The narrow neck of the Forth-Clyde isthmus followed by the Wall is an obvious potential Roman frontier location. According to the Roman historian Tacitus (writing around AD 98), “the isthmus was now firmly held by garrisons (praesidia) during the Flavian conquest of Scotland, although not necessarily as a ‘frontier’ (Agricola 23). The area was used for a temporary halt during Agricola’s fourth campaign, which was primarily one of consolidation and fort building. (Hanson 1991, 107). The reasons for this short-lived hiatus in that advance have been much debated, but may relate to the changes of emperor, with Titus succeeding Vespasian in AD 79, only to be followed by Domitian two years later.
Unfortunately, there is very little supporting structural evidence for a pre-Antonine frontier. Leaving aside forts beyond the eastern and western ends of the Wall at Elginhaugh and Barochan Hill respectively, the most obvious installation is the fort at Camelon (Maxfield 1980). Full publication of the excavations undertaken in the 1970s is still awaited, augmented by more recent work (Hunter 2012, 285; Kilpatrick 2016). The strategic importance of the site, however, seems to relate more to operations north of the isthmus, as confirmed by its re-use in the Antonine period and, importantly, its location north of the Wall. The only other confirmed installation on the isthmus is the fortlet or small fort at Mollins which, including the width of the ramparts, has an area of 0.4 ha. An aerial photographic discovery, its Flavian date was indicated on the basis of very limited ceramic evidence from small-scale excavations in the 1970s (Hanson and Maxwell 1980).
There is, however, a long tradition going back to 18th-century accounts of earlier, potentially Flavian, use of Antonine Wall sites. This hypothesis has been most extensively developed by Macdonald (1934, 267–73 and 466–8) and has been widely accepted thereafter. The most credible structural elements were the earlier enclosures that Macdonald recorded beneath the forts at Croy Hill and Bar Hill, but these were shown by subsequent excavation to be later in date (Hanson 2022 a; 8-16; Keppie 1985, 51–8).
Various other sites along the Wall have produced a few Flavian finds from early excavations such as Old Kilpatrick, Balmuildy, Cadder, Kirkintilloch, Castlecary and Mumrills, but without any associated structural evidence (Hanson 1980). However, none of these sites have sufficient dating evidence which would support Flavian occupation, as the recent analysis of coin finds has confirmed (Brickstock 2020).
If there were a series of Flavian deployments across the isthmus, with forts or fortlets at regular intervals, further installations would be expected. There is, however, no reason for them necessarily to coincide with other Antonine Wall installations, as the criteria for the location of a continuous linear barrier were not necessarily the same as those which determined the positioning of individual military sites. Recent work on the Stanegate, the Roman road linking Carlisle and Corbridge to the north of Hadrians Wall, suggests that Flavian sites were found at river crossings rather than on the later line of Hadrian’s Wall, and it is worth exploring a similar scenario for the area of the Antonine Wall.
- Review the dating evidence from those Antonine Wall forts where Flavian occupation has been postulated.
- Seek further military sites across the isthmus and at suitable river crossings, particularly through aerial reconnaissance and/or geophysical survey.
- Geoarchaeological work at any identifiable pre-Antonine sites in the region might provide some insight into the long-term developments in the landscape, including the Roman military impact and engagement with it.