PALACE ARCHITECTURE. The abundant variety of styles in Iranian domestic architecture conceals a basic functional system that has remained unchanged since the Achaemenid period (see ACHAEMENID DYNASTY) and is preserved in palaces and substantial residences. This tradition can scarcely be revealed by art historical methods. Architecture is not a purely artistic expression, but a complex cultural phenomenon, which is also shaped by functional, technical and social factors. The traditional type of Iranian residences shows two main elements: a representative part of one or more large and high rooms or halls, and a private part of at least two stories. The private ground floor compartments are used as secondary business, storage or junk rooms, if large enough, while the upstairs rooms are the personal living space. There is view contact between the private and the representative areas (Huff, 1978a; idem, 1993, pp. 50-53; idem, 1999; idem, 2005, pp. 371-95). Elevated private rooms are not by chance called the ‘noble floor’ (It. piano nobile), and are especially favored in hot climates because cool breezes are one of their advantages.

This functional system has been observed and studied in well-preserved palaces and houses of more recent periods, and consequently can be inferred from remaining traces and foundations of excavated archeological monuments. It was already developed in the Late Bronze Age culture of Central Asia, and seems to have survived throughout the Early Iron Age period (Sarianidi, pp. 79-132; Huff, 2001, pp. 181-97). It is present in excavated Median monuments in Iran (see MEDIA). In the small fortress of Tepe Ozbaki, probably the seat of a local chieftain, a block of narrow compartments without an access door could only have served as a substructure for rooms upstairs, accessible by still preserved stairs. One of the larger ground floor rooms, with benches along the walls, obviously for communal gatherings, has installations that seem to indicate cultic purposes (Majidzadeh, pp. 5, 9-10; Huff, 2005, pp. 380-81). The highly informative site of Godin Tepe (FIGURE 1), clearly the residence of an important baron, grew to a large fortress with two assembly halls, containing wall benches and podium, and successively enlarged blocks of substructures. These compartments were wide enough to be used as junk- or box-rooms, as suggested by archeological evidence, though their main purpose must have been to support the upstairs private rooms of the palace, as proven by the remains of a large and comfortable staircase (Young and Levine, pp. 30-36; Huff, 2005, pp. 378-80).

The most impressive complex of Median architecture at Tepe Nush-i Jan (FIGURE 2) has predominantly cultic character. A hypostyle assembly hall is attached to the older of two successive fire temples, while the extensions of the largest building, the so-called fort, nearly embrace the second sanctuary. Although the parallel compartments of 2.34 m were wide enough for use, their height of more than 5.5 m was out of proportion for store rooms, and the same goes for the sumptuous layout of the entrance with guard room and the comfortable winding staircase. Fragments of the strut vaultings, broken down into apparently empty magazine compartments, still preserve remainders of the upstairs floor and inventory on their upside, indicating that, like in Godin Tepe, the fort and its enlargements were primarily built as substructures, with the essential rooms of whatever purpose on top (Stronach 1969, pp. 11-15; Stronach and Roaf, pp. 11-24; Huff, 2001, pp. 193-94). The general scarcity of detailed knowledge of Median society and religious customs makes the interpretation of Nush-i Jan difficult. But as an example of elevated prestige architecture, its fort is of significant value. The architectural type seems to have been widespread in Median and Achaemenid times, as is testified by the donjon-like castles of Garri Kariz and Ulug Depe in Turkmenistan, and of Tell Gubba in Iraq (Pilipko, pp. 28-58; Boucharlat, pp. 479-87; Fuji, pp. 28,150-52).

A combination of congregation hall and two-story wings was already developed in the possibly cultic buildings of Early Iron Age Ḥasanlu (see ḤASANLU TEPPE; Young, pp. 48-71; Dyson and Voigt, pp. 219-36). Their model of an anteroom with a square winding staircase was a predecessor of Nush-i Jan. But only during the Achaemenid period was a successful solution for this combination established. Cyrus II (b. ca. 600, d. 530 BCE) may have taken the first step at his residence at Pasargadae (FIGURE 3). This hypothesis offers a conclusive explanation for the empty spaces between the long porticus halls on either side of the hypostyle hall of palace P. It would allow for fitting Pasargadae in the further development of Persian architecture, if these now empty spaces contained two-story blocks of mud-brick constructions with the private rooms of the royal family. Moreover, the enigmatic rectangular lateral mud-brick pillars inside the hypostyle hall could be explained as supporting not only the roof but also upper galleries that were connected with the lateral upstairs rooms and which, conversely, provided those with views into the assembly hall (Stronach, 1978, pp. 78-106; Huff, 2005, pp. 381-83). The definite step towards a satisfactory combination of the two originally separate architectural elements of a hall of representation and a two-story private area into one unified palace building occurred during the reign of Darius I (r. 522-486 BCE), with the creation of the so-called apadana, best preserved at Persepolis (FIGURE 4; Schmidt, pp. 70-106; Stronach, 1985, pp. 433-45; idem, “Apadana”). The four corner towers between the ayvāns, which surround the apadana’s square hypostyle hall are nearly identical with the so-called fort of Tepe Nush-i Jan. In the precise excavation report, the archeological evidence and the architectural analysis both indicate beyond doubt that the lower part of the corner towers and the small rear compartments were primarily substructures for sumptuous upstairs rooms, certainly the private area of the royal family. The ground floor compartments obviously served as guard-rooms and junk- or box-rooms, while the hall’s rear rooms probably contained equipment for state ceremonies (Schmidt, pp. 70-106; Huff 1978a, pp. 237-43; idem, 2001, pp. 194-95; idem, 2005, 375-78). In his foundation inscription, the tablets of which were excavated beneath the hall’s inner corner, Darius implored Ahura Mazdā’s protection “for himself and his house,” probably meaning not only his dynasty, but also literally referring to his house (Kent, pp. 136-37).

A peculiarity of all later, well-preserved Iranian palaces of the apadana type is the visual communication between the building’s private and representative areas because windows in the upstairs rooms allow for viewing the great halls from above. That this form of communication was already possible in Achaemenian palaces can only be presumed, although the intention of having people in the private area participate in events in the public area without being visible from below, may have been one of the rationales for the creation of a unified palace building. There is little conclusive information on Parthian palaces but already the first Sasanian palaces, which Ardašir I (d. 242 CE) at Firuzābād built in part during the last decades of the Parthian period, display the fully developed system of upstairs private rooms with view into the halls of representation (Huff, 1993, pp. 45-53; idem, 1999, pp. 154-56). Although the use of vaulting techniques and stone-mortar masonry considerably changed the appearance of the earlier mudbrick architecture, the functional system remained the same. In the older palace Qalʿa-ye Doḵtar (FIGURE 5), windows offer views into the ayvān, the domed hall, and its side halls from the upstairs rooms, which are spared out from the masonry at the squinch zone. Ardašir’s great palace on the plain, called Ātaškada (FIGURE 6) and built after his victory over the Parthians (224 CE), has a first upstairs floor with high window openings, through which people could look down into the ceremonial halls, and a throne seat in the wall between ayvān and central hall so that the king could give audiences to smaller or larger congregations in either hall. But the private area was on a second upstairs floor, between the domes and with suites of barrel-vaulted and domed rooms (Huff, 1971, pp. 136-50; idem, 1978b, pp. 137-40; idem, 1999, pp. 154-56; idem, 2005, pp. 373-74; idem, 2008, p. 48, fig. 10, pp. 53-54).

The highly informative condition of the early palaces at Firuzābād provides a safe basis for an interpretation of the three-dimensional composition of the late Sasanian main palace of Tāq-e Kesrā (see AYVĀN-E KESRĀ; FIGURE 7) at Ctesiphon. Reliable traces of an upstairs private story are the large windows in the rear corners with a few remainders of a brickwork grid through which one could look down into the ayvān without being visible from below. The excavated foundations of ground floor rooms of different shapes and sizes on three sides of the ayvān are further proof of the building’s apadana structure (Reuther, pp. 15-23; Huff, 1971, pp. 150-54; idem, 1999, pp. 157-58). Before destruction, it must have been a huge block of brickwork, its side facades pierced by windows and small upstairs ayvāns or loggias, a building much bigger and more massive but similar to ʿĀli Qāpu (FIGURE 8) of Shah ʿAbbās I (r. 1588-1629), the oldest and most important Safavid palace of Isfahan.

ʿĀli Qāpu (lit. sublime porte, cf. Galdieri, p. 9, n. 2) consists of three different structures, one built on top of the other. The ground floor is the actual gate of the palatial area, serving administrative functions (Huff, 2006b, pp. 342-43), while the palace begins at the level of the great open portico (Pers. tālār). The physician and traveler Engelbert Kaempfer (1651-1712) mentioned that the tālār and the adjacent large high hall were used for state receptions, Nowruz festivities, government councils, and assemblies of the nobility. The shah and his dignitaries stayed on the tālār whenever they attended sports events, public ceremonies, executions, etc. on the maydān. On the remaining three sides of the large hall, there were upstairs room to which only the shah and his eunuchs had access. These rooms had windows that allowed the women of the court to observe ceremonies in the great hall and on the tālār (Kaempfer, pp. 164-65). The third unit of ʿĀli Qāpu was also an area with restricted access, because the delicate pleasure palace with one floor level contained the private rooms of the shah (Ferrante, pp. 133-206; Galdieri, pp. 15-33).

The palaces of Isfahan are paradigmatic of the various developments, re-buildings, and changing functions of royal palaces in general. In ʿĀli Qāpu, an extensive enclosure with a separate entrance gate and various lodgings for women and eunuchs was built in the palace gardens (Kaempfer, pp. 25-26, 163, pl. VI). Although ʿĀli Qāpu was designed as the main palace of ʿAbbās I, combining public space and private living quarters, it was later also used as lodging for foreign visitors. The garden and pleasure palace of Hašt Behešt (FIGURE 9) were originally designed to accommodate festivities. The ground plan shows an amazing similarity to the Achaemenian apadana (Ferrante, pp. 399-420), and when the four two-storied corner towers were turned into eight apartments for eight favorite ladies of the court, the building even served the apadana’s original function. Palaces, like the small, one-level loft on top of ʿĀli Qāpu, were already built in Sasanian times, like the royal pilgrims hostel at the fire temple of Taḵt-e Solaymān. But at this site upstairs rooms were added, when after 1265 the Il-khanid sultan Abaqa (d. 1282) repaired the palace (Naumann, pp. 43-46, 80-97; Huff, 2006a, p.103).

After the fall of the Safavids, the apadana typecontinued to be used for both noble palaces and private residences. A reduced and simplified variant became the model of Zand and Qajar palaces, which are linear buildings with a central ayvān between two-story lateral aisles with the private quarters (Karapétian, e.g., pp. 235-52). This variant is similar to the suggested reconstruction of Pasargadae Palace P. As earlier in the late Sasanian period, the domed main hall was again abolished, and the ayvān became the only throne hall. Its function is perfectly documented by a drawing that Eugène Flandin (1809-89; see FLANDIN AND COSTE) made of the early Qajar Golestan Palace (Kāḵ-e Golestān) in Tehran before the ayvān got its mirror revetment: women stand in a lateral upstairs gallery and look down onto the shah’s empty marble throne (Flandin and Coste, pl. 32; FIGURE 10). A modern example of the traditional apadana type is the palace of the mother of Moḥammad-Reżā Shah Pahlavi (r. 1941-79) at Saʿdābād, Tehran, completed in the early 1970s. The central hall is surrounded by two stories of rooms, and the private suites upstairs are accessible from a gallery that offers an unrestricted view into the reception hall.

A particular Persian type of mausoleum became the final resting place of members of the ruling families, and this building tradition demonstrates how deeply the basic architectural idea of the royal palace was rooted in Iranian civilization. At Bukhara, the tiny useless galleries at squinch height in the Samanid tomb (early 10th century CE; see BUKHARA v ARCHEOLOGY AND MONUMENTS) can only be explained as a symbolic reminiscence of the upstairs private area of the king’s palace (Leisten, pp. 141-42; Huff, 1999). At Marv, the Saljuq sultan Sanjar (r. 1118-57) built a mausoleum with galleries that mainly open towards the outside, and the Arab historian Ebn al-Jauwzi (1126-1200) reported that he named it the “Palace of the Next World” (Ar. dār al-āḵira; cited in Leisten, p. 202). At Solṭaniya, the tomb-mosque of the Il-Khanid sultan Uljāytu (r. 1304-16) has upstairs rooms and galleries that resemble the upper floor of a veritable royal palace (Wilber, pp. 139-41).



R. Boucharlat and others, “The Citadel of Ulug Depe and the Iron Age Archaeological Sequence in Southern Central Asia,” Iranica Antiqua 40, 2005, pp. 479-514.

R. Dyson, Jr., and M. Voigt, “A Temple at Hasanlu,” in Yeki bud, yeki nabud: Essays on the Archaeology of Iran in Honor of William M. Sumner, ed. N. Miller and K. Abdi, Los Angeles, 2003, pp. 219-36.

M. Ferrante, “Dessins et observations préliminaries pour la restauration du palace de ʿĀlī Qāpū” and “Le pavillon de Hašt Bihišt, ou la huit paradis, à Ispahan: Relevés et problèmes s’y rattachant,” in Travaux de restauration de monuments historiques en Iran, ed. G. Zander, Rome, 1968, pp. 137-206 and pp. 399-420.

E. Flandin and P. Coste, Voyage en Perse:Perse moderne - Planches, Paris, n.d.; repeatedly reprinted, occasionally counted as volume VI.

H. Fujii, ed., Preliminary Report of Excavations at Gubba and Songor: Hamrin Report 6, Tokyo 1981.

E. Galdieri, Esfahan: Ali Qapu, Rome, 1979.

D. Huff, “Qalʿa-ye Dukhtar bei Firuzabad: Ein Beitrag zur sasanidischen Palastarchitektur,” AMI NF 4, 1971, pp. 127-71.

Idem, Wohn- und Repräsentationsbereiche in der traditionellen iranischen Architektur, Diskussionen zur archäologischen Bauforschung 3, Berlin, 1978a, pp. 237-43.

Idem, Ausgrabungen auf Qal’a ye Dukhtar bei Firuzabad 1976, Archäologischer Anzeiger 11, 1978b, pp. 117-47.

Idem, “Sassanidische architektuur/Architecture sassanide,” in Splendeur des Sassanides/ Hofkunst van de Sassanieden, Bruxelles, 1993, pp. 45-61; Fr. tr. defective.

Idem, “Traditionen iranischer Palastarchitektur in vorislamischer Zeit,” in Rezeption in der islamischen Kunst, ed. B. Finster and C. Fragner, Beirut, 1999, pp. 141-60.

Idem, “Bronzezeitliche Monumentalarchitektur in Zentralasien,” in Migration und Kulturtransfer: Der Wandel vorder- und zentralasiatischer Kulturen im Umbruch vom 2. zum 1. vorchristlichen Jahrtausend, Bonn, 2001, pp. 181-97.

Idem, “From Median to Achaemenian Palace Architecture,” Iranica Antiqua 40, 2005, pp. 371-95.

Idem, “The Ilkhanid Palace at Takht-i Sulayman: Excavation Results,” in Beyond the Legacy of ·enghis Khan, ed. L. Komaroff, Leiden, 2006a, pp. 94-110.

Idem, “Die Entwicklung der Stadt- und Palasttore in Zentralasien,” in Stadttore, Bautyp und Kunstform, ed. T. Schattner and F. Valdéz Fernбndez, Mainz, 2006, pp. 325-44.

Idem, “Formation and Ideology of the Sasanian State in the Context of Archaeological Evidence,” in The Idea of Iran: III - The Sasanian Era, ed. V. S. Curtis and S. Stewart, London, 2008, pp. 31-59.

E. Kaempfer, Am Hofe des persischen Grosskönigs, 1684-1685: Das erste Buch der Amoenitates exoticae 1712, ed. W. Hinz, Leipzig, 1940.

K. Karapétian, Isfahan, New Julfa - Le case degli armeni: Una raccolta di rilevamenti architettonici, Rome, 1974.

R. Kent, Old Persian, 2nd ed., New Haven, Conn., 1953.

T. Leisten, Architektur für Tote: Bestattung in architektonischem Kontext in den Kernländern der islamischen Welt zwischen 3./9. und 6./12. Jahrhundert, Berlin, 1998.

Y. Majidzadeh, The Ancient Ozbaki site, Savoudjbolagh, Iran, Tehran, 2001.

R. Naumann, Die Ruinen von Tacht-e Suleiman und Zendan-e Suleiman und Umgebung, Berlin, 1977.

V. I. Pilipko, “Poselenie rannezheleznogo veka Gyarry-Kjariz 1,” in Turkmenistan v иpochu rannezheleznogo veka, ed. V. M. Masson and E. Atagarryev, Ashkhabad, 1984, pp. 28-58.

O. Reuther, Die Ausgrabungen der Deutschen Ktesiphon-Expedition im Winter 1928/1929, Berlin, 1930.

V. Sarianidi, Margiana and Protozoroastrism, Athens, 1998.

E. F. Schmidt, Persepolis, 3 vols., Chicago, 1953.

D. Stronach, “Excavations at Tepe Nush-i Jan 1967,” Iran 7, 1969, pp. 1-20.

Idem, Pasargadae, Oxford, 1978.

Idem, “The Apadana: A Signature of the Line of Darius I,” in De l’Indus aux Balkans: Recueil а la mémoire de Jean Deshayes, ed. J.-L. Huat and others, Paris, 1985, pp. 433-45.

Idem, “Apadana ii. Building,” EIr II, pp. 146-48.

D. Stronach and M. Roaf, “Excavations at Nush-i Jan: Part 1 - Third Interim Report,” Iran 16, 1978, pp. 1-11.

D. N. Wilber, The Architecture of Islamic Iran: The Il Khanid Period, Princeton, 1955.

T. C. Young, “Thoughts on the Architecture of Hasanlu IV,” Iranica Antiqua 6, 1966, pp. 48-71.

T. C. Young and L. D. Levine, Excavations of the Godin Project: Second Progress Report, Toronto, 1974.

(Dietrich Huff)

Originally Published: July 28, 2008

Last Updated: July 28, 2008