Re-use of Roman stone including a carving of a Pegasus at the Crichton souterrain ©RCAHMS
The idea of memory in conceptualising landscape is relevant in two areas of study in Roman Scotland. The first is the positioning of forts in relation to the existing and past landscape, both cultural and natural. This has seen no sustained study, and when comments are made, they tend to be focused on Roman forts supplanting indigenous settlements (e.g. Hanson 2004, 146). Yet the landscape had a deep past, with certain locations having a resonance and significance to the local populations. Was this reflected at all in the disposition of Roman forces, in deliberate attempts to appropriate part of this, or was it immaterial? Is the location of Newstead (Trimontium), for instance, linked solely to a strategic crossing of the Tweed or control of a local power centre, or did the visual (and arguably religious) significance of the three Eildons play a role too? Is the conjunction of Roman sites and Neolithic ritual monuments, for instance at Raeburnfoot and Fourmerkland, just a reflection of the desirable properties of flat gravel terraces near water courses?
The examples of Cramond, Inveresk and Newstead show the extent of Roman activity beyond the fort walls. At Newstead and to a lesser extent Inveresk some of this was known from cropmarks, but the extent of activity at Cramond is only known from excavation and survey. At all three sites significant new finds have been made in previously blank areas some distance from the fort. There is no reason to think these forts are unusual in having activity around them; they are simply better-known than most other Scottish sites. This emphasises the need to evaluate any development within c. 1km of a Roman fort, in order to find the little-understood landscapes surrounding it. Inveresk © RCAHMS, Cramond © RCAHMS © Headland Archaeology Ltd © Ordnance Survey, Newstead, © RCAHMS
Ideas of memory are also relevant in considering the legacy of Rome (see also section 6.5). How were fort sites perceived and used in later periods? Were they foci for activity, or appropriated to acquire power by association, or ignored? There has been no systematic treatment, but random examples indicate a variety of processes: evidence from Kintore that the area of the camp was avoided for several centuries (as taboo, or perhaps contaminated land?) contrasts with the burial tumuli adjacent to the ramparts of the fortress at Inchtuthil (Winlow and Cook 2010), or the churches and occasional castles set in other forts (Cook and Dunbar 2008, 354-6; see Maldonado 2015 for re-use of the Antonine Wall). Place names may provide another pointer to later views; many Roman forts were called 'fort' in either British (caer e.g. Cramond, Cadder, Carriden) or Gaelic (cathair e.g. Stracathro) (Watson 1926, 365-71), but the significance of this merits renewed attention. There was also reuse of Roman stone (discussed below, section 6.5) for powerful Early Historic fort sites as well as souterrains and burials, again indicating a perceived significance to the legacy of Rome.
Neither the Roman perceptions of the pre-existing landscape nor the later impact of Roman forts and camps have received any sustained attention; both are ripe for research.