Combined Caesarea Expeditions: 1998 Summer Season


by: Kenneth G. Holum, Project Director


c98s1.jpg (14195 bytes) (1) Caesarea, 40 km. north of Tel Aviv, was Herod the Great's port city, seat of the Roman governors of Judaea and Palestine, and later a prosperous Muslim and Crusader town. (2, 3) The Combined Caesarea Expeditions conducted summer excavations from May15 to July 23, 1998. This was our Tenth Anniversary Season, and also the tenth season in a row. The project directors are Kenneth G. Holum of the University of Maryland, and Avner Raban and Joseph Patrich of the University of Haifa. Institutions participating in the 1998 excavations were the University of Haifa, the University of Maryland, the University of Oklahoma, Ardingly College, Trinity College, Dominican University, Concordia College, Temple University, and the University of South Dakota. The total of student and volunteer excavators on land and under water was more than 180. We are grateful for continuing support from the Joseph and Mary Keller Foundation, the National Endowment for the Humanities, the Dorot Foundation, and several other private foundations and individuals.

(4) In 1998 we excavated in two areas underwater, N and K, located within or adjacent to the main basin of King Herod's great harbor, and in two areas on land, area LL, located on the north side of the Inner Harbor basin, and area TP, the Temple Platform, site of Herod's temple to Roma and Augustus.

Underwater Investigations

I will mention only briefly our underwater work in King Herod's harbor (5), directed by Avner Raban and Ed Reinhardt, who will write soon on their excavations. In area N, located on the inner edge of the southern mole of King Herod's harbor (6), the divers studied a series of tilted ashlar blocks that were probably part of the substructure of the promenade on the mole mentioned by Josephus in his accounts of Caesarea's founding. In area K (6, 7) two probes were placed against the southern side of twin towers located just outside the harbor entrance to study their method of construction. These towers, also mentioned by Josephus, supported statues of members of Augustus' family. A large block of hydraulic concrete, part of the tower's structure, was found to have been made of two distinct types of concrete..


Area LL

(4, 8) Area LL, supervised by Martha Risser, is located on the north side of the Inner Harbor basin, where Lee I. Levine and Ehud Netzer of the Hebrew University of Jerusalem excavated from 1974 to 1976. Levine and Netzer identified four distinct strata in their excavations. Their stratum 1 was the latest Crusader phase, and stratum 2 was Early Islamic, characterized here by numerous walls 25-30 cm. in width representing one-storey dwelling units. This stratum closely followed the layout of stratum 3, earlier Early Islamic, when land use changed and the sector became a residential district. Levine and Netzer found the most marked difference between stratum 3 and stratum 4, which was the latest Byzantine occupation, characterized by a horreum or warehouse, about 20 x 45 m., fronting the Inner Harbor on its southern side and a wide road constructed of limestone pavers, running east-west on the north side of the warehouses (9). The width of the warehouse walls ranged from 65 to 95 cm. The street was likewise Byzantine but had at least three earlier pavements beneath it.

CCE renewed area LL investigations in 1996, hoping to study further an important commercial and residential district adjacent to the harbor, and to explore the site's occupation diachronically from as far back as the Herodian or even Hellenistic period. In 1996 and 1997 we confirmed and refined the Netzer and Levine stratigraphy, including identification of the Islamic dwellings and the Byzantine horreum or warehouse.

(10, 11) The earliest structures uncovered in 1997 and 1998 were heavy and broad foundations below the Byzantine warehouse that we believe represented a Roman or even Herodian building of the same type. (12, 13) So far, we have found these foundations wherever we have penetrated beneath the Byzantine walls, so the earlier building was apparently built on a similar plan.

(14, 15) Of the Byzantine warehouse we have now exposed most of two additional rooms east of the central corridor down to floor level, exposing seven or eight courses of solidly-built walls, and have dated the structure to about 400 C.E. Because of its overall dimensions, about 20 x 45 m., and heavy construction, we identify this as a state warehouse, probably connected with the annona, the state apparatus that supplied grain and other commodities to the army and civil service. (16) We have also exposed more of the limestone street onto which the warehouse apparently opened to the north. Beneath the street we exposed a capacious drain channel. The warehouse and the street appear to have served their original purposes through the rest of the Byzantine period and until the Muslim conquest in 640.

(17, 18) In our interpretation, the large warehouse room that we have been excavating, east of the central corridor, was reroofed and subdivided about 700, during the Umayyad period. A light curtain wall divided the room, and a new roof (or second storey) was supported by pilasters made of reused column segments. We do not as yet know the function of the remodeled warehouse room. Clear domestic occupation is attested only in the ninth century, when rooms and courtyards were built over the Byzantine street, and beneath them wells, cisterns, and sink-pits. (19, 20) In LL5, for example, the street was blocked in the ninth century by a wall between two pilasters and two phases of stone-paved courtyard, belonging to a courtyard residence, were laid above the Byzantine street. The courtyard incorporated the access shaft to a perfectly preserved cistern of the Fatimid period, about 1000 C.E., beneath the pavement. (21) Also part of the residence was a subterranean grain bin, built into a room of the earlier Byzantine horreum, once lined with carefully cut and fitted ashlar facing that would have kept the contents dry and safe from vermin. In the Fatimid period, about 1000 C.E. such grain bins were common at Caesarea, and were always associated in this period with residential units.

(17, 22) The most important discovery in LL, however, relates to the transition from Byzantine to Islamic. In both the Hebrew University excavations and our current campaigns, the warehouse has yielded clear evidence of use as a warehouse right up to the Muslim conquest, then abandonment, and then conversion to other purposes. In the room that we excavated, east of the central corridor, a new slab floor was laid in the eighth century, when the room was subdivided and a new roof installed. This eighth-century slab floor sealed what we call the abandonment layer, representing pottery in use just before the building was abandoned. (23) Our excavators extracted 54 baskets of pottery from this layer (L1242) and eighteen identifiable coins, mostly of the sixth and early seventh centuries, of which the latest precisely datable ones were two folles of Heraclius, minted 629/30. Seven lamp sherds were also found, all of the seventh century. The pottery assemblage, large sherds and whole vessels (24) deposited directly on the floor of the Byzantine warehouse, was virtually all jars and amphoras, with only a few residual sherds of cooking ware and tableware. The repertoire was 55% Riley 1 and Riley 2 jars, vessels manufactured in Palestine and used in the wine trade, 10% carrot-shaped amphoras from Egypt, and nearly 20% Riley 5 and 6 amphoras from the East Aegean and Asia Minor. This deposit, and a similar one recovered earlier in the Hebrew University excavations, is unambiguous evidence that this building was used for storage of exchange commodities until the Muslim conquest, and that it was abandoned soon thereafter.

Area TP

(4, 25) The Temple Platform, where Farland Stanley supervises, is the site overlooking the harbor where King Herod built his temple to Roma and Augustus, mentioned by Josephus. Avraham Negev of the Hebrew University cleared the site of modern buildings and excavated there in the early 1960s but discovered little. We began excavating in 1989, after Yoram Tsafrir and I recognized the foundations of an Early Christian (26) church among the remains that Negev left. The church, it turned out, was built squarely on the foundations of the temple. Clearly, the site has much to teach us about the process of Christianizing Caesarea, and the Roman Empire. (25, 27) Herod chose a spectacular setting for his temple. The Temple Platform is a low ridge of bedrock, artificially extended to form an esplanade about 110 x 90 m. In its original configuration, wings extended westward embracing the Inner Harbor, making one organic unit of temple and harbor.

(28) In 1998 we found more remains of the temple, of which only the deep foundations survive in situ. On the east, in TP25 (29) we exposed an inner corner of the cella foundation, confirming that the width of the foundations was 8 m. (30) On the west we exposed the inner edge of the pronaos foundation, confirming that its width too was 8 m. We also found evidence of quarrying in the underlying bedrock, either at the time the temple was constructed or earlier. The 1998 results confirm that the ground plan of the foundations, projected in 1997, is correct.

(31, 26) In TP25 and elsewhere, we likewise exposed further foundations of the octagonal church, in part heavily robbed. (32) The line of these foundations was already known, however, so other discoveries were more important. (33) In 1998 we exposed and studied a group of enigmatic foundations in TP19/23, TP1/2, and in TP25. (34) It is now clear that these foundations lay above remains of the temple and below the floors of the church. Thus the temple had been largely or completely destroyed up to a century before the church was built. Pottery and coins found below these foundations dates them to about 400-450, and we now think the church was built about 480-500. Nevertheless, numerous stones of the temple were incorporated into church (47), so many that we can say they shared the same fabric. From what we have recovered so far of the intermediate structure, or structures, there is no sign that they formed a monumental building, such as an intervening church.

(35) On the eastern side of the octagonal church, in TP9 and 25, we have located remains of the bema, the raised platform for the clergy and the altar. In our plan we give the bema an apsidal shape (36), since the church lacks an external apse. The evidence is a mortar foundation or matrix for marble floor slabs that Negev had already exposed (35). In it the impressions of the slabs can still be traced. If our identification is correct, the church in its final configuration had the normal bema on the east raised about 30 cm. above the floor level of the church. (37) We also sectioned the bema, and to our surprise discovered that in its original configuration, when built about 480-500, the church apparently had no raised bema. The leveling and matrix of the bema rest above a second sequence of leveling layers and matrix that represent the floor of the church, which therefore extended originally across the bema area.

(38) I suggest as a parallel the eastern church at Mamshit in the Negev, which had a raised bema, a chancel screen, and behind the altar table the stepped synthronon. The arrangement at Caesarea would have been similar, but perhaps grander and more elegant. At Mamshit there was also an ambo, the pulpit where the preacher stood elevated above the congregation to deliver his sermons. Small posts supported it, (39) as in the base of the ambo of the north church at Rehovot, also in the Negev. (40) During the 1998 season, we also discovered the base of the ambo of the Caesarea church. It was indeed larger and grander. The base was a circular disk of marble that was originally embedded in the church floor in front of the bema and slightly to the north (41). Although it is incomplete, we can see easily in the photograph (40) that the Caesarea ambo base had six square posts that supported the elevated pulpit.

(42) The six fragments of the ambo discovered so far had been buried, along with two monolithic marble columns from the church colonnade, in a pit dug beneath the bema. Apparently the church collapsed in the great earthquake of 749, and within the next century domestic occupation established itself on the ancient site of the temple and church. Rather than remove these heavy architectural fragments, the house builders simply buried them beneath the new floor level. The pit, like a similar one in the vicinity, contained pottery of the ninth century.

(43) When the church had fallen, domestic occupation established itself on the Temple Platform, represented mainly by cisterns, wells, and sink pits, the remnants of courtyard houses that in the later tenth century also had subterranean grain bins, as in area LL. From the eleventh through thirteenth centuries, however, monumental public buildings again existed on the Temple Platform, represented by parallel north-south foundations nearly 2 m. broad that presumably supported vaults (43, 44). The 1998 season did not resolve the issue of the function of these monumental buildings, but in TP24 (45) we did expose a massive wall at an angle that closed the largest vaulted structure on the north, and through it an entrance into the structure to the south. In the eleventh and twelfth centuries this wall was flanked on the north by a low bench that had animal ties. The search for an explanation continues.

Thus the 1998 excavations shed much further light on Herod's temple, on the buildings that succeeded it on the same site, and on the warehouse complex in area LL. We contributed new evidence for the impact of the Muslim conquest in 640 on the city and its economy. In the meantime, work continues on the material collected in 1998 and earlier seasons, on coins, ceramics, animal bones, and botanical samples (46). Debra Taylor of Tufts University has catalogued more than sixty fragments of the temple superstructure, (47) including more than one block that was used both in the temple and the church. We are also proud to announce that Caesarea Papers 2, edited by Kenneth G. Holum, Avner Raban, and Joseph Patrich, will appear soon as a supplement volume of the Journal of Roman Archaeology. It will take its place with other well-known publications of the Caesarea excavations, (29) such as King Herod's Dream (48), long out of print, and Caesarea Maritima: A Retrospective after Two Millenia, available from E.J. Brill, Leiden (49).