Difficult as it is to believe that we are now in spring owing to the continued presence of snow on the ground and the threat of more to come, today saw topsoil stripping for our Easter weekend small-scale project. It’s more in the nature of an evaluation exercise than a full-blown excavation: we have stripped a linear trench across the line predicted for the henge outer ditch on the basis of the 1996 magnetometer survey. In previous years, the location of the ditch has been masked by the presence of an unknown depth of colluvium; in 2012, an area of colluvium was removed by machine down to the chalk bedrock, at which point (to the east-north-east of the centre of the henge), it was more than half a metre deep. This was a cause for concern, as the depth meant that we would never get into the ditch fills using the 100% recovery and three-dimensional recording of all finds that we had been practising up to that point.
In addition to this worry, I was concerned that the scale of the ditch suggested by magnetometry (between 3.5 and 5.0 m wide at the top) might indicate a commensurate depth. I had visions of Alexander Keiller’s excavation of the outer ditch at Avebury, where the V-shaped ditch was more than 11 m deep, with a width of 21 m at the top: a similar feature at Stapleton’s field might be almost two metres in depth. Then I was puzzled by the results of the 2011/12 resitivity survey, where the spread of colluvium was visible but the ditch was not. On the one hand, I thought that maybe the colluvium was so deep (perhaps a metre or so) that the outer ditch could not be detected at the resolution achieved (at a depth of a metre, using the survey technique employed, most features would be invisible); on the other, I was heartened by the likelihood that a ditch two metres deep would show up regardless of the depth of colluvium sealing it. Either way, I was beginning to suspect that the ditch was of a more manageable size.
I arrived on site slightly late (about 8.40) to find Tony, Nigel and Jim chatting to the digger driver. Under the snow, it was possible to make out the outline of last summer’s trench, which made it reasonably easy for me to locate the rough centre of the henge. I had already decided that we would open our exploratory slot down hill from the centre of the monument, where the colluvium was likely to be thinner (we knew from last year’s project that it did not extend as far to the south-east as the Roman enclosure ditch). The machine could then cut a trench up to 20 m long from the outside to the inside of the henge. This way, we could minimise the potential damage to deposits within the outer ditch.
There was barely any colluvium at the south-eastern (downhill) end of the trench, although it began to show up within a metre or so of the end of the trench. In the conditions on site (it was -2° C and icy), it was difficult to see artefacts within the colluvium, although Tony managed to identify a piece of Roman tegula, with part of the flange visible. After cutting back for over five metres, with the colluvium becoming increasingly thick (it has reached a depth of perhaps 0.25 m at this point), the chalk bedrock suddenly dived down. This was the outer edge of the ditch, exactly where the magnetometer survey had predicted it to be. It was clear straight away that the side of the ditch was not steep, perhaps around 45°, which raised my hopes that it would prove to be reasonably shallow.
The machine was used to empty the ditch of its contents. Although it is too early to be certain, it appeared to contain only one fill, which was sterile (or, at least, contained too few artefacts and ecofacts to be visible during soil stripping). If true, this is intriguing. It suggests that the ditch filled in a single episode; in other words, it was deliberately backfilled. If this proves to be the case, it explains how the ditch was so full that a colluvium formed over it north-west of the henge from the Late Neolithic onwards (the infamous context (35) that we spent so long excavating in 2011 in the mistaken belief that it was indeed the upper ditch fill). If it is a deliberate backfill, it raises some intriguing issues: why would the outer ditch have been filled and how much effort would have been required for the undertaking? This might be yet another indication that the early ‘formative’ henge was converted to the classic type some time before the middle of the third millennium BC.
The ditch proved to be at the upper limit of size suggested by magnetometry, around 5 m at the surface of the chalk. However, its sides sloped quite gently. Although we stopped the digger at a depth of 1.2 m from the ground surface for health and safety reasons, the angle of slope of the ditch suggests that there is less than half a metre of deposit left in the bottom. We will therefore be able to excavate a slot through the base, where there ought to be some indication of primary silts (and, I hope, something that can be dated). This means that the ditch is no more than 1.7 m deep (and cut under 1.5 m into the bedrock). This is very encouraging for the summer as it means that it will be possible to excavate a complete section through it by hand whilst maintaining our three-dimensional finds recording system.
After locating the inner edge of the ditch, another 1.5 m of topsoil and colluvium was stripped. At this point, over the berm between outer ditch and bank, the colluvium appears to be at least 0.4 m deep, confirming the observation to the north-east of the henge last year. After tidying up the soil scattered by the machine around the top of the trench, we left it until Friday (there is little point in cleaning up inside the trench, as more soil will have fallen in by then). I left around 10.15, while Tony, Nigel and Jim began putting up the hazard fence around the trench as, although it’s in the middle of a field, we need to ensure that people don’t stumble into it in the dark.
The forecast for the weekend isn’t good: there is supposed to be a high risk of more snow on Thursday or Friday and temperatures are not due to get above around 5° all week. This is hardly what one expects for an early spring project, but it’s necessary for understanding the monument. I will be back to blogging on Friday, when we return to clean and record what has been exposed. I also hope that it will be possible to excavate the soil in the base of the ditch. If we are lucky, there may be a primary silt with organic remains in it.