Whitton Cross Roads

Whitton Crossroads: Iron Age and Roman Farmstead.   by John Etherington
  • The Iron Age and Roman Farm at Whitton Crossroads, South Glamorgan (ST 081713) is a little to the south-east of the cross-roads, at the highest point of the limestone plateau. 
The farmstead was discovered during 1956-71 and trial excavations were conducted by the Inspectorate of Ancient Monuments. It was more fully excavated between 1965 and 1970 by the Department of Archaeology of the (then) University College of South Wales and Monmouthshire, Cardiff. 

The most prominent and probably first-constructed feature is an almost square enclosure, 60 m x 56 m inside a bank and ditch. A few shards of Iron Age pottery amongst Roman finds suggests that there may have been a pre-conquest enclosure on the site. The ditch was originally v-shape and a maximum of two metres deep, with a bank three to five metres wide, and formerly about two metres high. The enclosure is 0.33 ha (0.8 acre) in area. The bank has been eroded by age and ploughing, now being less than 0.2-0.3 m high, and the ditch infilled with eroded material. Deep ploughing at the time of excavation was further damaging the foundations of some buildings. 

The entrance was on the east side throughout the 300 years of the farm’s life. It comprised a gap in the bank and causeway across the ditch. There were stone kerbs on bank ends with various timber gates and towers at different times. A trackway with rubble footings and variously, limestone flag paving, mortar and tufa surfacing, or cobbles, entered the farmyard.

When excavation started, there was little to suggest that much information would emerge or that dateable pottery would establish a detailed evolutionary sequence of buildings spanning some three centuries. 

In total, seven or eight timber roundhouses were built and rebuilt during the century following AD 30. These were conical structures, probably thatched or shingled, on low stone walls and of 9.5 to 14+ m diameter. There is not much evidence of post-holes for roof support, the base wall bearing the weight. This is the reverse of usual practice at the time.  

Between AD 115 and 135 some square timber buildings were erected, and rectangular stone buildings first appeared around AD 135. The later stone buildings post-date all of the timber roundhouses. They were  built mainly of limestone but incorporate some tufa in the early walls. A few roofing slabs of Pennant sandstone (probably from area north of Cardiff) and a few roof-tiles were found, but thatch or shingling was probably more common (though leaving no remains). 

The site was never deserted or razed to the ground. Rebuilding took place by replacement of ageing or obsolete structures in a continuous process leading from an Iron Age village of roundhouses to a final, fairly sophisticated group of rectangular stone buildings roughly occupying the four corners of the enclosure. 

At some time in the late second or early third century a "central heating system" (an underfloor hypocaust chamber) was constructed but never fired. The hypocaust chamber in the western range of buildings was filled with refuse and that in the eastern range used for storage.  At the same time a "verandah" was built. 

By the end of the third century the bank was of little consequence and the ditch silted but they were probably marked by a hedge or fence and no structure apart from a grain drying kiln was outside them. 

Occupation ended around AD 340, probably by a gradual decline rather than any dramatic event. Two and a half centuries later, Cadoc would almost certainly have seen the ruins of the Roman stone structures which might well have still been occupied in a low-key way, perhaps used as animal housing? Early medieval settlements left very few archaeological clues so little evidence of post-Roman occupation would remain. 

Some finds deserve special mention, for example evidence of cereal growing appeared in querns and a large millstone though there were no actual remains of grain. However, transport was expensive in Roman times so the milling equipment is circumstantial evidence for on-site cereal agriculture.  

The East Range of buildings included a stone tank and T-shaped drying kiln suggesting, perhaps, malting and brewing? A second (grain?) drying kiln was found outside the enclosure on the south side of the entrance. Both kilns have already been mentioned in Newsletter 108 in relation to grain drying and malting. 

Livestock and dairy farming were undertaken as analysis of animal bones showed that many animals were older than beef cattle would have been. Cattle made up 35% of all bone. Other stock comprised sheep + goat (40%) and pig (16%). The presence of deer (red and roe) bones at just 3% suggests that some of the animal's preferred habitat of open woodland, was available within a reasonable distance, as it still is today. 

Eighteen coins were found during excavation, another three in 1977, plus one previous discovery. Mostly mid-2nd century except for some early silver. The last was from 337-41. Most of the pottery was coarse earthenware and the imported Samian ware was relatively scarce, mostly from central Gaul. A clay brick and a broken half, stamped BOV was speculatively associated with a place name, perhaps Cowbridge (Bovium)? 

Thirty-three more or less complete enamel bronze brooches, some harness ornaments rings and bracelets made the bulk of ornamental metal work. A few glass beads posed an enigma as they were very old when the Whitton farmstead was established, dating from the1st to 3rd century BC, some originating from mainland Europe. A Roman Antiques Roadshow?! 

Functional items included iron nails, chisels and other tools but a particularly interesting find was a steelyard of four main divisions each subdivided into 12ths, presumably uncia and siliqua(ounces and grains). Evidence of iron smelting took the form of slag and analysis showed no reason why the ore should not have come from the immediate source of the Taffs Well/Llanharry deposits. 

Little can be said about the interior of the stone buildings but a small amount of wall plaster was found, invariably painted in white, red, olive green and dark green in various combinations. Despite the attempt to move upmarket with a hypocaust system, the dwellings here had no mosaic floors, which were a feature of wealthier sites.

The Roman Environment

We have previously speculated about the extent to which the countryside of Cadoc's Vale could be reconstructed from the writings of the 12th century monks who transcribed and embroidered the Lives of the Saints. A little more information comes from a few archaeological sites such as the “Dark Ages” Dinas Powis settlement (Newsletter 72 1996).

During the Whitton excavation, animal bones, land snails and insects and plant remains were found. These allow an ecological reconstruction of the landscape and are certainly more accurate than our speculation about the 6th century Vale. There is one drawback: most of the preserved material was found in an abandoned well which had been filled with refuse, a bit like 
interpreting the modern countryside from remains in a garden compost-heap! 

Small mammal and frog bones suggest an environment with open grassland close to the farm and scrub-land and brambles at a slightly larger distance, much as it is today. Nearby grassland is indicated by the field vole, which has a foraging range of only 10 m or so and the scrubbramble by common shrew, bank vole and woodmouse. Frog and water vole tell us that there 
was slow moving or still water nearby, also as now. All of these animals can still be captured today in the vicinity, except perhaps for the water vole, which declined so catastrophically in the late 20th century. 

Several groups of plants suggest that the Roman countryside was similar to ours, before modern agriculture seriously reduced grassland diversity (mid-20th century). Grasses and herbs of hay-meadows formed one group, with agricultural plants (emmer wheat and barley) and arable weeds as another. Many plants, which can be found today in neglected gardens and overgrown farmyards, were common. A third group was from hedgerow, scrub or open woodland including ash, wild rose, blackberry, elder and the herbs hemp agrimony and wood forget-me-not. One interesting tree was wild plum or bullace. We have previously mentioned the occurrence of bullace in Llancarfan hedgerows today. 

A very specific group of animals were the snails, of which there were many species, all suggesting rotting organic material, hence the reference to a compost heap. Many insects were found, again some associated with rotting vegetation, dung and carrion. However, an open well also serves as a pitfall trap and other insects were found, deriving from wasteland habitats 
(overgrown farmyard) and others from open grassland. 

Comments