Phase VIII: Modern Archaeological Excavations

Excavations sponsored by the American Academy in Rome began at Cosa in 1948, and thus it is not surprising that various areas of the House of Diana had already been the objects of archaeological investigation both formal and casual. Into the former category falls the excavation of the angle between the southeast wall of the house and the wall between O and I³⁶. The backfill of this excavation contaminated parts of the layers covering rooms I and O: in room I, where contamination included fragments of a WWII-era bombshell, the fill was designated 81, while in O the corresponding backfill was recorded as 79. It seems that this cut was intended to confirm BrownÌs hypothesis about the internal layout of the buildings around the forum, and thus only uncovered about two meters of the external wall and a meter of the cross-wall between the southeast ala and room O. The sounding must have revealed the presence of the quarter-round molding in O and is probably the basis for BrownÌs statement that the original public atrium became a house in the Augustan period³⁷.

While the excavation around I and O was duly recorded, two other soundings do not seem to have been included in any of the site publications. Around the northwest door to room E, a fairly regular cut, clearly backfilled with 60, seems to have been intended solely to clarify the position of the wall between this room and the atrium. Although this cut may also have been the result of an attempt to rob building material or a modern foray with a metal detector, the squarish form of the cut and the fact that the wall uncovered corresponds to wall placement on the hypothetical Atrium Building plan may argue for an archaeological origin. There also seems to have been a sounding in the east corner of room N, where we found that a precise rectangular cut descended through all three pavements and the neighboring wall; although the only explanation for this cut seems to be archaeological activity, there is no documentation of such a sounding in any of the previous Cosa publications. The cut was filled with a rocky, dark soil (346) that, given the absence of floor fragments, may represent natural sediment rather than deliberate backfill.

In the area of the House of Diana were a number of other traces of activity peripheral to archaeological excavation. Two of these traces were interpreted with the aid of photographs published in previous reports: two postholes originally thought to be associated with the sunken-floor building in room D were discovered to belong to a wooden fence located around the forum in the 1960s or 70s, while, thanks to an aerial photograph showing the placement of a previous spoil heap in the area of the house, a series of puzzling layers just below topsoil in the middle of the trench (54, 55, 56) were revealed to be stray piles of spoil from an earlier excavation. Other features seem to bear similar interpretations. A pair of what appear to be postholes, one to the south in the area of room E, the other further north, should probably interpreted as the result of Soprintendenza maintenance work, as should a patch of sandy gray soil (85) that may be the remains of cement-mixing. Naturally, the chronological framework of these activities is short enough to make fine-tuning unnecessary, but a clue to their relative chronology can be found in the fact that one of the postholes cuts into burnt red soil that may have once been a campfire (53), soil that in turn covered one of the backdirt piles mentioned above. By the beginning of our excavation in 1993, a thin layer of loose black topsoil (0) had accumulated over these recent traces and over the abandonment layers that covered the rest of the House of Diana.

36. Cosa III, 96 and fig. 34: here these walls were used to support the extrapolation of the plan of AB II onto the other buildings around the Forum.

37. Cosa III, 238.