Our understanding of the nature and date range of the bank-and-ditch monuments – of various sizes and shapes – that have traditionally been described as ‘henges’ and ‘hengiform monuments’ has evolved considerably since Anthony Harding and Graham Lee compiled their Britain-wide corpus in 1987 (Harding and Lee 1987). Gibson provides an excellent history of the concept of the ‘henge’, and of the problems attached to the use of this term (Gibson 2012). Richard Bradley’s work on four such sites in north and north-east Scotland has confirmed that this class of monument was built at different times over a very long period (Bradley 2011). Indeed, it has long been argued that ‘henge’ is a problematic term and should be abandoned (Gibson 2012, 17), but so far no adequate replacement term has been coined. For the sake of convenience, it will be used here, in quotes, with due acknowledgement that it is a problematic term.
The earliest dated henge in Scotland – a single-entrance, roughly circular, Class I monument, enclosing a ring of standing stones at the Stones of Stenness in Orkney – has been radiocarbon-dated to the 30th century cal BC (as discussed in Sheridan 2024). The dating of a Food Vessel-associated skeleton in a grave sealed under the bank of a two entrance, Class II, henge at North Mains, Perth and Kinross to 1971-1626 cal BC indicates that this henge was constructed during the Early Bronze Age (95.1%, 3490±65 BP, GU-1381: Sheridan 2004). Excavations elsewhere in Britain, including at Broomend of Crichie (Bradley 2011,74), have confirmed that other Class II two-entrance henges are most likely to be of Early Bronze Age date. Meanwhile, Kenny Brophy’s excavation of single-entranced ‘henges’ at Forteviot, Perth and Kinross, confirmed that some ‘Class I’ henges were constructed during the Chalcolithic period – probably 25th to 23rd century cal BC in the case of Henge 1 (Brophy and Noble 2020, 147) – and also during the Early to Middle Bronze Age (Brophy and Noble 2020,155).
It is with this chronological caveat that the topic of potential Late Neolithic ‘henges’ in the SESARF research area is approached.
In their nationwide review, Harding and Lee cited just two sites as candidates for single-entrance henges within the SESARF area: Overhowden (Cat. No. 244) and Rachan Hill (Cat. No. 245), both Scottish Borders. A sub-circular enclosure at Aytonlaw, Scottish Borders (Cat. No. 243) is dismissed as being a defensive or domestic structure, and other sites at Kirkland, Newlands and Thurston Mains, East Lothian (Cat. Nos. 285-287) are similarly dismissed.
Attention had been drawn to Overhowden as a potential henge by Stuart and Margaret Piggott in the 1940s, and two small exploratory trenches were cut across the ditch and bank in 1950 (Atkinson 1950). The monument measures 70 to 76 metres internally and has an internal, rock-cut ditch around 4.57 metres wide and an external bank 9.14 metres wide, separated from the ditch by a berm up to 4.57 metres wide (Harding and Lee 1987, 343). It is located on sloping ground on the west side of the Lauder valley, at an altitude of nearly 300 metres above sea level. While he found no artefacts within the site, Atkinson remarked on the abundance of lithic finds from the vicinity – particularly petit tranchet-derivative flint arrowheads, edge-polished flint scrapers and a macehead. These, along with another nearby concentration of finds at Airhouse, were subsequently studied by Torben Ballin (2011). The Overhowden finds concentrated on a small rise only 500 feet (about 150 m) away from the ‘henge’ but a small trial excavation by Atkinson revealed no features. While no direct connection between these Middle-to-Late Neolithic artefacts and the Overhowden ‘henge’ can be made, their proximity attests to activity in the vicinity within the time-frame in question.
Atkinson also speculated whether Rachan Hill (which he described as ‘Rachan Slack, Broughton, Peeblesshire’) – spotted by Richard Feachem on an aerial photograph – might also be a ‘henge’. However, he remarked that it was too badly damaged by cultivation for a positive identification, and no clear evidence for a bank could be found. Like Overhowden, it is on relatively high ground, in a gently sloping depression. Harding and Lee commented: ‘Insufficient information available, but cannot be ruled out as henge-related.’ (Harding and Lee 1987, 345). The fact that Atkinson claimed to have found evidence suggesting a second causeway diametrically opposed to the first when he probed the site raises the possibility, however, that if this is a ‘henge’, it could be a Class II, two-entranced example.
Two other potential candidate sites, both in Scottish Borders, are listed on the HES ‘Trove’ website. The first is Marygold, although all that is written about this is that ‘The location, classification and period of this site have been reviewed.’ The second is Lewenshope, although this has more than one entrance: one on the west side and a larger break on the opposite side.