Amadiya/Amedi Citadel

The mesa-like outcrop on which the Amadiya/Amedi citadel was built is roughly elliptical, with the occupied area measuring little more than 1 km at its greatest extent (north-south); its elevation is uneven, with areas of the east side up to 40 m higher than the west side. Traditionally, the city was only accessible by two monumental entranceways through the medieval circuit walls: one to the east and one to the west. The latter stands near the citadel’s southwestern end. Known as the Mosul Gate (or Bahdinan Gate), it is dated by an inscription in its outer frame to the reign of Badr al-Din Lu’lu’, 1233–1259 AD (see the entry on this monument). This entrance remains accessible only by foot or pack animal, by way of an ancient staircase that passes three Parthian-era reliefs carved into the citadel’s cliff walls. 

The Eastern Gate, also known as the Zibari Gate, once gave access to the high northeastern end of the citadel plateau and was closely connected to the palace (see below). After a modern road was built up to the east side of the citadel in the 1930s, it fell out of use and was subsequently destroyed. However, the gate—and a portion of its associated footpath—was photographed in the late 19^th century, providing an impression of its appearance at this time (see the historical photograph). The bulk of the structure was built of roughly hewn stone, with an arched portal of cut ashlars apparently without relief decoration. The window on the wooden second story is similar in type to those of the palace, which was just beyond the wall; this may mean that the gate was built at the same time, probably in the early 1700s.1 However, it is possible that it was preceded by an earlier structure. The stone staircase involves similar construction methods to that on the opposite side of the citadel and is likely ancient as well. Its remnants can still be found in the area’s upper slope, along with a few of the gate’s foundation blocks.

Although much of the historical architecture was lost during various military conflicts in the 19^th and 20^th centuries, a number of important monuments are preserved. A selection of these have been documented by the MMM team during its several stays in Amadiya/Amedi. In terms of antiquities, there has been very little systematic archaeological activity conducted within the citadel. However, one ancient structure is found in the southwestern area of the city, within the grounds of an Islamic cemetery.2 Rectangular in form, it measures 17 x 29 m, with its long axis oriented northwest-southeast; its walls are carved into the bedrock up to 3 m deep (see the ground plan and section). Three roughly parallel rows of pillar bases, also carved from living rock, are arrayed inside. Given the remnants of water channels along the perimeter, it likely functioned as a cistern—at least for a time—but certain features point to additional uses. A large niche dug into the southeast wall has led to the suggestion that the building was converted into a Christian church (though depending on the date, other interpretations are possible).3 Smaller niches in the northeast and southwest walls were probably added as arcosolia graves in the 5^th–6^th centuries AD.4 

The city’s main mosque stands toward the north of the city. This is probably to be associated with the historical Great Mosque, dating originally to the Zengid period (mid-12^th–early 13^th century) but with phases of reconstruction. Its longitudinal prayer hall—in a ruined state by the 19^th century—was rebuilt in the 20th century, though a historical minaret likely dating to 16^th century survives (see the full entry on the mosque and minaret). Near the Great Mosque is a ruined structure called the Tomb of Hazana, dedicated to a Jewish holy man of antiquity. It has also been associated with the prophet Ezekiel, who was the patron of the now-destroyed synagogue once attached to the tomb, dating to the early Bahdinan period (13^th century).5 The tomb was also sacred to the city’s Muslim and Christian residents (and remains so to this day). 17^th-century visitors report that Amadiya/Amedi had a large Christian church, though this seems to have been lost by the mid-19^th century.6 A modern church, serving the city’s small remaining Christian community, lies along the western edge of the citadel.

Amadiya/Amedi’s palace, the seat of power of the city’s successive rulers, was situated on the high ground of the citadel’s northeast, very close to the Zibari Gate. The building, which grew progressively ruined from the late 19^th century, was finally built over by a school in the later decades of the 20^th century. One feature that has survived, though partially reconstructed, is known as the “Snake Gate” or “Dragon Gate”. It now appears as if it were a freestanding monument, but Henry Binder’s 19^th-century photograph indicates that it was once integral to the facade of the palace.7 The gate, 3.5–4 m wide, consists of two large blocks held together by a keystone, forming an arch. The spandrel blocks are carved with antithetical dragons with long, twining bodies. On the keystone, the talons of a bird of prey grasp the dragons by the nose. The block above it, originating from another monument and placed here in the reconstruction, shows the upper body of another raptor. The archway lacks an inscription that would allow for a precise dating. Based on the style of the relief sculpture, the gate has been dated to the 13^th century or later; thus, it may have made reference to the comparable image on the earlier Mosul Gate.8 Notably, the imagery and style are related to Seljuk reliefs in Anatolia.9  

Binder’s photograph of the palace’s facade is also the only known visual documentation of the larger building. This edifice, as it appears here, can perhaps be dated to the 18^th century, though the incorporation of the Dragon Gate suggests that it was rebuilt over an earlier structure.10 19^th-century travelers provide some information on the layout of the palace beyond its facade. Henry James Ross, who visited shortly after the city’s annexation by the Ottomans, describes it thus: “The place as usual was built in a hollow quadrangle cut up by a wall passing through the centre into two courtyards, the outer being the Divan Khanah, or public apartments, the inner the hareem for the women. The ground floor was mostly occupied by roomy stables and lodgings for servants; above, the accommodation was I thought rather scanty; however, the apartment I was ushered into was large, and had been a fair specimen of Eastern internal decoration, but it was greatly dilapidated.”11 It is clear from Ross’s and other descriptions that the innermost rooms of the palace were aligned with the citadel walls, overlooking the northern valley toward Mt. Matin.

A royal cemetery was situated in the vicinity of the palace. Two mausolea constructed of limestone blocks survive on its grounds. The earlier of the two was built for the emir Hussein Wali (r. 1534–1573). Hexagonal in plan, it features elegant calligraphy over its segmental-arched entrance. The second was built for a family member of the emir Ishmail (r. 1768–1798); it is also hexagonal and is domed above. In addition, several gravestones of other members of the ruling family are preserved here.

Travelers visiting Amadiya/Amedi during the 17^th–mid-19^th centuries report of a busy bazaar and a hammam, as well as a caravanserai on the western edge of town. Much of this urban fabric was ruined during the Bahdinan-Ottoman military conflict of the mid-19^th century. However, photographs from the late 19^th and early 20^th centuries provide images of the city’s traditional houses at this time (see Henry Binder's photograph; see Mark Sykes's photograph). These one-story habitations were built of roughly hewn stone with flat roofs of wood and thatch. Due to subsequent destruction in the later 20^th century, relatively few such traditional habitations are now to be found, but some survive, for example in the vicinity of the mosque. 

A number of traditional homes remain along the citadel walls to the north, taking advantage of the views over the northern valley (as did the palace). A building that is possibly to be identified as a residence, once two stories high and now partially collapsed, is associated locally with the Kittani family. However, the building’s original function and construction date are not known, and the house appears to have repurposed much earlier structures into its fabric. The first story is faced with fine ashlar masonry; the blocks vary in size and are set in irregular courses, creating a complex interlocking surface. The arched entrance is framed by a shallow, pointed vault, with a window above. The stonework is articulated with elegant carvings: detailed colonettes and a band of interlace along the vault’s piers, and delicate flowers in each block of the inner arch. This basic type of portal design has a long history in eastern Anatolian/northern Mesopotamian architecture, with parallels ranging from the Seljuk era to the 19^th century.12  

Most of the city fabric that is visible today was built up in the latter half of the twentieth century, though its layout follows earlier patterns. For instance, the marketplace—covered in its central stretch—cuts diagonally through the center of the city, running between the Mosul and Zibari Gates. More broadly, the neighborhoods are divided into several districts (mahalle), including Servepki to the north, Meydan in the center, and the Christian Quarter along the citadel’s western limit. In recent decades, multiple settlement clusters have also sprung up below the slopes of the acropolis.

Although not on the citadel proper, several nearby places of historical interest should be noted here. Perhaps most remarkable is the Qubahan Medrese. During the Bahdinan Emirate, this was one of Kurdistan’s leading universities, attracting students from around the Muslim world. The nearby town of Sulav, about 1 km northwest of Amadiya/Amedi on the slopes of Mt. Matin, has seen increasing touristic development. A high waterfall, noted by travelers already in the 19^th century, serves as the community’s iconic landmark; the site also offers impressive views of the citadel (see the panorama).

• 1. Warkworth (1898, 183) confirms that the upper story was composed of woodwork. On the date, see Ammann 2004/2005, n. 196.
• 2. See Bachmann 1913, 2–3.
• 3. William Ainsworth, who first made note of the structure in 1840, considered it to be a Persian fire temple (Ainsworth 1842 (2), 200); Layard (1849 (1), 161) interpreted it as a church, while Badger (1852 (1), 204) identified it as a cistern. Bachmann (1913, 3) combined the latter two interpretations, and this has been followed by many scholars discussing the structure. The idea that it functioned as a fire temple during the Sasanian period has also remained in consideration (Marf Zamua 2008, 117).
• 4. See Boehmer 1976 (including a plan of one grave, fig. 7).
• 5. See the historical photograph in Brauer 1993. Another synagogue, dedicated to Ezra and also no longer extant, was located further south.
• 6. Domenico Lanza, visiting in the 1600s, notes a great church (see the quotation in Galletti 2001, 116–117); Badger (1852 (1), 200) claims that there were no formal churches were in existence at the time of his visit and that private houses were used for worship.
• 7. Further documentation from the early 1970s shows the gate standing, attached to the ruins of the proximal walls of the palace; it was moved by the 1990s (see the photographs in Gierlichs 1998).
• 8. For a discussion of the date and an interpretation of the imagery, see Gierlichs 1998.
• 9. For Seljuk comparanda, see Katharina Otto-Dorn, “Figural Stone Reliefs on Seljuk Sacred Architecture in Anatolia,” Kunst des Orients 12, H 1/2 (1978–1979), 103–149. See also the sources in al-Janabi 1982, n. 411.
• 10. Local oral tradition holds that there was an inscription on the palace attributing it to the Bahdinan emir Osman (r. 1700–1702); see Ammann 2004/2005, n. 196.
• 11. Ross 1902, 107–109. Layard (1848 (1), 156) also provides a brief description.
• 12. For parallels, see, e.g., the portal of the madrassa of Taskinpasa, 14th century; the madrassa of Mustafapaşa, 19th century.