Site Report

WoSASPIN 5161
NS10NE 3 196 072.

There is a local tradition that near to Turnberry Castle (NS 196 072) there was a town of the same name, of which no vestige now survives.
W Macfarlane 1907

Since the construction of a wartime airfield and Turnberry golf course the original undulating duneland has been transformed, and no trace of possible Medieval occupation is visible in the vicinity of the castle. From local information it was established that within living memory a farm and a group of cottages existed in area NS 197 071, but this is likely to have been 18/19th century occupation of which no trace now exists.
Visited by OS (JRL) 31 May 1977

NS 1970 0715
It is likely that the reported local information of a farm and cottages recorded by the previous authority referred to the former Warren Farm (see WoSAPIN 42955). It is possible that there was an early castletoun associated with Turnberry Castle (see WoSASPIN 5159) on land beyond its earthwork defences, but no reliable evidence of its existence has been discovered. From comparison with other castles of the period, it is also possible that servants' or tradesmens' buildings were present on the modern lighthouse headland site within the outer curtain wall of the castle, separate from a main lordly residence on the rocky outcrops to the E, and that this may have been the origins of the traditional belief in a vil or small settlement associated with the castle.
Entered WoSAS (HMcB) 22/06/2015

NS 1968 0713
Limited evaluation trenching inside and to the S of the former lighthouse walled garden in 2015-16 (see Event 6564) revealed no clear evidence of medieval settlement there, although undated localised anthropogenic deposits were found within the stabilised wind-blown lenses making up the sandy soils.

While a farm or small settlement is shown at or near the site of the 19th C farm of Turnberry (see WoSASPIN 69437) on both the Gordon manuscript map of Cuningham (sic) of ca. 1636-52 and on Blaeu's 1654 Atlas of Scotland map of the N part of Carrick which was produced in part from the lost manuscript map of Timothy Pont (ca. 1590-1600), no occupation is depicted at or near the castle site on Turnberry Head. The open headland is named "Turnbery head" by Gordon, while Blaeu's map uses the modern spelling of Turnberry and also appears to depict the castle with a conventional symbol representing ruinous towers.
General Roy's Military Survey of Scotland (1747-55) depicts settlement at Turnberry in a location on or close to the modern farm of the same name, but not close to the castle, which is the only structure shown within the bounds of the putative park (see WoSASPIN 67402).
Notwithstanding the above observation regarding structures in the possible park in the mid-18th century, a farmstead called "Clachanton" (see WoSASPIN 69438) is shown on the Ordnance Survey 1st Edition map of the area at NS 20790 07360, some 670m NW of Turnberry, and in a location similar to that depicted for a settlement called "Clachan of Turnberry" on Blaeu's 1654 Atlas of Scotland map referred to above.
Given the shifting pattern of settlement in and around the policies of the mediaeval castle after it fell into disuse, it is impossible to predict with any certainty which if any of the sites mentioned may have been the origin of the local tradition relating to a settlement near the castle. Given the observed incidence of wind-blown sand deposits on and around the headland, it is possible that settlement there at the time of the castle's occupation became unsustainable after the lordly residence fell out of use, and that any residents moved to one of more of the later settlement locations, or between them, as conditions allowed.
Entered WoSAS (HMcB) 06/01/2020