A season of intensive site survey just north of Lago di Chiusi at località Gioiella on the border between modern Umbria and Tuscany has revealed evidence for a rural site occupied from ca. the end of the 2nd c. BC through the late 3rd c. AD, based on finewares, wine amphorae, and coin finds. Concentrations of tile and building material suggest two to three buildings placed along a terraced hillslope overlooking the lake. Finds of boxtiles, fragments of monochrome mosaic, lead piping and tiles for pilae are evidence for heated rooms and probably a bath suite. Several pieces of marble (one sculpted) attest to imported decorative elements. There are clear indications of on-site food and cloth production, as well as storage vessels. Nearly a kilometer away, a likely section of ancient road is cut into the side of Poggio S. Maria, running down towards the lake and passing a well-preserved cistern of Roman date. Our investigations are focused on understanding this hydrologically sensitive landscape over time: the development of settlement, agriculture, and political control in the late Republic, the nature of trade and communication with other rural and urban sites in the region, and the relationship between humans and their environment—particularly the manipulation of water—during the site’s imperial lifetime. Finally, we seek to understand the site’s transition (given scattered evidence for medieval use), to its current buried, cultivated state.

The archaeological site "La Villa" is located between Gioiella and Vaiano on a hill just to the north of Lago di Chiusi. To the east of the site is a Roman cistern and what is likely a segment of ancient road. A surface survey of the Vaiano-Gioiella Villa site conducted in 2015 revealed that the site was occupied, but perhaps not continuously, from the 2nd century B.C. to the 3rd century A.D. The distribution of material recovered from the surface in 2015 also suggested that this was a large complex with at least two distinct building areas: one to the south, where fragments belonging to a thermal structure were recovered, and another to the north. The first season of excavation was conducted in 2016 with approximately 15 days of fieldwork between 6 June and 4 July. Based on the density and distribution of materials recorded during the 2015 survey, the project began excavation on the eastern edge of the site. Prior to excavation, a geophysical survey (resistivity) was completed in that area. The results indicated a possible 'surface' and helped determine the location of the four (4 x 4 meter) excavation squares for 2016. Towards the north of the excavated area, an accumulation of debris was uncovered. The deposition of this stratum indicates it was formed by agricultural activity on the site and likely spread from further up the slope of the hill (to the north and west). Below the debris a layer of natural sand was encountered, which was the 'surface' indicated in the geophysical survey and which likely represents the eastern limits of the site. To the south and west, however, a remarkably dense collapse from a building was uncovered. This deposit consisted almost entirely of roof tiles, a few floor tiles, and at least one piece of cocciopesto flooring. In 2016 a small area of this collapse was excavated, but over 153 kg of tile was recovered within an area of just 1x4 meters. However, the most interesting discovery was a channel that had been cut into the natural sediment (the sand) and then covered with tiles pitched to form a triangular covering. Two segments of the tiles were removed during excavation in order to recover the sediment inside the channel and to find the bottom. Surprisingly, there was no bottom tile. Nevertheless, this appears to be a drainage system constructed directly on the natural sediment and taking advantage of the natural absorptive properties of the sand. The channel is at least 6.5 meters long but its overall length remains to be discovered. Parallel and to the north of one segment of the channel is a short stone wall constructed in at least three phases. Stratigraphically, the channel was constructed first; the wall was a later addition, perhaps as a retaining effort against the sand on the north side. Material from the excavation confirms that there was a thermal complex on the site, as there are examples of suspensurae tiles, tubuli for heating the walls, and mosaic fragments. The excavations also uncovered several examples of Sigillata Italica (Arretina) with manufacturers' stamps. Finally, the recovery of numerous artifacts with traces of burning, including a large tile waster, indicates that ceramics and possibly metals were produced at this villa complex. (Rebecca Schindler)

Excavations took place at the Gioiella-Vaiano Villa site in the Comune of Castiglione del Lago from 1 June to 5 July 2017. The site, which is situated on a hill overlooking Lago di Chiusi, comprises a large villa complex spread out over at least three terraces. The 2017 campaign focused on two areas: the central terrace where a series of features of various date were uncovered, and part of a bath complex on a lower level to the south-east. In the central area, two adjoining excavation squares revealed two substantial north-south walls approximately 1.5 meters apart. These walls supported a staircase that likely connected the central area to a lower terrace. Large well-cut stones are keyed into the walls forming a series of supports along each side of the staircase; the steps themselves were likely made of wood. At the end of the 2017 campaign, an approximately 3-meter length of the staircase has been uncovered reaching a depth of 1.5 meters on the south. The bottom of the staricase is not yet visible. To the east of the staircase, there is a well-built cobble wall that forms an apse facing north; only a small fraction of the apse was excavated in 2017, and the function of this structure remains unknown. After the apsidal structure went out of use, a north-south concrete wall was constructed up against and over the apse. This feature runs the length of the 4 x 4 meter square and its overall length and function is not yet certain, however, it may have served as a retaining wall for the terraces. The pottery in the fill in this area is mixed, ranging in date from the 3rd century BC (black-glaze) to the 3rd century AD (sigillata) with a few small sherds of invetriata, suggesting that the fill is late. The most interesting find from the central area was part of a catilus from a Roman mill, confirming large-scale agricultural production at the villa in the Roman period. Excavation to the south-east, on a lower terrace approximately 3.50 m below the central area, confirmed the location of the bath complex, the presence of which had been supected based on the survey conducted in 2015. After removing the topsoil, which contained mostly building material related to the bath—including tiles, tubuli, suspensurae, and fragments of mosaic—we were able to identify the hypocausts for at least two heated rooms, with a possible third room to the west. The hypocaust walls appear to have been cut directly into the natural sand and lined with cocciopesto. The northern room is apsidal, and there appears to be an opening between the two rooms to allow for the flow of air; however, the location of the stoke hole has not yet been discovered. In both rooms, the lowest levels of the pilae for the hypocaust remain in situ in several places. Fragments of black-and-white mosaic were recovered along with many pieces of tubuli for heating the walls. In the area just to the west of the southern room, several nice examples of window glass were recovered. The bath complex would have had a pleasant southern aspect overlooking the Lago di Chiusi.

In 2018, excavations at the Gioiella-Vaiano Villa site continued in the two areas identified in 2017: the bath complex on the lower terrace and the central terrace where several features, including a large staircase, had been previously identified. In both areas we opened new excavation squares in order to determine both the horizontal extent of the structures and their depth. The villa’s history of use and reconstruction, from the 2nd c. BC to at least the late 3rd c. AD, and the long period of abandonment during late antiquity and the early Medieval period have resulted in significant strata of both debris and fill. Thus, the Roman occupation levels are much deeper in some areas of the site than we initially expected. In the bath complex, the 2018 excavations revealed the northern half of the apsidal room first identified in 2017. In this area of the bath the depth of the fill and accumulation of debris below the modern plough soil allowed for better preservation of the hypocaust system with several pilae preserved to their full height (63 cm). The extent of the bath to the east is still unknown. To north a passageway appears to extend from the apsidal room, but it is not yet clear if this was part of the original construction or a later addition. Against the outer wall of the apse on the northwest, a later north-south wall was constructed that was then subsequently modified with pan tiles that appear to be reused for drainage. It is possible that this addition is related to the drainage channel six meters to the north that was excavated in 2016. In the central area excavation focused on the two squares immediately to the south of the north-south staircase and the north-south late period concrete wall first identified in 2017. A brief exploration of the trench immediately to the north of the concrete wall revealed undisturbed clay sediment suggesting that we are at the northern limit of the features of the villa’s central terrace. On the other hand, to the south of both the stairs and the concrete wall we now have evidence of a large structure that likely went through at least three phases of reconstruction and reuse. Under the north-south concrete wall is a well-built limestone wall, the complete depth of which is not yet known, but which has an opening, like a window, framed by long tiles. This wall appears to be related to another limestone wall that runs east-west against the bottom of the stairs. That wall also uses long tiles, which are at the same elevation as the tiles in the north-south wall and which are precisely at a right angle to one another. Thus, the stairs (excavated in 2017) may have provided access to a large structure. To the south of and just inside the western edge of the north-south limestone wall, are two large ashlar blocks stacked on one another and above a segment of cocciopesto that seems to form the wall of a basin. At the bottom of the cocciopesto, on the last day of excavation, we uncovered a small portion of a tile floor, more than 2 meters below the the top of the north-south concrete wall. This type of construction suggests a production area, however, the depth of the fill made in this area made it difficult to reach Roman era occupation levels, and the precise relationships between all the features in the central area remain to be uncovered in future campaigns. The fill at the bottom of the stairs and above the east-west limestone wall included cooking wares and storage vessels, some of which can be reconstructed into almost complete forms. We also recovered more than 500 pieces of intonaco from this area, suggesting that at some point debris from elsewhere on the villa was being dumped into this central structure. In addition to the intonaco, other notable finds from the 2018 campaign included several well-preserved lamps and a bronze sestertius of Faustina II. From the floor of the bath house, below the pilae, we have a complete water jug.

The fourth season of excavations at the Vaiano-Gioiella Villa site took place during June and July of 2019. Excavation continued on the lower terrace where three rooms of a bath complex had previously been identified, as well as in the central area where a building with several construction phases, including a staircase leading from an upper terrace to a lower level, had been discovered. At the end of the 2018 season, a small section of a cocciopesto basin with a tile floor was uncovered, which at the time appeared to be for agricultural production (see 2018 FASTI summary). However, the 2019 campaign has revealed that this structure was most likely a type of nymphaeum. In 2019 we expanded the excavations in the central area to the south and focused on clearing the accumulated layers of fill inside the structure identified in 2018. Due to the dense layers of disarticulated Roman building debris in the upper strata, which appear to have been used to level this area for modern agriculture, excavation was limited to the east and north part of the building. Limestone ashlars stacked on top of each other, one set of which was first uncovered above the basin in 2018, form a system of pilasters that serve to create niches along both the north and east walls. On the north, there are two niches on either side of the “stairway” that descends into the structure from the north (see 2017 FASTI summary). On the east wall, four niches have been exposed with a possible fifth to the south. If the building was symmetrical, there were likely 12 niches in total. The overall dimensions of the building are estimated to be 6.5 m x 8.4 m—a substantial structure. The scale of the building is matched by the decoration. The walls of the niches are surfaced with cocciopesto modeled to look like an artificial cave, with rough limestone nodules creating a rocky surface. Set into this surface are small pieces of blue-glass tesserae. The outer face of the limestone pilasters appears to have been decorated with intonaco painted in red, yellow, black, and white. However, there are not yet enough large fragments to reconstruct a pattern. The “floors” of the niches are formed by the top of the thick walls of the cocciopesto basin, but the limestone pilasters each have a layer of clay fill between the bottom of the stone and the top of the cocciopesto basin. This has yet to be explained, especially since the basin now appears to be a large pool forming a water feature. Moreover, since the basin walls extend up to the niches and pilasters, it is unclear if and how a floor for supporting furniture or people standing would have been constructed. The tile floor of the basin is well preserved in the north-east section, but in a later construction phase appears to have been partially removed towards the south and center of the structure. Also during a later building phase, the “stairway” was blocked. Given the nature of the building, it is now possible that what were first identified as stairs, may actually be a water feature, perhaps a cascade, for the nymphaeum. Materials recovered from the fills inside the nymphaeum include a wide variety of ceramics and building materials of mixed dates, suggesting the most of the upper fill in the building is a relatively late accumulation. However, as noted in 2018, the levels closest to the floor of the basin include larger fragments of cooking wares and amphora, some of which can be partially reconstructed. These materials are concentrated towards the north and suggest the deliberate dumping of objects from a kitchen or storeroom into the open, but unused, nymphaeum building. From the upper fills the most notable finds of 2019 are two complete and identical brick stamps preserving the name L.ATALLIANI (“of Lucius Atallianus”). Three partial stamps bearing this name had been recovered in previous seasons. On the lower terrace, three rooms of a bath complex had previously been identified, and in 2018 excavations focused on the apsidal room where the lower part of a hypocaust floor is partially preserved. In 2019, excavation on the lower terrace was extended to the west for three reasons: 1) to determine the construction of the exterior wall of the apsidal building, 2) to understand the nature of the later reuse of the bath building, and 3) to connect stratigraphically the central and lower terraces. The exterior wall support for the apse was constructed of cobbles laid in rough rows. As we noted in 2017 on the southern side of the bath complex, this wall was constructed directly into a cut in the natural sand sediment. There was no evidence of a foundation trench and no datable material. In regard to the later reuse of the bath complex, just to the north of the apsidal structure is a channel composed of repurposed cover tiles turned upside down to create a drain running from the west into the pan tiles we uncovered in 2018. While the channel was likely used to drain water, a deposit of burnt material around and inside the cover tiles, suggests there may have been kiln activity in the area and/or the final destruction of the building happened due to fire. Unfortunately, at the level of the drain the site has been heavily disturbed by modern ploughing. As in previous seasons, there was almost no pottery from the bath area. On the other hand, some singular pieces of interest came from the fills in this area: a well-cut marble cornice piece, marble wall revetment, and a fragment of a millefiori glass vessel; tantalizing examples of the luxury this villa once enjoyed.