Chronicle of work in unit J1

February 2009 - G. Buccellati
Background
The initial sounding (2003)
The main excavations (2005-6)
Clarifications (2007)
Looking for the early interface (2008)
The Late Chalcolithic structure (2009)
Plans for the future
Synopsis of elements per season
This version

Back to top

Background

     In the very first season of excavations at Tell Mozan (1984), we opened unit B1, some 60 meters to the north of J1. B1 was the first of five units that yielded what we eventually labelled Temple BA. It was clear that the Temple stood above a high rise with a sharper slope to the south, the direction from which one would have approached to the Temple, given the location of its entrance at its southwestern corner.
     It was therefore our intention to explore further this area, and a new opportunity presented itself through the work planned as a result of our collaboration with Peter Pfälzner and Heike Dohmann-Pfälzner. Their goal was to excavate the area to the south of the Temple, in zone C, and as a first step it was agreed that they would open a long trench connecting the southern end of the Temple (B1) with their projected excavation (C2). This trench (labeled B6) would give us a stratigraphic link between the old and the new excavations.
     The B6 excavations (1999) were particularly successful because (1) they revealed the existence of a monumental staircase flanking the base of the Terrace. Subsequently, (2) the Pfälzners proposed to undertake a geo-physical survey that extended beyond the limits of the area (C2) where they had undertaken excavations. This survey (2000) showed that the staircase was flanked by a stone wall (of which only small portions had been exposed in the excavations), which suggested the possibility that the Terrace was oval in shape. It also supported (3) our initial hypothesis that zone J was a large open area.
Back to top

The initial sounding (2003)

     It was then as a result of our initial intent to pursue the exploration of Temple BA and the rise on which it stood, on the one hand, and, on the other, of the findings by the Pfälzners in B6, that we decided in 2003 to resume excavations in the area adjacent to the Temple. An additional reason had emerged in the intervening years. The excavations in the area of the Palace (AA) showed that the Palace extended east in the direction of the Temple Terrace and of the open area, or plaza, between the two. We assumed that the open area was at the same elevation as the Palace court(8500), and that the eastern perimeter of the Palace might be reached through a deep sounding at a location which we could only guess.
     Accordingly, unit J1 was conceived as a long trench which would join transversally the area of the Palace with the area of the Temple. The orientation of the long strip was askew vis-à-vis the standard physical grid, and it was so located as to give us the best advantage in our intent to put Palace and Terrace in relationship with each other.
     In the western portion of the long trench we reached a stratum with extensive brickfall: the quality of the bricks was similar to that of the Palace, and the elevation was that of the Palace court (8500). This suggested that we were just outside the eastern perimeter of the Palace, possibly just to the outside. Exposure was so limited that we could not plausibly test this hypothesis. But it seemed reasonable to assume that we had here the western end of the Plaza.
     The eastern end of the trench straddled, as intended, the revetment wall of the Temple Terrace. The trench was very narrow at this point, and in 2003 we were not able to reach the very bottom of the wall, though we were not far from it.
     Since we concentrated on reaching as low an elevation as possible in both the western and the eastern end, the central part of the long trench was never even started.

Back to top

The main excavations (2005-6)

     In 2005, after a year's intermission, we resumed excavations in J1. Two changes affected the overall organization of the unit. First, given the considerable difference in depositional history between the western and the eastern end of the long trench, it was decided that the western end would be renamed as A19 (hence the pertinent data are not now included in the current J1 digital book). Second, since we abandoned the idea of completing the trench for its full east-west length, we decided to change the alignment of the local physical grid, so as to conform to the orientation of the Temple Terrace.
     The main results obtained in 2005 and 2006 were as follows.
     (1) For the reasons indicated, we had anticipated that the bottom of the revetment wall would be at the same elevation as the Plaza (8500). If so, the wall would have stood to a height of 5 meters, and as a result we had stopped, in the 2003 sounding, at about 2.5 meters from the top of the wall. In 2005 we discovered that the wall was instead only 3 meters in height, and that it was fronted by an escarpment.
     (2) As a consequence, the next goal was to follow the escarpment to the presumed elevation of 8500. Even by 2006, we were not able to reach this elevation. One reason was the complexity of the stratigraphy: already in 2003 we had reached, next to the wall, early third millennium strata, but at some distance to the south we still had Mittani materials. We felt that the situation required great care - and a larger exposure.
     (3) The need for larger exposure resulted not only from (a) the desire to gain a better control on the stratigraphy, but also from (b) the interest we had developed in obtaining a more adequate view of the revetment wall from the same floor level where the ancients stood, and (c) the concern for safety. With regard to the latter, we decided to implement a suggestion made by Gionata Rizzi, a specialist in architectural conservation whom we had approached specifically with regard to the problems posed by our wall: he recommended that we created a series of stepped and sloped sections that would replace the stright vertical sections that, at projected heights of 7 and more meters, were both dangerous and difficult to maintain.
     (4) In the process, we gained greater insight into a number of structural features that characterize the Plaza, in particular the escarpment itself and a number of what we called "curtain walls."
Back to top

Clarifications (2007)

     Work in 2007 was limited to only a few days (July 24 to August 12), because the season was planned primarily as a study season. Work in J1 was part of a training stage for some of the junior members on the staff, and this seemed like a good opportunity to (a) begin to clarify the problem of the close juxtaposition of third millennium and Mittani material, (b) seek to identify what we expected to be the next (Khabur) stratum, and (c) possibly reach the 8500 elevation which was already fairly close at hand.
     As it happened, we were unable to reach any of these goals, but the season was important in that it gave us important additional material to set the stage for a better informed approach to just those problems.
Back to top

Looking for the early interface (2008)

     We knew, after the 2007 season, that we had to undertake a determined effort in order to excavate with the greatest care within the limited exposure frame, already established in previous seasons, of two 5 x 5 squares (essentially the two loci k126 and k128). The results were significant.
     (1) It became clear that the Mittani accumulations rested directly on accumulations of the early Ninevite V period. Surprising at first, this juxtaposition ("clash of centuries," as I called it in the panels that illustrate the site, see page 4) can be explained with certain assumptions relating to the developmental history of the site.
     (2) It was already clear from the earlier exposure that two successive escarpments abutted the revetment wall, but the excavation of the lower (early Ninevite V) accumulations cause us to revise substantially their original function and respective dates.
     (3) In the 2005 season we had opened a small sounding in order to expose the base of the revetment wall, and this had cut through both escarpments. Taking advantage of this pre-existing cut, we excavated, in 2008, a small sounding (k128) to explore the earlier levels, and we came across a stone structure which we presume to be a Late Chalcolithic antecedent of the third millennium revetment wall.
     (4) While examining closer the face of the revetment wall, we first noticed a triangular motif which is significant as an indication of the fact the wall remained unaltered since first erected (in early Ninevite V) until the middle of the Mittani period.
Back to top

The Late Chalcolithic structure (2009 - by Lorenzo Crescioli)

     The excavation's aims this year were to check the questions arisen at the very end of the last season, i.e., the nature of a possible Late Chalcolithic wall. The results are significant:
     (1) The full exposure of the wall (^wall2) yielded several important facts.
(a) This is indeed a wall, running in an easterly direction.
(b) It is bonded with a wall (^wall3) running in a southerly direction, which stands even higher that the first wall.
(c) This second wall is flanked to the NW by deposits that are early in date and much higher than the ones flanking it to the east.
(d) We have exposed the base of the first wall (^wall2).
(e) At this juncture, we have pure Late Chalcolithic accumulations f355 (even if for now with very few sherds, due to the small size of the sounding).
     These facts allow the following conclusions.
(a)The nature and location of the EW wall (^wall2) matches so well the later revetment wall (^wall1) that we may safely assume it served the same purpose.
(b) The date may be safely assumed to be LC3, even though some possible doubts still remain.
(c) The association with the NS wall (^wall3), plus the difference in the deposits that flank the latter, imply that this was the northwestern end of the Plaza in Late Chalcolithic times.
(d) The presence of LC3 materials coming from a much higher elevation a few meters to the northeast (J3) lead us to assume the existence of a Late Chalcolithic mound and a Late Chalcolithic plaza, involving a extraordinary continuity of meaning and use for this sacral area.
     (2) Another important result was the clear identification of both escarpments as dating back to the ED III period. So also the construction of the entire terrace complex date to the ED III, a great expansion period and spatial reorganization of the city of Urkesh.
     (3) Moreover, after this season, it seems very likely that the reconstruction of the escarpment after a so short period of time is due, not to the rise of the plaza that in fact was still clean in the ED III period, but to the massive water damage that eroded deeply the surface of the Early escarpment. The erosion, that probably did not affect the western portion where the base of the ED III wall was protected by a stone escarpment J5, forced them to find a solution, i.e. an almost complete rebuilding of a new escarpment, much higher than the previous one, to protect the base of the revetment wall.
     (4) Another question was regarding the absence of Ninevite materials in the northern half of J1, which are instead very considerable in the southern half. We could confirm last year's conclusions that this is linked with the ED III cuts that removed the earlier Ninevite V accumulations, which most probably abutted the Late Chalcolithic wall. The cut was due to the rebuilding of the second escarpment. This would mean that the Late Chalcolithic walls were in use for a long period of time.
Back to top

Plans for the future

     Unit J1 is so configured that it lends itself ideally to remain open indefinitely for continued excavations. At the moment, the main objectives we can identify are as follows
     (1) We plan to further widen the exposure to the south in order to gain a full view of the revetment wall and of the Temple atop the Terrace.
     (2) In so doing, we will be able to follow the early Ninevite V accumulations to a point (in the south) where we may find evidence of later periods.
     (3) The to west, we have already begun to establish a link with unit J5: while further work in this respect will especially be the task of unit J5, it will have to be coordinated closely with work in J1, in order to establish clearly the configuration of the Plaza floors as they rise (presumably) to the northwest along the face of the revetment wall.
Back to top

Synopsis of elements per season

     Following is a chart indicating the elements that were defined in the various seasons of excavation.
     In the first season of excavations in J1 (MZ16, 2003), the area opened came to include two very different sectors. The one to the west was eventually subsumed under unit A19. The numbers that are missing from the J1 list given below are the ones that were incorporated in A19.

season loci features items q-lots views
2003 k7-k11/k101,k102,k103 f1-f17,f19,f20, f23-f25,f28,f32,f38,f41-f43,f47,f61,f68, f71-f73,f75,f76,f81,f89,f98,f114,f124 i1,i2,i10,i12 q1-q35,q38,q39,q43,q45,q46,q50,q54,q55,q64,q68,q72, q74-76,q81,q82, q84-q86,q90,q96, q103-q105,q110,q116,q120, q123-q126,q129,q131,q138-q140,q146-q149, q157-q159,q162,q165,q166,q170,q171,q174,q180,q181,q189,q190,q192,q198,q204,q209,q212,q214,q215,q218,q219 v1-v7,v12,v13,v15,v19,v20,v22,v23, v25-v27,v30, v33-v35,v39,v40
2005 k12-k34/k106-k111 f140-f196 i15-i29 q290-q550 v49-v119
2006 k112-k119 f197-f242 i30-i41 q551-q1029 v120-v226
2007 - f243-f250 i42-i48 q1030-q1055 v227-v241
2008 k120-k128 f251-f300 i49-i64 q1056-q1209 v242-v294
2009 k129-k132 f301-f376 i65-i80 q1211-q1336 v334-v447
Back to top

This version

     The version published here is up to date as of December 7, 2009. Following updates will indicate additions.
     At regular intervals, given versions will be archived and their data publihsed in the Ephemeris portion of the Urkesh website.
Back to top