For me, the femaleness of the group was striking: Henry surrounded by Barbara, Gill, Cath, Helen, Mary and Pippa ...

I'm glad there are so few men. We can avoid the macho types who are so often attracted ...

We decided to excavate part of the terminal area of the Leskernick stone row because: i) it was a small entity; ii) an interesting number of ritual and landscape features were visible from it; iii) it was situated approximately half-way between the stone circles and might be considered as a focal point in the overall zonation of the ritual monuments in the landscape. We chose to investigate an area between one of the terminal stones and the first stone of the stone row because the final stages of the approach to the terminal may have been a zone of special significance, in the passage along the row and access to the terminal. The trench boundaries incorporated one of the three recumbent stones of the terminal setting, and was oriented ENE along the axis of the stone row ('pdf' stone row terminal: location of excavation trench).

Sue has brought four of her own fairly massive 2" grid pegs that form the initial basis of the grid. These seem ridiculously thick and cumbersome to me, but I'm told they won't be easily knocked over, or dislodged, unlike my own ...

An area excavation, rather than test pits was chosen to maximise the possibility of revealing the types of features and activities which we thought would be probable in the vicinity of a stone row terminal. These possible features included post-settings, the stone-holes of the now recumbent stones, evidence of human or animal burials or cremations, and artefact deposits or scatters.

lan and I set up the site grid, with Chris occasionally holding the end of a tape with a lost look in his eyes. Having got this underway Chris started setting up the fence around the trench area. It amused me that the person who least liked the 'rules' of excavation had fenced us in ...

Back at the camp site the warm sun had brought out the midges in droves. An unpleasant hour was spent battling with them before I retired to the bar ...

Tuesday 13 June

We deturfed the trench taking off grass root mat layer which was c. 70 mm thick.

It felt as if we were disturbing (mutilating) a landscape at rest ...

At the same time, the locations of each of the stones in the row, together with the centre of each of the stone circles, were marked out using white flags.

These flags drew our eyes out of the fenced-in trench and facilitated the consideration of wider sets of relationships between the 'trench' area, the rest of the 'ritual complex', and dominant focal points in the landscape. The waving flags made the 'trench people' feel part of a wider landscape and helped us focus upon the inter-relationships between the terminal setting and the other stones. The wild horses were magnetically attracted to the flags and completely ignored our trench, which we had specifically wired them out of ...

Wednesday 14 June

I am amazed at the neatness of the turf stack outside the trench, the obvious aesthetics of the straight lines and neatly numbered grid pegs. I remark on the professionalism of it all to Ian and Sue: 'It looks just like an excavation should look'. Eyebrows are raised. There is a green baize door through the fence into this interiorised little world. Highly incongruous in the middle of nowhere, little figures huddling behind it, peering down at stone and soil. The creation of another reality, dark and secret. Inside the door there is discipline: you cannot smoke in there, there are places one must not walk, lines, pegs, tapes, objects which should not be disturbed ...

Under the grass mat layer there was a dark brownish-black soft, silty horizon (Context 1), which stained the hands. Mixed in with this were coarse elements of occasional granite pieces and, in the south-east corner of the trench, some slate fragments. This silty layer (c. 0. 10 m in depth) was removed down to the top of a bleached soil horizon (Context 2), which comprised a friable mid greyish-black coarse sand.

Two features quickly became apparent. One was the top of an 'incomplete' circular arrangement (c. 0.30 m internal diameter), of five granite pieces (Context 4) with a 'missing' stone on the north part of its circuit, This 'circle' was just 0.20 m east of one of the recumbent terminal setting stones (Context 14). The other feature was a sub-circular concentration (c. 0.60 m across) of thin pieces of slate (Context 8). This concentration disappeared into the east edge of the trench and corresponded to where we had noted the presence of slate on deturfing. The slate fragments included tile-sized pieces and, due to its proximity to the surface, was interpreted to be a relatively modern dump of building material (see Appendix). Slate is not local to the site but slate stone occurs north of Bodmin and is widely-used for roofing.

Spent the afternoon helping at the excavation. Henry mattocked, I shovelled and barrowed. Tiring and fairly tedious. Not much for the imagination to work on, though Pippa discovered that one of the recumbent stones was 6 ft plus ...

The pre-excavation plan ('pdf') of the trench was commenced by planning the three recumbent stones of the terminal setting. The dimensions of two of these stones are evident since the stones lie wholly on, or very near, the present day land surface. The third stone lay about one metre outside the west edge of the excavation trench, aligned along the axis of the stone row, with its western end deeply buried. Probing established its length to be at least 1.95 m. This is 0.50 m higher than the other stones and it would have been the most spectacular of the stones in the terminal setting.

Thursday 15 June

The houses were being identified on the ground. This activity involved constant movement across the landscape, which created a sense of freedom which was difficult to achieve on the excavation. Indeed, by contrast the excavation was creating a sense of unnatural fixedness ...

With further trowelling, two more features became evident. The first was a large 'sausage shaped' area of dark brownish-black sandy silt (Context 6) situated in the middle of the trench, north-west of the slate concentration and oriented NNE-SSW. The second was a semicircular area of mid-brown friable silt (Context 7), which continued into the west edge of the excavation trench at the point that it cut across the end of one of the recumbent stones (Context 13). We thought that the sausage-shaped area might be either a natural depression which retained moisture or a footing trench for a semicircular structure such as a windbreak, but much more recent in date than the stone row. The two areas (Contexts 7 and 4), next to the recumbent terminal stones, remained the most interesting. The mid-brown silt (Context 7) was full of rooty material. We interpreted it as having accumulated in a waterlogged hollow (Context 11). The circle of granite pieces (Context 4) was clearly deliberately placed, and in alignment with the stone row. Perhaps we had a stone-lined hearth? Or a cremation pit? But the stones were not affected by heat. Alternatively it might be a specially lined pit for the deposition of offerings: pieces of white quartz, charcoal from non-local oak timbers - material known to be deposited under cairns in the area. The uppermost fill (Context 5) within (and possibly encircling: we were not sure yet) the circle of granite pieces (Context 4), was all that was revealed. This fill was dark and of soft consistency, A new possibility emerged: we might have the post-pipe of a substantial rotted post, c. 0.30 m wide, which could be part of a post alignment preceding the stone row. Perhaps the terminal area had incorporated timbers as well as standing stones?

Some of the rust that had been gradually accumulating on my new trowel was now worn away. I noticed how large it was compared with other trowels in use and was duly informed that all their trowels had started out this way. The more diminutive the size of your trowel the greater your status as an archaeologist since years of scraping were required to reduce the blade to an area little larger than a postage stamp. The trowel was a prized personal possession and a lengthy discussion ensued about the best place to carve one's initials, or name, on the wooden handle. This also would show signs of longevity - a sleek oiled surface produced by being pressed into the palm of a sweaty hand for months on end. Another quaint archaeological fetish. This ageing of objects, through appearance, and the clear relationship between use, time and status reminded me very much of Kula valuables but while these are given away it would be horrific for an archaeologist to give up a trowel as it was so obviously entangled with personal identity. I should have spent several days gradually filing down the blade of my trowel before the excavation commenced. But even if I had done that the handle would have given the game away. Taken to its logical extreme, the greatest status symbol of all would be to have no trowel at all. How foolish I had been to purchase one! ...

Friday 16 June

The southern half of the fill of Context 11 was emptied to produce a west-east section ('pdf' stone row sections). Under the root-filled layer (Context 7), there was a dark brown silty clay (Context 9) abutting the sides and base of the hollow, We interpreted Context 9 as a primary inwash into a cut or depression (Context 11) in the bleached soil horizon (Context 2), which had subsequently become waterlogged (the interpretation of the previous day), causing the build up of the rooty mass of Context 7. The fill thus came about by natural processes, but how the hollow had been formed in the first place remained unresolved.

It was good to walk across the settlement and feel the freedom of moving through a landscape. The lack of ambulation in an excavation trench closes down some of the senses, and also concentrates others, slight changes in texture, compaction, sound etc. as the trowel blade scrapes along and slices through fills. The excavation trench seemed part of a secret world which could not be seen from the western settlement ...

The edge of the cut (Context 3) in which the granite pieces had been placed was finally located. The cut was now seen to be wide of the outer edge of the packing by c. 0.20 m. This suggested a pit with an internal circle of stone pieces and not, as we had thought, a post-hole with stone-packing. We now believed that we had located the original stone-packed hole of one of the fallen terminal stones.

Sunday 18 June

Here one worked within a structured, overtly hierarchical environment. While the fencing off of the pit is, of course, necessary it gave the whole process of the excavation a certain alien character. The fence around the pit was like a metaphor for the divisions in attitude towards archaeology among the team ...

The 'sausage-shaped' feature was half emptied to create a west-east cross section. It was surprisingly shallow (c. 40 mm max. depth) with rounded irregular sides, concave profile, and no perceptible break of slope. It contained a single fill (Context 6). We think that it is a relatively modern scuff-hollow created by animals.

The smallness of people in the landscape. Looking up from the prehistoric stone row, people on the settlement seem tiny. Skylarks, yellow flowers, fluffy cloud ...

Excavation of the exposed part of Context 11 was completed to reveal a north-south section in the edge of the trench. This feature had to be earlier than the fall or removal of the recumbent stone (flat-topped at both ends), which projected over the point that the feature continued into the edge of the trench. Parts of this stone (Context 13) had broken away to rest at the interface between the two fills (Contexts 7 and 9), while the stone itself was resting at the interface between the 'topsoil' (Context 1) and the uppermost fill (Context 7). Our interpretation was that while it was still standing, a hollow had been created around the standing stone, perhaps by animals, and this had become waterlogged and silted up. There was no evidence of an original stone-hole, suggesting that we possibly had the uppermost end of the standing stone exposed in the north-south section, The hollow was oval-shaped indicating that the stone originally stood on a north-south axis. The recumbent position of the stone reflects this supposed original alignment.

Monday 19 June

The ignorant, including myself, are kept well away to minimise damage to certain areas ...

A west-east section was created across Context 3 by removing the southern half of the fill. The edge of the feature was still extremely difficult to define. The bluish fill of Context 5 (first thought to be a post-pipe fill) now seemed to be a shallow silting across the feature. Underneath this it gradually became clear that the fill (Context 12) within the granite circular arrangement of stones also continued for c. 0. 10 m on the outside of its southern circuit, where it adjoined the compacted edge of a different fill (Context 10). On the south-west side of the circuit the compacted edge was seen to mirror the projected imprint of three of the granite stones, if their uppermost ends were sloped further backwards to meet it. This indicated that the granite stones had shifted. As the sectioning continued, the 'missing' stone from the granite circular arrangement appeared in the middle of the fill ('pdf' section underlying stone). Neither the movement of the stones, nor the position of the 'missing' stone, suggested that they had been caused by a fallen standing stone. The stone must have been carefully removed with the packing stones shifting marginally, and the one stone loosened from the circuit being placed (deliberately?) in the void left by the removed terminal stone. A new, and exciting interpretative possibility now emerged: a deliberate, but careful, dismantling of the terminal setting in the Bronze Age.

This decommissioning of the site had been done in such a way as to preserve the essential character of the monument by: 1) only destroying its distant visuality, by taking down the conspicuous stones of the terminal setting; and 2) by selective disalignment of parts of the stone row only beyond a specific point - the boggy area marked by the leat created by later, post-medieval, tin-streaming.

Tuesday 20 June

This is the first day I have spent most of the time on the excavation and it does provide some relief from shouting out 'Fowey valley and hut 20 straight out, Brown Willy and hut 24 to left ...

The difficulties in understanding the relationships between the fills in the stone-hole (Context 3) were further resolved by quarter-sectioning the remaining fill. Context 10 was a grey-brown silty sand, c. 0. 10 m in depth and wholly rested on a ledge. The abutting Context 12 , a mid greyish-black clayey silt, partly lay on this edge, but towards the centre of the fill of the granite setting the edge of the ledge was reached, and the fill deepened considerably, Context 10 was interpreted as the silting or backfill around the granite packing which kept a stone upright in place. Context 12 was understood as the backfill after the stone had been removed. It became clear that the stone-hole was surrounded by a ramp probably to help set up the stone, but the orientation of the stone-hole and its dimensions remained unclear.

There is a distinction between the excavation and the settlement work. The excavation, because it is destructive, has to be more detailed, more obsessional. There are other differences. About looking down in an excavation, as opposed to looking out. About the specificity of the small area under excavation. The 'box' that one has created, and that, if one is not very careful, is divorced from the multiplicity of nested scales of action, movement, thought, sight, understanding ...

We began taking off the bleached soil horizon (Context 2). Because of little remaining time, we decided to concentrate on the western 3 m of the trench (an area of 3 x 5 m). This means that the slate concentration (Context 8) is unexcavated.

Wednesday 21 June

The excavation was proceeding at a snail's pace and I am longing to see what is at the base of the stones ...

Mattocking of the western part of the trench was continued down to a mottled dark orange, sandy, iron-rich soil horizon (Context 16) known locally as 'rabb'. No other features were discovered.

The excavation of the stone-hole was completed. Underneath the displaced stone from the granite packing circle there was a thin (c. 3 mm thick) layer of bright orange iron pan. Once the hole was entirely excavated it was possible to establish the original orientation of the long axis, west-east. The stone which it held would, therefore, have been aligned along the axis of the stone row. This stone-hole cut through the Context 2 horizon and penetrated the 'rabb' (Context 16) horizon to a depth of 0. 17 m. The weight of the stone had compacted the 'rabb' and gave some indication of the shape of the base of the stone that had originally stood in it. This corresponded with the square-edged southern end of the adjacent (Context 14) recumbent stone, So, the pointed end of the stone would have been uppermost end. It was now also evident that the west side of the stone-hole ledge went under this recumbent stone. The section created at the point this stone crossed the ledge or ramp showed that a final silting (Context 5) and a turfline (Context 1) had formed over the ledge/ramp before the stone came to be in its present position. At the thickest part of the stone its weight had pushed it through the turf to rest on top of the final silting. Three interpretations remain possible:

1. The stone seals a Bronze Age turfline which had already formed around it while standing;

2. During the Bronze Age the stone was dismantled and placed in some unknown location, and at some time after a turfline formed over the pit, the stone was moved to its present position; or

3. The stone was dismantled in more recent times and the stone seals a 'modern' turfline.

Pollen analysis might tell us whether the turfline is prehistoric or more recent, and sampling will be undertaken next year.

I went down and looked at the trench where the turfing had been finished in my absence. The excavation hole was hardly visible. It, and I, felt lost in the landscape ... The different axis of each terminal stone is interesting. It became too dark to think it through. It was a wrench to leave the place ...

Conclusion: August 1995

It seems very unlikely that the stone fell itself because of both the lack of disruption to the setting such a fall would cause and its present position, which could only have been achieved by lifting and placing the stone. It is now possible to reconstruct the original configuration of the terminal setting. A pointed-top stone stood at the end of the stone row with its axis oriented directly along the stone row alignment. The shortest terminal stone, which was flat-topped, would have stood just to the north of it on a north-south axis. If we presume that the tall stone, with one end still deeply buried, fell, or was dismantled, along its axis of gravity, as has happened with the other two terminal stones, it would have been transverse to the axis of the stone row, ie. north-south. This arrangement of the three stones would have created a 'cove', or a triangular space c. 2.6 x 2.6 x 4 m, with a terminal stone at each corner. The positioning of two of the stones on axes at right angles to the stone row alignment would have created a space in which the stone could act as 'backdrops', both partially concealing, and focusing attention on the activities performed within the setting, This configuration of stones would have allowed unobstructed access from the south-east side and the southern stone circle, with a visual focal point being created by the flat-topped stone being aligned along this path of movement. There would also have been unobstructed access from a north-westerly direction, from a spur of land with three small cairns (invisible from the stone row terminal), c. 1 km. away. Approach from the northern stone circle, and the southern settlement area at Leskernick would have been visually disrupted by the transverse orientations of two of the stones. An approach along the stone row would not only have been blocked by the terminal stone on the axis of the row, but additionally, beyond it, by a terminal stone placed transverse to the axis of the stone row. What the stone row, in fact, appears to be doing is dividing two zones of access to the terminal setting.