The Forth and Clyde estuaries determined the topographic location of the Antonine Wall. While there is an absence of data for the relative sea level (RSL) in the Roman period, this is generally considered to be broadly similar to current levels, perhaps with extensive mudflats (Tipping and Tisdall 2005, 444–6), though models of 1m height difference have been proposed elsewhere, such as in northern England (Bradley et al 2011). Late prehistoric shorelines are likely to have fallen (from about 15m to around 12m above sea level) before 2000 BC (Smith et al 2010). At present, sea level models are based on insufficient evidence, including limited data from specific locations. Existing models may be flawed as a result of sedimentation and erosion over the Holocene which has changed the form and structure of the seabed and land surfaces (Sturt et al 2013). Harbour sites in both estuaries are assumed, but not proven.
River systems and routes were likely not subject to much change in the mid-2nd century AD and the surrounding catchments were stable (Tipping and Tisdall 2005, 446–7). The navigability of rivers has not been studied, though it has been suggested that the Carron was navigable as far as Camelon (Tatton Brown 1980; Bailey 1992).
Soils contemporary with Antonine occupation are very well preserved in the Wall itself, which is a resource that has not been extensively exploited. Analysis of such soils can provide information on the natural environment, land use and land management prior to the building of the Wall as well as the sourcing of material for construction. The latter may cast light on why turves were seemingly used as cheeks, the vertical ends, in the Wall east of Watling Lodge rather than in the core as usual. In general, the Wall traversed some of the better soils suitable for arable agriculture in Scotland, particularly towards its eastern end (Bibby 1991), which may account for the relatively poor turf available for Wall building in this sector.
While it appears that much is known about land use around the Wall from pollen analysis, almost all of the data is from samples that come from uncertain locations with unresolved chronology (Tipping and Tisdall 2005). Both the pollen and macro-plant record provides evidence of woodland, which may have been managed rather than wild woods. (eg Hanson 1996). Despite some palynologists suggesting otherwise (eg Dumayne and Barber 1994), rapid and large-scale woodland clearance linked to the expansion and probable intensification of agriculture was a late Iron Age phenomenon (Tipping 1994; Ramsay and Dickson 1997). Thus, the establishment of the Antonine Wall took place within a well-established farmed landscape, in which crop-growing was probably important (Tipping and Tisdall 2005, 460). Defining how much and what types of cereals were grown is crucial for understanding how sustainable the food supplies were. Geoarchaeological analysis of soils from the Wall or the area around can tell us about both Roman resource management strategies and the use of the landscape in the pre-Roman Iron Age.
On-site sampling for pollen has taken place, predominantly from ditch fills or buried soil horizons such as those provided by rampart turves (eg Dickson and Dickson 2016). This approach tends to suffer from the general problem that bedevils the analysis of pollen from soils linked to differential pollen survival and uncertainty concerning the extent of the area from which it was derived. Because barley is consistently predominant in the macro-plant record from military sites, it has been suggested that assumptions about its use only for animal consumption should be reconsidered (Leslie et al 2007, 136–7 and 146–7). There is scope for more diverse approaches here, such as single seed isotope analysis.
Field systems are known in the immediate vicinity of several Wall forts (see Extramural activity) and contemporary sites nearby (eg Inveresk). These are generally not closely dated; most are assumed to be Roman, though some may have earlier origins.
Research Issues
- What were the sea levels during the Antonine Wall period? There is a need for more detailed relative sea level analyses for the Antonine period using archaeologically derived data.
- How navigable were estuaries and rivers in the Roman period? Can any harbours be identified? Wider hydrological issues should also be studied– particularly those relating to the identification of harbours.
- More use should be made of sections through the rampart to understand the local environment better, including soil development, nutrient loss, land capability and agricultural productivity. Any opportunity should be taken to employ a full range of geoarchaeological techniques, such as thin section, phytolith, dung spore, isotope and chemical analysis, on material from the Wall itself.
- Future pollen analysis should move the focus away from large-diameter peat bogs to arrays of small diameter peats which resolve landscapes at the human or the farm scale (Fyfe et al 2004), linked to systematic Bayesian refined “wiggle-matched” radiocarbon-dated chronologies which have the temporal resolution to define Antonine occupation and its environmental impacts in relation to landscape exploitation.
- Any opportunity to sample and analyse waterlogged deposits for environmental evidence on excavations should continue to be taken.
- Any opportunity to investigate field systems in the immediate vicinity of the Wall should be taken (see Extramural activity).