The 2018 excavation campaign of the University of Innsbruck at Ascoli Satriano, loc. Giarnera Piccola, took place between August 22 and September 12, 2018. The team was composed of archaeologists and students of the department of Archaeology (Classical and Provincial Roman Archaeology Section) of the Leopold-Franzens-Universität Innsbruck. An anthropologist was present to study the human skeletal remains. All works were carried out under the direction of the Soprintendenza Archeologia, Belle Arti e Paesaggio per le province di Barletta – Andria – Trani e Foggia.
Most of this year’s investigations focused on sectors that had already been studied in 2017: work continued in the north-eastern part of sector F, in the southern part of sector G and in the south-western part of sector H. Research concentrated on the Archaic period, and at the same time the study on the relation between burials and buildings in the Daunian period continued. In addition, a geomagnetic survey was carried out.

Sector F
In 2017, sector F had already been extended by approximately 130 m² to the north-east. The new part is characterized by a large number of (structural) elements, spread over two terraced areas. Some of the structures can be assigned to the Archaic period, while other structural discoveries allow a dating in the late Classical – Hellenistic phase. In the terraced area to the north, four phases of construction (phase A-D) can be identified at the current state of investigations. The most recent phase (phase A) was documented by a dwelling unit (building 20), partially excavated in the eastern corner of the extension of sector F, can be dated to the last third of the 4th century BC. The dating of the building is based on the use of clay bricks as a construction material and on pottery fragments.
The foundations of this structure cut through the levelling layer SE 1238 that covers a slightly earlier construction (phase B). The most conspicuous context of this phase is the layer of pebbles (SE 1261) located at the northern edge of the enlargement, undoubtedly continuing outside the excavation area.

Immediately below the phase B layer, a series of small pits have been identified that can be interpreted as postholes of Archaic date (phase C), most of which, namely SE 1241, 1243 – 1247, 1249 – 1255 and 1262, belong to a rectangular shed structure (structure A). The intermediate layer of levelling was only a few centimeters strong. The position of structure A above the southern part of hut 2 suggests a more recent date. To date, the width of the chronological gap between these two buildings remains unclear.
The postholes of structure A were dug into layer SE 1239, located above the oldest stratum of the eastern part of the terraced area (phase D). This use layer was identified already 2017 in the passage of the two terraced areas.

In the 2018 campaign, works concentrated on the northern strip of the terraced area. In the southern part only a small part NW of the small pits SE 1267 and SE 1268 was studied. In this section it was possible to identify a layer of cobblestones (SE 1286), with a rather wide chronological range within the 4th century BC according to the material recovered (fragments of pottery and bricks).
The most important finds in the northern part of the terraced area will be presented below, separated in phases (partly these are discoveries of this year together with some situations already identified in 2017 and completed in 2018).

Phase A
The remains of building 20 were not further investigated in 2018. Due to the very limited excavation area, the continuous excavation work will only make sense after a further extension of the excavation area. However, it was possible to assign the pits SE 1259, 1260 and 1285 (discovered in 2017) to this layer. With a maximum size of 1.95 × 1.25 × 0.15 m they showed a filling consisting of stones, fragments and splinters of bricks and mixed soil. Even after the completion of the works, it was not possible to define their function more precisely.

Phase B
Cobblestone layer SE 1261 was only a few centimeters strong and therefore could be interpreted as the basis of the walking surface. Similar findings from the Archaic, Classical and early Hellenistic periods have already been documented in other parts in the Giarnera Piccola.
To the same phase can also be assigned structure SE 1257. It is a shallow pit (0.15m) of which the bottom is covered with a layer of stones and brick fragments that could also be interpreted as the remains of a foundation that unfortunately cannot be specified with greater precision.
According to the evaluation of the various findings, the floor layer SE 1258 adheres to the structure SE 1257 – recognizable in particular on the west side – and presumably is also related to SE 1261. The initial level of the phase B layers is also tangible thanks to a third part of the cobblestone (SE 1239b), documented north of SE 1256.
In 2018, two additional pits were identified (SE 1301 e SE 1325), which were also cut in the upper edge of the levelling layer SE 1239. Pit SE 1301 is located north-east of the structure SE 1257, the pit SE 1325 south of SE 1261. Both pits are oval in shape, with dimensions from 0.6 × 0.5 to 1.0 × 0.7 m and a maximum depth of 0.15 m. Their filling material consisted of soil, stones and brick fragments.

Phase C
A main objective of this year’s excavation campaign was the excavation and documentation of the cavities interpreted as postholes SE 1240-1252, 1254-1256, 1260, 1262 and 1263, most of which belong to structure A. Additionally, ten more postholes were identified. Six of these (SE 1304, 1305, 1306 and 1307) were found below, to the east and north (SE 1289 and 1321) of SE 1261 (SE 1322). Two others (SE 1299 and SE 1300) have been found under a pile of stones (SE 1242), further two under pits SE 1259 (SE 1308) and SE 1260 (SE 1309) in the passage to the terraced area of the southern settlement. The holes SE 1299 and SE 1300 were located in the overlapping area of hut 2 and structure A. Their assignment to one or the other structure is still to be ascertained.
Below structure SE 1257 a cobblestone layer has been found, eventually to be interpreted as a phase C walking surface possibly belonging to structure A. In this case, the two smaller foundations might be regarded as the remains of a sort of fence.

Phase D
In 2018, the presence of a fourth floor layer (SE 1298) was confirmed. This layer had been found in some sections of the phase B postholes. It was particularly well distinguishable in the section of SE 1252 and characterized by the presence of stones and pottery fragments on the surface. In the sections of the phase C postholes, an elevation of the ground to the north (towards hut 2) could also be detected for this phase, documented both by the thinning of the levelling layer SE 1239, and by a slight elevation of the virgin ground SE 3/99.
The main objectives of future campaigns will be a reconstruction of the layout of structure A (phase C) and its overlap with hut 2. Further investigations will also be necessary for the chronological delimitation of the two buildings, since both the date of hut 2 and that of structure A have so far remained vague.
Another important objective for the 2019 campaign will be the excavation and documentation of the oldest activity layer (SE 1298) of this area (phase D), which has yet been discovered only in the sections of the postholes.
Important information for the chronological classification of structure A had already been identified in the leveling layer below (SE 1239). During the excavation, fragments of black-glazed pottery emerged, while the most recent material, such as tiles, is completely absent. This makes a 5th century BC date plausible both for structure A and for the ‘shed’.
In the western part of sector F, pit SE 1287/1288 was excavated. The oval-shaped context with a maximum diameter of 1.45 m was located west of the slope wall SE 835 of Building 11/Phase 1. The filling material consisted of stones, mixed soil and a large number of brick- and pottery-fragments. In addition, animal bones and remains of charcoal were recovered. In view of similar findings in the locality of Giarnera Piccola, pit SE 1287/1288 could have a ritual function. Inside it, an elongated stone formation was found (SE 1292), which continued also to the west and can probably be interpreted as the remains of the foundations of a wall. This finding, 1.05 m long and 0.25 m wide, belongs stratigraphically to the wall SE 781, i.e. to the previous phase of building 11/Phase 1, and was arranged perpendicularly to it.

Sector G

Early phase hut 1
Already in 2017 it was possible to document an earlier phase under hut 1. The oldest stratum was identified under the northern part of hut 1 and consisted of the approximately flat occupation layer SE 1190, the burnt patch SE 1183 and small postholes SE 1130 and 1189. These findings suggested an older shed-like structure under Hut 1.
This hypothesis could not be verified as the oldest floor stopped to the north and south and the southern part of SE 1130 was situated in the virgin ground. Therefore, in 2017 trial trench (1/17) was carried out in the southern area, to identify the presence of an alleged older phase. In this trench, at a slightly lower level, a presumably older layer was identified, SE 1190b, which was identified thanks to the discovery of pottery fragments widely distributed on this level. The same layer is associated with posthole SE 1326, identified in 2018, and a possible hearth (SE 1279). During the 2018 excavation campaign, this trench was extended under hut 1 to the north and west to clarify the stratigraphic affiliation of the previous contexts SE 1326 and SE 1279. In the western extension, two more postholes have been found (SE 1319, SE 1320) and to this older phase presumably previously excavated postholes such as SE 1061, 1062 and 1074 can be ascribed.
Thanks to this information gathered in the last two excavation campaigns, it has been possible to hypothesize one or even two phases prior to Hut 1. However, it has not yet been possible to completely define the arrangement of the structures and clearly assign their elements. For a definitive reconstruction, further investigations will therefore be necessary. Based on the subgeometric pottery fragments found in the last two excavation campaigns, the layer(s) underneath hut 1 can be dated within the 8th/7th century BC.

Pit SE 1019/1020
This large oval-shaped (about 6.0 × 5.0 m) pit was discovered as early as 2013, but was not yet completely excavated. In 2018, a section was made and half (west side) of the unit was excavated. The limits have a convex profile with a flat bottom. The pit reached a maximum depth of 0.75 m. At least three different layers of filling were identified. The central package showed a gravelly consistency, while the other two were of a humic character. All three layers contained a rather large number of ceramic material, mixed with fragments of burnt clay/daub and animal bones. The lower and upper layers also contained a considerable amount of charcoal. The uppermost filling layer could be interpreted as a pit dug independently of the other two as a waste deposit during the construction of a nearby shed structure. In the southern part of the SE 1019/1020 bottom, a further and deeper shaft-like pit was documented. It is not yet clear whether it was part of the initial pit or a subsequent intervention, since especially in this part of the section the stratigraphic order of the filling packages is blurred. It will be a future task to clarify this situation by fully excavating the structure. The earliest dating of the pit SE 1019/1020 according to the subgeometric pottery fragments dates back to the 8th/7th century BC while its function is still unclear.

Pit SE 968
This pit was identified several years ago together with pit SE 1019/1020, but until this year it was not possible to carry out in-depth investigations. It was cut in the area south of SE 1019/1020 but the discovery of ceramic material from the 4th century BC provides a much more recent date than the pit SE 1019/1020. On the surface, the pit was circular in shape with a diameter of about 1.70 m, the edge was reinforced by a circle of stones and pebbles. The filling material consisted of brown clay soil. On the northern, western and eastern outer sides, the pit was lined by gravelly soil (SE 1324), which presumably can be interpreted as filling material of an additional (building?) pit that was distinguished in the pit section SE 1019/1020 with almost vertical walls and a depth of at least 1,00 m. Since the bottom of the pit has not been reached and the stratigraphic relationship with the most recent pit SE 968 has remained unclear, further investigation will be necessary.

Tomb 1/17 (SE 1269 and 1294-1297)
Tomb 1/17 is located north of Hut 1. Its dimensions are approximately 1.05 × 0.5 m and it had a cover consisting of three tiles arranged in a rectangular manner, one of which was used entirely, a second one only half while the third tile has been preserved to about a quarter. The badly preserved remains of a prematurely born child (SE 1296) whose sex could not be ascertained were placed in the rectangular pit. The head was positioned to the south-west. The funerary equipment (SE 1295) consisted of six ceramic vases and a bone object deposited in the central area of the pit. The ceramic material consisted of two vases with banded decoration: a pitcher (no. 31, 37/18) and a single-handled cup (no. 38/18), a small bell crater in “mixed style” (no. 40/18), a black-glazed miniature Oinochoe with red overpainted decoration (no. 41/18), an undecorated cup (no. 42/18) and an impasto jug (no. 38/18). The assemblage dates to the second half of the 4th century BC.

Area east of tomb 1/17
Immediately to the east of the tomb 1/17 already in 2017 the floor layer SE 1270 was discovered that belonged to the oldest level of the hut 1. Its dating has been confirmed by two rim fragments of a subgeometric Daunian olla from the 8th/7th century BC found on the surface. This floor can be associated with two postholes, SE 946 and SE 948, located immediately to the east. To the south-east of these two holes, in the border area between sector F and sector G, another posthole (SE 1291) belonging to the same structure was discovered that for the time being has just been documented on the surface. The surface extends eastwards where it leads under the levelling package SE 961b and will be subject of further investigation in the next excavation campaigns.

Sector H
Area east of building 12
The already visible postholes SE 1037, 1041, 1049, 1052 and 1225 and the hearth-like structure SE 1284 were examined. During this work, three further units were discovered (SE 1303, 1332 and 1334) in the immediate vicinity. Posthole SE 1334 came to light by cutting through the pit SE 1303 and was covered by the material SE 961b. In the section of SE 1049 a layer of cobblestones (SE 1290) was documented that could be interpreted as a floor plan of an earlier stage. Both the SE 1334 posthole and the SE 1290 cobblestone layer can stratigraphically most likely be connected with the postholes SE 946, 948 and 1291 in Sector G and thus belong to the oldest Archaic layer under Hut 1.

Tomb 9/13
This tomb was located below the wall SE 909 in the eastern part of the polygonal courtyard structure, where the wall was collapsed and considerably sunken into the ground. After removing the collapsed part, the filling of the shaft was removed and the tomb cover was documented (SE 1302). It covered an area of 2.05 × 1.30 m and consisted of several limestone and sandstone slabs of dimensions of max. 0.80 × 0.50 × 0.10 m. The marginal area had a border of pebbles, flakes and stones.
The elongated rectangular pit measuring 1.6 × 0.75 m contained a single deposition (SE 1313). The deceased was placed along the western side, his face eastwards, in a crouched position with his contracted legs resting on the right side. According to the anthropological study, the remains can be assigned to a man between 35 and 45 years of age. The pottery was deposited along the eastern side of the pit and in the space at the foot of the deceased in the northern area of the pit. The grave goods consisted of 13 pottery vases and an amber pendant. Seven vases were of subgeometric Daunian ware: an olla (no. 117/18), a stamnos (no. 120/18), three jugs (nos. 124, 126 and 128/18), a high-stemmed cup (no. 121/18) and a bell crater (no. 125/18). Inside the olla a miniature black-glazed jug (no. 118/18) was found. In addition, a black-glazed kylix (no. 123/18), a black-glazed kantharos (no. 129/18), a brownish-glazed cup (no. 119/18), an undecorated krateriskos (no. 122/18) and an impasto-jug (no. 127/18) were recovered. The assemblage allows a dating in the second half of the 4th century BC.
In the central part of the pit, after the removal of the tomb floor and the grave goods, a previous burial (SE 1317) was detected, placed secondarily in a small cavity (0.70 × 0.40 m; SE 1314). The skull cap and the burial basin SE 1317 were located to the north, the lower jaw on the opposite side and the long bones in the middle. Among the bones were found fragments of different iron fibulae, a bead of glass paste and a bone ring (SE 1316, nos. 146-149/18).
The major part of tomb 9/13 is located under the entrance area of the polygonal courtyard, only the northern part of the burial extended under the wall SE 909. Similar situations of overlapping structures and burials have already been observed in other cases such as tomb 4/07 (building 2), tomb 10/13 (polygonal courtyard) and tomb 3/12 (building 9).

Geomagnetic survey
In addition to the excavation, a geomagnetic survey was carried out. First of all, the study focused on the western and eastern areas of the already open area (sectors F and G). In addition, the area to the south and east of the current excavation area were also examined, remaining within the property acquired by the municipality of Ascoli Satriano for archaeological investigation. The image of the magnetometer presented large elongated anomalies that extend diagonally from north-east to south-west, which are most likely of geological origin (furrows of erosion). In addition, dipolar disturbances could be detected in some places, probably caused by rather large metal objects inside the ground.
There are also circular-shaped anomalies of various sizes, probably partly interpretable (especially the smallest) as postholes. However, it was not possible to reconstruct coherent structures. Some of the larger circular anomalies could be interpreted as tombs. To the east of the current excavation area it was possible to identify a point with anomalies arranged at right angles that could indicate the remains of a rectangular building facing north-west/south-east.