IMTO Italian Mission to Oman University of Pisa PRELIMINARY REPORT (FEBRUARY-MARCH 2009)

IMTO Italian Mission to Oman University of Pisa PRELIMINARY REPORT (FEBRUARY-MARCH 2009)

SALUT PRELIMINARY REPORT (FEBRUARY-MARCH 2009) INTRODUCTION The archaeological excavation on the site of Salut was carried out by the team of the Italian Mission to Oman (IMTO), University of Pisa, from 07.02.2009 till 19.03.2009. The members of the team were: Prof. Alessandra Avanzini (Director of IMTO), Prof. Carl Phillips (archaeologist, field director), Costanza Odierna (archaeologist, trenchsupervisor), Michele Degli Esposti (archaeologist, trench-supervisor), Alessandro Massa and Raffaella Bonino (topographers and architects in charge for the archaeological park project), Federica Episcopo (student), Mr. Said al-salmi (representative of the Office of H.E. the Adviser of H.M. the Sultan for Cultural Affaires), Mr. Mohammad al-giahfali (representative of the Office of H.E. the Adviser of H.M. the Sultan for Cultural Affaires). THE EXCAVATIONS The 2009 excavation at Salut has proved quite exceptional. For the first time it is possible to speak with certainty about two distinct architectural phases during the Iron Age period. The earlier of these phases is represented by the burnt building in Area 1 which can be linked now with the basement excavated in earlier excavation seasons. The burnt building can also be related to parts of the outer wall, especially the tower that projects from the south-east side of the site and this enables further architectural reconstruction to be proposed. The second architectural phase is not so well defined but is characterised by the platform and terraces on the east side of the site. It is possible that some of these terraces incorporated parts of buildings also dating from the earlier phase, but this has yet to be confirmed. And in a few places in Area 1 it is possible that there is a third architectural phase dating from the Iron Age. Evidence for this is, however, very limited and comprises a few small sections of wall that are built on a different alignment to the walls of the previous two phases. Four new14c dates, taken from samples collected in 2008, further confirm that Salut was established at the very beginning of the Iron Age period (c. 1300 BC) and continued to be occupied until the latter part of the first millennium BC. Significant progress has been made also in defining the outer perimeter wall at Salut and almost the complete outline has now been planned. The problem of finding the

entrance to Salut is not yet fully solved but there is the strong possibility that an entrance was located on the south-east side of the site an entrance which would coincide with the main long axis of the buildings and terraces. There is also the clear doorway leading into the tower on the east side of the site. It was not practicable to investigate this during this season because of the restoration work taking place. It should, however, be possible to investigate this fully now that the nearby restoration has been completed. The restoration of parts of the eastern tower, to a height of approximately 5 metres, has proved the feasibility of further restoration of the walls that define the perimeter of the site and further work in Area 4 has provided some indication about the interior fill and construction of the outer wall. The 2009 excavations have also been exceptional in terms of the number and quality of the artefacts that have been recovered from primary Iron Age contexts. Other finds have been found in secondary contexts but are nevertheless of equal significance. A selection of the most important objects found in 2009 is presented in the section of this report following a description of the excavations carried out in Area 1, Area 4 and the clearing of the outer perimeter wall which includes the tower on the east side of the site. EXCAVATIONS IN AREA 1 (C. Odierna, trench-supervisor) I. THE BURNT BUILDING During this season the excavation continued in Area1, where important results for the understanding of the different building phases of Salut are evident. The trench, partially excavated during the last mission, to south towards wall M57 and west until the edge of the hill, was extended and enlarged, to cover an area of 6x4 m. Work started by removing the upper layer (US1) covering the area delimitated by the stone foundations of a modern structure which was above the upper part of the wall M138. When these foundations and their related floor (US307) were removed, an earlier deposit (US348), was uncovered, above which the foundations of the modern house were built, as shown in section 1. The pottery found in this deposit comes from different chronological periods. US 348 covered un another layer, US 356, characterised by the presence of burnt material mixed with crumbled mud-bricks. No Islamic pottery was found in this layer.

The removal of US 356 showed the presence of the upper part of three mud-brick walls perpendicular to M138, the profiles of which were just visible during the last mission in the section (fig. 1), and the mud-brick wall M130 that in a later period cut M117 (fig. 2). At this stage an area (the Burt Building) was evident, defined by the walls M117 at the north/north-east and M138 at north-west/south-west and divided into a series of rooms by the walls M57 and 127-129. These comprise: Room 1, between M57 and M129; Room 2, between M129 and M128; Room 3, between M128 and M127. In front of these rooms a corridor runs along the length of the wall M117. North-west of the Burnt Building, divided by a threshold, is an area designated Room4 (fig. 3). Except for Room 4, all of this area was covered by a layer with a characteristic very dark colour, a soft consistency and with the presence of a large amount of Iron Age pottery. The excavation of the single rooms has produced the following results: I.1 ROOM 1 As shown on the section 2 and in fig. 5, the burnt layer US394, is cut by a later pit US393 and covers a very thin, light brown layer (US395) that stands on top of floor US396. The latter was disturbed by a series of small pits filled by ash and a small amount of charcoal (US397). It is important to note that we found a similar deposit to US395 in the layers covering the floor in the corridor and in Room 4. I.2 ROOM 2 In this room the typical burnt layer (US384) was cut, near to M129, by a later pit (US385) in which were found a few sherds of Iron Age III diagnostic pottery (section 3; fig.6). The presence of the pits (US 393, 385 and 390 in the corridor) along with their related pottery material is important because it indicates the long period of use of the building s structures. US385 and US384 covered the earth packed floor US386. I.3 ROOM 3 The removal of the burnt layer US380 gave evidence that the wall M138 was built in a later period in respect to the foundation of the Burnt Building, and that it is not the

continuation of the outsider wall. It was, in fact, built on top of US380 as the fig.8 shows. Probably contemporaneous to the foundation of M138 is the building of the small mud-bricks wall M139 that, running between M128 and M129, closes Room 3. I.4 THE CORRIDOR A long, narrow space (corridor), runs alongside M117 and in front of Rooms 1-3 (fig.9). At northern end a threshold made by two superimposed rows of mud-bricks marks the area inside the Burnt Building and the outside (US376 is the filling that covered it. Fig.10). The stratigraphy of the corridor doesn t show big differences with that one of the Room1-3. Here the burnt layer, US377, covers a very thin layer, US378, which in colour and consistency can be compared with US395, in room 1, and US367 in Room 4. Near the south end of the corridor, in front of M129 and against M117, US377 is cut by a later pit (US390), which contained some sherds of Iron Age III pottery (section 3; fig. 11). From against M129, the removal of US378, which covered the floor, revealed the presence of a badly preserved burnt wooden upright, standing on a flat stone, and can be interpreted as a door timber (Fig. 12). I.5 OUTSIDE THE BURNT BUILDING, ROOM 4 Outside the threshold we have an area closed to the north-east by M117, at the northwest by M49 and at the west by M138. Under the upper layer, US356, was found a layer, reddish in colour and with a medium compact consistency (US366). The removal of this layer revealed the foundation of wall M138, built on top of US366, and the presence, almost in the middle of the area, of an alignment of a mud-bricks wall (M136) and the presence of a stone wall (M137), built on US366 and against M117 and M142. M137 covered a deposit (US389) characterised by an accumulation of big stones which sit on top of US367 and a deposit (US392) interpreted as the collapse of M136, and lying between M136 and M117, and covering US 368 Floor. It consisted of a big block of mud-bricks and was mixed with a very dark loam containing charcoals and ash (fig. 13). Under US366 a very thin deposit (US367) which covers the floor US368 was found. It is similar to US395 and US378 in colour and consistency.

After the removal of US392, the connection between the US368 Floor and the floor US100 (found during the 2006 excavation) was established. I.6 THE WALL M117 The extension of the excavations in Area 1 helped to clarify the nature of some structures excavated during the last season and which remained partially understood. At the north end of M117 a square (1,10x1.6 m) filled with organic materials (US337) was excavated and proved to be similar to one found during the 2006 season. The top was plugged with stones (US327. Fig. 14). The work done during this season clarified that M117 is part of the Burnt Building and that during a later period some other walls were added parallel to it on the north side. It was also cut at the south end by wall M130. These later walls, reduced the square space described and also covered earlier walls perpendicular to M117, that only became visible after the removal of the later walls (fig. 15). II. NEW TRENCH EAST OF M57 A new trench was opened at east of M57 to understand the connection between the Burnt Building and the area excavated in the 2006 mission, which had revealed a very well preserved pavement (US391). Under the superficial layer US1, different deposits were uncovered: US383 is composed of collapsed stones and a large amount of pottery of different periods. It covered US387, a very compact layer composed of mud mixed with a few pieces of pottery. This floor relates to the foundation of M130. It was cut on the west side of the trench by a pit (US388) lying just above the pavement US391, from where came different sherds of Iron Age III pottery (such as the tulip bowl and the lentoid jar). US387 probably covers four mud-bricks walls, the significance of which remain to be understood. They seem define an almost squared area where the presence of a floor was made visible by the imprint of woven basket (fig. 17). The function of this area remains to be investigated. III. THE SMALL TOWER Outside M138 the western and steepest slope of the hill was covered by mud-bricks, probably part of the construction of the outside wall. In correspondence with a badly preserved projecting stone wall, built against the outside wall, a long and narrow room measuring 4x1.40 m was opened (Fig. 18).

Different levels were recorded in this area: Initially it was probably part of the Burnt Building. Walls M129 and M128, before being cut by M138, projected towards the outside wall, perpendicular to the mudbrick wall M139 which is parallel the outside stone wall- and is probably contemporary with it. In the next phase walls M129 and M128 were substituted by walls M133 and M134, made from mud-bricks different from the earlier phase. Related to these walls, the latest floor was covered by a medium compact deposit, mixed with few sherds of pottery (US372). In the third and final phase, a higher floor (US371), was defined and enclosed by two roughly made stones walls, M131 and M132. This floor which is associated with Iron Age III pottery is not preserved on the south-west side of the room due to slope erosion. IV. NEW SOUNDING IN BUILDING 1 (C. Phillips, C. Odierna, trench-supervisors) Also in Area 1 a further sounding was excavated to see the level of the floor going with M10. The first deposits, US 399 were approximately 40 cm deep. Beneath this is a 10cm thick layer of burnt ashy deposits (US 400) which lie directly on a floor surface related to M10. M10 and the floor are clearly contemporary with the burnt building or first phase occupation of the site. US 399 is a deliberate fill and probably in preparation for the second phase of building. US 400 contained a good selection of Iron Age pottery and organic materials, namely charcoal and bones. EXCAVATIONS IN AREA 4 (C. Phillips, F. Episcopo trench-supervisors) In Area 4 one of the main objectives was to try and understand the connection with the tower on the east side of the site. In 2008 it had been observed that in the area directly above the tower there was a distinct concentration of stone debris between the Iron Age mud-brick wall (M102) and the steep slope leading down to the tower. A trench approximately 4 metres wide was excavated with the aim of removing the stone fill. This showed that the stone fill, which shows steep tip-lines, is two to three metres deep and is accumulated against the eroded face of M102. The fact that the outer face of M102 is distinctly eroded is of some significance because the stone fill (US 373) contained a large amount of Islamic pottery predominantly of 12 th -13 th century date. The stone fill is likely, therefore to be collapsed material from the Islamic period that accumulated against the already exposed face of M102. This further suggests that some major collapse and/or erosion of Iron Age walls took place in the period between c. 300 BC and the 12 th century or thereabouts. It is hardly

surprising that such erosion should take place over a period of c. 1500 years of abandonment (fig. 19-20). The stone fill with Islamic pottery appears to lie adjacent to the stone wall M123. On the opposite side of this wall, however, there is also a massive deposit of intentional fill that occupies the space between the outer face of the Iron Age mud-brick wall and the outer face of the main outer wall. In this instance the fill comprises small chips of greenstone. These deposits also slope steeply down and away from the face of the mud-brick walls. In contrast to US373 a relatively small amount of pottery was recovered from these deposits but all of Iron Age type. It seems therefore that the greenstone fill represents the original deposits used to fill the space between the Iron Age mud-brick walls that form the innermost part of the main outside wall. This begs the question whether wall M123 is of Iron Age date or if it is a result of remedial work done during the early Islamic re-occupation of the site. Further work is necessary to confirm this. Further work is also necessary to fully remove the stone fill (US373) and to see how the eroded mud-brick surface of wall M102 relates to the upper part of the tower. No trace of steps was yet found in this area. That said, there are still a lot of the later deposits that need be removed and it is also clear that in the period following the Iron Age abandonment erosion has taken place on a massive scale in this part of the site. Further work in Area 4 included the completion of earlier work, namely the excavation of US 298 and US369. US 298 is the fill from a L-shaped room, partly excavated in 2008A. The deposits are rich in bones and pottery and carinated bowls were especially abundant. US 369 is a deliberate deposit comprising mud-brick lumps and debris used to fill the area between M118 and M119. These two walls appear to have formed a terrace and the fill (U369) contained an abundance of Iron Age pottery, including a number of almost complete vessels. EXCAVATIONS INSIDE THE MAIN TOWER (M. Degli Esposti, trench-supervisor) One of the main goals of this campaign was the rebuilding of the frontal and northwestern part of the wall of the tower located on the east side of Salut. To allow the undertaking of these works some further cleaning and removal of the uppermost deposits inside the tower was necessary. In particular a clearer view of the internal face of MB39 (north-western portion of the tower s wall) was needed. In all the excavated area only what can be defined as US1 has been removed, reaching different heights depending on the initial situation. In Areas 5 and 6 the layout of

some rooms is already visible. In Area 7, however, a substantial heap of debris still remains, despite approximately 4 m 3 of it having been removed. In Area 5, which had been already partially cleared of debris during 2005B campaign, the walls M40, M44 and M43 have been further unearthed, though no floor has been reached. A new wall, NW-SE oriented and slightly curved eastward has been uncovered (M124), linking M43 to another new wall, SE-NW oriented (M125), which stands as the boundary between Area 5 and Area 6. In the latter a short NW-SE oriented wall, standing much higher than the others, has been brought to light in its upper courses (M126). It links M125 with the southeastern part of the tower s wall. Between walls M125 and M43 some large stones are scattered, probably linked with the first collapse of the walls, and will have to be given a new US when excavated. Two short walls, possibly reinforcement for the walls against which they are leaning, have been unearthed beside M126 and near the corner defined by M40 and the main wall M39. These walls have been left without number until further excavation in the area. While conducting these operations, all the internal line of the tower s main wall was revealed, and what seems a narrow passage that could eventually lead to the upper part of the site has been identified. This needs further investigation. Furthermore, it seems that the north-western part of the tower wall, near its junction with the upper main wall, was actually constituted by two adjacent parallel walls (fig. 21). FURTHER INVESTIGATION ON THE MAIN OUTER WALL (M. Degli Esposti, trenchsupervisor) Aiming to reveal the entire plan of the main walls surrounding the site, a few days were spent unearthing their south-eastern portion, namely the part included in SAS 7. As a result of this work, only a short part of the walls still remains buried, where a passage is needed for temporarily dumping soil from the excavation. Furthermore, the structure that was believed to be a projecting tower/ bastion, projecting southward from the top of the site, turned out to be almost aligned with the south-eastern part of the walls, that actually show a straight and deep corner in correspondence with the newly excavated southernmost room of the burnt building. The space between the big buttress wall first revealed during SL08B campaign and the above mentioned bastion was also cleaned, showing a complex layout comprising mud-bricks walls and the probable remains of some stone walls, actually in such a poor state of preservation that very little if anything can be said on their plan. In order to understand the construction technique of the external walls, some cleaning and digging has been carried out on the portion going from the walls M131/M133 to

the radial wall M56. This means that the part of the walls lying under the stone slabs floor unearthed between the radial walls M56 and M57 was investigated too. What is evident is that the radial walls M56 and M57 were built simultaneously, linking the internal and external faces of the site s main wall, and thus constituting real radial walls. The space between them, as well as it has been revealed for the smaller sector defined by the walls M131/M133 and wall M57, was filled by means of a series of mud-brick features, the voids between which were subsequently filled in with loam. These deposits contain a significant quantity of pottery sherds. The filling has been given just two numbers: US374 - the filling between M57 and M131/133, and US375 - the filling between M56 and M57. These mud-brick features were both little walls and small platform-like features, similar to those filling some of the compartments forming the platforms inside the site. In many cases, the mud-brick walls were made up of alternating layers of compact and complete bricks with layers of mud or irregular shaped mud-brick lumps. Many of these walls were built one above the other without a regular alignment between them. The most important data revealed by these works is the evidence that the outer face of the main wall had to be at least as high as to reach the level of the slab floor set between M56 and M57, thus giving an important indication for the possible restoration of the external walls.

SIGNIFICANT FINDS FROM THE 2009 EXCAVATIONS I. These included the first Iron Age seal to be found at Salut. Several of these are known from other sites in South East Arabia and represent the only indication that we have so far for any kind of administrative system that might have existed: The seal is made of dark gray soft-stone. It has a pyramidal shape and a slightly rounded apex, horizontally perforated (0,04 cm). The seal is 1 cm tall, its irregular squared face measures 1,5 by 1,6 cm. The impression shows a simple design formed by eight incised and pierced triangles and an incised dot and circle (star/sun motif?). This seal compare well with the ones found at the Iron Age site of Rumeilah and Qarn bint Saud (Abu Dhabi Emirate- U.A.E.), both stylistically and by virtue of their shape. Other two similar seals, one from a grave in Qarn bint Sa ud and one from Hili 2, are not published. SL09A; US373,1 S305 II. Also found this year was Salut s first cuboid incense burner. Despite its being broken and incomplete it is still possible to reconstruct the size and form of the original object and also to envisage the decoration of two of its four sides as well as how the top and bottom must have looked. The object was originally ten centimetres tall and the sides will have measured approximately seven centimetres. The incense burner is made of clay and the fabric is similar to that of some of the pottery found at Salut. The decoration is made by deep incisions and impressions which together produce a sharp, excised pattern. On one side the decoration comprises three raised lines that follow the outline of the object. A single raised line defines the arch formed by the two legs, and at the apex there are two short vertical lines. The

space within this frame is filled with horizontal and vertical zig-zag lines. On the other side the decoration comprises a single raised line around the outline of the object and also defining the arch formed by the two legs. Different from the other side, however, above the arch there is a half-rectangle pattern formed by five raised lines. The space between the half-rectangle and the outer frame is again filled with horizontal and vertical zig-zag lines. The space between the inner side of the halfrectangle and the apex of the arch is also filled with vertical zig-zag lines. Usually it is necessary to be cautious when inferring the function of an object based only on its shape. In this case, however, the definition of the object as an incense burner is assured by examples from South Arabia which have the names of various types of incense inscribed on them. The exact date when the cuboid form of incense burner was first adopted in South Arabia is difficult to determine despite there being numerous examples because few are from excavated contexts. The earliest securely dated examples of this shape of incense burner are in fact from southern Mesopotamia. Examples are known from the site of Ur and can be dated to the Eighth / Seventh century BC. Early examples are known also from the Levant where they have been found at Lachish and Tell Ajjul. Several examples of this type of object have been found in East and South East Arabia. In Bahrain comparable incense burners have been found at the Qalat al- Bahrain in contexts dated between 600 and 400 BC. Three examples have been excavated at Rafaq in the Wadi al-qawr (southern Ras al-khaimah, UAE) and one example was found at Rumeilah (UAE) in the area near the al-buraimi / al-ain oases. In both the latter cases the finds probably date some time between 600 and 300 BC. The cuboid incense burner found at Salut is the first of its kind to be found on an Iron Age site in the Sultanate of Oman. It has already been confirmed that the occupation at Salut began in the latter part of the second millennium BC and continued until the mid-first millennium BC. At Rafaq and Rumeilah the cubod incense burners were found in association with pottery dating to the Iron Age III period. This period is well attested at Salut and it is probable that the incense burner dates from this period, some time between 600 and 300 BC. Further excavations at Salut will hopefully help to define this period more precisely. By way of a more general point the discovery of the cuboid incense burner further adds to the growing evidence for the use of incense throughout the Arabian peninsular in the first millennium BC.

SL09A US1,176 Cl135 III. Several complete or almost complete pottery vessels were found and include types that would typically be designated to the Iron Age II period and others that would be considered of Ion Age III date. SL09A US367,1 Carinated bowl- Iron Age II

SL09A US388,2, Tulip bowl- Iron Age III SL09A US388,1,lentoid jar, Iron Age III IV. The range and variety of bronze objects found at Salut has also been significantly enhanced. The smallest of the bronze snakes yet found was recovered from inside the burnt building. Although very small it is nevertheless very important to find such an object in a clearly early Iron Age context.

SL09A; US394,2 MB193 V. The discovery of a bronze dagger, a probable spearhead and several arrowheads has increased the range of Iron Age weaponry known from the site and the region as a whole. Also important for dating purposes is an early bronze axe-head that was also found in part of the burnt building. VI. Also found in the burnt building was a complete soft-stone vessel of a type distributed widely throughout Oman and the Gulf in the early first millennium BC.

SL09A; US390,1 S312

Fig.1 The Burnt Building at the end of the 2008B excavation and the profiles of the walls M127-128. Fig.2 M110 and M117.

Fig.3 The Burnt Building after excavation. Fig.4 The Burnt Building, Room 1 after excavation.

Fig.5 The Burnt Building, Room 1, particular of the section. Fig.6 The Burnt Building, Room 2, after excavation.

Fig.7 The Burnt Building, Room 3, after excavation. Fig.8 The Burnt Building, Room 3, particular of the section

Fig.9 The Burnt Building, the corridor Fig.10 The Burnt Building, the threshold Fig.11 The Burnt Building, the corridor, particular of the section

Fig.12 The Burnt Building, the corridor, the door timber against M129 Fig.13 Room 4

Fig.14 M117 after the 2008B excavation. Fig.15 M117 after the 2009A excavation.

Fig.16 New trench east of M57, general view Fig.17 New trench east of M57, imprint of woven basket.

Fig.18 The small tower.

Fig.19 Area 4, general view. Fig.20 Area 4, general view.

Fig.21 SAS2, Excavation inside the main tower. Fig.22, Salut, view from the surrounding plain

APPENDIX Methodological note After all past campaigns, the overall excavated area of the site of Salut has become remarkably wide. That has on the one hand caused the joining between some of the originally defined excavation areas, on the other has forced us to envisage the rethinking of the original subdivision of the site in terms of SAS and areas, constituting these latter further subdivisions inside each SAS. Thus, we decided to keep the existing SAS 1 and 2 as they were defined, that meaning that SAS 1 includes the whole upper part of the site inside the main wall (the platforms system and the more ancient basement area), while SAS 2 corresponds to the protruding so-called tower. New SAS will be defined as follows: SAS 3: Corresponding to trench 3 dug during 2008B campaign, it goes south-eastward from in front of the tower, right in its middle point, encompasses its south-eastern side and reaches the first big terrace wall on the southern side of the hill, that s to say M103. SAS 4: Corresponding to trench 4 dug during 2008B campaign, starts in front of the tower and goes north-westward, climbing up the hill until the second buttress M114 and the second terrace wall M116, which defines its limit. SAS 5: Runs along the south-eastern side of the main walls, going from the terrace wall M103 to the second main rocky ridge visible on this side, nearly perpendicular to main walls, which it meets almost in the middle of this side, namely in a point currently not excavated. SAS 6: Covers all the north-western side of the hill outside the main walls, thus going from M116 up to the top of the hill and then encompassing the site until it reaches the long SW-NE buttress first retrieved during 2008B campaign. SAS 7: Covers the area between SAS 5 and 6, going from the long SW-NE wall to the main southeastern rocky ridge. Inside each SAS smaller excavation Areas will been redefined. Currently these Areas have been identified only inside SAS 1 and 2. SAS 1 Area 1: western half of the site, on the highest top of the hill. It is included between the walls M26 at north and the outside wall at east, west and south. Area 4: north-eastern half oh the site, on the lowest side of the hill. It is included between the walls M26 at north-west and the outside main tower wall at north east..

SAS 2 Area 5: northern half of the forepart of the tower, it is the area in which excavation already took place in 2005B campaign. Separated from Area 6 by wall M125. Area 6: southern half of the forepart of the tower, first removal of the uppermost deposits begun during this campaign. Area 7: includes the part of the tower close to SAS1, south-east of the postern. Here only the removal of part of the massive collapse heap has been undertaken so far.