Megaliths and Post-Modernism: The Case of Wales
Fleming, Andrew, Antiquity
'Our work here is open ... to new interpretations ... since anyone can visit these stones and experience these places themselves, make new observations and check old ones:
Eleven years ago, Christopher Tilley published A phenomenology of landscape: places, paths and monuments (1994). It has become a much-cited book. Tilley took the archaeology of landscape in a new direction, presenting a mode of field observation designed to explore his ethnographically based, persuasive characterisation of Neolithic sacred geography. He presented three case studies, two of which concerned the megalithic chamber tombs of south-west and south-east Wales. He suggested that significant numbers of these monuments were designed to refer to prominent hills, rock outcrops and watercourses, thus apparently offering evidence-based insights into Neolithic cosmological perceptions. Five years later, I argued that Tilley's findings could not be regarded as sustainable contributions to Welsh Neolithic studies (Fleming 1999).
More recently, in Places of Special Virtue (2004), written with Alasdair Whittle, Vicki Cummings has adopted Tilley's approach--with equally problematic consequences. I feel that I must now expound my critique at greater length. I will deal mostly with south-west Wales, and will use the abbreviation 'TC' to refer to the Tilley--Cummings approach where appropriate, abbreviating 'Cummings and Whittle' to CW.
In south-west Wales, the TC approach mainly involves chiming that megalithic tombs deliberately 'referenced' rock outcrops located at various distances; a significant outcrop may have been immediately beside a tomb, a few hundred metres away, or on the summit of a distant hill. Occasionally links are claimed with springs or water courses. TC also discuss relationships between tombs and the sea. Cummings considers 33 sites in total, comprising a minimum of 40 monuments.
In cases where tombs are immediately adjacent to rock outcrops (24 per cent of the total, or 8 sites out of 33; Cummings & Whittle 2004: 29), it is hard to deny some kind of deliberate association. Whether this has to do with Neolithic cosmological beliefs is another issue; I will deal with these tombs later. In the case of linkages with more distant outcrops, however--and much of the TC thesis is based on these--the arguments are lacking in rigour. For reasons mostly to do with the tombs' state of preservation, the positions of entrances--or the directions which the tombs 'faced'--are frequently unclear. So TC cannot work with indicated 'alignments' or directions. But instead of regarding this as a good reason for using better quality data, in another research area, they have chosen instead to argue that a rocky outcrop or hill may be significantly associated with a tomb if it is simply visible from the site (although significance is also claimed when a target is invisible from a site, as in the case of the outcrop to the west-north-west of Carreg Samson; pace both Tilley 1994:99 and Cummings & Whittle 2004: 56). At Garn Turne (Cummings & Whittle 2004: 148), the invisibility of the Preselis is claimed to be deliberately contrived. TC have chosen to do their fieldwork in an area where the location of many tombs on hillsides (Cummings & Whittle 2004: 37, 87) virtually guarantees wide vistas of striking hills, rock outcrops and sometimes dramatic coastlines. It comes as no surprise that rock outcrops may be seen on the skyline from 12 sites, that is 36 per cent of them, or that 61 per cent of the sites, 20 in total, have a view of outcrops (Cummings & Whittle 2004: 88). (But 13 tombs out of the 33 are not related to outcrops at all; 2004: 29.) Tilley writes for the most part as if it is unnecessary to demonstrate that the claimed associations with distant outcrops are more than coincidental, whilst Cummings makes little more than gestures in this direction. …