4 Tracing the Circuit Wall of the Roman City

A series of field observations during the summer campaigns of 2001 and 2002 aimed to check the validity of a recently propounded city wall hypothesis (Note 5). At an early stage of this research, our fieldwork led to the conclusion that this 'desktop-made' working hypothesis was completely unrealistic. In general, it does not comply with the topographic reality of the field and it must therefore be completely abandoned.

Our proposal is to start working from the premise of a smaller city enclosure, a reduction of almost a quarter, leaving most of the western part of the earlier proposed plan out of the circuit wall. This higher western part is in fact a continuation of the NW-SE orientated hill Malhadais, and is not an isolated hilltop as might appear when seen from a position more downslope. Its gradient, on average between 11 and 17°, is much too steep to allow normal house building in Roman times. We also noted that the bedrock outcrops at surface level almost everywhere here and that its undisturbed state does not suggest ancient human interference for building purposes. An inclusion of part of this higher area within the wall circuit (as a kind of 'acropolis' or strategic lookout) is altogether possible, but this should then be done according to 'logical topographic arguments'.

Detailed observations of the topographic, geomorphologic and archaeological nature on that part of the landscape around which most archaeological findings have so far been located, lead us to propose here a new working hypothesis for further study of the urbanisation of Ammaia. This geoarchaeological fieldwork comprised a full assessment of present-day terracing and field boundaries, a general evaluation of the dispersal of archaeological debris (e.g. building materials, pottery) in the fields, and careful mapping of still visible positive or negative remains of the circuit wall itself, some of which were discovered during recent excavation work by our Portuguese colleagues. In our analysis we gave due consideration to the argument that the Roman town builders took great care to use the topography of the terrain realistically. When evaluating this, we approached the state of the terrain today with the eye of geomorphologists, who take account of the sometimes intricate erosion history of a hilly landscape during a life-span of almost two millennia.

Our working plan for the exact location of the ancient urban area proposes the existence of a quite regular, almost rectangular walled area, only somewhat elongated in a south-westerly direction with a small hilltop extension. The surface area of the walled Roman town structure of Ammaia would then approach some 20.8ha. The location of this relatively small urban area was certainly well chosen. Apart from its general situation in a suitable agricultural area with good provision of water and other natural resources (see section 2), the Roman topographers chose their site well. The town lay on a predominantly west–east orientated footslope, developed between the base of a low hill culminating at +607m and the valley of the Rio Sever. Both north and south of this footslope a small tributary stream joins the Sever from the west (Fig. 3). The geological substratum of the site is composed of Devonian clayey schists, a metamorphic rock with subparallel orientation of the micaceous minerals (Fig. 4). The weathering product of the schists gives rise to sandy clay soils. The greater part of the enclosed city area was situated between +520 and +560m, on a gently sloping terrain. Only the small elongated western segment of the city area (see above) was on a steep slope.

The most distinctive element of the organisation of the site was, of course, its wall circuit. Let us therefore have a closer look at its proposed trace, starting from the most regular north-eastern side and moving counter-clockwise (Fig. 13).


© Internet Archaeology/Author(s)
University of York legal statements | Terms and Conditions | Last updated: Mon Nov 21 2005