Phase I: Republican Construction

Coins recovered by Brown from the foundation trenches of the external house walls dated the first building activity in the area to around 170 B.C. However, the black glaze in the earliest levels of the garden suggests a date around the middle of the century (part IV, p. 269). The discrepancy might be resolved by assuming that the house was built before the garden was prepared, but as the garden was certainly cut out at the same time as the rest of the house the later date is more likely. The earliest phase of construction involved massive excavation of the bedrock sloping up to the southwest: to create a more level layout, the bedrock in the rear of the house was cut down to a depth of up to two meters. In the south corner, the bedrock was left at a slightly higher level, creating an elevated platform. In the west corner of the house, where excavation was deepest, the sides of the cut were smoothed and straightened, and served as the lower levels of the walls of the building. In this area, a thin layer of black soil (412) just above the bedrock floor of the cut may bear witness to the initial construction phase. It is possible that this dark deposit is composed of the remains of fires used to crack the bedrock to facilitate its excavation. Since this area seems to have been used as a garden from the earliest phase of the house, however, the black soil may also represent the decomposition of an initial layer of organic material deposited as fertilizer. The composition of the layer was not analyzed, and thus it is impossible to confirm either interpretation.

During the same construction phase, several other cuts were made in the bedrock on which the house would sit. In the front of the house, two roughly square pits, each about two meters deep, were cut into the west corners of future rooms C and D (177, pl. 48, and 178). Although both pits lack outlets, they seem to have been designed to function as cesspits. In other parts of the site, similar cuts have usually been interpreted in the same way¹², and it seems that the local limestone bedrock was porous and absorbent enough to make drainage arrangements unnecessary. The pit in room D showed an outcrop about 1.5m from the bottom from which a vault may have sprung, but it seems more likely that both this pit and the pit in room C were originally covered with simple planking. In the center of the house, a large, deep rectangular cut was lined with hydraulic mortar and roofed with a vault of rough, rectangular limestone voussoirs (pl. 49). A short transverse passage was cut from the main chamber to a slightly truncated well shaft that opened in the atrium of the house. This structure was to serve as a cistern throughout the life of the house. A fourth cut was made in the south quarter of the house, forming a rectangular chamber about two meters long, 1.3m. wide, and about 1.5m. from floor to roof. Like the cistern, it was vaulted with rough, un-mortared voussoirs, but it was not lined. The lack of mortar lining suggests that this cut, like the cesspits in the front of the house, was intended for the disposal of wastewater. A square hole (ca. 0.5 x 0.5m.) left in the crown of the vault at its northeastern end may have been designed to allow periodic cleaning of the cesspit, though the long sequence of pottery deposited in this pit suggests that the primary use of the opening was rather the disposal of garbage (part IV, p. 278ff.). The chamber seems to have been designed to communicate through a natural (?) fissure with a large drainage area in room N; this area was partially covered by large, worked stone slabs that left an opening about 1m. square. The last cut in the initial phase of construction was an irregular squarish pit, about 1 x 1.5m., in the west corner of the platform of bedrock left in the south corner of the house, perhaps intended as a settling basin for what was to be a raised cistern.

Once the shaping of the terrain and the excavation of rock-cut features had been completed, the walls of the house were constructed, probably using the stones cut from the bedrock. In the front of the house, the builders used typical Roman construction techniques: dry-stone socles of roughly worked, medium-sized stones bore superstructures of rammed earth (pis» de terre). To the rear of the house, as noted above, the function of the socles was taken over by the bedrock sides of the original cut, topped by one or two courses of large stones. The form and phasing of the exterior walls is clear, as those cut into bedrock must represent the initial plan of the house. These walls enclosed a rectangular space about 17m. wide and 34m. long; their use as party walls by the "fish market" to the northwest and by AB VI to the southeast indicate that they and the space they enclose were part of the original central plan for this side of the Forum. Later modifications, especially in the rear third of the house, have made the original position of interior walls slightly more difficult to determine, but careful examination of construction technique and bonding relationships reveals a fairly clear picture of the plan of the Republican building.

[ 1 2 3 next ]

12. Brown, Richardson and Richardson 1993, 84 and 87.