Basic background figures
Table 1 below shows the figures for four West-country counties.
The table shows that in the counties with longer coastlines in relation to the area, there are fewer mills recorded in the Domesday survey than might be expected in comparison with other counties. Compare column 4 with column 5. It provides a prima facie case for a more detailed examination. This is however a very crude measure because, as will be shown below, not all coastlines are suitable for the construction of tide-mills.
Is there an anomaly to be explained?
Does the lack of mills in Cornwall in 1086 need any explanation? Might it simply be misleading to compare the figures county by county. Domesday mills are not evenly distributed in Domesday England. For instance in 1086 in Cambridgeshire in a total of 16 Hundreds, there are listed about 130 Domesday mills. But 89 of those mills are concentrated in six of those Hundreds. By contrast seven of the Hundreds have only one mill or no mills listed. The Domesday mills are not evenly distributed there or in other Counties. Might then the lack of mills in Devon and Cornwall be explained by a similar, perhaps random, uneven distribution? A study of the distribution of Domesday or more modern watermills in other Counties shows that the patterns are certainly not caused by random chance. In Cambridgeshire for instance the distribution can be explained. Those Hundreds in Cambridgeshire with no Domesday mills never later acquired any water-mills and there reasons for this. The available water-power in a low rainfall region, the shallow gradients of the landscape and additionally the geology prevented the construction of watermills in large areas of Cambridgeshire. The distribution pattern is entirely explicable. (Keith 2017) The landscape always determines where watermills can be established and where they cannot. In Devon and Cornwall it is unlikely that the main constraints to the establishment of fresh-water mills listed above apply. Devon and Cornwall have a much greater rainfall and many watercourses with steeper gradients.
Previous opinions and comments
Many, including Darby, have noted the anomaly, and been surprised by the scarcity of mills recorded in the Domesday survey for Devon and Cornwall. It appears no one suggested that it was mere chance. (Darby 1977 appendix 14)
The pattern is most striking in Cornwall where there are only six mills recorded in five locations. In Devon there are 96 recorded in 78 entries.
There seem to have been few attempts to explain why this is so. But Margaret Hodgen did directly address the issue and suggested that this anomaly resulted from the introduction of watermills to Britain having been comparatively recent in 1086 and the diffusion of the technical knowledge was at that time in the process of spreading from east to west in England. (Hodgen 1937) This thesis is now widely discounted as many have shown that watermill technology was well established in the Roman Empire and had then already spread to Britain. (Spain 2008 and Wikander 2000) It is widely accepted that that water mill technology was not new in 1086.
Another explanation for the lack of recorded Domesday mills could be that many of the Devon and Cornwall mills were not under manorial control and therefore not profitable enough to be recorded in Domesday. The rainfall and gradients could have supported many small mainly horizontal mills serving only one or two land holdings and therefore not worth manorial control. This was apparently so in later years in the 686 km2 Camel catchment area in which 197 mills or mill sites have been identified. (Bodman 2015) This theory cannot be dismissed but it is not clear that in 1086 feudal control of resources and society was any weaker in the West Country than elsewhere. The manors’ milling monopolies were of considerable worth and is the only class of asset with individual values recorded in the survey.