Journal / Summary:D-4-2007-Summary

Final Trench Summary, Year 2007 Season

Area D

Trench 4

July 5, 2007

The first week of actual digging (after removing the 2m of L4097) revealed relatively little. In L4099, we found a thin ephemeral surface of organic material. L4102 was a rock and pottery concentration resting on what we later realized was a thin mud packed surface (L4103) that was virtually impossible to see while digging down. L4101 was a pit containing a large concentration of sherds as well as a pot holder stand made of sherds in packed mud.

We finally started hitting some good context at the end of the first week when we came across the top of what later became the collapse layer L4111.

There was a little confusion here at the beginning before we knew what we had, so let me take a minute to explain the loci. I originally thought L4108 was the wall. Working under the assumption that L4108 was a wall, I defined L4016 and L4107 on either side. The "wall" L4108 however ended up being just the top of the collapse layer L4111. Thus L4106, 4107, and 4108 are actually the same fill above collapse locus. I later redefined the fill above the collapse locus as L4109. The wall from which the collapse occurred is locus 4110.

The collapse layer itself (L4111) was characterized by burnt mud brick of red, yellow and green varieties. There was a thick layer of large, whole mud bricks at the top of the collapse, followed by a higher concentration of thin burned mud roofing material. There were also architectural components containing the curvature of the roofing beams – the diameter of the beams tended to be between 6 and 9cm from the ones that I measured.

We later switched loci from L4111 to L4112 because we noticed a change in the characteristics of the locus. This new locus contained primarily burnt red brick material and the roofing material with scattered white chaffy bits showing either a plaster between bricks or inclusions within the bricks. We also figured that we were getting close to the floor now that we were under the roofing material.

We then divided L4112 into L4112 and L4113 when we noticed a white layer that we thought was a surface that became L4113. However, when we got below it, we realized that it was not actually a change but still part of the same collapse - possibly a plaster facing from a fallen wall? We kept the distinction however, as the remaining L4112 was showing some different characteristics than that of L4113. Primarily, the remaining section of L4112 is a more ashy fill like content than the red brick material showing up in the area later separated as L4113. This showed that the collapse was higher and more concentrated by the wall 4110 than on the west side of the trench.

Under L4113 and L4112, we found the floor towards the end of the week (L4115). The floor had several small finds on it including 3 andirons, various shaped pottery, and several other hardened mudbrick items theoretically for use in an oven? One of the most interesting finds was a carved mound of hardened clay (L4113 – KT#47) in the shape of a cone with a rounded top delineated with a groove. Similar material was used in a solid dome mound that was found in the SW corner of 4113. I think these were used in some fashion to shape or mold material around: possibly baskets or hides. We also found several unfired shallow clay vessels that would fall apart upon excavation. A couple samples of these are L4141-KT#4 and L4112-KT#28.

One of the most important things determined from the excavation of the collapse layer was the order of collapse. Most of the large small finds towards the floor L4115 were not lying flat on the floor showing that they fell into the place that we excavated them. Theoretically then, they could have been sitting on a shelf on the wall. The way the collapse must have occurred during the fire, due to the way the architecture came up, was that the roof would have caved in first, knocking the objects to the ground below it. Hence the objects, followed by a layer of roofing material. After the roof caved in, the walls then collapsed on top creating the highly concentrated layer of solid burnt mud bricks at the top of the collapse layer we excavated.

At this point, we knocked the East baulk into D8 in order to go down and find the connection between the house in my trench and the rest of the walls in D8, D5, and D6. We initially divided the baulk extension into two loci based on 4122 containing a slight bit more of the collapse characteristics that 4121.

We also straightened the North and West baulks which gave us half a meter more of trench space. Under the West baulk, we were able to trace wall 4118 all the way to the corner of the trench. We also found a surface just west of this wall with a pot submerged below the surface in the northernmost point of L4127. We found part of a wall that we had missed jut to the west of this submerged pot (possibly an extension of L4118) and theoretically this wall was separating out a room that continues into the West baulk with multiple rows of these submerged storage pots.

Under the North baulk, we were able to define the wall 4119 better and really determine that the area between that wall and 4116 was a storage place. We were under this impression anyways, when we found the organic material slump L4117 in between these two parallel running walls (L4116 and L4119). We also found the oven 4125 up against the wall 4119.

I believe the East Baulk is the real key to the puzzle however. The room associated with surface 4134 was extremely burnt on all four walls. Also, the surface itself (4134) was so burnt, that we only recognized it as a change in material / locus and not immediately as a surface itself. Thus we ended up switching locus when we hit it, but we never took a HAP sample. The surface was extremely burnt orange, to the point of being soft and crumbly to the touch. This room was clearly a storage room of some sort, having a crawl space / window access from a whole preserved in wall 4110. This crawl space was 54cm of the ground, 70cm wide, and 75cm from the NE corner of the room associated with the 4115 surface. It is important to note that the ledge of the crawl space showed wear in the shape of a curve from repeated use both coming and going from the storage room.

This storage room (associated with the 4134 surface) contained what we originally misdefined as a pit (4123). When we first hit this "pit" it had a rounded shape and the normal characteristics of a pit (soft, silty, ashy soil). However, as we went down, the pit ended precisely at the surface 4134 but continued out in a bell shape as we dug down – uncharacteristic of a pit. We now believe that this "pit" was a pile of wood or some type of material being stored in this room.

I also believe that the fire initially started here in this storage room. It is interesting to note, that the four walls of the storage room all had signs of burn, while on the other side of wall L4110 (directly on the other side of this room) the wall had no ash. There was significant burn on the rest of that side of wall 4110 however -- just not directly on the other side of the storage room. One theory is that the fire started in the storage room causing that part of the wall 4110 to collapse into the larger room. The higher section of that part of the wall then blocked the lower half of the wall from the flames and kept it from getting burnt. And it was when that section of the wall from the storage room collapsed, that the fire the spread to the larger room and caused the roof to collapse, etc. This would also explain why, even at the lowest part of the collapse, we were still getting some mixed content of mud brick and architectural components from the roof.

The wall L4137 is the wall that connects D4, D5, D6, and D8.

The last part of the trench that requires some explanation is the North end of the East Baulk extension. We originally defined a wall (L4131) running North to South down the center of the locus (4121) and dug it as such. However, the L4131 locus that we thought was the wall turned out to be contained collapse running through what appeared to be two parallel walls (L4139 and L4140). We determined that L4131 was actually a small bin that caught the collapse that fell into this area. L4140 is a normal large wall, while L4139 was a thin wall on the west side of the bin.