Phase VI: The Severan Reoccupation

Epigraphic material, discussed elsewhere, provides evidence for renewed interest in the much-reduced town of Cosa under Caracalla. There seems to have been an effort on the part of the imperial administration to revive some of the civic functions of the old city, an effort that is archaeologically recognizable in a number of construction activities around the forum. The House of Diana, now entirely collapsed, may have been one of the areas in which new building took place. The open storefronts of rooms C and D and the entrance to the fauces were blocked off, presenting a blank wall to the forum (3) and concealing the ruined state of the house behind. At the same time, a long, solid wall (32) was built through the collapse of the house. Only the lower courses, all in large, irregular stones, were preserved, but from these remains it was clear that the wall had been built to bond with the blocking of shop D. 30, the southeast wall of D, was probably destroyed at this time. The foundation trench did not cut any of the pavements, however, and simply cleared debris down to the level of the old floors.

Fig. 66: Forum V, the horrea.

The new wall ran northeast-southwest, parallel to the northwestern exterior wall of the house and 5 meters away from it, and stretched from the Severan blocking wall next to the forum portico to the Augustan blocking wall of the garden loggia (231). Apart from the blocking wall across C, D, and the fauces, no other structural remains are associated with 32.

It is only possible to explain the construction of this wall if it is assumed that at least the external wall of the house (33) and the wall between room K and the garden (231) were still partially intact at the time 32 was put up. If these walls were still in place, they could have been rebuilt in conjunction with the new construction to create a long, narrow building perpendicular to the long axis of the forum (fig. 66). This building is echoed by similar Severan-period constructions along the entrance to the forum; their narrow layout and lack of any visible floor surface suggests that they functioned as granaries³⁴. The new granary seems to have been the only part of the original house that was reoccupied; in the rest of the ruined building, activity in this period was limited to clearing and cannibalization of building materials. The main cistern seems to have still been open at this time, as it was used to dispose of large amounts of construction debris (tiles, pis», stones, even window glass). Parallels for a signet ring found in the cistern point to a Severan date for this dumping³⁵. The dating of the construction of the long walls to the Severan period rests on very tenuous evidence, as associated pottery of any sort is entirely absent. However, the links to similar, better-dated constructions elsewhere in the forum (below, Forum VI) seem to confirm the dating of this phase.

34. The elongated form of granaries has much to do with the need to provide an raised floor to protect grain stores from moisture and animals. These floors were usually made of wood, a fact that might explain the total lack of a use surface associated with the interior of the Severan building in the House of Diana.

35. Passalacqua in Fentress and Rabinowitz 1996, 236.