The 2024 Field Season at Caere, Vigna Marini Vitalini: A New Etruscan Temple and Its Surroundings

Abstract

The field known as the Vigna Marini Vitalini holds a special place in the archaeology of the city of Caere, beginning in the late 19th century, when a massive number of architectural terracottas were discovered there and eventually found their way to European and North American museums. In the summer of 2024, the Queen’s University team opened a new trench in the area where the 19th-century excavations had taken place. The foundations of a temple came to light, along with a building with at least three rooms parallel to its southwest side. Based on the finds from the associated layers and a deposit of architectural terracottas, the temple can be dated to the early fifth century BCE. When the monumental structure was built, however, the area already had a long prior history. The Etruscan builders dug the foundation trenches into a beaten earth pavement that extends over a large area, suggesting that this had previously been an open space. The floor covers a filled quarry and traces of even earlier wooden structures. This new evidence shows that the history of this area was by no means a linear and planned development, and that there were multiple changes of destination and transformations of the urban landscape.

1. Introduction (F.C.)

For more than 10 years, a Queen’s University team has been conducting fieldwork in the Vigna Marini Vitalini, in the center of the urban area of Caere (Figures 1 and 2), a site with an exceptionally rich and complex building history. The archaeological sequence has provided a wealth of new data on the development of the city in all periods of occupation, from the Iron Age to the Roman Imperial period. The so-called hypogaeum of Clepsina has been explored in its entirety, allowing for a new reconstruction and interpretation of this monument. In the 2024 season, a new excavation area (Area 4) was opened, corresponding to a concentration of anomalies revealed by the magnetometric survey. The massive foundations of a structure identified as a temple came to light as soon as the thick layer of modern ploughed soil was removed (Figure 3). This building was placed in a large paved open space, which covered a backfilled quarry and traces of even earlier habitation. To the immediate southwest, additional structures were uncovered, referred to as the Southwest Building. This report presents the results of this excavation, along with a discussion and preliminary interpretation of the recent discoveries.

Figure 1:

The urban area of Caere; in grey, the Vigna Marini Vitalini (Caterina De Angelis).

Figure 2:

Map of the Vigna Marini Vitalini with the areas excavated by Queen’s University and the neighboring fields with the excavations of Mengarelli and Consiglio Nazionale delle Ricerche. (Marco Di Lieto/Caere project).

Figure 3:

Ortophoto of the new excavation area (Area 4) with its sectors (Marco Di Lieto/Caere project).

2. The Southwest Building (A.F., A.S., J.T.)

A long wall (1302) runs parallel to the temple’s southwest side (Figure 4). Three perpendicular walls (1303, 1305, and 1306) define three rooms, labelled as sectors A, B, and C. Wall 1304, parallel to 1302, acts as the fourth side of sectors A and B and likely continues beyond the excavation limit. The space between the long wall and the temple foundations is sector D. A stone conduit runs along the temple foundation and then takes a slightly divergent path crossing sector D. Excavations are still underway in all of these sectors, but there is sufficient evidence to propose a preliminary reconstruction of the phases.

Figure 4:

Map of the Southwest building. Yellow: beaten earth floors pre-dating the temple; green: Late archaic building; blue: temple and drain; pink: wall 1304 (Marco Di Lieto/Caere project).

In sectors A and B are sequences of superimposed floor layers that are cut on all sides by the foundation trenches for the surrounding walls and therefore must predate their construction. The floor layers in sectors A and B roughly correlate to one another, seemingly once spanning both sectors before they were cut by the later foundation trenches. It is possible that they extend into sector C as well. Early analysis of the assemblages from these floor layers shows that the more recent material dates to the Archaic period, with abundant residual materials from the seventh century BCE and a significant presence of Iron Age finds. The deeper layers investigated in 2024 may date to the seventh century BCE.

The long wall, 1302, is built with more regular stonework and tightly constructed. In contrast, the perpendicular walls 1305 and 1306 are built of blocks of similar dimensions, but with more uneven spacing and placement. The extremities of these walls are not bonded with 1302 but simply abut it. The foundation trenches, which are very wide, appear to confirm that 1302 was built first; however, when the two cross walls were added there was also some work done along the long wall, possibly to repair the foundation. Wall 1304, which is barely visible along the excavation limit, is certainly a later addition: its blocks are different in size and its foundation trench intercepts and damages 1305. The finds from the foundation trench of 1304 indicate a terminus post quem in the fifth century BCE.

Excavation in sector C is not as advanced as in sectors A and B. Here, however, part of the subfloor associated with the building survives. Even though the preservation of the layer is not ideal, the assemblage seems to have a terminus post quem in the second half of the sixth century BCE. A gap in wall 1302, with two regular square cuts in the blocks on either side, shows where a threshold once sat. Beside the threshold, cut into the upper floor and along the wall, is a pit (−1317) that contained a ritual deposit (Figure 5). Although damaged by a later trench dug across sector C, the deposit (1313) provides valuable evidence of ritual activity in the area. The pit may have been covered with roof tiles, large fragments of which were found lying flat near and over the pit. There are numerous comparisons for the placement of tiles over pits with ritual deposits, such as at Pyrgi and Tarquinia. The deposit included an intact olla containing animal bones, a loom weight found just beside it (Figure 6), numerous fragments of a second olla, and fragments of Attic cups, including two black-figure eye cups found near the base and beneath the intact olla. In the part of the deposit disturbed by later intervention were fragments of an antefix and sima, the bottom of an impasto cup/lid, numerous fragments of bucchero and coarse ware, a black pebble, and earlier material. Two bucchero fragments bear inscriptions. The ollae, discovered approximately 60 cm from one another, are a common vessel for ritual deposits. Both contained osteological fragments, with other bones found beneath and around them. The deposit contained fish bones, the ribs and vertebrae of either a young pig or sheep, the humerus of a small wild bird, and what may be a hog’s phalanx. The variety of species represented in the deposit may provide insight into its purpose. The inclusion of fish bones is relatively rare, although we have comparanda in a layered deposit containing multiple ollae from the “monumental complex” at Tarquinia. Here fish, shellfish, and swine bones were associated in one layer, and fish and bird in the other. The top layer contained earlier pottery, which included a spindle whorl. In another deposit, fish and a piglet were associated with vegetal offerings. These deposits, all dated to the sixth century BCE, have been identified as foundation rituals for structures of the sanctuary. D. Maras has noticed the presence of fish, and particularly fish bones in association with swine, in Etruscan foundation rituals, although these observations are still based on limited evidence and require further investigation.

Figure 5:

Ortophoto of the Southwest building with the threshold in wall 1302; the asterisk marks the ritual deposit 1313. (Marco Di Lieto/Caere project).

Figure 6:

The ritual deposit under excavation, with the intact olla and the loom weight. (Caere project).

Athenian eye cups are common in Etruria and have also been found in sanctuaries and ritual contexts. Here less than half of one cup survives, while only a few fragments remain of the other. The exterior of the more complete cup (Figure 7) shows two large eyes and the hind legs and tail of two lions, while the lower part is covered in concentric rings. The most common scene with lions on eye cups is Herakles fighting the Nemean Lion, but most likely the lions stood on their own at either side of the handles. Another cup bearing this iconography has been found at Vulci. The interior of the cup is harder to identify. The depiction of a gorgon in the tondo may be expected, but here the decoration appears to take up the entire space.

Figure 7:

Fragments of the more complete Athenian eye cup from the ritual deposit. (Caere project).

Only a small part of the antefix remains (Figure 8). It belongs to a category of female heads with diadem that is well attested at Caere and dated to 530–520 BCE. Numerous antefixes of this type were excavated in the Vigna Marini Vitalini in the 19th century, showing variations in the painted decoration of the diadem, hair, and earrings. The deposit also contained two parts of a sima, decorated with a step pattern and meander (Figure 9). These patterns are common at Caere and in its territory between 540 and 500 BCE and typically contain panels with starflowers and waterbirds. The two fragments, with steps in opposite directions, must have come from the opposite sides of the pediment, which presented symmetrical patterns. These pieces indicate the presence of a late sixth-century BCE decorative system in the area, perhaps belonging to the Southwest Building itself. Since they were found in the disturbed section, it is not clear whether they belong to the ritual pit, and if so, whether they were originally more complete. There are, however, numerous other cases of deposits containing architectural terracottas in Etruria, particularly antefixes and other figured pieces. The pottery found in the deposit also includes two inscribed pieces of bucchero. One, which starts with the first-person personal pronoun mi, may be the beginning of an ownership or dedicatory inscription (Figure 10).

Figure 8:

Fragment of head antefix. (Caere project).

Figure 9:

Fragments of raking simae. (Caere project).

Figure 10:

Bucchero fragment with inscription. (Caere project).

Another notable object from the assemblage is a smooth black pebble, completely out of place on the city plateau, entirely made of yellow-red volcanic tuff. Black stones of a larger size have been found in numerous Etruscan ritual contexts, including a deposit in the Vigna Marini Vitalini that marked the closure of the whole area in the second century CE; this deposit had also included an antefix. G. Colonna has associated these stones to the cult of the chtonic deity Śuri. It is certainly possible that the black pebble was an item of religious/symbolic significance and might have played a role in the ritual.

This assemblage has characteristics seen in both foundational and votive deposits within sanctuaries. There is some significance in the placement of this deposit along a wall and beside a doorway: it appears to mark the door as a sort of transitional space. The presence of two eye cups and two ollae may reflect a pattern observed at other Etruscan cult sites, notably Pyrgi, where vases frequently occur in pairs and are possibly related to the worship of a divine couple. The loom weight, a typical artifact in ritual assemblages, is often assumed to be a gender indicator for the subject and/or the recipient of these offerings.

While both the Athenian eye cups and the antefix date to circa 530–520 BCE, the dating of the other fragmentary material from the assemblage is less precise but within the sixth century BCE. In the sanctuary of Pyrgi there are numerous deposits made of fragmented and incomplete pottery, often Attic, and sometimes architectural terracottas are also present. The fragmentary and incomplete cups and antefix may have been placed in the pit as a sort of relic, perhaps even broken on purpose, which would explain the damage. On the other hand, the ollae containing food, placed on top of the deposit, are contemporary with the ritual. One of them is an example of the so-called external slip ware, whose production starts at the end of the sixth century BCE and becomes significant only in the fifth century BCE. The deposit dug into the subfloor abutting wall 1302 may have been associated with a renovation, rather than with the foundation of the Southwest Building, and the architectural terracottas and the pottery could have been part of its previous décor and furnishings.

In sector D, along the Southwest Building, there is another sequence of floors, although only a single layer has been excavated. The latest material from this layer dates to the late sixth or early fifth centuries BCE. The trench for the long wall 1302 is wide and rather irregular on this side as well, suggesting later interventions on the structure and, in particular, the part corresponding to sector C. The beaten earth floors in sector D were cut for the construction of the temple foundation and for the stone drain that runs beside it.

Based on the excavation data, the construction of the Southwest Building can be dated to the sixth century BCE. It underwent at least one renovation, and most likely functioned in conjunction with the paved open space next to it, which was accessed through the doorway. Considering these findings, a broad sequence of phases can be proposed. There is significant evidence of occupation in the Iron Age and the Orientalizing period, documented by abundant residual material. The deeper excavated layers in the building may also date to the Orientalizing period, and earlier undisturbed layers may lie beneath them. The construction of the Southwest Building took place in the Archaic period. There may have been multiple phases, possibly with the addition of the two cross walls. The ritual deposit was placed in sector C, probably between the end of the sixth and the early fifth centuries BCE, possibly marking a phase of renovation. The very large and irregular trenches suggest that the foundations may have been repaired. The temple and its drain were built around 470 BCE, as discussed below. Lastly, wall 1304 indicates a rather invasive renovation of the Southwest Building that took place in the fifth century BCE.

In conclusion, the Southwest Building was a structure of considerable size and architectural quality, the full extent and plan of which cannot be appreciated at this stage of the excavation. It had a long straight side with at least one doorway and may have had a roof with architectural terracottas. A large open area with a beaten earth floor occupied the space to the immediate northeast. The construction of a temple, parallel with and so close to the building, suggests that both the paved space and the Southwest Building may have already been public facilities. Rather than a complete change of use for this area, the new temple may only mark the development of more monumental and specialized structures for communal religion, a trend that is well documented across Etruria and Latium. After the construction of the temple, the Southwest Building was either abandoned or continued to function alongside the temple as a service structure, like those seen at other Etruscan sanctuaries such as Pyrgi and Marzabotto. However, at this stage in the fieldwork, crucial information, most importantly the plan and extent of the building, remains unavailable until excavation resumes.

3. The Phases Preceding the Temple (S.K., H.M.)

The excavations in the northern section of Area 4, sector E, unearthed evidence of multiple phases of use of the area. The earliest phase of activity in Sector E is attested by circular cuts in the bedrock. These were discovered initially by means of modern interventions at the site, namely a large hole dug into the beaten earth floor that spans the majority of Area 4. When the modern fill was removed at the beginning of the 2024 season, the surface of the underlying bedrock was exposed, showing multiple cuts. A small test trench was opened into the floor, adjacent to the modern hole, and three more cuts were discovered, two large ones and a smaller one (Figure 11). Less than 2 m to the northeast of Sector E, four more circular cuts came to light in the bedrock bottom of a foundation trench of the later temple, visible because the blocks had been stolen.

Figure 11:

Ortophoto of the trench in sector E showing the bedrock with circular cuts and quarry cuts. On the right, the robbed foundation trench of the temple. (Marco Di Lieto/Caere project).

In total, 11 circular cuts were discovered, which can be divided into two size categories. Six of the cuts measure between roughly 70 and 95 cm in diameter, whereas four average between 25 and 30 cm in diameter. One of the cuts, located in the area of the modern interventions, is badly damaged and its original shape and size are difficult to determine. All bottoms of the cuts are flat, with vertical, smooth sides. The larger cuts in the main area have similar depths of around 50 cm, while the cuts in the wall trench are naturally much shallower due to the lowered level of the bedrock. The smaller cuts are all approximately 20–30 cm in depth.

Given the shape and placement of the cuts, it is assumed that these were the post holes for wooden structures. Fragments of clay plaster were found in the cuts as well as in other sectors of Area 4. There are two cases of the larger post holes bisecting one another, suggesting at least two phases of occupation. The archaeological deposit atop the bedrock, however, was completely removed during later phases of occupation, and consequently very few finds remain that can be associated with the structure. In addition, the small size of the area excavated thus far prevents us from identifying a pattern to these cuts.

The size of the post holes, however, can be used as a criterion to deduce a tentative dating and designation of the structures. The smaller post holes find good comparisons in the huts of the Iron Age and the earlier phases of the Orientalizing period. These huts typically have small to medium post holes, roughly 30–50 cm in diameter, and are often accompanied by thin canals encircling the perimeter of the structure or marking internal partitions. Examples of this style of hut in Etruria and northern Latium include those beneath the Portonaccio sanctuary in Veii, scattered across the Monterozzi plateau in Tarquinia, and on the Palatine Hill in Rome. It is not clear if the small holes in Area 4 belong to the same phase as the larger ones or to a different, possibly earlier time. On the other hand, large post holes that exceed 50 cm in diameter, such as the six found in Area 4, are naturally associated with larger, more substantial structures. Three such structures are found at Veii and Populonia.

On the Piazza d’Armi plateau in Veii, Oval Hut M has post holes ranging from 35–60 cm. The structure dates to the ninth century BCE and houses a human burial in the center, which is exceptional in Etruria. It was replaced by the rectangular timber Building H in the third quarter of the eighth century BCE with larger post holes measuring 50–65 cm in diameter. Even larger are the holes for the portico surrounding the later Building D, dating to the late seventh to early sixth centuries BCE.

Also at Veii, at the North-West Gate, the remains of at least two structures have been discovered with larger post holes. It is thought that there was a circular hut with post holes ranging from 60 cm to a meter in diameter, although the attribution of cuts to certain structures is less than secure, since the area had been cleared for the construction of the circuit wall. A later rectangular hut structure with post holes of similar sizes was built over the top of the previous circular hut. Although the proposed structures cannot be concretely dated, they may date from the eighth century and the first half of the seventh century BCE respectively.

Lastly, at Poggio del Telegrafo in Populonia, post holes of an oval hut structure measure from 70–90 cm in diameter, atop which a later rectangular structure was constructed with post holes ranging from 70 cm to over a meter in diameter. In one of the post holes from the rectangular hut, a deposition of nearly 100 single-handled cups suggests both ritual closure of a communal space in the early seventh century BCE, and the political/social importance of the building, informally dubbed the “King’s House.” The circular hut is dated to the ninth century BCE, while the rectangular hut is dated to the beginning of the seventh century BCE.

Given the comparable post hole sizes of the structures at Veii and Populonia, and the continuity of these spaces, it is possible that the large post holes found in Area 4 might belong to a similar structure or structures, possibly pertinent to the higher social levels of early Caeretan society. However, as mentioned earlier, the layers associated with the period of use of the structure(s) at Area 4 have been removed, and the area itself has been only partially excavated. The presence of few mid- to late seventh-century BCE fragments in one of the larger holes suggests that the posts may have been part of one of the larger timber-framed structures that developed in the Orientalizing phase before the appearance of architecture with stone foundations, as seen in Veii (Structure H) and Populonia (Rectangular Timber building). The smaller post holes, if they belong to the same phase, could have been part of the same structure. If they are earlier, as may be more probable, they document the presence of smaller huts at the site.

Small postholes and canals have also been found in other parts of the Vigna Marini Vitalini, in particular in Area 1 (later occupied by the hypogeum of Clepsina), along with remains of pebble floors and layers dating to the Late Iron Age. In Area 2, there is a larger post hole in an area of quarrying activity, which could be related to earlier dwellings (Figure 12). All this suggests the presence of a larger nucleus of habitation with multiple huts in the Vigna Marini Vitalini. The structures documented by the post holes were eventually abandoned to make room for a quarry that partially destroyed their traces. This was a radical “re-zoning” of the area, from a residential nucleus, possibly with elite structures, to a work site for the extraction of stone blocks. Angular vertical cuts are clearly visible near the post holes, indicating an open-pit quarry. On the surface of the bedrock, the outline of a block that had been delineated but never removed is also visible.

Figure 12:

Worked bedrock and large hole in Area 2. (Caere project).

Evidence of open-pit quarrying was discovered not only in the small test trench in Sector E, but in multiple locations in Area 4 (Figure 13). The almost complete removal of the blocks of one of the walls of the temple has revealed the bedrock underneath, with evident signs of step quarrying (Figure 14). The level of the quarry appears to become deeper moving from northwest to southeast. The period of the quarry’s activity must follow the abandonment of the structures with rock-cut post holes, which, as noted above, occurred at some point starting from the mid- to late seventh century BCE. The quarry was eventually closed when the area was filled, leveled, and a large beaten earth floor was laid down. The height of the block that is delineated in the bedrock is unknown. The visible side, however, measures roughly 85–90 × 40–45 cm, which is compatible with the temple’s foundation blocks and other monumental structures such as the temples of Vigna Parrocchiale and Pyrgi, but significantly different from those of the Southwest Building.

Figure 13:

Ortophoto of the excavation area with the vertical cuts for step quarrying visible in the robbed foundation trenches and in the trench of Sector E (Marco Di Lieto/Caere project).

Figure 14:

Bedrock with signs of step quarrying exposed in the foundation trench of the temple. (Caere project).

There are multiple cases of urban quarrying in Caere. In the same Vigna Marini Vitalini, quarrying activity that dates to the sixth century BCE was revealed in Area 2. In this area a smoothed bedrock surface can be found with a raised lip that separates this worked floor area from the deeper quarrying activity on the other side (Figure 12). A large round hole, discussed above, might also be the remains of an earlier structure. In the same area, a few meters northwest, the exposed bedrock shows the outline of multiple blocks carved into the bedrock (Figure 15). This area was covered with a thick fill rich in Archaic pottery.

Figure 15:

Delineated blocks and cuts in sector F2 of area 2. (Caere project).

In Area 1, around the later hypogeum of Clepsina, there is further evidence of Archaic quarries that were backfilled and built over. In this area we can also see post holes that were cut through by quarrying activity, as well as the remains of canals cut into the bedrock (Figure 16). In the adjacent Vigna Parrocchiale we can see a sequence similar to the one observed in Area 4 of Vigna Marini Vitalini, with a very deep quarry (up to 11 m). This quarry functioned along with some nearby structures until it was filled and partially built over in the early fifth century BCE, when a large temple was constructed. A “service quarry,” opened to supply nearby construction sites with materials, is also present in the Manganello sanctuary, where the space was artificially filled to allow for the construction of the temple, a project tentatively dated to the fourth century BCE but possibly as early as the fifth century. Along the line of the defensive walls of Caere on the northwest side of the plateau there is extensive evidence of open-pit quarrying to extract materials for the city walls as well as to create a steeper rock face, further strengthening the defenses of the city.

Figure 16:

Bedrock with quarry cuts, holes and canals in Area 1 (Caere project).

Outside of Caere, quarrying is attested at other southern Etruscan cities such as Veii. In the territory of Veii, Archaic rural dwellings featured underground rooms obtained by quarrying the bedrock, with the extracted blocks used for the upper part of the structure. The same technique has been observed in urban contexts at San Giovenale and Acquarossa. In other cases, small quarries were located next to roads that were used to transport the blocks to their destination, as seen at the site of via Cesaro. Here open-pit quarrying and delineated blocks were found, similar to what we see at Caere.

Outside of Etruria, this pattern of tuff quarrying is evident in Rome during the sixth century BCE on the southwest side of the Palatine Hill, where there is also evidence of an Iron Age nucleus of habitation. The quarried tuff bed was discovered beneath two large structures (A and B) dating to the fifth century BCE, located near the sanctuary of Magna Mater. This is thought to have been a service quarry for buildings on the Palatine. The sequence of Iron Age occupation, followed by the use of the bedrock for the extraction of construction materials and then for the construction of new monumental religious buildings, is clearly reminiscent of what we find in the Vigna Marini Vitalini.

The opening of multiple, relatively small open air service quarries in the urban area of Caere, mostly dated to the Archaic period, indicates a phase of intense building activity and urban renovation. The quarry in Area 4 evinces a complete change in the designation of the area. Thus, there is no direct connection between the earlier wooden structures and the later temple. The model of the so-called cultic hut, frequently put forward in the many cases in which traces of Iron Age structures are found beneath the foundations of temples, certainly does not apply here, and it should perhaps be considered with caution in other cases as well.

Having considered the phases of both the hut occupation and quarrying activity in Area 4, the next stage of the archeological sequence is the construction of a large, uniform, and densely compacted beaten earth floor. This floor would be placed atop the fill of the earlier quarry and would be then cut by the construction of the later temple. The full extent of the floor surface was not revealed during last year’s excavation. The uncovered dimensions currently sit at roughly 18 × 17 m, but the only visible boundary is the Southwest Building; the floor surface extends beyond the excavation limit in all other directions.

A cross section of the floor is visible in several parts of the excavation site, including on one side of the foundation trench of the temple, where the blocks were removed for modern re-use (Figure 17). A thick gray layer rich in clay is visible in the lower part and is found in other sectors of the excavated area (Figure 18), suggesting that the old quarry was first filled with this kind of soil. On top of this layer is a leveling fill of soil with higher draining properties, which becomes increasingly deep towards the southeast of Area 4 as the floor of the quarry becomes deeper. The top level of the floor has a high density of small chips of tuff, topped by a thin, smooth, tightly compact floor surface with a higher clay component, consisting of multiple superimposed thin layers. This floor extends beyond the limits of the quarry and covers the area with post holes.

Figure 17:

Section of the beaten earth floor and the clay fill underneath visible in the foundation trench of the temple. (Caere project).

Figure 18:

Aerial photo of area 4. The arrows indicate the points where the clay layer has been exposed. (Marco Di Lieto/Caere project).

There are several clues for dating the construction of the floor. The floor installment would have necessarily followed the abandonment of the timber structures as well as the period of activity of the quarry but would have pre-dated the construction of the temple. The few finds from the post holes provide a broad and generic date to the Orientalizing period. The pottery pieces from the floor layer date to the Archaic period, probably to the second half of the sixth century BCE. Lastly, pottery from the foundation trenches of the temple dates to the late sixth or early fifth centuries BCE. In addition, the interventions in Room C in the Southwest Building date to sometime in the late sixth century BCE, and it is not likely that the door led directly into an open and active quarry. Considering all evidence, the floor may have been constructed in the second half of the sixth century BCE.

Given the presumably large size of the floor and the great investment of labor and resources to fill the large quarry beneath, not to mention the construction of the surface itself, we can assume that the space was communally significant. The creation of a large open space for communal use finds compelling parallels at other sites of Central Tyrrhenian Italy. In the cities of Satricum and Tusculum in Latium, for instance, there is also evidence for the development of forum-like open spaces near sanctuaries and public buildings in the sixth century BCE.

The obvious exception would be the Forum Romanum, traditionally dated at least a century earlier. However, new research by N. Terrenato and his team at Sant’Omobono on fluctuating Tiber River levels in early Rome has led to a reexamination of the layers beneath the Forum Romanum, and to a reconsideration of when the Forum was first formalized. Based on environmental and archaeological data, the sixth century BCE now appears to be a more probable date. Thus, the floor surface found at Vigna Marini Vitalini might be the attestation of an Archaic Etruscan forum. The limited data retrieved from last summer’s excavation, however, is insufficient to confirm conclusively what, for the moment, remains a tantalizing hypothesis.

4. The Temple and Its Terracottas (F.C.)

The area underwent a major transformation when a large new structure was built only 2 m from the Southwest Building, occupying part of the area of the beaten earth floor (Figure 3). The structures within the excavation area consist of two parallel walls, which continue beyond the excavation limit to the northwest and southeast, connected by a perpendicular wall. The layout, size, and technique of the building and the architectural terracottas found in association with it strongly suggest its identification as a temple.

The foundations (Figure 19) are set in regular trenches cut into the Archaic floor and the clay level beneath to reach the bedrock, whose depth increases moving towards the southeast due to the quarry’s stepped floor. Thus, more courses of blocks are necessary in the southeast section and fewer and fewer in the northwest, so much so that the continuation of the northeast wall is made visible only by a slightly depressed canal on the surface of the bedrock (Figure 20). The blocks that rested here have been removed for reuse.

Figure 19:

Map of the foundations of the temple (Marco Di Lieto/Caere project).

Figure 20:

View from NE of the foundation trench of the NE wall (Caere project).

In the southeast section, the foundations are much wider (2.4–2.5 m northeast wall, 2.2–2.4 m southwest wall), while the cross wall and the continuation of the side walls to the northwest are thinner (1.4–1.5 m). This section of the southwest wall has been completely robbed, but the foundation trench shows that its thickness was about the same. The external side of the northeast wall is aligned in both sections, while the internal side of the southeast sections is visibly advanced. This difference is not due to two different building phases. In the best-preserved side wall, the same foundation trench with a homogenous fill is used for both the narrower and the wider sections. Where the two sections meet with the transversal wall, there are some small adjustments to the blocks for better fit and partial superpositions. Thus, the different thickness of the foundations may reflect a difference in their structural function.

The walls are made of square blocks of a roughly standardized size. The vast majority are 45–48 cm wide and 42–45 cm tall, while their lengths vary, typically around 90 cm but with numerous longer and shorter blocks. This is similar to the format used in Temples A and B of Pyrgi and the temple of Vigna Parrocchiale. In some cases, the blocks were modified on site to adapt them to their intended location, reducing their size or cutting vertical and horizontal indentations. Large pieces of tuff, likely the by-product of resizing blocks, were wedged into the foundation trenches. No binder was used, but a very thin layer of gray clay is visible, especially on the horizontal surfaces. The wider foundations are made of courses of blocks in either a stretcher or header pattern, simply laid next to each other and not staggered (Figure 21). Five blocks in the former pattern correspond to two and a half in the latter. The transversal wall and the thinner side walls are made of one row of stretchers and a row of headers for each course.

Figure 21:

Wider section of the NE wall showing the superposed courses of blocks. The thinner section is visible in the background. (Caere project).

The robbing of a good part of the southwest side wall allows one to see that the blocks of the lowest course are placed to fill the cuts of the quarry, which are at an angle with the direction of the wall and support the courses above. As the courses rise, their orientation turns increasingly until they reach the desired alignment.

Of the building, only the foundations survive, and they survive only partially. Almost nothing remains of the superstructure after centuries of destruction, agricultural work, and predatory excavations. The 19th-century excavations were particularly damaging. It appears that the workers followed the level of the Archaic beaten earth floor, digging deeper only at some points. They had apparently noticed that no valuable pieces were found in the almost sterile fill beneath. The same workers may be responsible for the removal of a large number of blocks from the southwest wall, reused to build the wall along the southwest boundary of the field. The resulting empty trench was filled up with soil from the excavation, which contains some 19th-century pottery and pieces of marble and opus reticulatum from a Roman building, possibly the nearby theatre.

The stone drain is probably part of the same project as the temple (Figure 22). It runs along the southeast side of the building, filling the same foundation trench, and then takes a slightly divergent direction towards the northwest, becoming also more steeply inclined. Corresponding to the transversal wall, the temple is 11.63 m wide, but its length remains unknown. The part within the excavation area exceeds 17 m. It is also possible that only the central part of the building has been uncovered and its full extent will be revealed in the next campaigns.

Figure 22:

Drain along the SW wall of the temple, from SE. (Caere project).

Based on the principle that foundations reflect a superstructure, it is fair to assume that the building was divided into two parts, without internal partitions or supports. The structure is not so wide as to make covering its span with a timber roof without additional support technically challenging. The narrower foundations may correspond to walls of similar thickness, while a line of columns might be very tentatively reconstructed resting on the massive foundations of the southeast side. The construction of the temple is certainly later than the sixth-century BCE beaten earth floor into which its foundation trenches were cut. The finds from their fill are relatively numerous and indicate a tentative terminus post quem in the early fifth century BCE.

A large assemblage of architectural terracottas was found at the bottom of a large, regular trench, probably evidence of 19th-century excavations (Figure 23). Over 90% of the pieces are homogenous in typology, fabric, size and decoration, and likely belong to the same roof, a system of the so-called Early Second Phase. All of the remaining pieces are later. These are evidently materials discarded by the excavators because of their low commercial value. In fact, antefixes and other figural pieces are absent. Given their findspot and assemblage consistency, it is reasonable to assign them to the building where they were found, albeit in a secondary context.

Figure 23:

Modern trench found filled with architectural terracottas. (Caere project).

A tentative reconstruction can be provided for two elements of this roof: the raking sima and the eaves tiles. The sima (Figure 24) features a large and hollow base torus painted in a black, white, and red zigzag pattern; a painted fascia with a cable pattern of red ribbons with white borders on a black background; a cavetto with oblique black, white, and red bands; alternating red and black convex strigils, separated by white spaces; and a deep channel for the attachment of the openwork cresting on top. There are some fragments of fascia in a cable pattern of different colors – white ribbons with alternating black and red borders – and of a cavetto with bands inclined in the opposite direction. These may come from the opposite side of the pediment or the other front of the building.

Figure 24:

Tentative reconstruction of the raking sima. (Caere project).

The cable pattern has a long tradition in Etruscan architectural decoration, but on Early Second Phase simae it becomes obsolete after the first quarter of the fifth century BCE in favor of the meander, which is virtually ubiquitous afterwards. The sima of the temple of Juno Moneta at Segni is one of the most recent, if we accept the dating of C. Carlucci, who places it after the Temple of the Castores. Conversely, Colonna and Cifarelli directly connect the group of terracottas with the (second) settlement of Roman colonists in 495 BCE.

The sima from Segni is an especially good comparison for the association of the fascia with cable pattern and a torus with zigzag. The latter originates from patterns that appear at the end of the sixth and early fifth centuries BCE and takes this form in the 470s–460s BCE; it is used, for instance, in the second system of Veii-Portonaccio, at Falerii-Sassi Caduti, and at Pyrgi on Temple A. In the ensuing period, horizontal garlands of leaves become the standard decoration for the torus. The sima from Segni may be one of the most recent of this kind with a zigzag torus.

Thus, the sima of Vigna Marini Vitalini appears to be later than the systems of the first 20 years of the fifth century BCE, but earlier than those of 470–460 BCE, where the zigzag torus is associated with a fascia with meander. All clues point to a dating in the 470s, which fully aligns with the dating of the structure indicated by the finds from the foundation trenches. No pieces of the cresting and the revetment plaques associated with this sima have been identified so far (Figure 25).

Figure 25:

Reconstruction of the pediment of the temple with the raking sima. (Caere project).

The other element that can be reconstructed is the eaves tile (Figure 26). The margin is painted with a simple red, white, and black meander, while the underside is decorated with a pattern with red and black bands framing a complex red and black meander that includes checkered squares. Based on the painted pattern, the eaves tiles were probably 48–50 cm wide, which matches the standard size of the tiles used in Temple A of Pyrgi. On each tile there were two complete checkered squares and two partial ones on either side, which matched those on the adjacent tiles. The pattern was painted with the aid of a grid of incised lines (Figure 27). The tile was completed by a hanging drip, attached with molten lead poured into rectangular holes near the margin (Figure 28).

Figure 26:

Reconstruction of the painted pattern of the eaves tiles. (Caere project).

Figure 27:

Fragment of eaves tile with incised guidelines. (Caere project).

Figure 28:

Fragments of eaves tiles with lead attachments for the hanging drips. (Caere project).

The painted decoration finds a good comparison at Caere in the Vigna Parrocchiale (ca. 480 BCE), on a tile with a one-piece integrated drip that is probably slightly earlier. This type of eaves tile is common at the beginning of the Second Phase and is soon replaced by the type with attached drip. The eaves tiles of Temple A of Pyrgi are similar in shape to those from Area 4 but with simpler decoration. A closer comparison comes from Vulci-Fontanile di Legnisina, although its dating is not firmly established. Tiles with various forms of complex meander and checkered squares are also found in Latium, for instance Lanuvium and Ardea. Thus, the most probable dating for the eaves tiles corresponds with that of the raking sima, approximately 470s BCE.

A large number of architectural terracottas were excavated from Vigna Marini Vitalini in the 19th century and dispersed to numerous museum collections. The trench where our terracottas were found was most likely produced by one of those excavations, and our pieces are those that were left behind. Therefore, it is only logical to look for other elements from the same roof, the ones that were removed and sold, among the pieces with traceable provenance from the Vigna Marini Vitalini. Among those pieces, three shell antefixes with a head of silen and maenad stand out (Figures 29–31). They are currently divided between the museums in Berlin and Copenhagen. They are dated stylistically to ca. 470 BCE and their fabric is macroscopically consistent with that of our pieces. Their provenance is certain: they were found in the excavations of M. Lazzari, A. Jacobini, and A. Lauri, with the assistance of W. Helbig. Their excavations took place “behind the theatre,” as confirmed by archival documents, in the same general area excavated in 2024.

Figure 29:

Berlin, Antikensammlung. Female head antefix TC 6681.2b, from Vigna Marini Vitalini (Antikensammlung, Staatliche Museen zu Berlin – Preußischer Kulturbesitz).

Figure 31:

Copenhagen, Ny Carlsberg Glyptotek. Silen head antefix HIN 55, from Vigna Marini Vitalini (Ole Haupt, Ny Carlsberg Glyptotek).

Figure 30:

Berlin, Antikensammlung. Silen head antefix TC 6681.2c, from Vigna Marini Vitalini (Antikensammlung, Staatliche Museen zu Berlin – Preußischer Kulturbesitz).

This type of shell antefix is rare and typical of Caere. The three pieces from Vigna Marini Vitalini show refinement, with an openwork shell and hook meanders on the base. The same types were used for the antefixes of Temple A of Pyrgi, where the shell is solid, the base is wider, the painted meander is simpler, and the dimensions are also slightly smaller. These are clearly derivative pieces and must be from a slightly later date. Rather than for economy of production, the modifications of the Pyrgi pieces may have been introduced to enhance durability, albeit at the expense of the refinement of the delicate original pieces. It is noteworthy that the painted meander on the base of the Berlin and Copenhagen antefixes complements nicely the intricate pattern on the eaves tiles found in 2024, whereas the base of the Pyrgi antefixes matches the margin of the associated eaves tiles and agrees with their underside, which bears a simpler decoration made of larger elements.

The excavations of R. Mengarelli in the north part of the Vigna Parrocchiale in 1912–1913 brought to light an artisanal compound with water reservoirs and kilns, which he called “the temple of Hera” after some inscribed vessels (Figure 2). This location, only 150 m north of the temple, is where these antefixes were produced, as indicated by a mold for the shell found by Mengarelli. The face of the maenad or silen was obtained from a different mold and then attached to the shell, which is the same for both types. In the Berlin and Copenhagen pieces, the ears of the silen were also made separately and applied by hand, as shown by their slightly different position on each piece and are completely detached from the shell. However, on those from temple A at Pyrgi, the ears are attached to the shell and therefore less exposed to damage. As previously mentioned, there are also fragments of other elements, mostly crestings and plaques, that can be dated to the late fifth and fourth centuries BCE. Rather than parts of other buildings, they may be replacements for damaged elements of the original temple.

In conclusion, the 2024 excavation has revealed a significant addition to the urban topography of Caere and more specifically of this area; it was certainly one of the hubs of the Etruscan community. The location is central and easily accessible, and the terrain is plain and regular, especially along the modern dirt road. It was also occupied by monumental structures of different functions and chronologies, from the Archaic period to the Roman era, all sharing the same orientation from the outset: in Area 1, the structures date back to the early seventh century BCE, if not earlier. This, however, was not a gradual development over a long period. As we have seen, the archaeological record from Vigna Marini Vitalini shows a trajectory of radical changes and sharp turning points. The public character of this space becomes fully established especially from the early fifth century BCE, as we can observe in the area of the temple, but also in Area 1 and in the adjacent Vigna Parrocchiale, where structures with different functions – both residential and productive – are eventually replaced by communal facilities.

The Vigna Marini Vitalini and the Vigna Parrocchiale should be considered an integrated system, together with the Vigna Grande, of which we know regrettably very little. However, we do know that fragments of the same types of Archaic and Classical architectural terracottas were found there, along with sculptures that must have adorned the forum in the Roman period, and even an inscription that mentions a curia. Thus, the excavation of 2024 has opened promising avenues for further research. An investigation of the site must be continued and refined in order to gain a fuller understanding of the situation and to validate these interim proposals.