Richard Frye. The Heritage of Persia

starting with pg. 198.

HEIRS OF THE ACHAEMENIDS

Ardashir and the Cycle of History

FOR THE PERSIANS solid history begins with the Sasanians. What transpired before Ardashir is vague and legendary, a heroic age; but this does not mean that after Ardashir we escape myth and uncertainty, for what happened and whal people believe should have happened are frequently confused even in that portion of Iran's history which is related by many different sources. The story of the founding of the Sasanian dynasty is not unlike the story of Cyrus or even Arsaces, both of which generally conform to epic norms.

In one Pahlavi source, the Kar Namak of Ardashir, or his 'book of deeds', it is related that Sasan was a shepherd of King Papak who ruled in the city of Istakhr near Persepolis. Sasan was a descendant of the Achaemenids, but he kept this a secret until Papak had a dream which was inerpreted that the son of Sasan would one day rule the world. So Papak gave his daughter to Sasan and from this union Ardashir was born. This story is repeated by Firdosi in the national epic and it was evidently widely believed since Agathias, living in the sixth century, gave a somewhat garbled version of the story, stating that Papak was an astrologer and Sasan a soldier who was a guest in his house. Recognising signs of greatness in Sasan Papak gave him his wife and Ardashir was born. Much later when Ardashir was king a quarrel between the two old men broke out, which was settled by calling Ardashir the son of Papak though descended from Sasan.

Another tradition found in Ibn al-Athir (ed. Tornberg I.272), in Eutychius (ed. Cheikho, foll. 65v) and others, has Sasan a princelet in Fars, Papak his son and Ardashir his grandson. This is the position adopted by most scholars today, especially after the discovery -of the famous trilingual inscription of Shapur I on the Ka'bah of Zoroaster which is the Sasanian counterpart of the OP Behistun inscription. This inscription, however, merely names Sasan with a title 'the lord' presumably as an ancestor, while Papak is here and elsewhere specifi- cally called the grandfather of Shapur. The mother of Papak is given as Denak, but it is not stated whether she was married to Sasan who is never named as Ardashir's grandfather, although this is a probable assumption. Therefore an obscurity does exist, even in the inscriptions, about the exact relationship between Sasan and Papak.

In the Syriaic chronicle of Arbela, we read that in the time of Vologeses IV (circa AD 191-207) the Parthians fought against the Persians, and later the same chronicle says: 'In earlier times the Persians tried to unseat the Parthians; many times they exerted them- selves in war but were defeated.' The chronicle further says that later the Persians and Medes made an alliance with the kings of Adiabene and Kirkuk and that together they overthrew the Parthians. The date and circumstances of the defeat and death of Artabanus V, the oppo- nent of Ardashir, are not clear; the usual dates have been given as either AD 224 or 226. The coins of the last Arsacids, however, confuse the matter, so much that a long joint rule of Vologeses V (207-227?) and Artabanus V (213-224?) has been proposed with the son of Artabanus, Artavasdes, ruling one year 226-227. Inasmuch, however, as Arsacid resistance did not end with the death of Artabanus one might suppose that coins of the last Arsacids were minted in his name even after the victory of Ardashir which may be dated from various sources probably as April 224.

In the titulary of the royal Sasanian inscriptions one may see the expansion of the state. Sasan, as noted, is referred to merely as 'lord' while Papak is 'king'. Ardashir is 'king or kings of Iran' and Shapur is 'king of kings of Iran and non-Iran'. An indication of how one might be misled in interpreting an inscription, is the appellation 'god' (bgy) for Papak in KZ, but 'Mazda worshipping god' for Ardashir and Shapur. This might induce one to assume that Papak held a different position or faith in religion than his son and grandson. Yet the same formula appears on later inscriptions in Taq-i Bustan, and one cannot conclude anything from the practice of omitting 'mazda-yasnian' from the name of a grandfather. The phrase 'whose seed (or origin) is from the gods', however, is a continuation of a Seleucid if not Achaemenid formula while the term 'god' applied to the ruler had probably by this time assumed the significance of 'your majesty' in protocol.

For the dating we fortunately have an inscription written in the Parthian and Sasanian Middle Persian languages on a pillar in Bishapur. The text says: 'in the month of Fravardin of the year 58, forty years of the fire of Ardashir, twenty-four years of the fire of Shapur, (which is) the king of fires'. On the reverses of Sasanian coins we have Aramaic NWR' ZY 'fire of-' until Shapur II; then we have the Iranian 'twr y until Yazdagird II (439-457) after whom it disappears. Each king apparently had his own fire, lighted at the beginning of his reign, and this fire was on a portable fire altar similar to those on the coins, as one would gather from Sebeos the Armenian writer, from Ammianus Marcellinus and from others. Shapur's fire was caled the king of fires possibly because it was identified with the Gushnasp fire of the warriors, which was later designated 'the victorious king of fires', but the text is not clear), or maybe the king's fire was called the king of fires simply as a manner of speaking. The date of accession and the date of the crowning of a king have usually differed in the ancient Orient, and these dates are not precisely known in regard to Ardashir and Shapur. From the inscription of Bishapur we would have three dates, the beginning of the Sasanian era, the accession of Ardashir and the accession of Shapur. Great controversy has raged over the date of Shapur's accession and crowning, but his first year must begin either at the end of 239 or 241. The coins of Artabanus V and Vologeses, mentioned above, would tend to favour the year 241 but they are not decisive.

If Papak had been the director of the Anahita shrine at Istakhr before he became king, afterwards he and especially his son were busy with other affairs, even though both may have retained the dignity as head of the temple. Papak had a small court, the most prominent members of which are named in Shapur's great trilingual inscription. There is only one title, the major domo (dnyk), mentioned and no religious designations, so one should assume that Papak's court was that of a small principality with no bureaucratic tradition. After Ardashir became the king of kings of Iran, the successor of the Parthians, the situation changed. Ardashir inherited the feudal organisation of the Arsacids which is clearly seen in the inscription. At the new court we find an order of protocol beginning with four powerful eastern kings, three of whom oddly have the same name as Ardashir. The first on the list is the king of Khurasan, the upper country and homeland of the defeated Parthians, while the second is the king of Merv who is called Ardashir. It would be natural to suppose that relatives or close friends of Ardashir were appointed to offices in the new empire, especially in the important posts in eastern Iran, but we do not know the relationship of these rulers to the king of kings. The next two kings of Seistan and Kirman are also both called Ardashir, the latter, according to Tabari, being a son of the king of kings. One may further assume that these 'kingdoms' were won by force of arms, and hence were free to be assigned to favourites, while rulers who submitted to the Sasanian monarch probably retained their principalities in a feudal relationship.

The inscription continues with three queens, probably the king's mother, grandmother and sister, the 'queen of queens'. Then follows an Ardashir the bitaxs and a Papak the chiliarch (hazarpat). From their names and high rank both were presumably members of the Sasanian family. The former was probably almost like an assistant to the king since the title as used earlier in Georgia implies that there the pitiaxsi was second to the king in rank and importance. At the Sasanian court this rank may have declined somewhat, so that the bitaxs and the chiliarch divided the civil and military direction of the affairs of the empire between them.

The heads of the great Parthian feudal families are next in the list, first the Varaz family which is new. The Varaz may have been essentially a northern Iranian family since the name appears frequently in connection with Armenia or Azerbaijan. Second in rank of the feudal families is a representative of the famous Suren family, while third comes the lord of Andegan, also called Indegan, presumably another feudal appanage. Two members of the well-known Karen family are followed by a name known elsewhere, Apursam, who bears the honorific 'glory of Ardashir', followed by the lord of the area around Mt. Demavend and a member of the Saphpat family which ends the list of families.

The chief of the scribes, chief of the armoury and other officials, as well as prominent persons with no offices named, complete the list of people in Ardashir's court who were honoured by having sacrifices performed in their names at the fires established by Shapur I at Naqsh-i Rustam. The court of Ardashir shows the same features of an unfixed central state and bureaucracy which also would have been characteristic of the Parthian court, and everything points to a continuity from the past. The early coins of Ardashir too are copies of those of Mithradates II, but the traditions of iconography of the various crowns worn by the early Sasanian kings are by no means clear. One must resist the temptation to see cultic or religious significance in every feature of ancient art and archaeology even though such ideas must have been frequently present.

Later Sasanian tradition, reported mainly in Arabic sources, traces the beginnings of all institutions of church and state back to Ardashir. He is the ruler who reinstated or resurrected the old Persian empire with its various institutions as well as the religion of Zoroaster which had been in eclipse under the Hellenistic kings and the Parthians. Apursam, the confidant of Ardashir, was credited with holding the office of prime minister (vuzurg framadar) while Tansar was the first chief mobad according to Arabic sources. The purpose of the later Sasanians in attributing an early origin for many offices was probably that they wished to seek authority for new developments by clauning that these were in fact not new, but dated from the beginning of the empire although they had fallen into decay. The antiquarian renaissance of the time of Chosroes I is well known and will be discussed below, and this was probably the period when the reference of institutions back to Ardashir was made. A writer in Arabic Mas'udl, for example, not only attributed the founding of certain offices to Ardashir but also the ordering of society into classes which, however, could not be the work of one king, Ardashir I.

From Shapur's inscription we can also infer the extent of Ardashir's empire. From Islamic and other sources scholars have proposed that Ardashir re-established the Achaemenid empire in the east including the Punja and did well in advancing the frontier against the Romans in the west. The same sources, however, tell us that Ardashir had much fighting to do to consolidate his rule, especially in Armenia where resistance was strong. The fact that in inscriptions Ardashir is called the king of kings of Iran, but not of non-Iran, would imply that he did not appreciably advance his boundaries outside of Eranshahr which, of course, included Mesopotamia but not Armenia (according to the Paikuli inscription, line 8) and probably not the Kushan empire in the east. On the other hand Tabari says that the kings of the Kushans, of Turan and Makran came to Ardashir, after his victories in the east, and offered their submission. It is possible that under Ardashir they stood only in a vassal relationship to him while under Shapur the Kushan kingdom and other areas were really included in the empire. This further implies wars by Shapur of which we have no evidence. The hegemony of Ardashir may have been light, based on a few victories over the allies of the Arsacids rather than actual conquest afterwards.

The Imperialism of Shapur

The Sasanian kings greatly favoured urbanism, a trait not in such evidence among their predecessors. The first two sovereigns of the house of Sasan were the greatest city founders of the line and most of the cities with royal names in them were founded or renamed by Ardashir or Shapur. The confusion of ancient native, Hellenistic, and Sasanian names given to cities frequently makes identifications of the cities difficult.

While their neighbours must have realised that the change of dynasties in Iran was not particularly to their interest, the Sasanians were soon to show the Romans and Kushans that a new Iranian nationalism and imperialism was a distinct peril to the peace. The Romans had won many victories in the last century of Parthian rule, so Ardashir was somewhat of a change while Shapur's conquests turned the balance of power in favour of Iran. We know much about his wars with the Romans because they were spectacular as well as victorious campaigns. His inscription of the Ka'bah of Zoroaster is both an important record and a paean of victory regarding his wars with Rome. Some scholars have accepted every word of his record of the struggle with the Romans as true, but have denied any other conquests of Shapur since they are not mentioned. Another inscription on the same structure, written by the order of an important religious figure Kartir, however, does tell of campaigns in Transcaucasia. We may also assume that victories in eastern Iran extended the empire to India, although we cannot exclude the possibility that some of these campaigns occurred towards the end of Ardashir's reign.

Shapur's inscription of KZ tells of three campaigns against the Romans, first at the beginning of his reign when Gordian marched against Shapur but was defeated and killed, whereupon Philip the Arab succeeded him as Roman emperor and made peace with Shapur. The second campaign resulted in the destruction of a Roman army of 60,000 men, after which the Persians ravaged Syria and Cappadocia, capturing Antioch on the Orontes as well as many other cities. In the third campaign the Emperor Valerian was defeated and captured after which Shapur again raided Syria and eastern Anatolia. Other sources tell us that in the third campaign King Odenath of Palmyra attacked and defeated the Persians, seizing much of their booty while they were on the homeward march. The first and third campaigns of Shapur can be dated in 243-244 and 259-260 and can be followed in literary sources. The second campaign presents problems in dating and identification because of the excavations at Dura-Europos. The evidence from Dura suggests that this Roman outpost on the Euphrates was captured by the Persians in 253 who held it for a few months and then again in 256 when the city was stormed and destroyed by Shapur. The question arises, which of the dates belong to Shapur's second campaign. Generallv speaking a 'campaign' in the Near East from ancient times has meant an expedition of one year. It is possible that the second expedition of Shapur lasted a number of years, including 253 and 256, as I have suggested elsewhere. More study has convinced me that this is unlikely and that 256 is the date of the second campaign while 253 was a minor raiding expedition not mentioned in KZ.

The capture of Valerian was an unparalleled event in history and Shapur made certain that the world knew about it through his inscription and rock reliefs at Bishapur and Naqsh-i Rustam. Although the interpretation of these reliefs is varied and disputed it may be true that they are a kind of counterpart in pictures of the inscription and the three Romans at Bishapur represent the three Roman emperors mentioned in the inscription, Gordian, Philip and Valerian. The prisoners captured by Shapur in his wars with the Romans included many technicians and from Antioch the bishop of the city who, with many of his flock, was settled in Khuzistan. The city of Gundeshapur ('the better Antioch of Shapur') was settled with Roman prisoners and the Caesar's dam at Shustar was one of their constructions. Prisoners were set led in Fars, Parthia, Khuzistan and elsewhere and they probably provided the basis of the later Christian communities in Iran.

The fortress town of Hatra which had repulsed the Romans on various occasions fell to Shapur, probably on his second campaign. One may also tentatively assign the ruin and abandonment of the towns of Hatra, Assur, Dura and other sites to the conquests of Shapur which thus must have changed the face of the Roman-Iranian frontier lands with the consequent end of certain trade routes and roads. The Romans contributed to this too when Aurelian conquered and destroyed Palmyra under Queen Zenobia in 272. Thereafter the Romans, and later the Byzantines, and the Sasanians maintained a system of border buffer states and limes between their two empires which were as often at war with each other as not.

Shapur was not only victorious against the Romans but also in the north in Transcaucasia and presumably in the east. According to KZ the Sasanian empire included 'Turan, Makuran, Paradan, India and the Kushanshahr right up to Pashkibur and up to Kash, Sogd and Shsh'. This passage has been discussed by several scholars, and I would interpret it to mean that first the land of Turan, probably in- cluding most of the province of Kalat in present-day Pakistan, was included in the empire. This Turan may well have some relation to the opposition of Iran and Turan in the national epic, especially when we know that many of the stories come from neighbouring Seistan. A further possibility, that the kingdom of Turan was created by invaders from Central Asia, cannot be dismissed. Next comes Makuan which is easily identified and then Paradan which presents a problem since we have no definite literary references to it and cannot locate it. I suggest that it may be located either in Arachosia or at the mouth of the Indus river rather than a small locality in Gedrosia. India or Hindustan is generally recognised as the Indus valley, but I suspect it is only the upper Indus here, north of present Sukkur into the Punjab. Exactly when this area submitted to the Sasanians is uncertain.

The Kushan empire at this time had already passed its prime and according to some numismatists may have split into two kingdoms, a Bactrian and an Indian kingdom, or even into more parts. It is tempting to think that the limits given in Shapur KZ refer only to the extent or boundaries of a northern Kushan kingdom, which submitted to Shapur after a defeat, since there is no evidence that the Sasanian armies actually reached the confines of the Peshawar region, Kashgar, Sogdiana, and Tashkent. It is not certain that Pashkibur is in fact modern Peshawar, but in any case a district or principality rather than the city is meant. The district either was possibly restricted to the Peshawar plain east of the present Khyber pass, or more likely comprised all of the lowlands which were the ancient Gandhara, including present Jalalabad. Kashgar surely means the kingdom which may have extended into Russian Turkestan north of the Oxus river, or we may have in the inscription the actual or the pretended extent of the Kushan empire up to the borders of the state of Kashgar which was more or less restricted to eastern Turkestan. I am inclined to favour this latter view since Sogdiana and Shash were probably states with their centres primarily and respectively in the Zarafshan and Ferghana valleys. In other words the boundaries of the Kushanshahr in theory, if not in practice, included the mountainous area of part of the Pamirs and present-day Tajikistan. The scanty archaeological and Chinese literary evidence would not contradict this view.

Thus in the north-east Ardashir and/or Shapur secured the sub- mission of the Kushan state. A good guess would put the first defeat and submission of the Kushans under Ardashir while the incorporation of the Kushanshahr in the Sasanian empire would date from Shapur's reign. In all probability the oasis state of Merv marked the military outpost of direct Sasanian rule under Shapur as it did later. In the eyes of the Persians what was beyond was no longer Iran but non-Iran. The archaeological evidence for the destruction of the city of Kapisa (hodie Begram) north of Kabul can be neither attributed nor denied to Shapur, but is probably earlier.

The extent of Shapur's hegemony in the east, on the whole, is now known from his inscription. From Shapur's inscription KZ we see that most of Transcaucasia was included in his empire, and from the inscription of Kartir at the same site we learn 'the land of Armenia, Georgia, Albania and Balasagan, up to the Gate of the Albanians, Shapur the king of kings with his horse(s) and men pillaged, bumed and devastated'. This indicates that Shapur did not inherit these lands from his father but had to conquer them, and for Kartir these are lands of non-Ian (Aniran) . Shapur re-created the Achaemenid empire and the Persians again ruled over non-Iranians. Yet Shapur was not the great innovator` or organiser that Darius was, since he continued for the most part in the path he had inherited, the legacy of the Parthians. A new feature, however, was the state church which will be discussed below.

The list of notables at the court of Shapur in KZ is both longer and more variegated than that of his father. From this and other inscriptions, the protocol and the social stratification of the Sasanian court are revealed. In the bilingual (Parthian and Middle Persian) inscription of Hajjiabad Shapur tells of an arrow he shot in the presence of the rulers (shahrdar, i.e. the kings of various countries in the Sasanian empire), the royal princes (BR BYT' or vispuhr), the great nobles (vazurkan) and the small nobles (azatan). In the Paikuli inscription of Narseh we find the expression, 'the Persian and Parthian royal princes, great and small nobles', which reveals the fusion of the Parthian and Persian nobility, perhaps similar to the Medes and Persians in the time of the Achaemenids. The court of Shapur, like that of Ardashir, does not show the developed forms of imperial bureaucracy characteristic of the later empire, for example the offices of the prime minister or chief of priests are not present. The functions of many of the listed posts are not known, but a number of considerations lead one to believe that the court differs little from the Arsacid court. A surprise is the presence of seven satraps, the latest appearance of this title, referring to the districts or provinces as well as the chief city which gave its name to the province. The satrapies depended directly on the king and the central government hence were located in western Iran and not on the frontiers. Subdivisions of provinces existed but apparently neither in a uniform system nor throughout the empire.

Although the Sasanians have been characterised as representing an `Iranian reaction to Hellenism, under Shapur we hhe Iast Greek used in inscriptions in Iran, and his patronage of Gr-e-esophers and savants has come down in Persian tradition. Likewise the mosaics of his new city Bishapur in Fars reveal a strong Western influence not to be attributed solely to artisans among the prisoners from Roman armies.2l One may suggest that under Shapur there is really a revival of Greek cultural influences in Iran which, however, hardly survives his death.

As the empire expanded so the bureaucracy also must have grown, but again the old traditions continued. We know from several sources that the royal seals were not personal seals, but were used by various officials of the king as had been true earlier. Just as in Seleucid times Sasanian official seals carried only legends or monograms but no figures. Representations of deities, personal portraits or animals were pictured on private seals. The official seals seem to have been important prerogatives of office, and later we find many seal impressions of mobads and other religious dignitaries as well as civil officials. Seals were used for all kinds of business and for religious affairs seals should be mentioned the insignia, coats-of-arms or emblems which were used by noble families as their signs of identification. Many of them were really stylised monograms or abbreviations, but Sasanian heraldry is a complicated subject which has been little studied. Insignia already existed in Parthian times and there is an interesting parallel between the signs or coats-of-arms on the headgear of Kushan notables on sculptures from Mathura, India and the signs on the helmets of the notables of Shapur's retinue pictured on the rocks of Naqsh-i Rajab near Persepolis. The proliferation of titles and honorifics in the course of Sasanian history was a tendency which lasted down to the twentieth century and the confusion of personal names, offices or titles, and honorifics was a problem for Byzantine writers in their day as it was for more contemporary foreign authors writing about Iran.

Social structure under the early Sasanians again most probably was an inheritance from Arsacid times. Divisions in society were normal in the Near East and by no means restricted to the caste-conscious Indians or the Zoroastrian Iranians. For example, Strabo speaks of four 'castes' among the Georgians: the rulers, priests, soldiers and the common people, and the importance of families where possessions were held in common. When the Zoroastrian church became firmly established in Iran it contributed to the fixing of social classes in accordance with religious tradition. As is well known society was later divided into four classes, the priests, warriors, scribes and common folk. The extended family has remained the basic unit of allegiance, trust and authority in Iran down to the present day, and while the centralisation of government in Sasanian Iran was a feature which distinguished it from Arsacid times, none the less the family remained paramount.

Shapur was known for his liberal spirit and in religion, if nowhere else, his liberalism apparently was in contrast to the policy of his successors. It is significant, I think, that the successor of Shapur, Hormizd Ardashir and another son, the future king of kings Narseh, are both mentioned prominently among those members of the royal family for whom special fires were instituted by Shapur; while another son Varahran, king of Gilan, does not have a fire instituted in his honor. The succession of Hormizd Ardashir seems to have been unopposed and under him the policy of Shapur was still in effect, but Hormizd did not rule long and he was succeeded by Varahran, known as Bahram in Islamic sources. A change in religious policy occurred which we shall discuss below and quite probably there were other changes too. Unfortunately our sources tell us little of this period of Sasanian history and Islamic authors give no hint of difficulties or important changes. Varahran was succeeded by his son of the same name, who after a reign of seventeen years was followed by his son, a third Varahran.

Then came a reaction and Narseh, son of Shapur and now surely advanced in years, revolted and seized the throne. Among other actions he had the name of a predecessor, Varahran I, chipped away from an inscription in Bishapur and his own name substituted for it. This, and his toleration of Manichaeism, in which he followed his father, indicate a change in the policy which had been followed by the Bahrams. Under his rule the Romans recouped their lost prestige and also some territory so that future relations were based on a kind of balance of power. The Sasanian empire was now more occupied with internal affairs than with external, and presumably a modus vivendi between the great feudal lords and the king of kings had been forged in such a way that a new allegiance to the house of Sasan was accepted by all.

Heresies and the Church

The development of the church during the early Sasanian empire is tied to the name of Kartir who was unknown to history before the discovery of his monolingual inscriptions in the Middle Persian language. One was carved below the Middle Persian verison of Shapur on the Ka'bah of Zoroaster, another on the cliff at Naqsh-i Rustam behind the horse of Shapur showing his triumph over the Roman emperor, a third at Nazsh-i Rajab and a fourth on a mountainside at Sar Mashhad south of Kazerun. At Naqsh-i Rajab accompanying the inscription is presumably the representation of Kartir himself with finger raised in a gesture of respect. At Sar Mashhad Bahram II is shown killing a lion while protecting his queen, and behind her is probably Kartir. The contents of these inscriptions are very much the same, except that Sar Mashhad and Naqsh-i Rustam are longer than the other two, while Naqsh-i Rajab is a kind of testament of personal belief. Unfortunately both the Naqsh-i Rustam and the Sar Mashhad inscriptions are badly weathered and large portions illegible. None the less, the story they tell of Kartir reveals a fascinating page of early Sasanian history, the establishment of orthodoxy and a state church.

Before turning to Kartir, an examination of Islamic and Pahlavi sources reveals that chief religious leader or mobadan mobad of Ardashir was a certain Tansar, whose name probably should be read Tosar. He is also called a herbad or "teaching priest" in some sources. There is no indication that Tosar is to be identified with Kartir, but his activities, including making a new recension of the Avesta according to the Denkart would make a veritable Kartir of him. The inscriptions, however, are more reliable than literary sources and they tell only of Kartir, although a person called Tosar may have been active under Ardashir before Kartir came to the fore. Kartir must be the real founder of Zoroastrian orthodoxy under the early Sasanian kings.

The longest inscriptions of Kartir are the eighty-one lines of Naqsh-i Rustam and the almost identical fifty-nine lines of Sar Mashhad, the first twenty-five lines of which latter inscription are the same as the inscription of Kartir Ka'bah of Zoroaster, while lines 52 to the end are almost a verbatim copy of his inscription at Naqsh-i Rajab. In the central part of Sar Mashhad Kartir *Han- girpe (hnglpy), as he calls himself, gives what almost seems to be an apologia pro vita sua. The early fragmentary passages contain interesting theological points, the interpretation of which is very difficult because of lacunae. Afterwards Kartir becomes more personal, but in the third person, telling of a trip of many nobles to Khurasan about a woman together with the man Kartir *Hangirpe, a place (?) called pwlsy and many other enigmatic details. Kartir goes to great pains to tell posterity that he first came to power under Shapur-when he was a herbad and a mobad, which implies at least the existence of different kinds of priests already in the Zoroastrian religion. Under King Hormizd he was given the title 'mobad of Ahura Mazda', probably the first to hold this later well-attested title. In the reign of Varahran II he received the rank of nobility, the headship of the religion, and was made chief judge of the empire, and chief of the royal fire at Istakhr at the imperial shrine of Anahita. The reason for these great honours is implied in the honorific given by the same king to Kartir, 'soul-saviour of Varahran'. Undoubtedly Kartir played the role of father confessor to the king and was th,ereby rewarded. The fact that he is called 'the lord' at the very end of Naqsh-i Rajab and that he notes his elevation to the nobility further suggests that the nobility were all powerful in this period. Kartir probably played an important political as well as re- ligious role in the empire.

Of great importance was the activity of Kartir outside of Iran in trying to establish both fire temples and orthodoxy among the Hellen- ised Magians and to convert those pagans who followed rites and beliefs similar to those of the Zoroastrians; in other words Kartir was a missionary.

At the same time he reacted strongly against both foreign religions and heresies within Iran, and this may well be one reason why Mithraism as we know it in the Roman Empire is not also found in Iran. Kartir (KZ 9-IO) specifically attacked Jews, Buddhists, Hindus Nazoraeans (Mandaeans?), Christians, Mktk (a Mesopotamian religion?), and Manichaeans, destroying their centres and proscribing them. The work of Kartir apparently was not an innovation, smce Armenian and Syriac sources tell of the zeal of Ardashir in establishing fire temples and destroying pagan temples, especially in Armenia. Kartir's action was militant Zoroastrian orthodoxy in Zoroastrianism, for he Magi were organised, hel esy was forbidden, and many Varah nres were insituted. These fires represented the backbone of the Sasanian fire cult for they were centres of teaching as well as rites in the various geographical areas of the land. The work of Kartir was impressive (KZ, line I4) for we see in effect the ordering of the state church in Iran, including the practice of consanguineous marriages, a feature of Zoroastrianism which adversely struck outside observers. He also laid the basis for the power of the clergy which was to rival, if not later surpass, that of the nobility.

The fanaticism of the period of Varahran II was tempered in the reign of Narseh (293-302) who revolted against the young King Varahran III, who is called the Saka king in Paikuli, and seized power in northern Iran. He marched on Ctesiphon and was met by a party at Paikuli, a site north of present Khaniqin, and there he was proclaimed king of kings, and a bilingual inscription was erected to commemorate this event. In line 16 of Paikuli the name 'Kartir, the mobad of Ahura Mazda' appears, but because of lacunae in the in- scription one cannot say whether he is a foe or friend of Narseh. He was surely quite elderly and must have died or retired shortly after- wards. Since Narseh did not mutilate Kartir's inscriptions, and there is no evidence of a clash between the two, we may assume that Narseh, who mentions in his inscription (Paikuli, line g) 'Ahura Mazda and all the gods and Anahita called the lady', did not overthrow the work of Kartir. The policy of toleration of Narseh towards the Manichaeans is generally known, but it is possible that a change began at the very end of the reign of Varahran II. The evidence of a complete about-face in religious policy under Narseh and a victory of herbads over mobads or Anahita over Ahura Mazda, is lacking; rather the change seems to be one of relaxation yet continuity.

The question of heresies within the Zoroastrian religion is complicated because our Pahlavi sources are all post-Islamic in date, when the minority religious comrnunities of the Zoroastrians were more concerned with correct beliefs than in Sasanian times when the religion was upheld by the state. I believe that orthopraxy was more important than orthodoxy under the Sasanians and Zurvanism, or time speculation, was not a heresy in the same manner as Mazdakism, which was a threat to the practices and the organisation of society as well as the church. But in the early days of the empire the Zandiks, as the Manichaeans were called, were the chief heretics. The exact dates of Mani's life are uncertain since they are tied to the chronology of Shapur's accession which itself is not certain; but he was killed either in the last year of the reign of Varahran I (274) or in the early years of his successor ( 277 ) . Manichaeism has been called an expression of universalism or syncretism in religion and it has been compared with Bahaism of the present day. It is perhaps not as representative of Iranian religious tendencies in its dualism as was the Zoroastrian state religion, but certainly the syncretic and 'international' features of Manichaeism found many ready supporters in Iran. We are not here concerned with the teachings of Manichaeism which are at present better known than before the discovery of original Manichaean writings in Coptic, Parthian, Sogdian and other languages. The Manichaeans suffered the same fate in Iran as in the Christian world; in both the arch-heretics were alwas Manichaeans and they were accordingly persecuted severely. After Narseh, however, Manichaean cornmunities continued to exist in Iran, especially in eastern Iran, and later, as is well known, Manichaean missionaries reached as far as China.

Perhaps the most striking development of Manichaeism was the social and economic movement led by Mazdak at the very end of the fifth century, about whom much has been written of late. It would seem that royal opposition to the nobility and their power was an important reason for the support of Mazdak by King Kavad. The Mazdakites preached a form of communism, the division of wealth including wives and concubines, which found support among the poor, but our sources are not clear and are contradictory about the course of events of this revolution. The Mazdakites, however, met the same fate the Manichaeans had suffered at the hands of Kartir and King Varahran. It happened at the end of Kavad's (second) reign, and the Crown Prince Chosroes Anosharvan was the chief instigator of the massacre of the Mazdakites circa 528. The death of their leaders, of course, did not end the Mazdakites as a sect but sent them underground. But a new pejorative had been coined and henceforth any social or religious reformer was usually branded as a Mazdakite by his opponents, and this lasted long into Islamic times when many Iranian revolts against the caliphate or the rule of the Arabs were designated as Mazdakite movements. The Mazdakite movement was known to such Islamic authors as Nizam al-Mulk in his Siyasatname.

Already, from the beginning of the Sasanian period, we are in a new religious world. The cults of the old Mesopotamian gods were long since dead and in their places new gnostic and ritualistic sects had arisen side by side with Christianity, Judaism and Zoroastrianism. Cabalistic beliefs and practices seem to have been widespread, and in the views of most Greek and Roman authors the Persians were the chief believers in magic and unusual religious practices. Zoroastrianism for the classical writers was the epitome of the mysterious, Oriental cult. Yet Kartir and his followers laid the basis for Zoroastrian ortho- doxy which probably opposed magic, demon worship, and the like as much as did Christian orthodoxy in the empire of the Caesars.

Belief in divine revelation and the recording of that revelation in books was in the air, and the Christians, of course, were the most widespread propagators of the idea of 'Holy Writ'. It may have been because of the example of the Christians that the Zoroastrian church assembled and canonised its writings. Zoroastrian tradition claims that fragments of the Avesta were assembled and presumably written down in Arsacid times and again under Shapur I. The written Avesta of the early Sasanians must have been really a mnemonic device to aid the memory of the priests who usually recited the Avesta in a traditional Oriental manner. In the beginning of the fifth century the present Armenian alphabet was devised mainly to propagate the Christiar religion in that land. Some have conjectured that the present Avestan alphabet was invented about the same time possibly as a forerunner or even as an imitation of the Armenian alphabet although the Avestan alphabet in phonetic completeness is more like the Devanagari alphabet of Sanskrit. It is not impossible to assume a religious motivation for the creation of this rather late alphabet which, as far as we know, was only used for texts of the Zoroastrian religion. It is a pity that this alphabet did not replace the incomplete Pahlavi alphabet, with its great deficiency of letters to represent sounds, for the Middle Persian language. It must be emphasized that we have no old manuscripts of the Avesta, none earlier than thirteenth or fourteenth century, but the existence of a written Avesta in Sasanian times much as we know it today seems assured in spite of the overwhelming importance of the oral tradition.

From Christian authors writing in Syriac and Armenian it would seem that the Sasanians primarily followed Zurvanism, a Zoroastrian heresy which, after the Islamic conquest, vanished in favour of ortho- doxy. I believe, as shown elsewhere, that Zurvanism was not a full- fledged heresy with doctrines, rites and a 'church' organisation separate from the Zoroastrian fold, but rather a movement to be compared perhaps with the Mu'tazilites of Islamic times. There were basically two features of Zurvanism which have been preserved for us, time speculation (eternity, etc.) and the myth of the birth of both Ohrmizd (Ahura Mazda) and Ahriman from their father Zurvan. The first was widespread and certainly by itself would not form the basis for a separate sect. The Zurvan birth story can be paralleled by the story of Chronos in Greek mythology and again, in my opinion, would not lead to the formation of a sect. Undoubtedly the Zurvan birth story was widespread among 'orthodo' Zoroastrians in Sasanian times, but after the Islamic conquest when Zoroastrians withdrew into tightly knit communities, Zurvanite elements were eliminated from the new orthodoxy which was concrned with 'orthodoxy' as well as 'orthopraxy'. In Sasanian times a Zoroastrian heretic was more one who broke away from orthopraxy and even became a Christian or Manichaean, while in Islamic times a Zoroastrian heretic was primarily a person who also broke with orthodoxy as, for example, Abalish (or 'Abdallah?) a Zoroastrian who became a heretic in the time of the caliph al -Ma'mun in the ninth century, and may have adopted Mani- chaean beliefs. Any kind of social heresy, of course, would be the concern of the ruling caliph.

From the acts of the Christian martyrs we learn much especially about the Nestorian communities in the Sasanian empire. In effect the consolidation and growth of the Zoroastrian church in Iran was paralleled by the growth of the Christian church and of the Manichaean communities. Undoubtedly the influx of Christian prisoners in Iran in the wake of both Shapurs' conquests gave a strong impetus to the spread of Christianity, but the religion naturally spread in Mesopotamia among the Semitic peoples. The first great persecution of Christians occurred under Shapur II, beginning about 339, and seems to have had political motivation since it began after Constantine had made Christianity the religion of the Roman Empire. Later there were periods of tolerance followed by more persecution, but after the break away of the Nestorians from other Christians at the end of the fifth century, the condition of Christians in Iran improved. The Nestorians elected a catholicos who had his seat in Ctesiphon and synods usually met there in deciding church problems. The ecclesiastical geography of the Nestorian bishoprics is also of importance for the civil geography of the Sasanian empire since the Church usually followed civil boundaries; thus we gain some knowledge of civil administrative divisions from the acts of the martyrs.

The Christianisation of Armenia and Transcaucasia in the fourth century provided a source of conflict betwen Armenia and the Sasanians even more than the struggle for influence in those areas between Romans and Persians. In the east, too, Christian missionaries made converts among the Hephthalites and Sogdians, so one may infer everywhere a growing Christian influence at the end of the Sasanian empire. The whole religious picture of Iran, however, was more complicated than we can know from the sparse records, and the interplay of various religions is matched by internal divisions within the Zoroastrian church which we perceive but dimly.

The Glory that was Iran

If one asks an ordinary Persian who had built an unknown, ancient, ruined mosque or other structure in some locality, the chances are great that he would reply it was Shah 'Abbas, the Safavid ruler who embellished with edifices the city of Isfahan. If the ruins were clearly pre-Islamic the reply might be Khusro or Chosroes Anosharvan 'of the immortal soul', the Sasanian counterpart of Shah 'Abbas. His very name became, like that of Caesar, the designation of the Sasanian kings for the Arabs (Kisra in Arabic) and almost a synonym for splendour and glory. But Chosroes ruled Iran less than a century before the Arab conquest and, as is not uncommon in history, the seeds of decay already existed in the period of greatest splendour in the Sasanian Empire.

Iran had not fared well in her external relations under the successors of Shapur I; under Varahran II the Romans regained lost territory in northern Mesopotamia as well as hegemony over Armenia. Narseh fared no better and further concessions had to be made to the Emperor Galerius. After him it seemed as though the Romans had regained the dominant position which they had held in Parthian times. Under Shapur II, who had an unusually long rule of seventy years, the Sasanians passed to the offensive both in the west and in the east where the Kushan state and other territories probably had proclaimed their independence during the minority of Shapur. On the whole Shapur II was successful in regaining both territory and lost prestige for the Persians. He followed the practice of Shapur I in settling Roman prisoners in various provinces of his empire, according to Ammianus who is a valuable source for the history of Shapur II and his wars with the Romans.

After Shapur his weak successors lost much of their imperial authority to the nobility which grew in strength and influence. Although there may be no causal connection it is interesting to note that as royal power declined in favour of the feudal lords, the heroic, or epic tales regarding the reigns of such kings as Varahran V or Bahram Gor (42I-439) the hunter of wild asses, increased or came to the fore. One may suspect that titles and offices increased in number and importance during the long period of weak monarchs. Concomitant with the new power of the nobility were struggles over the succession by opposing parties of the feudal lords. Such was the case with the crowning of Varahran V (in 421) and of Peroz (459).

In the fifth century a forrnidable new enemy appeared in the north-east as successor to the Kushans, a new wave of invaders from Central Asia called the Hephthalites. They are connected with the new order on the steppes of Central Asia which can be characterised best as the rise of the Altaic-speaking peoples or the Hunnic movement. Just as the first millennium BC in Central Asia was considered by classical authors as the period of Scythian dominance in the steppes, so the first half of the first millennium AD is the time of the Huns, while the second half and later is the period of the Turks and the Mongols. Of course the term 'Scythian' continued to be used by classical authors for various steppe peoples well into the Christian era just as the Ottomans were designated 'Huns' by several Byzantine authors. None the less the various terms 'Scythian, Hun and Turk' were general designations of the steppe peoples in Western sources including the Near East, though the Chinese had other names. Obviously not all peoples who lived in, or came from Central Asia into the Near East or eastern Europe in the first half of the first millennium AD were Huns, and the fact that Western and Near Eastern sources call a tribe Hunnic really only means that they came from the steppes of Central Asia, a vast area. The word 'Hun' has caused scholars great trouble as have other problems of Hunnic history, but this is not the place to discuss such questions as, for example, the iclentity of the Hsiung-nu of Chinese sources with various 'Huns' of Western, Near Eastern or Indian sources.

Although presumably the name of the Huns appears as early as the geography of Ptolemy, applied to a tribe in South Russia, we cannot find any other evidence for 'Huns' in the Near East or South Russia before the fourth century AD. The joining of the word 'Hun' to the Kidarites by Priskos is probably an example of the use of the general fifth-century term for an earlier history and no proof that the Kidarites were Altaic-speaking people. Presumably Kidara was the name of a ruler since the name appears on coins, but there is no evidence that he led a new Central Asian horde to conquer the Kushan realm. Several attempts to date a ruler Kidara have not been convincing and we may only hazard a guess that such a reign was in the fourth century.

Another name from eastern Iran or Central Asia seems to indicate a migration or invasion from the North. The newcomers are called Chionites in classical sources. In 359 the king of the Chionites, Grumbates, is mentioned by Ammianus as an ally with Shapur II and his army before the walls of Amida. It is generally believed that the Chionites, with the form OIONO=Hyon=Hun on their coins, were Central Asian invaders of eastern Iran connected with the Hunas of Indian sources and with their successors the Hephthalites. Unfortunately we have no sources for the history of eastern Iran in this period and the many and varied coins have not been properly classified, an extremely difficult task.

From the coins of certain Sasanian Kushan rulers one would conclude that the Persians were at least liege lords of part of the Kushan domains throughout most of Shapur II's rule. Some time, probably at the end of the fourth or early fifth century, a new ruler Kidara appears as an independent southern Kushan ruler. The Chionites probably moved into the northern Kushan domains (north of the Oxus river) some years before Kidara whose power seems to have been based mainly in lands south of the Hindu Kush since he has coins with Brahmi legends. This division between lands north and south of the mountains is important. The Chionites probably expanded over Kushan domains and independent rulers of them appeared in Bamiyan, Zabul and elsewhere, the coins of which are very difficult to classify. The confusion in our sources between Kidarites, Chionites and Hephthalites may well reflect a real mixture of peoples and rulers. One may say, however, that the name of the Chionites is followed by that of the Hephthalites in history.

It is difficult to determine the ethnic composition of Chionites or Hephthalites, but there is no evidence that the Chionites were different from the Hephthalites; rather the meagre evidence indicates that the Hephthalites may have stood in the same relation to the Chionites as the older Kushans did to the Yueh-chih. In other words, the Hep thalites may have been a prominent tribe or clan of the Chionites. One may well expect Altaic, i.e. Hunnic, elements among the Hephthalites, to use the later name, but again the evidence points primarily to Iranians. It is possible that some of the early rulers were Huns, but there were still many Iranians in Central Asia, and the people of eastern Iran among whom the Hephthalites settled were also Iranian, so we may consider the Hephthalite empire in eastern Iran and north-west India as basically an Iranian one. Zoroastrian as well as Manichaean missions in Central Asia must have increased the West Iranian cultural elements among the people. Undoubtedly by the time of the Arab conquests, however, the Turkic elements among the Hephthalites had increased, but that was after the Turks themselves had appeared in the Near East. It is, of course, possible to construct theories of history and of ethnic relationship on the basis of suggested etymologies of one or two words, but the lack not only of sources but of reliable traditions in the fragmentary information about Central Asia and eastern Iran in classical sources makes any theory highly speculative.

The Persians in the last half of the fifth century suffered a series of defeats at the hands of the Hephthalites and King Peroz lost his life in 484 in battle with them. After him the nobles waxed even stronger, placing several rulers on the throne in succession and finally Kavad I, who then maintained his throne only with Hephthalite aid. This was a period of low ebb for the Sasanians when their eastern neighbours exercised influence even in internal affairs. The Mazdakite revolution already has been mentioned, but the great change or revolution in Iran came with Chosroes I who, as we have said, was the greatest pre-Islamic ruler in the minds of the Persians.

The far reaching reform of taxation under Chosroes has been discussed by several scholars, notably F. Altheim, whose merit was to show repeatedly that the model for the new system of taxation was the system in force in the eastern Roman Empire which in turn had been built on the reforms of Diocletian. The unrest and social changes of the Mazdakite period made a new assessment of property and of taxes necessary, but we cannot say with certainty what the situation was before Chosroes. What is reported by later authors of Sasanian times refers to the post-Chosroes period. We may assume that Chosroes wanted stability, and in terms of taxation, of course principally on the land, a fixed sum rather than a yearly variation according to the yield, which seems to have been the old system. A survey of the land was made including a census and a counting of date palms and olive trees. The land tax of the later Roman empire was based on the land unit the iugum, but the amount of taxation was already determined by the indictio and divided among the various plots of land. This became the system of Sasanian Iran, of course with many different details into which we cannot go. The Sasanian head tax, like the Roman capitatio, was under Chosroes assessed in a number of fixed categories according to the productive capacity of a man. In both empires state employees were exempt from paying the head tax, and in Iran the Magi, soldiers and the high nobility were exempt as well. Certain details of the taxation are disputed but the main lines are clear; Chosroes sought stability and a fixed income for government coffers.

From the Talmud we learn that ancient practices in regard to the payment of taxes still continued under Chosroes. If one could not pay his land tax and another paid it, the latter received the land. By paying the land tax of someone who could not pay, one could obtain the debtor as a bondsman or slave. According to one source, if a Jew declared he was a Zoroastrian he could escape the head tax. This was rather a special tax, or a heavier head tax, placed on Jews, Christians and other minorities. The bishop for the Christians and the head of the Jews for the Jewish communities collected taxes from their followers. This continuity of tax practices in Iran continued into Islamic times. The Sasanian system provided the background of the well-known but also in part different and complicated system of the Islamic kharaj and jizya.

In addition to a tax and financial reform, there was a social and bureaucratic revolution, but again many details escape us or are subject to various interpretations. Certain innovations may be the work of Chosroes' predecessors, but one may say that after him they appear as a characteristic feature of Sasanian Iran. The most important was perhaps the growth of the lower nobility or the dihqans (literally village lord) as the Arab conquerors called that backbone of Persian provincial and local administration. This lower nobility really possessed and ruled the land at the end of the Sasanian empire and it would seem that they owed their positions to the ruler and were an effective counter-weight to the few great families who became progressively less important. In line with his policy of stability Chosroes may have sought religious support for a social stratification of four classes or castes, which, however, may have developed throughout earlier Iranian history so that by the time of Chosroes it was full-fledged.

There is considerable material in Islamic works, such as the Kitab al-Taj of Jahiz, and countless anecdotes and stories which refer to the activities of Chosroes I. The sources agree in their assessment of the empire of the Sasanians after Chosroes as a tightly organised structure with the king supreme at the top of the hierarchy. The 'mystique' of the king of kings was reinforced, and books of protocol, mirrors of princes and other writings, laid down the duties of monarchs to their subjects and subjects to their ruler. It would seem that there was a considerable activity in fixing rules of behaviour, prerogatives and obliga- tions for various classes of society in this period. The ofices of mobadan-mobad or chief of the clergy, dabiran-debir, or chief of the scribes, and similar titles, in imitation of the king of kings, indicate the ordering of society by imperial and religious sanction. The fascinating picture of society under the later Sasanians is one of a people who have seemingly reached a social and religious stability in religion, class structure and general culture but continuing with the seeds of decay in the resultant stagnation.

The age of Chosroes was one of conquest too. Antioch was briefly captured in 540 and in the east the Hephthalite power was crushed by a joint Persian and Turkish attack circa 558 when the Western Turkish khanate and the Sasanians ended a united Hephthalite rule replacing it with at least nominal Turkish hegemony north of the Oxus river and Sasanian overlordship over many of the Hephthalite principalities south of the Oxus. Chosroes, as Shapur I and II, was known also for his systematic transport and settlement of prisoners of war in various parts of Iran, an age old custom followed in Iran by Shah 'Abbas and Reza Shah in more recent times. It was under Chosroes that the unusual but not really important Sasanian conquest of Yemen took place which had echoes in the Quran. Under Chosroes we find the frontiers of the empire secured by a system of limes in the Syrian desert, in the Caucasus by Derbend and east of the Caspian Sea in the steppes of Gurgan. The institution of a system of four spahbads or generals of the realm in north, south, east and west is also attributed to Chosroes, and one hears more of the importance of marzbans or 'wardens of the marches' in this later period of Sasanian history. The city building activity of Chosroes already has been mentioned. One town he built with the aid of Byzantine prisoners was the better Antioch of Chosroes near Ctesiphon, with a name similar to the better Antioch of Shapur I of Gundeshapur. The seal of Chosroes was a wild boar which symbol was very widespread in Sasanian art. The reorganisation of the bureaucracy by means of a system of divans or ministries by Chosroes is generally regarded as the prototype of the 'Abbasid divans by many Islamic authors and while proof of direct continuity is sometimes difficult to establish beyond doubt, there were many influences.

There is so much written about Chosroes that one may omit a discussion here and refer to various writings about him. The internal reforms of the king of kings were more important than external changes in the frontiers, and their overall result was a decline in the power of the great nobility and the subkings in favour of the bureaucracy. The army too was reorganised and tied to the central authority more than to the local officers and lords. While one could continue with a long list of reforms attributed to Chosroes, some of the lesser known developments in that period of Sasanian history might be of interest.

It is well known that names which we find in the national epic appear at the end of the fifth and the beginning of the sixth century arnong the royal family and presumably also among the nobility although we hear little about the latter. The old title of kavi in its Middle Persian form kay, written kdy, appears on coins of Peroz and Kavad, another indication of an antiquarian revival. It is highly probable that the lays and legends of ancient Iran were gathered together in the days of Chosroes I and that the national epic as we know it in Firdosi was much the same then as now. Whether there was any great remaking of the epic, such as weaving events of Chosroes' life into those of Kai Khusro, cannot be proved but it is not impossible. Some scholars would even attribute the introduction of the highest offices of the empire, such as mobadan-mobad, first to the reign of Chosroes, but the wholesale assignment of innovations to him is probably an exaggeration. Likewise the contention that Chosroes founded a new hierarchy of fire temples with the introduction of a Gushnasp fire, tied with the crowning of the king in Shiz or Ganzak, is possible but unproved.43

Chosroes' name is also connected with a revival of learning with both Greek and Indian influences coming into Persian intellectual activities. Agathias has a well-known passage about the Greek philosophers (presumably neo-Platonists) who came to the Persian court after the closing of their academy in Athens in 529 and who were well received. The question of the extent of Sasanian learning is unsettled in its details, some scholars attributing a Persian origin to much of later Islamic science and learning, others denying the existence of a large Pahlavi scientific literature. We know of Burzoe, the famous physician of Chosroes, who reputedly was sent to India by the king and brought back the game of chess plus many Sanskrit books such as the fables of Bidpay and works on medicine which he translated into Pahlavi. Other Persian authors are known only by later references. Many Arabic and New Persian works on astronomy such as star tables (especially the Zt-j-i Shahriya-r) betray Sasanian prototypes, and one may suspect that much Pahlavi profane literature was lost because the mobads were not concemed to preserve it, while men of learning were content to use Arabic rather than the difficult Pahlavi form of writing for their works of science and literature.

On the other hand it is virtually certain that various Greek scientific works were translated into Pahlavi and then later from Pahlavi to Arabic, an indication of the existence of scholarly activity in Pahlavi. This learning, however, would seem to be more compilation than original, and the literary renaissance in the time of Chosroes also was primarily concerned with writing down and fixing various stories and legends including the national epic. The letter of Tosar, which has been mentioned, the Kar Namak of Ardashir and other tracts of Pahlavi literature have been attributed to this period. Some scholars have maintained also that the Avestan alphabet was created under Chosroes rather than earlier. The changes and additions which must have occurred in both epic and religious literature make datings difficult, but the great activity under the rule of Chosroes cannot be denied.

Sasanian art can be characterised as the culmination of a millennium of development. For one may discern Greek and Roman elements, ancient Oriental archaising motifs and purely Iranian subjects, such as the investiture of the king on horseback, in later Sasanian art. The brief Greek revival under Shapur I hardly interrupts the development of Iranian art from the Parthian period and Ardashir down to Chosroes. Just as in late Gandharan art so in Sasanian art stucco and plaster are supreme as the medium of expression. The widespread use of monograms, symbols and complicated designs is typical of late Sasanian art and as such is a forerunner of Islamic art . The more naturalistic emphasis in earlier Sasanian art seems to give ground before more stylised and even geometric art at the end of the period. The anthropomorphic representation of the god Ahura Mazda, perhaps a residue from the 'messianic period' of the religions of the Near East, is not attested at the end of the empire. Although ancient motifs of the hunt, investiture of the king or battles on horseback, appear on rock carvings or on the wonderful silver platters, they are all distinctive and could not be mistaken for anything other than Sasanian. The Sasanian hallmark or 'stamp' may be considered another evidence of the freezing of culture and society. What has remained of the architecture, sculpture, metalwork, ceramics and silks of the Sasanian period, however, is enough to testify to the grandeur and richness of Iranian culture.

The Sasanian empire seemed stronger than ever after Chosroes but in spite of his changes and reforms the age was not one of innovation. Rather the period in a truer perspective might be characterised as a summation of the past, of gathering-in and recording, when history becomes important as a justification for the state and the religion. The past which was revived in epic, in traditions and in customs, however, was a heroic past of great and noble families and of feudal mores, not of a centralised, bureaucratic state which Chosroes wanted to establish. Were the successors of Chosroes somewhat like Don Quixote while the people were ready for the new message of the followers of Muhammad? The noble families kept alive the heroic traditions of Iran and they survived the Islamic onslaught while the empire went down in ashes. Local self-interest and fierce individualism have been both the bane and the glory of Iran throughout its history, but through triumph and defeat the culture and the way of life of the Persians have unified the population of the country more than political or even religious forms unless they too were integrated into the heritage of Persia.