t: +20 12 5113357 (mobile)
t: +20 27955666 (office)
The initial sorting of potsherds from the excavation at Grid 12.

Home/Recent Projects/Excavation/Grid 12/

Grid 12 - 2004


Progress of excavation
Plans and sections
The general condition of the site and its processes of formation
Selection of finds

Progress of excavation

Barry Kemp
Each new Amarna excavation is conducted within a grid of five-metre squares laid out to correspond with prominent permanent features, to assist in maintaining the grid’s alignment from one season to the next. Each new grid is given a number, and the new one is no. 12. It is aligned to the main north wall of Ranefer’s house, which forms part of the grid. The grid squares are identified by letter and number combinations, the numbers running from south to north, and the letters from west to east. The first task of the season was to lay out a portion of the grid, the corners of the squares intended for excavation marked by iron spikes hammered into the sandy ground, and the sides of the squares defined by orange string (click here for gridded contour plan). When the time came to start the excavation, four small teams of workmen were hired; two to work with the Qatari team and Barry Kemp, and two with Anna Stevens assisted by Andrew Bednarski. By the end of the season a continuous block of seven of the squares had been excavated.

Grid 12 prior to excavation, facing south-east. Note the undulating spoil heaps of the early 20th-century excavations in the background.
Grid 12 prior to excavation, facing south-east. Note the undulating spoil heaps of the early 20th-century excavations in the background.

The centre of the excavations was occupied by the walls of a house towards the smaller end of the range at Amarna, with a courtyard (area 8) attached to the east side (click here for final top plan). It lies within square N50 of the original 200-metre grid of squares originally created by the German expedition. On consulting the list of house numbers used in the past (it is published in the Amarna Survey volume) the next unused number in the N50 sequence is 36. So the new house becomes N50.36. (The house of Ranefer is N49.18.) For ease of description the principal spaces that constitute the house have been given separate area numbers (click here for plan showing area numbers).

Most of the walls of the house were of the thickness of a single brick laid lengthwise, and such walls are prone to collapse down to the last few courses, as had happened here. As with Amarna houses generally it centred on a square living-room (area 2, 3.55 x 3.87 m). The faces of the surrounding walls had been given a thick coat of mud plaster [10596] which, on the west and north sides, and the northern part of the east side, had been made of pale desert clay (the remainder being of darker alluvial clay). Parts of the original floor [10597] remained, made from a thick layer of mud apparently without the inclusion of mud bricks. Embedded within the surface were many tiny fish bones, of which samples were collected for identification. On the south side of the room a platform or dais [10598] had been built on top of the floor. It was made from an edging of one or two lines of mud bricks, the remainder of the space being filled with ashy soil, thinly plastered over with mud. Much of the dais has been robbed out, but the continuation of the brick edging is marked by traces of mud mortar [10741].

House N50.36, facing west
House N50.36, facing west

Three or four narrow doorways were set into the sides of the central room. That on the east led in from an outer room (area 1, 3.4 x 2.1 m). The original wooden pivot-block for the door was still preserved in position. In the south-west corner of the central room a door opened to a small chamber (area 7, 2.95 x 1.95 m). A part of the threshold survived, consisting of a wooden beam 5 cm wide [10582] set within bricks and mortar. The adjacent corner between walls [10591] and [10592] is made not from mud bricks but from a dense mass of mud containing numerous cavities [10583]. This could represent where a vertical wooden post, acting as a doorjamb, has been turned into a termite nest, with the loss of the original wood. Nothing remained of the original floor of the room. Area 3 (1.80 x 5.15 m) is a longer room at the rear of the house, entered from the central room by another doorway in the south-west corner, between two brick buttresses [10594] and [10595], the latter forming the end of wall [10739] which had been strengthened by a second buttress [10740]. Part of the original floor [10738] survived, made of mud bricks laid somewhat irregularly in mud mortar.

It was not possible in the available time to excavate every single deposit, and in the north-west corner of the house (square R6, area 5) one such deposit [10633] is probably covering the foundations of a staircase of which the brickwork [10670] would have been part. It might have been entered from a doorway in the north-west corner of area 2 although this remains to be properly ascertained. Area 4 (4.15 m long) is an L-shaped space in the north-east corner of the house which has no clearly marked doorway, although the buttress part of the way along the north face of wall [10647] might mark one side of a doorway leading from the north-east corner of the central room. Area 1 was the first room of the house to be entered from outside, through a doorway (found poorly preserved) in its east wall [10320]. It also contained the door to area 6 (4.77 x 2 m) which occupied part of the southern side of the house and had been subdivided by means of a wall [10593], which reached only part of the way across the room and ended in a buttress.

The house opened from a courtyard (area 8, 7.35 x 3.7 m) that extended along the east side of the house. Set into the floor was a limestone mortar [10679] of the kind that we know was primarily used for the pounding of the cereal emmer as a way of dehusking it prior to grinding on a quern. Part of a small brick box-like structure [10418] does protrude into the excavated part of the court on the east side, and this could be part of a standard brick quern emplacement. A cylindrical oven [10465/10501] of standard size and design stood in one corner of the courtyard. The courtyard could therefore have been the place where the household prepared its cereal food. The entrance from outside seems to have been located along the south wall, to judge from the presence of a limestone door pivot block [10638], which was exposed at the end of the excavation season. It lay within a deposit of ashy sand and rubble [10637].

Square T5, with blackened pits <10444>, <10512>, <10519> and <10544> cut into the sandy gebel surface. Note also the circular oven in the lower right. Facing east.
Square T5, with blackened pits <10444>, <10512>, <10519> and <10544> cut into the sandy gebel surface. Note also the circular oven in the lower right. Facing east.

As the excavation progressed downwards to the original desert surface several pits were discovered. Four shallow ones occupy square T5 ([10444], [10512], [10519] and [10544]). They were floored with mud stained with ash and probably derive from a specific common activity performed in all of them, perhaps the use of ovens. Ash had been banked against wall [10330]. This is evident from the even ashy stain, which covers the face of the wall, even though the original deposit is no longer present, having probably been removed when the area was turned over. The deeper pit [10413] in the middle of S6 (area 8), however, is so inconveniently placed as to suggest that it belongs to an early moment, before the house and its court were fully in use. The loss of the floor surface in the court prevents us from seeing the precise relationship between the two.

Not every wall exposed in this year’s excavations belongs to house N50.36. In the south side of squares R5 and S5 are the walls of another building, presumed to be a house, from the older excavations (from 1923 where the plans are not only unpublished but lost). Two rooms are defined, by walls [10590] and [10611], both built against the south wall [10589] of house N50.36. In square T5 are further walls and floors from other structures the nature of which is not clear from the limited area exposed, although the group on the south side of the square (units [10307] and [10408]) form the corner of a building.

Diagonally opposite, principally in square Q6, are thicker walls from a separate building to which we have not yet assigned a number. Two of these walls ([10276] and [10277]) come together to define the north-east corner of a rectangular space, bounded on the south by wall [10581] in square Q5 (not excavated this season). These walls are of two bricks in thickness. Against the north-east corner of this space, and occupying the north-east corner of square Q6, lay a room or enclosure, of single-brick thickness. One of its walls [10288] remained unaltered for the duration of the site’s occupation, forming a boundary to the rear part of house N50.36. The other wall of the pair [10523] had been demolished to its foundations and replaced by another thicker version [10289] built parallel to it, but at 65 cms to the west and related to a floor at a higher level. The earlier wall [10523] had possessed a pair of buttresses [10599/10732] on the inside face and these had flanked a doorway in front of which a broad step of mud bricks [10294] had projected. The interior of the room had originally been floored with a layer of mud bricks over which mud plaster had been spread. A small patch of this floor [10585] remained beside buttress [10732]. The replacement wall [10289/10731], although thicker, had kept the same design of doorway, with internal buttresses. Nothing remained, however, of any step or threshold, the original brick step [10294] by this time being buried beneath a layer of sand and rubble.

Square Q5 facing south, showing the later thicker wall [10289] west of the earlier wall [10523]. Note also the circular cast of a pottery vessel in the mud surface. Facing south-west.
Square Q5 facing south, showing the later thicker wall [10289] west of the earlier wall [10523]. Note also the circular cast of a pottery vessel in the mud surface. Facing south-west.

A small area of the floor of this second phase (R6 [10737]) remained in the south-east corner of the room. It lay about 40 cm above the desert, but the continuation of the floor westwards might have sloped down as it approach the doorway. This fragment of floor was part of a larger deposit of mud [10736] which had been laid down, presumably filling a hole dug into the later floor, to fix in place a buried container (presumed to be a pottery storage jar). All that survived of the container was a rough cast of its shape in the mud [10580], perhaps the result of forming a hollow shape by hand before inserting the pot itself.

In front of the doorway lay an area, perhaps originally open, which contained no structural elements. The floor deposits were no longer preserved except in a narrow strip beneath the walls ([10731] and [10289]) which flank the doorway. This consisted largely of ash and suggests that originally this area had been floored with a layer of ash, which had subsequently compacted down to a thickness of about 10 cm. (click here for profile).

View of squares Q6, R6 and R7 showing walls belonging to houses other than N50.36. Facing west.

Plans & section

Final top plan
Gridded contour plan
Section of east baulk of Q6 and north baulk of R5
Section of north baulk of Q6
Profile of west face of walls [10730] and [10289] in Q6
Plan showing area numbers
Plan showing wall units
Plan showing fill units

The general condition of the site and its processes of formation

Barry Kemp
On reading the reports of the older excavations one gains very little idea of the condition in which they found it. The British plans show only walls with blank white spaces in between. The German plans are more detailed and often indicate what the final floor or ground surface looked like, but add no descriptions. It is easy to be lulled into thinking that Amarna was a bit like Pompeii , and that the removal of the sand and rubble will reveal more or less intact floors and the deposits left behind at the moment of abandonment. Far from it. Amarna stood as an open abandoned site close to the river and centres of habitation. It quickly became apparent this year that, although the modern surface of the ground shows little sign of disturbance, much of the site had been turned over and floors destroyed in some past era. The most striking evidence for the antiquity of some of this activity was provided in square Q6. The upper levels preserved large patches of a wall, which had collapsed, in a single sheet of brickwork over a layer of sand, both of them undisturbed. Yet the entire floor of the room had been removed. It was particularly clear in the section face that the digging up of the floor had occurred whilst the adjacent wall was still standing about two and a half metres high.

The implication is that, once the site had been abandoned, scavenging began and was thoroughly pursued. As to what people were looking for, again Q6 provides a clue. In one corner there remains the impression in a deposit of mud of a large bulbous storage jar, which had been buried to its neck beneath a mud floor constructed at a higher level when the room was remodelled. It was in such buried jars that ancient Egyptians stored small valuable things. The famous ‘crock of gold’ discovery, which Pendlebury made in 1930 in the North Suburb, is an example. The lure of the occasional discovery of treasure of that kind would have been enough to propel ancient hunters to very thorough investigations.

What does this mean for the future? Even the thoroughness of ancient treasure hunters was not total. The intensity of disturbance varies from one area to another. The small sample-area of housing excavated in 1987 (published in Amarna Reports VI, house P46.33) had survived much better, though it was not free from disturbance. Ranefer’s own house, itself within grid 12, is another and much closer example. Here and there pockets remain where more has survived undisturbed, and it is in these pockets that information about living-conditions is better preserved.

Selection of finds

Anna Stevens

34119 Monkey figurine 34119 Monkey figurine

T5 [10317]
Length 3.0 cm. Height 1.8 cm. Thickness 0.95 cm. Diam. of hole 0.4 cm. Limestone

Small figure of monkey standing on all fours and straddling a pile of rounded objects, probably fruit (grapes?). Carved in the round. The figure stands on a low, flat base. A sub-circular hole has been bored through the figure, beneath the stomach. A small portion of the base is chipped. Patches of red paint are preserved on the body, head and limbs of the monkey, fruit and base. There is a possible patch of black paint on the back. The figure is slightly blackened from direct burning. The figure is smaller that most monkey figures from Amarna. The presence of the hole is also somewhat unusual. A mark has been roughly incised on the base.

34124 Monkey figurine 34124 Monkey figurine

T5 [10348]
Length 4.6 cm. Height 3.35 cm. Thickness 1.35 cm. Limestone

Lower part of a figurine, probably of a monkey. Carved in high relief. The figurine has a slight pedestal with slightly convex base. On either side, a foot and lower leg is shown, straddling an upright, curved object in front, probably a pot or basket. A further upright object is carved behind the feet, perhaps a seat? Traces of red paint appear on the vessel, the feet and the ‘seat’. The paint on the vessel is possibly in a gridded pattern, perhaps to represent basketry? There are possible patches of black paint on the vessel, base and behind the legs. A patch of blackening on the base is certainly the result of exposure to heat, the interior of the stone having turned pinkish.

34125 Lotus flower inlay 34125 Lotus flower inlay

T5 [10348]
Length 5.4 cm. Breadth 3.85 cm. Thickness 0.65 cm. Faience

Fragment of inlay with floral motif, the leaves and petals of an open lotus flower. The material has a sandy grey core, 0.58 cm thick, to which the glaze has been applied directly. The glaze, green, red, blue and white, is even and thin. The back surface is relatively flat and has a vaguely ‘gridded’ surface (from resting on cloth?). The glaze is slightly pitted, not the result of weathering.

34126 Seal impression 34126 Seal impression

Q6 [10281]
Length 1.55 cm. Breadth 1.3 cm. Thickness 0.5 cm. Mud

Fragment of a small seal impression, bearing the name [Neferkheperure] Waenre. The mud has a fine-grained matrix with no visible inclusions. It has been burnt. Linear impressions (wood grain?) occur on the reverse. The reverse surface contains a distinct change of angle: the seal was not applied to a flat surface. Similar, but not identical, to Petrie types 61–4.

34126 Seal impression 34144 Fragments of shoes

T5 [10445]
Length (a) c. 18 cm. Breadth c. 6 cm. Thickness c. 0.2 cm
(b) c. 17 cm. Breadth c. 5 cm. Thickness c. 0.2 cm. Leather

23 fragments from the soles of two leather shoes, apparently a pair. Two larger pieces (a) and (b) are from each sole. One piece of strapping was joined to the sole at the time of discovery but subsequently separated. The fragments are hard and curled, some pieces being folded over. Some pieces have a resinous appearance. The large pieces are oblong, both being broken at one end. One or two rows of stitching-holes occur around the perimeter; in places the leather stitching survives. Made from a single thickness of leather?

34126 Seal impression 34148 Fragments of leather sheet

T5 [10466] (lower horizon)
Length (largest) 6.5 cm. Breadth 5.8 cm. Thickness c. 0.1 cm. Leather

About 66 fragments of leather sheet with short linear scores across the surface. Most pieces are flat or slightly bent; some are folded over. There is some variation in orientation and spacing of the score marks; for example, some are in a zig-zag pattern; some are in straight lines. The leather is hard but fragile, and slightly resinous in section.

Progress Map 2006


Website first posted September 2000; last updated October 2017 | enquiries concerning website: email bjk2@cam.ac.uk