Room 8 [N: 1088.00 N, S: 1077.00 N, E: 129.00 E, W: 121.00 E]
Room 7 [N: 1086.30 N, S: 1078.20 N, E: 123.90 E, W: 119.40 E]
Rodríguez-Álvarez, Emilio
Session III

Introduction:
This is the final report for the third session of excavations in the 2015 season for Rooms 8 and 7, Unit II, in the area of Temple E SE. Room 8 was first excavated in 1995 (NB 877: 73–198; NB 845: 33–75) and then between 23rd and 25th of April 2014 by Rous and Worsham. Work was briefly resumed between the 19th and the 21st of May 2015 by Tanaka and Rodríguez-Álvarez and carried out by Rodríguez-Álvarez in June 2015. Room 7 was first excavated in 1992 by Schmalz (NB854) and briefly explored in 1994 (NB 864). Excavation was resumed between the 5th of May and the 29th of June in 2014 by Burr and Rogers and continued by Rodríguez-Álvarez between the 16th and the 19th of June 2015. Dr. Guy Sanders (Director) and Larkin Kennedy (Field Director) supervised. Emilio Rodríguez-Álvarez recorded. Kostas Arberoris excavated and Thanassis Notis and Vassilis Kollias sieved.

Goals of excavation:
The excavation of this area had two main aims: first, to explore the use of the area through time, before and after the space was defined as Room 8 by the erection of Wall 154 (1085.91–1086.65N, 124.04–126.17E) and the abutting Wall 166 (1086.27–1087.30N, 126.20–130.82E) at the north, Wall 720 (1081.10-1079.20N, 128.55-127.50E) and 830 (1084.10-1081.10N, 128.30-126.70E) at the east, Wall 156 (1077.50–1078.97N, 123.58–128.40E) at the south and Wall 157 (1078.37–1085.05N, 122.55–124.63E) at the west. Secondly, to assess the correlation between Room 8 and some of the finds made by Smotherman and Brannelly in Room 9 during the previous session, specially the pit (Context 733/717/726/747, N. 1083.88; S. 1079.99; E. 130.80; W. 127.50) sectioned by Wall 830, which seemed to extend into Room 8. When the excavations in room 8 revealed a second pit shared by Rooms 8 and 7, the correlation of finds with the material recovered by Burr and Rogers in 2014 and the relationship between the pit (Cut 842 1081.65-1078.40N, 127.80-123.90E) and Wall 157 (1078.37-1085.05N, 124.63-122.55E) became the third aim of the session. Finally, it is intended to backfill the area and to prepare it for visitors once the area has been fully studied.
Frankish Period (AD 1210-1450)
Mid 13th century:
The first signs of activity in Room 8 are associated to the erection and ultimate dismantlement of Wall 850, a limestone structure made with roughly squared blocks and bounded with a mixture of mud and lime (1085.80-1081.40N, 125.55-123.30E). This structure was found in the NW quadrant of Room 8, not aligned, joined or abutted with any of the other walls of the room. A white clay surface was excavated in association to this wall (870 1086.10-1084.20N, 126.60-124.10E) and dated by the pottery finds to the 3rd quarter of the 13th century. Its characteristics are reminiscent of those of the so-called Frankish floor located and excavated in several locations of the Frankish area.
Context 845, the deposit that covered the wall, was dated to the 4th quarter of the 13th century, but we know the wall was carefully dismantled before 1270±10 CE, since the digging of Cut 842 truncated the south section of the structure and this event was dated to this date. No traces of blocks belonging to the wall or any other sign of collapse have been found on the contexts laid up against both sides of the wall. These contexts make up a large number of small deposits, with almost no compaction of the soil, scarce artifacts and clustered dates. Most of them have been dated to the mid 13th century. Contexts 862 and 864, located in the lowest part of the stratigraphic sequence, have been dated to the 3rd quarter of the 13th century. Based on this evidence, I concluded that around the central years of the 13th century this wall is erected. Then, around 1270 CE the area is backfilled and leveled with small throws of soil which are accumulated on both sides and, finally, top of the wall (Context 845).
C. 1270 CE:
The next traces of anthropic action in the area are defined by two large pits that were excavated and filled in the last quarter of the 13th century. These pits were later truncated by the walls that defined Rooms 7, 8 and 9. Cut 823 (1083.80- 1080.80N, 127.50- 126.45E, depth: 0.62) is the west side in room 8 of a large pit also excavated in Room 9 (as Cut 733). The filling of the pit removed in Room 8 (Contexts 806, 813, 815, 819 and 821) was dated by pottery to the 4th quarter of the 13th century, and lotted with the material recovered in Room 9 (Contexts 717, 726 and 747) as Lot 2015-04. Context 806 included a cubic bone die with numbers drilled and painted in black (MF- 2015-57).
A second pit, Cut 842 (1081.65- 1078.40N, 127.80- 123.90E, depth: 0.47), was located on the South section of Room 8 and extending to the west into Room 7. This pit was also truncated by a structure (Wall 157), the part located in Room 7 already excavated by Burr and Rogers in 2014. The filling of this pit (Contexts 825 and 839) was dated by pottery to 1270±10 CE; Context 274 (1078.20-1081.30 N, 121.85-123.25 E, cf. Burr and Rogers, 2014) was dated to the late 3rd quarter of the 13th century. However, the earliest fill in the west portion of this pit, Context 887, was excavated this year and dated by pottery to the 4th quarter of the 13th century, providing further evidence for the equivalence of the pit cut to the east and west of Wall 157 (Cut 889=842; 1081.70- 1078.50N, 123.80- 121.40E).
4th quarter of the 13th century:
In the last decades of the 13th century the space north of the church gets delimited by a series of walls defining what is now labelled as Room 7, Room 8 and Room 9. Room 8 is separated from Room 7 by Wall 157 (124.63-122.55E/1078.37-1085.05N; Wall 13 NB 877), from the corridor north of the church by Wall 156 (128.40-123.58E/1077.50-1078.97N; Wall 16 NB 877 and 845) and from Room 9 by two structures, Wall 720 (1081.30-1079.20N, 128.55-127.50E) and Wall 830 (1084.10-1081.30N, 128.30-126.80E).
The data obtained from the excavation and subsequent study of the two pits described in the previous section, as well as the chronologies established for these structures by Rous and Worsham on one hand, and Burr and Rogers on the other, have been used to establish a sequence of construction for these structures. The first wall erected in the area in this period was Wall 156, since Walls 157 and 720 abut rather than join with it. This structure was dated by Rous and Worsham, (Context 184), to the 3rd quarter of the 13th century. Bennett, in 2015, dated the wall to the last years of the 3rd quarter of the 13th century, but based solely on a lens of soil left unexcavated in 2014 on the corridor between the church and Room 8 by Swalec and Wilson (see also NB 864). The pit shared by Room 7 and Room 8 was truncated completely on its southern edge by the erection of Wall 156, with the exception of a small portion of the edge preserved in the southeast corner of the room as Context 855, and dated by pottery to the 3rd quarter of the 13th century. If we also consider the date provided for the fill of the pit is 1270±10 CE (Context 825 and 839), we obtain a TPQ for the erection of the wall of very late 3rd or early 4th quarter of the 13th century.
The structure that divides Room 7 and Room 8, Wall 157, was built against 156. This wall 157 also divides Cut 842=889, the pit located between Room 8 and Room 9, in two different sections. The fill of the cut excavated in Room 7 was dated by Burr and Rogers to the late 3rd quarter of the 13th century (Context 274) and this session to the 4th quarter of the 13th century (Context 887); the fill of the pit in Room 8, as indicated above, to 1270±10 CE. A possible bench (cf. Context 175, Rous and Worsham 2014) laid along Wall 157 was likewise dated to the late 13th century.
Room 8 is separated from Room 9 on the east by two structures, Wall 720 and Wall 830. These two structures were first recorded together as Wall 155 in 2014 by Rous and Worsham, based on the records for Wall 18 (NB 877, 845). In May 2015 the record of Wall 155 was duplicated by Smotherman and Brannelly as Wall 720, still comprising in its extension the structures I define here as Wall 720 and 830. In June 2015 it was discovered that this single structure was in fact two different walls abutting. The study of previous documentation of the structure pointed out the duplicity of this record. Since a new nomenclature was needed, it was decided to keep the duplicated record of 720 for the south section of the wall and assign 830 to the north. Thus, Wall 155 is in fact Wall 830 plus Wall 720 (as recorded by Rodríguez-Álvarez), and the original record for Wall 720 (as recorded by Smotherman and Brannelly) equates Wall 155 but it is now obsolete.
Coming back to the analysis of the area, former Wall 155 was dated by Rous and Worsham based on the stratigraphic relationship with two possible foundation trenches (Context 176 and 177) to the late 13th or early 14th century. The excavations of this season showed that its south section (now Wall 720), also truncated Cut 842 on the east edge, preserving only part of it in Context 855 (as explained above in relation to the dating of Wall 156). Further work in June revealed that a second wall, 830, was constructed on the north end of Wall 720. Traces of what appears to be plaster were spotted on the surface where 830 abuts 720. Wall 720 truncated Cut 842=889 on its east edge, and can be dated to a time after 1270±10 and the construction of Wall 156 on which is abutted. Finally, Wall 830 divides the pit between Room 8 and Room 9 (Cut 823=733), and can be dated to at least the 4th quarter of the 13th century and after the erection of Wall 720.
The erection of these walls defined the space of Room 8. Former excavation in the area revealed a series of deposits that have been interpreted as the use surface of this room. The so-called "Frankish floor" (NB 877 and 845, see NB 845 p. 53), also excavated as Context 140 and the earlier surface excavated in Context 168 seem all to have been laid against the wall. These deposits were dated to the 14th century. These surfaces rest on two deposits, Context 786 and 796, which have been interpreted as garbage piles that were levelled in the area as sub-floor of for these use surfaces. The date provided by the pottery is the 4th quarter of the 13th century. Two bone cubic dice with numbers drilled and painted in black (MF-2015-24 and MF-2015-25) were recovered from this garbage context. A lead seal (MF-2015-56) was recovered from Context 844 (now equated to 786).

Conclusion:
The succession of structures, spaces and use in the area took place in a relatively short period of time. The presence of the garbage pits in the space later occupied by Room 7, 8 and 9 led different excavators in the past to interpret the space as an open air garbage dump. The discovery, however, of Wall 850 opens the possibility that this open area was a transitional phase between two different construction phases in the area: the first carefully dismantled before 1270±10 CE and the second one carried out in the last quarter of the 13th century.
With regard to the first phase of construction, Wall 850 and the possible clay floor associated to it (Context 870) are the only evidence, and more work is needed to assess the extension both in time and space of this structure. The presence of pits truncated by walls was already noted by previous excavators in the area (e.g. Rous and Worsham, Burr and Rogers, Smotherman and Brannelly) and led them to conclude that the area was an open space where garbage was deposited. The analysis of the materials recovered from the pit deposits (Lot 2015-04: 806, 819, 821; Lot 2015-38: 825; Lot 2015-37: 839) and the leveled garbage deposits (Lot 2015-22: 786, 796) clearly indicate that the disposal of refuse is taking place in the area. One pattern observed in this behavior is that alongside the pottery and animal bones, those deposits interpreted as garbage layers are characterized by a high presence of coins (17 in Context 786, 3 in 796, 5 in Context 806, 7 in Context 825). These coins, however, are never contemporary to the date provided by the pottery finds for the deposit, with all legible coins belonging to the Byzantine period, often to different issues of Manuel I (1143-1180). Coins contemporary to the Frankish dates of these contexts tend to appear in isolation and associated with leveling fills (e.g. Coin 2015-584, a CORINTVM issue of William Villehardouin found in Context 858, a leveling fill associated with Wall 850). Coinage of the Byzantine period seems to have lost any monetary value during the Frankish period, and it is in consequence discarded as a worthless object with the rest of the refuse produced in the area. These garbage deposits were also characterized by an abundant presence of iron remains, mostly square and round shaft nails, and small fragments of glass. Shards of prunted beakers, similar to MF-1992-38 and MF-2009-29, are especially abundant in these deposits. In conclusion, the study of Cut 823/733 and Cut 842/889 reinforces the interpretation of former excavators of the area being used as an open air garbage dump. But if the chronology of the area exposed in this summary is correct, that would mean that the church was already functioning while the space was working as a dump area. The discovery of Wall 850 not only indicates the presence of a former constructive phase but also clarifies the interpretation of the role of the pits and the leveling of the garbage in the area. The dismantlement of Wall 850 and the rapid succession of deposits in Room 7 and Room 8 might belong to a sequence of events in which the former structure is torn down to leave space for the rooms defined by Wall 156, 157 and 720. In the brief interim of construction garbage was disposed in this area north of the church, but just for a brief period of time, as the homogeneity of the fills of the pits (Contexts 825-839 and Contexts 806-819-821), seem to indicate.

Recommendations for Future Excavation:
- Continue with the excavation of Room 7 and Room 8, in order to reach the deposits corresponding to the 2nd quarter of the 13th century, in order to explore into more detail the early phase of construction in the area related to Wall 850.
- Analyse the sequence of construction of the walls shared by Rooms 7, 8 and 9, in order to assess whether the lack of joins among these walls is the product of an extended lapse in their construction or the result of the building techniques of the period. The relative chronology offered in this report is based on dates provided by artefacts and the visible relationships among the structures. Further excavation in the area and, especially, a detailed study of masonry styles and techniques in the Frankish period could corroborate or invalidate the present interpretation.