EXCAVATIONS AT COSA (1991-1997), PART 2: THE STRATIGRAPHY
previous pagenext page

IX D North

(Teresa Clay, 1991, and Joanna Hambly, 1992)

Objectives

The cleaning of the Basilica in 1991 and the discovery that the structures on it were all Byzantine led us to ask the extent of this occupation, which clearly depended on the substantial road running up from the northeast gate, the 'Porta Romana'. Further, 1951 excavations in the trench 'isolating' the Basilica (ENE 5, 6, 7) had revealed apparently late walls in this area. The trench was thus started in 1991, and its excavation continued in 1992.

Dimensions

The trench measured roughly 10 x 13m., but was rather irregular in shape: an olive tree occupied the center, and another two the northeast edge. Further, the central part of the trench was only excavated down to the collapse of the Basilica wall: at the suggestion of Dr. Ciampoltrini of the Soprintendenza this was conserved. In no area was bedrock reached, although the cleaning of the 'isolation trench' and the features associated with it gave us a clear idea of the stratigraphy.

Stratigraphy

IX D north, Republican walls cut by the construction of the Basilica
Fig. 74: IX D north, Republican walls cut by the construction of the Basilica
The earliest features on the site were the walls of two Republican houses (2 and 35), separated by what is probably a road running northeast - southwest (this road appears to have been confirmed by the excavation of the sample trench P7). Their walls were of roughly coursed stone and earth, founded on the bedrock without construction trenches (fig. 74). To the northeast of 2 a small patch of surface was uncovered (16), which was probably contemporary with the use of the building.

The earlier houses were destroyed during the construction of the Basilica, leaving the pis» de terre deriving from wall 2 in a large heap (25). The construction trench for the Tribune of the Basilica was 50cm. wider than the wall, and was filled with stone footings, whose mortar laps onto wall 2 (36, 37). A similar consÇtrucÇtion trench for the Basilica cut wall 35 (41).

During the later Republican period and the Augustan occupation the area remained open, becoming a dumping ground during the first quarter of the 2nd c. A.D., when a rich midden deposit, 17, accumulated. This consisted of gray silt containing discrete lenses of other material (sand, clay) and huge quantities of amphorae, fine wares, glass, bronze and iron objects. The deposit appears to be a general rubbish tip of household and destruction debris. It is not, however, associated with anything structural in the trench.

IX D north Third-century walls.
Fig. 75: IX D north Third-century walls.
The third century saw major rebuilding in the area of the Basilica, with the reconstruction of the Odeon under Maximinus (A.D. 232) and the construction of the corner buttresses supporting the Curia. In the excavated area, two adjoining walls, 4 and 40 were constructed (fig. 75). These were well built and mortared, 0.45m. wide, faced on both sides and oriented northeast-southwest on the alignment of the Roman city. Wall 40 is slightly off-alignment, however, and seems to run slightly north so as to avoid the Curia buttress: it seems logical to suppose that it post-dates it. It is probable that these are house walls, and that the excavated area lies outside the house. 3, in unmortared stone, takes up the line of the pre-Basilica 2, butting the corner of the building and blocking the line of the earlier road. Taken together, these walls seem to enclose a yard, surfaced with a compacted layer of earth containing mortar and crushed stone, 12. A fragment of third-century pottery, ARS form 32, dates the whole to the 3rd c.

These features should probably be associated with RichardÇson's 'Level III': (8/6/1951) ÏExcavation of tribune vault reveals firm rammed earth floor over a fill of debris through and just within doorway. This level (II) followed outside and found to continue as rammed earth floor...Ó (26/6/1951) ÏDirectly beneath level II appeared the floor of Level III, in part (SE) of rammed earth in part (NE of irregular rough cement of broken stone and friable mortar). To this floor corresponded a rough rubblework sill across the archway of the tribune vault. At this level a drainpipe traversing the NE wall of the enclosure (3) discharged into a diagonal channel that ended in a box-like depression before tribune vault arch. In angle of tribune and Basilica square stone manger set in cement floor. This arrangeÇment evidently involved use of trib. vault as stable.Ó RichardÇson is evidently correct in interpreting the whole ensemble as an area for the enclosure of animals, and other apparently agricultural activities.

IX D north, the collapse of the basilica or church wall.
Fig. 77: IX D north, the collapse of the basilica or church wall.
The area was apparently abandoned by the end of the 3rd c.: no 4th or 5th c. pottery was found there, nor are there any later domestic features within the area of the excavation. The next surface, corresponding perhaps to RichardÇson's ÎLevel IIÌ, was that of a small cemetery, excavated only in the northwestern sector of the trench, where it was not covered by the collapse of the Basilica wall (fig. 77). This consisted of six graves: a seventh skull was visible in section, while a 'superficial burial' was recorded by Richardson near the tribune vault. A summary table of these burials is found in part I, p. 79. The cemetery appears to have been bounded to the east by a wall running diagonally to the line of the trench, 11. Extensively root-damaged, the wall was preserved over 3.70m. It was 0.85m. wide, and faced on both sides. The core was of unmortared rubble stone. This appears to have separated the cemetery from a road running diagonally across the site, surfaced with small rounded stones tightly packed in a pale gray sandy matrix. The surface of the road (15) was slightly cambered, and its makeup (38) was up to 30cm. deep in the center, tapering out to 7m. on both sides. The western edge was well defined by a kerb of closely set stone and broken tile. The eastern edge lay under the section, giving it a width of at least 5m. It is thus impossible to tell whether it was enclosed on its eastern edge by a wall similar to that found in trench Forum II.

With the exception of a small metal ornament, no contemÇporary finds were associated with any of these features. It is thus impossible to be certain of their date, and we must proceed by association: the cemetery with the 6th c. church lying just above it, and the road with the 6th c. ramp running up through the Curia.

The collapse of the wall of the Basilica/Odeon, which must have been used by the church, occurred at sometime after the abandonment of the cemetery. It fell outward from the platform, and its collapse is relatively coherent in the area of the trench, where the string courses of bricks can be easily traced (5b: pl. 65 and fig. 77).

Lapping over the collapse of the wall was a very compacted yellowy-brown surface, 8, containing large quantities of stone, mortar and tile, suggesting a heavily trampled exterior surface such as a yard or sheep-pen. 12th and 13th c. pottery was found in these layers, but there is some possibility of contamination from the layers above.

Modern layers included 9, composed of a series of dumps of tile, stone and mortar and much mixed debris, increasing gradually in depth downslope towards the northeast. Traces of a track, path or barrow-run (7) and a compact mortar spread just under topsoil (6,10) are probably connected with the 1951 excavations, as were the successive tips of earth.


previous pagenext page