Roman road at Whitton Edge in the Scottish Borders. The irregular quarry pits are often the best surviving indicators of a road line. © RCAHMS
Roman roads had a lasting impact on the infrastructure of Britain, but the last systematic study of the Roman road network of the whole of Britain was undertaken by Ivan D Margary (1955; 3rd edition 1973). The Ordnance Survey Archaeology Branch produced detailed files of all known and speculated Roman roads from the 1950s - 1970s; these files are now held by RCAHMS and the information on them has been digitised but is not currently available through Canmore.
Since these studies, a number of surveys have looked at the road network in various areas of Scotland (eg. Allan Wilson's work in Dumfries and Galloway - Wilson 1989; 1999; and that by Frank Newall and William Lonie in various parts of southern Scotland, regularly reported in Discovery and Excavation in Scotland; see also Lonie 2004). A limited amount of work has also been undertaken on the Military Way, the Roman road running to the rear of the Antonine Wall. Clarke (in press) shows the potential complexity and variety in the road system over time around key fort sites in his analysis of the roads around Newstead.
North of the isthmus, the Roman road is known running from Camelon, with a break in the Forth valley at Stirling, to Ardoch, Strageath and Bertha on the River Tay. This was a well-engineered highway and significant lengths remain in use; it has been subject to recent survey by the Roman Gask project (e.g. Woolliscroft and Davies 2002; Woolliscroft 2005). The line is accompanied by a chain of fortifications and towers and is regarded as part of a Roman frontier (Woolliscroft and Hoffmann 2006) or supply line (Dobat 2009). The road has produced no dating evidence, but is usually assumed to be contemporary with its forts, which were built in the later 1st century. This may now be less certain, however, as the road has been argued to post-date a camp at Innerpeffray West (see above), although there was no dating evidence to confirm that the relevant road surface was Roman in date.
There are antiquarian accounts of a route continuing the known road beyond the Tay to Strathmore, and recent air photographic traces may hint at corroboration (Woolliscroft, pers comm). There are also stub roads through the gates of a more northerly line of forts from near Loch Lomond to Strathmore, but no roads are known linking them to each other or the known line. No road is known to the legionary fortress of Inchtuthil, the key base of the entire 1st century occupation, although one is known from the fortress to the quarry and camp at Steeds Stalls, Gourdie, to the north. It is possible that future work may yield traces of a network of light track-ways to serve these sites.
In 2009, John Poulter published a monograph looking at the planning of Roman military structures in northern Britain, including a stretch of Dere Street from the Vale of York to Newstead (see also Poulter 2010). His fieldwork methodology has produced interesting results relating to the planning and layout of Roman roads and linear frontiers, and could be applied to other Roman roads in Scotland. It provides insights into the direction and process of planning, the views which were sought from linear frontiers, and thus something of the motives behind them.
A systematic overview of the road network in Scotland, considering all lines claimed as Roman, from aerial and ground survey and excavation evidence, is required. This should also consider the application of Poulter's methodology to other Roman roads in Scotland and the study of the post-Roman history and influence on known roads, and of any pre-Roman antecedents. Such an assessment would allow targeted aerial and field survey then to attempt to fill gaps.
The Gask Ridge road requires further assessment - can the current known road be dated? If it is 2nd century in date, there is still likely to have been a road between the 1st century fortifications and, as these have their entrances oriented on the known line, this may prove to overlie a Flavian predecessor. The fact that this has gone undetected suggests it was little more than a path that would have been easily ploughed out, raising the possibility that other early roads existed that have not been discovered because they were never upgraded.