- Roman presence in South West Scotland was fleeting, with their military forces occupying rather than settling within the region. Roman material culture occurs in small quantities on indigenous sites, but larger quantities do turn up from time to time (for example, the Carlingwark Loch hoard). To date, no evidence has been found for re-organisation of the landscape on a scale which might suggest that areas were parcelled up into farms, then distributed amongst either favoured local elites or retired military personnel. Nor are there any other indications of ‘Romanisation’ amongst the local population, such as the replacement of monumental roundhouse structures by villas laid out in the Roman style.
- Roman settlement was entirely military in character. We know a great deal about the character of such sites, in part because the Romans left written accounts which explained all sorts of aspects, including, for example, layout, positioning, the character of defences and the number of men expected to be stationed there. Armed with this basic information, archaeologists have -for over a hundred years – been testing the validity of these accounts through excavations of Roman sites distributed all over their Empire, which extended from the north of Scotland to the Middle East, from the Rhineland to North Africa. Such detailed examinations continue to this day, as new sites are discovered and excavated, allowing us to make at least some basic premises about how the various types of Roman military ‘settlement’ would have been arranged and how their component buildings would have looked.
- Roman military sites vary in character. The most striking difference is surely between the more permanent military installations, the forts, and their temporary, more ephemeral, counterparts, the camps. ‘Permanent’ does not mean upstanding and occupied throughout the entire duration of every Roman military campaign in Scotland. Instead, forts were built by a particular legion or auxiliary regiment, then occupied by them for the duration of each campaign. The fort would then be dismantled and anything left behind destroyed to prevent useful items falling into enemy hands.
- Because forts were sited in important strategic locations, the sites could be reused during successive campaigns. Birrens, for example, sat on an important routeway heading north through the Nith valley. They were permanently garrisoned – ‘permanent’ within the context of a military campaign, which is – and often the same site was reused during successive military campaigns. Different architectural styles coupled with dateable material culture types allowed different phases of occupation and use to be identified and interpreted. Birrens, arguably the most extensively studied Roman fort within the region, revealed three phases of construction/occupation occupying virtually the same spot, though the internal area increased for the second and third phases. The first phase, with timber buildings (presumably combined with clay daub walls), has been dated to the earlier Flavian campaign, suggesting that the fort formed part of the wider Tyne-Solway frontier system as constructed during the reign of Hadrian. The second and third phases were stone-built, but the quality of the stonework varied. The second phase was better built than the third, perhaps indicating that the second phase was built by legionaries, the third by a detachment of auxiliaries.
- On some Antonine forts, the use of timber construction continues. At Carzield, for example, built during the Antonine period, the fort’s buildings use timber posts combined with daub and wattle walls and clay roof tiles. Window glass is often found on the larger, permanent sites, too. Roofs could be made of clay tiles, manufactured on site, or thatch. Carzield also differed from Birrens in that it was constructed on a brand new site, rather than re-using the site of the earlier Flavian fort located nearby at Dalswinton.
- The layout and internal arrangements of these forts remained largely consistent both across the empire and throughout all of the various military campaigns. The external defences defined a square or rectilinear space with rounded corners (the familiar so-called ‘playing card’ shape), with opposing gateways on two or four sides. Roads bisected the interior. Occupying a roughly central position was the principia. This housed accommodation and offices for senior officers (sometimes equipped with underfloor heating), the strongroom, where wages were stored, and the shrine, where the honours and standards, and a statue of the emperor, were held. Arranged around the principia were lines of barrack blocks, stables, granaries (with raised floors to assist air circulation), and ranks of cooking pits and ovens where the soldiers cooked their food, usually located around the margins of the site. There was also a communal latrine, often placed in a way where wastewater from the baths could be used to flush away sewage. The site of Birrens, which remains the most extensively excavated and studied of the region’s forts, shows all of these elements, though the presence of multiple enclosure ditches confounded early researchers.
- Larger forts like Birrens and Carzield formed important nodal points placed at important strategic locations in the landscape. Linking them was a network of roads, along which – interspersed at regular intervals – were the much smaller fortlets (like those at Beattock (Barnhill), Kirkland, Outerwards, Raeburnfoot and Sanquhar) and the even more diminutive watch towers like Craik Cross. Their size was dependent on the number of soldiers deployed there. They would have been always occupied, providing a constant presence within the surrounding countryside and ensuring safe passage for those using the road network. This means that even sites which now survive as little more than a visual trace on aerial photographs can be assessed as part of the wider network of Roman sites, as we have a good idea of how they would have been manned and how they would have functioned strategically.
- In addition to the more permanent infrastructure provided by the forts, fortlets and signal towers, there are a number of temporary installations known broadly as ‘camps.’ Like the forts, these have been extensively studied by researchers who include Rebecca Jones (Jones). Camps provided short-term accommodation, which may have lasted for weeks or even days. Some may even have been constructed to provide overnight rest for soldiers engaged in a long march between permanent stations, for example. Sometimes, the same sites were repeatedly chosen for placing these camps, resulting in a complex palimpsest of features. At Beattock (Barnhill) a total of four camps of Antonine date have been identified, the earliest of which overlies the site of a Flavian fortlet. Two phases were noted at Ellisland and Girvan Mains, with evidence for camps also identified at Annan Hill, Beattock (Bankend), Carronbridge, Drumlanrig, Innerfield, Ward Law and, most recently, Ayr Academy.
- The military personnel billeted in these camps stayed in leather tents, with ten men occupying a single tent. The construction of a substantial defensive ditch – square or rectilinear on plan, with the obligatory rounded corners – was always undertaken. Sometimes there are more elaborate defences around the entrance (see Conflict and Violence). Apart from these substantial ditches, the only features which tend to leave tangible traces of temporary camps are the ovens where soldiers cooked their food. These characteristic dumbbell-shaped pits can, on occasion, be the only indication that a temporary camp ever existed in a given location. In forts, ovens tend to be laid out in specific places, around the margins. It has been surmised that the situation would have been similar in the camps, with each unit of men pitching their tents and digging their cooking pits in the same place each time. A degree of organisation does appear to be evident at Dalswinton, where remote sensing has shown lines of pits close to the external ditch and also in the interior (Jones et al 2016). However, the excavators at Carronbridge suggested that the apparently random distribution of the cooking ovens across the interior of the camp might instead suggest that on occasions, such a degree of formality and regimentation may not be the case.
- The important role of the cooking pit in identifying Roman temporary camp sites is highlighted by recent excavations at Ayr Academy. Here, 26 such cooking pits were found on a site which had never previously been known for having any Roman military presence. The discovery was important in providing new insights into how far the Romans had penetrated into the area now occupied by the modern Ayrshires: previously, Roman sites had been known at Girvan Mains in the south and Loudon to the east, with an apparent absence throughout North Ayrshire and East Ayrshire, with sites identified further north and closer to the Clyde Valley.
- Although the camps leave scant traces in the modern landscape, enough information is left in the literature for us to have a solid understanding of how they functioned. Four categories of camp are recognised. These include marching camps, where troops pitched tents and rested overnight while travelling over long distances. Practice camps, where troops stayed while on training or manoeuvres, could have been occupied for a longer duration. So too could siege camps, where the troops found shelter while besieging an enemy stronghold (such as, potentially, Burnswark). Finally, there were construction camps, which would provide accommodation for troops while they were building more permanent structures such as the fort at Birrens. Traces of the original camp can sometimes survive as a trace on aerial photographs onto which the later fort is superimposed over much of its extent. According to Rebecca Jones (Jones 2006), most examples of temporary camps occurring in Britain would have been either marching camps raised by men moving between garrisons, or camps used by soldiers undertaking reconnaissance or policing an area or moving supplies. The different sizes of camps would be dictated by the number of men stationed there and would therefore reflect their different roles.
- Temporary camps appear empty and featureless today, but Jones reminds us that they would have been busy places. As well as the soldiers themselves, there would be pack animals, wagons, and horses. It is possible, too, that civilians would have been present. Accommodation would have been needed for camp followers. These might include traders, prostitutes and diviners, as well as unofficial family members attached to rank-and-file soldiers who were not allowed to marry whilst serving in the military. It has been suggested that the annexe, which can be found attached to some forts and camps, would have provided space to accommodate these hangers-on. Jones has, however, commented on the consistent size of these attached enclosures and suggests instead that they may house additional military personnel who arrived at a later date or who had a different role (Jones 2006).