WATER AND RECENT EXCAVATIONS AT AQABA
By Donald S. Whitcomb, Research Associate (Associate Professor),
The Oriental Institute and the Department of Near Eastern Languages and Civilizations
The University of Chicago
(This article originally appeared in The Oriental Institute News and Notes, No. 141, Spring 1994, and is made available electronically with the permission of the editor.)
The recent excavations at Aqaba, from November 1st through December 23rd, 1993, marked the sixth season at this early Islamic port. The goal of the excavations continues to be the multifaceted exploration of the history of this site, from the earliest Islamic period ca. 650 A.D. through the Fatimid period (abandoned at the end of the eleventh century). This year the actual process of digging brought forcibly to our attention the variety of interactions between the site and water.
Water is one of the key factors for settlement in most parts of the Middle East. An examination of the occurrence of groundwater around the Gulf of Aqaba points to the limitations on settlement and opportunities for development. The subterranean flow of water from the Wadi Araba in the north swings to the east, beneath the modern town of Aqaba; this leaves its sister city, Eilat, reliant on desalinized seawater. The direction of this groundwater has determined the location of ancient settlement-Bronze Age, Iron Age, Nabataean, and Byzantine sites are found only on the eastern, Jordanian side of the gulf. The abundant groundwater here has meant that the yearly Hajj caravan, bearing thousands of Muslim pilgrims from North Africa, Egypt, and Palestine, found a refreshing rest in Aqaba, or Ayla as it was then called.
Groundwater, somewhat brackish but potable, may be scooped from the sands right on Aqaba's beaches. On the archaeological site, it lies about 4.0 m below surface. We first encountered the groundwater in 1987 in front of the Egyptian Gate; water covered the threshold of the gate, just as Guillermo Algaze uncovered it. Last year, we reached a very early floor next to the Syrian Gate; several paving stones were removed and a quantity of sherds obtained before the water came rushing into the hole. This year we intended to explore the Large Enclosure, the largest structure known on the site and located over 4.0 m above the water.
The Large Enclosure was first excavated in 1987 and touched upon in 1992, as reported in the Oriental Institute's Annual Reports for those years. This preliminary work indicated a special structure with thick, well-constructed exterior walls, and a pier going down very deep. The upper stratum was a thick gravel layer. Most importantly, several column drums were found, apparently standing upright in situ. In 1992, the discovery of the Syrian Street and its stratification confirmed the suspicion that the Large Enclosure was built in the early Abbasid period (phase B, 750-850 A.D.). This structure was part of the reconstruction of the city after the 748 earthquake, the building of which encroached on one of the axial streets of the original city. Alteration of an urban street design is, and was, extremely rare; this suggested that the Large Enclosure was a very important structure that needed to be enlarged for the Abbasid city. Venturing into the realm of hypothesis, it was suggested that this might be the Congregational Mosque originally built by Uthman ibn Affan about 650. Such a find would be an important addition to the history of Ayla and indeed to early Islamic architecture.
During the first weeks of November 1993 we succeeded in revealing a peristyle courtyard. The arcade was supported by a complex configuration of columns and plastered piers, representing two phases of building. Unfortunately, all that remained was the basal elements and the foundations, around which were multiple layers of gravel floors, just beneath the modern surface. The gravel contained layers of plaster that suggested plaster runoff from walls and ceiling more than actual plaster flooring. A trench was placed next to one of the foundation piers to examine its depth, attendant stratification, and dating. Beneath the floors the pier descended one, two, three, four meters-into the water table. Just above the water was the foundation of a building made of red and white sandstone, which had turned to sand-mud. This early structure lined up with the edge of the original axial street. Between this early building and the late floors was 3.5 m of intentional deposition, soil fill with artifacts of the Umayyad period (phase A, 650-750 A.D.).
As the excavations extended to the east, we encountered further piers and round foundations, both increasingly associated with running foundations of limited depth (generally 1.5-2.0 m deep). These foundations were in turn associated with earlier walls of mudbrick as well as stone. Two students, Ra'd El-Shara and Clemens Reichel, mastered the difficult detection of mudbrick and robbed out stone walls. Some of the extant stone walls had been used as foundations for the Abbasid Large Enclosure, suggesting there existed an earlier form of this building. All this made for an increasingly complex architectural history, one in which structural elements were deeply buried and revealed only as isolated fragments. From this frustration came an inspiration, to extend the excavations into the wadi.
The wadi is a typical feature of the region, a wide flood channel; this channel, cutting across the ruins of Ayla, was first thought to be excavated by the British army in 1942. More recently, a hypothesis has been advanced (Oriental Institute Annual Report 1992-93, p. 19) that the wadi represents a geological fault, most likely a result of the earthquake of 1068. In any case, it was assumed that archaeological remains would have been removed long ago. Despite this, two trial trenches were placed in the bed of the wadi-both revealed substantial walls, and even a standing column, beneath about 1.0 m of silt. This encouraged us to attempt to follow the city wall across the wadi bed, hoping to find a disjuncture (lateral movement or sudden subsidence) that would confirm the existence of a fault. Impatience proved disastrous: we introduced a bulldozer to scrape off the top meter of silt; this worked well until we approached the bridge. As the site supervisor and Egyptology student John Nolan shouted a memorable "shilu (remove it!)," the bulldozer smashed a pressurized sewer pipe. Needless to say, this occurred on a Thursday afternoon and sewage ran through the wadi over the weekend. In developmental terms, this was a transfer of technology problem; while one might say this to a hundred Egyptian fellahin with no ill effect, one doesn't use such language to a bulldozer operator.
Toward the end of the excavations, when memory (and the smell) of this misfortune had dissipated, we returned to the wadi. This time we cleaned the bank of the wadi and found major walls in this section; we followed these walls across the wadi bed, almost to the eastern bank. This trench was 2.0-3.0 m wide and contained substantial architectural fragments, with disjunctions that might have been evidence of subsidence (faulting?). The area was cleaned for planning and section drawing, but the risk of working in a wadi became all too clear that evening-we had horrible, heavy rains. The Corniche Road was knee-deep and the wadi was a raging torrent of waters climbing the banks toward our trenches. The next day, staff and workmen all huddled dispirited in the wet morning chill. Our surveyor, Hugh Barnes, quickly laid out a grid on the highest remaining portions of the Large Enclosure; everyone pitched in and work continued-the soil was wetted only about 10 cm deep due to the rapid runoff typical of desertic conditions. In fact, rain is rare in Aqaba and makes no contribution to agriculture other than generally replenishing the groundwater (and flooding the improvident).
Thus water impinged on our excavations in three forms: groundwater, rainfall, and water disposal (a piped form). There is a fourth type of water source in Aqaba, manifest on the beaches less than 100 m away from the trenches. The seashore was for us a source of pleasant evening walks and occasional swimming. Always present offshore were numerous large freighters, reminding us of the commercial role of the port of Aqaba. The Red Sea was a route of trade and cultural connections in the early Islamic period as well. A special type of seventh century amphora was used as a cargo container by the traders. Such amphoras have been found near Aden and even as far as Axum in modern Ethiopia. We found startling evidence of this Ethiopian connection in an Axumite coin, the second from our excavations.
The expansion in Red Sea commerce after the Muslim conquest meant a dramatic need for more amphoras. Large kilns were reported to us in 1987 and the threat of their destruction sent us off the site, about 500 m into a residential part of modern Aqaba. We were lucky to have two specialists in ceramic kilns, Dr. Khairieh Amr and Ansam Malkawi, who began to dig up the garden of an apartment building. The trenches were oddly shaped, as we avoided several lemon trees, but the result was most satisfactory. Two kilns were partially uncovered; the larger was almost 4.0 m in diameter and over 2.0 m in preserved height to the firing floor. The second kiln was only slightly smaller with the firing floor completely preserved; the firing chamber was a strange cave with green glassy walls and arches made of spoiled amphoras. Amused neighbors constantly looked in and often mentioned other kilns discovered during construction of the neighborhood. This must have been an industrial area in the seventh century, indicating a vital increase in occupation and commerce. The amphoras and other pottery forms completely paralleled the ceramics found beneath the Large Enclosure.
While the kiln excavation proceeded with almost surgical precision, we expanded operations in the Large Enclosure, which was proving to be a large hypostyle building with a central courtyard. The excavation of the southwestern wall was not without its moments of excitement, due mainly to the talented digging of May Shaer. She carefully recovered numerous tesserae located quite close to the surface, showing that the inner face of the wall was once covered with glass mosaics. A small cache of gold coins was found within the upper layers of the gravel floor. These six dinars of the Fatimid period were minted in North Africa (in modern Tunisia) and, like the Sijilmasa hoard found in 1993, they testify to the economic connection between this region and Aqaba. The wall seemed without special features until we uncovered a niche with a platform stretching before it and a semi-circular salient projecting from the exterior face. The niche had all the characteristics of a mihrab, the focal feature of all mosques indicating the qibla or direction of prayer.
The architectural features of the Large Enclosure are consistent with early mosques:
- The size is within the most common range for urban mosques.
- The building had at least three entrances, approached by platform stairways.
- In the northern corner was a square structure that might have served as a tower (or sauma'a, an early minaret).
- The peristyle of columns has an additional row on the southwestern side forming a covered area of two riwaqs, i.e., the sanctuary.
- The multiple layers of flooring composed of clean gravel without any artifacts (excepting the six dinars).
- Finally, the single niche in the south wall, its form very similar to early Islamic mihrabs.
Yet we hesitated to label this the Congregational Mosque (masjid Jami') of early Islamic Ayla. The problem in this identification is the southwest orientation of the qibla wall; common qibla in southern Syria and Jordan is due south, and the actual direction of Mecca is slightly east of south. Did the early Muslims of Ayla pray in the direction of Upper Egypt? Now for various reasons, early mosques are often oriented incorrectly, as for instance Wasit in southern Iraq, which was off by some 34º. Another example of an excavated mosque is that of Qal'at 'Ana on the Euphrates, where the ninth century mosque had a qibla direction some 35º in error. Northedge (1988, p. 17) suggests this orientation was due to a lack of space on the island.
One might argue that, due to the configuration of the head of the gulf of Aqaba, southwest is the direction of the Arabian coast. More interesting is the possibility that the original mosque was given the same orientation as early mosques in northwestern Arabia (the Hijaz). Nevertheless, despite numerous comparisons and speculative reasons, this qibla represents a variation to a strong Muslim tradition and the identification as a mosque must remain somewhat tentative.
We began this season at Aqaba amidst pervasive rumors that this was our last year of excavations; this was a misunderstanding-this was the last season of funding under a USAID grant, administered by the American Center of Oriental Research. Reports of the death of this project have been "greatly exaggerated." In point of fact, it is slightly embarrassing, but by no means unusual, to come back from a major excavation with good results and more questions. (I like to imagine that now we have better questions.) Our efforts in the bed of the wadi were hampered by water but provided a dramatic realization. The wadi can be excavated quickly and efficiently and will provide a large exposure, perhaps 1/10th of the walled town; this architecture will reveal the organization of the Umayyad and Rashidun city, evidence which has been lacking due to the deep (3.0+ m) overburden of later periods. Architecture in the wadi will also provide a tangible hope for clarification of the Large Enclosure, the identification of which will remain problematic until an inscription or the earlier Umayyad period mosque is discovered. The waters of Ayla have guarded these monuments, but with new sources of funding Ayla will soon yield important pieces of history for early Islamic civilization.
Donald S. Whitcomb has conducted the excavations at Aqaba, Jordan, since 1986. This year's Annual Dinner will be held to benefit Don's future excavations at the site. The excavations at Aqaba were also featured on the cover and divider pages of the Oriental Institute's 1992-1993 Annual Report.
Revised: February 7, 2007