.bk A12 .fl K705jl.j .fd journal entries .ei jl .ed K705 .rd K703 .ri jl -dy Since the raising of the tents prevented us from digging in k23, we started a pickrun across k27 instead. There is some confusion for me about the feature. I am not sure whether we are in f316, the ashy layer, or 317, the brown later (or if f317 even exists). ed practiced digging; she has a very sensitive hand, and noticed some important things. f 317 nd ed noticed that the surface was not of uniform hardness all across its length. Also, I noticed a number of sherds along the S edge which were not lying flat, but upright, similar to the gullywash in f314. This signifies that the ^a was not regular, certainly not a floor. On the other hand, there are many lying flat, as well. I think that there are too many upright sherds for them to be only anomalies. f 318 ds gullywash with red bricky chunks. Perhaps this is the same as f217 and 218, found above it in k18 and 19. There are a lot of small stones, especially above the a31 wall. Perhaps these are the remains of the fill within the wall? I don't think so, though, since it would have been very strange for the gully to have washed away the giant building stones and to have left behind the much smaller fill. f 319 ds brickfall. This appears to be the same brickfall from above, f219, 215, 205, and 189. It covers the S excavation area in k23. Within it, in the SE quadrant, there was a pocket of ashy soil that may have come from a tannur or other fire source. However, right next to that were several upright or oblique sherds, suggesting still random accumulation. Although this is clearly a new feature, I haven't quite figured out these sherds. f 320 ds the ^a in k26. This ^a is clearly separate from that in k27. While excavating a few pickruns in k27, we discovered that it pressed against the very smooth face of k26. We still don't know the cause of the fissure. However, because the face of k26 is so smooth, we are presuming for the moment that it is older, and will therefore dig in k27. f 316 dy There is a clear vertical fissure around k27 on all sides. Towards k26, the ^a is pressed up against a very smooth face, and flakes off quite easily, suggesting that the ^a in k27 is later than that in k26. On the sides facing the stone walls, there is a fissure of soft, loose dirt down which a trowel slides with ease. I am reminded of a pit in A7 from three seasons ago, which had separated by approximately 2cm from the dirt on all sides, making us think for a time that it was a column. Could k27 be a pit, later than k26 and the apsidal structure? The contents of the ^a, large quantities of broken sherds, bones, stones, and also ash, also support the theory that it might be a pit. On the other hand, the very smooth face of k26 is unusual for a pit, according to gb. We will know more as the excavation continues. f 317 nd In the final analysis, f317 does not exist as a separate feature, and was simply f316 while still moist and recently uncovered. =l f316 f 316 nd Due to the non-existence of f317, f316 has now been extended to include all of the accumulation in k27 from its starting height to the bottom of the feature, which has not yet been reached. .rd K704 .ri gb .bk Z1 -lb kw and sg: see A12-lb for alternative abbreviations referring to kiln waste and slag, used often ambiguously. .bk A12 v 99 ph a detail view of a large animal bone lying flat, to document nature of deposit. f 316 cp it has been observed in recent days that the ^a is very dirty, in the sense that it shows many bones and sherds, often not lying flat; also pebbles, stones, and kiln waste, tannur chunks. v99 documents one aspect of this type of accumulation. v 99 Behind the bone, note a sherd lying vertically and a stone. f 316 H3 in spite of the nature and alignment of the components, I regard this feature as accumulation rather than fill, because there is an overall horizontal banding of matrix and even components. In other words, this is not a single fill, much less a massive discard, as we notice often in h4 and h5 hollows at the edge of the post-palace settlement. a 22 H1 Several possible processes are contemplated (see note H2 for description of characteristics that have led us to these possibilities). (1) The fissure reflects a lower structure, which caused settling. It is not clear then why the section does not allow a clear connection of layers on either side of the fissure. (2) A rodent may have opened a way through pre-existing deposition, but this would be possible only for part of what we observe, in particular the soft brown fill at elevation matching q397. (3) There may have been a wooden structure (or wood and matting) that disintegrated (e.g., a curtain wall). But there is no impression along the face of the fissure, nor is there any obvious remanent (though this may have fallen at the bottom). H2 The fissure 321 has two clear faces and very soft dirt, the type found as a secondary fill (e.g., from rodent activity). One face shows scalloping in at least one place. The excavated portion of f303 in section shows a clear upper portion of the same element that corresponds to f321, but it slopes inward and it leans a smooth face on the convex face. Also, tho color appears to be different. Note that the fissure continues all around a33, along the face of the stones. We noted this from the beginning, but it is becoming more and more marked. f 321 =l f303 nl we are equating f321 and f303 because the latter seems to be the upward continuation of the former, with 321 being vertical and 303 curving inward. But note that the interstice in faces between f303 matches (as least in size) the interface between faces in f321. a 33 H3 The marked fissure all around this aggregate suggest that the accumulation inside it shrank towards the center, possibly as a result of subsidence. .ri jl -dy Today we continued excavating in k27. The fissure continued down, and at points even reached 10cm in width. Also, gb spent most of the morning in the square with us, discussing encoding practices and the continuing excavating in k27. We also looked at the recent uncovered brickfall in k23, f319. gb decided that a picture documenting the new observation was needed, so we scraped and cleaned in preparation for it. The rest of the day, we performed most of a second pickrun across k27, with the help of a few extra workmen from A10 (who did not excavate today). Work ended an hour early, since gb decided to speak to the workmen about the project today. Tomorrow, A13 will not be digging, and we will have as many workmen from there as we need. Therefore, once the picture of k23 is taken, we will excavate at both ends of the square, hopefully to get k23 completed so that we can establish this season the S extent of the structure and return to focusing on a22. -sg Due to complications with the tent in the past few days, we have been unable to complete excavation in k23. Now that those problems have been solved, and given that we will have A13's workforce tomorrow, I hope to photograph and excavate the last meter of ^a tomorrow. This would serve two purposes: first, it would make the most efficient use of the available manpower, and second, it would let us (hopefully) define the S edge of the apsidal structure. Even if we do not finish digging within a22 this season, I would consider this one accomplishment almost as important, especially given the limited goals of this season. f 319 nd This ^bf came as a complete surprise to us, especially since we had not observed it in the baulk and had thought that the whole ^a was gullywash. Partly, this is because it was also missed last season: our scraping today suggests that it only extends 10 or 20cm to the north of the current excavation before dropping sharply off, so it is easy to see how it could have been missed, hidden at the edge of the square. Thankfully, lr noticed it this year in between pickruns, and we were able to document it. We will photograph it in context before continuing to excavate in that section. .rd K705 .ri ed v 100 This photo shows the brickfall f319 which is now clearly seen in the S section and from the top. The stones f322 appear to be contained in this feature. v 100a Brickfall f319 slopes in the W baulk and is covered by gullywash. These sloping layers appear to follow the original slope of the gully. v 100b closeup of the brickfall in section; also shows floating stones in this layer. f323 the ashy pocket is also visible. v 101 Shows the relationship between the floating stones a34 and a32 the apsidal structure. a 34 H2 Floating stones appear to be in a line parallel to the S part of the apsidal structure. Stones are the same type used for the walls of the apsidal structure. We see that stones are probably missing. .ri jl f 323 ds ashy pit in f318 and 319. At first, we thought that this was just an ashy pocket, such as from a tannur, but as we dug more, it became clear that it was actually a pit. f 318 ds Now, looking more closely at the section, I can see that what we thought were bricky chunks mixed in the gullywash was really the top layer of the brickfall. =l f319 f 324 ds This bright red layer began to appear yesterday, and as we followed it today, it came out much more. It seems to lie underneath f316, at least in the E side of the locus. f 325 ds pit? A white layer, in almost a complete half circle, pressed up against the face of the fissure. It looks like a very shallow pit, since a significant while lip/rim has been preserved, at least in parts. The white material appears to be a phytolith layer, and samples were taken for our bioarchaeologist. H99 The discovery of this pit raises some important questions about the deposition of a22. We were operating under the assumption that k27 had subsided and was therefore later than k26, but if so, where is the other half of the pit? It is unlikely that they would have dug through so clear a layer of white material last season, so it might be the case that k26 is the more recent accumulation. But if k26 cut away the rest of the pit and leaned against k27, what exactly did it lean against, i.e. what created the fissure? Once again, we come back to the same puzzling question, except now even more confused than before. -sg for the moment, our strategy is to leave k27 alone and ask the rest of the team what they think about our conundrum. Soon, we will have to decide whether to continue in k27 or switch to k26. f 320 ds For some reason, I remembered this accumulation as being mostly free from pottery, but it is in fact quite dirty in terms of pottery and bones, much like the ^a in k27. sg I will have to ask mkb to analyze this pottery as well, to find the relative dates of the two features. -dy Today did not begin as planned. Not only had A13 decided to work and hadn't told me, but our best pickman, Abdul Rackman, was missing. Adnan Omar was promoted from shovel to replace him for the day, since he had impressed me last season as a pickman for jlw and myself in A9. Also, A13's change of plans proved fortuitous, since we soon struck the white pit f325 and had to quit digging in k27, and the extra workforce would have been useless. Before we found the pit, we also traced the red layer f324 in the E half of the locus. After a strategy session with gb, we photographed and covered the pit and switched instead to removing f319 brickfall and leveling f320 in k26.