ḴOSROW I i. LIFE AND TIMES

Known as Anōširavān (Mid.Pers. anōšag-ruwān, q.v.), “of immortal soul,” Ḵosrow I (r. 531-79 CE) was given this title after death, to differentiate him from Ḵosrow II (590-628; q.v.), Parviz (Mid.Pers. Abarwēz, “triumphant”).  The Middle Persian form of the name is read as Husraw/Husrōy or Xusrow/Xōsro(y) (Av. Haosrauuah-, *hu-srauuah- “good fame, renowned”; Gk. Chosroēs, Ar. Kesrā [q.v.]; Gnoli, 1995, pp. 120-21; Gignoux, p. 100, n. 465). The clear reference to immortality in his title probably evoked a resemblance with the mythic immortal king, Kay Ḵosrow (q.v.; see KAYĀNIĀN vii). In particular, both kings were distinguished in public memory for their religious zeal as the restorer of the weh dēn, the “good religion” of Zoroastrianism, and for destroying many pagan idols in Iran.

Ḵosrow I was considered as the ideal king throughout Islamic times (de Fouchécour, pp. 38-58; Vitalone, pp. 435-39; Wiesehöfer, 2007, pp. 206-15; see also the contributions in Jullien, 2015) as the “just” (ʿādel) king, and he certainly was one of the main protagonists of Late Antiquity. According to Arthur Christensen’s historiographic view (1944, pp. 363-440), Ḵosrow inaugurated the most brilliant period of Sasanian Iran, through a series of important economic, fiscal, administrative, religious and military reforms, whose global effects probably escape to us (see below, ii. REFORMS). 

Ḵosrow I (b. ca. 496), was the son of Kawād I (q.v.) and the sister of Astabed (q.v.) (Aspebedes, in Procopius, De bello Persico [hereafter, BP] 1.11.5, 23.6), the Sasanian general (see spāhbed) who negotiated peace with Celer, in 506. This high official is likely to be identified with Persian Bōē (Joshua Stylite, tr. Wright, p. 59; Procopius, BP I 12.10). Thus the Xwadāy-nāmag’s tradition (Ṭabari, p. 884, tr., V., p. 128) that Ḵosrow was a son of “Nēwāndukht,” the daughter of a noble dehqān (q.v.) that Kawād encountered during his flight to the Hephthalites (q.v.), might be legendary (Christensen, 1925, pp. 118-19 and p. 95 note 3; 1944, p. 350). 

Following the Sasanian dynastic right, Kāvus (Kaosēs), being the eldest son of Kawād, should have succeeded him, but was disliked by his father, probably for his Mazdakite sympathies (Procopius, BP 2.9.12; see Börm, 2007, pp. 111-19). Zamasp (Zamēs), his other son who was second in age, however, was excluded from succession because he had lost an eye in battle (Procopius, BP 1.11.3-5; Theophylact, AM 6013).  

Kawād feared that after his death some Persian noble could try to eliminate his descendants, in particular, his younger and preferred son Ḵosrow; for this reason, he proposed to the Roman emperor Justin I (ca. 525), to formally adopt him. But this proposal was refused due to the intervention of the Quaestor Proclus (Procopius, BP, 1.11.1-30; Theophylact, AM 6013; Christensen, 1925, pp. 12-4), who offered an adoption “by arms,” according to the barbarian custom, without any written document (Pieler, pp. 427-29; Greatrex and Lieu, p. 81; Börm, 2007, pp. 312-17). This procedure, which patently implied poor care for the future of the next Sasanian monarch, was rejected by Kawād.  Thus, prince Ḵosrow, who was waiting at Nisibis (q.v.) for the powerful Persian ambassadors, Seoses (Siyāwaxš/Siāwoš) and Mebodes (Mahbod), as soon as he heard about the negative meeting with the Romans, had to come back home very enraged (Gariboldi, 2009, pp. 110-11). 

In 531, when Kawād was fatally ill and close to death, the problem of the succession occurred again. Kāvus made an alliance with the Mazdakites (Manicheans, in Theophylact, AM 6016, where Kāvus is indicated as king of Ṭabarestān: see Mango and Scott, pp. 259-61; Christensen, 1925, p. 117; 1944, pp. 353-55), hoping to ascend the throne instead of Ḵosrow. There was a real plan to annul Kawād’s will in favor of Ḵosrow, but he was able to persuade many nobles to vote for him and was legitimately elected king (Börm, 2007, pp. 111-19). Afterwards, he killed all his own brothers, together with the notables and the priests who had taken part in the plot against him (Procopius, BP, 1.23.6; Malalas, ed. Dindorf, pp. 444, 470-72; Histoire nestorienne 2.23, p. 146; Ebn Esfandiār, pp. 147-50). Thus the slaughter of the Mazdakites could have been connected with the bloody removals wanted by Ḵosrow in occasion of his enthronization. He soon started the reforms, but the process was slow (see Crone, pp. 30-2, even if her hypothesis that the revolt of the Mazdakites was a reaction to the reforms is not acceptable; Rubin, 1995, pp. 230-31; Gnoli, 2004, pp. 441-42, with full bibliography on Mazdakism; Fiaccadori, 2006, pp. 128-29; Pourshariati, pp. 83-118, underlines that Ḵosrow tried to divide the power of the Parthian families; Gariboldi, 2009, 85-142; Wiesehöfer, 2009, pp. 402-05).

According to Malalas (ed. Dindorf, p. 471), Ḵosrow was crowned after the death of Kawād on 13 September 531, but a very precise Arab horoscope of his accession to the throne, embedded in the Ketāb al-masāʾel, composed by the Persian astronomer and astrologer Qaṣrāni (Taqizadeh, pp. 128-30), dates the coronation of Ḵosrow to 18 August 531 and tells that he reigned for 47 years and eight months, so till March 579 CE (see also Ṭabari, I, p. 899, tr. V, p. 160; Nöldeke, pp. 429-30). Agathias (4.29.6) and John of Ephesus (Historiae Ecclesiasticae 3.6.20-21), give in total 48 years of rule; Eutychius of Alexandria (Annales, col. 1077) says 47 years and six months; Histoire nestorienne 2.38, p. 197, reports 47 years and some months. The testimony of the horoscope is trustworthy, since it is reasonable that Kawād abdicated before his death caused by an ictus (Gariboldi, 2009, p. 113). The physical and patrimonial persecution of the Mazdakites, anyway, continued long after 531/32 (Yarshater, pp. 998-1006). 

Zoroastrian texts, in general, exalt the pōlāwadēn xwadāyīh, “the reign of steel,” inaugurated by Ḵosrow in defense of religious orthodoxy, because he kept away the “accursed Mazdak” and sustained many theological councils with the most authoritative Magi of the time (Zand ī Wahman Yasn III, 28, II, 2; Cereti, 1995, pp. 134-35, 174-76). Furthermore, a passage of the Bundahišn (q.v.), which probably contains an original fragment of the Xwadāy-Nāmag, tells that Ḵosrow was much praised because he killed Mazdak, restored the true religion, drove away the enemies and finally made the Iranian realm safe (chap. 33.24; Arzarnouche, 2015, pp. 236-45, 250; Pakzad, pp. 367-68; Shapira, pp. 65-66; Messina, pp. 266, 288-89). These actually were the main themes of the Sasanian propaganda concerning the long reign of Ḵosrow.

The Letter of Tansar (tr. Boyce, pp. 33-39) well documents that under his rule the tie between state and religion was strengthened, and Sasanian society was rigidly divided into four estates led by the king (Tafazzoli, p. 2). It is likely that in this period also occurred the written compilation of the Avesta, with the invention of the Avestan alphabet and the codification of many Pahlavi texts (Bailey, pp. 169-72; Gnoli, 1971, pp. 226-27; Cereti, 2008, pp. 179-82). Procopius says that Ḵosrow was “fond of innovations” (BP, 1.23.1), and he liked Greek and Indian cultures (Panaino, 1999, pp. 47-58; Azarnouche, 2013, pp. 36-38; Huby et al., 2016), in particular philosophy, astral sciences, and medicine (John of Ephesus, Historiae Ecclesiasticae, 3.6.20). About 543, he suffered from a serious plague, and Justinian sent to him a learned doctor by name Tribunus (Procopius BP, 2.24.8, 28.8 [545 CE]; Goth. 4.10.10-16; Histoire nestorienne 2.27, pp. 161-62). He ordered the translation of many scientific works from Greek or Syrian into the Pahlavi language (Bailey, pp. 80-88, 157-58; Garsoïan, 1983, p. 577; Panaino, 2001a, pp. 125-28; 2001b), but, according to the ironic words of Agathias (2.28), he actually did not understand Aristotle’s doctrines. 

After the closure of the Academy in Athens by Justinian in 529, Ḵosrow welcomed at his court the seven philosophers who decided to leave the Roman empire, but soon they were disgusted by Persian customs, which appeared fairly “barbarous” to them (Agathias, 2.28-32. On this famous episode and its cultural implications, see Cameron, 1969-70, pp. 165-76; Pugliese Carratelli, 1971, pp. 599-602; Fiaccadori, 1983; Melasecchi, pp. 21-35; Hartmann, pp. 127-38; Börm, 2007, pp. 277-83; Frendo, 2008, pp. 100-106). 

To face the initial difficult political situation, Ḵosrow signed with the Romans the so- called “eternal peace” (Procopius, BP, 1.22.16-19; Malalas, ed. Dindorf, pp. 471, 477; cf. Blockley, 1985b, pp. 70-71; Greatrex and Lieu, pp. 96-97). He quickly eliminated all his internal political opponents and started to improve the Persian economy, continuing the administrative and fiscal reforms that had been carried out by Kawād (Rubin, pp. 229-31; Pigulevskaja, 1937, p. 145; Ognibene and Gariboldi, p. 71; Jackson Bonner, pp. 71-78). The importance of his military and administrative reform is now definitively established by Sasanian primary sources (Gyselen, 2001; 2007, pp. 47-53). Furthermore, a Middle Persian juridical text, Mādayān ī Hazār Dādestān (q.v.) clearly attests (93.4-9; Macuch, pp. 14-15, 596, 600) that Ḵosrow implemented the use of seals for many civil and clerical administrations, such as judges operating in small districts or villages, and he even instituted a special tribunal in defense of the poor (Gyselen, 1989, p. 3; Perikhanian, p. 214). Then he turned against the Romans, claiming that some Arab troops, allied with them, attacked the Lakhmid (q.v.) ruler Monḏer III b. Noʿmān (Ṭabari, I, p. 958, tr., V., pp. 252-53. For the events of the war between Persia and Byzantium see Byzantine-Iranian Relations; Justinian I; Martindale, pp. 303-306; Martindale, s.v. “Chosroes I”; Frye, pp. 153-62. A useful collection of sources is in Greatrex and Lieu, pp. 96-162; Dignas and Winter, pp. 38-42).

With a sudden military campaign, Ḵosrow invaded Syria and pillaged many cities, conquering Antioch (q.v.), in 540. He deported its inhabitants to a new town, purposely built for them near Ctesiphon (q.v.), named Wēh-Andiyōk-Husraw, “the good Antioch of Ḵosrow” (Börm, 2006, pp. 303-305; Gariboldi, 2007-8, pp. 28-29). According to the Šahrestānīhā ī Ērānšahr (q.v.), Ḵosrow built five cities named after him (Daryaee, 2002, p. 14, par. 19). 

We also know some Sasanian sealings with the toponym Ērān-abzūd-Husraw, “Ḵosrow (has) increased Ērān,” an area probably located in upper Mesopotamia (q.v.) under the fiscal control of an āmārgar (q.v.), “accountant” (Gyselen, 2002, pp. 140-42; 2007, p. 43). An influential secretary of Ḵosrow was Pābak (Bābak, q.v.), who carried out the land reform and enjoyed great prestige at court (Ṭabari, I, p. 963, tr. V., p. 262; Dinavari, p. 74; see Browne, p. 231; Tafazzoli, pp. 20-33). Christian prisoners were divided into artisan corporations under the direct control of the king (Jullien and Jullien, p. 161). He deported farmers also from Callinicus in 542 to acquire labor supply for the agricultural development of Persia (Morony, 2004, pp. 173-76; see DEPORTATIONS). In the city of Sura on the Euphrates (q.v.), he found among the captives he had taken a beautiful woman, Euphemia by name, and married her (Procopius, BP, 2.5.28), but he had many other wives. On the way back to Persia, Ḵosrow exacted a large amount of gold from the cities he encountered to spare their inhabitants (Gariboldi, 2006, p. 422; 2009, pp. 68-69; Iluk, pp. 90-92; Altheim and Stiehl, 1954, pp. 46-50. Curiously, the third episode of the Sirat Anuširwān, in Grignaschi, 1966, p. 17, reports that the king gave back to the poor people of Syria part of the gold tribute he received from Justinian).

In 541, by invitation of Gubazes, king of Lazica, he invaded this strategic region, which offered to the Persians an important outlet on the Black Sea (q.v.). The Sasanians maintained the possession of Lazica till 548, when Gubazes returned to a policy in favor of the Romans (Greatrex and Lieu, pp. 115-22; Bertinelli Angeli, pp. 126-32; Braund, pp. 298-311). Warfare continued in Lazica up to 557, when a truce was concluded on the basis of the status quo (Agathias 4.30.7-10).  Ḵosrow then sent to Constantinople his powerful chamberlain Zīg Yazd-Gušnasp, to negotiate with care the terms for peace (Constantine Porphyrogenitus, 1.89 [p. 405]; see Manini, pp. 130-33; Diebler, pp. 189-200), which was finally signed at Dārā (q.v.), on the border of the two great empires, in 561/62 (Menander Protector, frag. 6.1-3; cf. Blockley, 1985a, pp. 55-91; Dignas and Winter, pp. 138-48. On the letter sent by Ḵosrow to Justinian, illuminating Sasanian royal ideology, see Huyse, pp. 194-96; Panaino, 2009, pp. 230-31). 

This “fifty years” peace treaty provided, in substance, the cession of the Lazica to the Romans, in change of an annual payment of 30,000 nomismata to the Persians (Güterbock, pp. 61-65), which mainly served to sustain military expenses for the control of the Caucasus passes, even if a final agreement about the Suania region was not found. Ḵosrow strongly fortified the Darband (q.v.) pass with long walls and military posts (Howard-Johnston, pp. 191-97; Harmatta, 1996, 81-82; Mahamedi, pp. 150-52). Nisibis and Dārā were confirmed as the main trade centers with the Romans, and they also included an international agreement regarding the tolerant treatment of religious minorities in both empires (Winter, pp. 67-72; Guillaumont, pp. 48-49).

Under Ḵosrow, we know of just a few persecutions against Christians: Usually they were Persian nobles converted to Christian faith, as the case of martyr Šīrīn (Devos, pp. 6-11; see MARTYRS, CHRISTIAN) or the katholikos Mār Abā (q.v.; 540-52), who was, on the contrary, protected many times by the king himself against the Magians (Peeters, pp. 88-111; Garsoïan, 1973, pp. 131-33). Ḵosrow cunningly took advantage of the strong influence of Mār Abā to repress the riot headed by his eldest son Anōšagzād (see ANŌŠAZĀD), “born immortal”, in 550-51, who, being a Christian, had been banished to Bēṯ Lapaṭ (q.v.; Histoire nestorienne 2.27, pp. 162-63; Dinavari, pp. 70-71; Ferdowsi, ed. Mohl, VI, pp. 225-43. See Nöldeke, pp. 467-74; Labourt, pp. 189-90; Christensen, 1944, p. 383; Pigulevskaja, 1963, pp. 221-28, connects this episode with the Mazdakite upset in Khuzestan; Hutter, p. 169; Panaino, 2004, pp. 815-19; Jullien, pp. 222-24; Gariboldi, 2009, pp. 138-39). When Anōšagzād learned that his father was ill, he tried to usurp the royal power (Procopius, Goth. 4.10.8-22), thus Ḵosrow scalded his eyelids to prevent him from becoming king. Procopius (BP 2.24.8) says that another son of Ḵosrow revolted against him in 543 CE, but it is likely a duplicate of the same story.

Ḵosrow was also engaged with the eastern borders of the Empire, for the Western Turks invaded Central Asia. The Sasanians made an alliance with them around 557-61 to defeat the Hephthalites. Ḵosrow got hold of the territory south of the Oxus river (see AMU DARYĀ), and in order to consolidate the relations with the Turks, he married the noble daughter of the khaqan (q.v.), who gave birth to his future successor to the throne of Persia, Hormozd IV (q.v.; Masʿudi, Moruj, II, p. 211, ed. Pellat, I, p. 312, says that this woman was named Fāqom; Ṭabari, I, p. 988, tr. V., p. 295; see Widengren, 1952, pp. 76-78; Mochiri, pp. 238-39). 

Some Bactrian documents testify that the Persians, in about 560, were the opponents of the Hephthalites in northern Hindu Kush (q.v.), and the further evidence of a Sasanian bulla indicates that the region of Kadagistān (q.v.) was under the control of an ōstāndār, a “provincial administrator,” in the late 6th century (Sims-Williams, 2000, p. 197; 2008, pp. 96-99; Gyselen, 2002, p. 152). Probably in 570, Ḵosrow launched a successful expedition to conquer Yemen. The Sasanian army ended the Abyssinian occupation and installed Sayf b. Ḏi Yazan as Ḵosrow’s client king. Thus, Ḵosrow secured a new maritime way to avoid the Byzantine commercial block (Ṭabari, I, pp. 948-49, tr. V., pp. 239-42; Morony, 2002, p. 34; Potts, pp. 206-9).

Yet the peace with the Romans lasted only ten years, since in 572 a revolt against the Persians broke out in Armenia and the marzbān Surēn was killed at Dvin (q.v.; Sebēos, ed. Thomson, p. 6; Garsoïan, 2009, p. 108). Thus the emperor Justin II (520-78 CE) refused to pay the annual tribute established under the treaty terms and gave military help to the Armenians. In addition, the Turkish-Byzantine connection was an important cause for the renewal of the war between Byzantium and Persia (Menander Protector, frag. 16.1; cf. Blockley 1985a, pp. 150-55; Turtledove, pp. 298-99). In 573 the Sasanians conquered Dārā. After vain diplomatic contacts in 575/6, Ḵosrow, notwithstanding his old age, personally led his troops through Mesopotamia and Armenia, but, in the end, the incursion was not a great success (Menander Protector, frag. 18.6; Theophylact, 3.14. 11; John of Ephesus, 6.8-9; Evagrius, 5.14-15; Whitby, 1994, pp. 227-31; 1986, pp. 262-67). During new negotiations for peace, he seriously fell ill with chest complaint (Histoire nestorienne 2.38, p. 197), and consequently died in Ctesiphon in February/March 579 CE, before the ambassadors sent by emperor Tiberius could reach Persia (Agathias, 4.29.9-10; Menander Protector, frag. 23.9. Cf. Blockley, 1985a, pp. 207-9; Theophylact, 3.16.7). The tale reported by Sebēos (I, pp. 9-10) that Ḵosrow would have been converted to Christian faith on his deathbed is fanciful, and it rather reflects the general positive trend of Christian sources towards him.

ḴOSROW I ii. REFORMS

A series of reforms in the taxation of the Sasanian monarchy and in its military organization, probably initiated already under Kawāḏ I, were carried out to their full extent by his son, Ḵosrow I Anuširwān (r. 531-79).

Sources. The two most detailed surviving accounts in Arabic are Ṭabari’s Taʾriḵ al-rosul wa’l-moluk and the Nehāyat al-erab fi aḵbār al-fors wa’l-ʿarab, falsely attributed to Asmaʿi. In New Persian, Ferdowsi’s Šāh-nāma is likewise valuable. Balʿami’s modified translation of Ṭabari into Persian adds at least one important detail. These sources appear to derive, through different lines of transmission, from various redactions of the lost Sasanian Xwadāy-nāmag (Khwadāy-nāmag). A literary source which seems to have used good sources independent of the Xwadāy-nāmag tradition is the so-called Sirat Anuširwān, professedly an autobiography of King Ḵosrow I, embedded in Meskawayh’s Tajāreb al-omam. Other sources in which some valuable details about the reforms have been preserved are the geographical surveys of Ebn Ḵordāḏbeh, Ebn Ḥawqal, and Ebn Rosta (see “Bibliography”).

Background of the reforms. There are clear signs that the economy of the Sasanian kingdom had never been based on full-scale monetary circulation, in spite of its fairly stable silver drahm denomination (see DIRHAM i). The absence of a convenient, universally applicable small change circulation (Göbl in Altheim and Stiehl, 1954, pp. 96-99; Göbl, 1971, pp. 25-30; Simon, 1976, pp. 150-51) entailed an economy based to a large extent on the barter of natural produce. The little that we know about the military organization of the kingdom before the reign of Ḵosrow I appears to reflect this situation. From our meager information about remuneration for the professional core of soldiery, we may conclude that it was supported through land grants or through rations in kind rather than paid in money. This has given rise to the notion of a feudal army based on enfeoffment, entailing bonds of trust and dependence, either to the king himself or to the other grandees of his realm, which may be described as ties of vassalage (Widengren, 1956).

This situation explains a good deal about the Sasanian system of taxation before the beginning of the sixth century. It was based on crop-sharing, the exaction of agricultural produce proportionate to annual yield as assessed by royal tax-collectors on the spot, and levied in kind. In addition, a poll tax was imposed on most subjects, which may have been paid mostly in money, though part was perhaps commuted to goods. It must be conceded that this picture, based on the later Islamic sources, is hardly consonant with the one that emerges from details scattered in the Babylonian Talmud. Apparent discrepancies may, however, be explained by the assumption that, whereas the Talmud reflects the experience of Jews (in places where most of the Jewish population of the realm was concentrated) regarding the manner in which the system was actually applied, the later Islamic sources record mainly the principles according to which it was supposed to have been run (see mainly Newman, 1932, pp. 161-86; Solodukho, 1948; and Goodblatt, 1979, for the evidence; cf. Rubin, 1995, pp. 231-33, for a tentative solution of the problems it creates).

The system was inefficient and wasteful, especially with regard to the land-tax; it was subject to frequent fluctuations, and allowed little scope for financial planning. The necessity of waiting for the tax-collector with the crops untouched in the field or on the tree involved the risk that some would be damaged or destroyed before being enjoyed by farmers or the king (as exemplified by one particular episode preserved in Ṯaʿālebi, p. 595; Ebn Ḥawqal, pp. 203-4; and Balʿami, p. 148; see also Pigulevskaya, 1937, pp. 145-46). Only lands held directly by the king could be effectively taxed in this manner, but even on royal domains the avarice of corrupt tax-assessors will have hampered collection.

The tax reform. Towards the end of the 5th century, the burden of taxation on the peasantry seems to have become increasingly oppressive: the complex relations with the Hephthalite khanate, whose threat loomed in the east, resulted in heavy demands, though recurrent famines were compelling kings to grant occasional, and not entirely adequate, tax remissions. This oppression contributed significantly to the popularity of Mazdak (on whom see IRAN ix. RELIGIONS IN IRAN, XIII/4, pp. 437-38), and to the temporary success of his revolt. For some time he managed to enlist the support of King Kawāḏ I himself, who appears to have used this movement to humble his recalcitrant nobility.

When the king ultimately turned his back upon the movement and allowed his son, Ḵosrow I Anuširwān, to put it down, the battered nobles were in no position to form a viable opposition to the one serious attempt to introduce a tax reform in the Sasanian realm. It was apparently begun towards the end of Kawāḏ I’s reign (Ebn Ḵorrdāḏbeh, p. 14; Ebn Ḥawqal, p. 234; and Ebn Rosta, p. 104, ascribe it mainly to him) and continued by his son (for a more detailed discussion, see Rubin, 1995, pp. 229-31, 290-91). The reform involved the following features:

(1) The land-tax (ḵarāj in the Arabic sources). On the basis of a general land survey, a new system for exacting the land tax was devised. Fixed money rates of taxation were imposed on agricultural land according to its size and according to the kind of crops raised. According to Ṭabari (pp. 990 -91; cf. Bosworth, pp. 255-56) these rates were, 1 drahm for every jarib of cereals; 8 drahms for every jarib of vine; 7 drahms for every jarib of clover; and 1 drahm for every four Persian date palms, or for every six plain date palms (on the misleading suggestion of Grignaschi that the calculation was made according to area units, applying one fixed rate to all sorts of land and produce, see Rubin, 1995, pp. 266-68). The tax was calculated in the drahm currency, but at least some probably continued to be levied in kind, calculated according to the current value of the produce in drahms (thus, e.g., Pigulevskaya, 1937). This new system, if efficiently applied, would enable a monarch to anticipate incomes and budget expenses. It might be seen as oppressive on the peasantry, primarily because the fixed drahm rates allegedly disregarded fluctuations in agricultural yield caused by drought, other natural calamities or war (thus Grignaschi, 1971, followed by Crone, 1991), but this is to ignore its almost unanimous positive evaluation in the sources. A control mechanism allowing for rebates and remissions whenever and wherever needed, based mainly on mobads (mūbadān) acting as district judges, was incorporated in the system (Rubin, 1995, pp. 261-66; 293-94).

(2) The poll tax (jezya in the Arabic sources). It is Ḵosrow’s revised poll-tax which bears out, by the very nature of its progressive rates, according to the taxpayer’s means, his basic intention to devise a fundamentally fair system. It is true that it did not do away with the exemptions granted to members of the higher classes, but even these were somewhat counterbalanced by the exemption of people below twenty and above fifty of all classes (Ṭabari, p. 962; cf. Bosworth, pp. 259-60).

If a distinction is drawn between the tax reform’s institution and operation in Ḵosrow’s reign, and what it subsequently became, the system appears reasonably efficient and fair (Nehāyat al-erab, p. 330, describing it as unfair and oppressive, appears to be the one discordant voice among the sources we possess). It considerably augmented crown revenues, but, as already stated, it also included a mechanism that enabled constant revision and adaptation.

(3) The rehabilitation of agriculture. The fiscal reform was accompanied by an agricultural reform. Dispossessed farmers were restored to their lands, financial help was available to enable them to restart cultivation, and a mechanism was instituted to assist farms affected by natural disasters (Rubin, 1995, pp. 254-66). The overall intended result seems to have been to maintain a system of small farms that might be easily taxed, and to prevent the growth of huge estates whose powerful owners might accumulate privileges and immunities, and obstruct effective taxation.

The question whether Ḵosrow I’s system was inspired by that of Diocletian in the Roman Empire (see Althein and Stiehl, 1958, pp. 41-42, and Pigulevakaya, 1937; cf. Hahn, 1959) may be left aside. More than two centuries separate the institution of the two systems, and there was enough in the Iranian background to account for devising the former without recourse to the latter (Rubin, 1995, pp. 295 -96).

(4) The military reform. Ḵosrow’s reform was meant to have a lasting impact on Sasanian military organization by providing the king with a standing army of crack units of horsemen (asavarān; see ASWĀR), under his direct command and permanently at his disposal, who received a salary, at least when on foreign campaigns (for the evidence and its analysis, see Rubin, 1995, pp. 286-91). This body of choice warriors was recruited among young nobles as well as from the country gentry (dehkanān; see DEHQĀN) who wished to embark on a military career. On the frontiers, other troops recruited from the nomadic periphery of the Sasanian Empire, such as Turks (Meskawayh, pp. 102-3, cf. pp. 106-7), might be employed to repel invasions or check them until the arrival of the mobile crack units. Yet other forces, recruited from semi-independent enclaves within the kingdom, might even be used as special task forces, as the one sent to occupy Yemen (see ABNĀʾ), which was recruited in Daylam in the mountainous region of Gilan. The need to resort to such additional sources of manpower, without a clear notion of how they were to be remunerated in the long run, exposes one of the system’s major weaknesses (for the evidence see Rubin, 1995, p. 272, n.122, pp. 284-85).

The supreme command was brought under stricter direct royal control: the single commander in chief, the Erānspāhbed (Artēštārānsālār; see ARTĒŠTĀR), was replaced by four spāhbeds, each one at the head of the forces of one-quarter of the realm, and each one beholden directly to the king (Ṭabari, p. 894, with Bosworth, p. 149, n. 385, updating Nöldeke, 1879, p. 155, n.2; cf. Christensen, 1944, pp. 131-32). The commanders of the frontier marches—the marzbāns—were likewise supposed to take their orders directly from the king (Christensen, 1944, pp. 521-22; Rubin, 1995, pp. 292-93, commenting on Meskawayh, pp. 106-7).

The operation of the reforms and their long-term effect. Ḵosrow’s system appears to have enjoyed moderate success for a few decades, until the difficulties that beset the Sasanian monarchy exposed its weaknesses. According to Ebn Ḵorrdāḏbeh (p. 14) the amount levied in the Sawād alone (i.e., the province of Āsōristān), already under Kawāḏ, was 150,000,000 weight-unit drahm (derhām meṯqāl). Effective central rule was, however, as necessary for the maintenance of the system as its proper functioning was for the endurance of effective central rule. The Sirat Anuširwān indicates that towards the end of his reign, Ḵosrow was hard put to keep his system functioning (Meskawayh, pp. 104-6; Rubin, 1995, pp. 237-39, 279-84). The control mechanism he had instituted proved to be as susceptible to corruption as the taxation machinery that had to be controlled. Furthermore, the strained relations between soldier and civilian, especially in the remoter zones, exacted their toll. In effect, the king could restrain only the soldiers under his direct command from despoiling the rural taxpayers, as is shown by the restrictions imposed by Hormozd IV (r. 579-90) on a journey to Mah (Ṭabari, p. 989; cf. Bosworth, p. 296). As time went on taxation in the Sasanian realm became more and more extortionate. Even so, the huge sums of money reported to have accumulated in Ḵosrow II’s coffers (Ṭabari, p.1042; cf. Bosworth, p. 377) were by no means due to effective taxation alone (see Altheim and Stiehl, 1958, pp. 52-53; cf. Rubin, 2000, pp. 656-57). It may be concluded that assertions like the one made by Ṭabari (p. 962; cf. Bosworth, p. 260), can hardly be used to support the prevalent notion that the Muslim conquerors took over Ḵosrow’s system in its original, fully fledged, and uncorrupted form.

The long-run failure to maintain the financial resources on which Ḵosrow’s reformed standing army was to be based inevitably led to its gradual degeneration. Already in his own lifetime the asavarān were increasingly reverting to an enfeoffed estate, even though the tendency of such allotments to become hereditary, with the consequent problems caused by alienation, had to be faced (Rubin, 1995, pp. 294-95, with n. 159 for additional bibliography; Rubin, 2000, pp. 658-59).

The high nobility was a more serious problem than the cavalrymen. Restored by Ḵosrow himself after its humiliation during the Mazdakite revolt, it soon returned to its pristine positions of power. The notion that the supreme military commanders (such as the spāhbeds and the marzbāns whose offices were created by Ḵosrow himself), and ministers of state were now salaried civil servants is contradicted by the limited available evidence (Rubin, 2000, pp. 657-58). Dinawari’s statement (p.102) that Besṭām remitted half of the land tax in territories under his control during his rebellion against Ḵosrow II suggests what other potentates, not in direct revolt, some of them in royal employ, may have done, less openly but less generously.

ḴOSROW I iii. COINAGE

The reign of Ḵosrow I is generally regarded as constituting the heyday of the Sasanian Empire. His coinage, rather to the contrary, marks the nadir of Sasanian coin art. The appalling quality of Ḵosrow’s coins are caused, to some extent, by serious production problems, since the flans were too thin for the impression of the dies, with the result that many parts of the coin image remain obscure (so-called “dead” or “blind spot”). Even worse are the artistic insufficiencies as the average portraits look more like caricatures than the dignified depictions of the King of Kings of Iran. Even if Ḵosrow I introduced some new typological features, in general there is much more numismatic continuity between him and his father Kawād I than one would expect by following the written sources.

Typology. While the reverse typology of Ḵosrow I is to some extent innovative, his obverses are conservative. With only minor variations, the last obverse type of Ḵosrow’s predecessor Kawād I is used for the entire reign. The crown also fully corresponds to his father’s; thus, the normal type shows Ḵosrow’s right-facing bust wearing a crown consisting of a crescent above the forehead and two mural elements (type SN [Göbl, 1971] I, II; see PLATE I. a-d, g). On both shoulders, a crescent is depicted, and in the left and right field one star respectively; both features were introduced by Kawād I. The main alterations of Ḵosrow I, compared to his father’s coinage, are the less prominent depiction of the upper ribbons and the replacement of the star/crescent combination at 3h, 6h, and 9h, outside the rim by just a crescent. It seems very likely that this had propagandist reasons, since Hormozd IV, who like Kawād I had a hostile attitude towards the nobility, returned to this king’s astral symbols, while Wahrām VI, who rebelled against Hormozd IV, employed the crescents following the pattern of Ḵosrow I. On rare gold dinars, a special obverse type showing Ḵosrow’s frontal bust is attested, a direct takeover from Kawād I (type SN III, PLATE I. e,f). Ḵosrow’s obverse legends in the first five years of rule simply cite the royal name, following the patterns of the early issues of Kawād I. In regnal year 5, the acclamation ʾpzwny “may he prosper” is added—another parallel to Kawād. Two spelling variants of Ḵosrow’s name exist: the more common one reads hwslwb (PLATE I. b-e), and the earlier and more rare reads hwslwdy (PLATE I. a, f).

PLATE I.  Coinage of Ḵosrow I (RY = regnal year).a. AR Drachm, type SN I/1, mint LD, RY 5. Bibliothèque nationale de France, Cabinet des Médailles, Paris. b. AR Drachm, type SN II/2, mint AYLAN, RY 6. Bibliothèque nationale de France, Cabinet des Médailles, Paris. c. AR Drachm, type II/2, mint HWC, RY 27. Bibliothèque nationale de France, Cabinet des Médailles, Paris. d. AR Drachm, type II/2, mint BYS, RY 47. Bibliothèque nationale de France, Cabinet des Médailles, Paris.e. AU Dinar, type SN III/3, no mint, RY 21. Busso Peus, Auction sale 368, 2001, no. 396.f. AU Dinar, type SN III/4, no mint, RY 13. Mosig-Walburg,1993, fig. 7.g. AE, type SN II/2, mint and RY illegible. Bibliothèque nationale de France, Cabinet des Médailles, Paris.View full image in a new tab

The most noteworthy features of Ḵosrow’s coinage are his innovations in reverse typology. The first type (type SN 1, PLATE I. a), struck from regnal year 1 to 5, has two major differences compared to the reverses of the period from Pērōz until Kawād. Firstly, the assistant figures are shown frontally, a totally new depiction in the repertoire of Sasanian typology. Secondly, they hold what appears to be a spear, an attribute encountered previously under Wahrām II (276-93). It is, in the context of this so to speak “archaic” depiction, no mere coincidence that during Ḵosrow’s early years one can observe attempts aimed at a naturalistic rendering of the reverse figures in the fashion of 3rd-century issues. The attendants’ bodies are shown much finer and more detailed than under Kawād, and, very rarely, quite elaborate depictions of the crowns can be found (see PLATE I. a). The quality, however, quickly depreciates. Another typological modification of the early years is the treatment of the altar shaft that is formed by a thin vertical line crossed in the center by three horizontal strokes without the usual altar ribbons. From regnal year 5 to 48, Ḵosrow employed another reverse type that was to become—with minor modifications under Ḵosrow II—canonical until the end of Sasanian-style coinage (type SN 2; PLATE I. b-d, g). It features both assistant figures, frontal, leaning on a sword. Also, the treatment of the fire altar is modified compared to type SN 1, with the altar ribbons again depicted and—for the first time in Sasanian typology—showing with ends upwards instead of hanging downwards. The gold dinars from Ḵosrow’s regnal years 21 and 44 (the latter ones formerly incorrectly believed to refer to what was supposed to be the last regnal year of Kawād I) show the standing figure of Ḵosrow holding a diadem wreath to the right (type SN 3, PLATE I. e), once again a direct takeover from Kawād I. Also, a dinar dated regnal year 13 showing on the reverse the frontal figure of the king leaning on a sword belongs to Ḵosrow I, rather than to Ḵosrow II, because of style and flan size (Schindel, 2006) (type 4, not in Göbl, 1971; PLATE I. f). Minor typological modifications of the basic type SN 2 occur on copper coins, but most specimens are too badly preserved to allow a conclusive typological attribution. The gold dinars with reverse types 3 and 4 feature special legends, respectively gyhʾnprʾknyt “who bestows splendor on the world” and gyhʾnʾpybym krtʾl “who frees the world from fear.” Apart from these, all other coins display the mint signature at 3h, and the date at 9h, save for some copper coins, on which due to lack of space or careless engraving this administrative information is missing. Sometimes, additional marks outside the rim occur on the drachm reverses, such as the letters hw (most likely for “good”) in regnal year 8, as well as various combinations of dots (cf. PLATE I. d).

Denominations. The narrowing of the denominational spectrum already came to an end under Kawād I. Thus, under Ḵosrow I—as well as under all his successors—only three different denominations exist: gold dinars, silver drachms, and small copper coins. The dinars (PLATE I. e, f) are extremely rare: only five genuine specimens are known. All employ special types on both sides. The authenticity of gold coins which feature regular drachm typology and bear the mint signature AT is questionable; in any case, they are not fully official products. Drachms of Ḵosrow I (PLATE I. a-d) are very common, even if not as abundant as those of Ḵosrow II. No silver fractions exist. Copper coins (PLATE I. g) are very scarce, with most specimens being not well enough preserved for a full attribution by mint and date.

Mints. At the present state of research, about 45 different mints are attested (see SASANIAN COINAGE). Most were in existence already under Kawād I (Schindel, 2004, pp. 470-71), but some mints are additions of Ḵosrow I. The most important among these are HWC (Xūzistān), LAM (Rām-Hormozd in Xūzistān) and NAL/WAL (uncertain). The mint signature AS (Āsūristān), one of the most common signatures from Wahrām IV onwards, is last attested in regnal year 22 of Ḵosrow I. In regnal year 23, WYHC (Weh-az-āntiyok-Xusro) starts operating on a large scale, which proves that both signatures refer to the same location, i.e., Ctesiphon. Some mints that were founded by Kawād I, but which were rather insignificant during his reign, become increasingly important under Ḵosrow I, the most prominent examples being AYLAN and WYH. Perhaps this is to be connected with the taxation reform begun by Kawād and brought to completion by Ḵosrow. In any case, the Sasanian mint organization was always flexible. Seen as a whole, the most common mints are AY (Erān-xwarrah-Šāpur), AYLAN (uncertain), LD (Ray) and BYŠ (Bišāpur) as well as Ctesiphon if one adds AS and WHYC. Detailed mint statistics can be found in Schindel (forthcoming). In general, the patterns of coin production show an organic development based on the system of Kawād I, as no drastic alterations of Ḵosrow I are recognizable. The dies were produced centrally for the entire Sasanian realm