| ISSN 1357-4442 | Editor: Simon Denison |
|---|
| REGIONS |
Cornish farms in prehistoric farmyards
In the first of a new series on recent advances in the archaeology of different regions of Britain, Nicholas Johnson looks at Cornwall
It is often said that Britain has an ‘ancient landscape’, in which modern features – settlements, sacred sites, boundaries, and the rest – are built on top of, or near to, precursors that date back centuries or even millennia.
Nowhere is this truer than in Cornwall; and perhaps the most remarkable achievement of archaeology and historical research in the county over the past decade or so has been to demonstrate how much of the landscape remains essentially unchanged since late prehistory – and in some places since even earlier than that.
A number of strands of evidence have come together to produce this conclusion. In 1985, we obtained a clearer idea of which settlements in Cornwall have at least a medieval origin, when the Cornish Place-Names Index (produced by the Institute of Cornish Studies at Exeter University) was marked up on maps, and all medieval settlement place-names entered onto the Sites and Monuments Record (the SMR). It was known that settlement names with the prefix Tre, and several other forms, were pre-Norman in origin; while settlements that were first recorded in documents or on maps before 1540 were regarded as having a medieval provenance. The place-names themselves were useful – the Cornish ones helped to indicate the relative antiquity of the settlements, with English place-names spreading westwards over time.
Using this and other evidence, it has been possible to identify the areas in Cornwall and Scilly which were enclosed and farmed in the medieval period. Fields associated with farms and hamlets of this date have a distinctive character, often the result of the late medieval enclosure of former strip fields. (By contrast, settlements and fields of 18th or 19th century origin, mostly enclosures of former heathland or downland, are predominantly straight-sided and rectilinear.)
For some years, also, members of the Cornwall Archaeological Unit have been documenting the number and distribution of ‘rounds’ (enclosed prehistoric farmsteads with a single ditch and bank) and other variations of late prehistoric enclosures. About 650 of these are known already, with many surviving as earthworks, but the number is likely to almost double to 1,000 following the systematic examination of all aerial photographs of the county as part of the English Royal Commission’s National Mapping Programme project.
What has become increasingly clear from this work is that areas of land that were enclosed and unenclosed, farmland and heathland, in the medieval period, were used in largely the same way in late prehistory. In general terms, this has been shown elsewhere in the country too – that the zone of medieval settlement (ie, the farming zone) is also, by and large, the zone of late prehistoric and Romano-British farming.
However, in Cornwall, we can go further. It has been suggested that the pattern of rounds is related to the pattern of medieval farms – in fact that rounds appear to fill up and complement the medieval pattern. This seems to suggest that many medieval farms sit on top of, or very close to, prehistoric rounds, and that therefore a substantial number of prehistoric sites have been perpetuated up to today beneath medieval settlements.
The hypothesis is supported by placename evidence (elements such as Ker, meaning fort), as well as earthwork evidence. Examples of farms inside prehistoric enclosures include Tregear, south of Camborne, where the farm buildings lie within the ramparts of a classic round, and there are several others where the farms are built immediately outside the round.
Fogous (stone-built underground chambers) at farms provide similar evidence, such as the one underneath a slurry tank at Pendeen Farm on the Land’s End Peninsula. Fogous are associated with Iron Age settlements, and it is quite likely at Pendeen and other similar sites that an Iron Age settlement developed into a Romano-British ‘courtyard house’, and then later into a medieval farm. Some early churchyards, too, lie within late prehistoric enclosures.
On the other hand, many hundreds of rounds were abandoned as settlement sites and their remains can now be found as earthworks or crop marks in the midst of fields.
Moving still further back in time, there is evidence of Bronze Age settlement within the late prehistoric farming zone: for example, at Trevisker Round and Penhale Round, where substantial Bronze Age settlement remains were found beneath or next to later settlement. Although the general patterns of settlements and fields are often different in this earlier period, there are some indications that continuity of field patterns may occur in some areas such as West Penwith, parts of the Lizard, and around Bodmin Moor. The important point is that the basic farmland/heathland zones appear to stretch back into the 2nd millennium BC, even if we can’t trace continuity of settlement back that far.
A total of 5,349 medieval settlements are recorded in the SMR. We can be certain that this figure is the minimum number of medieval settlements in Cornwall. If we assume generously that over half (about 3,000) of the known medieval settlements were new in the medieval period, related to the colonisation of rough ground (heath or moor) and woodland (much less so), or were new locations within the existing farming zone, then we are left with around 2,500 settlements which might be perpetuating the location of later prehistoric settlements.
A total of 1,000 abandoned prehistoric enclosures and 2,500 perpetuated prehistoric settlements gives a grand total of at least 3,500 late prehistoric settlements in Cornwall. This does not seem excessive, bearing in mind that they may not all have been inhabited at the same time.
However, the possible consequences of this estimate on the likely population level of prehistoric Cornwall, and of Britain as a whole, is remarkable. Given that there may also have been large numbers of unenclosed settlements, few of which have yet been identified, it would seem that the density of rural settlement in the late Iron Age/Romano-British period was approaching that of medieval Cornwall.
Moreover, each round contained at least two or three roundhouses, and sometimes more. Assuming, then, that many of Cornwall’s 3,500 rounds were inhabited at the same time, and that each was home to about 30 individuals, we have a population of Cornwall in excess of 100,000 people. The unenclosed Iron Age sites (of unknown number) would increase this figure, though by an unquantifiable amount. The population of Cornwall in 1801 (192,281) was 1:46 that of England (8,893,000). Using this ratio as a rough indicator of relative population, then this would give England a population at the beginning of the Roman period of about 5 million – roughly in line with generally accepted current estimates.
The antiquity of the Cornish landscape can be seen not only in the continuation of settlement sites over millennia, but also, in more detail, in the survival of the medieval landscape intact.
In land enclosed during the medieval period, where the settlements have medieval credentials, we now think that the majority of lanes, roads and field boundaries are at least medieval in origin. If this is combined with all the medieval sites in the SMR (such as settlements and buildings, churches, chapels, crosses and bridges) then it begins to seem that a substantial percentage of at least the late Tudor landscape pattern and detail survives today. Many field boundaries, or ‘hedges’ – our massive earthen banks with stone facings – have been removed over time to increase field sizes, of course, but there has been no general re-ordering of the landscape.
Incidentally, if we consult the tithe maps of the 1840s, our understanding of the medieval landscape can be further enhanced with the addition of boundaries, structures and buildings lost since the 1840s. As a result it may be possible to begin producing the equivalent of ancient OS Pathfinder maps for Cornwall at a scale of 1: 25,000.
So what does this remarkable landscape continuity say about Cornwall? The dominant impression is that until the 19th century, the system ‘worked’. Although there were periods of change (principally Bronze Age to Iron Age, post-Roman to pre-Norman, and late medieval), when field patterns and a proportion of settlement locations moved around, the change was largely to enclose the areas of heathland, not to reorganise the existing enclosed farmland. There was no incentive to change it. One might regard this as evidence of economic inertia or lack of enterprise; but in the natural sciences, such a state would be regarded as a success, suggesting that the landscape had reached its ‘successional climax’, or that there was a balance between the physical resource (the quality of land, and the climate) and the aspirations of those who used it.
Cornwall’s traditional links, in the medieval period and earlier, were not exclusively with England but with Brittany, Wales and Ireland. Devon and parts of Ireland have rounds, as also do parts of Wales (where they are known as raths); in Devon, too, there are similarly substantial earthen field boundaries that, once built, are not easy to move. I suspect that future archaeological work in these and other parts of Britain’s ‘highland zone’ will find a similar conservatism, and continuity of location, to that which has now been established for Cornwall.
Nicholas Johnson is the County Archaeologist in Cornwall
Fact file
Return to the British Archaeology homepage
© Council for British Archaeology, 1998