Charles George, Rossana Valente
Nezi Field Excavations
N: 1015.90 N, S: 1006.90 N, E: 264.20 E, W: 255.50 E
28 May – 22 June, 2012

This is a final summary of the third season of excavation (28th May – 22th June) in 2012 in the northwest area of Nezi field. Guy Sanders (director) and Heather Graybehl (field director) supervised. The blue excavation team consisted of Charles George and Rossana Valente (recorders), Athanasios Sakellariou (pickman), Athanasios Notis (pickman), Christos Sakellariou (shovelman), and Pavlos Sennes (barrowman).
Excavation began in the area bounded by the Giambouranis house (NB 252, NB 262, 1015.90 N) to the north, Wall 540 to the east (264.20 E), Wall 366 to the south (1006.90 N), and the balk of Nezi field (255.50 E) to the west. The lack of preserved architecture, except the partially preserved Wall 747, led us initially to interpret our area as exterior space associated with the Byzantine room to the south, bounded by Wall 366 and Robbing Trench 497 to the north (1006.00 N), Wall 365 to the south (1002.10 N), Wall 332 to the east (262.07 E) and Wall 306 to the west (258.01 E). The goal of this session was to further explore the Byzantine habitation layers found during earlier excavations in the area (2008 and 2009, sessions 1st and 2nd 2012), with the hope of better understanding the use of space in this area during Frankish and Byzantine times and also of reaching Late Roman material by the end of the season.

Late Antique (5th-7th AD)

We have reached Late Antique contexts in the northeast of the area which are overlaid by Wall 540, dated to the late 11th c. For this reason, we cannot see the eastern ends of the contexts, and we suggest close cooperation and contact with the Pink Team’s excavations of the area east of the wall.
We have identified a mid 6th- early 7th c. AD floor (S 936, removed as 957), which seems to continue under wall 540, and we argue that it is equal to a floor in the Pink Area (S941). This context contained two fifth century nummi (coin no. 2012-159, 160), with five fifth century nummi (coin no. 2012- 162-166, 169) also found in what we interpret as the continuation of the floor to the north (959), and two fifth century nummi (coin no. 2012-161, 164 ) found in a suggested floor surface below S 936 (958). The excavation of the floor has revealed another possible floor, which itself was laid over another possible floor surface. This may suggest that we have a stratification of floors indicating several successive habitations during the Late Antique period. The section of the floor called 959 seems to be laid on an E-W wall (S 918), therefore dating the wall earlier. It is unclear when and under what circumstances the wall later was robbed out. However, further excavation in the area and removal of wall 540 should elucidate the relationship.

Above floor S 936 is a tile destruction layer (934), which is likely equal to the tile destruction layer across wall 540 on the Pink Team’s tile destruction layer 929. The elevation of Pink 929 is roughly equivalent to that of Blue 934 and part of Blue 931, a leveling fill directly above the destruction layer. As supporting evidence for this relationship is an AFRS form 99, which has joining fragments both in Blue 931 and Pink 929.

Vast quantities of sherds of Late Antique pottery have been found in the fill of a Late Byzantine lime pit (the eastern part of pit is 878 and western part is 917), an ovular Byzantine pit (888), and adjacent cleaning contexts and leveling fills (895, 906, 907, 908). Also, a Late Antique glass bottle (MF 2012-50) and jar (MF 2012-51) have been found in the lime pit (917). While we have interpreted these all as redepositions, the large size of the glass sherds and the size and density of the pottery suggest that the Late Antique material in these fills does not show signs of frequent redeposition.

Late Byzantine (1059-1210 AD)

The most important feature identified as Late Byzantine is the western portion of wall called 747, which dates to the late 11th c. The excavation of the fill to the east of western wall 747 has revealed a finished surface to the wall and an apparent end to the foundation beneath the wall. We think that in this space East of Western wall 747 there was an entrance. As the Western and eastern portions of wall called 747 are on the same axis, despite the gap and differences in construction, we maintain that they are contemporaneous and of the same structure. We still hope to determine whether there was an interior space north or south of the wall. North of 747 may have been exterior as suggested by the massive pit there (C 870, fill 844 and 868, W-E 258.43-261.89, S-N 1011.63-1014.56), which we have identified as a bothros, since it contains a mix of pottery ranging from the Geometric Period to the late 11th c.

Western Wall 747 seems to be surrounded by several leveling fills that are also dated to the Late Byzantine period. These leveling fills are both north (854, 855, 857, 858, 859, 863, 868, 869, 875, 876, 925, and 927) and south (897 and 904) of Wall 747. These leveling fills characterized by a yellowish brown soil, are poorly sorted, and there are not many material inclusions. Small quantities of pottery are found across these contexts. These small contexts are overlaid in several areas by various lenses of leveling fill, often consisting of harder-packed and different colored soil, but the pottery dates from all of these contexts suggest that they are all part of a relatively contemporaneous leveling of the area. None of these contexts has a precise date for the Late Byzantine period; we can relate this context to the Late Byzantine leveling activity of this area mainly by stratigraphic relationships. In the middle Byzantine dumped fill of a robbing trench along the W Scarp (950), we have found a likely-Christian clay bread mold (MF 2012-47). These fills seem bounded by Wall 747 in the south but extend as far as the Northern Balk.

We have found a lime pit (C 882, W-E 257.54-260.87 S-N 1008.88-1010.10, filled by 878 and 917) directly south of Western Wall 747. All the preserved sides of this pit are covered with lime. It is possible that the walls of this cut were intentionally covered with lime, for building purposes. For example, the lime pit may have been used for the construction of walls 747 and/or 540, which both also seem to date to the late 11th c. In a later action, this cut was filled by a dumped fill (878 and 917). The filling of this cut, dated to the late 11th c, is a secondary deposition of large quantities of Late Antique pottery (5th – 6th C). Excavation of the oven (S 910) and the area between the oven and the pit will elucidate the issue and will provide confirmation as to the western boundary of the pit, which is not yet certain.

Last but not least in importance is pit 888 (W-E 262.21-263.10, N-S 1009.06-1007.06). Partially excavated during session 1 (888=680), this pit seems to be a bothros. Its dumped fill, dated to the 11th C, seems to be a secondary deposition of much earlier material including large quantities of Late Antique (5th – 6th C) pottery, Late Antique coins (2012 22-29, 105-126), and a 2nd-3rd c. incised intaglio (MF 2012-33).

Frankish (1210-1458 AD)

During the Frankish period E - W Wall 747 was partially robbed out. As noted during session 2, certain fills (833, 744) excavated in the robbing trench for this wall were distinguished both in materials excavated and soil composition, and should be considered the first robbing episode of this structure.
S of this robbing trench, we excavated a Frankish pit dated by the pottery to 1270 (C 926 filled by 883, 884, 885, 887, 889, W-E 261.88-264.03, S-N 1010.35-1008.90). The pit was particularly rich in finds, including luxurious items such as gilded bone bands (MF 2012-43, 45), incised bone knife handles (MF 2012-48), and gilded bronze flower-shaped clamps (MF 2012-53A, 53B, 54, 55) as well as stone- and metal- working tools such as a lead cupellation bowl (MF 2012-38) and an iron chisel (MF 2012-40). Also found was a pecten shell used by pilgrims travelling on the Way of St. James to Santiago de Compostela (MF 2012-34).

Below the pit, we came on the head of a well, a cistern, or a manhole (S 902 W-E 262.23-262.71, S-N 1009.55-1010.05). The structure is lined with plaster, and goes down 2.35 m, without any visible fill except for a small accumulation of debris, much of which is from our excavation of proximate contexts. The hole seems to have been intentionally closed with blocks. That it is a manhole is suggested by the slightly ovular shape of the hole, but it appears to be self-contained in all horizontal directions, which makes it more likely that it was a cistern or a well. That it is a well is suggested by the nearly intact water jug that was found at the bottom of pit 889, just centimeters away from the hole. Further excavation is needed to
Another Frankish pit (C 871, filled by 860, W-E 255.51-256.95, S-N 1011.92-1013.63) was excavated during the second quarter of the 13th C, as its dumped fill suggests. This pit, located next to the W scarp of the excavation area has been interpreted as a bothros. Interestingly, the N side of the pit scarp is full of tile. A few large, unworked stones (possibly mined bedrock) are visible at the bottom on the E side.
These pits may have been created for the purpose of storage. Generally Frankish pits were made along walls, as the surface of the wall offered an easy ready-made boundary for the pit. Our Frankish pits C 871 and C 926 correspond to this interpretation as do two Frankish pits excavated by the Pink Team (C 847, C 867).

Early Modern (1831-1945)

Context 880 is the only context that has been dated to the Early Modern period, specifically Turkish II. We have interpreted the Turkish finds as contamination from the robbing trench for wall 747 (759), and therefore we interpret the context as a disturbance of the Frankish pit (C 926).

Recommendations

We recommend to future excavators the following:

1. Take down Wall 540. We are now certain that the Late Antique floors in the northeast of our area (S 936, 939, and those stratigraphically below) continue under the wall and into the Pink area (see above). The excavation and dating of the wall would remove an obstacle to excavating in the area and would be helpful for dating in the area. Lastly, there are many walls in the Blue (S 747 and S 918) and Pink areas that misalign by as few as 0.10-0.30 m, and the excavation of the wall would help in seeing more clearly whether they are related and how.

2. Excavate oven S 910 and the surrounding contexts in order to determine the relationship between the oven and the Byzantine lime pit to the east (C 882).

3. Excavate floors S 911 and S 912 to discern their dates as well as the relationship between the two floors and between the floors and the surrounding area

4. Take down the contexts east of 949 and then proceed to finish its excavation. Digging in this area is important for understanding the boundary between the Late Antique material to the east and the 11th c. lenses of fill 844 to the west.
We hope that in the future, interior and exterior areas will be able to be identified, and that the Late Antique levels, now appearing in the NW, will become apparent throughout the area.