WOMEN

i. IN PRE-ISLAMIC PERSIA

Few primary literary texts document the activities of royal and non-royal women in pre-Islamic Persia. To learn about these, we are largely dependent on the often hostile secondary sources of the Greek and Roman periods which, however, are of limited historical value, as they tend to focus on particular aspects of the lives of royal Persian women or use specific descriptions for historiographical purposes. Likewise, historical texts of the early Islamic period have to be used with caution. In contrast, the archeological material available to us underscores the fact that depictions of women were available on a variety of media throughout the pre-Islamic empires, highlighted in the importance of the representation of royal women at the Sasanian court.

i. Achaemenid period.
(1) Titles for royal women.
(2) Activities at court.
(3) Marriage alliances.
(4) Female workers.
(5) Representation of women.

ii. Parthian period.
(1) Introduction.
(2) Titles for Parthian royal women.
(3) Activities at court.
(4) Representations of Parthian women.

iii. Sasanian period.
(1) Introduction.
(2) Titles for royal women.
(3) Marriage alliances.
(4) Activities at court.
(5) Representations of Sasanian women.

i. ACHAEMENID PERIOD

(1) Titles for royal women. The most invaluable source for the discussion of women in the Achaemenid period are the Persepolis Fortification Tablets (PFT). They shed light on the titles for royal women, as well as on their economic position. The PFT also offer unique insight into the position of female laborers who were recipients of food rations at Persepolis. While other Near Eastern sources can complement some of the information provided in the PFT, it is predominantly the Greek historical sources and, occasionally, biblical references, which allow a discussion of the position of women at the Achaemenid court.

The PFT attest to the fact that the Elamite title dukšiš (princess) was used collectively for Achaemenid royal women. Their individual status was determined by their relation to the king, and accordingly the women were referred to as “the king’s mother” (Elam. *sunki ammari), “the king’s wife” (Elam.*sunki irtiri), and “the king’s daughter” (Elam. sunki pakri). These terms of reference follow Assyrian and Babylonian usages, attested in the terms ummi šarri, aššat šarri, and mārat šarri. Babylonian sources dated to the Achaemenid period also referred to a woman belonging to the royal household as “woman of the palace” (Bab. ša ekalli; BE 9 28.50).

Greek sources frequently mention the existence of royal concubines (Gr. pallakai) at the Persian court, especially emphasizing their considerable number: 300 according to Heracleides of Kyme (FGrH 689 F1), and 360 according to Dinon (FGrH 690 F27). Yet the status of these women was in no way compatible to that of Greek concubines. The royal concubines, or women of the king, often were of high rank, but due to their non-Persian descent they were excluded from marrying the king, and their offspring—in principle—could not succeed to the throne. Other women entered the king’s palace as captives:

“The 14th [year] (345 BCE) of Umasu who is called Artaxerxes (Artaxerxes III): In the month Tishri the prisoners which the king took [from] Sidon [were brought] to Babylon and Susa. … On the 16th day the […] women, prisoners from Sidon, which the king sent to Babylon, on that day they entered the palace of the king.” (ABC, Chronicle 9).

(2) Activities at court. While the accommodation of the women of the court in a designated area within the palace complex is likely, no written or archeological evidence allows the suggestion that they lived in the seclusion of a harem. On the contrary, royal women were among the few people at the court who had direct access to the king and joined him at breakfast and dinner (Heracleides of Kyme FGrH 689 F2 apud Athenaios, 4.145c; cf. Xen., Cyr. 1.3.4; Plut., Art. 5.5;). They enjoyed economic independence and were able to travel in their own right, as well as in the king’s entourage, accompanying the king on campaigns and possibly on hunts (Heracleides of Kyme FGrH 689 F1). Together with the royal concubines, the wives of the Persian nobles, and their female attendants, they took a designated position within the royal train and were provided with their own personnel and royal guards. Presumably the women’s tents were also set up in an appropriate space during encampment. Information about women’s participation in royal banquets is contradictory. While Herodotus (5th century BCE) relates the story of Persian envoys at the Macedonian court, demanding the presence of women during a banquet, claiming it to be a custom in their own country (5.18.2), the Book of Esther (1:9-12; cf. Dan. 5:2,10) implies that the king’s wife and the women of the court celebrated a feast separate from the king’s banquet.

One of the most intriguing issues revealed in the archeological sources is the fact that royal women are depicted holding audiences which are reminiscent of the king’s audience scene from the apadāna reliefs. As is known from the Book of Nehemiah (2.6), the king’s wife could be present at royal audiences, but the archeological evidence reveals that they also held their own council. This may be a custom adopted from the Elamite court, as is suggested in a seal impression on some PFT. The seal PFS 77* (Garrison and Root, in press) is carved in Neo-Elamite style and shows a woman enthroned, accompanied by a female servant standing behind her and receiving a female visitor, who is separated from the enthroned woman by an incense burner. A similar scene is depicted on a cylinder seal carved in Achaemenid court style (FIGURE 1; cf. Spycket, fig. 7; Brosius, 2005), and on an unpublished seal in the Metropolitan Museum of Art, New York (acc. no. C16496). Further examples can be found on funerary stelae from Asia Minor dated to the Achaemenid period (Bakir; Brosius, 2005). The representation of women in audience scenes could only have found artistic expression if high-ranking women indeed held audiences themselves.

Royal women owned land and estates in Persis (Fārs) as well as outside the Persian heartland, e.g., in Babylonia, Syria, Egypt, and Media. They employed their own workforce and it also appears that certain administrative officials were assigned to them (e.g., BE 9 28, 9 50; Hdt., 2.98.1; Xen., An. 1.4.9; 2.4.27; Plato, Alc. I 121C-123CD; Plut., Art. 19.10). Most notable is Irdabama, a royal woman of the court of Darius I (r. 522-486) and unknown to Greek sources. She possessed her own workforces, mainly centered on Tirrazziš (Shiraz), which could include up to 480 laborers (PF 1028; PF-NN 1068, 1146). A small workforce known as matištukkašp also was assigned to her. With her own seal (PFS 51) Irdabama authorized the transactions of foodstuffs, while officials, such as Uštana and Rašda, using seal PFS 36 and 78 respectively, carried out her orders (Brosius, 1996, pp. 135-44). Estates are attested as belonging to Irtašduna, identified as Artystone, wife of Darius I. She also used her own seal (PFS 38) to authorize transactions and ration payments for her workforce.

Greek historiography predominantly depicts Persian royal women as intriguing and interfering individuals at the Persian court, and they thus appear as instigators of rebellion and upheaval. However, few incidences attest to the fact that the women’s rank and close proximity to the king allowed them to intervene between the king and members of the royal family and the nobility. They were thus able to lessen a punishment or ask for the reinstatement of a disgraced family member, as in the case of Amestris and Amytis saving the life of Megabyzus (Ctesias FGrH 688 F 13). Overall their actions seem to be determined by their ambition to protect the royal family and consequently avenge the death of a family member or endeavor to save his life.

Apparently following Babylonian practices, an official mourning period was declared at the announcement of the death of a female member of the royal family, and the burial rites were carried out in accordance with her royal status. Like the body of the king, that of the king’s wife had to be returned to Persis (Ctesias FGrH 688 F 14.15; cf. Hdt., 2.1.1; ABC, 7, col. III: 23-24).

(3) Marriage Alliances. Following Near Eastern practices, female relatives of the king were used in political marriage alliances, such as Nebuchadnezzar’s marriage to the daughter of the Median king (ABC, 3, l.29) and the marriage between the Median king Astyages (r. 584?-550?) and Aryenis, daughter of the Lydian king Alyattes (Hdt., 1.74.4; cf. Röllig, p. 22). Royal daughters were given in marriage to foreign kings in order to affirm political alliances or to confirm a peace treaty. Astyages is said to have married his daughter Mandane to Cambyses I (6th century BCE) in an apparent attempt to prevent the downfall of his realm (Hdt., 1.107.1), while according to Ctesias (d. after 397 BCE), the first link between the two royal houses was established only after Cyrus I (6th century BCE) conquered Media and subsequently married Astyages’ daughter Amytis (FGrH 688 F9). Persian kings also established their connections with the Persian nobility through a deliberate marriage policy. Cyrus II (r. ca. 559-530) married Cassandane (d. 538? BCE), daughter of the Achaemenid Pharnaspes (Hdt., 2.1.1). Phaidyme, daughter of the noble Otanes was said to have been the wife, first of Cambyses II (r. 530-522), and then of Bardiya. In order to legitimize his accession, but, more importantly, to prevent rival claims to the throne, Darius I entered a series of marriage alliances which remain unparalleled within Achaemenid rule. His alliances included the daughters of the previous kings, Atossa, Artystone, and Parmys (Hdt., 3.88.2-3). In addition, he created close familial links with his two closest allies who helped him to secure the throne, Gobryas and Otanes. He married daughters of both nobles (Hdt., 7.2.2, 3.88.4), while they themselves were married to sisters of Darius (Hdt., 7.5.1, 7.82). Their respective offspring also intermarried, with Darius’ daughter Artazostre marrying Mardonius son of Gobryas (Hdt., 6.43.1; PFa 5), and his son Xerxes I (r. 486-465) marrying Otanes’ daughter Amestris (Hdt., 7.61.2). Such interfamilial marriages were again concluded by the offspring of Darius II (r. 423-404), albeit on a much smaller scale, when Amestris was married to Teritouchmes son of Hydarnes and Artaxerxes II (r. 404-359) to Hydarnes’ daughter Stateira (Ctesias FGrH 688 F15). Artaxerxes’ own daughter was to be married to Tissaphernes, satrap of Sardis, who might himself have been a son of the same Hydarnes (Diod. Sic., 14.26.4; cf. Lewis, pp. 83-84).

Marriages between family members, including half-siblings, nieces, and cousins also occurred, and as is important to emphasize, these were not regarded as incestuous. Darius, whom Xerxes I had designated as heir to the throne, was married to Artaynte, the daughter of his uncle Masistes (Hdt., 9.108.1), while Ochus, who succeeded to the throne as Darius II, had married his half-sister Parysatis (Ctesias FGrH 688 F15) when he was still satrap of Hyrcania (Gorgān). Artaxerxes III (r. 359-338) is said to have married his niece (Val. Max., 9.2. ext. 7). Greek sources claim that incestuous brother-sister and father-daughter marriages were concluded within the royal family, but it is difficult to determine their veracity. Cambyses II allegedly married two of his sisters (Hdt., 3.31.1), which even at the time was regarded as an unlawful act, and Artaxerxes II is said to have married two daughters (Plut., Art. 23.5-6). Against the accusation of Cambyses’ incestuous marriages stands Herodotus’ own statement that Cambyses was married to Otanes’ daughter Phaidyme, as well as Ctesias’ reference to a wife named Roxane (FGrH 688 F13), whom he does not identify as a sister. Furthermore, the fact that the accusation of incest is listed in a series of sacrilegious acts committed by Cambyses, all of which are to emphasize his insanity and hubris, should caution against their acceptance. They derived from a common Egyptian source hostile to Cambyses, and some of these atrocities, such as the killing of the Apis bull, have been proved to be untrue (cf. Kuhrt, Sherwin-White 1987). The accusation made by Plutarch (46-after 119 CE) that Artaxerxes II married two of his daughters cannot be validated either way, but considering Plutarch’s penchant for court gossip, these marriages seem to emphasize the decadence of the Persian court manifested in the stereotypical portrayal of Persian women as power-hungry and scheming. If these marriages were concluded, we know of no offspring resulting from them. The reference that Darius III (r. 336-330) married his sister Stateira (Plut., Alex. 30.3) is contradicted by the passage in Curtius Rufus (fl. 1st century CE?), according to which only two children of Sisygambis’ family had survived, namely Darius III and his brother Oxathres (Curt., 10.5.23, 3.11.8). We also have the statement by Diodorus Siculus (fl. mid-1st century BCE) that Darius’ wife was a sister of Pharnaces (Diod. Sic., 17.21.3). Royal daughters were frequently used as pawns in the perpetual re-affirmation of the nobility’s loyalty to the king, and were given in marriage to Persian nobles as the highest royal reward for loyal service (Hdt., 5.116, 121-2; Ctesias FGrH 688 F13; Xen., An. 2.4.8; Plut., Art. 27.7; Diod. Sic., 14.26.4; cf. Brosius, 1996, pp. 72-77).

(4) Female workers. The PFT provide unique insight into the position of female laborers. Although women do not appear to have held office within the royal administration or were involved in conducting religious ceremonies, they are amply attested as workers laboring alongside men and children. Frequently workforces were headed by a woman called arraššara pašabena who – with one or two exceptions – received the highest monthly ration among the labor force, including 50 quarts of grain, 30 quarts of wine, and the third of a sheep. These women were evidently known to the administration by their personal name (PF 1790). The f pašap were a special workforce which is well attested in the villages of Persis and Fahliān. Identified by Hinz (1973a, p.172) as an occupational designation meaning “Teppichknüpferinnen, Schneiderinnen,” closer investigation suggests that f pašap is a collective term which refers to a particular kind of a predominantly female workforce and which includes a number of different professions (Brosius, 1996, pp. 155-66, esp. 165-66).

For labor requiring less skilled workers, reflected in the limited range of ration scales, women received one third less food than men. In contrast, for skilled labor women and men were given equal rations according to their level of qualification. As mothers of newborn babies, women were given extra rations, with mothers of boys receiving twice the amount of food than mothers of baby girls (Brosius, 1996, pp. 171-78). The almost even balance between male and female births, as well as between male and female working children, attests to the fact that infanticide was not practiced.

(5) Representations of women. There are few depictions of women but they nevertheless demonstrate that women could be shown in a variety of media, and thus were an accepted subject in Achaemenid art. According to Herodotus, Darius I had a statue of his wife Artystone made of gold (Hdt., 7.69.2). Although no statues of women have come down to us, the surviving evidence suggests that royal and non-royal women were depicted. The high relief figure of a high-ranking Persian woman found in Egypt (FIGURE 2) shows her in a many-folded Persian dress belted in the front (see CLOTHING ii. In the Median and Achaemenid periods) and wearing an elaborate necklace. She is wearing a turreted crown, and her bobbed hairstyle follows Achaemenid fashion (cf. the lapislazuli head from Persepolis and the limestone head from Masjed-e Solaymān in Spycket, figs. 6, 9).

Similarly the ivory statuette from Susa (Musée du Louvre, SB 3728/9188; see Spycket, fig.1) depicts a crowned woman in Persian dress. The Persian women on a tapestry from Pazyryk are also wearing crowns, with a long veil falling down the back (Rudenko, pp. 296-97 and pl. 177). Their depiction is evidently based on an known and accepted artistic form that can be repeatedly found on Achaemenid or Achaemenizing art of the satrapies, including gems and rings from Asia Minor, funerary stelae from Dascylium, stands of incense burners and kohl tubes from Syria-Palestine (Spycket, pp. 43-45 and pls. 21-26; Starr, pp. 49-115; Boardman), as well as gold plaques (Dalton) from the Oxus Treasure. 

ii. PARTHIAN PERIOD

(1) Introduction. The problematic situation of the written and archeological sources for the Parthian empire extends itself to the discussion of women in the Parthian period. Few documents have come down to us which enable us to gain an understanding of Parthian women, and these reveal little more than the names and titles of royal women. Classical sources provide some information on the marriage policy of Parthian kings and on the women’s presence in the king’s entourage. It can safely be assumed that the position of Parthian noble women mirrored that of royal women, albeit on a more modest scale. It certainly appears to have been the case that Parthian nobles were accompanied on campaigns by their female household, exemplified in the 200 wagons which transported the women of Surena (Plut., Crass. 21). The sources do not permit a detailed discussion of the ranking order among the women of the court, nor of their possible political influence and economic independence. On the basis of Achaemenid and Seleucid practices one can only hypothesize that Parthian royal women, too, owned land and estates, as well as manufactures. The influence of the Seleucids on the early Parthian kings, perceptible in the expression of kingship (e.g., the diadem, Greek epithets in the royal title, the influence in art and architecture), may have extended to the public presentation of Parthian royal women. No information has as yet come to light which would allow us to discuss the legal and economic situation of non-royal women in the Parthian period.

Evidence from Nisāya alludes to the Greek influence in the depiction of women in statues, sculptures, stucco work, and in art objects made of precious metal and ivory (Invernizzi, pp. 45-60). To gain insight into the way women were depicted in Parthian court art we need to turn to the artistic legacy of the regional kingdoms of the Parthian empire, especially to Elymais, as well as to cities such as Dura-Europus and Hatra which modeled their art and architecture on those of the Parthian royal court. Their art provides us with a rich impression as to how royal and noble women dressed according to Parthian fashion and could be publicly presented in their stance, expression, their coiffure, dress, jewelry and other adornment.

(2) Titles for Parthian royal women. Parthian kings were polygamous. In addition to the king’s wives, royal concubines also resided at the court and formed part of his entourage. The royal wives were chosen from among the female members of the royal family itself as well as from Parthian noble families. Among the non-Parthian women at the king’s court we find the daughter of Demetrius Nikanor (Justinus, 38.10.10), who was captured after the death of Antiochus VI (r. 145-142) and taken to the court of Mithridates I (r. 171-139/38). One of the foreign women who secured a high position at the court was Musa (Josephus, Ant. Jud. 18.2.4). Augustus (63 BCE-14 CE) sent the Italian woman as a ‘gift’ to Phraates IV (r. 38-3/2). She was the mother of the heir to the throne, Phraates V (r. 2 BCE-2 CE), and by marrying her own son, became queen. A Greek woman (Dio Cass., 62, 63.5; Josephus, Ant. Jud. 20.74; Tac., Ann. 12.44) is known as the mother of Vologeses I (r. 51-76/80 CE).

The wives of Parthian kings are mentioned in official documents, a practice which may have been adapted from the Seleucids (OGIS 219). Their personal names and titles appear in Babylonian documents of the Parthian period, as well as in the Greek documents from Avroman in Kurdistan. The king’s wife bore the title “queen” (Bab. šarratu, Gr. basilisse,), while the term “lady” (Bab. beltu) was used as a form of address. Greek references to royal wives as “sisters (of the king) from the same father” (Gr. homopatriai adelphai) are known from the Avroman documents which mention Siace (Minns, no. I: 2-3), the wife and sister of Mithridates II (r. 124/123-88/87), who is named alongside two other wives, “Aryazate surnamed Automa, daughter of the Great King Tigranes, and his wife” and “Azate his compaternal sister and wife” (Minns, no. II: 3-5). A sister-wife is also attested for Orodes I. Her name is given in Babylonian documents as Isbubarza (Strassmeier, p. 112; Potts, p. 392). While these references point to an emphasis on the siblings’ common paternal descent they allow no conclusions as to whether they descended from the same mother, and thus to their identification as full siblings. Not only do we lack conclusive evidence that Parthians entered incestuous marriages between full siblings, these reference do also not allow us to draw any conclusions about such practice within the Parthian dynasty.

In most known cases women are only mentioned as the king’s wives. A Babylonian text mentions Ri-[x]-nu, the mother of Phraates I (r. 176-171), and probably the wife of Mithridates I. As the mother of the heir to the throne she took an eminent position at the court when she acted as regent for the young Phraates after the king’s death (Debevoise, p. 29; Oelsner, pp. 147-8). Both wives of Gotarzes I (r. 91/90-81/80), one of whom was named Aši’abatum, were referred to as his “wives and ladies” (Bab. aššati(?)-šu beltu; cf. Strassmeier, p. 112; Minns, p. 34), while Pi-ir(?)-ri-ta-na-a, the wife of Phraates III (r. 71/70-58/57), was called his “wife and queen” (Bab. aššatišu šarratu; cf. Strassmeier, p. 112; Minns, p. 36). Phraates IV had four wives: Olennieire, Cleopatra, Baseirta and Bistheibanaps (Minns, p. 30).

(3) Activities at court. Female members of the king’s family were used as pawns in the establishment and confirmation of political alliances. This was a practice attested especially between Parthia and Armenia, though undoubtedly political marriage alliances with other principalities of the Parthian empire also occurred. A daughter of Phraates IV was married to Tigranes of Armenia, and Orodes II (r. 58/7-38) affirmed his alliance with the Armenian king Artavasdes with a marriage between his son Pacorus and a sister of the Armenian king (Plut., Crass. 33).

The royal wives and the women of the palace accompanied the king on military campaigns and thus were frequently in danger of capture by the enemy and even of death. Among the Parthian princesses captured by Romans, the daughter of Osroes (r. 108/9-127/28) was taken hostage by Trajan (r. 98-117) after the occupation of Ctesiphon. She was returned to the Parthian king by Hadrian (r. 117-38) in 128/29 CE (SHA, Hadrian 13.8). To prevent the women’s falling into enemy hands, Parthian kings did resume to extreme measures. In 26 BCE, threatened with an advance of Tiridates, Phraates IV killed the women in his entourage (Isidore of Charax, Parth. Stat. 1), and in 52 CE the wife and children of the Armenian king Mithridates were killed by Rhadamistus, son of Pharasmanes of Iberia (Tac., Ann. 12.44-47). When Rhadamistus himself was pursued, he stabbed his pregnant wife and threw her into a river. She was found alive and rescued (Tac., Ann. 12.51; 13.6). After the Alans captured the female entourage of Vologeses I’s brother Pacorus, he was forced to pay ransom for his wife and his concubines (Josephus, Ant. Jud. 7.7.4). (4) Representations of Parthian women. Public presentation of royal women seems to have been limited. They do not appear on the few extant Parthian reliefs, while statues and sculptures of royal Parthian women which – like statues of the kings – will have existed, did not survive. Recent scholarship has moved away from the identification of the two seated figures next to the reclining king in the Elymaean investiture relief from Tang-e Sarvak II as Mithra and Anāhitā/Athena or Artemis and Athena (Vanden Berghe and Schippmann, pp. 70-73). Both figures are male, wearing long tunics and trousers, rather than dresses, and at least one of the seated figures is sporting a moustache in Parthian fashion (Vanden Berghe and Schippmann, p. 72).

Only in exceptional cases were women depicted on coins. Thus Musa is depicted as the queen of Phraates V, wearing a crown and a diadem (Alram, p. 385 pl. 3 no. 27). Not merely this marriage, but also her appearance on coins was a unique occurrence. At a regional level we find Anzaze depicted on coins together with her husband Kamnaskires II (ca. 145-139 BCE), king of Elymais (Alram, p. 386, pl. 4 no. 37).

As the king’s power and exalted position were reflected in the splendid display of the court, the public presentation of his entourage formed an important element in the expression of kingship. As part of that expression royal women were undoubtedly dressed in fine and richly embroidered garments, long-sleeved, many-folded dresses tied by a belt. Their jewelry included multiple necklaces, bracelets and earrings, made of large-sized pearls or worked in gold with precious and semi-precious stones. Royal women probably wore a crown, but they could also wear the upright tiara or simply a veil which hung down the back. This is the impression given by the depictions of high-ranking women in sculpture, relief, and painting in the art of cities such as Palmyra, Dura-Europus and Hatra, which was influenced by, or modelled on, Parthian court art (Gall, pp. 75-84). The depiction of these women, as well of local female deities, testifies that it was not a restricted artistic subject, but rather, that female statues and other representations of women were an expression of their high social status (FIGURE 3).

iii. SASANIAN PERIOD

(1) Introduction. The evidence for the Sasanian period reveals a more discernible public role played by royal women. In continuity with Parthian practices, female members of the royal family were mentioned in official documents, most notably in trilingual inscription (ŠKZ) of Shapur I (r. 239-270) from the Kaʿba-ye Zardošt at Naqš-e Rostam. But moreover they received public honor in royal fires lit for members of the king’s family, as well as in the sacrifices ordered by the king. Numerous seals and finger rings depict royal and high-ranking women. The king’s consort appears on royal reliefs and on silver plates and bowls. Their depiction on coins (FIGURE 4) remains exceptional and is only attested for the wife of Bahrām II (r. 276-93) and the reigning queen Bōrān (r. 630-31).

The discussion of Sasanian women has been impeded by two major assumptions: Firstly, that ’ordinary’ women were not depicted in Sasanian art but that any representation of females must be that of the goddess Anāhitā (Shahbazi, 1983, p. 261), and secondly, that the title “Queen of Queens,” used in ŠKZ as a reference to a king’s daughter or sister, signaled the woman’s status as a wife of the king, and thus gave proof of incestuous marriage alliances within the Sasanian royal family. These alliances were understood as a reflection of the Zoroastrian practice of close-kin-marriages (MPers. xwēdōdah). Yet scrutiny of the sources should caution against the equation of a royal title with a particular marriage, even more so since the evidence for such alleged marriages is extremely scant. The fact that Šāpurduxtak, the wife of Bahram II, bore the title, although she was not his sister, gives further weight to the argument that the title was not restricted to a king’s sister. In her case we need to consider whether her status as the king’s wife coincided with that of queen of queens. Equally, Kardēr’s statement that he promoted close-kin-marriages (ŠKZ para. 14) neither allows a conclusion as to the kind of familial marriages he fostered, nor whether he referred specifically to the royal family or to Zorostrian Persians in general. In consequence of recent criticism and rejection of the existence of a state religion in the Sasanian empire it also can no longer be argued that such marital unions were sanctioned at highest political level (Wiesehöfer, 2001, pp. 210-12; cf. Macuch, p. 152). Analysis of the written evidence for the Sasanian period does not permit the conclusion that the Sasanian kings favored incestuous marriages. (2) Titles for royal women. The main source for the titles of royal women is ŠKZ. Referring to royal women of the immediate family of the king, as well as of the extended royal family, the inscription provides us not only with personal names but also with different titles which indicate an order of rank among the royal women. The king’s mother, referred to was the “Mother of the King of Kings” (MPers. šāhān šāh mād), and the king’s wife, who bore the title of the “Empire’s Queen” (MPers. šahr bāmbišn), probably held the most senior rank among the women at the court. The title “Queen” (MPers. bāmbišn) was held by royal women of the court who were the king’s daughters and sisters, as well as the wives of Sasanian regional kings. The title “Queen of Queens” (MPers. bāmbišnān bāmbišn) probably has to be understood within that context as a title for the royal woman who took the highest rank among the queens (Gignoux, “Ādur-Anāhīd”; Wiesehöfer, 2001, p. 175). According to ŠKZ it was given to a king’s daughter, and is attested for Dēnag, the daughter of Pāpak, and for Ādur-Anāhīd, the daughter of Shapur I. The title “Princess” (MPers. duxš), mentioned only once for Rōdduxt, daughter of Anōšag (ŠKZ para. 37), might indicate young age and/or her premarital status. One form of address was “Lady” (MPers. Bānūg). The wife of Ardašir I (d. 242 CE) and mother of Shapur I is referred to as “Lady Murrōd, Mother of the King of Kings’ (ŠKZ para. 37), and the same address is used for the goddess Anāhitā in the inscription of Narseh (r. 293-302) from Paikuli (NPi B).

In his inscription Shapur I ordered royal fires to be lit in honor of himself as king, as well as for his children, beginning with his daughter Ādur-Anāhīd and also mentioning three of his sons, Hormozd I (r. 270-71), Shapur, and Narseh (ŠKZ para. 36). Yet the daily sacrifices of a lamb, 1.5 measures of bread and 4 measures of wine were to be given for each member of his family, including the Mother of the King of Kings, the Empire’s Queen, the king’s daughters, as well as the king’s male and female ancestors. These sacrifices for the souls of the living and deceased members of the royal dynasty, including the royal women, attest to the recognized official status these women held at the Sasanian court.

(3) Marriage alliances. The polygamy of the Sasanian kings was on the one hand due to the king’s need to produce numerous male offspring in order to select the best heir to the throne, and on the other hand due to the need to enter marriages with the daughters of Sasanian nobles and political allies in order to cement political allegiances. Since the heir’s descent was not dependent on his mother’s descent, ethnicity, or religious belief, the claim that incestuous marriages, which were allegedly concluded in order to preserve the ‘pure bloodline,’ determined the choice of the heir to the throne, is undermined further.

The beginnings of the Sasanian dynasty are as ambiguous in regard to royal women as they are in regard to Ardašir’s descent from Sāsān. While the Muslim historian Ṭabari (839-923) claims that Pāpak was the son of a union between Sāsān and Rāmbehešt, a daughter of the noble family of the Bāzrangi clan (pp. 4-5), Shapur I calls in his inscription Dēnag mother of Pāpak, and refers to her as King’s Mother, without linking her to Sāsān (ŠKZ para. 23). Pāpak himself was married to a woman named Rōdak, who, as the mother of Ardašir I, held the title of Mother of the King of Kings (ŠKZ para. 41). Pāpak’s daughter Dēnag was given the title of Queen of Queens, a title which now appears parallel to the title of King of Kings held by Pāpak’s son and heir. A seal (Hermitage Museum, St. Petersburg, Russia, Inv. GL. 979; cf. Gignoux and Gyselen, 1989, p. 882) depicting a woman in profile, and identifying her as “Dēnag, Queen of Queens, head of the eunuch section” (MPers. Dēnag ī bāmbišn bāmbišnān mahist pad tan šābestān) has conventionally been ascribed to her. What precisely this title meant cannot be ascertained, but it might be suggested that as head of the women’s section of the court she was in charge of its personnel, which presumably included not only eunuchs, but also female servants.

As mother of Shapur I, the heir to the throne, the title of Mother of the King of Kings was transferred to Ardašir’s wife Murrōd, while Shapur’s daughter Ādur-Anāhīd became the Queen of Queens (Gignoux 1985: 472). A seal is also attested for the Empire’s Queen Xwaranzēm, most likely the wife of Shapur I (Gignoux, 1978, 4.77). The obvious distinction between a king’s wife – identified as in the cases of Pāpak, Ardašir I, and Shapur I – and a royal daughter who was given the title of Queen of Queens, as different from the title of Queen bestowed upon other women of the court, forces us to dismiss the commonly held view that the title must automatically refer to the king’s wife. Evidently it was used to signal the highest rank among the queens of the Sasanian court, which included the wives of regional kings.

An unusual reference to a daughter’s female descent (ŠKZ para. 37) is difficult to interpret, since Xwaranzēm is mentioned as the mother of Warāzduxt, though the name itself does not signify that she was not a daughter of Shapur I. In his inscription Shapur I mentions his sons Hormozd I, Shapur, Narseh, and Bahrām I (r. 271-74) without indicating their maternal descent. However, according to Ṭabari (p. 40), the mother of Hormozd I was a daughter of king Mihrak. The mention of Hormozdag as the son of Hormozd I is preceded by a reference to a queen Staxryād, who may be identified as the wife of Hormozd I (ŠKZ para. 38). For Shapur I’s son Shapur, king of Messene, a son Pērōz and a daughter Šāpurduxtak are attested, while his wife may have been queen Dēnag, identified as “Queen of Messene, dastgerd of Shapur” (ŠKZ para. 44). Šāpurduxtak, Queen of the Sakas and Lady Narsehduxtak are named respectively as the wife and presumably the daughter of Narseh, king of the Sakas. Shahbazi (1983, p. 267) has rightly cautioned against the identification of Šāpurduxtak as the daughter of Shapur I. That the personal name itself allows no conclusion as to the woman’s paternal descent becomes clear in regard to another daughter of Narseh, Hormozdduxtak (ŠKZ para. 37). Finally, a woman referred to as Lady Čašmag and most likely a member of the Sasanian royal family (ŠKZ para. 37) may be linked to “Čašmag brave-of-Shapur” (ŠKZ para. 46).

Ṭabari (pp. 309, 316-17) mentions the brother-sister marriage between Bahrām VI Čōbīn (r. 590-91) and Kurdiyah, but it is not stated whether they were full siblings. Nothing is known of the descent of Dēnag, the wife of Yazdegerd II (r. 439-57) and mother of Pērōz (Firuz, r. 459-84) and Hormozd III (r. 457-59), though her name indicates that she was a member of the Sasanian royal family. She is said to have governed the kingdom around Ctesiphon (Ṭabari, p. 109). If this information is historical, it indicates that royal women could be given political positions in the organization of the empire.

Several cases attest to the marriage of the Sasanian king with non-Persian women due to political alliances, and, intriguingly, the sons of these unions often succeeded to the throne. The wife of Bahrām V Gōr (r. 420-38) is said to have been an Indian princess (Ṭabari, p. 868), but nothing further is known about her or her offspring. The marriage between Kawād I (r. 488-96) and Nēwanduxt, the daughter of the Khāqān of the Hephtalites, resulted in her becoming the Mother of the King of Kings Ḵosrow I Anoširvān (r. 531-79). It has been suggested that she might have been a daughter of Pērōzduxt, who had previously been taken captive by the Hephtalite king, and was returned after Suḵrā had negotiated her release (Ṭabari, pp. 119-20, 129 n. 334). The plausibility of this report however, depends on the feasibility of Pērōzduxt’s bearing a royal child whom she then had to leave behind at her release. For equally political reasons (Masʿudi, para. 632), Ḵosrow I married a daughter of the Khāqān of the Turks who became the mother of Hormozd IV (r. 579-90). Ḵosrow II (r. 590-628) is said to have married two Christian wives: the Byzantine princess Maryam and Širin. According to Christian sources, the latter was also a Byzantine princess (Theophylactus Simocatta, Hist. 5.14.2-11; cf. Whitby, pp. 305, 234), though of Aramean origin (Fowden, p. 136). According to Ṭabari (p. 379-80, cf. 312 n. 729), Maryam was the mother of Kawād II (r. 628), while Širin was the mother of Mardānšāh and the (foster) mother of Šahriyār. Ḵosrow II married Širin in 592, and, hoping for the birth of a son, dedicated gifts to the Shrine of St. Sergius at Rusafa (Sergiopolis) a year later (Evagrius, HE 6.21; Theophylactus Simocatta, Hist. 5.14.2-11; cf. Fowden, pp. 137-39). In the royal couple’s donation of a gold cross, as well as in the building of three churches, one of which was dedicated to St. Sergius, Fowden (pp. 139-40) sees a parallel to the previous offering of a cross and the building projects undertaken by Justinian (r. 527-65) and Theodora (d. 548), the Byzantine emperor and empress. The church for Širin was built together with a palace in Beṯ Lašpar, which has been identified as Qaṣr-e Širin in northwest Iran (Chronicle of Seert, 58: 146-7; cf. Fowden, p. 140).

During the political turmoil of the late Sasanian period, two daughters of Ḵosrow II, Bōrān and Āzarmīgduxt (r. 631), succeeded, if only for brief periods of time, to the Sasanian throne. According to the Muslim historian Hamza Eṣfahāni (893/4-970/1) Bōrān was a daughter of Khosrow’s Christian wife Maryam (p. 54). Her reign was marked by a benevolent rule and the hope to end the war by peaceful means. New coins were issued, and a tax relief and the rebuilding of stone bridges and bridges made of boats were ordered to improve the economic situation in the empire (Ṭabari, pp. 403-405). Coins identify her in the legend and show her portrait, wearing a crown and crescent moon (FIGURE 4; cf. Göbl, pp. 54-55, pl.15, nos. 228-29). Bōrān’s most important political gesture was the return of the Holy Cross to Jerusalem.

Āzarmigduxt’s reign lasted for only six months before she was apparently killed by Rostam, the son of Farroḵ-Hormozd, the spāhbed of Khorasan (Ṭabari, pp. 406-407). She minted coins, which bore her name but showed the image of a king, identified either as her father or Farroḵ-Hormozd (Mochiri, pp. 241-44; Gignoux, “Āzarmīgduxt”; cf. Sellwood, pp. 21, 169-70). All known coins with her name are from the mint WYHC.

(4) Activities at court. As was the custom, royal women accompanied the king on campaigns, traveling in carriages for greater comfort (Ṭabari, p. 130 n. 335). The women’s presence was probably to instill the army with confidence in a victorious outcome of the impending battle, but in reality it exposed the women to potential danger. This was no more apparent than in the Hephtalites’ capture of the women’s train of Pērōz, including his wife and daughter. The women were only freed when the Sasanian general Suḵrā entered ransom negotiations with the Hephtalite king (Ṭabari, pp. 119-20). Previously, the women in the train of Narseh suffered a similar fate when they were captured by the Romans, and released only after Narseh agreed to a peace treaty (Peter Patricius, frgs. 13-14). Ḵosrow II ordered his women and children to be brought to safety when he came under threat from Bahrām VI Čobin his rival to the throne (Tabari, p. 304). (5) Representations of Sasanian women. The representation of royal women attests to their official recognition as being closely related to the king. Royal women like Xwaranzēm, Dēnag, and the wife of Shapur III (r. 383-88) had their own seals which identified them by their portrait and an inscription. The mere fact that royal women owned seals is a reflection of their authority which allowed them to conduct transactions and official affairs (Gignoux and Gyselen, 1989). Their representation on reliefs and on silver bowls and plates demonstrates that the depiction of females was part of the catalogue of royal Sasanian art. Their appearance in investiture reliefs signals a new variant to the genre. The subject is first introduced by Ardašir I whose investiture relief at Naqš-e Rajab includes two females (FIGURE 1). Facing away from the investiture scene they nevertheless perform the same gesture of respect as the Sasanian nobles witnessing the event.

At Naqš-e Rostam, the king’s consort now appears together with other members of the king’s family in the investiture relief of Bahrām II (Herrmann, p. 168; Shahbazi, “Bahrām II”). Other artistic subjects were introduced. On the rock reliefs at Sarab-e Qandil and Barm-e Delak the royal couple is depicted exchanging a flower. At Sarab-e Qandil the flower is held in the woman’s right hand, while her left hand is hidden in her sleeve and held up in a gesture of respect, while at Barm-e Delak the king offers a flower to his wife. It is now thought that these scenes depict royal marriages (Shahbazi, 1998; cf. Vanden Berghe, “Barm-e Delak”; for different identifications, see Hinz 1973b, pp. 201-212). This artistic theme finds further expression on seals which depict couples holding a ring or a flower (Gignoux and Gyselen, 1982, p. 50). At Sar Mashad a relief ascribed to Bahrām II shows the king protecting his wife and two other family members from a lion’s attack (Herrmann, p. 167; Shahbazi, “Bahrām II”). Bahrām’s wife is also identified on a silver cup from Zargveshi, Georgia, which depicts her portrait alongside with those of the king and the son (Harper, 1974, pp. 63-64, 70-71). Exceptionally, too, is her depiction on coins. On the obverse she is appearing either with the king as the royal couple, or with the king and the heir to the throne. The reverse shows the royal couple flanking a fire altar, and the legend identifies her as Šāpurduxtak, Queen of Queens (Lukonin, p.112; Shahbazi, 1983, pp. 265-66). The identification of this queen as the daughter of Shapur, king of Messene (Lukonin, p.112; cf. Frye, p. 304) is a mere assumption and impossible to substantiate. But as Queen of Queens she ranked above all other queens of the royal court.

A further image of the Sasanian royal family is found in Narseh’s relief at Naqš-e Rostam. It depicts the king and queen standing opposite each other in semi-profile, both holding a ring. Between them stands the small male figure, most likely to be identified as their son and heir Hormozd II (r. 302-309), and two nobles stand behind the king. While in the past scholars tended to identify the woman as the goddess Anāhitā, serious doubt has been expressed towards the association of Anahita with royal investiture (Shahbazi, 1983). Her identification on the royal reliefs was simply the logical conclusion from the erroneous assumption that mortal women were not depicted. The woman next to Narseh has her left hand covered by the sleeve of her dress. This stance is a gesture of respect given to the king by his court, as can be seen on several Sasanian reliefs depicting members of the Sasanian nobility, including Ardašir’s relief at Naqš-e Rajab. The woman, therefore, is most likely Šāpurduxtak, Narseh’s wife and queen. The identification of Šāpurduxtak as the daughter of Shapur I is supported by her crown, which is identical with her father’s (Shahbazi, 1983, pp. 266-67). Her portrait was also added to Shapur’s triumph scene at Darabgird (Shahbazi, 1983, p. 267 fig. 5).

In light of this evidence, the identification of the woman in the investiture relief at Ṭāq-e Bostān, ascribed to Ḵosrow II, as the goddess Anāhitā needs to be reconsidered. As none of the women depicted in the early Sasanian reliefs can be identified as Anāhitā, there is no convincing evidence to suggest that she was associated with investiture. It is most likely that the woman in the relief at Ṭāq-e Bostān is the king’s consort, and thus ought to be identified as either Maryam or Širin.

High-ranking women were depicted on a variety of silver vessels (Harper 1971; 1975; 1981) and on seals and bullae (Gignoux, 1978; Gignoux and Gyselen, 1982; 1989). They are shown as full figures or busts, with their face in profile facing to the right. They are characterized by long, curly hair, which is either worn in ringlets or braids, and/or with part of it gathered in a round hairdo on top of the head. The considerable variety in the headdresses of royal women invites the suggestion that they, too, wore a personalized crown. The headdresses include simple high caps, caps embroidered with pearls, beaded diadems, turreted crowns similar to that of Shapur I, and crowns made in the shape of rams’ horns (FIGURE 2).

Their jewelry includes several necklaces, using large pearls, as well as pearl-shaped earrings. They are depicted in this style on seals, e.g., the seal of Dēnag and of the wife of Shapur III (Gignoux and Gyselen, 1989, p. 896, fig. 24). The latter is wearing a round beaded headdress decorated with rams’ horns and an oak-leaf. She may be compared to the queen depicted in a banquet scene on a silver dish, wearing an identical crown (Walters Art Gallery, inv. no. 57.709; Vanden Berghe and Overlaet, fig. 65). Female dancers and musicians are frequently depicted on Sasanian silverware. These objects, drinking cups, vessels, and especially jugs, probably have to be placed in the context of royal banquets. Female musicians playing flutes and harps are also depicted on the hunting reliefs at Ṭāq-e Bostān. These motifs continued into the art of the early Islamic period.