (probably unruly elements), armed groups of popular origin. In this, he relied upon the nobles who previously had been in command. A ·police chier is mentioned; there is no information as to who the ·policew wereM (personal followers of this chief1). Isma 'n himself apparently did not have significant troops of his own. He then had to deal with an enemy from the outside. Again, Isma 'n wassembled as much of an army as he could· ," so he obviously did not have a professional army at his disposal. After that, he tried to weaken the position of the leading families in Bukhara; he could not confront them militarily, but had to use trickery.96 It is in Isma 'n's dealings with his brother ｎｾｲＬ＠ head of the Samanid family and ruler of Samarqand, that we first see a figure who may have been a ·slave general·, Simi '1Kabir;9'7 but volunteer (ghiJzJ) troops from Bukhara also participated in the fighting. Since this was more than ten years after Ismi'TI's takeover in Bukbara, we cannot exclude the possibility that he had built up a military following of his own in the The Abbasid slave mny was created by the caliphs al-Ma'miio mel ｡ｬＭｍｵＧｾ＠ in the 9th ceotury A.D. It bas to be noted that both of them spent coasidenbIe time in n .......... So, eveD if tho Abbasid -slave army- is prior to the Samaaid ODe, tho model itaelfmay very well have Eastem Iranian roots. Cf. Beckwith. - For the Samanid use of slave mnies. tho SiytbtllIU'Ima is frequently quoted as a source; so it is for the Ismi'iU episode mentioned move. I penoaally believe that Nqim al-mulk was to such an extent guided by political intenst that his text IIIU8t be treated with extreme cautioD. The stories about the Samaoids are evidently told for didactic reuoas; the sultan is warned apiDst the 1Jitinis who flourished in his day, and is admonisbcd to keep hi8 troops in JOOd order. 9l bcaiDninI of the " Bosworth in many places. e.g. -ghulllm. ii: Persia-, where be describes the structure of an umy comprisioa ghilmlln: -[T]be particular coacem of the ghulilms was normally to act .. a dcpadable 6lite force within the widei' body of the umy and as a penoaal bodypanl-. M l;Iusain b. al-'Ali', ｾｦｉｽｩ｢Ｍ＠ 95 Frye's trmsIatiOD. ＤｨｾＬ＠ Narshakbi. ed. ｾｷｩＮ＠ 9Sf., traosl. Frye. 81. 81; ed. Ri4awi.96. • See DOte 34. Frye'. 9'/ He -bad been • slave of their father, and was (Ismi 'U's) ｣ｯｭ｡､･ｲＭｩｮｾｦＢＬ＠ tnosIation, 85. Ri4awi, 100. Slave is ghul4m. In another context, Simi'I-Kabir i. called -client of his father-, Frye, 28, Ri4awi, 34; client is nuJWltJ. 26 meantime and that these troops were made up of -slaves- or -clients-. His brother Nqr's army, on the other hand, comprised the personal followings of the rest of the family as well as -the Turks from Isfijib-. 9I Other names of military leaders from Isma 'U's reign include Bars al-Kabir who is called a ghulilm of Isma'D's." He was governor of Jurjln. We are told that he collected enormous sums as taxes from the districts of Jurjan, Tabaristin and Rayy and had already sent them to Bukhara when he learnt that his -master- Ismi 'n had died. He then called the caravan back and took the money himself. When after a while Isma 'n' s son Al}mad succeeded in stabilizing his rule and came to Nishapur, Bin fled to Baghdad with 4000 horse. loo Another one is the founder of the Simjurid dynasty, Simjur al-Dawati. He is called Isma 'n's mawliJ10l and was later named to various leading positions, among them the governorate of SIstan. We know next to nothing about the background of one of Ism 'n's most loyal commanders, I;{amawaih. He fought at Isma 'n's side in the very beginning (in the year 262, when the Samanid's position in Bukhara was not stable at all).lCl'l A man also referred to as l:Iamawaih b. 'Ali appears in a similar position some 40 years later, quelling several revolts on behalf of Isma 'n's grandson, Nap- b. ａｾｭ｡､ＮｬＨｮ＠ There can be doubts about these men's identity, but not about their (or his) loyalty. He is not referred to as a ghulilm or mawlll, but on the other hand, his genealogy is not given. A genealogy is furnished for yet another of Isma'n's commanders, Atlmad b. Sahl. He descended from a family of dihqllns in the Marw region. Having been • Probably there was 80IDe kind of coveoant with the Muslim Turks of that region. In I!IMlIkl al'tJlom, they are called tJshN Turks, those with • treaty (117. Minorsky. 118). In 299. duriDa • Wrevolt Win S'""1stio. wtreaty Turks w 11I1'kIJn-i zinhiJrr sided with the rebels. TiJrl1ch-i SI.WIJIt. 298f. " b. aI-Athir. VDI:45 (year: 296). - A mID named Birs-i Baybndi is meDtioaed . . . military lelld« undel' Ismi'il in Narsbakhi. cd. Ri4awi. lOS. He took part in the campaign apiDst 'Amr b. al-Laith• .. Nomination .. governor in Jurjill: b. ai-Athie. VD:527. year: 290. Flight to Baabdad: VID:7. Arrival there: VID:45. He was employed .. lovemor in diyar IWbr'a. 101 TtJrtkh-i SfSllJn, 293. See also MClVil. ·Simcurilcc w I. ICD b. ai-Athie. Vll:281. ICD b. aI-Athir. VllI:77; Narsbaklii. tnnsJatioo Frye. 95. ed. Ric;lawi. l1lf. 27 molested and even imprisoned by 'Amr b. Layth, he took the Samanid's side of his own accord. He was influential all along and remained loyal until the amir Nafr ·made him promises he did not keep afterwards· (m the context of a ·revolt- wbell Al)mad b. Sahl's support was needed). The proud dihqan took offense and revolted, using his considerable influence in Khurasan, above all in the Marw region. lot These five examples will have to suffice to show that Isma 'n's commanden (and those of his successors down to his grandson ｎｾｲＩ＠ came from different backgrounds. Some represented the old leading families, some the -new- system of ghulilms and mawlas. lQ5 Rather than insist on the juridical status of these men (whether they were slaves, freedmen or freeborn), we will have to look at their relationship to the ruler. But before turning to that question, a few summarizing remarks seem in order. *** With Isma 'n, a change took place in the way rulers saw their position. Up to then, the early Samanids had more or less followed the pre-Islamic model: there was a nominal overlord in Samarqand, but he did not interfere much with his ·vassals·, the dtlhiJqtn. Military power depended on their voluntary support. A new element had been introduced by Islam; the new religion provided a religious legitimizing rationale for frontier warfare (although frontier warfare must have been a familiar phenomenon in the region). This was now conducted as ghozw, ribIl! or in similar forms, all of which presupposed voluntary participation and self-organization to a high degree. The ruler, not to speak of the state as an abstract principle, was the foremost ghiUJ' fighter, but not really in command as is shown by such campaigns as did not involve any initiative on the part of the state. Isrni 'n seems to have been the first Samanid for whom kingship was an aim and a necessity. He certainly was the first really to rely on troops led by men personally loyal to him, not to a legitimizing purpose. In order to ensure his own position, he needed such forces, and he had - up to a point - to discard those who previously had made up the core of such military power as was at the dynasty's disposal. His attitude regarding the leading Bukharan families from the pre-Islarnic ... b. aI-Atbir. VIll:1l7f. year:306. HI5 In KhUrasaD, Rifi' b. Hartbama is said to have introduced -military sJaveJ')'w 00 • laqe acaIe. Wheo he was defeated and besieged in Nisbapur. -his companions and slaves left him. He IuId 4000 slaves. No perDOf of Khurasan before him bad so many·, b. aI-AtIir, Vll:4S7f. 28 period is indicative of this; they all but disappear from the historian' s record in the times of Isma 'n. That kingship was indeed what Isma 'n had in mind is perhaps best illustrated in the well-known story about the Outer Wall: When the wall began to fall into disrepair, he refused to order it to be built up again, saying: -While I live, I am the wall of the district of Bukhara 106 In a way, he still brought about a synthesis of two very different systems of military organization: the wold w one in which the ruler was only primus inter pares, and the wneww one in which the ruler was princeps in his own right. W• Isrna 'n's endeavour to build up an army led by his personal retainers amounted to the elimination of the previously prominent groups of intermediaries, the gentry as well as the religious dignitaries. He wanted the military to be personally loyal to him, not to a legitimizing purpose he could not control, and not to nobles who had a power basis of their own. The system of a ruler who runs a state through his household (or: as if it was his household) has been described as patrimonialism. 107 The ties linking the ruler to his retainers will be discussed presently. *** The most important tie between a military leader, either a wslave w or a man who had joined the ruler out of his own will, was personal loyalty. Mottahedeh has discussed this at length in his fascinating book on Loyalty and Leadership in a BUyid . context. 101 This is a mutual affair: the ruler has not only rights, but also obligations; if these are not met, the relationship changes. The ruler can expect to be obeyed only if he himself proves loyal to the wslave w• The examples cited above are instances of how this relationship worked. The military commander and governor of Iurjin felt loyal only to Isma 'n, not to the dynasty; even though he had sent all those riches to his master, he felt that he had the right to withhold them from his master's son (as long as no new compact of mutual loyalty had been concluded). In this, there is no difference between this governor (who may have been a slave in that .. Narsbakbi. eel. RMJawi 41, tnmslatioo Frye 34. He is praised because be lived up to his promise, but certainly this shows a cbaoge in defense politics: the center of gravity shifted away to Ibe rulet and hi. military ..,.,.,..... from the gentry and the volunteers. lin The term is, of course, Weber's. For a detailed acx:ount of patrimonialism as conceived by Weber in a Muslim context see 1naIcit. Hodpoo bas decided to keep a reserve about Weberiaa of IsUun, vol. I (preliminary remarks) . terminology 00 general grounds in his ｖ･ｮｴｾ＠ .. See bihlioarapby. 29 he was bought by Isma 'n and brought up as a soldier) and the freeborn noble Al}mad b. Sahl who likewise, after a long life of loyal service, became a -rebel· after having been slighted in a way he found intolerable: the ruler did not keep his promises, so the personal link was severed. Now, Ismi 'n is praised because he cared extremely well for his mawDlr. 109 He is the ruler who kept his promises, the ideal patrimonial ruler. It has been argued that mawaJr in a Central Asian or even Eastern Iranian (Khurasanian) context does not mean ·client· if this implies an unfree status at some point of the mawllJ's career. ItO Rather, -vassals· are to be understood, the overlords being either the Caliph or else the local ruler; in any case, the Central Asian concept behind it is the status of chlJ/cQr as discussed by Beckwith.1II Whether this relationship involved a background of ·sacral kingship· even in a Muslim context (Beckwith is in my view right in linking its pre-Islamic origins to such an institution) cannot be discussed here. 112 An army led by personal retainers lOt Narsbakhi, transl. Frye, 28, ed. Ri4awi, 34. lie Beckwith, 31; Crone, 78. III In biB study quoted more than 0DCe. Numerous examples could be added to biB. 1 will reatrict myae1f to two of them. First, in the hagioaraphic sources not nol'lDlllly coasuIted in theao matters. chtJk4r is fouacl in the """";01 of -penonal retainer (bound by penooal loyalty 01' oath), but DOt • slave-. The vazir Bil M. . . . Warqini kept his master, the Seljuqid Tugbnl, waitinl, bectluse be had not yet finished his prayer: -I am God's slave and your chiJJ«u. As lon, as 1 have DOt fulfilled God's orders and have not completed what God'. slave has to do. I sbaU not begin with my obligatioas as your dIIJJcar. If you want. vazir who is not God's slave but entirely your ch4kar, I sbaU ao home DOW.(M''';wmlJUld b. al-Munawwar. 319). The same source shows a sayyid and • baud of hi. foUowers who are called his cIuJkarrJn who are tryin, to confront sbaykh AbU Sa'id. ibid.. 22lf. - Second, in the TtJrlkh-i St#iJn. IfIIZWI4 clearly deaotes • notable. A prominent case is Mul;tamlJUld b. Hurmuz, called -Mawli SaadaIi-. 297ff. Bee aIao b. aI-AtIir, vm:69. Another instance TtJrlkh-i SlsliJn, 299-301. Both theee examples are dated around 300 A.H. 112 But certainly the Dumber and quality of reIaiDets fIockin8 to a ruler or a preteader .... to do with that penon's chari.ma. lDalcik, SS, stresses that the oriJins of the patrimonial household (ovee in the Ottoman case) are to be traced back to the -warband of nlJUrs or yoldtuh- typical of every coaqueror in the steppe. See aIao the studies of B. MlIDZ and J. Wood. 30 in this understanding seems to have been the only structure available to Ismi 'n in his endeavour to build an empire (meaning conquest): it was the only one he could expect to be loyal to him given the purpose he had in mind. Considetations of effectiveness and loyalty were thus instrumental in his choice of such a military format. What, then, about considerations of expense'? ••• An army led by personal retainers certainly was more expensive than the army of the "old" format made up of volunteers and led by gentry commanders. In the first place, a standing professional army is, as a general rule, more demanding on the state coffers than a non-professional one, since it has to be paid even while not on campaign. But above all, personal retainers acquire claims on their master he will have to meet in order to ensure their loyalty. It is useful in this context to remember that the personal tie between retainer and ruler was a mutual one. The central concept here was ni'ma, "benefit" . III This is quite material, "benefit" meaning more often than not money or opportunities of making money. Even more abstract "honours" expressed in the system of ranking going with them not infrequently involve positions which either can be translated into material terms or else increased influence which boils down to the same thing. The Qwadgu Bilig more than once or twice underlines that a ruler has to be generous in order to have loyal troopS.114 Even in the most ideal of cases, when both the ruler and the retainers are faithfully keeping their obligations, problems are liable to emerge. Every individual retainer will strive to earn as much "benefit" as possible and to get the ruler to prefer him over the others. Such built-in competition can have a positive effect on the army's loyalty and effectiveness as long as resources are plentiful and their allocation is in finn control. But as soon as resources become scarce and/or control weakens, 113 See the extensive discussioo in Mottahedeh. The master is called wall tIl-ni '1fUJ. the retaiDer owes -,ratitude- shuler, if he falls short of this. he is accused of Icqfrtln til.,.; '1fUJ aod 110 00. All tt.e.e expreaioas are technical terms for certain types of behaviour. II. E.,.• verses 2057.2808,3031.5476.5514.5591. The source also differentiates betweaa troops loyally serving • ruler who coaforms to the principles of kinpbip as laid down in the book OIl the OM band and slaves on the other. The ruler is warned: -Free-bom men win gladly serve OM with . . qualities [that is. ODe who treats the fighters well] aod sacrifice themselves for his bappi.....-. vene 4406 in Daakoff's traDslatioo. The ori&inal (and with it the Turkish translatioo) is more explicit in sayinl that free-bom men win become such. ruler's slaves. After that, the ruler is told DOt to spead his treasure on slaves but on such free..bom men (who will serve him better). 31 distribution coalitions between retainers and groups of retainers are likdy to form. II! The last decades of Samanid rule are, as is well known, chamcterized by more or less continuous struggle over provinces, governorships and the like; most of the acting parties were commanders of -slave troops-. Every time a ruler (or the civil administration in Bukhara) tried to regain control over the allocation of resources, the answer was -revolt- .116 Resources could no longer be enlarged by conquest, and thus, from a shrinking pie more and more commanders demanded ever increasing slices. In more than one case, military leaders distributed the coveted posts among themselves without bothering to ask the ruler or the administration in Bukbara. Very quickly, the possibilities of extraction available to the Samanids proved unsatisfactory for military needs. From NUl} b. N . (r. 331 - 343/943 - 54) on at latest, there was a financial crisis. 117 The increasing demands on the fmancial apparatus are illustrated by a story from Sistin. Shortly after the Samanids had conquered this rich region (the conquest was a difficult and protracted affair which took place in the reign of AI,mad b. Ism'n around 300 A.H.), they had to increase the taxes because the salaries of the·. Khurasani army were higher than the 1 million dirham Sistin yielded formerly. With that, the soldiers had to be accomodated with the inhabitants. 1I8 Revolt was not slow to break out. Its leader complained that -it is not customary in Sistin to liS See ·Olsoo's theory· OIl distribution c:oaIitions. I" This is e.,. recounted by Mirkhwind. IV:70f. Since ｍｾ＠ (the later Ghamavid ruler) is UMble to regain his ancient po8ition after his father's death, be tbiDks that he bas faII_ iDto disJnIce. He CIIDDOt IICcept abe asww aiv-. namely. that Beatuzun (another ·slave· .....) wbo ... heeD pv. abe JOVernorate of Khurasan (indeed ODe of the IDOIIt important spoils) bas _ ｾ＠ fNfIIr him because of his seaiority. The real reasoo for preferriDa Beetuzun was that he wu the ODIy military collUlWlCler at band to repel the meGllCin, Twb. Beatuzuo himself wu DOt satisfied with wt.t he JOt aDd, havina formed • coaIitiOll with the -turncoat Faiq· (Frye, ·Samlmida-. 156). ctepo.ed the amir, • mere boy. After thai, Malpniid was coavinced that the Samanid dynasty wu 011 lite decliM. - Foe' aevenl cues of continuity in the inner cin:le of Samanid .d Qankbanid nile see DOW Kodmev. 117 See Bosworth, ·N. b. N ....·: ｾ＠ costs of quelling internal rebelliousoeaa .d of the WU'8 in DOfthem Penia caused. fiDmciai crisis during ｎｾＧｳ＠ reign. with the army often.goin, unpaid.d the subjects complainio, of increased tuatioa·. Bosworth adds that subsequent rulers were DOt 8ble to improve this situatiOil. II. This is one of the most odious thin.. _ army c:.n do since it fundamentally disturt. people·. lives. Bvea if there is DO outriabt tnDspessioo. women have to behave very diffilraltly if ..... is • 8tnopr in the bouse. People expect that .... army stay in ita camp. ｾＮ＠ If. camp is DOt at ｾ＠ barncks have to be coastructed. See mo the legeod of bow •AbdaUib b. 'fihir built lite prrisoo town of Sbidyikh near Nisbapur. e.,. in b. al-Athir. XI:273. mo in Yiqiit. m:3OSb S.v. ShIJdyakh. 32 increase taxes w•1I9 The revolt was quelled, but it is not clear whether the Samanids ever were able to take in the increased taxes. This example also shows that the troops in the outer regions (far from Bukbara) had to levy the taxes needed for their upkeep where they were without help from the center. Later on, the military commanders flatly refused to pay the central divan anything. This was maybe not only insubordination, but a necessity: they bad so many troops they had to provide for that the resources available in their province were hardly enough.12O As a rule, military commanders were assigned a province as governors. Only in some cases did they have to deduct anything from the province's income for such purposes as the upkeep of fortresses and the like.121 Thus, the retainers were quite rational in demanding an ever larger part of the empire's resources. They had to pay their troops in order to keep them. In sum, they had evolved into a new group of intermediaries: the ruler held military power only as long as he could get one of the slave generals to provide him with troops. But the wslave generals w were not rooted in society in the same way the gentry or the religious dignitaries were; on the contrary, their mobility has often been stressed. Their services had to be bought, as no legitimizing rationale could convince them. Their loyalty depended more and more on the extraction capacities provided by the civilian administration. *** To come to a conclusion: In my view, slave soldiery is by no means a direct consequence of Islam as a religion. In Eastern Iran and Transoxiana at least, Muslims did not abandon military affairs; most men, town dwellers and rurals alike, were skilled (to a degree) in the use of weapons. The dohilqtn, as intermediaries, were instrumental in mobilizing and organizing local troops and leading them afield. Garcin seems to agree with Crone: w[1]f a military aristocracy [ ... ] had developed in the . Abbisid era, the constitution of regiments of slave origin might have been II' T4I1kh-i SfsI4ll, 297 f. 121 AbU •Ali S"unjiir b.d taken over all of Khurasan -and divided the tax monies between his followers. N-. b. Mm,iir ubd him. to leave some of the province to the administration of the PenoaaI DiVlUl (dlwdn-i kh4f.f). He refused- and pw as a I'e8SOD that his foUowen (1!ashDnt) were so numerous that he DOt oaIy could DOt waive auy of his income. but he also b.d to ask for 1DI:n. 'Utbi/Judidbaqiai, 91. See also Meavil. -Simcuriler-, IV:SSOf. III 'UtbilJurfidhaqini, 6S. 76. 33 avoided- .122 The point is: there was a military aristocracy, at least in the area under study, and it was active well beyond that area. In this respect, there was no need to use slaves as soldiers. They were pre/erred because rulers thought they would be more dependable than volunteer troops led by gentry intermediaries. Not considerations of effectiveness, but of loyalty were the issue. In 1be Samanida could at first well afford to have troops loyal to the ruler only, and they did all they could to separate themselves from the intermediary strata they previously had depended on. The possibilities of extraction their civilian governmental apparatus commanded were however not large enough to ensure permanent loyalty. 1be reasons for this are to be looked for in the economy of their time. l24 The Samanid state broke up because it no longer controlled the links between military format (the increasing demands of the -slave generals-) on the one hand and extraction and allocation of resources on the other hand. The Samanid rulers had tried to build an autonomous state in order to exclude intermediaries from power; these intermediaries, however, were not eliminated. In the end, they not only turned their backs on the last Samanid rulers, but actively supported the next dynasty. 122 Garcia, 117. 1:D Slave armies were DOt more effective than other armies of the day. Time and apia. they suft'eIed cruabiDl defeats at the band of nomads, the first iostance beiDa the Gbazaavid cIisut« 8t l)endiqaqiD ia 1040 (A.D.) apinst the Seljuqs. 1:14 This is DOt the place to discuss $amanid economics. Their resources comprised Imd tax, reveaue from trade, 8bove all slave trade, 8Dd income from silver mines. One of the most _ _ features is the CODStaot drain of specie to Eastern Europe, where enormous amounts of Sanvmid silver have beea found. Which kind of income was ID08t important is hard to guess. Bibliography ａｮｾＬ＠ 'Abdallah: Tabaqllt ｡ｬＭｾｦｩｹＮ＠ Ed. lJabibi. Kabul, 1962. AyaIon, David: "Aspects of the Mamliik Phenomenon: The Importance of the MamlUk Institution". - In: Der Islam 53 (1976), 196 - 225. AyaIon, David: IfL'esclavage du Mamelouk". - In: Oriental Notes and Studies 1 (Jerusalem, 1951). Ayalon, David: "Preliminary Remarks on the MamUik Military Institution in Islam". - In: War, Technology and Society in the Middle East. SOAS, London, 1975, 44 - 58. Bat 'ami, Abu 'Ali Mutmnmad: 1rehran, 1353/1975. TIlrIkh-i BaJ 'amf. Ed. GuoibadI, 2 vols., BalkhI, 'Abdallih b. 'Umar; Persian version by 'Abdallah l;Iusaini Balkhi: FtlI/Il 'il-; Balkh. Ed. l:Iabibi, Tehran, 135011951. BartoI'd. V. V.: "1rurkestan v epokbu mongol'skogo Ddestviya". - In: Sochineniya I, Moscow, 1963. Beckwith, Christopher: "Aspects of the Early History of the Central Asian Guard Corps in Islam. If - In: Archivum Eurasiae Medii Aevi 4 (1984), 29 - 43. Bosworth, C.B.: • An Alleged Embassy from the Emperor of China to the Amir N . b. AIpnad: a Contribution to Simanid Military History". - In: Minovi, M. & I. Afshar (eds.): YIld-1IIJma-yi lWlnf-yi Minorsky. Tehran, 1969, 17 - 29. Also in: 1he Medieval History of Iran, A/glumistan and Central Asia. Variorum Reprints, London, 1977 (text XXII, pages run from 1 - 13). j 35 Bosworth, C.E.: ·The Armies of the ｾ｡ｦｩｲ､ｳﾷＮ＠ - In: BSOAS 31 (1968),534 - 54. Bosworth, C.E.: "The City of Tarsus and the Arab-Byzantine Frontiers in Early and Middle Abbasid Times". - In: Oriens 33 (1992),268 - 86. Bosworth, C.E.: ｾ＠ Ghaznavids. Edinburgh, 1963. Bosworth, C.E.: "Ghulim, ii: Persia". - In: E12. Bosworth, C.E.: "The Heritage of Rulership in Early Islamic Iran and the Search for Dynastic Connections with the Past". - In: Iran 11 (1973), 51 - 62. ｎｾｲＢＮ＠ Bosworth, C.E.: "Nul} b. - In: E12. Bregel, Yuri: ｾ＠ Role o/Central Asia in the History o/the Muslim East. n.p., 1980 (1be Afghanistan Council Occasional Papers : 20). BUchner, V.F.: • Samaniden " . - In: Ell IV: 130 ff. Bii'l-F3f;\l: TIJrCkh-i Bayhaqr. Ed. GhanilFayyi4, Tehran, 1324/1945. Cahen, Claude: "Mouvements populaires et autonomisme urbain dans l' Asic musulmane du Moyen Age". - In: Arabica 5 (1958), 225 - SO; 6 (1959), 2S 56 and 233 - 65. Castagne, Joseph: ·Le culte des lieWt saints de l'Islam au Turkestan". - In: L 'Ethnographie n.s. 46 (1951), 46 - 124. Chabbi, Jacqueline: "Remarques sur Ie developpement historique des mouvements ascCtiques et mystiques au Khurasan·. - In: Studia Islamica 46 (1977), 5 -72. Claessen, Henri J. (ed.): ｾ＠ Early State. The Hague, 1978. CJaessen, Henri J. (ed.): Early State Dynamics. Leiden, 1987. Cook, Michael: (Comments on Garcin). - In: Baechler et al.: Europe and the Rise 0/ Capitalism. Oxford, 1988, 133f. Crone, Patricia: Slaves on Horses. 1980. ｾ＠ Evolution of the Islamic Polity. Cambridge, 36 Dodkhudoeva, Lola: Epigrqficheskie pamyatniki Samarkonda XI - XIV w. Dushanbe, 1992. Donner, Fred M.: (Review of Crone). - In: JoU17Ull of the American Orienlal Society 102 (1982), 367 - 71. Finer, Samuel E.: ·State- and Nation-Building in Europe: The Role of the Military·. - In: Tilly, Charles (ed.): The Fomu:uion of National SIllIeS in Westem Europe. Princeton, 1975, 84 - 163. Frye, Richard N.: Bukhara. The Medieval Achievement. Norman (Oldaboma), 1965. Frye, Richard N.: ·The Samanids·. - In: Cambridge History of Iran IV, Cambridge 1975, 136 - 61. Frye, Richard N: ·'Tarxun - Tiirxiin' and Central Asian History·. - In: Harvard Journal of Asiatic Studies 14 (1951), 105 - 29. Frye, Richard N. (see Narshakhi). Garcfa-Arenal, Mercedes: "La pnictica del precepto de al-amr hi 'I-mo 'rlif wa '11U1hy 'an al-mun/cQr en la hagiograffa magrebf". - In: al-Qantara 13/1 (1992), 147 - 70 • .Garcin, lean-Claude: ·The Mamliik Military System and the Blocking of Medieval Moslem Society·. - In: Baechler et ale (eds): Europe and the Rise of Capitalism. Oxford, 1988, 113 - JO. Gar01Zi, Abu Sa'id 1347/1969. ' Abd al-l:Iayy: Zain al-akhb4r. Ed. lJabibi, Tehran, Goitein, Shlomo D.: A Mediterranean Society. Vol. II, ·The Community·. Berkeley, 1971. Gorodlshche Paylcend. Tashkent, 1988. Havemann, Axel: Ri 'lJsa und qa4lJ. Institutionen als Ausdruck wechselnder Kr4fteverhlJltnisse in syrisclum Stl1dten rom 10. his zum 12. Jahrhunden. Freiburg/Br., 1975. Hodgson, M.: The Venture of Islam. 3 vols. Chicago & London, 1976. 37 Homani, Albert: -Ottoman Reform and the Politics of Notables-•- In: The F.mergen« of the Modem Middle East (Collected Papers). London, 1981. lfudtJd al- 'Ilkun. Ed. Suwda, Tehran 1340/1962; translation and copious annotation VI. Minorsky: The Regions of 1M World. London, 1937 (OMS n.s. : 11). b. al-AthIr: al-KlJmiljt'l-tlJrlkh, Tornberg's text reprinted Bairiit, 1965 ft. b. al-Faq'ih: K. al-buldlJn. Ed. M. J. de Ooeje. Leiden, 1885 (BOA: S). b. Funduq: TIJrIkh-i BayhmJ. Tehran, 1317/1939. b. I;Iawqal, Abu'l-Qasim: K. al-masiJIilc wa 'l-mamIJlik. Ed. M.J. de Ooeje, Leiden, 1883, reprint Leiden, 1938/9 (BOA : 2). lnalcik, H.: -Comments on 'Sultanism': Max Weber's Typification of the Ottoman Polity-. - In: Princeton Papers in Near Eastern Studies 1 (1992), 49 - 72. I,tabD, Abu IsI}aq Ibrahim b. ｍｵｾｭ｡､Ｚ＠ K. al-mtlSiJlik wa'l-mam/Jlilc. Ed. M.J. de Goeje, Leiden, 1870, reprint Leiden, 1927 (BOA: 1). Kochnev, B.: Karakhanidskie monery: istochnikovedcJu!skoe i istorlcheskoe issledovanie. Avtoref. diss., Moscow, 1993. livshits, V.A.: -Praviteli Pancha-. - In: Narody ADi i Ajiiki 1979 nr. 4, 56 - 68. Madelung, Wilferd: ReligiOUS Trends in Early Islamic Iran, Albany, NY, 1988 (Columbia Lectures on Iranian Studies : 4). Manz, Beatrice F.: 71u! Rise and Rule ofTamerkme. Cambridge, 1989. Mas'iidi, Abu'l-IJasan 'An: Murqj al-dhahtlb. Ed. Mut,ammad Mut,yI al-dIn •Abd ｡ｬＭｾｩ､Ｌ＠ Bairiit n.d. Miwardi, Abu'l-1;Iasan •An b. Mu1}ammad: al-AI}Icilm al-,fui(lJniyya. ｍｾＬ＠ ｍｩｾＬ＠ ｍｾＬ＠ Erdogan: Gazneliler Devleti Tarihi. Ankara, 1989. Erdogan: -Simcuriler-. (Series of S articles, published as follows:) I: Tarih Dergisi 32 (1979), 71 - 88; H: Tarih EnsIitllsQ Dergisi 10/11 (1981), 91 - 6; HI: Tarih Dergisi 33 (1981), 115 - 32; n.d. 38 IV: Belleten Sayt 195, 49 (1985), 547 - 67; V: Tanh EnstiIflsIl Dergisi 13 (1983-7), 123 - 38. Mikhaylova, I.B.: Srednevekovyj Bagdod. Moscow, 1990. Mirkbwind: Rawt/Ill ｡ｬＭｾｱｦｉｊＮ＠ Ed. in 7 vols., Tehran, 1339/1961. Mottahedeh, Roy: Loyalty and Leodership in an Early Islamic Society. Princeton, 1980. Mul}ammad b. al-Munawwar: AsrDr als-tawlpd ft maqlJmlll al-laikh Aba &z ltd. Ed. Bahmanyir, Tehran, 132011941. Mu'in al-fuqari: Kitilb-i muJ/il-ZiJda. Ed. GulchIn-i rna'iIli, Tehran, 133911961. Muqaddasi, ｾ｡ｭｳｵ､ｩｮ＠ Abu 'Abdallah: K. al}san al-taqlJslm ft ma lrifat al-aqalDn. Ed. M.J. de Goeje, Leiden, 1906 (BGA : 8). Nadim: Fihrist. Translation Dodge, New York & London, 1970. Narshakhi, MuIJmnmod: TIJr1kh-i Bukhilra. Ed. Ri4awi, Tehran 1939. Ed. Scheler, Paris 1892. Translation Frye: The History of Bukhara. Cambridge, 1954. Notb, Albrecht: wDas Ribit der Almoraviden w• - In: Der Orient in tier Forschung (Festschrift O. Spies), Wiesbaden, 1967,499 - 511. Olson, Mancur: The Rise and Decline ofNations: Economic Growth. S/(lgj1tllion, and Social Rigidities. New Haven, 1982. Paul, JUrgen: Herrscher, Gemeinwesen, Venninler: Ostiran und Transoxanien in vonnongolischer Zeit. (forthcoming.) Paul, JOrgen: WThe Histories of Samarqand w • - In: Studia Iranica 2211 (1993), 69 - 92. Paul, Jiirgen: wNachrichten arabischer Geographen aus Mittelasien w • - In: Fragner, Bert G. & B. Hoffmann (eds.): Bamberger Mittelasienstudien. Berlin, 1994, 179 - 91. Pipes, Daniel: Slave Soldiers and Islam: the Genesis of a Military System. New Haven, 1981. 39 Poonawala, I. -Nasafr. - In: El2. Qandiyya (Arabic). Ms. Istanbul, Turban Valide 70 and Paris, ｂｩ｢ｬｯｾｵ･＠ Nationale, Arabe 6284. Qarshi, Jamil: MulIJI.IIlt21 ｡ｬＭｾｵｲＴｽＮ＠ Ms. St Petersburg, IV RAN C-2S6. ai-Rashid b. aI-Zubair aI-QicJi: K. al-dJuJJchJl 'ir wa 'l-tulJof. Ed. Mnl)ammad ｾ､｡ｉｬｩｨＬ＠ Kuwait 1959. Raspopova, V.I. Zhilishche Pentkhikenta. Moscow, 1990. Rokkan, Stein: -Dimensions of State Formation and Nation Building: A Possible Paradigm for Researcb on Variations within Europe-. - In: Tilly, Charles (ed.): 'I'M FomuJlion of National States in Westem Europe. Princeton, 1975, 562 - 600. SahmI, Jp.mza b. Yiisuf: TiJrtkh Jurjiln. Ed. N4am aI-dIn. HydarabadlD., 1950. Sam'ini, Abu Sa'd 'Abd ai-KarIm: HyderabadlD., 1962 - 82. al-Ansilb. 13 vols. Ed. Yam3ni, Shosban, Boaz: -The 'Politics of Notables' in Medieval Islam-. - In: Asian and African Studies 20 (1986), 179 - 215. Sbue, Vivienne: 17Je Reach of the Slale: Sketches of the Chinese Body Politic. Stanford, 1988. Spuler, Bertold: Iran infrilhisltlmischer Zeit. Wiesbaden, 1953. Taban, AbU Ja' far Mul].ammad b. Jarir: TIJrikh al-l'USul wa '/-multIk. 3 parts, Leiden, 1879-90. TilrIkh-i StstlJn. Ed. Malik aI-lu 'ara Bahar, Tehran, 1314/1935. Tilly, Charles: Coercion, CopittJl. and European States. AD 990- 1990. Princeton, 1990. 'Utbi, Abu Nqr: Tartkh yamrnr. Recension ManW: SariJ al-yamrnr al-musammil bi'I-fad1 al-wahbr 'alil tilrtkh Abt ｎｾｲ＠ al- 'Utbt. aI-Qabira 126811869 (2 vols.) - Persian translation Junadhaqaru: Tarjuma-yi tlJrlkh-i yamInt. Tehran, 1334/1955. 40 Weinberger t J.: -The Authorship of Two Twelfth Century Transoxanian Biographical Dictionaries-. - In: Arabica 33 (1986), 369 - 82. Woods, John: The Aqquyunlu. Minneapolis & Chicago, 1976. Va 'qUbi, AJ}mad b. Abi Va 'qub: K. al-buldlJn. Ed. M.J. de Goeje. Leiden, 1891 (BOA : 7). Viqiit, !ihib al-din Abu 'Abdallih: Mu Jam al-buldlln. S vols. Bairiit, 137476/19SS-S7. VOsuf KhiM ｾｩ｢Ｚ＠ QUladgu BiDg. Transl. R. Dankoff: Wudom of Royal Glory. Chicago & London, 1983. Original: Vusuf Has lDcib: KUladgu BlUg. I: Metin. ed. Rqid Rahmeti Arat. istanbul, 1947. n: TercOme (Rqid Rahmeti Arat). Ankara, 19S9.