°Carwyn Farm,203, Carwin Farm, Carwin, Phillack

 ROMAN (?)

Carwin in Phillack OS 25” 62.10, no. 4; NGR SW 584398; at about 160’ O.D. Dis-
covered by the writer in field-work, 1953 (see Dudley, 1954, no. 7: Fowler, 1962, 37,
no. 5). Surveyed by a team under Mr, P. J. Fowler in April, 1956. See fig. 12.

How this prominent and very remarkable site escaped notice until so recently is a
mystery. It is a rectangular camp, possibly a Roman military work, whose internal
dimensions were of the order of 700 ft. NE-SW and 350 ft. NW-SE. Until recently,
almost the entire northern (recte north-western) side, and much of the south-western
end, were intact. At the moment (the site having now been scheduled under the Ancient
Monuments legislation) the eastern half of the north-western side alone remains. This
is, from the internal ground level, up to nine feet high, and measures some 30 feet across
its base. The external glacis descends straight into a filled, but perceptible, ditch, from
which a modern stone-faced hedge rises to form the boundary of OS field no. 2.

The soil capping of the interior is only 6 in. thick, and covers clean blown sand.
Recent burials of deceased farm livestock in the bank have revealed that this, too,
consists entirely of blown sand without turf-lines. The field which now engulfs the
south-western half of the site was formed by levelling the south-western portion (some
250 ft.) of the surviving rampart and by depositing and mechanically spreading about
thousand tons of soil. The south-eastern (long side) is detectable over much of its length,
however, as a Slight linear depression (the ditch ?).

Linguistically this is an extremely interesting example of nomenclature, Carwin Is a
farm, and a very ancient tenement, in Phillack parish, situated some little distance
away at SW 582387, and separated from this camp by the modern farms known as
Treeve and Pulsack, Earlier forms are Caerwyn 1335, Kaerwen 1317, and Kaergwyn
in 1311. This represents O.Co. *caer guyn (or possibly the suspected fem. form guen,
since *caer, Ker, is a fem. noun) and can only mean ‘the white camp’. There is no other
round or earthwork of any description within at least a mile and certainly none which,
historically, could ever have stood on Carwin tenement. The name must then refer to
this particular fort, since such it seems to be, and is a safe example of the extension of
ker to a non-native work,

Despite the apparent impermanency of a work constructed entirely in sand (and it
must be recalled that the north-western rampart is still standing today to an impressive
height) a work of this size—S} acres internally—is large enough for a milteniary cohort

45