t: +20 12 5113357 (mobile)
t: +20 27955666 (office)
The lower end of the Stone Village at an early stage in the excavation season, viewed to the south

Stone Village Project 2006-2007


Anna Stevens and Wendy Dolling

Contents

Acknowledgements
2006–7, progress and results
Objects and environmental remains
Charcoal from the 2005–6 and 2006–7 seasons
Discussion
Publications cited


Acknowledgements

A second season of survey and excavation at the Stone Village was undertaken from 20th November to 30th December, 2006, with a supplementary season of survey conducted from 19th April to 3rd May, 2007. The results of both are reported here. The fieldwork was carried out with the assistance of SCA inspectors Mr Helmi Hussein and Mr Gamal Abu Bakr, and workmen Ahmed Mokhtar and Mahmoud Sayed Nasser el-Din. The work was funded by the Egypt Exploration Society’s Centenary Fund, the G. A. Wainwright Fund, Oxford University and the Mulvey Fund, Cambridge University. The support lent the project by the EES (including the loan of a total station), Professor Barry Kemp, and the SCA is acknowledged with much gratitude.

2006–7, progress and results


Survey
In the combined 2006–7 season, approximately four weeks were devoted to survey and four to excavation, with the main site again the focus of work. The surface survey was extended eastwards from the south-west corner of the site. As in 2005–6, planning points were positioned with a Leica TCR307 total station, but all planning done by hand at a scale of 1:50. The survey was brought to a stage where it is approximately a quarter completed.

Excavation
Two areas were chosen for excavation: Trench 1 was extended northwards, and a new trench (Trench 2) opened along the southern margin of the site. All excavation was done by the authors, with 100% of each fill unit sieved through a 4 mm sieve. At the close of excavation, plastic sheeting was laid over the trench floors, and each filled with excavation spoil.

Aerial photograph of the main site taken by Gwil Owen in April 2007. Trenches 1 and 2 are shown, both largely backfilled but with the tops of the walls in Trench 2 still visible. Site north is to the top of the image.
Aerial photograph of the main site taken by Gwil Owen in April 2007. Trenches 1 and 2 are shown, both largely backfilled but with the tops of the walls in Trench 2 still visible. Site north is to the top of the image.

The northern extension to Trench 1 prior to excavation in 2006–7. Facing south.
The northern extension to Trench 1 prior to excavation in 2006–7. Facing south.

Trench 1, northern extension
In 2005–6, Trench 1 had been positioned in part to define the northern limit of the main site. At the close of excavation, a thick surface, 11238, covered the northern margin of the trench, continuing beyond the north baulk. In 2006–7, therefore, the trench was extended with the aim of identifying where, and how, this surface ended. A 3 x 4 metre extension to the trench was laid out and planned, with excavation itself contained to the eastern half of this area.

The trench surface prior to excavation sloped downwards to the north and slightly to the east, the surface deposit, (11456), comprising loose sand with minor ceramic and mid-sized limestone boulders. This blurred slightly into a layer of sand with increased ceramic and boulders, and minor chaff, (11458). There were also a few small nodules of the orange-brown ‘marl’ used in antiquity to make mortar – often irregularly shaped ‘plugs’ placed between limestone boulders – and brick. It is virtually impossible to distinguish, visually, highly degraded fragments of marl-based mortar/brick from nodules of marl dug fresh from the desert floor, but when encountered archaeologically across the main site the marl fragments are probably the former. (11458) covered most of the trench, although in the south-west corner there was a small mounded deposit of silty sand with minor chaff and ash, (11457), and in the south-east corner a concentration of large loose limestone boulders and marl-mortar clumps, (11459). The removal of (11458) across the northern and central margins of the trench revealed a slightly more compact horizon of almost sterile sand. This was left untouched for several days, during which wind-blown debris and baulk collapse accumulated over it. This intrusive material, along with the now intermixed surface of the sand, was in turn removed as unit (11493), to reveal sterile sandy gebel (11499) across the entire northern and central margins of the trench.

The stratigraphy across the southern margin was more complex; here, piles of stone and mortar architectural collapse were interspersed with layers of ancient rubbish and wind-blown sand. Removal of (11458) further defined the mound of silty sand (11457) and the deposit of boulders and marl, (11459), and elsewhere exposed a slightly friable deposit of coarse sand (11460) containing fragments of marl mortar, sherds, and a few lenses of silt/ash. The stone-by-stone removal of (11459) indicated that it was a pile of architectural collapse. It overlay (11460), which now formed a distinct mound along most of the southern margin. (11457) was removed to reveal a friable deposit of silty and chaffy sand, (11462), also restricted to the south-west corner. Excavation through (11460) then exposed a lower horizon of stones and mortar, (11463), which rested upon a compacted mound of fine sand and silt with moderate marl nodules, (11464). This was further exposed under (11462), so that it extended across the entire width of the trench. (11463) was contiguous with a concentration of boulders and mortar, (11254), that had been exposed in the north-east corner of the original trench in 2005–6. It was not clear, at the end of that season, whether these were in situ, and they were left unexcavated. A 0.75 x 1.5 metre span of these stones were now included in the excavation area. The careful removal of (11463)/(11254), confirmed that these were not in situ. The underlying fill, (11464), comprised silty sand with minor marl nodules, straw, ashy lenses and a few small fragments of gypsum. This, and the overlying fill (11460), were probably original rubbish dumps: discarded material associated with the occupation of the site. Both extended beyond the eastern edge of the excavation area.

View of the northern extension to Trench 1 (partially covered by fill in the background) after excavation. Facing south.
View of the northern extension to Trench 1 (partially covered by fill in the background) after excavation. Facing south.

Beneath (11464) in the south-west corner of the trench extension was a small area of loosely compacted ashy sand, rich in chaff, (11466). This in turn covered a relatively thin layer of compacted fine sand with minor charcoal and chaff (11467), running east-west across the trench. This passed beneath the northern edge of the thick laminated surface 11238, which spanned the northern margin of the original trench; the sand overlay a surface, 11494, left unexcavated but with a matrix largely of silty sand with much chaff and moderate charcoal. It sloped down to the north, over the sandy gebel (11499), and was in effect a lower lamina of 11238. Given the small span of 11494 exposed, it was impossible to confirm that it had not been truncated by post-occupation looting. But its sloping profile is generally consistent that of floor 11238 as a whole, and we can be relatively confident that the edge of 11494 marks the original northern edge of the site here. The general lack of cultural material in deposits excavated north of this edge provides some support for this. There is no indication that the surface was contained to its north by a wall, so it probably belonged to the site exterior, or at least a space that was not completely walled and roofed. Here, therefore, the transition between built and unbuilt space appears to have been marked initially by a plastered or trampled exterior surface, over which rubbish and wind-blown sand accumulated during the course of occupation.

Final top plan of the northern extension to Trench 1
Final top plan of the northern extension to Trench 1.

Trench 2
A slightly mounded area along the southern margin of the site was selected as the second excavation area. The surface stones here lay in roughly linear arrangements and it was speculated that they marked the location of a multi-roomed structure of roughly rectangular shape running north-south beneath the sand. A 4 x 5 metre trench was opened over the southern end of this area. The general aims in working here were to better assess the preservation of the subsurface structures, and of how the surface remains related to these. The work proved fruitful on both points.

Trench 2 prior to excavation, facing west.
Trench 2 prior to excavation, facing west.

Work began with the removal of the surface sand, (11455), along with those stones lying loose within it, across the entire trench. As the sand was scraped away, it revealed further limestone boulders, with several distinct areas of soft orange-brown deposit. The latter were removed individually, as units (11468) – (11471); each comprised sand with a high concentration of small nodules of orange marl. When these were removed, the entire trench was covered by limestone boulders. No meaningful lines or clusters of stones could be distinguished, so they were removed as a single unit, (11472), divided into quadrants.

Trench 2 after removal of the surface deposits, showing the dense horizon of boulders, (11472). Facing west.
Trench 2 after removal of the surface deposits, showing the dense horizon of boulders, (11472). Facing west.

The stones were picked out individually. Most lay in a loose matrix of sand and small marl nodules, with some larger marl-mortar plugs. In the north-east quadrant lay a few large fragments of marl-based brick, and one intact brick. As excavation progressed, lines of walls began to emerge: [11490] running north-south along the eastern baulk; [11481]/[11473] passing east-west through the north margin; [11489] running through the south baulk; and [11482]/[11488] roughly bisecting the trench north-south. All were formed from unworked limestone boulders bonded by marl-based mortar, and faced with marl-based plaster. They were preserved up to 60 centimetres high – considerably higher than the walls in Trench 1. The loose stones continued to a depth of 50 – 80 cm, to approximately 20 cm below the surviving wall tops; below this horizon, the density of stone decreased noticeably.

Trench 2 after removal of the horizon of loose stones (11472), with the lines of walls emerging. Facing east.
Trench 2 after removal of the horizon of loose stones (11472), with the lines of walls emerging. Facing east.

The western half of the trench, west of wall [11482]/[11488], was now covered with a fill of ashy sand, (11477). In the eastern half, a line of stone scatter (11780) ran approximately east-west between walls [11488] and [11490], and subdivided the sandy fill here into (11478) to the north and (11483) to the south. Work continued in the western space with (11477), a fill of fine to medium sand with ash, largely as a diffuse element but in patches denser and charcoal rich. It contained plant matter, largely chaff, nodules of marl-based mortar and flattened clumps of chaff-rich sand, later identified as pieces of floor build-up. As (11477) was removed, a further line of wall, [11491]/[11501], became visible running through the western baulk and projecting beyond the northern baulk. (11477) continued to a depth of around 20 – 40 cm, where the sandy gebel surface (11509) began to emerge. Patches of plaster flooring, 11507 and 11510, were also revealed, but it was soon evident that most of the floor no longer survived. It was clear that the damage to them was deliberate: it was too extensive and the breaks too abrupt to have been caused by natural wear. The best preserved patch of floor contained a distinct ashy lamina, suggesting that activity in this room generated ash; hence, the ashy nature of (11477), which was largely churned-up floor and over-floor deposits. The lower parts of walls [11481] and [11484] were also slightly blackened, and slightly hardened. Ash was perhaps once banked up in the north-east corner of the room, and the source of the heat may also have been nearby. The latter is not represented directly. Just west of the entranceway between the western and eastern rooms, the edge of the plaster floor was curved and the adjacent gebel surface slightly depressed, so there was perhaps a jar or similar (a pottery hearth?) set into the floor here, although it appears an inconvenient location.

In the eastern room, (11478) was removed to reveal gebel (11509) with patches of truncated flooring plaster, 11504 and 11508. (11478) comprised largely sand with frequent small to mid-sized marl clumps. The fill was relatively rich in chaffy plant material, and contained low to medium density ceramic. A notable feature of the unit when compared to (11477) was its lack of ash. With the exception of living debris incorporated into the floor layers, it should be noted, Trench 2 yielded no undisturbed deposits from the primary occupation phase of the main site.

Stone scatter (11780) was removed to reveal a line of in situ limestone boulders [11482] running roughly east-west between walls [11488] and [11490]. A few fragments of gebel-based mortar ‘plugs’ were found around the stones, but none adhered to them, and there was no trace of facing plaster. The construction was preserved to three courses of stone, but much of the overlying stone of (11780) had presumably slumped from it. Originally, it was probably just one stone wide. Its original height is impossible to determine, although it seems unlikely to have been a floor-to-ceiling wall, given the rough nature of its construction and apparent absence of any means of access into the small space it defined in the south-east corner. Here, excavation through (11483), revealed another patchy plaster floor 11502, which seems originally to have abutted [11482]. The floor surface was partially overlain by a slightly compacted accumulation of mortar and sand, (11503), adjacent to the south face of [11482], and left unexcavated. (11483) itself was a loose fill comparable to (11478).

Roughly constructed stone wall [11482], running east-west through the eastern room. Facing south.
Roughly constructed stone wall [11482], running east-west through the eastern room. Facing south.

Excavations in Trench 2 therefore exposed the back two, intercommunicating, rooms of a narrow building that is projecting northwards. The easternmost room is subdivided by a stone construction, [11482], of uncertain function. The full width of this part of the building has been exposed.

Trench 2 after excavation, showing two intercommunicating rooms of a building running north-south. Note that walls run through the east, south and west baulks. Facing east.
Trench 2 after excavation, showing two intercommunicating rooms of a building running north-south. Note that walls run through the east, south and west baulks. Facing east.

Final top plan of Trench 2.
Final top plan of Trench 2.

The greater preservation of the walls in Trench 2 than in Trench 1 shows that the level of damage is not uniform across the site. There is real potential at the Stone Village for the recovery of quite substantial in situ structures, and the exploration of architectural space. It was also reassuring that the layout of the structure as speculated from observation of the surface features was generally consistent with the in situ remains. Less predictable, but perhaps unsurprising, was the discovery that many walls (namely, [11491]/[11501], [11498] and [11490]) all lay not beneath linear stone ridges visible at surface level, but in the slightly depressed spans of clean sand running alongside these ridges. These walls, therefore, have largely toppled sideways rather than slump along their foundations. There is little reason to doubt that the situation in Trench 2 is not broadly typical of the site; a similar pattern was encountered in Trench 1. It implies that, in at least some cases, the lines of surface stone do reflect the alignments of walls, but displaced laterally. This pattern now needs to be taken into account when reading the surface remains.

Trench 2 after excavation, showing the relationship between surface features and in situ structures. Facing west. The wall running through the southern baulk of the trench extends beyond the western baulk, and is presumably continuing beneath the span of clean surface sand here.
Trench 2 after excavation, showing the relationship between surface features and in situ structures. Facing west. The wall running through the southern baulk of the trench extends beyond the western baulk, and is presumably continuing beneath the span of clean surface sand here.

The excavation in Trench 2 has demonstrated that the use of stone in walls was not restricted to the immediate foundation courses. It is possible, in fact, that the walls were built considerably higher in stone than their preserved height, given the large quantity of stone loose in the fill. The upper parts of the walls were presumably built of brick. All of the brick fragments yet encountered at the site, which have all been floating loose in fill, appear to be entirely marl-based. There are certainly no examples of the grey-brown silt-rich bricks often used in houses in the Main City. Interestingly, though, the facing plaster preserved on the walls in Trench 2 was often made from a slightly browner marl than the brighter orange marl used for the bonding plaster, which perhaps reflects an attempt to mimic silt plaster. One small fragment of marl-based plaster with very faint traces of probable red and blue pigment from the north-west corner of unit (11472) also suggests the presence of painted wall decoration. It is not yet clear how extensively elements such as wood or cut stone were also used as construction materials. As yet, no cut-stone architectural elements have been encountered. A patch of mortar with a possible vertical impression on the northern end of the east face of wall [11491] might mark the location of a wooden door frame. Yet the door jamb with the best-preserved facing plaster, the northernmost of the entrance between the eastern and western rooms, had no sign of such an impression. Several fragments of marl-based mortar from fill (11477) bore smooth right-angled impressions that may have been formed by cut wooden planks. Trench 2 also yielded the first evidence for roofing, in the form of fragments of marl plaster with impressions of bundles of grass, namely from units (11477), (11478) and (11483). The quantity of such fragments was not high, but probably enough to suggest that both the western and eastern rooms were once roofed. This raises the possibility that the roofs were substantial enough to support activity on them, and perhaps a second-storey proper.

This brings the issue of the function of the building. Although very little of it has yet been uncovered, the general layout of the exposed rooms does recall the back two rooms of small houses of the kind best known from the Workmen’s Village, but also found elsewhere at Amarna. The full excavation of this building, which will allow us to test this idea, is scheduled as a priority for the forthcoming season.

Trench 2 under excavation, facing north-west.
Trench 2 under excavation, facing north-west.

Objects and environmental remains


Excavated objects from the 2006–7 season included faience finger rings and pendants; glass and stone beads; small pieces of worked wood; textile fragments; a few pieces of worked chert; and a very small quantity of copper/bronze fragments. As yet, the site has not proven particularly rich in finds (a situation that may itself be significant in terms of its original function), although the material that has been found is generally consistent with that from housing areas elsewhere at Amarna. Whilst the sample size of objects from the two trenches is too low to yet allow extensive comparison across different areas of the site, both the range and quantity of items from Trench 1 seems slightly greater than from Trench 2. Part of the explanation for this may lie in the likelihood that some of the fill from the former was undisturbed ancient rubbish.

Pottery was sorted on site, with all sherds counted but only diagnostics retained; these await processing. Several sherds bear incised potmarks, whilst a particularly interesting find from Trench 2 was a group of fragments from a Canaanite amphora (object 37776), bearing both painted decoration and a hieratic jar label mentioning the port of ‘Peru-nefer’ (reading courtesy of Dr Marc Gabolde, who is preparing the label for publication).

Both trenches yielded a moderate amount of fragmentary animal bone. Some was slightly salt-effected and brittle, and the majority was unburnt. Preliminary examination suggests that it was largely mammalian, and there were indications of butchery on examples from both trenches. All plant materials, including charcoal, recovered during on-site sieving have been retained and samples of the matrix from each fill collected. In March 2007, the charcoal from both the 2005–6 and 2006–7 seasons was processed by Rainer Gerisch, whose report on this work follows.

Charcoal from the 2005–6 and 2006–7 seasons


Rainer Gerisch
This season, all wood charcoal from the Stone Village excavations in 2005–6 and 2006–7, recovered manually and by dry sieving, was analysed to determine the shrub and tree species which are present in the charcoal assemblages.

The material (wood and bark charcoal) comprises 23 samples, 6,018 pieces, 4,578.8 ml for the year 2005, and 38 samples, 4,551 pieces, 3,758.5 ml for 2006. 35 charcoal samples came from Trench 1 (2005 and 06), 26 samples from Trench 2 (2006).

The largest amounts of charcoal were found in Trench 1, feature [11239] with 1,105 pieces, 782.1 ml, feature [11230] with 641 pieces, 551.3 ml and feature [11228] with 625 pieces, 504.8 ml and in Trench 2, feature [11477] with 878 pieces, 814.4 ml.

The wood anatomical examination of 10,569 charcoal pieces has allowed the identification of 14 woody taxa. In comparison with other areas the obtained results showed no higher taxonomic diversity. More wadi plants occur due to the remote location of the settlement, on the eastern face of the low plateau in the desert to the east of the Main City. The other part of the identified taxa came from the floodplain and gardens. The studied material from the Workmen’s Village, on the western face of the low plateau, revealed 22 taxa.

Trench 1: Acacia nilotica-type: 7,085 pieces (35 samples, 5,331.2 ml), Acacia sp.: 112 pieces (26 samples, 144.8 ml), Chenopodiaceae, concentric type: 3 pieces (3 samples, 0.5 ml), Chenopodiaceae, foraminate type: 2 pieces (2 samples, 0.6 ml), Faidherbia albida: 16 pieces (10 samples, 8.7 ml), Ficus sycomorus: 27 pieces (15 samples, 10.5 ml), Leptadenia pyrotechnica: 2 pieces (2 samples, 0.6 ml), Mimusops sp.: 11 pieces (5 samples, 5.5 ml), Salix sp.: 1 pc (1 sample, 0.6 ml), Tamarix sp.: 402 pieces (33 samples, 291.4 ml), Zilla spinosa: 20 pieces (12 samples, 8.6 ml), Zygophyllum sp.: 1 piece (1 sample, 0.3 ml).

Trench 2: Acacia nilotica-type: 2,417 pieces (26 samples, 2,262.9 ml), Acacia sp.: 63 pieces (17 samples, 48.5 ml), Balanites aegyptiaca: 2 pieces (2 samples, 2.1 ml), Chenopodiaceae, concentric type: 2 pieces (2 samples, 1.5 ml), Faidherbia albida: 5 pieces (5 samples, 3.4 ml), Ficus sycomorus: 3 pieces (3 samples, 0.9 ml), Mimusops sp.: 4 pieces (4 samples, 2.2 ml), Palmae: 1 piece (1 sample, 0.2 ml), Tamarix sp.: 67 pieces (21 samples, 67.5 ml), Zilla spinosa: 8 pieces (5 samples, 4.6 ml).

Like in all samples from Amarna the assemblages are highly dominated by the wood of the Nile acacia, which represents an excellent fuel burning steadily and slow with high calorific value. The other main source for fuel in ancient Egypt was the tamarisk which is of less quality in comparison with acacia wood burning more quickly and with smoke.

No remnants of timber imports were identified from the charcoal material of the Stone Village excavated so far. But two little pieces of dessicated cedar wood were found among the charcoal from features [11228] and [11230] of Trench 1.

Charred fragments of dom palm nuts often appeared in the samples (both trenches) together with some charred and uncharred date stones (Trench 1). One charred pod segment of the Nile acacia was found in Trench 1, feature [11242].

Discussion

We are now in a position to make some very broad observations about the site. The excavations have demonstrated that the Stone Village had facilities such as food-preparation areas and relatively substantial roofed buildings. It had the means to support a permanent population; although how much of a village ‘character’ it had remains to be seen. One important element still missing is a water supply, but we might suppose that water was brought in from the city proper, as was done at the Workmen’s Village.

We can begin to piece together a general site history. At present, it appears simply that there was a clean piece of desert upon which, sometime during the Amarna period, a small occupation area was built. Walls [11489] and [11490] are built over a layer of ashy sand, (11511), but this can probably be dated to the construction of the walls themselves rather than an earlier phase of occupation. A patch of trampled gebel surface, 11505, immediately beneath the floor laminae in the eastern room, is also probably a ‘working surface’ associated with the construction of this building. It is too early to determine the extent to which the site was laid out and built during one main construction phase, or grew more organically, although Trench 2 has yielded two possible examples of internal modification during the occupation of the building here. The irregularity of the southern jamb of the entranceway between the eastern and western rooms could indicate it has been altered, whilst the irregular line of stones bisecting the eastern room, [11482], was potentially an addition to it, although this is far from clear.

Both the date at which the site was established, and the duration of its occupation, remain unclear. Trench 2, however, yielded the first directly dated item yet recovered from the site: a jar label (obj. 37808) inscribed ‘Year 4, wine of the domain […]’ (translation again courtesy of Dr Marc Gabolde). It is an unusual date for an Amarna jar label, and all the other published examples of this date have been found in the Central City (see Griffith in Petrie 1894, 32; Fairman in Pendlebury 1953, 159). It begs the difficult question of to whom the year date should be assigned. In Year 4 of Akhenaten’s reign, the royal family were probably not in residence at Akhetaten, but it would be a relatively high date for any of Akhenaten’s successors at Akhetaten. The use of the docket as a dating device is further complicated, of course, by the possibility that the vessel was reused and/or remained in circulation for some time at the Stone Village. The significance of the year date is returned to below.

One other major phase, or phases of activity, is represented in the archaeological record. This is disturbance, which presumably took place after the abandonment of the site. The floors seem to have been the target of the destruction. They were perhaps dug over for organic-rich fertiliser, although it seems more likely that they were destroyed by looters seeking goods buried beneath the floors.  But who was responsible? One possibility is that the residents of the Stone Village themselves dug through the floors to retrieve goods that they had buried there. But if this was standard practice, it is difficult to explain why the residents of the Workmen’s Village appear to have left their floors untouched; when the EES worked here in the 1920s, the floors were apparently well preserved. After the 2005–6 season, we speculated that the looting took place within the last century. This now finds some support in the form of two pieces of modern paper from unit (11478) in Trench 2, a deposit of largely churned-over floor sealed by the thick horizon of limestone boulders, (11472). One was part of a packet of ‘Hollywood Cigarettes’, a brand no longer available in Egypt, but seemingly in circulation around 20–30 years ago. The other was part of a magazine or newspaper, which appears of similar age. It is possible that these pieces were brought in by burrowing animals; there was at least one possible animal hole in the eastern face of wall [11484]. Yet there was no trace of burrows or nesting material within (11478) itself. Since the positions of the ridges of stones on the surface of the trench bore a relationship to the subsurface structures, it is likely that the main deposit of loose stones (11472) is largely collapse from the latter. These must have fallen, therefore, after the floors were damaged and (11478) accumulated. It seems unlikely that burrowing animals penetrated this dense horizon of stones, and it is probable that the two pieces of paper were deposited before this phase of structural collapse. It is not unreasonable to associate them with the main phase of disturbance to the site. This implies that the walls were standing intact above the desert surface within the last century, and also that much of the damage to the site occurred within this time frame. The digging out of wind-blown sand that had helped supported the walls may have increased the rate of degradation of the mortar and increased the rate of wall collapse. The question of why the site remained unrecorded until the 1970s, however, is still unanswerable.

The work is also gradually shedding light on the general appearance of the site, including the issue of whether it had a retaining wall similar to that at the Workmen’s Village. It was noted at the close of excavations in 2005–6, that there was no clear trace of a continuous perimeter wall in Trench 1 (although the damage to this part of the site needs to be borne in mind). The southern wall, [11489], exposed in Trench 2, however, does continue to the west and east beyond the trench boundaries. Its line coincides with a band of surface sand running through the southern margin of the stone scatter. Possible ‘corners’ can be discerned at either end, beyond which the density of surface stone decreases. The length of this band of sand (some 40 metres) may be the original east-west length of the site. The excavated segment of the wall, however, is just 30 cm wide, less than half the width of the perimeter wall at the Workmen’s Village. It may be, therefore, that the sharing of outer walls by several structures resulted in a degree of containment of the site – reflected in aerial views – but that there was no rigid boundary around it. The fact that garbage was accumulating on the external surface exposed in Trench 1 could also suggest permeability across the site exterior and interior. On the issue of the internal layout of the site, it is worth noting that moderately well defined bands of surface sand project northwards from the main east-west band of sand just discussed, at what appear to be fairly regular intervals of some five metres. Do these mark separating walls between repeating building units of a type exposed in Trench 2? Again, this is a point that only more excavation can now clarify.

It is too early to say a great deal about the purpose of the Stone Village vis-à-vis the city of Akhetaten. One approach to this issue, though, is to consider its relationship to its closest ‘neighbour’, the Workmen’s Village. It is striking that although ancient roadways sweep around both sites, none runs between them (see Fenwick 2004). Nor is there any surviving trace of a less formal track across the top of the plateau, which is in relatively pristine condition. This would appear to lessen the likelihood that one site served as an adjunct to the other – as a temporary workers’ encampment, for instance, or a settlement for servants – as communication between them would surely have left some trace on the landscape. It could imply, however, that the functions of the two sites were largely unrelated; or that they did in fact overlap in purpose but were not contemporary, one site perhaps replacing the other. The inscriptional evidence from the Workmen’s Village indicates activity here during the reigns of Akhenaten’s successors, particularly Tutankhamun, although the village itself may still have been built during Akhenaten’s reign (see Kemp 1987). It is tempting, therefore, to place the Stone Village within either the very early, or even the very late, history of the city; both scenarios could fit the presence of a Year 4 docket, if deposited close to its date of inscription. Only more fieldwork is going to bring us any closer to the answer.

Publications cited

Fenwick, H, 2004. ‘Ancient roads and GPS survey: modelling the Amarna plain’, Antiquity 78, 880–5.

Kemp, B. J., 1987. ‘The Amarna Workmen’s Village in retrospect’, Journal of Egyptian Archaeology 73, 21–50.

J. D. S. Pendlebury, 1951. The City of Akhenaten III, London: Egypt Exploration Society.

W. M. F. Petrie, 1894. Tell el Amarna, London: Methuen & Co.