Notes

Introduction

1For general surveys of Gardīzī and his work, see C.E. Bosworth, ‘Early sources for the history of the first four Ghaznavid sultans (977–1041)’, IQ VII (1963), pp. 8–10; EIr art. ‘Gardīzī’ (C.E. Bosworth).

2Ḥabībī’s text, p. 252.

3V. Minorsky, ‘Gardīzī on India’, BSOAS XII (1947–9), p. 625 and n. 3. However, in his chapter 19 ‘On the sciences and tenets (maՙārif) of the Indians’, Gardīzī’s expressly acknowledged source is the Samanid author Abū ՙAbdallāh Jayhānī (presumably Abū ՙAbdallāh Muḥammad b. Aḥmad b. Naṣr, vizier to the Samanid Amir Naṣr (II) b. Aḥmad (II), see above, Part Three, p. 55 and n. 18), with material going back to the tradition of the mid-third/ninth century geographer Ibn Khurradādhbih. See Minorsky, op. cit., p. 626; Louise Marlow, ‘Some classical Muslim views of the Indian caste system’, MW LXXXV (1995), pp. 16–17.

4Fragner, ‘The concept of regionalism in historical research on Central Asia and Iran (a macro-historical interpretation)’, in Studies on Central Asian History in Honor of Yuri Bregel, ed. Devin DeWeese (Bloomington, Ind., 2001), pp. 244–7; and cf. Julie S. Meisami, Persian Historiography to the End of the Twelfth Century (Edinburgh, 1999), p. 68.

5See for al-Sallāmī and his work, W. Barthold, Turkestan down to the Mongol Invasion, 3rd ed. (London, 1968), pp. 10–11, 21; Fuat Sezgin, Geschichte des arabischen Schrifttums, I, Leiden 1967, 352 no. 5; and EI² art. ‘al-Sallāmī’ (C.E. Bosworth).

6Material from al-Sallāmī seems also to have been used in the seventh/thirteenth century by Ibn Khallikān and ՙAṭā’-Malik Juwaynī; see Barthold, op. cit., p. 10.

7See below, Ḥabībī’s text, p. 131, and Barthold, op. cit., p. 7.

8Op. cit., pp. 69ff.

9Ibid., pp. 74–5.

10Ed. Nazim, pp. 61–2, ed. Ḥabībī, pp. 173–4.

11A.C.S. Peacock, ‘ՙUtbī’s al-Yamīnī: patronage, composition and reception’, Arabica LIV (2007), pp. 519–20.

12See for them, C.E. Bosworth, The Later Ghaznavids: Splendour and Decay. The Dynasty in Afghanistan and Northern India 1040–1186 (Edinburgh, 1977), pp. 25–47.

13Ibid., pp. 41–7.

14On Gardīzī as a source for Khurasanian history, see Barthold, Turkestan, pp. 20–1, and as one for Ghaznavid history specifically, see Bosworth, ‘Early sources for the history of the first four Ghaznavid sultans’, loc. cit. Gardīzī’s place in the historical writing of the time is penetratingly discussed by Meisami in op. cit., pp. 66–79.

15See C.A. Storey, Persian Literature. A Bio-bibliographical Survey (London, 1927–53), Vol. I/2, pp. 65–7.

16Ibid., p. 72 n. 5.

17Cf. Minorsky, ‘Gardīzī on India’, pp. 625–6. Attempts at elucidating the Turkish names in Bayhaqī have been made by Bosworth in his ‘Notes on some Turkish names in Abu ’l-Faḍl Bayhaqī’s Tārīkh-i Masՙūdī’, Oriens XXXVI (2001), pp. 299–313, and his ‘Further notes on the Turkish names in Abu ’l-Faḍl Bayhaqī’s Tārīkh-i Masՙūdī’, to appear in a Festschrift for Dr Farhad Daftary.

18Bahār, Sabk-shināsī yā tārīkh-i taṭawwur-i nathr-i fārsī (Tehran, 1337/1958), Vol. 2, p. 50.

19G. Lazard, La langue des plus anciens monuments de la prose persane (Paris, 1963), pp. 71–3.

20See e.g. Ḥabībī’s text, p. 133, regarding Hārūn al-Rashīd’s death.

21In the present translation, these are in any case usually omitted.

22Ḥabībī’s text, p. 131.

23Cf. Bosworth, The History of Beyhaqi (see above, pp. 9–10), Vol. 1, Introduction, pp. 59–60, 72.

24I am grateful to Mr Mel Dadswell for providing me with an English translation of Barthold’s introduction to his section of this work on Gardīzī.

25For surveys of the printed editions of Gardīzī’s work, see Storey, op. cit., Vol. I, pp. 66, 1229 (who could at the time only list the edition of Nazim and a Tehran one of the mid-1930s), supplemented and updated by Yuri E. Bregel, Persidskaya literatura, biobibliograficheskii obzor (Moscow, 1972), Vol. I, pp. 288–9, noting subsequent Tehran prints up to and including that of Ḥabībī. There is also much information in Lazard, loc. cit. Concerning the most recent edition by Riḍāzāda Malik, see below and n. 27.

26Czeglédy, ‘Gardīzī on the history of Central Asia’, AOHung, XXXVII (1973), 257–8; Martinez, ‘Gardīzī’s two chapters on the Turks’, AEMAe II (1982), pp. 109–12. It may be noted that, as well as Martinez’s English translation, facsimile texts of the relevant parts of the manuscripts and discussion of Gardīzī’s two sections on the Turks, there is now a German translation of and commentary on these passages by H. Cöckenjan and I. Zimonyi in their Orientalische Berichte über die Völker Osteuropas und Zentralasiens im Mittlealter. Die Ğayhānī-Tradition, Veröffentlichungen der Societas Uralo-Altaica, Bd. 54 (Wiesbaden, 2001), pp. 95–190 (I am grateful to Dr Pavel B. Lurye for this reference).

27This became known to me only from a review article by Muḥammad Gulbun in Āyīna-yi Mirāth/Mirror of Heritage, NS V/1–2 (Tehran, Spring-Summer, 2007), pp. 367–75. However, work on this present translation and its commentary was virtually complete when Dr Farhad Daftary kindly procured for me from Tehran, after some difficulty, a copy of Riḍāzāda Malik’s book, and in the short time available a thoroughgoing comparison of Ḥabībī’s and Riḍāzāda Malik’s two texts has not been possible.

28It is Riḍāzāda Malik’s restored numbering of the component chapters, and their headings, which is followed in the translation below.

29Storey, op. cit., Vol. I, p. 66.

30I am most grateful to M. Étienne de la Vaissière for bringing Arends’s work to my notice in the first place and for providing me with a photocopy of it, and also grateful to Mr Mel Dadswell and his expertise in Russian bibliography for help with tracing the book’s history up to its publication in 1991.

31Hodivala, Studies in Indo-Muslim History. A Critical Commentary on Elliot and Dowson’s History of India as Told by its Own Historians, 2 vols. (Bombay, 1937–57); Ahmad, ‘A critical examination of Bayhaqī’s narration of the Indian expeditions during the reign of Masՙud of Ghazna’, in Yād-nāma-yi Abu ’l-Faḍl-i Bayhaqī, ed. Mashhad University Faculty of Letters and Human Sciences (Mashhad, 1350/1971), English section, pp. 34–83.

Part One: The Arab Governors

1On this division of the Sasanid realm into ‘quadrants’, see C. Brunner, ch. ‘Geographical and administrative divisions: settlements and economy’, in CHIr, Vol. 3/2, The Seleucid, Parthian and Sasanian Periods, ed. Ehsan Yarshater (Cambridge, 1983), pp. 747ff.

2As noted by Gardīzī, above, p. 14, this southerly route was the normal one to Sistan and Khurasan at this early period, since what became the more usual, northerly route along the southern rim of the Elburz chain was menaced by the un-Islamised mountain peoples, Daylamīs, Jīlīs, etc. to the north. See C.E. Bosworth, Sīstān under the Arabs, from the Islamic Conquest to the Rise of the Ṣaffārids (30–250/651–864) (Rome, 1968), pp. 13–15.

3See on this outstanding warrior, H.A.R. Gibb, The Arab Conquests in Central Asia (London, 1923), pp. 15–16; J. Walker, A Catalogue of the Muhammadan Coins in the British Museum. Volume I. A Catalogue of the Arab-Sassanian Coins (Umaiyad Governors in the East, Arab-Ephthalites, ՙAbbāsid Governors in Ṭabaristān and Bukhārā) (London, 1941), pp. xlvi–xlvii; M.A. Shaban, The ՙAbbāsid Revolution (Cambridge, 1970), pp. 16–25; EI² art. ‘ՙAbd Allāh b. ՙĀmir’ (H.A.R. Gibb); EIr art. ‘ՙAbdallāh b. ՙĀmer’ (J. Lassner). Al-Yaՙqūbī, Kitāb al-Buldān, French tr. G. Wiet, Les pays (Cairo, 1937), begins his list of the governors of Khurasan (pp. 114–38) with Ibn ՙĀmir and goes up to the last Tahirid governor in Khurasan, Muḥammad b. Ṭāhir. Ḥamza al-Iṣfahānī, Ta’rīkh Sinī mulūk al-arḍ wa ’l-anbiyā’ (Beirut, 1961), pp. 160–72, who dismisses the governors of Khurasan in the Umayyad period as contemptible tyrants, begins essentially with Abū Muslim and has a fair amount of detail on the governors of the early ՙAbbasids, the Tahirids and Saffarids, but does little more than to list the names of the earlier Samanids, ending with ՙAbd al-Malik (I) b. Nūḥ (I), the author’s contemporary.

4See on this Inner Asian people, foes of the Sasanids and Arabs alike, EI² art. ‘Hayāṭila’ (A.D.H. Bivar); EIr art. ‘Hephthalites’ (A.D.H. Bivar).

5G. Le Strange, The Lands of the Eastern Caliphate (Cambridge, 1905), p. 405.

6Riḍāzāda Malik’s reading of sipanj for an unclear consonant ductus seems better than Ḥabībī’s basīj ‘arms, gear, equipment’.

7According to al-Balādhurī, Futūḥ al-buldān, ed. M.J. de Goeje (Leiden, 1866), pp. 409–10, Umayr was the first to settle Arabs at Merv and thereby begin the process whereby the Arabs acquired taxable land in the Merv oasis, settling down and intermarrying with the Persian population. There thus began a process of assimilation there very different from elsewhere, in which the Arabs kept themselves apart from the indigenous population by keeping within fortified garrison cities and encampments. Gardīzī, however, attributes this programme of Arab settlement in the Merv oasis to Muՙāwiya’s governor Saՙīd b. ՙUthmān (see above, p. 18), and this seems more likely. Cf. EI² art. ‘Marw al-Shāhidjān’ (A.Yu. Yakubovskii and C.E. Bosworth).

8Conjecture of Ḥabībī of a possible reading for the text’s b.s.tām.

9For this invasion of Sistan in 31/651–2, see Bosworth, Sīstān under the Arabs, pp. 16–17.

10For these places along the Pilgrimage route from Iraq to Medina and Mecca, see Yāqūt, Muՙjam al-buldān (Beirut, 1374–6/1955–7), Vol. 1, p. 414, Vol. 2, p. 111, Vol. 5, pp. 255–6, 278–9; Abdullah Al-Wohaibi, The Northern Hijaz in the Writings of the Arab Geographers 800–1150 (Beirut, 1973), pp. 102–12 (for Juḥfa).

11In early Islamic times, kharāj seems to have been the land-tax paid by the non-Muslim dhimmīs in conquered lands, whilst the Muslims there paid ՙushr on their land. The subjects who were to pay the kharāj in the Merv oasis were obviously, at this time, substantially the Persian, non-Muslim population. Only in ՙAbbasid times does kharāj become the general, official term for land-tax. See F. Løkkegaard, Islamic Taxation in the Classic Period, with Special Reference to Circumstances in Iraq (Copenhagen, 1950), pp. 72ff.

12Lit. ‘he turned their heads’, sar-īshān bar gardānīd, unless one should read sirr-īshān . . ., with a meaning like ‘he revealed their secrets’.

13In pre-Islamic Bedouin society, fay’ (meaning something like ‘what is brought back’, i.e. to God and the Muslim community) was a general term for ‘booty, plunder’. In early Islamic times it took on the specialised meaning of ‘income from the conquered lands whose revenue went to the state, which then paid back a proportion of it to the Arab warriors’ (see Løkkegaard, op. cit., pp. 32ff.; EI² art. ‘Fay’’ (Løkkegaard). However, apart from this instance where the words of al-Ḥasan are quoted, Gardīzī uses fay’ with the older, general meaning of ‘plunder’, hence as a synonym for ghanīma.

14Sūrat al-Anbiyā’, XXI, vol. 111.

15Following Riḍāzāda Malik’s reading sarmā rather than Ḥabībī’s ānjā.

16There may be some confusion here with the Arab historians’ mention of Mujāshiՙ’s pursuit of the fugitive Sasanid emperor Yazdagird III; see Bosworth, Sīstān under the Arabs, pp. 15–16.

17For Ziyād in the East, see J. Wellhausen, The Arab Kingdom and Its Fall, English tr. Margaret G. Weir (Calcutta, 1927), pp. 119–30; Gibb, The Arab Conquests in Central Asia, pp. 16–17; Walker, A Catalogue of the Muhammadan Coins in the British Museum, I, pp. xlii–xlv; Bosworth, Sīstān under the Arabs, pp. 20–1; Shaban, The ՙAbbāsid Revolution, pp. 29–34; EI² art. ‘Ziyād b. Abīhi’ (I. Hasson).

18For Muhallab, see EI² art. ‘al-Muhallab b. Abī Ṣufra’ (P. Crone).

19Ḥabībī, p. 105 n. 9, thinks that this may refer to the Oxus.

20See on him, Walker, op. cit., pp. xlvii–xlix; Bosworth, op. cit., pp. 42–4; EI² art. ‘ՙUbayd Allāh b. Ziyād’ (C.F. Robinson).

21This seems the best translation here, given that Muhallab’s opponents here were Sogdians rather than Persians.

22Gibb, op. cit., pp. 17–19.

23See on Saՙīd, Shaban, The ՙAbbāsid Revolution, pp. 37–8.

24Gibb, op. cit., pp. 17–19; Shaban, op. cit., pp. 38–9. See also above, p. 15 and n. 7.

25Thus in Nafīsī’s text (chand sāl) for the reading of the mss.’ ad sāl ‘a hundred years’ adopted by Ḥabībī and Riḍāzāda Malik. Al-Yaՙqūbī, Buldān, tr. Wiet, p. 148, says that ՙAbd al-Raḥmān functioned in Khurasan for four months only, and was then dismissed for his feebleness and supine attitude.

26See on him, Gibb, op. cit., pp. 21–2; Walker, op. cit., pp. xlix–li; Bosworth, Sīstān under the Arabs, pp. 44–5, 48–9; Shaban, op. cit., pp. 39–41; EI² art. ‘Salm b. Ziyād b. Abīhi’ (C.E. Bosworth).

27This was the Queen-Mother (khātūn = ‘lady’, Sogdian γwt’ynh) acting as regent for the infant son of the ruler of Bukhara, the subsequent Bukhār-Khudāh Ṭughshāda. See Gibb, op. cit., pp. 17–22; Narshakhī, Tārīkh-i Bukhārā, English tr. R.N. Frye, The History of Bukhara (Cambridge, Mass., 1954), pp. 9–10.

28A mawlā of this Arab governor was Ruzayq, progenitor of the Tahirid governors of Khurasan in the third/ninth century, see above, pp. 42ff.

29On Ṭalḥa in Sistan, see Walker, op. cit., p. liv; Bosworth, Sīstān under the Arabs, p. 45.

30For ՙAbdallāh’s lengthy governorship, during which he, as a Sulamī, headed the Qaysī ascendancy in Khurasan and struck both dirhams and dinars in his own name, see al-Yaՙqūbī, Buldān, tr. Wiet, pp. 120–1; Barthold, Turkestan, p. 184; Wellhausen, The Arab Kingdom, pp. 416–21; Walker, op. cit., pp. lii–liii and index at p. 221; EI² art. ‘ՙAbd Allāh b. al-Khāzim’ (H.A.R. Gibb).

31Words added here by Nafīsī to complete the sense.

32Phrase supplied here from al-Yaՙqūbī, Buldān. The Khurāsāniyān mentioned here are, of course, the Arab troops and settlers in Khurasan rather than the population at large, as are likewise the ahl-i Khurāsān that Gardīzī often mentions below in various relationships with the governors and caliphs, becoming after the ՙAbbasid Revolution one of the mainstays of the regime. See on them Farouk Omar, ‘The composition of ՙAbbāsid support in the early ՙAbbāsid period’, in his ՙAbbāsiyyāt. Studies in the History of the Early ՙAbbāsids (Baghdad, 1976), pp. 42–5.

33Various words supplied here to complete the sense.

34On the roles of these two tribal leaders in Khurasan, Baḥīr b. Warqā’ and Bukayr b. Wishāḥ, and their involvement with ՙAbdallāh b. Khāzim, see Wellhausen, The Arab Kingdom, pp. 418ff.

35There is some uncertainty in the sources concerning the exact form of his name; he is not mentioned in al-Yaՙqūbī’s list of the governors.

36Baḥīr was from Ṣarīm, a sub-clan of Saՙd of Tamīm; see Wellhausen, op. cit., p. 421.

37See on him al-Yaՙqūbī, Buldān, tr. Wiet, p. 121; Wellhausen, op. cit., pp. 421–2, 426; Walker, op. cit., pp. lvii–lviii; Bosworth, Sīstān under the Arabs, pp. 49–50; Shaban, The ՙAbbāsid Revolution, pp. 44–6.

38For al-Ḥajjāj’s policy in the East, see Wellhausen, op. cit., pp. 426ff.; Shaban, op. cit., pp. 53–75; EI² art. ‘al-Ḥadjdjādj b. Yūsuf’ (A. Dietrich).

39For the sons of al-Muhallab, see EI² art. ‘Muhallabids’ (P. Crone).

40Ratbīl or Rutbīl or Zunbīl was more properly the title of the indigenous rulers of Zamīndāwar and Zābulistān, to the southeast of Kabul; see on them below, Part Two, n. 18.

41For `Ibn al-Ashՙath and his rebellion, see Bosworth, Sīstān under the Arabs, pp. 55–63; EI² art, ‘Ibn al-Ashՙath’ (L. Veccia Vaglieri).

42Ḥabībī reads an undotted consonant ductus as bakhtiyān, ‘Bactrian camels’. But Riḍāzāda interprets it with greater plausibility as najībān ‘camels of noble breed’, hence ‘swift-running’, supported by the fact that al-Yaՙqūbī, in his History, when recounting this incident, speaks of najā’ib; this translation is accordingly followed here.

43On Qutayba’s governorship, see in general Barthold, Turkestan, pp. 184–7; Wellhausen, The Arab Kingdom, pp. 429–44; Gibb, The Arab Conquests in Central Asia, pp. 29–56; Shaban, The ՙAbbāsid Revolution, pp. 63–75; EI² art. ‘Ḳutayba b. Muslim’ (C.E. Bosworth).

44See for this place below, Part Four, n. 15.

45tāzagī. Tāzīg was the Middle Persian term for ‘Arab’; see D.N. MacKenzie, A Concise Pahlavi Dictionary (Oxford, 1990), p. 83.

46On the fall of Qutayba, see Wellhausen, op. cit., pp. 439–43; R. Eisener, Zwischen Faktum und Fiktion. Eine Studie zum Umayyadenkalifen Sulaimān b. ՙAbdalmalik und seinem Bild in den Quellen (Wiesbaden, 1987), pp. 91–7.

47On Wakīՙ’s tenure of power, see Wellhausen, op. cit., pp. 442, 444–6; Eisener, op. cit., pp. 97–8. His governorship does not appear as such in al-Yaՙqūbī’s list, and seems to have been a provisional one; this historian in fact says (Buldān, tr. Wiet, p. 123) that Wakīՙ expected to be made governor, but the new caliph Yazīd b. ՙAbd al-Malik preferred to give the post to Yazīd b. Muhallab, thus temporarily restoring the ascendancy of the Yemenis in Khurasan.

48i.e. that laid down in the Qur’ān; see EI² art. ‘Ḥadd’ (B. Carra de Vaux, J. Schacht).

49On this second governorship of Yazīd, see al-Yaՙqūbī, Buldān, tr. Wiet, pp. 123–4; Wellhausen, op. cit., pp. 445–8; Shaban, The ՙAbbāsid Revolution, pp. 77–83.

50dar-i āhanīn. This must refer to the narrow valley and gorge connecting the upland plain of Arghiyān and Juwayn in northern Khurasan with the Caspian lowlands of Gurgān, in which flows the Gurgān river; the valley is called in the Ḥudūd al-ՙālam, English tr. V. Minorsky, 2nd ed. (London, 1970), p. 64, comm. p. 200, the Dīnār-zārī, now the Dahana-yi Gurgān.

51This is the ‘Ṣūl the Turk’ of al-Ṭabarī, Ta’rīkh al-Rusul wa ‘l-mulūk, ed. M.J. De Goeje et al. (Leiden, 1879–1901), Secunda series, p. 1323, English tr. D.S. Powers, The History of al-Ṭabarī. Vol. XXIV. The Empire in Transition (Albany, 1989), p. 48, located by him in Dihistān, to the north of Gurgān in the steppelands on the eastern shore of the Caspian Sea. The Turks in question were possibly Oghuz who had infiltrated southwards from the steppes to the north of the Aral Sea. See Barthold, A History of the Turkmen People, in Four Studies on the History of Central Asia, English tr. V. and T. Minorsky (Leiden, 1962), Vol. 3, pp. 87–8. Members of the family of Ṣūl, now Islamised, were to have a career in ՙAbbasid service and to produce a celebrated literary figure of the fourth/tenth century, Abū Bakr Muḥammad al-Ṣūlī. On the name Ṣūl/Sol and its possible meaning, see L. Rásonyi and I. Baski, Onomasticon turcicum. Turkic Personal Names as Collected by László Rásonyi (Bloomington, Ind., 2007), Vol. 2, pp. 665–6.

52For Yazīd’s military operations in Gurgān, see Eisener, op. cit., pp. 100–1.

53See on his term of office, al-Yaՙqūbī, Buldān, tr. Wiet, p. 124; Barthold, Turkestan, p. 188; Wellhausen, The Arab Kingdom, pp. 450–2; Shaban, The ՙAbbāsid Revolution, pp. 86–7; EI² art. ‘al-Djarrāḥ b. ՙAbd Allāh’ (D.M. Dunlop). Jarrāḥ has a claim to fame in the history of Arabic palaeography and diplomatic as the addressee of the only Arabic document from the Eastern Islamic world extant from this early period, sc. a letter found at Mount Mugh in what is now the Tajik Republic. See Geoffrey Khan, Studies in the Khalili Collection. Volume V. Arabic Documents (London, 2007), p. 15 (Khan here edits and translates a cache of Arabic documents from the mid-eighth century AD, apparently emanating from northern Afghanistan; these have greatly swelled the previously very exiguous extant corpus of early Arabic documents from the East).

54i.e. the great-grandson of the Prophet’s uncle al-ՙAbbās. Gardīzī is here registering the beginning of the ՙAbbasid daՙwa in Kufa when Muḥammad b. ՙAlī took over the claims of Abū Hāshim, son of ՙAli’s son Muḥammad b. al-Ḥanafiyya. See Shaban, op. cit., pp. 150–1.

55For Saՙīd, an Umayyad from the Abu ’l-ՙĀṣ branch of the clan, and nicknamed Khudhayna ‘Little Lady’ by the Khurasanian troops, who regarded his mildness as misplaced, see Barthold, op. cit., pp. 188–9; Wellhausen, op. cit., pp. 451–2; Shaban, op. cit., pp. 99–101.

56Following Nafīsī’s reading, mujāmalat.

57ՙUmar was indeed governor for Yazīd, but was replaced by Hishām on his accession in 105/724. See EI² art. ‘Ibn Hubayra’ (J.-C. Vadet).

58Khālid was governor of Iraq and the East for the greater part of Hishām’s twenty-years’ caliphate, with his brother Asad as governor in Khurasan for almost as long, dying in 120/738. See Wellhausen, The Arab Kingdom, pp. 455–6; Gibb, The Arab Conquests in Central Asia, pp. 67–89; Shaban, The ՙAbbāsid Revolution, pp. 106–27; EI² art. ‘Asad b. ՙAbd Allāh’ (H.A.R. Gibb); EI² art. ‘Khālid b. ՙAbd Allāh al-Ḳasrī’ (G.R. Hawting).

59Ḥabībī interprets the text here as shumā hamī irjāf afgandīd.

60See on him Barthold, Turkestan, pp. 189–90; Wellhausen, op. cit., pp. 456–9; Shaban, op. cit., pp. 109–12.

61This is apparently the Abū Muzāḥim of Arabic historical sources, the Qaghan of the Türgesh or Western Turk empire at this time, whose actual name is only known from its form in the Chinese annals, Su-lu. See Barthold, op. cit., pp. 190–1; Gibb, The Arab Conquests in Central Asia, pp. 73ff.

62khārijī, not, however, to be taken here in the sectarian sense of ‘Khārijite’. Al-Ḥārith seems to have been in theology a Murji’ite and, in practice, an ascetic calling for just rule according to the Qur’ān and Sunna, as noted by Gardīzī below, and for allegiance to one acceptable to the whole Muslim community (al-riḍā). His anti-government activity in Khurasan continued till his death in 128/746. See Wellhausen, The Arab Kingdom, pp. 459–98; Barthold, Turkestan, pp. 190–3; EI² art. ‘al-Ḥārith b. Suraydj’ (M.J. Kister).

63See on him Barthold, op. cit., p. 191; Wellhausen, op. cit., pp. 461–2, 466–7; Shaban, op. cit., pp. 118–21.

64On this second spell of office for Khālid and Asad as his deputy, see Barthold, loc. cit.; Wellhausen, op. cit., pp. 467–74; Shaban, The ՙAbbāsid Revolution, pp. 121–7.

65This lay in the westernmost part of Khurasan, to the west of Nishapur and adjoining the province of Qūmis; see Le Strange, The Lands of the Eastern Caliphate, p. 430.

66i.e. for a hostel or caravanserai.

67See on him Barthold, op. cit., pp. 192–4; Wellhausen, op. cit., pp. 474–91, 523–40; Gibb, op. cit., pp. 89–93; Shaban, op. cit., pp. 127–37, 159–61; E.L. Daniel, The Political and Social History of Khurasan under Abbasid Rule, 747–820 (Minneapolis and Chicago, 1979), pp. 44–5, 54–8; Moshe Sharon, Black Banners from the East II. Revolt. The Social and Military Aspects of the ՙAbbāsid Revolution (Jerusalem, 1990), pp. 34–7, 42–7, 63, 107ff., 147ff.; EI² art. ‘Naṣr b. Sayyār’ (C.E. Bosworth).

68Naṣr introduced significant financial reforms in Khurasan in 121/739 aimed at redressing the grievances of the Arab settlers in the Merv oasis (Gardīzī’s ahl-i Khurāsān here) against the local Persian dihqāns who had been collecting taxation from the oasis. See D.C. Dennett, Conversion and the Poll Tax in Early Islam (Cambridge, Mass, 1950), pp. 124–8; Shaban, op. cit., pp. 129–30.

69See on this ՙAlid, a descendant on his mother’s side of Muḥammad b. al-Ḥanafiyya, Daniel, op. cit., pp. 38–9; Sharon, Black Banners from the East. The Establishment of the ՙAbbāsid State – Incubation of a Revolt (Jerusalem/Leiden, 1983), pp. 175–9, 181; EI² art. ‘Yaḥyā b. Zayd’ (W. Madelung).

70See on these agents, Shaban, The ՙAbbāsid Revolution, pp. 151–2; Sharon, Black Banners from the East, pp. 188–97; Sharon, Revolt. The Social and Military Aspects of the ՙAbbāsid Revolution, pp. 20–1; EI² art. ‘Naḳīb. I. In Early Islamic History’ (C.E. Bosworth).

71As Ḥabībī remarks, n. 11, copyists seem to have split the name of the fifth naqīb into two separate persons, with the two brothers given as the twelfth one making up the full number.

72Text, z.r.y.q. This is the first appearance in Gardīzī’s pages of a family – probably of mawlā origin in Basra, and of unknown ethnic affiliation – which was to have a glorious future in both Iraq and Khurasan. Both Ṭalḥa and his brother Muṣՙab were, as appears here, involved with the ՙAbbasid daՙwa, and the patronage of the new caliphal dynasty subsequently favoured the family’s rise, with Muṣՙab’s grandson Ṭāhir Dhu ’l-Yamīnayn becoming al-Ma’mūn’s commander during the civil war with al-Amīn (see above, pp. 40–1) and founding the line of Tahirid governors of Khurasan. See Mongi Kaabi, ‘Les origines ṭāhirides dans la daՙwa ՙabbāside’, Arabica, XIX (1972), pp. 145–64; Kaabi, Les Ṭāhirides au Hurāsān et en Iraq (III^ème H./IX^ème J.-C.) (Tunis, 1983), Vol. 1, pp. 62–3.

73Text, ’.n.y.sū; cf. Ḥabībī’s n. 5 and EI² art. ‘Yaḥyā b. Zayd’ (W. Madelung).

74This explanation, and the more usual (and plausible) one that Marwān was called al-Ḥimār because of the wild ass’s proverbial hardihood and endurance, are given by al-Thaՙālibī in his Laṭā’if al-maՙārif, English tr. C.E. Bosworth, The Book of Curious and Entertaining Information (Edinburgh, 1968), p. 61.

75See on Jahm and his doctrines, W. Montgomery Watt, The Formative Period of Islamic Thought (Edinburgh, 1973), pp. 145–8; EI² arts. ‘Djahm b. Ṣafwān’ and ‘Djahmiyya’ (Montgomery Watt).

76Shaybān b. Salama had been an adherent of the Khārijite rebel in Iraq and al-Jazīra, al-Ḍaḥḥāk b. Qays, and had fled from there to Khurasan, where he became an associate of Judayՙ al-Kirmānī until he was killed by Abū Muslim in 130/748. See Daniel, The Political and Social History of Khurasan, pp. 78–9.

77This was presumably the first protective ditch constructed at Mākhān or Mākhwān (see below, n. 80) by Abū Muslim as a defence against possible attack by Naṣr, see Daniel, op. cit., pp. 52–3. He then had a second ditch made at Gīrang or Kīrang, which lay on the Murghāb river, presumably to disrupt communication between Naṣr and his supporters in Marw al-Rūd; see Le Strange, The Lands of the Eastern Caliphate, p. 400, and Ḥudūd al-ՙālam, tr. Minorsky, p. 105, comm. p. 328.

78The origins of Abū Muslim are in fact very obscure, making him an enigmatic figure in Islamic history; see Sharon, Black Banners from the East, pp. 203–7. It is not clear whether he was a free man or a mawlā, or of Persian ethnos, as is usually assumed.

79For Qaḥṭaba, the second most important person of the ՙAbbasid daՙwa in Khurasan after Abū Muslim, and one of the nuqabā’ (see above, pp. 27–8), see P. Crone, Slaves on Horses. The Evolution of the Islamic Polity (Cambridge, 1980), p. 188; EI² art. ‘Ḳaḥṭaba’ (M. Sharon).

80This is described by Yāqūt as a village three parasangs from Merv. Various traditions make Abū Muslim’s father, as Abū Muslim himself, a native of Mākhān. See Irène Mélikoff, Abū Muslim, le <> du Khorasan, dans la tradition épique turco-iranienne (Paris, 1962), p. 93 n. 5.

81On the apocalyptic and prophetic sayings which seem to have been put into circulation by what Sharon has called the ‘ՙAbbasid official propaganda machine’, see his Black Banners from the East, pp. 87ff.

82In Bīrūnī’s original (see next note), ṣalāt.

83This passage about Bihāfarīd is clearly taken from al-Bīrūnī’s al-Āthār al-bāqiya ՙan al-qurūn al-khāliya, ed. E. Sachau (Leipzig, 1878), English tr. E. Sachau, The Chronology of Ancient Nations (London, 1879). See on this Zoroastrian reformer, Gh.H. Sadighi, Les mouvements religieux iraniens au II^e et au III^e siècle de l’Hégire (Paris, 1938), pp. 111–31; B. Scarcia Amoretti, ch. ‘Sects and heresies’, in CHIr, Vol. 4, The Period from the Arab Invasion to the Saljuqs, ed. R.N. Frye (Cambridge, 1975), pp. 489–90; EI² art. ‘Bih’āfrīd b. Farwardīn’ (D. Sourdel). Meisami, Persian Historiography, pp. 74–5, noted Gardīzī’s concern – as shown also by other historians in medieval Islam – with the movements of various pseudo-prophets and religious reformers, the implication being that it is the ruler’s duty to eradicate such dissidence and to support the true faith.

84Actually at Ushmūnayn in Upper Egypt; see EI² art. ‘Marwān II’ (G.R. Hawting).

85On this pro-ՙAlid revolt of Sharīk b. Shaykh al-Mahrī, centred on Bukhara, see Barthold, Turkestan, pp. 194–5.

86The background to this is explained by al-Ṭabarī, Ta’rīkh, Secunda series, p. 1891, English tr. J.A. Williams, The History of al-Ṭabarī. Vol. XXVII. The ՙAbbāsid Revolution (Albany, 1985), p. 211: that al-Saffāḥ had been fearful of an army of several thousand Khurasanian troops descending on Iraq and, on the plea that the Pilgrimage Road across the Arabian desert of Najd could not feed and supply such a throng, instructed Abū Muslim to bring 1,000 troops only. See also H. Kennedy, The Early Abbasid Caliphate. A Political History (London and Totowa, N.J., 1981), p. 54.

87On ՙAbdallāh b. ՙAlī’s revolt with the support of the Syrian army, see J. Lassner, The Shaping of ՙAbbāsid Rule (Princeton, 1980), pp. 35–8; Kennedy, op. cit., pp. 58–60.

88Reading thus, following Nafīsī’s text rather than the completely implausible Ḥarra of Ḥabībī and Riḍāzāda Malik, Ḥīra being an ancient centre of Christianity in central Iraq, where this tarsā’ī was presumably a monk or anchorite.

89This seems to be a reference to the dispute over the plunder captured from ՙAbdallāh b. ՙAlī’s army when Abū Muslim defeated it at Nisibīn. According to al-Ṭabarī, Ta’rīkh, Tertia series, pp. 102–3, English tr. Jane D. McAuliffe, The History of al-Ṭabarī. Vol. XXVIII. ՙAbbāsid Authority Affirmed (Albany, 1995), p. 23, for the distribution of this booty, the caliph sent his own agent, the mawlā Abū Khaṣīb (on whom see Crone, Slaves on Horses, p. 190), with Abū Muslim protesting that the caliph was only entitled to a fifth (khums) of captured spoils. See also Kennedy, op. cit., p. 61.

90Abū Mujrim, punning on Abū Muslim’s name.

91Salama, Manṣūr’s mother, being a Berber slave.

92The detailed Arabic historical sources (e.g. al-Yaՙqūbī, al-Ṭabarī) list further petty accusations hurled at Abū Muslim by the caliph.

93sar-i kas; Nafīsī’s text has sarhang ‘senior officer’; al-Ṭabarī has ṣāḥib al-ḥaras. See on ՙUthmān b. Nahīk, Crone, Slaves on Horses, p. 189.

94Meisami, Persian Historiography, pp. 70–1, partially translates these exchanges between al-Manṣūr and Abū Muslim, noting the apocalyptic and prophetic overtones in this story of the unfolding of the ՙAbbasid Revolution and its consequences for a protagonist like Abū Muslim.

95Abū Dāwūd had been one of the original twelve naqībs (see above, p. 26) and one of Abū Muslim’s most trusted lieutenants. See on him al-Yaՙqūbī, Buldān, tr. Wiet, p. 129; Daniel, The Political and Social History of Khurasan, pp. 86, 132, 158–9.

96There are, however, in the sources differing accounts of his death, including one which makes Abū Dāwūd an object of suspicion, because of his previous closeness to Abū Muslim, for al-Manṣūr, who accordingly had him murdered. See Daniel, op. cit., pp. 158–9, who makes the point that the killing of Abū Muslim ushered in a period of sharp factional strife in Khurasan between the erstwhile supporters of Abū Muslim and the partisans of al-Manṣūr and the ՙAbbasids.

The ‘Wearers of White’ (in Arabic contexts appearing as the Mubayyiḍa) mentioned here by Gardīzī must have been a ghulāt group, one of several who claimed to be avenging the death of Abū Muslim and whose activity in Khurasan antedated that of the rebel al-Muqannaՙ (‘The Veiled One’) of some twenty years later (see above, p. 34). This Saՙīd Jawlāh (?) does not seem to be mentioned in other sources, but there was at this time ghulāt agitation in Khurasan led by one Isḥāq al-Turk (the latter component of his name does not seem necessarily to imply Turkish ethnicity, but that he had worked among the Turks on the fringes of Transoxania); see concerning Isḥāq, Sadighi, Les mouvements religieux iraniens, pp. 150–4; Scarcia Amoretti, ch. ‘Sects and heresies’, pp. 496–7; and Daniel, op. cit., p. 132.

97Ḥamza al-Iṣfahānī, Ta’rīkh Sinī mulūk al-arḍ wa ’l-anbiyā’, p. 162, mentions here with some diffidence (‘He [sc. God] knows best the truth of matters’) that the commander of the police guard, one Abū ՙIṣām ՙAbd al-Raḥmān b. Salīm, functioned as governor for a year and a month after Abu Dāwūd’s death.

98See on him and his subsequent revolt, Yaՙqūbī, Buldān, tr. Wiet, pp. 129–30; Daniel, The Political and Social History of Khurasan, pp. 159–61; Crone, Slaves on Horses, pp. 173–4.

99Ibrāhīm and his brother Muḥammad were grandsons of al-Ḥasan b. ՙAlī b. Abī Ṭālib. In 145/762 the two of them led an unsuccessful revolt in Basra against al-Manṣūr, being killed in the attempt. See Lassner, The Shaping of ՙAbbāsid Rule, pp. 72, 74, 76–7, 81–4.

100This agitator may have carried on the work of the Isḥāq the Turk mentioned above; see Daniel, op. cit., pp. 132–3.

101Zam(m) was a crossing-point on the Oxus, on the left bank roughly midway between Āmul and Tirmidh. See Le Strange, The Lands of the Eastern Caliphate, pp. 403–4.

102On this episode of ՙAbd al-Jabbār’s rebellion against the caliph – for whatever reason – see Daniel, op. cit., pp. 159–62, who stresses, as one element at work here, the tension between governors and other officials who represented Khurasanian Arab interests, and the centralising policies of the ՙAbbasids.

103Much obscurity surrounds Ustādsīs and his revolt. He was, according to some sources, a local ruler (malik) in the Bādghīs region, and he may originally have assisted the ՙAbbasid daՙwa; others connect him with the Khārijites of neighbouring Sistan; less probable, given his obvious links with Muslim groups, even if somewhat heterodox ones like the Khārijites, is that, like Bihāfarīd, his movement had elements of a reformed version of Zoroastrianism. See Sadighi, Les mouvements religieux iraniens, pp. 155–62; Scarcia Amoretti, ch. ‘Sects and heresies’, pp. 497–8; Daniel, op. cit., pp. 133–7; EI² art. ‘Ustādhsīs’ (W. Madelung).

104taqdīrī kun, perhaps, in the light of what follows, with the implication ‘allot us a share in captured booty!’

105Thus Ḥabībī, chīzī ba-dād; Riḍāzāda Malik has the opposite, chīzī na-dād, ‘he gave them nothing’.

106On Marājil, Hārūn’s slave concubine, see Nabia Abbott, Two Queens of Baghdad. Mother and Wife of Hārūn al-Rashīd (Chicago, 1946), pp. 110, 141; according to Ibn Shākir al-Kutubī, she died in childbirth, having borne al-Ma’mūn. Daniel, op. cit., p. 136, suggests that this putative connection between al-Ma’mūn and Ustādsīs may have been put into circulation by pro-Arab, anti-Persian partisans during the Shuՙūbiyya controversies in order to cast aspersions on al-Ma’mūn and his alleged pro-Iranian proclivities.

107Ḥabībī, n. 6, observes that none of the standard chroniclers, such as al-Yaՙqūbī, al-Ṭabarī and Ḥamza al-Iṣfahānī, mentions this ephemeral governor.

108Ḥumayd had been one of the nuẓarā’, i.e. stand-ins or deputies for the twelve nuqabā’, in the ՙAbbasid daՙwa. See Crone, Slaves on Horses, p. 188; Sharon, Black Banners from the East, p. 195.

109Al-Muqannaՙ’s rebellion probably began ca. 160/777, and constitutes the most important of the anti-ՙAbbasid social and religious movements in the Islamic East during this period. See on this heresiarch, E.G. Browne, A Literary History of Persia (London and Cambridge, 1908–24), Vol. 1, pp. 318–23; Barthold, Turkestan, pp. 198–200; Sadighi, op. cit., pp. 163–86; Scarcia Amoretti, op. cit., pp. 498–500; Farouk Omar, ‘A point of view on the nature of the Iranian revolts in the early Abbasid period’, in ՙAbbāsiyyāt. Studies in the History of the Early ՙAbbāsids, pp. 81–3; Daniel, op. cit., pp. 137–47; EI² art. ‘al-Muḳannaՙ’ (ed.). The adoption of a white banner, and the white garments of his followers, sapīd-jāmagān or mubayyiḍa, were conscious repudiations of ՙAbbasid authority, with its characteristic colour of black for banners, uniforms, etc.; see Omar, ‘The significance of the colours of banners in the early ՙAbbāsid period’, in ՙAbbāsiyyāt, pp. 148–54.

110This idea of ‘transmigration’ (tanāsukh) involves the transference, or the immanence (ḥulūl), of a divine element from one prophet, spiritual leader or (in the case of the Shīՙa) imam to another. In early Islam, the concept was especially characteristic of the Kaysāniyya, partisans of the claims of ՙAlī’s son Muḥammad b. al-Ḥanafiyya, and of other subsequent ghulāt sects. See EI² art. ‘Tanāsukh’ (D. Gimaret). Here in Central Asia, where there was a ferment of religious ideas and beliefs, it is likely, as Daniel suggests (The Political and Social History of Khurasan, pp. 139, 144) that Muqannaՙ’s movement drew in adherents of other heterodox currents. Adoption of the name Hāshim would obviously connect al-Muqannaՙ, whatever his real name was, with messianic currents in the original ՙAbbasid daՙwa, although his own movement was distinctly hostile to the political authority of the ՙAbbasids.

111This may refer to Turks settled on the fringes of Transoxania (cf. below, Part Four, n. 10), or those Oghuz which Ibn al-Athīr, obviously utilising older sources, mentions as coming at this time from the farthest fringes of Inner Asia; see P.B. Golden, An Introduction to the History of the Turkic Peoples. Ethnogenesis and State-Formation in Medieval and Early Modern Eurasia and the Middle East (Wiesbaden, 1992), pp. 206–7. This mention by Gardīzī is nevertheless interesting in showing that al-Muqannaՙ was able to attract a wide spread of support from all the local peoples of Transoxania, especially those in rural areas; and it is further worthy of note that Narshakhī, Tārīkh-i Bukhārā, tr. Frye, p. 72, states that Muqannaՙ had a commander in Bukhara with the name, redolent of the first Turkish empire on the Orkhon, of Kül Er Tegin; see Frye’s nos. 66, 259.

112The readings and sense of the mss. here are unclear. Nafīsī and Ḥabībī read kūy-hā, but the latter, n. 6, suggests a possible meaning for the kūs-hā (‘drums’) of the mss.: that important towns had drums that were beaten at the government headquarters as a manifestation of sovereignty, citing the historian of the Delhi Sultanate period Minhāj al-Dīn Jūzjānī, who states that up to 617/1220 and the coming of Chingiz Khān the ‘fanfare of drums (nawbat) of Abū Muslim’ were beaten at Merv (Ṭabaqāt-i nāṣirī, ed. Ḥabībī [Kabul, 1342–3/1963–4], Vol. 1, p. 107). The meaning here would accordingly be ‘and they beat the ceremonial drums’.

113This obviously lay in the Bukhara region; Yāqūt, Muՙjam al-buldān, Vol. 5, p. 309, reads the name as Nuwajkath but does not pinpoint its exact location.

114This is the obvious sense of Ṣughdiyān, pace Frye, op. cit., p. 71, followed by Daniel, The Political and Social History of Khurasan, p. 142, both taking it, improbably, as a personal name.

115This is the revolt of Yūsuf al-Barm. Whether he was a Khārijite or was simply a rebel representing local interests in eastern Khurasan against the ՙAbbasid state apparatus is unclear. See Daniel, op. cit., pp. 166–7.

116Muՙādh was a mawlā of Rabīՙa or of Dhuhl of Bakr. See on him al-Yaՙqūbī, Buldān, tr. Wiet, pp. 131–2; Daniel, op. cit., pp. 142–3; Crone, Slaves on Horses, pp. 183–4.

117Ṭawāwīs (‘The Peacocks’, so-called, says Narshakhī, because the invading Arabs first encountered peacocks there) was one or two days, journey from Bukhara. See Ḥudūd al-ՙālam, tr. Minorsky, p. 113; Barthold, Turkestan, pp. 98–9; Le Strange, The Lands of the Eastern Caliphate, p. 462.

118This appears to be a personal name, although nothing further is known of the man. It is a not uncommon name in medieval Arabic onomastic, as has been pointed out to me by Professor Van Gelder, citing e.g. the many Khārijas in the index to Caskel and Strenziok-Ibn al-Kalbī’s Ğamharat an-nasab. An emendation of the text to khārijī ‘a Khārijite’, seems nevertheless not impossible, since although al-Muqannaՙ himself was not a Khārijite, as observed above, nos. 108–9, he attracted many heterodox elements to his standard.

119Musayyab was a mawlā of Ḍabba who had been a naẓīr in the ՙAbbasid daՙwa. see Daniel, op. cit., p. 168; Crone, op. cit., p. 181.

120Since the story of al-Muqannaՙ seems early to have attracted fantastic, semi-legendary accretions, there are in the sources various accounts of his death, including self-immolation as well as mass poisoning (this last reminiscent of the mass suicide of Jim Jones and the adherents of his ‘The People’s Temple Full Gospel Church’ at Jonestown, Guyana, in 1978).

121The author of the Ḥudūd al-ՙālam, writing two centuries after al-Muqannaՙ’s time, states (tr. Minorsky, p. 117, comm. p. 356) that the people of Īlāq on the middle Syr Darya mostly professed the creed of the sapīd-jāmagān.

122Barthold, Turkestan, p. 205, read the second component of this name as d.h.da/Dahda, admitting that no governor of that name is mentioned in the sources but suggesting that he was an Arab financial official in Transoxania.

123Concerning these alloy coins, introduced, so Narshakhī says (Tārīkh-i Bukhārā, tr. Frye, pp. 35–7, 129), for local circulation when the price of silver had become excessively high in Khwarazm and Transoxania, see Barthold, op. cit., pp. 204–7.

124Both Barthold, op. cit., p. 202, and Daniel, The Political and Social History of Khurasan, p. 168, note that the governorships of Abu ’l-ՙAbbās and then of the Barmakid al-Faḍl b. Yaḥyā were ones in which beneficial policies for the people of the East were pursued.

125Khwārazmī defines bast (lit. ‘something which is bound up, blocked’, hence a barrier) as a measurement used at Merv for distributing irrigation water, essentially a board with a hole in it of determined size. See Bosworth, ‘Abū ՙAbdallāh al-Khwārazmī on the technical terms of the secretary’s art. A contribution to the administrative history of mediaeval Islam’, JESHO XII (1969), p. 152.

126Presumably this was a ribāṭ for travellers crossing the Qara Qum. It is unmentioned t. by the geographers, and may have had only an ephemeral existence.

127This was on 15 Rabīՙ 170/14 September 786.

128This dating is clearly erroneous, since Ghiṭrīf was governor in this year (see below). We know from other sources that ՙAbbās’s raid against Buddhist shrines at Kabul and Shābahār (the latter place to be distinguished from the homonymous place outside Ghazna mentioned in Ghaznavid history) actually took place in 171/787–8; see Bosworth, Sīstān under the Arabs, pp. 85–6. Riḍāzāda Malik has the date Muḥarram 170 [/July 786], but this is clearly too early.

129In 173/789–90. Ḥamza al-Iṣfahānī, Ta’rīkh Sinī mulūk al-arḍ wa ‘l-anbiyā’, p. 164, has, as governor between Jaՙfar and Ghiṭrīf, Ḥasan b. Qaḥṭaba, but again with the disclaimer ‘God knows best’.

130In 175/791–2. Ghiṭrīf was Hārūn’s maternal uncle, the brother of the caliph’s mother Khayzurān, wife of al-Mahdī. See Abbott, Two Queens of Baghdad, p. 29.

131This Jabūya/Jabghūya must have been a Turkish chief from some tribal group like the Qarluq, Yabghu being an ancient Turkish princely title. See EIr art. ‘Jabḡuya. ii. In Islamic sources’ (C.E. Bosworth).

132See above, p. 36. Narshakhī relates that Ghiṭrīf’s debased coinage placed a heavy burden on the people of Bukhara (and doubtless elsewhere in the East) because taxes had now to be paid at rates always manipulated in favour of the administration and to the detriment of the taxpayers; see his Tārīkh-i Bukhārā, tr. Frye, pp. 36–7. It is clear from various sources that Ghiṭrīf’s governorship was regarded as oppressive.

133Ḥuḍayn seems, from the parallel source of the Tārīkh-i Sīstān, to be the correct form of the name rather than the text’s Ḥuṣayn. See on his revolt, Bosworth, Sīstān under the Arabs, p. 85.

134al-Yaՙqūbī, Buldān, tr. Wiet, p. 133, and Ḥamza al-Iṣfahānī, op. cit., p. 165, record Ḥamza b. Mālik b. Haytham al-Khuzāՙī as Ghiṭrīf’s brief successor as governor in 176/792–3 or at the beginning of the next year.

135For Faḍl’s governorship, regarded in the historical sources and in the adab literature as a benevolent, even idyllic one, see Barthold, Turkestan, p. 203; Daniel, The Political and Social History of Khurasan, p. 169.

136For the etymology of the name of this Iranian magnate, see F. Justi, Iranisches Namenbuch (Marburg, 1895), p. 170: ‘Felsen-hahn (Tetraogallus)’. He is mentioned in al-Ṭabarī as the father of Kāwūs, father of the Afshīn Khaydhar or Ḥaydar, who was prominent in al-Muՙtaṣim’s caliphate as the vanquisher of Bābak Khurramī (see above, p. 43 and Part Two, n. 5). These princes of Ushrūsana held the title of Afshīn, for which see C.E. Bosworth and Sir Gerard Clauson, ‘Al-Xwārazmī on the peoples of Central Asia’, JRAS (1965), pp. 7–8.

137Nafīsī has in his text ‘to Shahrazūr, and from there to Asadābād, and there’.

138ՙUmar b. Jamīl did not, so far as we know from historical literature, found a line in Chaghāniyān of any significance, but the history of this province is particularly obscure until the emergence of the Muḥtājids in the early fourth/tenth century; see Bosworth, ‘The rulers of Chaghāniyān in early Islamic times’, Iran JBIPS XIX (1981), p. 3.

139The story of Hārūn and his breach of faith with the Barmakids, who had been loyal servants of the ՙAbbasids, has been already given by Gardīzī in his section on the caliphs, ed. Ḥabībī, pp. 69–70, ed. Riḍāzāda Malik, pp. 128–9, with an emphasis on the whole episode as a phase in ՙAbbasid decline; cf. Meisami, Persian Historiography, p. 73.

140Manṣūr was thus, like Ghiṭrīf, a maternal relative of the ՙAbbasid royal family, through the brother of al-Manṣūr’s Ḥimyarite wife. See Crone, Slaves on Horses, pp. 255–6 n. 580.

141If the referent in wa bāz gasht is Manṣūr, then presumably he returned to Iraq; if it is Ḥamza, then he would have gone back to his native province of Sistan. Ḥamza’s revolt was the most protracted and serious Khārijite outbreak in the East, beginning in Sistan ca. 179/795–6, engulfing neighbouring provinces like Khurasan and Kirman, and only ending thirty years later with Ḥamza’s death in 213/828. See Sadighi, Les mouvements religieux iraniens, pp. 54–6; Bosworth, Sīstān under the Arabs, pp. 91–4; EIr art. ‘Ḥamza b. Ādarak’ (C.E. Bosworth).

Other historians (e.g. al-Ṭabarī, Ta’rīkh, Tertia series, p. 644, English tr. C.E. Bosworth, The History of al-Ṭabarī. Vol. XXX. The ՙAbbāsid Caliphate in Equilibrium [Albany, 1989], p. 162; Ḥamza al-Iṣfahānī, Ta’rīkh Sinī mulūk al-arḍ, p. 165) record that Jaՙfar b. Yaḥyā al-Barmakī was briefly appointed to Khurasan after his successful campaignings in Syria, but he does not seem ever to have gone there personally.

142On ՙAlī’s governorship, regarded in all the sources as repressive and exploitative, see Barthold, Turkestan, p. 203; Daniel, The Political and Social History of Khurasan, pp. 170–4; EI² art. ‘ՙAlī b. ՙĪsā’ (Ch. Pellat). His family was of mawlā origin, and had been prominent in the ՙAbbasid daՙwa; see Crone, op. cit., pp. 178–9.

143Nothing is known of this work.

144Following the reading of Nafīsī given in Ḥabībī’s n. 13, ba-shikast.

145i.e. Ṭāhir b. al-Ḥusayn b. Muṣՙab b. Ruzayq, whose family power base was in Pūshang.

146The affluent stream of the Oxus, on whose banks Balkh stood, is called by Ibn Ḥawqal a century-and-a-half later, Dih Ās ‘[that which drives] ten mills’; see Le Strange, The Lands of the Eastern Caliphate, p. 420.

147Harthama was of Khurasanian mawlā origin, who had risen high in caliphal favour. See on him Crone, op. cit., pp. 75, 177; EI² art. ‘Harthama b. Aՙyan’ (Ch. Pellat).

148Rāfiՙ’s revolt probably reflected popular discontent in Transoxania with ՙAlī b. ՙĪsā’s misgovernment there, but it spread also to Khurasan, where ՙAlī’s policies were equally rapacious. See Barthold, Turkestan, pp. 200–1; Daniel, The Political and Social History of Khurasan, pp. 172–5; EI² art. ‘Rāfiՙ b. al-Layth b. Naṣr b. Sayyār’ (C.E. Bosworth).

149This section on Harthama’s governorship is apparently considerably confused in the mss., with an inconsequential chronology of events. This is unnoted by Ḥabībī, but has been put right by Riḍāzāda Malik, and the latter’s rearrangement of the entry is followed here.

150Thus the suggestion of Ḥabībī in his n. 16 for the term niwishta, although one would not expect volunteers like these to be entitled to regular stipends on the dīwān al-jaysh rolls; such volunteers normally brought their own arms, equipment and mounts but shared in captured plunder.

151On the confusion in the sources about the manner and place of Ḥamza’s death, and the briefness of Abū Isḥāq Ibrāhīm’s headship of the Khārijites of Sistan, see Bosworth, Sīstān under the Arabs, pp. 103–4.

152The sources are likewise confused about the events surrounding the end of Rāfiՙ’s uprising. Some state that he accepted a pardon from al-Ma’mūn, after which he drops out of historical mention; see Daniel, op. cit., pp. 174–5.

153Following here Riḍāzāda Malik’s ba-dādand.

154This man is presumably the ancestor of the petty dynasty of the Bānījūrids or Abū Dāwūdids, apparently originally from Farghāna but ruling in Balkh and Ṭukhāristān in the later third/ninth century and possibly in the fourth/tenth one. See C.E. Bosworth, The New Islamic Dynasties. A Chronological and Genealogical Manual (Edinburgh, 1996), p. 174 n. 85; EI² Suppl. art ‘Bānīdjūrids’ (C.E. Bosworth).

155According to the anonymous (or conceivably by one Ibn Shādī of the Hamadān region?) Mujmal al-tawārīkh, ed. Malik al-Shuՙarā’ Bahār (Tehran, 1318/1939), p. 349, the astrologer Dūbān had been sent to al-Ma’mūn by the king of Kabul. Kabul was at this time very much within the Indian cultural world, and India was famed as a home of magical and astrological lore.

156qaḍīb, text q.ṣ.b. This and the other two items were insignia of kingship traditionally handed down from caliph to caliph.

157Amongst a considerable literature on the succession struggle between the two brothers, see the exhaustive study by F. Gabrieli, ‘La successione di Hārūn al-Rašīd e la guerra fra al-Amīn e al-Ma’mūn’, RSO XI (1926–8), pp. 341–97.

158See on al-Ma’mūn’s personal rule in Khurasan, Roy Mottahedeh, ch. ‘The ՙAbbāsid caliphate in Iran’, in CHIr, Vol. 4, pp. 72–4; Daniel, The Political and Social History of Khurasan, pp. 175–82.

159al-Yaՙqūbī, Buldān, tr. Wiet, p. 136, and Ḥamza al-Iṣfahānī, Ta’rīkh Sinī mulūk al-arḍ wa ’l-anbiyā’, p. 167, record that al-Ma’mūn in 203/818–19 briefly appointed Faḍl b. Sahl’s kinsman Rajā’ b. Abi ’l-Ḍaḥḥāk to the governorship before Ghassān.

160Ibn Khallikān, Wafayāt al-aՙyān, ed. Iḥsān ՙAbbās (Beirut, 1968–72), Vol. 4, p. 44, has for this name the much more plausible one of ‘Ghālib [al-Masՙūdī al-Aswad]’. On the very doubtful connection of this person with the heresiarch Ustādsīs, see above, n. 106.

161On Ghassān’s governorship, see Bosworth, Sīstān under the Arabs, pp. 101–2; Daniel, op. cit., p. 181; and on Faḍl b. Sahl, EIr art. ‘Fazl b. Sahl’ (C.E. Bosworth).

Part Two: The Tahirids and Saffarids

1Thus correctly in Ḥabībī for Nazim’s Shabīb. For Naṣr’s raising of the Qays or North Arabs of Jazīra in rebellion against al-Amīn and then al-Ma’mūn, apparently a movement with anti-Persianising currents within it, see EI² art. ‘Naṣr b. Shabath’ (C.E. Bosworth).

2On Ṭāhir’s appointment to Khurasan and his death soon afterwards, see D. Sourdel, ‘Les circonstances de la mort de Ṭāhir I^er au Ḥurāsān en 207/822’, Arabica V (1958), pp. 66–9; Daniel, op. cit., pp. 181–2; Bosworth, ch. ‘The Ṭāhirids and Ṣaffārids’, in CHIr, Vol. 4, p. 95; Mongi Kaabi, Les Ṭāhirides au Hurāsān et en Iraq (III^ème H./IX^ème J.-C.), (Tunis, 1983), Vol. 1, pp. 139–86.

3See above, pp. 38–9.

4On Ṭalḥa’s governorship, see Bosworth, op. cit., pp. 95–6; Kaabi, op. cit., pp. 186–90.

5On Bābak’s movement and the Khurramiyya or Khurramdīnān, see Sadighi, Les mouvements religieux iraniens, pp. 187–280; Scarcia Amoretti, ch. ‘Sects and heresies’, pp. 503–9; G.-H. Yusofi, EIr art. ‘Bābak Korramī’.

6On ՙAbdallāh’s governorship in general, see Bosworth, ch. ‘The Ṭāhirids and Ṣaffārids’, pp. 97–101; Kaabi, op. cit., pp. 236–58; EIr art. ‘ՙAbdallāh b. Ṭāher’ (Bosworth); and on the Khārijites in Khurasan and the revolt of Ḥamza b. Ādharak specifically, above, Part One, n. 141.

7The towel was presumably poisoned.

8Cf. on ՙAbdallāh’s circumspection, Barthold, Turkestan, p. 209.

9The text here re-arranged in chronological order by Riḍāzāda Malik.

10On Māzyār’s revolt, see Sadighi, Les mouvements religieux iraniens, pp. 290–303; Kaabi, op. cit., pp. 253–5; EI² art. ‘Ḳārinids’ (M. Rekaya).

11Following here Ḥabībī’s qunī wa q.n.yāt, there being presumably some difference between the two terms not apparent to us today.

12Unidentified, but presumably an adīb or boon-companion.

13On his governorship, see Bosworth, ch. ‘The Ṭāhirids and Ṣaffārids’, pp. 101–2; Kaabi, op. cit., pp. 293–5.

14Gardīzī omits in his listing al-Muՙtazz, who reigned for three years (252–55/866–69) between al-Mustaՙīn and al-Muhtadī.

15See on his governorship and his subsequent history, Bosworth, op. cit., pp. 102–3, 114–15; Kaabi, op. cit., pp. 299–311.

16On the career of Yaՙqūb in general, see Bosworth, Sīstān under the Arabs, pp. 112–21, and Bosworth, The History of the Saffarids of Sistan and the Maliks of Nimruz (Costa Mesa and New York, 1994), pp. 67–180; EI² art. ‘Yaՙūb b. al-Layth’ (Bosworth).

17For a recent reinterpretation of Yaՙqūb’s role as an ՙayyār leader, see Deborah G. Tor, Violent Order: Religious Warfare, Chivalry, and theՙAyyār Phenomenon in the Medieval Islamic World (Würzburg, 2007), pp. 85ff.

18On Yaՙqūb and his contacts with the Ratbīls/Zunbīls, see Bosworth, The History of the Saffarids of Sistan, pp. 76, 85ff. The correct form of the name or title of this ruler in Zābulistān, Zamīndāwar and Rukhwad/Rukhūd/al-Rukhkhaj, whom the first Arab raiders to Bust and beyond encountered in the later seventh century AD (see above, n. 40), and who was probably one of the southern Hephthalites, has long been enigmatic. On account of the ruler’s association in the sources with the local god Zūn or Zhūn and its cult, Josef Marquart in a classic article (J. Marquart and J.J.M. de Groot, ‘Das Reich Zābul und der Gott Žūn vom 6–9 Jahrhundert’, in Festschrift Eduard Sachau zum siebigsten Geburtstage von Freunden und Schülern gewidmet, ed. G. Weil [Berlin, 1915], pp. 248–92) read the first part of the name as zun, although the second element –bīl was never satisfactorily explained. The Arabic sources, both historical and poetical, generally have Ratbīl, and it now seems fairly certain that this is the most likely form, derived from the Old Turkic title used for a tribal ruler subordinate to the Great Qaghan, élteber (concerning which see A. Bombaci, ‘On the ancient Turkic title Eltäbär’, in Proceedings of the IXth Meeting of the Permanent International Altaistic Conference [Naples, 1966], pp. 1–66; Sir Gerard Clauson, An Etymological Dictionary of Pre-Thirteenth Century Turkish [Oxford, 1972], p. 134). Forms related to it are known from various parts of northwestern India and what is now northern and eastern Afghanistan, including the uilotobēr = hilitiber of the recently discovered documents in the Eastern Middle Iranian language of Bactrian. See N. Sims-Williams, ‘Ancient Afghanistan and its invaders. Linguistic evidence from the Bactrian documents and inscriptions’, in N. Sims-Williams (ed.), Indo-Iranian Languages and Peoples (Oxford, 2002), p. 235; Bosworth, ‘The appearance and establishment of Islam in Afghanistan’, in Islamisation de l’Asie centrale. Processus locaux d’acculturation du VII^e au XI^e siècle, ed. E. de la Vaissière (Paris, 2008), pp. 97–114.

19See Bosworth, ‘Notes on the pre-Ghaznavid history of eastern Afghanistan’, IQ IX (1965), pp. 18, 20, 22.

20The text has Māhjūr for the last element of this name. See on Dāwūd b. al-ՙAbbās’s line, above, Part One, n. 154.

21Fīrūz b. K.b.k or K.b.r is mentioned in the sources variously as the son of the Ratbīl/Zunbīl (see above, n. 18) and as ‘the ruler of Zābulistān’. See Bosworth, The History of the Saffarids of Sistan, pp. 99–100.

22This place in Bādghīs was an enduring centre of the Khārijites. See Le Strange, The Lands of the Eastern Caliphate, p. 410; EI² art. ‘Karūkh’ (ed.).

23Thus restored in the text of Ḥabībī, p. 140 n. 6. Farhādhān or Farhādgird, a place on the road between Herat and Nishapur, in the district of Nishapur called Asfand, for the F.r.hād of the mss. See Le Strange, op. cit., p. 388.

24On this annexation of Nishapur and Khurasan by Yaՙqūb (short-lived, as it proved), see Bosworth, The History of the Saffarids of Sistan, pp. 109–21; Bosworth, ch. ‘The Ṭāhirids and Ṣaffārids’, pp. 114–15. The episode is noted in an anecdote of Bayhaqī; see Bosworth, The History of Beyhaqi (The History of Sultan Masՙud of Ghazna, 1030–1041) (New York, 2009), Vol. 1, pp. 352–3.

25This place is frequently mentioned in the chronicles of the Caspian coast region; see, e.g., Ibn Isfandiyār, English tr. E.G. Browne, An Abridged Translation of the History of Ṭabaristán Compiled about AH 613 (AD 1216) (Leiden and London, 1905), p. 182.

26See Bosworth, The History of the Saffarids of Sistan, pp. 110, 123–7; and for the background of the Ḥasanid daՙwa in Gurgān and Ṭabaristān, W. Madelung, ch. ‘The minor dynasties of Northern Iran’, in CHIr, Vol. 4, pp. 206–12.

27Nazim has, apparently erroneously, Ḍallālī for this name, whilst Ḥabībī has Ḍallābī; it is most probably the Ṣallābī of Riḍāzāda Malik which is correct. The nisba in fact remains mysterious; G.C. Miles, The Numismatic History of Rayy (New York, 1938), p. 129, registers al-Ṣalānī; Al-Samՙānī in his Kitāb al-Ansāb does not record it.

28Thus, du barādar, whereas a plurality of brothers, ՙAbdallāh and an unspecified number, are mentioned previously.

29Reading, with Ḥabībī and Riḍāzāda Malik, manfadh rather than Nazim’s m.n.q.d (? munqadd). An ancient canal, the Nahr al-Malik, running from Fallūja on the Euphrates, joined the Tigris just above Dayr al-ՙĀqūl; see Le Strange, The Lands of the Eastern Caliphate, p. 68.

30For Yaՙqūb’s defeat at the hands of the caliphal army and its aftermath, see Bosworth, ch. ‘The Ṭāhirids and Ṣaffārids’, p. 113; Bosworth, The History of the Saffarids of Sistan, pp. 158ff. Gardīzī telescopes these events after the Dayr al-ՙĀqūl battle. During the two years before his death, Yaՙqūb was in Khuzistan and involved in a three-way struggle with ՙAbbasid forces and the Zanj rebels; see ibid., pp. 162–8.

31Following Ḥabībī’s text for these two names, see his n. 3.

32I.e. because of the ending of the threat to Sistan (this phrase only in Ḥabībī). On Yaՙqūb’s and ՙAmr’s prolonged struggles with this ambitious rival commander, Aḥmad Khujistānī, for control of Khurasan, see Bosworth, The History of the Saffarids of Sistan, pp. 127–33, 194–201.

33Otherwise gazīt in early New Persian (= Arabic jizya); see F. de Blois, ‘A Persian poem lamenting the Arab conquest’, in Studies in Honour of Clifford Edmund Bosworth. II. The Sultan’s Turret: Studies in Persian and Turkish Culture, ed. Carole Hillenbrand (Leiden, 2000), p. 87 and n. 20.

34sabal, thus suggested by Habibi, n. 4, for an uncertain reading of the mss. For this unusual word, said to be from the name of a celebrated mare in ancient Arabian times, see Lane, Lexicon, p. 1301c.

35On ՙAmr’s administrative and financial system, see Barthold, Turkestan, pp. 220–1; Bosworth, The History of the Saffarids of Sistan, pp. 358–9.

36On his organisation of the army, see Bosworth, ‘The armies of the Ṣaffārids’, BSOAS, XXXI (1968), pp. 545, 549–51; idem, The History of the Saffarids of Sistan, pp. 351–4. This description of the ՙarḍ, with inspection of everyone’s military preparedness, from the Amir himself downwards, is also given in the lengthy biography of Yaՙqūb and his brother in Ibn Khallikān’s Wafayāt al-aՙyān, at Vol. 6, pp. 421–2, English tr. McG. de Slane, Vol. 4, p. 322, both accounts being obviously derived from the common source of al-Sallāmī’s Ta’rīkh Wulāt Khurāsān (see above, Introduction, p. 2). Ibn Khallikān has the further detail of a comparison of the ՙarḍ under ՙAmr with that of the Sasanid emperor Khusraw Anūshīrwān; Barthold, who in Turkestan, p. 221, also gives an account of this Saffarid ՙarḍ, remarked that the resemblance between the two military procedures, though separated by some three centuries, could hardly be coincidental.

37On the diplomatic exchanges between ՙAmr and the caliph, leading up to ՙAmr’s ill-fated decision to march against the Samanids, see Bosworth, The History of the Saffarids of Sistan, pp. 223–8.

38For ՙAmr’s attack on Ismāՙīl b. Aḥmad, his defeat and his imprisonment and consequent death see Bosworth, ch. ‘The Ṭāhirids and Ṣaffārids’, p. 121; Bosworth, The History of the Saffarids of Sistan, pp. 229–35.

Part Three: The Samanids

1Much of this elaborate genealogy, with over fifty generations between Sāmān Khudāh and Kayūmarth, is unintelligible, doubtless at least in part through corrupt transmission. Al-Bīrūnī, in his al-Āthār al-bāqiya, tr. Sachau, p. 48, merely traces Sāmān Khudāh’s great-grandson Ismāՙīl b. Aḥmad back through four more generations to the father of the emperor of Sasanid times, Bahrām VI Chūbīn (r. 590–1), sc. Bahrām Gushnāsp. The late sixth-early seventh/late twelfth-early thirteenth century author Ibn Ẓāfir, citing the fifth/sixth century ՙAbbasid historian Ghars al-Niՙma Hilāl al-Ṣābi’, takes the genealogy five generations beyond Bahrām Chūbīn’s father to Yazdagird II (r. 399–420) (Luke Treadwell, ‘The account of the Samanid dynasty in Ibn Ẓāfir al-Azdī’s Akhbār al-duwal al-munqaṭiՙa’, Iran JBIPS XLIII [2005], pp. 136, 152). It is clear that Gardīzī was far less informed about the pre-Islamic Persian kings than e.g. his predecessor by about a century, Ḥamza al-Iṣfahānī, in the first chapter of the latter’s Ta’rīkh Sinī mulūk al-arḍ wa ’l-anbiyā’. Mr de Blois has observed to me that none of the unintelligible names in Gardīzī’s genealogy here has any obvious connection with Bactrian or Sogdian onomastic.

2Narshakhī, Tārīkh-i Bukhārā, tr. Frye, pp. 76–7; Barthold, Turkestan, p. 209–10; Frye, ch. ‘The Sāmānids’, in CHIr, Vol. 4, p. 136; Frye, Bukhara, the Medieval Achievement, 2nd ed. (Costa Mesa, 1996), p. 35.

3Narshakhi, op. cit., tr. Frye, pp. 82–6; Barthold, op. cit., pp. 209–10; Frye, ch. ‘The Sāmānids’, pp. 136–7.

4The name of this rebel Samanid commander is rendered unintelligibly in Nazim, but Ḥabībī has the correct Ūk.r.t.m.sh, Turkish ögretmish ‘he taught, instructed’. See Clauson, An Etymological Dictionary of Pre-Thirteenth Century Turkish, p. 114; Faruk Sümer, ed. Turan Yazgan, Türk devletleri tarihinde şahis adları (Istanbul, 1995), Vol. 2, pp. 493–4; Rásonyi and Baski, Onomasticon turcicum, Vol. 2, p. 591.

5I.e. the Daylamī lord Justān (III) b. Wahsūdān, involved at this time in the confused struggles of the ՙAbbasids, the Zaydī Imāms of Gurgān and Ṭabaristān and the Samanids for control of northern upland Iran and the Caspian coastlands. See Madelung, ch. ‘The minor dynasties of northern Iran’, pp. 208–9; Bosworth, The New Islamic Dynasties, p. 145 no. 69.

6Text, Pārs, i.e. Turkish bars ‘leopard’, also used for other large felines; see Clauson, op. cit., p. 368; Sümer, op. cit., Vol. 2, s.v.

7See on Ismāՙīl’s reign, Narshakhī, Tārīkh-i Bukhārā, tr. Frye, pp. 86–94; Frye, ch. ‘The Sāmānids’, pp. 138–41; Frye, Bukhara, the Medieval Achievement, pp. 38–49; Treadwell, ‘The account of the Samanid dynasty in Ibn Ẓāfir al-Azdī’s Akhbār al-duwal al-munqaṭiՙa’, p. 153. Concerning the title Amir-i Māḍī, with the epithet conventionally translated by the common meaning of the word māḍī as ‘late, one who has passed away’, Treadwell has recently cited Van Gelder’s opinion that a better rendering in this context would be ‘incisive, penetrating, efficacious’, referring to Ismāՙīl’s numerous military successes (see op. cit., p. 153 n. 20). Since the epithets applied posthumously to the succeeding Samanid amirs all refer to such kingly qualities as good fortune, piety, righteousness, justice, divine guidance, etc., this suggestion seems a sound one.

8I.e. Khurasan in its widest sense, all the lands east of Ray and Qūmis, including Transoxania.

9The investiture diploma sent by the caliph in 298/910–11 included a renewal of the grant of governorship over Sistan and its dependencies which al-Muՙtaḍid had given Ismāՙīl b. Aḥmad ten years previously; see Bosworth, The History of the Saffarids of Sistan, p. 263. This grant now became the legal justification for Samanid attempts to annex Sistan to their lands; a policy ultimately unsuccessful, however, given the distance of Sistan from the Samanid centre of power in Transoxania, as the events recorded below show.

10Muՙaddal and his brother Muḥammad, sons of a brother of Yaՙqūb and ՙAmr, had succeeded to power in the truncated Saffarid dominions in the confused years after ՙAmr b. Layth’s capture; see Bosworth, op. cit., pp. 260–6.

11On this superannuated former soldier of the Samanids, Muḥammad b. Hurmuz, see Bosworth, ‘The armies of the Ṣaffārids’, p. 539, and Bosworth, The History of the Saffarids of Sistan, pp. 269–71.

12Sīmjūr Dawātī was a Turkish ghulām commander of the Samanids whose family was to play a notable role in the warfare of the Samanids with their rivals in northern Persia, at times acting as governors of Khurasan and acquiring estates in Quhistān. See Bosworth, The New Islamic Dynasties, p. 175 no. 86, and EI² art. ‘Sīmdjūrids’ (Bosworth).

13For the episode of Samanid intervention in Sistan, followed by an abortive attempt by al-Muqtadir to recover the province for the caliphate, see Bosworth, The History of the Saffarids of Sistan, pp. 263–75.

14Firabr lay just north of the right bank of the Oxus, opposite Āmul-i Shaṭṭ, and on the road to Bukhara. See Le Strange, The Lands of the Eastern Caliphate, p. 443.

15This ՙAlid made firm the Zaydī imāmate which had been established in Ṭabaristān in the latter half of the third/ninth century, and ruled there till his death in 304/917; the Samanid army sent against him under Muḥammad b. Ṣuՙlūk was disastrously defeated by Ḥasan just before Amir Aḥmad’s murder. See EI² art. ‘Ḥasan al-Uṭrūsh’ (R. Strothmann) and Madelung, ch. ‘The minor dynasties of northern Iran’, pp. 208–10.

16See on Aḥmad’s reign, Frye, ch. ‘The Sāmānids’, p. 141, and specifically for his murder, Treadwell, ‘Ibn Ẓāfir al-Azdī’s account of the murder of Aḥmad b. Ismāՙīl al-Sāmānī and the succession of his son Naṣr’, in Studies in Honour of Clifford Edmund Boswortḥ II. The Sultan’s Turret. Studies in Persian and Turkish Culture, pp. 397–419. Treadwell notes (pp. 414ff.) that this account seems to be independent of that of al-Sallāmī or a source of his, and was probably written by a contemporary of the events in question. Some of the sources state that the Amir was murdered by his guard because of his excessive favour to scholars and the religious classes, and possibly because he had favoured the use of Arabic at court over Persian; see Barthold, Turkestan, p. 240.

17See the references in the previous note, with once again an especially detailed account in Ibn Ẓāfir apud Treadwell, pp. 405–13. Naṣr was clearly intended to be the puppet of the elements who had conspired to kill his father.

18There is much confusion over the members of the Jayhānī family, at least three of whom served the Samanids as viziers in the course of the fourth/tenth century, in particular, over the exact forms of their names (Muḥammad and Aḥmad being especially liable to be mixed up with each other). The one mentioned here by Gardīzī was apparently the first Jayhānī, serving Amir Naṣr till ca. 310/922 and being then replaced by Abu ’l-Faḍl Muḥammad b. ՙUbaydallāh Balՙāmī (see above, p. 57), who was in turn replaced, at the end of Naṣr’s thirty years’ reign, by the second Jayhānī, Abū ՙAlī Muḥammad b. Muḥammad, very probably the son of the first Jayhānī. Gardīzī expatiates here on Jayhānī’s great learning and intellectual curiosity, and this is confirmed by the geographer al-Maqdisī’s description of Abū ՙAbdallāh as skilled in philosophy and astronomy and as one who sought out persons from all the the world in order to get information on their homelands, their specialities and characteristics (partial Fr. tr. André Miquel, Aḥsan at-taqāsīm fī maՙrifat al-aqālīm (La meilleure répartition pour la connaissance des provinces) [Damascus, 1963], pp. 12–14 §§ 10–11). There seems to have been a taint of Ismāՙlism, perhaps more of that faith’s philosophical aspect than of it as a radical, politically activist creed, surrounding the three main members of the family involved in political life during Samanid times (see Patricia Crone and Luke Treadwell, ‘A new text on Ismailism at the Samanid court’, in Chase F. Robinson (ed.), Texts, Documents and Artefacts. Islamic Studies in Honour of D.S. Richards [Leiden, 2003], pp. 54–5). Ch. Pellat suggested that the celebrated, lost geographical work attributed to a Jayhānī, a Kitāb al-Mamālik wa ’l-masālik, was not the work of a single person but was added to or remodelled by more than one member of the family, a process not unknown in Arabic literature. See Frye, ch. ‘The Sāmānids’, pp. 142–3; EI² Suppl. art. ‘al-Djayhānī’ (Pellat).

19This seems to be the correct reading for this place name. According to Yāqūt, Muՙjam al-buldān, Vol. 2, p. 365, it was three parasangs from Samarqand and famed as the site of the great traditionist Muḥammad b. Ismāՙīl al-Bukhārī’s tomb; see also Barthold, Turkestan, p. 126.

20Thus in both mss.; later sources have varying forms for this doubtful name.

21Thus in the surmise of Ḥabībī, p. 151 n. 8 for the z.rāh of the mss.

22I.e. of Yazdagird III (r. 632–51), the last Sasanid emperor.

23Ḥabībī, p. 151 n. 10, identifies this with Kīrang, which he says was a well-known place in Mongol times.

24Text, s.b.k.r.y. Sebük-eri (this could be, in Turkish, something like ‘beloved man’, but the etymology remains far from sure) was to play a dominant role within the Saffarid state after the capture of ՙAmr and the succession in the capital Zarang of his weaker grandson Ṭāhir b. Muḥammad b. ՙAmr in 287/900. See Bosworth, The History of the Saffarids of Sistan, pp. 243ff.

25This is Qarategin al-Isfījābī (d. 317/929), who towards the end of his career in the service of the Samanids withdrew to Bust and al-Rukhkhaj on the far southern fringes of the Samanid amirate and established there a virtually independent line of Turkish ghulāms who held power until the Ghaznavid founder Sebüktegin expanded into that region in 367/977–8. His son Manṣūr later appears as commander-in-chief in Khurasan for the Samanids, see above, p. 60, and also his grandson Aḥmad b. Manṣūr, see above, p. 67.

26Ḥabībī, p. 152 n. 10, points out that this place, defectively written in the mss., is mentioned by the geographers as being a district of Marw al-Rūd (which would accord with the mention here of a river; in this case, the river would be the Murghāb).

27Naṣr’s youth obviously allowed these rebels and rival claimants to the throne to rear their heads. See Narshakhī, Tārīkh-i Bukhārā, tr. Frye, pp. 95–6; Barthold, Turkestan, p. 241; Frye, ch. ‘The Sāmānids’, p. 141; Treadwell, ‘The account of the Samanid dynasty’, pp. 155–7.

28I.e. the Amir’s vizier, see below, n. 37.

29See Barthold, op. cit., p. 242, who notes that Abū Bakr al-Khabbāz must have had a charismatic personality and have had a great influence amongst the populace of Bukhara for there to have arisen this story of the miraculous preservation of his body from effects of the flames.

30Thus the suggested reading in Ḥabībī, cf. p. 153 n. 3, for the manuscripts’ s.n.jāb.

31The Muḥtāj family were hereditary lords of the principality of Chaghāniyān on the north bank of the upper Oxus river, probably of Iranian or Iranised Arab descent, and members of them served the Samanids as commanders and governors until the mid-fourth/tenth century, when the family lapsed out of public life and were apparently of significance only in Chaghāniyān itself. See Bosworth, ‘The rulers of Chaghāniyān in early Islamic times’, pp. 1–10; Bosworth, The New Islamic Dynasties, p. 177 n. 88; EI² art. ‘Muḥtādjids’ (Bosworth).

32As Nazim, p. 30 n. 2 notes, there seems to be a lacuna here.

33Other sources like Masՙūdī, Miskawayh and Ibn al-Athīr confirm that the leading assassin here was Bajkam, but do not expressly attribute to him, as here, a connection with Mardāwīj’s enemy and rival for power in northern Persia and a fellow-Daylamī, Mākān b. Kākī. Nazim does not specify the name Bajkam amongst the assassins but interprets the text here as ba-ḥukm-i Mākān ‘at the instigation of Mākān’; what Gardīzī originally wrote here is thus unclear.

The name Bajkam was a name well known amongst Turkish slave troops of this time, see Sümer, Türk devletleri tarihinde şahis adları, Vol. 2, pp. 482–3. It is, however, most probably in origin an Iranian word with the sense of ‘tassel, fringe or tail of horsehair, etc., tied on to a standard’, see G. Doerfer, Türkische und mongolische Elemente im Neupersischen (Wiesbaden, 1963–75), Vol. 2, Türkische Elemente im Neupersischen, p. 425 n. 840.

34Once Amir Naṣr had grown to maturity and had mastered the various threats to his throne, he adopted a forward policy in northern Persia, aiming at exerting Samanid authority over the Caspian provinces and securing control of Ray; in the later part of the century, this would bring them into prolonged conflict with the Buyids. As a result of these ambitions, Samanid forces clashed at this time with the Zaydī Imāms and with various Daylamī and Jīlī soldiers of fortune, eager to carve out principalities for themselves. The most notable of these were the commanders mentioned here, Mākān and Mardāwīj, the latter and his son Wushmgīr being the founders of a long-lasting line of amirs in Gurgān and Ṭabaristān. See Madelung, ch. ‘The minor dynasties of northern Iran’, pp. 208ff.; EI² arts. ‘Mākān b. Kākī’; ‘Mardāwīdj’; ‘Wushmgīr b. Ziyār’ (C.E. Bosworth).

35See on Naṣr’s reign, Narshakhī, Tārīkh-i Bukhārā, tr. Frye, pp. 95–6; Barthold, Turkestan, pp. 240–6; Frye, ch. ‘The Sāmānids’, pp. 141ff; Frye, Bukhara, the Medieval Achievement, pp. 51ff.; Treadwell, ‘The account of the Samanid dynasty’, pp. 155–7; EI² art. ‘Naṣr b. Aḥmad b. Ismāՙīl’ (C.E. Bosworth). A notable feature of Gardīzī’s account .of this Amir and his reign is the fact that, as with the other strictly historical sources, he has no mention of what seems to have been the temporary, if only limited, success of an Ismāՙīlī Shīՙī daՙwa at the Samanid court, with converts to this heresy (as Sunnīs regarded it), or at least sympathisers towards its philosophical aspects, from the Amir himself downwards. It does appear that the episode has been edited out of the historical sources, since the evidence for its existence comes from the basically adab and literary sources of Ibn al-Nadīm, al-Thaՙālibī and Niẓām al-Mulk. This is not the place for discussing what is a complex matter, but amongst recent writing on the topic, mention should be made of Crone and Treadwell, ‘A new text on Ismailism at the Samanid court’, pp. 37–67, and A. C.S. Peacock, Mediaeval Islamic Historiography and Political Legitimacy. Balՙamī’s Tārīkhnāma (London and New York, 2007), pp. 25–31.

36Following the conjecture of Ḥabībī, p. 154, see n. 7: wa ḥadd wa du gurūhī andar uftād, for a disturbed text here.

37For Balՙamī, see EIr art. ‘Balՙamī, Abu’l-Fażl Moḥammad’ (C.E. Bosworth), and for the Jayhānī family, see above n. 18.

38I.e. Abu ’l-Faḍl Sulamī, noted for his piety and vizier until 335/946, see below.

39Thus in Ḥabībī, following Ibn al-Athīr’s Ṭughān al-ḥājib for the incomprehensible Ṭ.gh.y al-māj.t of the mss.

40Following the text in Ḥabībī, ba-Marw.

41The interpretation of Ḥabībī and Riḍāzāda Malik, ḥīlatī karda būd, seems better than the ḥamlatī karda būd of the mss. and Nazim.

42Yāqūt, Muՙjam al-buldān, Vol. 1, p. 291, says that Ayghān was a village of Panj-dih, which lay on the Murghāb.

43The Sinj of ibid., Vol. 3, p. 265.

44In the mss., r.kh.ta ḥ.m.w.y, which Ḥabībī, p. 156 n. 3, reads as possibly R.khna, Rukhna being the name of one of the gates of the suburb of Bukhara according to Ibn Ḥawqal, Kitāb Ṣūrat al-arḍ, ed. J.H. Kramers (Leiden, 1938–39), Vol. 2, p. 484, French tr. J.H. Kramers and G. Wiet, Configuration de la terre (Paris, 1964), Vol. 2, p. 465.

45Thus in Nazim, but Ḥabībī and Riḍāzāda Malik have Khartang. The first seems nevertheless correct; see Barthold, Turkestan, pp. 248, 259, remarking that this place is not to be confused with the Khartang mentioned above, see n. 19.

46This seems the most probable rendering, since the name is attested in Turkish onomastic; the mss. have q.t.gīn. Qut means ‘spirit, life, vitality’ and ‘fortune, success’, so that the name could mean ‘fortunate prince’. See Rásonyi and Baski, Onomasticon turcicum, II, pp. 505–6.

47Ḥabībī, n. 13, suggests that this nisba (appearing thus in the mss.) comes from a place in Zābulistān (presumably the present-day Uruzgan, in the province of the same name in eastern Afghanistan).

48This was the celebrated gorge, now the Buzgala defile, through which the road from Tirmidh to Kish and Nakhshab in Sogdia passed. See Le Strange, The Lands of the Eastern Caliphate, pp. 441–2; Barthold, Turkestan, p. 138.

49Or Rāsht? See Ḥudūd al-ՙālam, tr. Minorsky, pp. 63, 120, comm. pp. 361–3. Zhāsht or Rāsht lay in the Buttamān mountains which separated the upper basins of the Oxus and, to their north, the Zarafshān, and was the haunt of the predatory Kumījīs, who were probably the remnants of an earlier Inner Asian people like the Sakas. See Bosworth and Clauson, ‘Al-Xwārazmī on the peoples of Central Asia’, pp. 8–9.

50Apparently the Wayshagirt or Bishgird (< *Wēshgird), the modern Fayḍābād, of udūd al-ՙālam, tr. Minorsky, pp. 115, 120, comm. pp. 353–4.

51Thus correctly in Ḥabībī and Riḍāzāda Malik. See on it Ḥudūd al-ՙālam, Barthold’s Preface, pp. 38–9, tr. Minorsky, p. 114; Barthold, Turkestan, p. 74.

52Although Gardīzī, in company with other historians, may have deliberately omitted mention of the presence of Ismāՙīlism at the Samanid court in Amir Naṣr b. Aḥmad’s reign (see above, n. 18), he clearly had a considerable interest in messianic and millenarian movements, if this was indeed one; and earlier in his history he had taken note of such movements as those of al-Muqannaՙ, Ustādsīs and Bābak. See above, Introduction, pp. 3–4.

53Nazim, p. 36 n. 1, implausibly suggests that this is the Warduk of al-Maqdisī, Aḥsan al-taqāsīm, p. 264, listed there as a dependency of Binkath in the district of Shāsh, i.e. north of the Syr Darya; but this region is a long way from the upper Oxus mountain region, the locale of this episode.

54I.e. Abū ՙAlī Ḥasan b. Būya, Rukn al-Dawla, the Buyid ruler in Jibāl; see on him EI² art. ‘Rukn al-Dawla’ (H. Bowen and C.E. Bosworth). Gardīzī usually writes his name with the Persian iḍāfa, Ḥasan-i Būya, but occasionally with the Arabic ibn.

55See on Nū’s reign, Narshakhī, Tārīkh-i Bukhārā, tr. Frye, pp. 97–8; Barthold, op. cit., pp. 246–9; Frye, ch. ‘The Sāmānids’, p. 151; Treadwell, ‘The account of the Samanid dynasty’, p. 157.

56The texts have Abu ’l-Fatḥ, confusing the father Abu ’l-Faḍl Muḥammad with his son and successor Abu ’l-Fatḥ ՙAlī, both of whom were viziers to the Buyids and celebrated literary stylists. See EI² art. ‘Ibn al-ՙAmīd’ (Cl. Cahen); J.L. Kraemer, Humanism in the Renaissance of Islam. The Cultural Revival during the Buyid Age (Leiden, 1986), pp. 241–59.

57mayzad, mīzad, thus conjectured by Ḥabībī, p. 160 n. 1, for an obscure text here. Riḍāzāda Malik proposes the equally possible reading mabarrat-hā ‘acts of beneficence’.

58The Khurasanian family of the ՙUtbīs, of Arab descent, provided two viziers for the later Samanids and a secretary, the famed historian, author of the Ta’rīkh al-yamīnī, Abū Naṣr Muḥammad b. ՙAbd al-Jabbār al-ՙUtbī, for Maḥmūd of Ghazna. See EI² art. ‘al-ՙUtbī’ (C.E. Bosworth).

59Abū Manṣur was a dihqān of Ṭūs and from a family that traced itself back to late Sasanid times. As well as his military role in Samanid affairs, he also has fame in Persian literature as the patron of a now lost New Persian prose version, made from the Pahlavi original, of the national epic, the Shāh-nāma. See V. Minorsky, ‘The older preface to the Shāh-Nāma’, in Studi orientalistici in onore di Giorgio Levi Della Vida (Rome, 1956), Vol. 2, pp. 162–3; EIr art. ‘Abū Manṣūr … b. ՙAbd al-Razzāq’ (Dj. Khaleghi-Motlagh).

60Reading here, with Ḥabībī and Riḍāzāda Malik, liyāqat.

61I.e. the son of the Abu ’l-Faḍl Muḥammad Balՙamī who had served Ismāՙīl b. Aḥmad and his grandson Naṣr as vizier (see above, p. 57). Abū ՙAlī was to achieve lasting literary fame as the author of the Persian translation and remodelling of Ṭabarī’s History, the Tārīkh-i Balՙamī. See EIr art. ‘Amīrak Balՙamī’ (Dj. Khaleghi-Motlagh); Peacock, Mediaeval Islamic Historiography and Political Legitimacy, pp. 31–5. The sources imply that Abū ՙAlī was inferior to his father in political wisdom and a better littérateur than statesman.

62Reading here, with Ḥabībī and Riḍāzāda Malik, kamtar.

63See on ՙAbd al-Malik’s reign, Narshakhī, Tārīkh-i Bukhārā, tr. Frye, p. 98; Treadwell, ‘The account of the Samanid dynasty’, pp. 157–8 (mentioning that, during his lifetime, this Amir was known as al-muwaffaq ‘The Divinely-Assisted One’); Barthold, Turkestan, pp. 249–50.

64This sentence not in Ḥabībī’s text. The son whose succession Alptegin favoured was, according to the geographer al-Maqdisī, Naṣr, who only held the throne, however, for one day; see Barthold, op. cit., p. 250.

65The words of a lacuna here have been supplied by Nazim and Ḥabībī.

66Ḥabībī, n. 3, opines that this otherwise unknown place, obviously lying to the northeast of Nishapur in the direction of Sarakhs, Merv and the Oxus, may be the S. āha mentioned by al-Maqdisī, Aḥsan al-taqāsīm, p. 351.

67In Narshakhī, op. cit., p. 99, the general deputed to attack Alptegin is named as Ashՙath b. Muḥammad; in Niẓām al-Mulk’s recounting of this episode in his Siyāsat-nāma, ch. XXVII, no name is given.

68Alptegin’s withdrawal to Ghazna after the failure of the putsch mounted by him and Abū ՙAlī Balՙamī (who apparently made his peace with the new regime, since he continued in office as vizier during the early part at least of Amir Manṣūr’s reign) marks the beginning of a line of former Samanid Turkish ghulām commanders in Ghazna, culminating in Sebüktegin’s assumption of power in 366/977 and the eventual foundation of the Ghaznavid sultanate. See on these ghulām commanders in Ghazna, Bosworth, ‘Notes on the pre-Ghaznavid history of eastern Afghanistan’, pp. 16–18.

69This phrase in Nazim’s text but omitted (by inadvertence?) from Ḥabībī’s one.

70al-Maqdisī, Aḥsan al-taqāsīm, p. 352, lists N.m.kh.k.n as a stage between Nishapur and Herat.

71This is the reading of the texts, and would ostensibly refer to the town of Gurgānj in Khwarazm, the region at that time ruled by the Shāhs of the Banū ՙIrāq line and in theory tributaries of the Samanids (see M. Fedorov, ‘The Khwarazmshahs of the Banū ՙIrāq (fourth/tenth century)’, Iran JBIPS XXXVIII [2000], pp. 71–5); but no other source refers to a raid into Khwarazm at this time.

72There is some uncertainty over the date of Balՙamī’s death, since ՙUtbī states that he served briefly as vizier to Nūḥ (II) b. Manṣūr (I) in 382/992 when the latter reoccupied Bukhara on the death of the Qarakhanid Hārūn or Ḥasan Bughrā Khān (see above, p. 72). See Peacock, Mediaeval Islamic Historiography and Political Legitimacy, pp. 32–3.

73This lay in the western part of Ghūr, to the east-south-east of Herat, and still in fact exists today as a small town; see Bosworth, The History of the Saffarids of Sistan, p. 409 n. 1252.

74For the expedition into Ghūr of Abū Jaՙfar (who came from a prominent family of Bayhaq and had been made governor of that town by Abu ’l-Ḥasan Sīmjūrī, see Bosworth, ‘The early Islamic history of Ghūr’, CAJ VI (1961), p. 122.

75Khalaf b. Aḥmad, of the so-called ‘second line’ of Saffarid Amirs of Sistan, had for long been at odds, first with Abu ’l-Ḥusayn Ṭāhir (apparently connected maternally with the Saffarid dynasty) and then with the latter’s son Ḥusayn, over control of Sistan. Sources like al-ՙUtbī state that the Samanids intervened on the side of Ḥusayn because Khalaf had cut off the customary tribute and presents to Bukhara. See Bosworth, The History of the Saffarids of Sistan, pp. 302ff.

76recte, in 363/974.

77The texts have Abu ’l-Faḍl, i.e. the reverse of the confusion of names noted in n. 56 above; the father Abu ’l-Faḍl had in fact died in 360/970.

78I.e. only five months after Balՙamī’s death, if this did indeed take place in Jumādā II 363.

79See on Manṣūr’s reign, Narshakhī, Tārīkh-i Bukhārā, tr. Frye, pp. 98–9; Barthold, Turkestan, pp. 250–2; Frye, ch. ‘The Sāmānids’, pp. 152–6; Treadwell, ‘The account of the Samanid dynasty’, p. 158. Niẓām al-Mulk mentions a recrudescence of Ismāՙīlism as occurring during Manṣūr’s amirate, but Crone and Treadwell suggest that, rather than being an organised daՙwa on the scale of that in Naṣr b. Aḥmad’s time (see above, n. 35), this was more probably sporadic provincial disturbances by Ismāՙīlīs, Khurramīs or other sectaries and not a major outbreak in the Samanid heartland of Sogdia. See their ‘A new text on Ismailism at the Samanid court’, pp. 48–52.

80The Farīghūnids, apparently of Iranian stock, were local rulers of the region of Gūzgān in northern Afghanistan and tributaries of the Samanids; when Maḥmūd of Ghazna succeeded to the Samanid heritage in Khurasan, Gūzgān was speedily absorbed into his domains. See Minorsky, Ḥudūd al-ՙālam, comm., pp. 173–8; EIr art. ‘Āl-e Farīḡūn’ (C.E. Bosworth).

81The Samanids were normally abstemious in their personal use of titles and the granting of them, but from the middle years of the century, the Amirs were often in effect compelled to award grandiloquent laqabs to powerful governors and commanders. See Bosworth, ‘The titulature of the early Ghaznavids’, Oriens XV (1962), pp. 214–15.

82Abu ’l-Ḥasan Sīmjūrī represented, of course, the powerful military commanders’ growing ascendancy in the state that Abu ’l-Ḥusayn, like others of his predecessors in the vizierate, could be expected to try and curb.

83Ḥabībī, p. 165 n. 7 is unable to make sense of this word; the first syllable looks however like the negative particle na-. The sense must in fact be something like ‘inadvisable’.

84This is something like a colloquial English equivalent to the text’s idiomatic phrase wallāhi ki man sitāra ba-rūz bad-īshān namāyam, lit. ‘By God, I’ll make them see the stars by daylight’, i.e. turn their bright day into deep night; cf. Riḍāzāda Malik’s taՙlīqa at pp. 592–3.

85Member of the influential Nishapur family of Mīkālīs, which seems to have been of Sogdian origin; various of them served the Samanids, then the early Ghaznavids in administrative and secretarial posts, and one of them at least served the Seljuq Ṭoghrïl Beg. See EI² art. ‘Mīkālīs’ (C.E. Bosworth); EIr art. ‘Āl-e Mīkāl’ (R.W. Bulliet).

86From 371/981 to 387/997, Qābūs’s family patrimony of Gurgān and Ṭabaristān was occupied by the Buyids ՙAḍud al-Dawla, his son Mu’ayyid al-Dawla and then ՙAḍud al-Dawla’s brother Fakhr al-Dawla, with Qābūs as an exile in the Samanid lands. See EI² arts. ‘Ḳābūs b. Wushmagīr b. Ziyār’ and ‘Ziyārids’ (C.E. Bosworth).

87I.e. to don civilian dress, the garment of secretaries and officials, instead of the garb and panoply of war.

88In the final phase of his struggle with Ḥusayn b. Ṭāhir, Khalaf b. Aḥmad sought help from Sebüktegin, who had been in control of Bust since ca. 367–8/977–9, and received from him a force of the troops formerly in the service of Sebüktegin’s predecessor there, Bāytūz. It was presumably these who were now employed against Abū ՙAlī. See Bosworth, The History of the Saffarids of Sistan, pp. 311–12.

89The vizier’s kinsman, the historian Abū Naṣr al-ՙUtbī, with justice considered Abu ’l-Ḥusayn to be the last vizier of the Samanids worthy of the name. See Barthold, Turkestan, pp. 252–3; EI² art. ‘al-ՙUtbī. 2.’ (C.E. Bosworth).

90This seems the most probable interpretation of the text’s ’.n.j; ïnanch Turkish ‘belief, trust’, is wellattested in onomastic from Orkhon Turkish times onwards. See Rásonyi and Baski, Onomasticon turcicum, vol. 1, pp. 318–19; Sümer, Türk devletleri tarihinde şahis adları, vol. 1, pp. 39, 59.

91I.e. the son of Sulaymān b. Satuq Bughra Khān. See O. Pritsak, ‘Die Karachaniden’, Isl. XXXI (1953–4), p. 26.

92See above, n. 19.

93For these complex operations in Khurasan and Transoxania, now signalling the end of Samanid independent power, see Barthold, op. cit., pp. 252–4.

94This royal estate just outside Bukhara, with palaces and beautiful gardens, and dating back to the pre-Islamic princes of Sogdia, was a favoured residence of the Samanids. See the description of Narshakhī, Tārīkh-i Bukhārā, tr. Frye, pp. 26–9.

95The Ilig’s nomination of ՙAbd al-ՙAzīz to the Samanid throne, such as it was by now, was based on the fact that the latter’s father Amir Nūḥ (I) b. Naṣr (II) had nominated ՙAbd al-ՙAzīz as fourth in line of succession, as explained by Gardīzī, above, ed. Nazim, p. 39, ed. Ḥabībī, p. 159, but this son’s rights had been usurped by Manṣūr (I) b. Nūḥ (I) when he nominated his own son Nūḥ (II) as his successor. When Amir Nūḥ reoccupied his capital after the Ilig’s departure, he took vengeance on his uncle and had him blinded. See Barthold, Turkestan, p. 260; Treadwell, ‘The account of the Samanid dynasty’, p. 160 and n. 78.

96This place probably lay somewhere in the region of Aulié-Ata, medieval Ṭarāz, modern Dzambul (hence near the northwestern border of modern Kyrgyz Republic with that of the Kazakh Republic), in the surmise of Barthold. See his Zur Geschichte des Christentums in Mittel-Asien bis zur mongolischen Eroberung (Tübingen and Leipzig, 1901), p. 35.

97On this first Qarakhanid invasion of the Samanid lands, see Barthold, Turkestan, pp. 257–60; Pritsak, ‘Die Karachaniden’, p. 26. Bayhaqī has some information on it; see Bosworth, The History of Beyhaqi, Vol. 1, pp. 294–5.

98See EI² art. ‘Sebüktigin’ (C.E. Bosworth).

99Thus in Ḥabībī’s text, yaqīn dārand; in Nazim’s one, t.ՙ.y.n dārand.

100Dārā presumably shared in his father’s exile from the ancestral Buyid lands whilst the Buyids were in occupation of them (see above, n. 86), and obviously became a soldier of fortune in the struggles for control of Khurasan in these last days of Samanid rule there. His later career is somewhat obscure. From the information of sources like al-ՙUtbī, Ibn Isfandiyār and Ibn al-Athīr, he was the rival of his brother Manūchihr for power in the Caspian lands, being sheltered by Sultan Maḥmūd at his court as a check on the loyalty of Manūchihr. Ibn al-Athīr mentions him in connection with Abū Kālījār, the ruler in Gurgān and Ṭabaristān during Masՙūd’s reign, but it is strange that neither Gardīzī nor Bayhaqī mention him at this time. See Bosworth, ‘On the chronology of the Ziyārids in Gurgān and Ṭabaristān’, Isl. XL (1964), pp. 32–4.

101Following the definitions given by H.G. Farmer, in his EI²art. ‘Ṭabl’.

102See Barthold, Turkestan, pp. 261–2.

103The celebrated vizier of the Buyids Mu’ayyid al-Dawla and Fakhr al-Dawla, called Kāfī al-Kufāt ‘The Supremely-Capable One’, also famed as a littérateur and Maecenas. See EI²art. ‘Ibn ՙAbbād’ (Cl. Cahen and Ch. Pellat); Kraemer, Humanism in the Renaissance of Islam. The Cultural Revival during the Buyid Age, pp. 259–72.

104Thus plausibly surmised by both Nazim and Ḥabībī; the mss. have turkān-i ṣ.l.ḥ., though it would be possible to read this phrase as turkān-i ṣulḥ ‘Turks in a treaty relationship’, such Turks being mentioned on the northern fringes of Transoxania by the Ḥudūd al-ՙālam, tr. pp. 118–19, as turkān-i āshtī. The Khalaj Turks are known to have nomadised at this time in what is now eastern Afghanistan, possibly brought there some centuries previously as part of the following of the Sakas or Hephthalites, and were utilised as auxiliary troops by Sebüktegin and later by his son Maḥmud. See Bosworth, The Ghaznavids, Their Empire in Afghanistan and Eastern Iran 994:1040 (Edinburgh, 1963), pp. 35–6, 109; EI²art. ‘Khaladj. i. History’ (Bosworth).

105This Shāh was Abū ՙAbdallāh b. Aḥmad (r. not earlier than 366–7/977 to 385/995), last of the Banū ՙIrāq or ՙArāq line of Kāth. As Gardīzī goes on to relate, Abū ՙAbdallāh was at this point overthrown and killed by the Ma’mūnid line of Gurgānj who were later to fall victims to the aggression and violence of Maḥmud of Ghazna, see above, p. 85–6. See Fedorov, ‘The Khwarazmshahs of the Banū ՙIrāq (fourth/tenth century)’, pp. 71–5.

106For these events leading to the capture and eventual death of Abū ՙAlī Sīmjūrī, see Barthold, Turkestan, pp. 261–3.

107This victory of Ma’mūn b. Muḥammad of Gurgānj inaugurated a new, but short-lived, line of Khwārazm Shāhs. See Bosworth, The New Islamic Dynasties, pp. 178–9 no. 89, and above, p. 74.

108For this name, see Rásonyi and Baski, vol. 1, p. 139, vol. 2, p. 813; Sümer, Türk devletleri tarihinde şahis adları, vol. 1, p. 125, vol. 2, p. 544. Tüzün can mean ‘smooth, self-controlled, well-behaved’, see Clauson, Etymological Dictionary, p. 576, but the second element of the name may also, but less probably, reflect the Orkhon Turkish official function of the todhun, see ibid., p. 457.

109The three texts have 13 Rajab for this date, but al-ՙUtbī, Vol. 1, p. 255, has 14 Rajab, giving the correct correspondence.

110See on Nū’s reign, Narshakhī, Tārīkh-i Bukhārā, tr. Frye, p. 99; Barthold, Turkestan, pp. 252–64; Frye, ch. ‘The Sāmānids’, pp. 154, 156–8; Treadwell, ‘The account of the Samanid dynasty’, pp. 158–60.

111Text here incomprehensible, with a defectively written word in the text.

112See below, n. 118.

113On Maḥmūd’s struggle with IsmāՙI l for the succession, see Muḥammad Nāẓim, The Life and Times of Sulṭān Maḥmūd of Ghazna pp. 38–41.

114According to Nazim, text, p. 59 n. 2, there is a lacuna in the text here.

115Thus making the reading ‘al-Jayhānī’ of Barthold, Turkestan, p. 265, and Nāẓim, text, p. 59, unnecessary.

116See for it, Yāqūt, Muՙjam al-buldān, Vol. 2, p. 391.

117See on Manṣūr’s reign, Barthold, op. cit., pp. 264–6; Frye, ch. ‘The Samanids’, pp. 158–9; Treadwell, ‘The account of the Samanid dynasty’, pp. 160–1; and for Bayhaqī’s account of these events, Bosworth, The History of Beyhaqi, Vol. 2, pp. 339–42.

118Correctly, Arslān Ilig Abu ’l-Ḥasan Naṣr b. ՙAlī of Uzgend or Özkend (Arslān Ilig being the designation of the subordinate Khān in the Qarakhanid hierarchy of rule, in this case, subordinate to his brother the Great Khān Abū Naṣr Aḥmad). He was the nephew of the first Qarakhanid to invade Transoxania, Hārūn or Ḥasan Bughrā Khān. See Pritsak, ‘Die Karachaniden’, p. 27; Doerfer, Türkische und Mongolische Elemente im Neupersischen. II. Türkische Elemente im Neupersischen, pp. 210–13 no. 661.

119Uzgend in Farghāna, now a mere village in the Andijān district of the Kyrgyz Republic, was at this time the main town of the region and the Ilig’s capital, later becoming that of the eastern branch of the Qarakhanid confederation. See EĪ² art. ‘Özkend’ (C.E. Bosworth); Valentina D. Goriatcheva, ‘A propos de deux capitales du kaganat karakhanide’, in Cahiers d’Asie centrale No. 9, Études karakhanides (Tashkent/Aix-en-Provence, 2001), pp. 104–14.

120See on ՙAbd al-Malik’s brief reign, Barthold, Turkestan, pp. 266–8; Nāẓim, Sulṭān Maḥmūd of Ghazna, pp. 43–5; Frye, ch. ‘The Samanids’, p. 159; Treadwell, ‘The account of the Samanid dynasty’, p. 161.

Part Four: The Early Ghaznavids

1Thus in Riḍāzāda Malik, kaldāniyān, for the kalāniyān of the mss., followed by Nazim and Ḥabībī but of no obvious meaning.

2See Barthold, op. cit., p. 271; Bosworth, ‘The titulature of the early Ghaznavids’, pp. 217–18; Bosworth, ‘The imperial policy of the early Ghaznawids’, Islamic Studies, Journal of the Central Institute of Islamic Research, Karachi, I/3 (1962), pp. 62–3.

3See Nāẓim, op. cit., p. 67; Bosworth, The History of the Saffarids of Sistan, p. 322.

4It is unclear to what this laconic reference refers.

5al-ՙUtbī details the rich presents of the specialities of the Inner Asian steppes and East Asia given to Maḥmūd on this occasion by the Qarakhanid, including khutuww. (In this context, probably narwhal or walrus ivory or possibly even fossilised mammoth tusks from Siberia, but the term is also used for rhinocerus horn; see, on the ambiguity of the term, Minorsky, Sharaf al-Zamān Ṭāhir on China, the Turks and India [London, 1942], pp. 82–3, and on this whole episode, Barthold, op. cit., p. 272.)

6This last of the Samanids was a son of Nūḥ (II) b. Manṣūr (I), hence brother of the two preceeding ephemeral Amirs. See Bosworth, The New Islamic Dynasties, pp. 170–1 no. 83.

7As Ḥabībī, p. 175 n. 4 says, the specific significance of this designation is mysterious, and the parallel sources (al-ՙUtbī, Ibn al-Athīr) throw no light on it; but it presumably points to the presence already of Indian troops, and here an Indian commander, in Maḥmūd’s army, already known from their mention at Zarang in 393/1003, according to the Tārīkh-i Sīstān. See Bosworth, The Ghaznavids, p. 110.

8Clearly a copyist’s error; a reading here like Kūchān would make sense geographically.

9On this Turkish name, tüz tash, lit. ‘straight, level stone’, see Sümer, Türk devletleri tarihinde şahis adları, Vol. 2, p. 551.

10Ibn Ẓāfir’s source seems to be the only other one mentioning this aid to Muntaṣir by the Yabghu of the Oghuz, and Ibn Ẓāfir specifically identifies him with Arslān Isrā’īl b. Seljuq, as Barthold also was inclined to do. See Treadwell, ‘The account of the Samanid dynasty’, p. 161; Barthold, Turkestan, p. 269. However, Pritsak suggested that this person was the original Yabghu of Jand on the lower Syr Darya, whose tenure of this ancient Turkish title, one going back to the succession of the Uyghur to the Orkhon Turkish state in 744, was continued into the time of Sultan Masՙūd of Ghazna, when the then Yabghu, Shāh Malik of Jand (son of the Yabghu who gave aid to Muntaṣir?) conquered Khwarazm as an ally of Masՙūd; this seems more probable. Arslān Isrā’īl’s assumption of the title of Yabghu was done in rivalry with the original Yabghu, there being strong hostility between the two branches of the Oghuz. See Cl. Cahen, ‘Le Malik-nâmeh et l’histoire des origines seljukides’, Oriens II (1949), pp. 46, 53–5; Bosworth, The Ghaznavids, pp. 221–2; Golden, An Introduction to the History of the Turkic Peoples, pp. 206. 218; and on earlier appearances of the title in Islamic sources, above, Part One, n. 131.

11Following the reading of Ḥabībī, cf. n. 2, bar ān kuh, presumably referring to the previously mentioned Kūhak.

12This is how Barthold, Turkestan, p. 269, interpreted the manuscripts’ wa Ghuzzān wa asīrān burdand.

13In both Nazim and Ḥabībī given as w.rghān, but this is the well-known town of Darghān on the left bank of the Oxus at the point where one first entered Khwarazmian territory, as noted by the Arab geographers. See Barthold, op. cit., pp. 142–3, 270; Le Strange, The Lands of the Eastern Caliphate, pp. 451–2.

14Cf. Rásonyi and Baski, Onomasticon turcicum, Vol. 2, p. 532: marïs, marïš.

15According to the Ḥudūd al-ՙālam, tr. Minorsky, p. 105, and other geographers detailed in Le Strange, op. cit., p. 400, Kushmayhan or Kushmīhan was a large village of the northeastern part of the Merv oasis, on the road across the Qara Qum desert to the Oxus and across it to Bukhara.

16This phrase concerning the accession of strength to Abū Ibrāhīm’s forces is supplied by Nazim in his text from parallel sources, cf. Barthold, Turkestan, p. 270.

17Ḥabībī, n. 10, states from al-ՙUtbī’s Yamīnī that Ibn (sic) Surkhāk was a member of the Samanid family.

18Muntaṣir’s Samanid kinsman Ibn Surhak had lured Muntaṣir into returning to Transoxania, whilst at the same time conniving with the Ilig against him. See Barthold, loc. cit.

19See on Muntaṣir and his brave but futile attempt to restore the fortunes of his house, Barthold, Turkestan, pp. 269–70; Nāẓim, Sulṭān Maḥmūd of Ghazna, pp. 45–6; Frye, ch. ‘The Sāmānids’, pp. 159–60; Treadwell, ‘The account of the Samanid dynasty’, pp. 161–2. Gardīzī telescopes events here considerably, with al-ՙUtbī giving more detail of Muntaṣir’s actions (he in fact made no fewer than four forays into Transoxania seeking to regain power there before he was finally killed).

20I.e. of the Hindūshāhī dynasty, whose powerful kingdom in northwestern India was based on Wayhind, modern Und near Attock. See EI²art. ‘Hindū-Shāhīs’ (C.E. Bosworth).

21See Nāẓim, op. cit., pp. 86–7; H.C. Ray, The Dynastic History of Northern India (Early Mediaeval Period) (Calcutta, 1931–6), Vol. 1, pp. 85–7.

22Following the interpretation of the text in Ḥabībī, p. 177 n. 8, mīrak to be read as marg?

23A stage on the road between Zarang and Bust, with a caravanserai, according to Ibn awqal, Kitāb Ṣūrat al-arḍ, Vol. 2, p. 422, tr. Kramers and Wiet, Configuration de la terre, Vol. 2, pp. 409–10. On this ending of Saffarid power in Sistan, see Bosworth, The History of the Saffarids of Sistan, pp. 325–7.

24On its identification and location of the text’s Bhāṭiya with Bhatinda, now in the southern part of the Indian Union’s Panjab province, see Nāẓim, op. cit., pp. 200–1, pace S.H. Hodivala, Studies in Indo-Muslim History, Vol. 1, pp. 138–9, identifying it with Bhera on the Jhelum river. The date of this expedition, unspecified by either al-ՙUtbī or Gardīzī, was probably the winter of 395/1004–5, see Nāẓim, Sulṭān Maḥmūd of Ghazna, pp. 202–3.

25The consonant ductus for the title is confused and seems to be trying to represent both Rāō and Rājah.

26The correction of Ḥabībī and Riḍāzāda to āb-i Sind ‘river of Sind’, i.e. the Indus, seems unlikely, given the distance of Bhatinda from that river; Nāẓim, op. cit., p. 100 n. 8, suggests that this Sāsind was the old name of a branch of the River Hakra.

27Thus in the mss. and Nazim; in Ḥabībī, Arg, both being possible, although ḥiṣār-i arg sounds tautologous. Ūk would be a variant spelling for the place to the north of Zarang, Ūq, frequently mentioned in the historical and geographical sources for Sistan; see Bosworth, The History of the Saffarids, pp. 77–8 and n. 217.

28See Nāẓim, op. cit., pp. 96–7; Bosworth, The Ghaznavids, pp. 52–3. Only five years later, Maḥmūd was to lead a further expedition against Multan, ostensibly because it was a nest of Carmathians but in reality because of the lure of the city’s great riches; see above, p. 83.

29In 390/999, at a time when the Samanids had not yet been totally defeated, Maḥmūd and the Ilig Naṣr had made an agreement to divide up the Samanid lands, with the Ilig having Transoxania and Maḥmūd the lands south of the Oxus (Barthold, Turkestan, pp. 266–7, 271–2; Nāẓim, op. cit., pp. 47–8). As is evident, this agreement did not last long.

30The text is dubious here: Nazim has s.m.r.w, and Ḥabībī has b.m.r.w : ?ba-marr-i u. Riḍāzāda’s interpretation ba-marw seems improbable.

31Barthold, op. cit., p. 273 n. 2, observed that the text here is badly mutilated. He translated here ‘[Maḥmūd’s soldiers] sang a Turkish song to a Khotanese melody’. Nazim’s text is unintelligible, and the translation here follows Ḥabībī, p. 179, cf. n. 1: dabdaba-yi tākhtan va āyīna ba-zadand.

32Thus according to the reading in Ḥabībī, p. 179, cf. n. 2, but there is obviously a lacuna and/or corruption in the text here. A difficulty about Ḥabībī’s reading is that the commander Ghāzī turns up alive and well in Sultan Masՙūd’s reign some two decades later as one of the Maḥmūdiyān, men of the former regime whose destruction the new Sultan then engineered, as related by Bayhaqī.

33Ḥabībī reads here chashm rasad where the mss., followed by Nazim, have ḥasham rasad. The whole sentence is difficult, and the translation here follows Barthold’s interpretation of this passage; he apparently read jān-rā ba-zadand for the khān-rā ba-zadand of Nazim, Ḥabībī and Riḍāzāda Malik, which last does not easily make sense.

34The vocalisation of this name as two syllables is indicated in Farrukhī’s mention of this victory over the Qarakhanids in Dīwān, ed. Muḥammad Dabīr-Siyāqī (Tehran, 1335/1956), p. 71 n. 35, cf. Ḥabībī, p. 179 n. 5. The site of Katar must have lain in Ṭukhāristān between Balkh and the Oxus river.

35 For this campaign, see Barthold, Turkestan, pp. 272–4; Nāẓim, Sulṭān Maḥmūd of Ghazna, pp. 48–51.

36Nāẓim, op. cit., p. 98, takes this to be a copyist’s mistake for Kh.w.ra, Khewra, the common name for the Salt Range in northwestern Panjab.

37Thus read by Ḥabībī for an unintelligible name in the mss.

38See Nāẓim, op. cit., pp. 98–9; H.C. Ray, The Dynastic History of Northern India, (Calcutta, 1931–6), Vol. 1, pp. 89–91.

39Ibid., p. 99.

40Ibid., pp. 103–4,

41Ibid., pp. 60–2.

42Ibid., pp. 192–3; Bosworth, ‘The imperial policy of the early Ghaznawids’, p. 68.

43Nāẓim, Sulṭān Maḥmūd of Ghazna, p. 164; Bosworth, op. cit., p. 60; Heinz Halm, ‘Fatimiden und Ghaznawiden’, in Studies in Honour of Clifford Edmund Bosworth. Vol. I. Hunter of the East, ed. I.R. Netton (Leiden, 2000), pp. 213–15.

44Nandana or Nārdīn (thus in ՙUtbī) lay on a northern spur of the Salt Range, in what is now the Jhelum District of Pakistani Panjab, commanding the route from the region of Attock into the Ganges Do’āb. See Nāẓim, op. cit., p. 91 and n. 4.

45Nāẓim, op. cit., pp. 91–3; Ray, The Dynastic History of Northern India, Vol. 1, pp. 94–101, 136.

46Nāẓim, op. cit., pp. 104–5; Ray, op. cit., Vol. 1, pp. 136–7. z.

47The mss. have Jaՙfarband, but the geographers situate the town of Jakarband or Jaqarband between Ṭāhiriyya and Darghān; see Le Strange, The Lands of the Eastern Caliphate, p. 451.

48See for his later career, EIr art. ‘Altuntaš’ (C.E. Bosworth).

49See for Bayhaqī’s detailed account of the the Ghaznavid invasion and annexation of Khwarazm (actually derived from al-Bīrūnī), Bosworth, The History of Beyhaqi, Vol. 2, pp. 369–400; also E. Sachau, ‘Zur Geschichte und Chronologie von Khwârazm’, SBWAW LXXIV (1873), pp. 292–311; Barthold, Turkestan, pp. 275–9; Nāẓim, Sulṭān Maḥmūd of Ghazna, pp. 56–60. Gardīzī does not mention that it was Maḥmūd’s brutal ultimatum to the Khwarazmians, demanding that they recognise him as first in their khuṭba, which precipitated the events giving him a convenient pretext for invading Khwarazm.

50The modern Bulandshahr, in Uttar Pradesh.

51See Nāẓim, op. cit., pp. 110–11; Ray, The Dynastic History of Northern India, Vol. 1, pp. 598–600; Hodivala, Studies in Indo-Muslim History, Vol. 1, pp. 146–8.

52Ray, however, in op. cit., Vol. 1, p. 606, see also Vol. 2, pp. 690–3, proposed to read n.n.da as a corrupt form of Bīdā = the first part of the name Vidyādhara, son and successor of Ganda/Gaṇḍa.

53The numerals here are confused in the texts, but the figure given here is Riḍāzāda Malik’s interpretation.

54Nāẓim, Sulṭān Maḥmūd of Ghazna, pp. 111–13; Ray, op. cit., Vol. 1, pp. 602ff.

55Nāẓim, op. cit. pp. 74–5. These river valleys ran down to the Kabul river, in the regions of Lamghān/Laghmān and Kunar lying to the north of modern Jalālābād. If Islam was indeed permanently implanted in these valleys, it must only have been in their lower reaches, since the highly mountainous, hardly penetrable interior parts of Kāfiristān continued to follow their blend of animism and polytheism till the campaign into Kāfiristān (modern Nūristān) of the Amir of Afghanistan ՙAbd al-Raḥmān Khān at the end of the nineteenth century. See EI²art. ‘Kāfiristān’ (C.E. Bosworth).

56I.e. the sub-Himalayan region of the Panjab west of the Chenab. See Nāẓim, op. cit., p. 105 n. 7, and al-Bīrūnī, Taḥqīq mā li ’l-Hind, English tr. E. Sachau, Alberuni’s India (London, 1910), Vol. 1, p. 208.

57Nāẓim, op. cit., p. 105; Ray, op. cit., Vol. 1, pp. 137–8.

58Nāẓim, op. cit., pp. 113–14; Ray, op. cit., Vol. 2, p. 689 n. 3. See also EI²art. ‘Gwāliyār’ (K.A. Nizami).

59See on the Ghaznavid ՙarḍ or review of troops, Bosworth, The Ghaznavids, pp. 122–4; EI²art. ‘Istiՙrāḍ, ՙArḍ’ (Bosworth).

60The ductus of this word is unclear. Ḥabībī, p. 186 n. 3, following a suggestion of Mīrzā Muḥammad Qazwīnī, suggests reading jast, and this is followed here in the translation.

61arash, i.e. the breadth between two outstretched arms. The reading here ‘two or three . . .’ follows Nazim’s text.

62This seems to be how the bridge of boats over the Oxus was constructed, but Barthold confessed, Turkestan, p. 282 n. 7, that the technical details here given were not entirely clear to him.

63Possibly the Muḥtājid prince Abu ’l-Muẓaffar Muḥammad b. Aḥmad, an early patron of the Ghaznavid poet Farrukhī. See Bosworth, ‘The rulers of Chaghāniyān in early Islamic times’, p. 12.

64The ductus here without any dots.

65Barthold, op. cit., pp. 282–3.

66Reading here, with Ḥabībī, p. 188 n. 8, ṭabarīhā.

67Following the texts’ muṣaffarī; Barthold, op. cit., p. 284 n. 5, read here maqāṣīrī (from Makassar?).

68The first word in this phrase is obscure, but khutuww is a familiar exotic product, see above, n. 5.

69Barthold, Turkestan, p. 284 n. 6, found this term incomprehensible. Ḥabībī, p. 189 n. 5, suggests a possible connection with khārchīnī, perhaps an arabisation of the word, said to be a hard substance from which such things as bells, cooking vessels, etc. could be made.

70Barthold, op. cit., pp. 283–4, translates this passage detailing the gifts that the two sovereigns presented to each other, noting various uncertainties in the texts.

71For the capture of Arslān Isrā’īl, see Nāẓim, Sulṭān Maḥmūd of Ghazna, pp. 63–4; Cahen, ‘Le Malik-nâmeh et l’histoire des origines seljukides’, p. 52.

72It seems that these Turkmens, who included the Seljuq family and their followers, had been auxiliaries in the service of the Qarakhanid ՙAlī b. Hārūn or Ḥasan Bughrā Khān, called ՙAlītegin, in the neighbourhood of Bukhara, but were now compelled to move towards the fringes of Khurasan when ՙAlītegin was temporarily driven out of Transoxania by his brother Yūsuf Qadïr Khān; see Bosworth, The Ghaznavids, pp. 223–4. The course of the relations between the ՙAlītegin and the Seljuqs is not, however, entirely clear; see the discussion in Cahen, op. cit., pp. 48ff.

73Thus in all the sources, but Jādhib, from the Arabic root j-dh-b ‘to drag, pull’ gives no sense here, and it is probable that we have here an early misspelling by copyists of the title Ḥājib ‘general, commander’.

74See Nāẓim, op. cit., pp. 64–5; Bosworth, op. cit., p. 224.

75This short section on the march to Nahrawāla and taking on supplies there is omitted from Nazim’s text, perhaps through inadvertence.

76In an Indian context, the Shamanīs/Sumaniyya often designate the Buddhists, but cf. Minorsky, ‘Gardīzī on India’, p. 627 n. 3, where H.W. Bailey is quoted as observing that śramaṇa means simply ‘ascetic’. It is indeed, highly improbable that there were any Buddhists in Kathiawar at this time.

77Thus written in the text; according to Ḥabībī, p. 191 n. 1, this is a Sanskrit word meaning ‘shrine’.

78Thus in Nazim’s text, to be preferred to Ḥabībī’s Jayḥūn; the latter has in fact Sayḥūn a few lines below. The name, normally used in Arabic geographical literature for the Jaxartes/Syr Darya, seems to have been used analogically for the Indus because this large river also divided off the Muslim lands from those of the unbelievers; see EI² art. ‘Mihrān’ (C.E. Bosworth).

79This ancient Indo-Aryan people of Sind and the southern Panjabi parts of the Indus valley had been fiercely hostile to the incoming Arabs of Muḥammad b. al-Qāsim at the opening of the eighth century AD, but are little mentioned until now, when they at times barred the advance of Ghaznavid raiders into the Indus valley lands. See EI² art. ‘Djāt’ (A.S. Bazmee Ansari).

80Nāẓim, Sulṭān Maḥmūd of Ghazna, pp. 115–21, 209–24.

81Barthold, Turkestan, p. 286; Nāẓim, op. cit., p. 56. It would seem that these two envoys represented Inner Asian potentates from beyond the lands of the Qarakhanids, whose ruling strata, at least, had not by this time become Muslim. If Qitā is a tribal name and not a geographical one, it may refer to the Qitan, who were probably Mongolic in ethnos and language and who occupied the Manchurian-northern China-Mongolian region after earlier Turkic peoples in Mongolia had migrated westwards towards the fringes of the Islamic lands (this last movement may be dimly referred to in Ibn al-Athīr’s laconic report, under the year 349/960, that 200,000 tents of the Turks converted to Islam). The Yughur are clearly the Uyghur, who had had a powerful state in Mongolia as successors of the Eastern Turk empire, but who had, after ca. 840, fallen into disunity, with new Uyghur political groupings forming in Eastern Turkestan, sc. the Tarim and Turfan basins, and in Kan-chu, the Kansu Corridor leading from eastern Turkestan to China. The faiths of these peoples ranged from animism and Buddhism to Manichaeism and Nestorian Christianity, hence their princes would not be acceptable to Maḥmūd as husbands for his womenfolk. See Golden, An Introduction to the History of the Turkic Peoples, pp. 162ff., 183ff.

82Bosworth, ‘The imperial policy of the early Ghaznawids’, p. 63; Bosworth, ‘The titulature of the early Ghaznavids’, p. 219.

83The campaigns against the heretical Muslims of Multan and against Kashmir not being counted here.

84Ḥabībī has here ‘four and a thousand hundreds’, presumably a clerical error since ‘1,400’ is correctly given a few lines below.

85Nāẓim, Sulṭān Maḥmūd of Ghazna, pp. 121–2. z.

86Nāẓim, op. cit., pp. 64–6; Cahen, ‘Le Malik-nâmeh et l’histoire des origines seljukides’, pp. 52, 55–6. Balkhān Kūh was the mountainous refuge region to the east of the Caspian Sea, now in the western part of the Turkmen Republic.

87See on this Ḥudūd al-ՙālam, tr. Minorsky, p. 64, comm. p. 200.

88Of unclear etymology, unless the first element is related to the kinship terms ekē ‘female relative older than oneself but younger than one’s father’, or eke ‘father, uncle, elder brother’ (Clauson, An Etymological Dictionary of Pre-Thirteenth Century Turkish, p. 100). Sümer, Türk devletleri tarihinde şahis adları, Vol. 2, p. 545, merely registers this name as Eygü (?) Tigin without attempting any etymology or meaning for it.

89Nāẓim, op. cit., pp. 80–3; Bosworth, ‘The imperial policy of the early Ghaznawids’, pp. 70–2; Bosworth, The Ghaznavids, pp. 53–4.

90Nāẓim, op. cit., pp. 123–4.

91On Maḥmūd’s favourite, his cup-bearer Ayāz, see EIr art. ‘Ayāz, Abu ’l-Najm Ūymāq’ (J. Matīnī), and for his role in later Persian literature, Gertrud Spiess, Maḥmūd von Ġazna bei Farīdu’dīn ՙAṭṭār (Basel, 1959).

92Dāya, presumably cognate with older Turkish tagāy, dagāy, Ottoman dayï ‘maternal uncle’ (Clauson, op. cit., p. 474); it may be that ՙAlī was connected in some way by marriage with Sultan Maḥmūd.

93On this Indian commander’s name, see Bosworth, The History of Beyhaqi, Vol. 2, p. 237 n. 32.

94The two mss. have w.l.ḥ, which Ḥabībī suggested should be amended to Rukhkhaj, the classical Arachosia, the region in which Tegīnābād was situated; from Bayhaqī we learn that Muḥammad’s place of incarceration was a fortress called Kūhtīz.

95On this first sultanate of Muḥammad, see R. Gelpke, Sulṭān Masՙūd I. von Ġazna. Die drei ersten Jahre seiner Herrschaft (421/1030–424/1033) (Munich, 1957), pp. 31–43; Bosworth, The Ghaznavids, pp. 228–9; EI²art. ‘Muḥammad b. Maḥmūd b. Sebüktigin’ (Bosworth). It may be noted that Gardīzī’s account of the sultanate is not unfavourable to Muḥammad, rather more so than that of Bayhaqī.

96The arrest and incarceration of these two commanders marks the beginning of a systematic campaign by the new sultan Masՙūd, extending even to his own uncle Yūsuf b. Sebüktegin, to remove from positions of power those connected in any way with his brother Muḥammad’s sultanate or, indeed, connected at all with the ancien régime of Maḥmūd, these last being stigmatised as the Maḥmūdiyān or Pidariyān. See Gardīzī’s own words concerning this vendetta, above, p. 100; Gelpke, op. cit., pp. 48–53; Bosworth, The Ghaznavids, pp. 101–2, 230–4.

97This is the celebrated vizier of Maḥmūd, who was now brought back to serve as Masՙūd’s minister until his death two years later (see above, p. 101). See Nāẓim, Sultān Maḥmūd of Ghazna, pp. 130–1, 135–6; Gelpke, op. cit., pp. 59–63; Bosworth, The Ghaznavids, pp. 60–1, 71–3; EI²art. ‘Maymandī’ (M. Nazim, C.E. Bosworth).

98See on this governor and castellan, Ray, The Dynastic History of Northern India, Vol. 1, pp. 137–8; Hodivala, Studies in Indo-Muslim History, Vol. 1, pp. 145–6.

99See on the Mīkālī family, above, Part Three, n. 85.

100I.e. from the Fatimid caliph al-Ḥākim.

101For this episode of Ḥasanak’s death, graphically and movingly described by Bayhaqī, see Gelpke, op. cit., pp. 67–71; Bosworth, The Ghaznavids, pp. 71, 182–4; Meisami, Persian Historiography, pp. 88–94; Halm, ‘Fatimiden und Ghaznawiden’, pp. 218–19.

102Text, Il-Yārūq, but the correct form is ascertainable from Bayhaqī, see Bosworth, The History of Beyhaqi, Vol. 1, p. 165 etc.

103See above, pp. 103–4.

104Earlier, it was stated that Maḥmūd held Majd al-Dawla at Ghazna.

105For these events in Makrān, see Bosworth, ‘Rulers of Makrān and Quṣdār in the early Islamic period’, St.Ir. XXIII (1994), pp. 206–7.

106Cahen, ‘Le Malik-nâmeh et l’histoire des origines seljukides’, p. 56.

107See on the new vizier, from a family which, like so many of the Ghaznavids’ officials, had apparently served the Samanids, Bosworth, The Ghaznavids, pp. 58, 61–2, 72.

108See Nazir Ahmad, ‘A critical examination of Bayhaqī’s narration of the Indian expeditions during the reign of Masՙud of Ghazna’, pp. 39–44, for a discussion of this expedition and the location of Sarastī.

109Thus in the texts of Ḥabībī and Riḍāzāda Malik, following the zāwi-hā of the mss.; Nazim has wādī-hā ‘river valleys’, and also the Tehran edition (presumably that of 1315/1936–7), according to Ḥabībī.

110Following Riḍāzāda Malik, khishtī ba-zad.

111Abū Kālījār is here the son of Surkhāb, but is said by the local historian Ibn Isfandiyār to have been the son and successor of Manuchihr b. Qābūs when the latter died in 424/1033, reigning till his own death in 441/1049–50. See Ibn Isfandiyār, tr. Browne, An Abridged Translation of the History of Ṭabaristán, p. 235. The history of these later Ziyārids is, however, only sketchily known.

112Bayhaqī describes Masՙūd’s campaign in the Caspian provinces in considerable detail, but does not conceal the fact that the Sultan’s violent and avaricious behaviour created a most unfavourable impression on contemporaries, with complaints being carried as far as the caliph’s court in Baghdad and to Mecca. See Bosworth, The History of Beyhaqi, Vol. 2, pp. 106–7, 109–23; also Bosworth, The Ghaznavids, pp. 74–5, 90–1; Meisami, Persian Historiography, pp. 95–8.

113This seems to be the meaning here of tabāhī īn az sālār bisyar ast, i.e. effort has been dissipated by sending several commanders at various times, with no unified plan of campaign.

114Khwāja Ḥ usayn was actually to be captured by the Seljuqs in the course of this ensuing battle (see above, p. 103), and thereafter appears to have entered their service; in the reign of the Great Seljuq Sultan Ṭoghrïl, successor to the Ghaznavids in Khurasan and Persia, he fulfilled various high offices, possibly as vizier. See H. Bowen, ‘Notes on some early Seljuqid viziers’, BSOAS XX (1957), pp. 105, 107–8.

115Possibly to be identified, according to Ḥabībī, n. 1, with a place S.w.s.qān or S.w.sn.qān near Merv.

116This is how Ḥabībī interprets the text’s jāmiՙ-i ՙarabī.

117These must have been captives and hostages brought back from the previous Ṭabaristān campaign, see above, p. 102.

118On Begtughdї’s defeat, see Cahen, ‘Le Malik-nâmeh et l’histoire des origines seljukides’, p. 49; Bosworth, The Ghaznavids, pp. 248–9; Meisami, op. cit., pp. 98.

119I.e. of the Indian division within the main Ghaznavid army, which Suvendharāy had commanded during Muḥammad’s brief reign, see above, p. 000.

120According to Hodivala, Studies in Indo-Muslim History, Vol. 1, p. 163, the name of Tilak or Tilaka’s father Jahlan/Jalhaṇa/Jahlaṇsi is an old one found in dynastic lists and inscriptions. Presumably Tilak was at his juncture appointed commander of the Indians in Bānha’s place.

121There are some chronological discrepancies in the accounts of Gardīzī and the other two sources describing Aḥmad’s revolt, his final vanquishing and his death, sc. Bayhaqī and Ibn al-Athīr. See for a discussion of the whole episode, problems involved with it and identification of the places mentioned in its course, Nazir Ahmad, ‘A critical examination of Bayhaqī’s narration of the Indian expeditions’, pp. 53–6.

122Bayhaqī has a description of the new Masՙūdī Palace, which had taken four years to build and for whose construction and decoration a total of fourteen million dirhams had been expended on the wages of craftsmen and payments to forced labour levies. See Bosworth, The History of Beyhaqi, Vol. 2, pp. 216–17.

123This phrase supplied from Bayhaqī and Firishta.

124Hānsī is situated in the modern Hariyana province of the Indian Union, to the northwest of Delhi. After Masՙūd’s conquest of the fortress from its Chawhān ruler, it became briefly a forward base for Ghaznavid armies launching raids against Delhi, but was recaptured during Mawdūd’s reign by the Rājā of Delhi, Mahīpāl, and a coalition of other princes and not regained by the Muslims till Ghurid times. See Nazir Ahmad, op. cit., p. 59; EI² art. ‘Hānsī’ (J. Burton-Page).

125Bayhaqī calls this place ‘The Virgin Fortress’, Qalՙat al-ՙAdhrā’.

126dīra; in the mss., dara ‘valley’.

127There are problems arising out of the divergent chronologies of Gardīzī and Bayhaqī for these campaigns of Masՙūd in India. These are discussed by Nazir Ahmad in his detailed study of the whole episode, in op. cit., pp. 58–66; he opts for the more careful chronology of Bayhaqī, for which see Bosworth, The History of Beyhaqi, Vol. 2, pp. 208–9.

128Gardīzī does not record the fact that, whilst the Amir was campaigning in India, his general Sübashï, ordered to march out from his base at Nishapur, had been defeated by the Turkmens and had had to withdraw to Herat, allowing the Turkmens to occupy Nishapur unopposed. See Meisami, Persian Historiography, pp. 99–100.

129This is Qarakhanid prince Böritegin, son of the Ilig Naṣr b. ՙAlī, and later better known in the sources as Ibrāhīm Tamghach Khān (d. 460/1068); he was from the Western branch of the family whose possessions were centred on Transoxania. At this time, however, Böritegin had recently escaped from custody of his enemies, the sons of his distant kinsman ՙAlītegin, and had come, via a stay in Uzgend/Özkend, to the upper Oxus region, where he had gathered together a force of the predatory Kumījīs and Kanjīna Turks of the Buttamān mountains region (see above, Part Three, n. 49) and was harrying the Ghaznavid dependencies of Wakhsh and Khuttal. See Barthold, Turkestan, pp. 300–1; Pritsak, ‘Die Karachaniden’, pp. 36–7.

130See above, n. 34. Ḥabībī in his text, p. 202 n. 1, quotes a verse of Farrukhī on Maḥmūd’s earlier meeting there with the Qarakhanid Yūsuf Qadïr Khān.

131The Sultan’s invasion of Transoxania was undertaken at the worst time of the year for weather, the winter months of 430/1038–39, as his advisers had pointed out but to no avail. Whereas Bayhaqī’s account does not conceal the terrible hardships endured by the army, Gardīzī relates the episode in a comparatively neutral fashion; the expedition was in fact a disaster, achieving nothing and diminishing Masՙūd’s prestige. See Barthold, op. cit., pp. 301–2; Meisami, op. cit., pp. 77, 100–1.

132The operations against this bandit chief are described in detail by Bayhaqī; see Bosworth, The History of Beyhaqi, Vol. 2, pp. 241–3.

133The Yabghu here is clearly the Seljuq Mūsā Yabghu, successor in this title to his close kinsman Arslān Isrā’īl (whom Maḥmūd had imprisoned in India towards the end of his reign, see Nāẓim, Sulṭān Maḥmūd of Ghazna, pp. 63–4) holding this title in rivalry with its original holder amongst the Oghuz, the ruler in Jand, see above, n. 10. The problems regarding the various holders of the title are discussed by Cahen, ‘Le Malik-nâmeh et l’histoire des origines seljukides’, pp. 53–5. Here in this passage of Gardīzī, Masՙūd seems here to have regarded Mūsā Yabghu as nominal head of the Seljuqs (it was to him that the Sultan had the Turkmens’ severed heads delivered, see below), although insofar as there was any central direction at all amongst the Seljuqs, it was Ṭoghril Beg and Chaghri Beg who in practice exercised this. See Bosworth, The Ghaznavids, pp. 244–5.

134āncha mā khwāstīm ba-kard amīr khwud ba-kard.

135There is thus no mention by Gardīzī of the several months’ occupation of Nishapur by the Seljuqs, whose arrival there is described in detail by Bayhaqī.

136Masՙūd’s forces had reoccupied Nishapur towards the beginning of 431/end of 1039 and opening of 1040, see Bosworth, op. cit., p. 251.

137Following Ḥabībī, ū-rā andar na-yāft.

138The people of the towns and cities of northern Khurasan like Bāward/Abīward and Sarakhs were clearly despairing of securing any protection from the Sultan against the Seljuqs, and were making their own terms with the Turkmens. The most notable instance of this was the ease with which Nishapur passed temporarily under Seljuq control in 429–30/1038–9 (an episode unnoted by Gardīzī but described in detail by Bayhaqī; see on this Bosworth, The Ghaznavids, pp. 244, 252ff.). See also ibid., pp. 264–5.

139Bayhaqī relates that 500 defectors from the Ghaznavid side had joined the Seljuq group of Yināliyān (i.e. those under the leadership of Ibrāhīm Ināl) at some point before the battle of Dandānqān, and that before this, former troops of Ghaznavid commanders whom Masՙūd had vindictively hunted down and killed, such as his uncle Yūsuf b. Sebüktegin, the Ḥājib ՙAlī Qarīb, Eryārūq, Ghāzī, etc., had joined the Turkmens.

140This was one of the decisive battles in the history of the eastern Islamic world, for the way was now open for the Seljuqs to sweep westwards, confront the powers of western Persia like the Kakuyids and Buyids, and establish the Great Seljuq sultanate, whilst the Ghaznavids, whilst still a great power, were forced to concentrate their vision eastwards to northern India. Bayhaqī has a very detailed account of the course of the Dandānqān battle, see The History of Beyhaqi, Vol. 2, pp. 309ff. See also Meisami, Persian Historiography, pp. 102–4.

141Thus in Ḥabībī’s text, ihmāl warzīda; Nazim has iḥtimāl warzīda ‘had been wavering and uncertain’.

142Bosworth, The Later Ghaznavids, pp. 9, 139.

143Bayhaqī’s narrative shows, however, that several of the Sultan’s advisers were against this retreat to India and thought that the situation in eastern Afghanistan could be held, see ibid., pp. 15–16,

144Ibid., pp. 9–13. The exact location of Hupyān (if this is the correct vocalisation of the name) is uncertain; see ibid., pp. 12–13.

145Bayhaqī calls these disruptive elements, in what is now eastern Afghanistan, Khalaj; these were almost certainly Turks but may well have become mingled with Afghan tribesmen, i.e. Pashtuns. See above, Part Three, n. 104, and Bosworth, The Later Ghaznavids, pp. 13–14 and n. 27, where it is said that the first mention in Islamic sources of Afghans appears to be in the Ḥudūd al-ՙālam. However, we now have mention of abagano, abgano, pl. abaganano, from some two centuries earlier in the recently discovered letters in the Middle Iranian Bactrian language from the Rōb/Ro’b and Gūzgān regions of northern Afghanistan (information from Professor N. Sims-Williams, and cf. above, Part Two, n. 18).

146See on these places, Bosworth, The Later Ghaznavids, pp. 16–17.

147In Bayhaqī this fortress is named as Naghar.

148Barthold, Turkestan, p. 303; Bosworth, op. cit., pp. 14–17. It is at this point that the ninth volume of Bayhaqī’s Mujalladāt ends, so that Gardīzī’s narrative assumes a particular importance as the sole contemporary source on the end of Masՙūd’s sultanate.

149This is modern Marigala, in a pass of the low hills separating Attock from Rawalpindi; Masՙūd had thus just crossed the Indus at Wayhind when his troops mutinied. See Bosworth, op. cit., pp. 19, 140.

150Ibid., pp. 20, 140.

151Text in the two mss., dīnawar, as also in Bayhaqī. Peshawar has been conjectured as a possible reading, but Ḥabībī, p. 205 n. 3, suggests that we should read d.n.pūr, d.n.būr, this being a site near the later Jalālābād, and this is more plausible; Jūzjānī, in his Ṭabaqāt-i nāṣirī (see above, Part One, n. 112) says that the battle took place in Nangrahār, i.e. in this same region of the middle Kabul river valley. Dunpūr is in fact known as a town, the resort of merchants, from the Ḥudūd al-ՙālam, tr. Minorsky, pp. 63, 72, 92, being placed on the opposite bank of the Kabul river from Lamghān, cf. comm. p. 252. The town is mentioned, in a later form, in the Bābur-nāma, and its name is possibly from the Sanskrit Udyānapūra.

152This duՙā or invocation of long life for ՙAbd al-Rashīd implies that Gardīzī was writing his history during that Amir’s reign, i.e. between ?440/1049 and 443/1052, thus confirming the lauding by Gardīzī of ՙAbd al-Rashīd as reigning monarch in the texts of Nazim, pp. 61–2, and Ḥabībī, p. 174, above.

153It would be interesting to know the circumstances in which ՙAbd al-Rashīd was present near the battlefield. Did he have supporters of his own with him, and was there perhaps a feeling amongst the Ghaznavid troops that he, as son of Maḥmūd and the eldest representative of the family, had a superior claim to the succession? Mawdūd obviously felt a strong need to conciliate him. But despite Mawdūd’s promises to him in return for his neutrality in the struggle, ՙAbd al-Rashīd was after the battle arrested and imprisoned by Mawdūd during the whole of the latter’s reign; see Bosworth, The Later Ghaznavids, pp. 23, 25.

154For the battle, which took place in Rajab or Shaՙbān 432/March or April 1041, see Bosworth, op. cit., pp. 22–4, 140, and for Mawdūd’s vengeance on Muḥammad and his supporters after it, ibid., pp. 24, 140.

SELECT BIBLIOGRAPHY

Note. This bibliography does not include some items which are mentioned only once or so in the text of the book and are of peripheral relevance to Gardīzī’s work; but where such items occur in the book, full bibliographical details are given there.

Abbott, Nabia, Two Queens of Baghdad. Mother and Wife of Hārūn al-Rashīd (Chicago, 1946).

Ahmad, Nazir, ‘A critical examination of Bayhaqī’s narration of the Indian expeditions during the reign of Masՙud of Ghazna’, in Yād-nāma-yi Abu ’l-Faḍl Bayhaqī. Majmūՙa-yi sukhanrānī-hā-yi Majlis-i Buzurgdāsht-i Abu ’l-Faḍl-i Bayhaqī, Mashhad, 21 tā 25 shahrīwarmāh 1349, ed. Mashhad University Faculty of Letters and Human Sciences (Mashhad, 1350/1971), English section, pp. 34–83.

Anon., Ḥudūd al-ՙālam, English tr. and commentary by V. Minorsky, with a Preface by V.V. Barthold, as Ḥudūd al-ՙālam ‘The Regions of the World’, a Persian Geography 372 AH–982 AD, GMS NS XI (London, 1937), 2nd ed. with additional material by the late Professor Minorsky [and with comments by Dr Manuchihr Sotudeh and ‘The second series of Addenda’ by V. Minorsky], ed. C.E. Bosworth (London, 1970).

Anon. (by Ibn Shādī?), Mujmal al-tawārīkh wa ’l-qiṣaṣ, ed. Malik al-Shuՙarā’ Mīrzā Taqī Bahār (Tehran, 1318/1939).

Anon., Tārīkh-i Sīstān, ed. Malik al-Shuՙarā’ Mīrzā Taqī Bahār (Tehran, 1314/1935), English tr. Milton Gold as The Tārikh-e Sistān, IsMEO, Serie Orientale Roma, Vol. XLVIII/Literary and Historical Texts from Iran, 2 (Rome, 1976).

Bahār, Mīrzā Taqī, Malik al-Shuՙarā’, Sabk-shināsī yā tārīkh-i taṭawwur-i nathr-i fārsī, 3 vols. (Tehran, 1337/1958).

al-Balādhurī, Aḥmad b. Yaḥyā, Futūḥ al-buldān, ed. M.J. De Goeje as Liber expugnationis regionum (Leiden, 1866).

Barthold, Wilhelm (V.V. Bart’old), Turkestan down to the Mongol Invasion, GMS N.S. V (London, 1928), 3rd ed. with an additional chapter, ed. C.E. Bosworth (London, 1968).

Bayhaqī, Abu ’l-Faḍl, Tārīkh-i Masՙūdī, English tr. C. Edmund Bosworth as The History of Beyhaqi (The History of Sultan Masՙud of Ghazna, 1030–1041), 3 vols. (New York, 2009).

al-Bīrūnī, Abū Rayḥān, al-Āthār al-bāqiya ՙan al-qurūn al-khāliya, ed. Eduard Sachau (Leipzig, 1878), English tr. E. Sachau as The Chronology of Ancient Nations (London, 1879).

——, Taḥqīq mā li ’l-Hind, English tr. E.C. Sachau as Alberuni’s India. An Account of the Religion, Philosophy, Literature, Geography, Chronology, Astronomy, Customs, Laws and Astrology of India about AD 1030, 2 vols. (London, 1910).

Bosworth, C. Edmund, ‘The early Islamic history of Ghūr’, CAJ VI (1961), pp. 116–33, reprinted in C.E. Bosworth, The Medieval History of Iran, Afghanistan and Central Asia, Variorum Reprints, Collected Studies Series CS 56 (London, 1977), no. IX.

——, ‘The titulature of the early Ghaznavids’, Oriens XV (1962), pp. 210–33, reprinted in The Medieval History of Iran, Afghanistan and Central Asia, no. X.

——, The Ghaznavids, their Empire in Afghanistan and Eastern Iran 994:1040, History, Philosophy and Economics 17 (Edinburgh, 1963), 2nd ed. with additional bibliography (Beirut, 1973).

——, ‘Early sources for the history of the first four Ghaznavid sultans (977–1040)’, IQ VII (1963), pp. 3–22, reprinted in The Medieval History of Iran, Afghanistan and Central Asia, no. XIII.

——, ‘On the chronology of the Ziyārids in Gurgān and Ṭabaristān’, Isl. XL (1964), pp. 25–34, reprinted in The Medieval History of Iran, Afghanistan and Central Asia, no. II.

——, ‘Notes on the pre-Ghaznavid history of eastern Afghanistan’, IQ IX (1965), pp. 12–24, reprinted in The Medieval History of Iran, Afghanistan and Central Asia, no. XIV.

——, Sīstān under the Arabs, from the Islamic Conquest to the Rise of the Ṣaffārids (30–250/651–864), IsMEO, Centro Studi e Scavi Archeologici in Asia, Reports and Memoirs XI (Rome, 1968).

——, ‘The armies of the Ṣaffārids’, BSOAS XXXI (1968), pp. 534–54, reprinted in The Medieval History of Iran, Afghanistan and Central Asia, no. XVII.

——, ‘Abū ՙAbdallāh al-Khwārazmī on the technical terms of the secretary’s art. A contribution to the administrative history of mediaeval Islam’, JESHO XII (1969), pp. 113–64, reprinted in Medieval Arabic Culture and Administration, Variorum Reprints, Collected Studies Series CS 165 (London, 1982), no. XV.

——, ch. ‘The Ṭāhirids and Ṣaffārids’, in The Cambridge History of Iran. Vol. 4. The Period from the Arab Invasion to the Saljuqs, ed. Richard N. Frye (Cambridge, 1975), pp. 90–135.

——, ch. ‘The early Ghaznavids’, in The Cambridge History of Iran. Vol. 4. The Period from the Arab Invasion to the Saljuqs, ed. Richard N. Frye (Cambridge, 1975), pp. 162–97.

——, ‘The rulers of Chaghāniyān in early Islamic times’, Iran JBIPS XIX (1981), pp. 1–20, reprinted in C. Edmund Bosworth, The Arabs, Byzantium and Islam. Studies in Early Islamic History and Culture, Variorum Collected Studies Series CS 529 (Aldershot, Hants., 1996), no. XXI.

——, The History of the Saffarids of Sistan and the Maliks of Nimruz (247/861 to 949/1542–3), Columbia Lectures on Iranian Studies 8 (Costa Mesa and New York, 1994).

——, ‘Rulers of Makrān and Quṣdār in the early Islamic period’, St.Ir. XXIII (1994), pp. 199–209.

——, The New Islamic Dynasties. A Chronological and Genealogical Manual (Edinburgh, 1996).

——, ‘Notes on some Turkish names in Abu ’l-Faḍl Bayhaqī’s Tārīkh-i Masՙūdī’, Oriens XXXVI (2001), pp. 299–313.

——, ‘Further notes on the Turkish names in Abu ’l-Faḍl Bayhaqī’s Tārīkh-i Masՙūdī’, to appear in Festschrift for Dr Farhad Daftary (London, 2010).

——, ‘The appearance and establishment of Islam in Afghanistan’, in Islamisation de l’Asie centrale. Processus locaux d’acculturation du VII^e du XI^e siècle, ed. E. de la Vaissière, Cahiers de Studia Iranica, 39 (Paris, 2008), pp. 97–114.

Bosworth, C. Edmund and Sir Gerard Clauson, ‘Al-Xwārazmī on the peoples of Central Asia’, JRAS (1965), pp. 2–12, reprinted in C.E. Bosworth, The Medieval History of Iran, Afghanistan and Central Asia, Variorum Reprints, Collected Studies Series CS56 (London, 1977), no. XX.

Bowen, H., ‘Notes on some early Seljuqid viziers’, BSOAS XX (1957), pp. 105–10.

Cahen, Cl., ‘Le Malik-nâmeh et l’histoire des origines seljukides’, Oriens II (1949), pp. 31–65.

Clauson, Sir Gerard, An Etymological Dictionary of Pre-Thirteenth Century Turkish (Oxford, 1972).

Crone, Patricia, Slaves on Horses. The Evolution of the Islamic Polity (Cambridge, 1980).

Crone, Patricia and Luke Treadwell, ‘A new text on Ismailism at the Samanid court’, in Texts, Documents and Artefacts. Islamic Studies in Honour of D.S. Richards, ed. Chase F. Robinson (Leiden, 2003), pp. 37–67.

Czeglédy, K., ‘Gardīzī on the history of Central Asia (AD 746–780)’, AOHung XXVII (1973), pp. 257–67.

Daniel, Elton L., The Political and Social History of Khurasan under Abbasid Rule 747–820, Bibliotheca Islamica (Minneapolis and Chicago, 1979).

Doerfer, Gerhard, Türkische und mongolische Elemente im Neupersischen, unter besonderer Berücksichtigung älterer neupersischer Geschichtsquellen, vor allem der Mongolen- und Timuridenzeit, Akademie der Wissenschaften und der Literatur, Veröffentlichungen der Orientalischen Kommission, Bd. XVI, XIX–XXI, 4 vols. (Wiesbaden, 1963–75).

Eisener, Reinhard, Zwischen Faktum und Fiktion. Eine Studie zum Umayyadenkalifen Sulaimān b. ՙAbdalmalik und seinem Bild in den Quellen (Wiesbaden, 1987).

Fedorov, Michael, ‘The Khwarazmshahs of the Banū ՙIrāq (fourth/tenth century)’, Iran JBIPS XXXVIII (2000), pp. 71–5.

Frye, Richard N. ch. ‘The Sāmānids’, in The Cambridge History of Iran. Vol. 4. The Period from the Arab Invasion to the Saljuqs, ed. Richard N. Frye (Cambridge, 1975), pp. 136–61.

——, Bukhara, the Medieval Achievement, 2nd ed. (Costa Mesa, 1996).

Gelpke, Rudolf, Sulṭān Masՙūd I. von Ġazna. Die drei ersten Jahre seiner Herrschaft (421/1030–424/1033), Inaugural Dissertation . . . der Universität Basel (Munich, 1957).

Gibb, Hamilton A. R., The Arab Conquests in Central Asia (London, 1923).

Golden, Peter B., An Introduction to the History of the Turkic Peoples. Ethnogenesis and State-Formation in Medieval and Early Modern Eurasia and the Middle East, Turcologica Band 9 (Wiesbaden, 1992).

Halm, Heinz, ‘Fatimiden und Ghaznawiden’, in Studies in Honour of Clifford Edmund Bosworth. Vol. I. Hunter of the East: Arabic and Semitic Studies, ed. Ian R. Netton (Leiden, 2000), pp. 209–21.

Ḥamza b. al-Ḥasan al-Iṣfahānī, Ta’rīkh Sinī mulūk al-arḍ wa ’l-anbiyā’ (Beirut, 1961).

Hodivala, S.H., Studies in Indo-Muslim History. A Critical Commentary on Elliot and Dowson’s History of India as Told by its Own Historians, 2 vols. (Bombay, 1937–57).

Ibn Ḥawqal, Muḥammad b. ՙAlī, Kitāb Ṣūrat al-arḍ, ed. J.H. Kramers, 2 vols. (Leiden, 1938–9), French tr. J.H. Kramers and Gaston Wiet as Configuration de la terre, Collection UNESCO d’oeuvres représentatives, série arabe, 2 vols. (Paris, 1964).

Ibn Isfandiyār, Muḥammad b. Ḥasan, Tārīkh-i Ṭabaristān, partial English tr. E.G. Browne as An Abridged Translation of the History of Ṭabaristán Compiled about AH 613 (AD 1216) by Muḥammad b. al-Ḥasan b. Isfandiyár, GMS II (Leiden and London, 1905).

Ibn Khallikān, Aḥmad b. Muḥammad, Wafayāt al-aՙyān, ed. Iḥsān ՙAbbās, 8 vols. (Beirut, 1968–72), English tr. Baron McGuckin de Slane as Ibn Khallikan’s Biographical Dictionary, 4 vols. (Paris, 1842–71).

Kaabi, Mongi, Les Ṭāhirides au Hurāsān et en Iraq (III^ème H./IX^ème J.-C.), 2 vols. (Tunis, 1983).

Kennedy, Hugh, The Early Abbasid Caliphate. A Political History (London and Totowa, N.J., 1981).

Kraemer, Joel L., Humanism in the Renaissance of Islam. The Cultural Revival during the Buyid Age (Leiden, 1986).

Lassner, Jacob, The Shaping of ՙAbbāsid Rule (Princeton, 1980).

Lazard, Gilbert, La langue des plus anciens monuments de la prose persane, Études linguistiques II (Paris, 1963).

Le Strange, Guy, The Lands of the Eastern Caliphate. Mesopotamia, Persia, and Central Asia from the Moslem Conquest to the Time of Timur, Cambridge Geographical Series (Cambridge, 1905).

Løkkegaard, Frede, Islamic Taxation in the Classic Period, with Special Reference to the Circumstances in Iraq (Copenhagen, 1950).

Madelung, Wilferd, ch. ‘The minor dynasties of northern Iran’, in The Cambridge History of Iran. Vol. 4. The Period from the Arab Invasion to the Saljuqs, ed. Richard N. Frye (Cambridge, 1975), pp. 198-249.

al-Maqdisī, Muḥammad b. Aḥmad, Aḥsan al-taqāsīm fī maՙrifat al-aqālīm, ed. M.J. De Goeje as Descriptio imperii moslemici, BGA III² (Leiden, 1906), partial French tr. André Miquel as La meilleure rēpartition pour la connaissance des provinces (Damascus, 1963).

Marquart, Josef, and J.J.M. de Groot, ‘Das Reich Zābul und der Gott Žūn vom 6–9 Jahrhundert’, in Festschrift Eduard Sachau zum siebigsten Geburtstage gewidmet von Freunden und Schülern, ed. Gotthold Weil (Berlin, 1915), pp. 248–92.

Martinez, A.P., ‘Gardīzī’s two chapters on the Turks’, AEMAe II (1982), pp. 109–217.

Meisami, Julie S., Persian Historiography to the End of the Twelfth Century (Edinburgh, 1999).

Minorsky, Vladimir, Sharaf al-Zamān Ṭāhir Marvazī on China, the Turks and India, Royal Asiatic Society, James G. Forlong Fund Vol. XXII (London, 1942).

——, ‘Gardīzī on India’, BSOAS XII (1947–9), pp. 625–40, reprinted in Iranica. Twenty Articles/Bīst maqāla-yi Mīnūrskī, Publications of the University of Tehran no. 775 (Tehran, 1964), pp. 200–15.

Mottahedeh, Roy, ch. ‘The ՙAbbāsid caliphate in Iran’, in The Cambridge History of Iran. Vol. 4. The Period from the Arab Invasion to the Saljuqs, ed. Richard N. Frye (Cambridge, 1975), pp. 57–89.

Narshakhī, Muḥammad b. Jaՙfar, Tārīkh-i Bukhārā, English tr. Richard N. Frye as The History of Bukhara. Translated from a Persian Abridgement of the Arabic Original by Narshakhī (Cambridge, Mass., 1954).

Nāẓim, Muḥammad, The Life and Times of Sulṭān Maḥmūd of Ghazna (Cambridge, 1931).

Omar, Farouk, ՙAbbāsiyyāt. Studies in the History of the Early ՙAbbāsids, publication subsidised by the University of Baghdad No. 104 (Baghdad, 1976).

Peacock, Andrew C. S., Mediaeval Islamic Historiography and Political Legitimacy. Balՙamī’s Tārīkhnāma, Routledge Studies in the History of Iran and Turkey (London and New York, 2007).

Pritsak, Omeljan, ‘Die Karachaniden’, Isl. XXXI (1953–4), pp. 17–68, reprinted in O. Pritsak, Studies in Medieval Eurasian History, Variorum Reprints, Collected Studies Series CS 132 (London, 1981), no. XVI.

Rásonyi, Láslo and Imre Baski, Onomasticon turcicum. Turkic Personal Names as Collected by László Rásonyi, Indian University Uralic and Altaic Series, Vol. 172, 2 vols. (Bloomington, Ind., 2007).

Ray, H.C., The Dynastic History of Northern India (Early Mediaeval Period). 2 vols. (Calcutta, 1931–6).

Sadighi, Gholam-Hossein, Les mouvements religieux iraniens au II^e et au III^e siècle de l’Hégire (Paris, 1938).

Scarcia Amoretti, B., ch. ‘Sects and heresies’, in The Cambridge History of Iran. Vol. 4. The Period from the Arab Invasion to the Saljuqs, ed. Richard N. Frye (Cambridge, 1975), pp. 481–519.

Shaban, Muhammad A., The ՙAbbāsid Revolution (Cambridge, 1970).

Sharon, Moshe, Black Banners from the East. The Establishment of the ՙAbbāsid State – Incubation of a Revolt, The Max Schloessinger Memorial Series, Monographs II (Jerusalem, Leiden, 1983).

——, Black Banners from the East II. Revolt. The Social and Military Aspects of the ՙAbbāsid Revolution, The Max Schloessinger Memorial Series, Monographs V (Jerusalem, 1990).

Storey, Charles A., Persian Literature. A Bio-bibliographical Survey. Vol. I. Qur’ānic Literature; History and Biography (London, 1927–53).

Sümer, Faruk, Türk devletleri tarihinde şahis adları, ed. Turan Yazgan, 2 vols. (Istanbul, 1995).

al-Ṭabarī, Muḥammad b. Jarīr, Ta’rīkh al-Rusul wa ‘l-mulūk, ed. M.J. De Goeje et al., 15 vols. (Leiden, 1879–1901), English tr. The History of al-Ṭabarī, an Annotated Translation. Vol. XXIV. The Empire in Transition. The Caliphates of Sulaymān, ՙUmar, and Yazīd AD 715–724/AH 97–105, tr. David S. Powers (Albany, 1989); Vol. XXVII. The ՙAbbāsid Revolution. AD 743–750/AH 126–132, tr. John A. Williams (Albany, 1985); Vol. XXVIII. ՙAbbāsid Authority Affirmed. The Early Years of al-Manṣūr AD 753–763/AH 136–145, tr. Jane D. McAuliffe (Albany, 1995); Vol. XXX. The ՙAbbāsid Caliphate in Equilibrium. The Caliphates of Mūsā al-Hādī and Hārūn al-Rashīd AD 785–809/AH 169–193, tr. C. Edmund Bosworth (Albany, 1989).

Tor, Deborah G., Violent Order: Religious Warfare, Chivalry, and theՙAyyār Phenomenon in the Medieval Islamic World, Istanbuler Texte und Studien, Band 11 (Würzburg, 2007).

Treadwell, W. Luke, ‘Ibn Ẓāfir al-Azdī’s account of the murder of Aḥmad b. Ismāՙīl al-Sāmānī and the succession of his son Naṣr’, in Studies in Honour of Clifford Edmund Bosworth. Vol. II. The Sultan’s Turret: Studies in Persian and Turkish Culture, ed. Carole Hillenbrand (Leiden, 2000), pp. 397–419.

——, ‘The account of the Samanid dynasty in Ibn Ẓāfir al-Azdī’s Akhbār al-duwal al-munqat.iՙa’, Iran JBIPS, XLIII (2005), pp. 135–71.

Walker, John, A Catalogue of the Muhammadan Coins in the British Museum. Vol. I. A Catalogue of the Arab-Sassanian Coins (Umaiyad Governors in the East, Arab-Ephthalites, ՙAbbāsid Governors in Ṭabaristān and Bukhārā) (London, 1941).

Wellhausen, Julius, The Arab Kingdom and Its Fall, English tr. Margaret G. Weir (Calcutta, 1927).

al-Yaՙqūbī, Aḥmad b. Abī Yaՙqūb, Kitāb al-Buldān, ed. M.J. De Goeje, BGA VII (Leiden, 1892), French tr. Gaston Wiet as Les pays (Cairo, 1937).

Yāqūt b. ՙAbdallāh al-Ḥamawī al-Rūmī, Muՙjam al-buldān, 5 vols. (Beirut, 1374–6/1955–7).

Zarrīnkūb, ՙAbd al-Ḥusain, ch. ‘The Arab conquest of Iran and its aftermath’, in The Cambridge History of Iran. Vol. IV. The Period from the Arab Invasion to the Saljuqs, ed. Richard N. Frye (Cambridge, 1975), pp. 1–56.