The Lavant Caves
The existence of a network of subterranean tunnels at Lavant had for many years been suspected due. Confirmation of the presence of underground workings, came in around 1890 when an unnamed shepherd lost two hurdles he had been carrying through an opening in the roof of the buried feature. Realising the potential significance of the find, the landowner, the 6th Duke of Richmond, forced a brick and mortar stairway into the caves and commissioned two members of the Sussex Archaeological Society, Charles Dawson and John Lewis, to conduct a further investigation.
Having gained access to the Caves, Dawson and Lewis set about examining the subterranean workings. Dawson busied himself with the investigation of floor debris, whilst Lewis began recording the nature and extent of all tunnels. A total of five irregular interconnecting tunnels and three domed chambers were partially cleared during the 1893 season. The maximum height of the galleries were between 4 and 5 feet (1.2 and 1.5m).
The lower layer of chalk rubble in the caves contained the bulk of the cultural and datable artefacts. These were formally presented to the landowner, the Duke of Richmond on after the exploration. It is fair to say that the assemblage was diverse, containing pieces of Roman mosaic, bronze ware and pottery, sixteenth century metalwork, a possible Georgian halfpenny, human teeth, some prehistoric worked flints, a miner’s chalk lamp, a red deer antler and objects of amber, lead and silver.
The first and most obvious problem surrounding the finds assemblage is the lack of coherency. In fact the overall feeling is of “the sweepings of rubbish thrown out of some antiquarian’s collection”. The prehistoric artefacts recovered, the flint, antler and worked chalk, have at times been used to propose that the site was in origin a Neolithic flint mine. Looking at the Lavant Caves, however, it is clear that they do not fall into the category of flint mine: the most serious concern being that they do not cut through any major seam of subterranean flint. The shape and scale of the galleries also do not conform to the standard type of Neolithic shaft, which usually possesses cramped tunnels and restricted areas of extraction.
If the Caves are not prehistoric then perhaps the artefacts suggest a period of later use? The Roman finds are neither uniform nor convincing. If the Cave had been a place of Roman activity, it is surprising that the people who left “these numerous bronze ornaments” left nothing substantial. The Cave could of course have possessed some religious significance, though the clear absence of coins would require some explanation. The post Medieval artefacts pointed to a yet later period of utilisation.
The poor levels on site recording at Lavant and the absence of a definitive publication, when combined with the diverse and disparate nature of the artefact assemblage (not to mention Dawson’s apparent reticence in discussing it) has led some to suggest the possibility of fraud. That the Cave itself existed, there can be no doubt, but there remains a significant question mark over the finds that Dawson and Lewis claimed to have made.
Verdict
The worked flints are certainly Neolithic or Early Bronze Age in date, and would be compatible with the Cave being a prehistoric flint mine. Unfortunately, the form of the Cave is in direct contrast to every known example of a Neolithic flint mine, and the fact that no subterranean seams of flint were detected during the cutting of the shaft, would seem to argue conclusively that this was not its primary function. We must ask, therefore, if the Cave is not prehistoric, then how did the worked flint arrive inside? The flint could have been residual: ancient finds accidentally incorporated within the fill of a much later shaft. It could have fallen through fissures in the roof of the Cave. It could have been brought in deliberately by an unscrupulous member of the 1893 excavation team, hoping to push back the date of the site by a couple of millennia.
Given the small size of the artefacts recovered, any piece could easily have been smuggled in and planted. The question here is whether such material would have been brought in to fool the site director Charles Dawson, site recorder John Lewis, or one of their largely anonymous workforce. But, one has to ask, why would anyone actually want to do this? What would be the point of fabricating the artefactual evidence of the Lavant Caves? Who would hope to benefit from such a convoluted deception?
A convincing case can be built up against the chief archaeological investigators, Charles Dawson and John Lewis. Both would have benefited from an impressive site, well stocked with artefacts. Both would have had easy and unfettered access to the Caves. Neither would be under suspicion on behalf of the landowner or from any visiting Archaeological Society’s or members of the general public. From their point of view, a ‘few worked flints’, a simple ‘chalk cup’ and a piece of deer antler would help to establish a Neolithic date for the Lavant site. Other, later artefacts would have provided additional finer detail, demonstrating that the flint-mine possessed greater currency than those normally investigated, having been reused in the Roman period and again in the sixteenth century. The chronological depth certainly made the Lavant Caves appear interesting, more so than a mere post Medieval chalk quarry which is what, in reality, the Caves appear to have been.
Of all the suspects in the case, Dawson would appear to have gained the most from any potential sabotage. His reticence in publishing the details of his work at Lavant could thus be easily explained. The site was an ‘important’ discovery which was much heralded in the pages of the local press and amongst national archaeological societies such as the Royal Society of Antiquaries. All this helped establish Dawson as an antiquarian of some repute, but it would not have been a site that he would have wanted to dwell on for long. Publication would have required interpretation, perhaps drawing attention to any inconsistencies in the original data set.
