The political breakdown of the caliphate in the 3rd/9th and 4th/10th centuries, although disastrous for the finances of the state and for agriculture in ʿErāq-e ʿArab and, perhaps, also in Ḵūzestān and parts of western Persia, did not have ill effects immediately on the economic life of Persia as a whole.




The political breakdown of the caliphate in the 3rd/9th and 4th/10th centuries, which has been briefly touched upon above, although disastrous for the finances of the state and for agriculture in ʿErāq-e ʿArab and, perhaps, also in Ḵūzestān and parts of western Persia, did not apparently have ill effects immediately on the economic life of Persia as a whole. ʿErāq-e ʿArab was not only the political center of the ʿAbbasid empire, but also its economic center. The Persian Gulf was, therefore, the main route for the Indian and Far Eastern trade as it had been in Sasanian times. For this reason, a series of flourishing towns and ports were to be found along the northern littoral of the Persian Gulf, some of which were later to disappear with the movement of trade in favor of the Red Sea to the detriment of the Persian Gulf. Among them was Sīnīz, which had a flourishing textile industry. It was plundered and virtually destroyed by the Carmathians in 321/933 or 322/934. Another was the important port of Sīrāf; it was partially destroyed by an earthquake in 366 or 367/977 (see EARTHQUAKES iv) and finally ruined by the movement of trade. Arrajān, situated inland on the borders of Fārs and Ḵūzestān, was also an emporium for Indian goods (Gaube,p. 45). Similarly, Narmāšīr, an important town with a numerous population, according to Moqaddasī (p. 463), lay on the pilgrim road from Sīstān to Mecca and was a mart for Indian goods.

The Muslim geographers of the 3rd/9th and 4th/10th centuries give, on the whole, the impression of a prosperous countryside based on agriculture and local crafts in Persia and the eastern provinces of the Muslim world. The towns were predominantly commercial centers but many of them also had a thriving manufacture. Ahvāz, according to Moqaddasī, possessed great warehouses, where merchandise was collected from inland towns and stored for transfer to Baṣra for final sale and export (Le Strange, Lands, pp. 233-34). Many of the major cities of inland Persia were also distribution centers for other parts of the country. Moqaddasī (p. 318) records that the markets of Toršīz (Ṭorṯīṯ) were renowned, whence merchandise was exported to and from Fārs and Isfahan. Nīšāpūr was described by the Arab geographers in the latter part of the 4th/10th century as a great trade center, the resort of merchants from ʿErāq-e ʿArab and Egypt, the depot (maṭrahá) for Ḵᵛārazm, Ray, and Gorgān, the entrepot (forża) for Fārs, Sind, and Kermān, and as having inhabitants who were the richest in Khorasan (Bosworth, The Ghaznavids, p. 150). Jīroft was another great city in the 4th/10th century. Goods from Sīstān and Khorasan were collected there. It also had a large export of dates. Moqaddasī states that nearly 100,000 camels set out every year for Khorasan with dates. Isfahan, in spite of disorders in Buyid times, became a flourishing city. The Ṭabara (Ṭabarak) quarter was added with a fortress built by Rokn-al-Dawla or Moʾayyed-al-Dawla; and in 429/1037-38 ʿAlāʾ-al-Dawla Moḥammad built a wall round the city, for which purpose he laid heavy impositions on the people. It contained splendid private and official residences, stables, baths, gardens, and fine well-stocked bāzārs (Māfarrūḵī, pp. 81, 83 ff, 113). Ebn Ḥawqal mentions its wealth and trade as well as its export of silks and textiles to other provinces. No other city between ʿErāq-e ʿArab and Khorasan had more trade except Ray (tr. Kramers, II, pp. 362-63).

Tolls (żarāʾeb) were in some cases levied at city gates. Moqaddasī, writing of the Jebāl, states that they were not heavy except in Isfahan and its districts. There 30 dirhams were taken for every load entering Yahūdīya (p. 400).

However, what is true of one city, town, or village at a particular moment cannot be taken as typical of other cities, towns, or villages at all times. Settlements rose and fell and a clear distinction cannot always be made between a city, town, or village. What might be designated as a city in one period might be merely a town or village in another, or even have disappeared. Eṣṭaḵr, once a great city, was laid in ruins in the reign of the Buyid Ṣamṣām-al-Dawla b. ʿAżod-al-Dawla, and became a mere village. Some towns decayed because they lost their population to a neighboring town. This was the case with Šāpūr (Bīšāpūr). According to Ebn Ḥawqal, it had been as large as Eṣṭaḵr and more populous, but Moqaddasī, writing in the latter part of the 4th/10th century, states that it had already gone to ruin for the most part, its population having migrated to the neighboring and rising city of Kāzerūn. Dīnavar, an ancient foundation, the revenue of which was allocated to the payment of the pensions of the inhabitants of Kūfa, became in the 4th/10th century the capital of the small kingdom of the Kurd Ḥasanūya (d. 369/979). Ebn Ḥawqal (p. 354) states that it was two-thirds the size of Hamadān. Sīrjān, which had been the chief city of Kermān under the Sasanians, continued to be the capital of the province until the middle of the 4th/10th century, when the Buyid governor Ebn Elyās moved his residence to Bardasīr (also known as Govāšīr and later as Kermān), relegating Sīrjān to being the second city of Kermān.

The essential difference between a town and village was that the town was the place where the villager paid his tax and delivered his crops. This distinction was not, however, hard and fast because all towns were not tax-collecting centers and some towns might themselves have to pay their taxes in the neighboring city. The majority of towns were market towns, closely integrated with their hinterlands, from which they drew their food and to which they sold, at least in part, their manufactures (see COMMERCE v; Le Strange, Lands). The bigger towns or cities, which produced a large range of goods, sold these to a wider market. Like the smaller towns, they needed a large commercial hinterland and to be the center or terminus of trade routes. They usually were administrative and tax-collecting centers as well as commercial centers. Many of them were also centers of learning and religion.

In the Ghaznavid period. Toward the end of the 4th/10th century conditions on the borders of the Samanid empire were changing. The Qarakhanids, who had been consolidating their power in Kāšḡar and Bālāsāḡūn, were pressing down into the Syr Darya basin. By 389/999 they had divided the Samanid empire with Sultan Maḥmūd of Ḡazna (r. 388-421/998-1030), with the river Oxus forming the frontier between them. In 396/1006, while Maḥmūd was absent on one of his Indian campaigns, the Qarakhanid Ileg Naṣr invaded Khorasan, but Maḥmud returned, repulsed his attack, and in 398/1008 defeated him near Balḵ. Maḥmūd, having consolidated his power in Khorasan, gradually brought under control various regions which had been on the periphery of the Samanid empire, Sīstān, Ḡarčestān, Jūzjān, Čaḡānīān, Ḵottal, and Ḵᵛārazm. The last named was important to him economically, because it was the Islamic terminus for caravans coming from the Oḡūz steppe and Siberia and, strategically, because it enabled him to turn the flank of the Qarakhanids in Transoxania (Bosworth, in Cambr. Hist. Iran IV, pp. 174-75).

Maḥmūd undertook numerous campaigns into India, from which he obtained immense booty and large numbers of slaves—according to ʿOtbī, 53,000 slaves were brought back from his Kanauj expedition in 409/1018 (Bosworth, in Cambr. Hist. Iran IV, p. 179). Warriors for the faith (ḡāzīs) and volunteers from all parts of the eastern Islamic world joined his army to descend on India. On his death his empire extended from the borders of Azerbaijan and Kurdistan to the Upper Ganges valley and from Ḵᵛārazm to the Indian Ocean. His Indian campaigns and the bullion he obtained from his temple raids reversed for a while the normal drain of specie to India and enabled him to maintain a good standard of gold and silver coinage, while the extra currency in circulation stimulated trade in the eastern Islamic world, but his reign was not accompanied or followed by economic growth. The countryside was drained of its wealth by the heavy imposts imposed to pay for his military expeditions, and the revenues of the territories under his control and the plunder he gained from his Indian raids went mainly to the equipment and support of his army. He erected a number of magnificent buildings, but their upkeep imposed a heavy burden on the population. The situation was further aggravated toward the middle of his reign in 401/1011 by famine in Khorasan. It was particularly severe in Nīšāpūr, where over 107,000 persons are said to have died and where instances of cannibalism were alleged (Bayhaqī, p. 176; Barthold, Turkestan, pp. 287-88).

In 407/1016-17 Maḥmūd attempted, without lasting success, to install his nominee in Kermān, where, since the death of ʿAżod-al-Dawla, Buyid princes had been fighting among themselves to the detriment of the prosperity of the province. Towards the end of his reign, Maḥmūd turned his attention towards Buyid domains in the Jebāl, where, as in Kermān, internecine strife and the exactions of the Daylamite soldiery were damaging the already fragile economy. In 419/1028 Maḥmūd marched on Ray, deposed the Buyid Majd-al-Dawla, and sacked the city (420/1029). He then sent his son Masʿūd to undertake operations to the west. On Maḥmūd’s death in 421/1030 Masʿūd hastened back to Ḡazna.

Masʿūd (r. 421-33/1030-40), having overthrown his brother Moḥammad, who had succeeded Maḥmūd, carried on his father’s policy of military conquest but with notably less success. His fiscal policies were equally oppressive. Khorasan in particular suffered from much disorder in the late Ghaznavid period. Control over Kᵛārazm, which guarded the approaches from the steppes to northeast Persia, was lost on the death of the Ghaznavid governor, the Ḵᵛārazmšāh Altuntaš.

Toward the end of the 4th/10th century the Oḡoz tribes (or Ḡozz as they are known in the Arabic and Persian sources) had begun to move westwards. A group crossed the Oxus into Khorasan in 416/1025. They created disturbances and were subsequently dispersed. Some went to Isfahan, some to Ray, and others to Azerbaijan. Another group under the sons of Saljuq, Mūsā, Mīkāʾīl, Arslān Esrāʾīl, and Mīkāʾīl’s two sons, Ṭoḡrel Beg Moḥammad and Čaḡrī Beg Dāʾūd, had meanwhile occupied winter pastures in Nūr Boḵārā and summer pastures in Sogdia. In 426/1034-35 some 700 or 900 of them crossed the Oxus and asked permission to live under the protection of Masʿūd. This was refused. They nevertheless moved into Khorasan. Their numbers meanwhile increased, and during the next few years they were constantly on the move in search of new pastures, harried by and harrying the Ghaznavids. In 429/1038 Marv, Herat, and Nīšāpūr submitted to them, although the last named was subsequently recovered by Masʿūd. Finally in 431/1040 a Saljuq force decisively defeated Masʿūd at Dandānqān and brought Ghaznavid rule to an end in Khorasan (Bayhaqī, ed. Fayyāż, pp. 832-40). After consolidating their conquests in that province the Saljuqs moved westward, and as they did so, bands of Ḡozz who had preceded them into Persia became associated with them, although outlying groups, while acknowledging the nominal overlordship of the Saljuqs, continued to act independently.

In the Saljuq period. Although the independent bands of Ḡozz, especially those who had preceded the Saljuqs into Persia, raided the countryside, the Saljuq movement into Persia appears, on the whole, to have caused remarkably little dislocation. This was perhaps partly because of the smallness of their numbers, which were probably to be counted in tens of thousands. Nevertheless, as a result of their advent into Persia, the distinction between the settled and the semi-settled elements of the population became sharper, though the immediate effect of this was localized. Many of the regions which formed the Saljuq empire were too hot and dry for the Turkmens and their flocks. Apart from Anatolia, the major settlements were in Azerbaijan, parts of Dīār Bakr, northern Kurdistan, Gorgān, Dehestān, and Marv, where many of the Ḡozz had remained. There were also isolated settlements in Fārs and Ḵūzestān. Presumably, the Ḡozz were mainly wide-ranging nomads owning two-humped camels, which were able to withstand cold winters and travel long distances, although they may well have combined this with nomadism based on sheep and other livestock, which required only local movements for pasturage (on the spread of the two-humped camel, see Bulliet, 1975, pp. 141 ff.; Cahen, “Nomades,” pp. 93-104).

Prior to the Saljuq period Transoxania and Ḵᵛārazm had bred and exported large numbers of sheep and cattle, but within the Persian provinces the balance was with arable farming; and it is not unlikely that the introduction of a limited number of flocks into Persia and the provision of meat, milk products, wool, and skins to the towns contributed to the prosperity of the country. The primary need of the Ḡozz was for pasture for their flocks, and the extensive practice of agriculture, apart from districts round the villages and towns, made it possible for them to be accommodated. However, there were indigenous herdsmen in the Kurdish districts of western Azerbaijan and the Jazīra, and, significantly, it was there that the early bands of Ḡozz were opposed by the local population.

With their victory over Masʿūd at Dandānqān, the Saljuqs became in effect the rulers of a territorial empire, the wardens of the northeastern marches of the Islamic world, heirs to the civilization which had developed in the lands of the Eastern Caliphate and the defenders of Sunni Islam. From the beginning of their transformation into rulers of a territorial empire, they had settled capitals. But once they had become the rulers of a territorial empire, they needed a more stable base for their power than that provided by the Ḡozz and the Turkmen tribes, whose numbers were, perhaps in any case, too small to support them in the long term. Slaves and freedmen became an increasingly important component in their armies (see Lambton, Continuity, pp. 6 ff.).

Although the Saljuqs were, by origin, foreign to the Perso-Arab population of the ʿAbbasid empire, their Muslim upbringing prepared them for a rapid acceptance of the cultural institutions of the Muslim world and the imperial tradition in its broad outlines. As a result of over-taxation by previous rulers and repeated military operations, the economic situation was precarious when the Saljuqs came to power. Having no bureaucratic machinery to rely upon for the collection of taxes to pay their officials and military forces other than that already in place, they continued the Buyid practice of paying them by assignments of revenue. However, the usages of the steppe and the nature of the Saljuq armed forces transformed both the imperial structure and the eqṭāʿ system. In the first place, the existing concept of rule was modified by the tribal idea that rule belonged to the family and was exercised by the eldest or the most competent member of the family. So the sultan granted to members of his family whole regions as eqṭāʿs, much as the ruling khan of a nomadic tribe might allocate to his family and his followers pastures and camping grounds. Such eqṭāʿs were not hereditary, but very soon the tendency arose for different branches of the family to regard certain regions as their own eqṭāʿ. Secondly, and more importantly, the eqṭāʿ became associated with governmental functions and immunities from taxes and dues. Thus, it became the dominant administrative, social, and economic institution of the empire.

The military eqṭāʿ did not disappear but was no longer the characteristic type of eqṭāʿ as it had been under the Buyids. The standing army was paid partly in cash and partly by eqṭāʿ, but the holders had no rights of jurisdiction in the districts assigned to them. In Khorasan Sanjar (511-52/1118-57) apparently exercised close control over such eqṭāʿs. The administrative eqṭāʿ, in contradistinction to the military eqṭāʿ, was a delegation of authority in the matter of the collection of taxes or of the administration, generally. It was normally tied to a district or to a whole region and might include villages and towns. Such eqṭāʿs were ad hoc grants and subject to renewal at irregular intervals. The government was thus enabled to dispense with the need to provide for and pay a provincial bureaucracy and to transfer funds to and from the provinces—no small convenience in view of the long distances which were often involved and the possibility of insecurity on the roads. The system was not without some benefit to the provinces also: political control was to some extent localized and local resources were for the most part spent locally, even if they were not strictly speaking under local control, since the moqṭaʿ was not usually a local man (though he might strike roots in his eqṭāʿ). Responsibility for security was placed on the moqṭaʿ and, on the whole, satisfactorily maintained. Except where the moqṭāʿ assumed that his tenure would be short and thus extorted as much as he could from his eqṭāʿ, self-interest demanded that he should exert a modicum of good government and good husbandry, and this continued to be the case even when the control of the central government declined. Although a strong moqṭaʿ might himself commit extortion against the peasants, he would prevent outsiders from acting with violence toward them. Theoretically, the possibility of a demand for redress from the sultan against the malpractices of a moqṭaʿ remained open, but it was largely illusory, partly because of the long distances often involved. The major disadvantage of the system at the local level was that it fostered the growth of a subject peasantry and the existence of private armies. The peasants, however, were not serfs.

Large and small estates continued to exist side by side, both in districts assigned as eqṭāʿs and outside them. In some districts, as would seem to be the case from Ebn Fondoq’s account of Bayhaq, there appears to have been a prosperous class of medium-sized and small landowners (Lambton, Continuity, pp. 130 ff.). Around the towns and cities land was probably owned in small parcels though not necessarily or primarily by peasants (as witness the documents recording the sale of landed properties, attestations of ownership, and requests for reductions of taxation in the neighborhood of Isfahan contained in Afšār, 1355 Š./1976, pp. 272, 277, 281-82, 285-91).

Crown lands do not appear to have been extensive, although there are references to property belonging to the dīvān-e ḵāṣṣ of Sanjar in Basṭām, Marv, and Ray (Jovaynī, pp. 56, 57, 72), and Rāvandī (p. 171) alleges that Sanjar made estates in ʿErāq-e ʿArab and other important regions into crown property. Ebn al-Balḵī states that the Rāhebān kārīz near Kāzerūn belonged to the royal dīvān (pp. 145-46).

The first call on the funds of rulers was the standing army. Rulers were also expected to carry out public works and to keep major irrigation works, roads, bridges, and caravanserais in repair. In the provinces these duties devolved on the governors and moqṭaʿs. There are various references to the building and repair of irrigation works by the Saljuqs and their governors, a fact which implies a recognition of the importance of agriculture for the economy of the country. Ḥamd-Allāh Mostawfī mentions a number of qanāts made by the Saljuqs in Qazvīn (Tārīḵ-e gozīda, p. 781). They also increased dams and canals on the Morḡāb river (Le Strange, Lands, p. 402). Čavlī Saqao, the Saljuq governor or moqṭaʿ of Fārs (d. 510/1116-17) repaired several dams in Fārs (Ebn al-Balḵī, pp. 128, 130, 151-52). Of the caravanserais extant from the Great Saljuq period, probably the best preserved is the Rebāṭ-e Šaraf on the Marv-Nīšāpūr road, restored in 549/1154 by Sanjar’s wife, but built perhaps in the 1120s by one of his viziers (Rogers, The Spread of Islam, p. 69).

The establishment of a reasonably efficient and enlightened government was vital for the well-being of agriculture, on which the economy of the state largely depended. Security was the basic requirement both for agriculture and trade and could only be maintained by a strong governmental authority, although not necessarily one which sought to exercise control over the daily affairs of its subjects. Broadly speaking, under the first three Saljuq sultans security was achieved without a high degree of centralization except in the tax administration, although the effect of this was modified by the delegation of local administration to the moqṭaʿs. It is, perhaps, partly because much of the economic life of the country was conducted outside governmental spheres that information of the details of its operation are lacking. There are few, if any, collections of private papers from the period.

The Saljuq empire reached its height under Malekšāh (r. 465-85/1072-92) and included most of western Asia, extending from Toḵārestān in the northeast to the frontiers of the Byzantine empire in Asia Minor and the Fatimid caliphate of Egypt in the west. The sweeping away of local dynasties, or their incorporation into the Saljuq empire, and the removal of local frontiers contributed to freedom of movement and trade over a wide area. In 479/1086 Malekšāh ordered the abolition of mokūs levied on merchants for all kinds of merchandise in ʿErāq-e ʿArab (Ebn al-Aṯīr, X, p. 105; see further CITIES iii, p. 615). However, the fact that there are later references to the abolition of mokūs indicates that the attempt to abolish these taxes was abortive.

The campaigns of Ṭoḡrel Beg, Alp Arslān, and Malekšāh were on a considerable scale but were mainly directed externally. Nevertheless, the gathering of armies for war if only because of commissariat problems involved some degree of local dislocation. After the death of Malekšāh, the civil war between Barkīāroq and Moḥammad caused disturbances throughout the country, as also did the struggles between the various Saljuq princes after the death of Moḥammad in 511/1118. Further, moqṭaʿs not infrequently had to take possession of their eqṭāʿs by force, but, on the whole, these operations were on a small scale. Outbreaks of disorder and factional strife in the large cities also became more frequent and are indicative of social, and possibly economic, discontent (see ʿAYYĀR), as also is the spread of the Nezārī (see BĀṬENĪYA) movement of the Ismaʿilis.

The speed with which towns and cities, and presumably their hinterlands, recovered from sieges, disorders, and natural calamities is notable and presumes a widespread use of credit, which enabled society to maintain itself and survive disasters. Nāṣer-e Ḵosrow, who left Marv in 437/1045 and traveled through Persia on his way to Egypt, reports that there was a famine in Qūha, a few days’ journey from Qazvīn and that barley bread had risen to 2 dirhams a man (ca. 3 kg). But its incidence appears to have been local. He makes no mention of shortages in Qazvīn, where he arrived in Moḥarram 438/July 1046. He states that Tabrīz, which he reached shortly thereafter, was a flourishing town (šahr-e ābādān), although an earthquake four years earlier (434/1042), in which 40,000 persons were said to have perished, had destroyed part of the town (Nāṣer-e Ḵosrow, pp. 4, 6; see also EARTHQUAKES iii and iv).

Nāṣer-e Ḵosrow reached Isfahan on his return journey in 444//1052. He remarks on the high level of security enjoyed by the people of Ḵān Lenjān near Isfahan and on the flourishing condition of Isfahan. He states that the city had many fine bāzārs. There were 200 ṣarrāfs in the bāzār of the moneychangers and in one street known as the Kū-ye Ṭerāz 50 good caravanserais, in each of which many merchants (bayyāʿān) and traders (hojradārān) were established. The caravan in which he had traveled carried 1,300 ḵarvārs of merchandise. Two years before his arrival the city had been invested for the second time by Ṭoḡrel Beg and had capitulated because of famine. Ṭoḡrel Beg had appointed a new governor and ordered him to remit the taxes for three years. As a result, those who had left the city had returned. Nāṣer-e Ḵosrow states that he had seen no finer or more flourishing city in all the Persian-speaking lands than Isfahan. This is all the more surprising in that shortly before his arrival there had been a severe famine, perhaps partly occasioned by Ṭoḡrel Beg’s siege. The effects of this, whatever its cause, were apparently still being felt, for Nāṣer-e Ḵosrow states that 1 1/2 mans of wheat bread was selling at 1 dirham ʿadl and 3 mans of barley bread at the same price, although barley was at the time being harvested. He was told that formerly 1 dirham bought at least 8 mans of bread (Nāṣer-e Ḵosrow, pp. 92-93).

Another place where, according to Nāṣer-e Ḵosrow, great security prevailed was Ṭabas, which was under a certain Abu’l-Ḥasan Gīlākī b. Moḥammad, who had taken it by the sword. The nearby city of Tūn, on the other hand, was mainly in ruins, although it still had 400 establishments which wove zīlūs. Qāʾen, through which he also passed, appears to have been a large and well-to-do prosperous city; the maqṣūra of the mosque was the largest he had seen in Khorasan (pp. 94-96). He reached Balḵ later in the year (444/1052). He remarks on the insecurity of the roads between Šabūrqān and Balḵ (p. 96). Shortly after Nāṣer-e Ḵosrow’s journey famine and pestilence (wabāʾ) raged in 448-49/1056-57 over a wide area from Transoxania and Khorasan to Azerbaijan, the Jebāl, southern Iraq, the Ḥejāz, and Egypt (Lambton, Continuity, pp. 165-66). Its effect on the Persian provinces appears to have been transient.

Although Baghdad retained its importance, it was no longer the political and commercial center it had once been. The establishment of the capital in Isfahan by Ṭoḡrel Beg led to a growth of population in that city and an increased demand for foodstuffs and other goods and services, all of which contributed to the prosperity of the region. The anonymous author of the Mojmal al-tawārīḵ wa’l-qeṣaṣ (ed. Bahār, p. 525), who wrote in the 6th/12th century, states that there was no greater city in ʿErāq-e ʿArab or Khorasan than Isfahan. Textiles, together with luxury articles, were carried from it to all parts of the world. Although Nīšāpūr was somewhat overshadowed by Isfahan once the latter had become the capital, it nevertheless remained one of the great cities of the empire until the end of the reign of Sanjar. It was partially destroyed by earthquake in 540/1145, plundered by the Ḡozz in 549/1154-55 (or according to Yāqūt, in 548/1153-54), and damaged by another earthquake in 605/1209. Marv, Sanjar’s capital, was a flourishing city in his day. It was plundered by the Ḡozz in 548/1153 and fell to Ay Aba after a siege in 557/1162. It appears to have recovered its prosperity under the Ḵᵛārazmšāhs. Yāqūt, who left it in 616/1219 after spending some three years there, remarks on its prosperous condition (Meynard, p. 529). Bardasīr, the capital of the Saljuqs of Kermān, flourished until decline set in under Ṭoḡrel Shah (551-65/1156-70), as also did Jīroft, their winter capital. Moḥammad b. Ebrāhīm, writing in the 6th/12th century, refers to Qamādīn, just outside Jīroft, where travelers coming by land or sea could store their goods and states that precious goods from China, India, Abyssinia, Zanzibar, Asia Minor, Egypt, Armenia, Azerbaijan, Transoxania, Khorasan, Fārs, and ʿErāq-e ʿArab were to be found there (Houtsma, Recueil, pp. 49, 83). However, from about 574/1178-79 the province of Kermān was overrun by the Ḡozz and was subject also to incursions by the rulers of Fārs and Khorasan. Agriculture was ruined and trade brought to a standstill. In the following year 575-76/1179-80 (the ḵarājī year 569) there was famine in the province. Afżal-al-Dīn describes conditions in Govāšīr (Bardasīr) in lurid terms and alleges cannibalism. He states, “Because of the piling up of the bodies of the dead in the districts [of the city], the living could not get by. No respect was shown by anyone to the dead, and they were not laid out or covered with shrouds” (Tārīḵ-e Afżal, p. 91; idem, ʿEqd al-ʿolā, p. 97, he puts the famine in the ḵarājī year 570).

Yazd, on the other hand, enjoyed on the whole a period of prosperity in the 5th/11th and 6th/12th centuries. When Ṭoḡrel Beg took Isfahan from the Kakuyid Abū Manṣūr Farāmarz in 443/1051, he gave him Abarqūh and Yazd as an eqṭāʿ (Ebn al-Aṯīr, IX, pp. 384-85), where he resided more or less undisturbed. He built a palace, a Friday mosque, and a wall round the city. His successors continued for some years as local rulers in Yazd, and under them new villages were built and qanāts dug in the vicinity of the city. They were succeeded in the late 6th/12th century by the Atabegs of Yazd, under whom development continued (Jaʿfarī, pp. 19-20; Bosworth, “Dailamīs in Central Iran,” pp. 89 ff.).

Hamadān, which became the political center of the Jebāl under the Saljuqs of Iraq in the second half of the 6th/12th century, similarly enjoyed a brief period of cultural and economic growth. Under Arslān-šāh (556-73/1161-77) it was an important city, but its prosperity did not last. The Jebāl suffered much damage in the struggles between the Ḵᵛārazmšāhs and the caliph.

That many of the sultans, their ministers and amirs, the rulers of the Saljuq succession states, and their women disposed of considerable wealth is evident from the numerous buildings, mosques, madrasas, hospices, rebāṭs and other religious foundations erected by them, the awqāf which they constituted for their upkeep, and their generous patronage of the ʿolamāʾ. Some but not all of these awqāf were constituted out of private property (Lambton, Continuity, pp. 149-51). While these various activities on the part of rulers and others were in part designed to strengthen and control the religious institution and, above all, to win the support of the religious classes because of their influence over the common people, the motive of piety cannot be excluded. We do not know how the masons and skilled craftsmen were collected or organized for these undertakings. The wholesale movement of skilled craftsmen by rulers was a later phenomenon. Nor do we know whether there was a surplus of labor. It may be that labor which would otherwise have been unemployed worked on these projects. In any case, such building programs are likely to have contributed in some measure to local prosperity and, so far as awqāf escaped confiscation, to a measure of capital accumulation.

The urban communities in the major cities contained many rich men, some of whom appear to have had large sums available in cash (cf. the story of Abū Hāšem, the raʾīs of Hamadān, who died in 502/1108-09, in Bondārī, pp. 89-90; Nīšāpūrī, pp. 42-43; see also Fragner, pp. 123 ff.). The leaders of the religious rites were sometimes very wealthy; and some of the ʿolamāʾ as administrators of awqāf accumulated, or disposed of, large sums. Many of them devoted their wealth to charitable purposes, but they also put their money out to work with merchants as also did government officials.

The jahābeḏa as bankers appear to have virtually disappeared, partly perhaps because the state was based primarily on a land economy and speculation arising from the fluctuation of exchange rates was more limited than had been the case formerly. Further, with the spread of the eqṭāʿ system, tax-farming had become very much less common and the services of the jahābeḏa for the transmission of funds were in less demand (Cahen notes the decline of the role of the jahābeḏa from Buyid times onwards; see “Quelques problèmes,” p. 355). The main moneylenders to private persons and to the government appear to have been merchants as was the case in medieval Europe (Pirenne, p. 127). Interest rates were high to compensate for unpaid debts. Demands for loans from those who were politically and socially influential could not easily be resisted or repayment be insisted upon.

ʿAlāʾ-al-Dīn Atsïz b. Qoṭb-al-Dīn Moḥammad b. Anūštigin (521-51/1127-56), who governed Ḵᵛārazm on behalf of the Saljuqs, made various attempts to throw off his allegiance to Sanjar, and when the latter was defeated by the Qara Ḵitays at Qaṭvān in 536/1141, he temporarily occupied parts of Khorasan. On Sanjar’s death in 552/1157, Il-Arslān, who had succeeded his father Atsïz as Ḵᵛārazmšāh in 551/1156, emerged as the most powerful ruler in the eastern provinces. The agricultural and economic prosperity of Ḵᵛārazm enabled him and his successors to maintain a large army recruited mainly from Qïpčaqs and Qanglï Turks. The ebb and flow of the struggles among the Ḵᵛārazmšāhs, Qara Ḵitays, Qarluqs, Ghurids, and Qïpčaqs in Trans-oxania, Ḵᵛārazm and Khorasan, the operations between the Ḵᵛārazmšāhs and the Ḡozz, who had taken possession of Khorasan on the death of Sanjar, the ensuing disorders, especially in Khorasan and the increasing infiltration of Qara Ḵitays, Qïpčaqs and Qanglï Turks into the eastern provinces of the dār-al-Eslām interrupted the continuity of economic life (Barthold, Turkestan, pp. 323 ff.; Morgan, 1988, pp. 46-50). The brutality and perfidy of the Qara Ḵitay are attested by most of the sources. With the weakening of pressure from the north-east, Il-Arslān’s successors intervened increasingly in the struggles among the Saljuq maleks, their amirs, and atabegs in the western provinces. The first occasion appears to have been in 562/1166-67 when Ïnanč (Īnānj) Sonqor, governor of Ray, who had been defeated by Arslān b. Ṭoḡrel and his atabeg Eldigüz, had recourse to the Ḵᵛārazmšāh for help. The latter’s army committed much disorder in Abhar and Qazvīn before returning to Ḵᵛārazm (Rāvandī, p. 294). The death of Il-Arslān in 567/1172 was followed by a civil war, from which Tekeš finally emerged as Ḵᵛārazmšāh. When the caliph al-Nāṣer (575-622/1180-1225) appealed to him for help against Ṭoḡrel b. Arslānšāh, the last of the Saljuq sultans of Iraq, he advanced through Māzandarān, conquered Ray and Hamadān, and defeated and killed Ṭoḡrel in 490/1194. His demand that the ḵoṭba be read in his name in Baghdad led to a rupture with the caliph. Fighting took place between them in 592/1196 to the disadvantage of the caliph (see EBN al-QAṢṢĀB), and skirmishes continued between them until Tekeš’s death in 496/1200. Further struggles between the caliph and Tekeš’s son and successor ʿAlāʾ-al-Dīn Moḥammad continued, during which the army of the Ḵᵛārazmšāh is alleged to have committed much devastation. Thus, when the first Mongol invasion took place under Jengiz Khan (see ČENGĪZ) in 1219 C.E., the Persian provinces were affected by political and economic weakness.

Between the Mongol invasion and the fall of the Il-khanate. The Mongol invasions were distinguished from the earlier Arab and Saljuq invasions by the numbers and military organization of the Mongol forces: they were a horde prepared and organized for war, a horde whose aim was political domination and whose sedentary subjects “were deemed to exist solely in order to be exploited” (Morgan, 1988, p. 53). During the first invasion, which was resumed under Ögedei after the death of Jengiz Khan, the empire of the Ḵᵛārazmšāh collapsed, and northern and eastern Persia were subjugated. The great cities of Transoxania and Khorasan, Marv, Herat, Balḵ, and Nīšāpūr were devastated and their population put to the sword. The irrigation system of the Moṟḡāb river was destroyed and as a result Marv, according to Ḥāfez-eá Abrū, was turned into a desert swamp (Le Strange, Lands, p. 402). Nīšāpūr, sacked in 618/1221-22, never again achieved its former prosperity. ʿErāq-e ʿAjam, Māzandarān, and Azerbaijan were laid waste by the Mongol generals Jebe and Sübedei as they pursued the Ḵᵛārazmšāh. Western Persia was conquered; Georgia, Armenia, and northern Mesopotamia taken; and in 1243 the Saljuq sultan of Rūm was defeated. No sustained effort, however, was made to incorporate Persia fully into the Mongol empire (Morgan, 1988, p. 58). Fārs under the Salghurids, who had ruled since 543/1148, and Kermān under the Qutlugh-khanid (Qara Ḵitay) dynasty, who had established themselves after the fall of the Ḵᵛārazmšāhs, did not lie in the main line of the Mongol advance and escaped devastation by acceptance of the obligation to pay tribute. This did not, however, exempt them from interference by Mongol officials and their forces and ultimate absorption into the Il-khanate.

The second invasion took place under Hülegü (Holāgū) in 654/1256. His commission was to extirpate the Ismaʿilis of Alamūt and to reduce the caliph of Baghdad. By the time he died in 663/1265, Hülegü had made himself master of most of the provinces of Persia, Iraq, and most of Asia Minor, although he failed to take Syria or defeat decisively the Golden Horde, who were pressing down from the Qïpčaq steppe to attack Persia via Darband and Šīrvān. The Il-khanid dynasty which he founded lasted some eighty years and became the warden of the marches in the northeast and northwest over against the Chaghatay Khanate and the Golden Horde, respectively. Khorasan once more became a frontier province. Together with Transoxania it was devastated from time to time by the rivalries of the Il-khans and the Chaghatay khans—a devastation which contributed further to the demographic and agricultural decline, and the frontier districts, both in the northeast and northwest, tended to be depopulated because of the repeated passage of armies (Rašīd-al-Dīn, Tārīḵ-e ḡāzānī, p. 350; Tārīḵ-e Waṣṣāf, pp. 78, 80; Sayf Heravī, pp. 408, 416). Devastation was, moreover, not confined to the frontier regions. The civil war, which broke out in 680/1282 on the accession of Aḥmad Tegüder (Takūdār, q.v.), Hülegü’s son, who was overthrown and killed by Arḡūn in 683/1284, caused much destruction as did later disputes over the succession. There was also much pillage by independent marauders, notably the Negūdarīs, who invaded Kermān and Sīstān in or about 677/1278-79 and thereafter made many raids both into Kermān and Fārs.

The Mongol invasions affected the economy of Persia in a number of ways, though not uniformly in all provinces. The immediate result of the invasions was depopulation, although this was not felt equally everywhere. There was also a change in the balance between the Turkish and non-Turkish elements in the population and between the settled and the nomadic sectors. The consequence of this was an agricultural recession, a diminution in the amount of land under cultivation, an increase in dead lands, and the disruption of the land tenure system. Azerbaijan became the center of the Il-khanate, and it was there, and to a lesser extent in Arrān and Central Asia, that the tribes who came with the Mongols were mainly settled.

Ebn al-Aṯīr, Nasavī, and Jūzjānī, who were contemporary with the invasion of Jengiz Khan, all give a similar story of devastation and their estimates of the number killed are enormous. Jovaynī, who was an eye-witness of Hülegü’s invasion, also reports very heavy loss of life. The figures given can hardly be accepted at face value, but there is little doubt that the casualties were on an unparalleled scale, especially in the northeast. Some of the most fertile regions—Baghdad, Nīšāpūr, and Herat among others—were laid waste and their cities devastated. The peasants, so far as they were not massacred or carried off into slavery, fled the land (Petrushevsky, pp. 485 ff, 496-97; Aubin, 1967, pp. 190-95; idem, 1976-77, pp. 130-01, 176-77).

There are no figures to show what the acreage under tillage and pasture was either before or after the Mongol invasions, but the general inference from the sources is that there was a relative increase in pasture land after the invasions. There may also have been a reversion to grain farming in many parts of the country. Bread was the essential food of everyone and the staple diet of the majority. When poor, peasants grow grain rather than other crops because it is what keeps them alive. Moreover, with the destruction of markets and the prevailing insecurity on the roads, there was little incentive to grow cash crops. The demand for grain by the Mongol armies is also likely to have reinforced a trend toward grain farming. Ashtor, on the other hand, states that the demographic decline in the Middle East during the latter Middle Ages resulted in a decrease in the demand for cereals and that the peasants replaced the cultivation of wheat and barley by other crops especially cotton (1981, p. 262). However, the evidence does not support this in the case of Persia, although by the end of the Il-khanate cotton was cultivated in many provinces.

There does, however, appear to have been a grain shortage in the reign of Ḡāzān (see below). Moreover, ʿOmarī, whose account is based partly on information from travelers who had been in the Il-khanate during the reign of Abū Saʿīd (716-36/1316-35) or his successors, remarks (p. 92) on the shortage of grain in ʿErāq-e ʿArab, notwithstanding the favorable nature of the soil of the Sawād for agriculture, and states that he had been told that the reasons for this were the destruction which occurred in ʿErāq-e ʿArab during the campaigns of Hülegü and the outflow of grain to neighboring regions. Discussing the crops and fruits of Tabrīz, he also remarks that wheat, barley, lentils, peas, and beans were found in moderate quantity around Tabrīz, but that, if agriculture were to be more intensive, the crops could be increased and the taxes raised. He continues, however, “Its kings (molūk) pay no attention to this” (p. 89).

In the early period of Mongol domination in Persia the conquered lands were divided into dalay (dalāy), a term originally applied to the subjects of the great khan and then to the land, which belonged immediately to him, and injü (īnjū), a term applied to the persons whom the great khan had given to his relatives and others, as well as to the land granted to his relatives as appanages. In Persia injü covered both the subjects of the Il-khan and his land and came to be loosely used as a general term for crown land. It is not certain from the evidence whether landed estates owned by private individuals continued to exist as independent categories of land or became subordinate titles in dalay and injü land, and if so, what the effect on their holders was. It may be that only land actually conquered by the sword became dalay and injü. From Waṣṣāf’s account it is clear that Fārs was not wholly injü or dalay, whatever may have been the situation in other provinces. The status of waqf land so far as it escaped devastation appears to have been little changed. Jovaynī states that awqāf and charitable foundations were exempted from dues and contributions (ed. Qazvīnī, I, p. 11). Waṣṣāf (Tārīk-e Waṣṣāfò, p. 284) records that Bāydū reaffirmed a yāsā which exempted awqāf from taxation.

Like other conquerors before them, the Mongols considered themselves entitled to the revenue of the conquered lands. They also brought with them certain new principles and practices. Basic to the regulation and exploitation of the subjects was the registration of the population, and there are references to partial censuses (šomāra) being carried out in Persia during the invasions. Their tax system was “occasional,” whereas the system prevailing in Muslim Persia was “regular” and rested on the land and its exploitation. It was, therefore, possible for the two systems to exist side by side, and it seems that a double system did for some time prevail, although there was no uniform tax administration during the early years of the Il-khanate. The relationship of the two systems to each other is obscure. Jovaynī mentions a census in Khorasan and Māzandarān about the year 637/1239-40 made by Korgüz (Korkūz), who, he states, counted the people and assessed the taxes (ed. Qazvīnī, II, p. 229; Rašīd-al-Dīn, Jāmeʿ al-tawārīkò, Baku, I, p. 169). The Tārīḵ-e Sīstān (ed. Bahār, p. 397) records that a census was conducted in Sīstān for the first time in 639/1241-42 for the purpose of imposing qobčūr and qalān. Later censuses do not appear to have been made in Persia, at least not on a country-wide basis (Lambton, Continuity, pp. 200 ff.).

Three new tax terms, qobčūr, qalān,and tamḡā, are used in the Persian and Arabic sources with reference to taxes introduced by the Mongols. All three mean different things at different times and in different places. The origins of qobčūr are complicated. In the Il-khanate it seems at first to have been a levy, or additional cess, imposed on the conquered population as a poll tax. It appears also to have been levied on flocks on the basis of the number of animals owned (qobčūr-e marāʿī), and on crafts (qobčūr-e motaḥarrefa). Qalān probably designated under the Il-khanate occasional levies, although it may also have covered some sort of labor service. The consensus of the sources is that the levy of both was extremely oppressive; qobčūr was often demanded several times in the same year (Lambton, Continuity, pp. 199 ff.; idem, “Mongol Fiscal Administration”). Tamḡā was, or came to be, a tax or toll on commercial goods, levied in the towns at varying rates. It furnished a substantial part of the revenue of the Il-khanate. Mostawfī distinguishes in his account of the revenues of the different provinces and districts between the revenues of certain towns and their surrounding districts (welāyāt), the former being provided by tamˊḡā taxes and the latter by dīvān dues (ḥoqūq-e dīvān). Among the towns and districts described in this way, all of which were situated on the main trade routes, were Baghdad, Kūfa, Wāseṭ, Ḥella, Isfahan, Solṭānīya, Qazvīn, Qom, Kāšān, Hamadān, Yazd, Tabrīz, Ojān, Ahar, Šūštar, Āva, Sāva, Zanjān, Marāḡa, and Shiraz. The tamˊḡā taxes of the last five were farmed. Writing of Fārs, Mostawfī states that the dues amounted to 2,871,200 dīnār-e rāʾej (currency dinars). In the provinces (welāyāt) they were levied for the most part in kind (maḥṣūl), but in the towns they were fixed as tamḡā (Nozhat al-qolūb, ed. Le Strange, p. 113). Occasional information of the incidence of tamˊḡā taxes is to be found in Nabipour’s edition of Falak ʿAlāʾ Tabrīzī’s Saʿādat-nāma and Qānūn al-saʿādat (see index, s.v. tamğa), but their relation to other taxes and the proportion they formed of the total revenue is obscure.

In addition to the above taxes, one of the most ubiquitous levies made for the ordūs and Mongol officials was purveyance (soyūrsāt), which bore very heavily on the local population. There are also references to extraordinary levies known as nemer (nemerï, nemārī). Another feature of the tax administration was the proliferation of dues and additional cesses on the regular taxes.

After Hülegü sacked Baghdad in 656/1258, he began to make permanent arrangements for his conquests and to allot specific districts as yūrts to his followers and to appoint for them ʿolūfa (provisions and fodder) in the conquered lands. Once the Il-khans became the rulers of a settled empire with more or less defined frontiers as was beginning to be the case under Hülegü, money as well as milk and meat was required, and garrisons, for which provisions and pay had to be provided, were needed to hold the conquests. Gradually, a distinction between the Mongol horde and the Mongol army began to emerge, although exactly when this took place is not clear. The ordūs of the Mongol princes and princesses, however, remained a disruptive element. They were often large and acted with impunity bringing ruin to the countryside. Under Hülegü and Abaqa (663/1295-680/1281) their provisions were provided according to Mongol custom: they had flocks and received a share of the booty taken from enemies and possibly also slaves, and they participated in ortaqs (trading partnerships). Towards the end of Abaqa’s reign the ordūs received a small allowance of provisions (āš) and under Arḡūn (683-80/1284-91) each ordū was given a special sum allocated on the provinces. Arḡūn is alleged to have exercised careful control over these allowances and also over the allocations of provisions (taḡār) to the army. Waṣṣāf mentions that at the beginning of Arḡūn’s reign there were three years of drought and famine (683-85/1284-87) in Fārs, during which he alleges 100,000 people died (p. 209; Aḥmad Zarkūb, p. 95). This must have contributed to the difficulties of the state.

Concluded in Part 3.

(Ann K. S. Lambton)

Originally Published: December 31, 1997

Last Updated: December 8, 2011

This article is available in print.
Vol. VIII, Fasc. 1, pp. 107-112 and Vol. VIII, Fasc. 2, pp. 113-132