University of Chicago
The idea of Jerusalem was fundamental in the earliest formulation of Islam, a part of the Abrahamic faith begun in the Hijaz during the early 7th century c. e. The Muslim conquests of Byzantine territory brought the Arab population to the Holy City (al-Quds) and possession of the Temple Mount (the Haram al-Sharif). The religious importance of this occupation was expressed in the building of a mosque and a shrine. While economic and administrative aspects are elements that define the foundation of an Islamic city, it was these ritual aspects that made Jerusalem the first and most important Islamic city after Medina.
The urban fabric of Jerusalem in the 6th century c. e. had been altered by Justinian to accommodate great numbers of pilgrims, as part of its ritual identity within the Byzantine Empire. The form of the Late Antique city is readily recognized in the mosaic famed as the Madaba Map, usually dated ca. 580 c. e. (Donner and Cuppers 1977: Abb. VI, VII).232 This map presents a startlingly clear urban conception (fig. 1), because it was intended as a means of instruction for pilgrims and, as such, the Church of the Holy Sepulchre becomes the focal element. The plan of the city shows the two colonnaded streets from the Damascus Gate on the north, one to the pilgrim basilica of Nea Haghia Maria and the other to the Episcopal seat of the Zion Church, each building marked by their tile roofs. If one superimposes these visible features upon a plan of the city (fig. 2), one sees an idealized circle focused on the Holy Sepulchre and an extension to the south to include the two principal churches.
The walls of Jerusalem in the Madaba Map have towers and three visible gates: the western or Citadel Gate, the northern Damascus Gate, and the eastern Jericho Gate. There is no indication of a south gate, though damage to the mosaic may have obscured this feature; more importantly, there is no indication of the eastern portion of the city and especially the Temple Mount, which was consciously ignored during the Byzantine period. It seems fair to conclude from this map that the eastern and southeastern portions of Jerusalem were unimportant in the Byzantine period and probably more so during the subsequent Sasanian occupation.
Scholars are taking increasing notice of the catastrophic fall of Jerusalem to the Persians in 614 c. e. The heavy cavalry of Khosrow II, the Sasanian emperor, swept across Syria, the Levant, and Egypt, not unlike the Crusader knights about 500 years later. Like these later counterparts, urban resistance provoked destruction and
Fig. 1. Madaba map (after Peters 1985).
Massacres; layers of burnt destruction, which archaeologists usually attributed to the Muslim conquest, are now increasingly blamed on the Persians.233 M. Ben-Dov (1982: 239-41; 2002; 162) has suggested that the obliteration of the Nea basilica may have been a special objective of Jewish auxiliaries, who found offense in its building materials (spolia from the nearby ruins of the Temple). This attack and occupation may have been a destructive and final break with the Classical city; it was a disruption that lasted some five years until the treaty with the conquering Muslim forces (Shahid 1989; 208-43). One may expect that during the retrenchment and repopulation of Jerusalem under Heraclius, prime attention would have been paid to the western sections and their churches.
Historical periodization underwent a major break with the Muslim conquest of Jerusalem in 638 c. e., signaling the end of Antiquity. The negotiated capitulation to the Muslim army and Caliph 'Umar, with its guarantees of church and population in Jerusalem, marks a contrast to the Persian experience (Grabar 1980; 340; Peters; 1985; 185-86). There is the story, perhaps in part a pious fiction, of the archbishop Sophronius guiding the caliph over the abandoned Temple Mount, which was covered with trash and debris. It appears that Muslim belief in the ritual importance of this site and long-standing Christian disinterest made a real estate deal easily
Achieved. What follows will be an attempt to reconstruct what the early Muslims did with this and other abandoned property.
Jerusalem, an Arab City
The urban plan reconstructed from the Madaba Map allows a preliminary understanding of the changes introduced in the early Islamic period. Cartography is important for evaluations of urbanism in this period. If one begins with D. Bahat's
Fig. 3. Jerusalem map (after Bahat 1983: 49).
Popular Carta map (1983: 49), one sees the attentions paid to the Haram al-Sharif, the contemporaneous Jewish quarter southwest of the Haram, and the Mosque of Omar sitting in a wasteland vaguely labeled “Christians” (fig. 3). The early Islamic city fairs better in Bahat's map (1996: 81), which acknowledges a continuity in streets and an infilling of principal buildings (fig. 4). Symptomatic of this change is the complex south and west of the Haram al-Sharif, newly excavated and identified as early Islamic (see below). In fact, it is not too much to imagine new architecture spread across much of the cityscape within its walls. Some of this urban development from the time of 'Umar's visit (636 c. e.) until the 10th century may be deduced from the descriptions of al-Muqaddasi (ca. 985 c. e.).
h
Fig. 4. Jerusalem map (after Bahat 1996: 81)
Ahmad ibn Muhammad al-Muqaddasi wrote a masterful geography of the entire Islamic world entitled Ahsan al-Taqasim fi ma'rifat al-Aqalim (“The Best Divisions for Knowledge of the Regions”).234 This innovative scholar devised hierarchically organized systems for each province and for each city and gives a methodological and detailed account of its features. One part of his method, perhaps a mnemonic
Fig. 5. Plan of Jerusalem, according to Muqaddasi.
Device, was to sequence his presentation of toponyms in a counterclockwise fashion, beginning (usually) in the southwest (fig. 5).235 He applied this system to Jerusalem (al-Quds), which was his native city. He lists the gates of Jerusalem as: Bab Sihyun (1, near the Zion church), Bab al-Tih (or al-Nih, 2, associated with the Nea Church), Bab al-Balat (3), Bab Jubb Armiya (4), and then Bab Silwan (5). He continues the list with the gates previously seen on the Madaba Map: Bab Ariha (or Jericho Gate on the east, 6), the Bab al-Amud (or Gate of the Pillar, seen in the mosaic as the north gate, 7) and finally the Bab Mihrab Daud (or Citadel Gate on the west, 8). A curious feature of this list is immediately apparent. While the east, north, and west have only one gate each, there are five gates on the south.
This concentration of gates in the southern part of the city has elicited speculations on their locations by Tsafrir (1977: 152-61) and Bahat (1986: 429-35).236 Their debate focuses on whether the gates reverted to the inner wall of the Roman period or expanded again to the outer (Eudocian) wall in the early Islamic period. Although repair of the walls is reported under the rule of Abd al-Malik, attention to urban walls would be unusual until the Abbasid period at the earliest (Magness 1991: 208-17; Whitcomb 2006: 61-74).237 The “iron gates” mentioned by Muqad-dasi raise the question of whether these were literal structures or conceptual urban limits—that is, protected localities. The accepted location of the Jewish settlement south of the Islamic administrative quarter on the Ophel Hill would be within this urban space of Jerusalem in the Early Islamic period.238 This distribution of “gates” in the south strongly indicates that the Early Islamic city developed this abandoned sector of the city during the first century of Islamic rule (Bahat 1996: 83).
The importance of this part of the city has taken on a more vivid meaning since the excavations, between 1968 and 1978, of an “administrative compound” consisting of “six large palaces [with] wall paintings, stucco, and vast dimensions,” in the words of Rosen-Ayalon (fig. 6; 1989: 8). She has shown that this palatial complex fits with the plan of the Haram al-Sharif and original Aqsa Mosque; its date should be at least to the time of Abd al-Malik and more likely to Mu'awiya (ca. 660, see below). One of these palaces has been described as the Dar al-Imara (the governor's residence), and another the diwan, a fiscal bureau and treasury. Ben-Dov has described a third element, the bath (hammam) with hypocausts, a caldarium of over 100 square meters, and a diwan (claimed to be similar to Kh. al-Mafjar).239
Fig. 6. Excavations south of Haram, 1977 (Ben-Dov 1982: 331).
With the Aqsa Mosque, one sees the four institutions typical of an Islamic city of the 8th-9th centuries.
More recently, J. Magness (2010) has offered a reassessment of the archaeological history of these buildings—more valuable, in the absence of any final reports, due
Fig. 7. Arab urban model (after Whitcomb 2007: fig. 2).
To the inclusion of post-1995 excavations by Y. Baruch and R. Reich (1999). She suggests a later Umayyad construction based on coinage found, perhaps during the reign of Hisham, and continuation of finished buildings through the 9th century Abbasid period.240 She notes (1992: 67-74) the suggestion of Reich and Baruch that the buildings fronted the Cardo, which may have “separated the administrative or governmental quarter from the civilian part of the town [to the west].” The key building for this interpretation is Building III, west of the largest building described
Fig. 8. Arab urban model applied to Jerusalem.
As the Dar al-Imara (Building II) and the axial street separating these two structures. The identification of this building as the diwan, bearing the descriptive name of al-Balat, follows a comparative study of early Islamic urban foundations.241 242 If this hypothesis is accurate, the Balat of Jerusalem, proximate to the Bab al-Balat, gave its name to the Islamic urban center. 11
Ben-Dov notes the resemblance of these buildings south of the Haram to the qusur or 'desert castles' of Umayyad Bilad al-Sham (1971: 38). The conjunction of these large buildings with a bath suggests a locational pattern indicative of an urban, or in many cases a proto-urban, settlement type (Whitcomb 2007). If one compares Khirbat al-Mafjar, Qasr al-Hayr al-Sharqi, and 'Anjar (fig. 7), one will notice the bath within each of these complexes bears a constant relationship in distance and direction to the 'palace' (or administrative center). The bathhouse is located about 50 m to the north. The 'palace' is proximate to the primary mosque and, further, this complex of bath/palace/mosque is located in the eastern half of the city. The western half seems to be primarily residential. A pre-Islamic and perhaps Arab aspect of this arrangement may be revealed in the very similar plan of the Nabatean town of Kurnub (Mampsis) in the Negev. There would seem to be a lo-
Fig. 9. Reconstruction of Haram and surrounding buildings (after Bahat 1989: 82-83).
Cational consistency with the north gate as well. This arrangement, and no doubt other structural elements, suggest that the planning of these Umayyad complexes was not random but in accord with a consistent plan, a conceptual model for urban organization. While the 'desert castles' seem to be incipient urban entities, Jerusalem reveals the completely developed manifestation of an early Islamic city (Fig. 8).
Jerusalem is an exceptional phenomenon due to the massive platform, the site of the Temple and now the Haram al-Sharif. The Dome of the Rock (Qubbat al-Sakhra) stands as a continuing symbol of Jerusalem (fig. 10). This structure is a focal point for the city, architecturally based on the octagonal martyrium, popularized in the Byzantine period. What may be memorialized on the Haram al-Sharif is a vestige of Mt. Moriah, a ritual high place, a concept not unfamiliar to those coming from Arabia. For the Muslims, it was (and is) a sacred space, a haram; the qubba may commemorate the Temple and raise an awareness of apocalyptic expectations (Rosen-Ayalon 1989). S. D. Goitein (1982: n. 22) seems to have been the first to suggest that Mu'awiya, with his special interest in Jerusalem, was the originator of the Dome of the Rock. Further, Mu'awiya may have found inspiration for this shrine in Caesarea (Qaysariya), the capital of Byzantine Palestine.
Only 10 years had passed after the peaceful evacuation of the Persians from Caesarea before Mu'awiya seems to have inherited not only the physical administrative apparatus but much of the bureaucratic system as governor (ca. 640 c. e.) (Whitcomb, in press). As Magness has shown (1994), the Octagonal Building on the
Fig. 10. The Dome of the Rock.
Fig. 11. Octagonal church at Caesarea (Holum 1999: fig. 13).
Fig. 12. Comparison of octagonal buildings at same scale.
Temple Platform dominated the skyline of early Islamic Qaysariya well into the 8th century, if not later (fig. 11). A comparison of their plans suggests that this complex may have inspired Mu'awiya to reproduce such a building on the platform of the Haram al-Sharif (fig. 12).243 This antecedent at Caesarea may corroborate the forceful, if speculative, argument by O. Grabar (1990: 156; 1996: 115) that the concept of the Dome of the Rock may be attributed to Mu'awiya, while the building was then finished by his successor Abd al-Malik in 692 c. e. (fig. 6). It may be coincidental—but of great interest—that south of the Temple Platform of Caesarea were
Fig. 13. Jerusalem, axial grid and communities.
Buildings described as a bath, a palace, and a tax office (skrinion, functionally not unlike the diwan).244
The political role of Jerusalem for Mu'awiya seems to have been an exceptional phase, for the city was only briefly, if ever, the political capital for Filastin. Rather, this city was, and for many still is, the apex of a sacral hierarchy of holy places (Wheatley 2001: 92-94). Its political balance was Ramla; this was a situation not unlike the dichotomy of Mecca and Madina, the political and sacral centers in the Hijaz. Jerusalem may have received more careful planning consideration as a reflection of cosmological order. The Aqsa Mosque was located south of this central maidan of the city and in axial orientation with the Dome and street below the platform, as Myriam Rosen-Ayalon first noticed. Alternatively, the Dome of the Rock may be taken as a nodal center (a sort of tetrapylon) marking the axial juncture to the south and west, what might have been principal directions within the new Islamic city (fig. 13). The Haram with its two structures, the Dome of the Rock and Aqsa Mosque, may be seen as one ritual complex, and in a wider context, elements in a wider plan of a ritual city. That the tetrapylon became a shrine may indicate the special character of this city, location of one of the three mosques important for Islam and an omphalos of the universe (Wheatley 2001: 295-8).
Conclusions
Goitein (1982: 175) has described the Arab imigration into early Islamic Jerusalem as a Yemenite colony, and at the same time states that Jerusalem never became an “Arab city.” On the contrary, the patterns delineated here suggest that just such an urban identity was imposed on the southern and eastern portions of the ancient city. Grabar (1996: 157-61) suggests that Mu'awiya's impulse might stem from an Arabian context; as he puts it, “much too easily we have all assumed that Arabia was forgotten once the visual riches of the conquered lands became available,” yet this cultural background was not confined to the oases and tribes but should include the cities of south Arabia.
The “Arab city” was a settlement pattern of some antiquity in South Arabia, a pattern that introduces important complications to the traditional conception of tribal settlement in Arabia (Whitcomb 1995b: 38-51). One may suggest that there was an Arabian concept of urbanism that proved appropriate as the material referent for the theocratic state, which began in the Hijaz in the early 7 th century and continued as the evolving expression of an Islamic settlement. Conrad (n. d.: 22) has observed, with apparent surprise, that early Islamic foundations “differ from the traditional Arabian tribal settlement. . . [in that] implicit in all of [these settlements] was the notion of a genuine town center, already a major departure from the traditional norm.” The model advanced herein emphasizes the primacy of the mosque and palace complex in new Islamic foundations and that this nucleation of a “genuine town center” is an innovation characteristic of the incipient Islamic state.
Fig. 14. Aerial photo of Jerusalem from the south (Peters 1985).
The plan of Jerusalem in the early Islamic period is a twin city (fig. 13) in that the Christian town with its religious foci remained inviolate on the west and the eastern and southeastern sections formed a new Muslim city. 245 New religious structures dominated the high platform of the Haram al-Sharif, administrative facilities were clustered in large buildings south of the Haram, and a bath was located along the old Cardo, which may have formed the primary locus of economic activity (i. e., the suq). Muslim residential areas may have been located to the west of the Cardo, spreading perhaps to the south. Located literally under the protection of the governmental center was probably the Jewish community, returned to the ancient City of David (fig. 14).
The evidence for Jerusalem in the early Islamic period, whether or not the specific model for the Arabian city finds acceptance, indicates that the Aqsa Mosque and the Dome of the Rock were elements in a larger conception of an Islamic city. The city was exceptional in its importance for this new religion and subjected to a new experimental development. This religious community strove for a structural setting to express the living of this new faith and to create relationships with past and contemporaneous religious communities in this Holy Land.
References
Bahat, D.
1983 Carta's Historical Atlas of Jerusalem: An Illustrated Survey. Jerusalem: Carta.
1986 Les portes de Jerusalem selon Mukaddasi: Nouvelle identification. Revue biblique 93: 429-35.
1996 The Illustrated Atlas of Jerusalem. Jerusalem: Carta.
Baruch, Y.
2002 Jerusalem: the Old City. Excavations and Surveys in Israel 114, 75*-76* [89-92].
Baruch, Y., and Reich, R.
1999 Renewed Excavations at the Umayyad Building III. Pp. 128-40 in New Studies on Jerusalem: Proceedings of the Fifth Conference, Dec. 23, 1999, ed. A. Faust and E. Baruch. Ramat Gan: Bar-Ilan University Press.
Ben-Dov, M.
1971 The Omayyad Structures near the Temple Mount. Pp. 37-44 in The Excavations in the Old City of Jerusalem near the Temple Mount: Second Preliminary Report 1969-70 Seasons, ed. B. Mazar. Jerusalem.
1976 The Area South of the Temple Mount in the Early Islamic Period. Pp. 97-101 in Jerusalem Revealed: Archaeology in the Holy City 1968-1974, ed. Y. Yadin. New Haven: Yale University Press.
1982 In the Shadow of the Temple: The Discovery of Ancient Jerusalem. Jerusalem: Keter.
2002 Historical Atlas of Jerusalem. New York: Continuum.
Conrad, L. I.
N. d. The Early Arab Urban Foundations in Iraq and Egypt: Implications for Trade and Exchange, in The Byzantine and Early Islamic Near East V: Trade and Exchange, ed. L. I. Conrad and G. R. D. King. (forthcoming).
Donner, H., and Cuppers, H.
1977 Die Mosaikkarte von Madeba, pt. 1. Wiesbaden: Otto Harrassowitz.
Foss, C.
2003 Persians in the Roman East (602-630 a. d.). Journal of the Royal Asiatic Society 3: 149-70.
Goitein, S. D.
1982 Jerusalem in the Arab period (638-1099). Pp. 168-96 in vol. 2 of The Jerusalem Cathedra, ed. L. I. Levine. Detroit: Wayne State University Press
Grabar, O.
1980 al-Kuds: B. Monuments, The Encyclopedia of Islam. 2nd edition. Leiden: Brill.
1990 The Meaning of the Dome of the Rock. Pp. 151-63 in Studies in Arab History: The Antonius lectures, 1978-87, ed. D. Hopwood. Oxford: Clarendon.
1996 The Shape of the Holy: Early Islamic Jerusalem. Princeton: Princeton University Press.
Luz, N.
1997 Construction of an Islamic City in Palestine. The Case of Umayyad al-Ramla. Journal of the Royal Asiatic Society 7: 27-54.
Magness, J.
1991 The Walls of Jerusalem in the Early Islamic Period. Biblical Archaeologist 54: 208-17.
1992a Reexamination of the Archaeological Evidence for the Sasanian Persian Destruction of the Tyropoeon Valley. Bulletin of the American Schools of Oriental Research 287: 67-74.
1994
The Pottery from Area V/4 at Caesarea. Annual of the American Schools of Oriental Research 52, 133-45.
2010
Early Islamic Urbanism and Building Activity in Jerusalem and at Hammath Gader. Pp. 147-63 Money, Exchange and the Economy in the First Century of Islam: An Interdisciplinary Review of Current Debates, ed. J. Haldon. Burlington, VT: Ashgate.
Al-Muqaddasi, Muhammad ibn Ahmad
1906 Kitab Ahsan al-Taqasim fi Ma'rifat al-Aqalim. M. de Goeje, ed. Bibliotheca Geogra-phorum Arabicorum, vol. 3, 2nd ed. Leiden: Brill.
1994 The Best Divisions for Knowledge of the Regions: A Translation of Ahsan al-Taqasim fi ma'rifat al-aqalim. B. A. Collins and M. H. al-Tai. Reading: Garnet.
Peters, F. E.
1985 Jerusalem: The Holy city in the Eyes of Chroniclers, Visitors, Pilgrims, and Prophets from the Days of Abraham to the Beginnings of Modern Times. Princeton: Princeton University Press.
Rosen-Ayalon, M.
1989 The Early Islamic Monuments of al-Haram al-Sharif: An Iconographic Study. Qedem 28. Jerusalem: Hebrew University.
Shahid, I.
1989 Heraclius and the Theme System: Further Observations. Byzantion 59: 208-43.
Tsafrir, Y.
1977 Muqaddasi's Gates of Jerusalem: A new Identification Based on Byzantine Sources. Israel Exploration Journal 27: 152-61.
Wheatley, P.
2001 The Places Where Men Pray Together: Cities in Islamic lands, Seventh through the Tenth Centuries. Chicago: University of Chicago Press.
Whitcomb, D.
1995a Islam and the Socio-Cultural Transition of Palestine, Early Islamic Period (6381099 c. e.). Pp. 488-501 in The Archaeology of Society in the Holy Land, ed. T. Levy. London: Pinter.
1995b
2006
Urbanism in Arabia. Arabian Archaeology and Epigraphy 7: 38 -51.
The Walls of Early Islamic Ayla: Defence or Symbol? Pp. 61-74 in Muslim Military Architecture in Greater Syria, from the Coming of Islam to the Ottoman period, ed. H. Kennedy. Leiden: Brill.
2007
An Urban Structure for the Early Islamic City: An Archaeological Hypothesis. Pp. 15-26 in Cities in the Pre-modern Islamic World: The Urban Impact of Religion, State and Society, ed. A. K. Bennison and A. L. Gascoigne. London: SOAS/Routledge. Qaysarea as an Early Islamic Settlement. Pp. 65-82 in Shaping the Middle East: Christians, Jews, and Muslims in an Age of Transition, ca. 400-800 c. e, ed. K. G. Holum and H. Lapin. Bethesda, MD: University Press of Maryland.
World History
