It is evident from other parts of the Roman Empire (Derks 1991, 1998), that the environment around Roman forts contained what were considered sacred places, both in terms of the natural landscape (for example springs and bogs) and built monuments, such as temples, shrines and cemeteries. In Scotland, whilst there is little evidence of a mixing of Roman military and indigenous Iron Age religious practices, possibly due to the relatively short and sporadic nature of the Roman occupation, there are indications of Roman religious practice.
Altars and inscriptions found in Scotland have revealed a range of deities and cults – both official (of the Emperor and Jupiter) and those brought by soldiers from their homelands (Keppie and Arnold 1984; RIB I; Hanson and Maxwell 1986, 182–86, 191–22). In south east Scotland, two altars were found at Cramond, one dedicated to Jupiter and the other, erected by a cohort of Tungrians, was dedicated to the Altervan Mothers (Rae et al. 1974). The carving on Eagle Rock, near Cramond is thought to date to the occupation of that fort, although there is a suggestion that the severely eroded sculpture is of Mercury in his horned helmet, rather than an eagle (Ross 1967).
A probable Mithraic altar, discovered outside the fort at Inveresk provides the most northerly example of a building relating to the cult of Mithras in Europe (Hunter et al. 2016). A stone built, apsidal-ended building, thought to be a monument dating to the end of the 2nd century AD may have existed at Easter Langlee. Unfortunately, the foundations were destroyed during gravel quarrying in 1965 and no archaeological monitoring was undertaken. However, investigations of the stones identified several inscriptions, described as being typical of Roman work (Steer 1966). However, more recent discussion has suggested the monument may have been an early medieval structure made of re-used Roman stone from the nearby Newstead fort (Ewart 2011). This merits further investigation, including re-examination of the surviving stones.
The discovery of an oversized pinecone finial from a roadside at Inveresk may demonstrate the existence of a large tomb or shrine there. Further investigation and geophysical assessment along the routes heading out of forts may yield more examples. The Cramond lion was discovered at the mouth of the River Almond near the fort (Hunter and Collard 1997). A tombstone of the cavalryman Crescens, was found at Carberry (Hunter and Keppie 2008), also a distance from the nearest fort and possibly close to the Roman road which ran south from Inveresk. The traditional focus of research on fort sites rather than their hinterlands may have contributed to the paucity of evidence in south east Scotland for Roman temples and burials. The need for wider landscape surveys, including geophysics, beyond the boundaries of settlements is very relevant here.
The deposition of some metalwork hoards in southern Scotland has often been identified as a form of votive offering, although by whom is less clear (Hunter 1997) and might be linked to the possible foundation deposit at Elginhaugh (Bateson and Hanson 1997; see also Woolliscroft 2017). Perhaps the most well-known and large-scale deposition in south east Scotland are the pits and wells at the Roman fort of Trimontium at Newstead which contained a huge range of material. While this has been analysed in several publications (e.g. Ross and Feachem 1976; Clarke and Jones 1996), some of the approaches now seem dated and more recent approaches to structured deposition may be helpful. Re-examination of the material excavated from these pits to ascertain its condition when it was buried could also be informative.
Moving further away from formal religion, it is also worth exploring issues around landscape knowledge and ideas of place. While the actual Roman occupation was intermittent, engagement with the landscape has a longer history and should include the influence of ways in which the existing Iron Age communities already saw their landscapes. The positioning of Roman sites, and often their repeated use over extended periods, reflected deliberate choices. While strategic considerations would certainly have been be critical, possible wider meanings of visually striking locations such as the Eildons may also be relevant. Certainly, the repeated use of the Newstead area for campaign camps suggests that the location had a firm place in Roman mental (and physical) maps of south east Scotland.