| Entry: | JHB continued working on cluster 16981, following it into Space 341; it ends in the doorway between Spaces 450 and 341, though – the cluster therefore is located north and northwest of buttress F.5057. In the doorway he uncovered a dense layer of phytoliths looking like straw or weed, probably a very dense bundle discarded here, as no weaving was apparent. The botany and phytolith specialists came out to have a look, and took several samples of phytoliths and ash from the cluster. They also encouraged us to take more samples, especially from ashy soil. We will do that, and it is good to know that the samples are actually being looked at, only days after we take them.
EUR continued room fill layer 16980, taking out more artefacts. The walls of Sp.341 are extremely difficult; the have multiple layers of plaster, are altogether not straight, buttress F.5057 lacks plaster on its western face altogether, and finds are going in deep into/under wall F.2413/5055. The room fill is therefore not complicated, but the walls are, so we try to go down to the floor, which was already visible in a little window, as soon as possible (another ca.10-15cm).
PTW spent the day removing the block of fill (16990) walls in the NW corner of B.107/Sp.462. We are now quite sure to have identified buttress F.3355 that sits on a badly visible older buttress F.3356 that really only shows because its material has no large finds in it, different from the fill around. The older buttress F.3352 has the shape of a very amorphous stump that F.3355 was built on top of with a therefore very irregular outline. The northern end of those two features is yet unclear, to we started removing soil from the supposed gap between the buttresses and the wall F.5074, but soon came upon soil that looks like it might be part of construction because of its homogeneity, because of lumps of the crumbly red mortar, and because of a large dark grey brick in horizontal position. We have to work on this further tomorrow.
KTX scraped off (16991) the dried upper 3-5cm of walls F.2427, F.5068, and the easternmost bits of F.3351 and F.3354. The goal was to follow the gap between the parallel walls F.2427 and F.5068 towards south, and find the outer edge between F.2427 and F.3351, being the outer SE corner of B.106. This obviously has to be defined before taking out the building. The undertaking was quite successful. Where we would expect the corner we are looking for, we only ever saw reddish-brown fill that was interpreted as late disturbance mainly because it has the fragments of orange burnt brick that often fill the late pits and that come from Byzantine grave linings. Today we had the idea, though, that it might not be a disturbance, but rather a package of Chalcolithic fill or a construction that walls F.2427 and F.3351 were built against to the never touched to form an outer corner. The inner corner was formed by a slightly weirdly shaped half-moon bit of the grey brick of B.106. This might all not be true, though, we will spray it tomorrow and have another look at it.
Towards the end of the day we had a long discussion about phases in the trench and strategy for the next days and weeks, started by Ian coming over for a nice long discussion in the trench, which we then continued among our team in the afternoon. While several member of our team are eager to get going on the wall removal, which is one of the major goals of this season, in the end we agreed that all three buildings, B.98, B.106, and B.107, are almost ready to come out but not quite yet. In all three of them, some issues still have to be solved, but this should be possible in 1-2 days and then we can get going.
1. Building 98: The walls faces towards the bottoms of the walls are still unclear in most places, finding them is tedious and often not possible because of rodent holes. Buttress F.5057, wall F.2413/5055 in Space 340 and all walls surrounding Space 341 seem to be sitting on fill. While I am about 70% sure that we will still find the walls when scraping back fill another few centimetres, my team mates, especially JHB and EUR who are working in the areas, think that these walls are sitting on fill or on older walls that were much thinner and that we would find when we go in further which we cannot really do because the upper walls would collapse. If this were true, this would still not say anything definite about the phasing of the room fill, and really altogether open more phasing problems. The floor 16977 definitely belongs to some features defined to form B.98, the clearest part is where the plaster floor runs up buttress F.3326. It is possible that not all construction features in the building were constructed at the same time, but that we now see the final state in a complicated building and rebuilding process. The people living in this final building might have been surrounded by walls whose lower parts were partially made up from cut old walls topped up with younger walls (F.2413/5055, F.5057, F.5056), as probably also happened in B.106 and B.107. In this case they would all have formed one building at some point, so do they not also belong to one phase? Do you give separate building numbers to each of the phases of a building, when these included cutting and topping up walls?
A completely different issue is the stratigraphic position of various fill units; artefacts lying on a floor that runs up one wall, but under another wall, and clusters that are partially inside the rooms and partially “under” walls – where do they belong to? It is possible that the floor was initially used, together with features such as F.3326 which are physically connected to it, and this initial building was abandoned and filled, and then walls such as F.2413/5055 were built on top of this fill to form one building together with the old features – that seems like a strange scenario to me, though, you would imagine they cleaned these many sherds and spiky other things off a perfectly usable plaster floor such as 16977 and used it again for their later building.
We will tomorrow try to remove the rest of the remaining room fill, including cluster 16981, and scrape the walls and check the fill-under-wall hypothesis, and then have an empty building from which we can remove walls and parts of walls and phases of walls all we want to checking phasing and construction techniques.
2. Building 107: While all walls and buttresses are mostly well defined, apart from the buttresses of the older building (F.3307, F.3308, F.3309, F.3355), their phasing is still unclear. The upper building, or building phase, B.107, is separated from the lower one by its building material, B.107 has perfectly visible bricks and lines of crumbly reddish-brown mortar. All of its walls and buttresses are physically connected (mind that only a stump of buttress F.3339 is left, and nothing of F.3337 and F.3338; they were removed by a mixture of late disturbances and excavation errors and only recognised after their removal) except for buttress F.3356 that will be discussed below. The wall bases of these walls wobble so violently, best visible in F.3344, that a floor or walking surface following these bases would have looked like a mini-gulf court. My interpretation is that some people decided to re-occupy a building that had suffered major erosion before, they cut the existing walls to where they were still intact and topped them up with newly built wall parts – or some similar scenario. Buttress F.3356 seems to verify this, as is base is heavily sloping and sitting on top of an amorphous stump of F.3355. The preserved top of F.3356 is about 1m under the base of F.3344, which it is supposed to belong to, and the back of F.3355 is not yet visible to were it is supposed to abut a wall which is probably further back west and might be found when removing wall F.3344 on top – we cannot cut back fill under 3344 to find this possible wall without having the wall collapse on us.
These people therefore lived in a building surrounded by walls with two different phases – is this now one building altogether, or two buildings?
The fill left around buttress F.3355/3356 and between the buttress and wall F.5074/3306 has to come off before we start taking out parts of the building. As phasing is so unclear, we will document the situation will all construction features present and then start to take them out based on our current interpretation of phasing. Should this prove wrong, the revised version can still be reconstructed from the 3D-model which contains all construction features.
3. Building 106: Apart from the southeast corner, the building is ready to go. We will brush and spray the area tomorrow, check again on the corner situation and draw what we see. Should the red soil be a disturbance, we will have to take that out first. B.106 has the same issues with uneven wall bases as B.107, so defining what a building is is also tricky here.
So we’ll have a busy day tomorrow and hopefully can start cutting walls of B.98 on Thursday. We also discussed different strategies of wall removal, layer by layer, making a section, taking out a quarter, and we’ll probably apply several of those for different parts of the buildings, depending on the various questions to be solved (do buttresses abut walls or bind into them? Do walls bind into each other? How do bricks look like??). Also, we discussed documentation and had the idea to use video and 3D-photography in addition to the usual photos and drawings.
As the reason to remove these buildings, apart from the fact that they have to go because we want to excavate deeper, is to find out more about their construction techniques and phases, the 3D-recording comes in especially handy. In a complicated building such as B.107 and the one below, we can take out one construction feature after the other, documenting each step of the destruction; when during destruction inscha’allah the phasing becomes clear, or maybe only later during our excavation and post-ex analysis, we can modify the model to show what we interpret to be phases. This will take more time than having several teams going at several walls at a time, but it will be rewarding, I hope. |
