KANGAVAR (Kengavar, Kangāvar), a town in the easternmost part of Kermanshah Province, on the modern road from Hamadan to Kermanshah, identical with a trace of the Silk Road, located at the distance of about 75 km from Hamadan and 96 km from Kermanshah. Isidorus of Charax in the first century CE referred to it as Congobar and mentioned a temple of Anāhitā (Anaitis) there (Isidorus of Charax, nos. 6, 7; Ziegler and Sontheimer, col. 327; see ANĀHĪD). Early Muslim geographers referred to the site as a small place with a mosque (menbar) and a columned building made of rocks that was standing on a platform; they called it Qaṣr-al-Loṣus, or Robbers’ Castle (Eṣṭaḵri, pp. 195, 197; Ebn Ḥawqal, p. 359, tr. Kramers and Wiet, p. 353; Moqaddasi, pp. 29, 393, 401; Ebn Rosta, p. 167; Ebn al-Faqih, p. 267). According to Ṭabari (I, p. 2649), the town received this appellation when, at the time of the Arab conquest, people here stole the pack animals of the Arab army stationed there on its way to Nehāvand. Eugène Flandin and Pascal-Xavier Coste identified the ruins on a hillside at the eastern edge of the settlement of Kangavar as the Temple of Anāhitā and gave the first detailed drawings and plans of the site (Flandin and Coste, p. 411).

Up to the first half of the 20th century, scholars pleaded for an ideal reconstruction of a columned temple on the “Greek” ground plan standing on a high platform or terrace with a monumental staircase on the south side of the terrace (Ghirshman, p. 24, plate. 31). This temple was dated to the 2nd or 1st century CE. During the 19th century and the beginning of the 20th century, the ruins were the object of vast devastation as the site was used as a quarry to get building material for the expanding settlement of modern Kangavar. It has been excavated by Persian archeologists since 1968, who have tried to prevent further destruction and to get more precise knowledge of the ruins. So far a general plan of the complex can be drawn, but still it is not sufficient for determining the function and shape of the terrace and the buildings that stood on the top of it (FIGURE 1; see Kāmbaḵš-e Fard, 1972, pp. 2-12; Azarnoush, pp. 69-94).

The natural hill of Kangavar is framed by a roughly quadrilateral massive platform built of unworked, large rubble stones, smaller stones, and gypsum mortar (PLATE I, PLATE II). The platform measures 224 m on the west side and 209 m on the south side, and its outer walls are 18 m thick (Fig. 1). The northeastern corner of the platform is not clear; perhaps a staircase on the eastern side of the 18-m thick wall has to be supposed. In the middle of the platform stands a 93-m long and 9.30-m wide central construction of rubble stone with mortar, and north of it there are further walls of so far unidentified function and date. Remains of an Islamic settlement were excavated in the northern part of the platform; some lime and gypsum-kilns in the southern and southeastern part destroyed parts of the stone platform, especially the southern side with the staircases. A Parthian cemetery was excavated in the hillside east of the platform, Parthian and Sasanian pottery was collected on the platform, and Islamic pottery was found in the excavated houses upon the northern part of the platform (Kāmbaḵš-e Fard, 1972). On the northwestern corner of the platform stands a mosque and an emāmzāda, partly constructed in the Il-khanid and Safavid periods (Azarnoush, p. 72).

On the south side of the Kangavar complex the platform of a possible height of around 20 m is reached by two staircases, beginning in about 30 m distance from the southwest and southeast corners of the platform (FIGURE 1, PLATE II, PLATE III, PLATE IV). The staircases are 4.18 m (western) and 4.08 to 4.15 m (eastern) wide, and the steps measure 12 to 15 cm in height and are 27 to 30 cm deep (Azarnoush, pp. 69-94, pl. 13).

The cornice along the top of the platform has the profile of a cymation and shows light Hellenistic influence (Azarnoush, pp. 69-94, fig. 6, pls. 14.1, 15.). On the cornice stood, at a distance of about 2.90 m from plinth to plinth, a row of columns, six of which are still in situ near the northwestern corner of the platform. The column base is 65-66 cm high, the column shaft 234-36 cm high, and the capital 53-54 cm, for a total height of 354 cm. Arches, spanning from capital to capital, formed a colonnade, possibly closed by a battlement. The colonnade is reconstructed to a height of about 8 m (Azarnoush, figs. 6-9 and pls. 14.2, 16-18).

The construction and position of the two staircases are reminiscent of the Achaemnenid staircase ramps of Persepolis and of Sasanian staircases, like the one at the ʿEmārat-e Ḵosrow in Qaṣr-e Širin (Kleiss). The construction of the walls of the platform and staircases at Kangavar is much less careful than that of Persepolis and is similar to that of Qaṣr-e Širin. The idea of constructing  the staircase ramps in the Achemenid manner is doubtless an Iranian inheritance, mixed with a slight Hellenistic-Roman influence concerning the details of the cornice and the pseudo-Roman colonnade.

It is questionable whether the Temple of Anāhitā described by Isidorus of Charax is identical with the ruins of Kangavar. Isidorus evidently described a different, earlier temple of the first century CE, either somewhere else in the region of Congobar (Kangavar) or at the location of the later platform. The latter, according to the results of the excavation, seems to have been built in Sasanian times, perhaps under Ḵosrow II Parvēz (r. 590-628; Azarnoush, p. 85; cf. Ebn Rosta, p. 167; Nozhat al-qolub, ed. and tr. Le Strange, p. 108, tr. p. 107).

The function of the platform of Kangavar is not clear yet, and further excavations are necessary to throw light on it. The older reconstructions on the terrace of a “Greek Temple” surrounded by columned halls (Ghirshman, pl. 31), cannot yet be confirmed as such, because there is not enough archeological evidence for a temple-like building.

The suggestion that it is a palace-like building constructed by Ḵosrow II Parvēz at the end of Sasanian period sounds convincing, with respect to the careless construction of the platform and its stone setting, the mason marks, and the details of columns, bases, capitals, and the cornice; but there are no traces (ground plan) of a palace architecture. So this idea also remains pure theory, like other ideas concerning the function of the Kangavar complex, for instance an open-air sanctuary on the terrace overlooking the plain of Kangavar to the south of the platform. Sanctuary terraces have been used in nearly all periods of the history of Iranian architecture (Kleiss, pp. 227-68). Until detailed further excavations are carried out, no definite judgements may be declared on the function of Kangavar platform; one can only date it to the late Sasanian time.



Massoud Azarnoush, “Excavations at Kangavar,” AMI 14, 1981, pp. 69-94, pls. 12-19.

Eugène Flandin and Pascal-Xavier Coste, Voyage en Perse, Paris, 1851, p. 411.

Roman Ghirshman, Iran: Parthes et Sassanides, tr. as Iran: Parther und Sasaniden, Munich, 1962, p. 24, pl. 31.

Isidorus of Charax, Mansiones parthicae, ed. and tr. with commentary Wilfred H. Schoff as Parthian Stations:An Account of the Overland Trade Route between the Levant and India in the First Century, Philadelphia, 1914, p. 7, no. 6.

Sayf-Allāh KāmbaḴš-e Fard (Kambakhsh Fard), “Kāveš dar maʿbad-e Ānāhitā (Kangāvar),” Barrasihā-ye tāriḵi 3/6, 1968, pp. 11-46.

Idem, “Kāvešhā-ye ʿelmi dar Kangāvar, Maʿbad-e Ānāhitā,” Bāstān-šenāsi wa honar-e Irān 9-10, 1972, pp. 2-24; French version as “Fouilles archéologiques à Kangavar: le temple d’Anahita,” ibid., pp. 2-6.

Idem, “Kavešhā-ye ʿelmi dar Maʿbad-e Ānāhitā, Kangāvar,” in Gozārešhā-ye dovvomin majmaʿ-e sālāna-ye kāvešhā wa pažuhešhā-ye bāstān-šenāsi dar Irān/Proceedings of the 2nd Annual Symposium on Archaeological Research in Iran, Tehran, 1974, pp. 1-20.

Idem, “Kangavār [sic] (Plates VI-VII),” Iran 11, 1973, pp. 196-97, pls. VIb and VIla.

Wolfram Kleiss, “Terrassenanlagen in der iranischen Architektur,” Archäologische Mitteilungen aus Iran und Turan 30, 1998, pp. 227-68.

Konrat Ziegler and Walther Sontheimer, Der Kleine Pauly: Lexikon der Antike, Munich, 1979, p. 327.

(Wolfram Kleiss)

Originally Published: December 15, 2010

Last Updated: April 20, 2012

This article is available in print.
Vol. XV, Fasc. 5, pp. 496-497