Map 8 Britannia Superior Compiled by A.S. Esmonde-Cleary, 1996 with the assistance of R. Warner (Ireland) Introduction Britain has a long tradition of antiquarian and archaeological investigation and recording of its Roman past, reaching back to figures such as Leland in the sixteenth century. In the eighteenth and nineteenth centuries the classically-educated aristocracy and gentry of a major imperial and military power naturally felt an affinity with the evidence for Rome’s presence in Britain. In the twentieth century, the development of archaeology as a discipline in its own right reinforced this interest in the Roman period, resulting in intense survey and excavation on Roman sites and commensurate work on artifacts and other remains. The cartographer is therefore spoiled for choice, and must determine the objectives of a map with care so as to know what to include and what to omit, and on what grounds. British archaeology already has a long tradition of systematization, sometimes based on regions as in the work of the Royal Commissions on (Ancient and) Historic Monuments for England (Scotland and Wales), but also on types of site or monument. Consequently, there are available compendia by Rivet (1979) on the ancient evidence for geography and toponymy; Wacher (1995) on the major towns; Burnham (1990) on the “small towns”; Margary (1973) on the roads that linked them; and Scott (1993) on villas. These works give a series of internally consistent catalogs of the major types of site. Maps of Roman Britain conventionally show the island with its modern coastline, but it is clear that there have been extensive changes since antiquity, and that the conventional approach risks understating the differences between the ancient and the modern. The east coast in particular has seen major erosional and depositional changes. The latter are the easier to assess. Recent work has shown that in the Roman period the Wash was a larger body of water than now, and that it was also surrounded by large areas of marsh and swamp, possibly subject to marine transgression (Waller 1994). In general these areas were drying out through the Roman period, perhaps in part because of the Car Dyke/Mid-Fen Dyke catch-water to the west (Simmons 1979). The map shows the best available estimates of coastline and the edge of dry land; the river courses through the area are even less certain. Smaller areas such as around the Humber estuary, the Waveney estuary (Wade-Martins 1994), the Wantsum Channel, Romney Marsh (Cunliffe 1980), the Pevensey Levels and the Somerset Levels have likewise been restored to the best current estimate of the ancient coastline. Some indication has been given, too, of the limited land reclamation along both sides of the inner part of the Severn estuary (Fulford 1994). Much more difficult to gauge is the extent of coastal erosion from Flamborough Head (the easternmost point of Map 9 G6) south along the coast of East Yorkshire, Lincolnshire and East Anglia. In the absence of any agreed reconstruction, it has only been possible to indicate that the modern coastline is not to be relied upon. Our evidence for the Roman names of peoples, places and features derives principally from Ptolemy, ItAnt and GeogRav (cf. Rivet 1979, chaps. II-V). Despite their widely-varying dates of composition, all three depend essentially on early imperial sources. Care has been taken to adhere to attested forms of names, so that those found only in Ptolemy are not latinized, and reconstructions of corrupt forms in GeogRav are treated with caution. Because of the wealth of known sites, there is particular difficulty in choosing what to mark on the map, and what to omit. In general, the aim is to present longer-lived sites, and thus to convey the overall impact of Rome on Britain. All sites with a Latin name and a known location are shown; these do include a few short-lived sites, but their number is too slight to affect the distribution patterns for classes of site as a whole. Otherwise, the difficulty of making choices is at its most acute in the marking of military sites. No “marching camps” (temporary campaign fortifications) are shown, nor are the semi-permanent garrison fortresses and forts which succeeded them as conquest turned to consolidation. What are shown in Wales, and in the north of England (Map 9), are the long-term garrison forts (re-)built in stone, which persisted, sometimes down to the end of the Roman period. The Late Roman “Forts of the Saxon Shore” on the southeast coast are also marked. Coloniae and civitas-capitals are shown. A selection of 112 MAP 8 BRITANNIA SUPERIOR the “small” towns (roadside settlements and market centers), is shown, omitting some of the less substantial. Nothing is marked as a “road-station,” since all known sites of this type coincide with towns. The main Roman roads on which the towns lay are indicated. Roads featured in ItAnt, or certain to have been long-distance routes, are marked as major; others of more local significance are left as minor. No attempt has been made to indicate the myriad local lanes and tracks. Rural settlement presents the most difficult choices. There is archaeological evidence for thousands of such settlements of the Roman period in Britain; the map cannot accommodate more than a fraction. The most characteristically “Roman” rural sites of the period are the villas, the Romanized farms. These could have been omitted altogether, but that would have been to ignore a major aspect of the Roman impact on Britain. So some 200 have been selected out of a total somewhere in the region of 1,000. The main criterion has been to select enough to show the overall distribution pattern, almost absent from the north and west, common, but unevenly spread, in the south and east. More specifically, preference has been given to villas which are prominent in the literature on Roman Britain, as well as to those which are less well known but were evidently of some substance, judging (for instance) by the presence of mosaic. Much the same criteria have been employed for rural temples. Omitted are the even more numerous “farmsteads” and “villages,” those isolated or nucleated settlements whose plan, structures and artifacts show much less Roman influence. Thus the likely majority of the rural populace of Roman Britain is absent from the map. Their omission stems in part from the scale, in part from the focus of this atlas upon Greek and Roman civilization and culture. Directory All place names are in the United Kingdom unless otherwise noted Abbreviations TIR BritSep Tabula Imperii Romani, Britannia Septentrionalis, London, 1987 TIR Condate Tabula Imperii Romani, Condate–Glevum–Londinium–Lutetia, London, 1983 Names Grid Name Period Modern Name / Location Reference F3 Abbotts Ann RL Scott 1993, 81 C3 Abercyfor RL Scott 1993, 57 E3 Abona fl. L Avon Rivet 1979, 239-40; TIR Condate 22 E3 ‘Abone’ RL Sea Mills TIR Condate 93; Ellis 1987 G1 Abos fl. R Ouse / Humber Rivet 1979, 240-41; TIR BritSep 44 E2 Acton Scott RL Scott 1993, 163 H3 Ad Ansam RL Higham Rivet 1979, 241 I4 Ad Lullia? FRA See Map 11 G1 Ad Pontem RL Thorpe-by-Newark Burnham 1990, 272-73 G2 Ailsworth RL Scott 1993, 31 D3 Alabum RL Llandovery Nash-Williams 1969, 95-96; TIR Condate 65 F2 Alauna? RL Alcester Burnham 1990, 92-97 F3 Alchester RL Burnham 1990, 97-103 G3 Alfoldean RL Smith 1987, 275-77 I3 Alresford RL Scott 1993, 60 G2 Ancaster RL Burnham 1990, 235-40 H4 Anderidos L Pevensey Maxfield 1989, 157-60 § Anderelium § Anderitos A1 ‘A(n)dros’ Ins. R Howth Head IRE TIR BritSep 42 G4 Angmering R Scott 1993, 181 MAP 8 BRITANNIA SUPERIOR 113 Grid Name Period Modern Name / Location Reference B4 Antiouestaion/ R/ Land’s End Rivet 1979, 252-53, 266-67; Bolerion Akron L TIR Condate 63 § Belerion HR G2 Apethorpe RL Scott 1993, 139 F1 Aquae *Arnemetiae RL Buxton Burnham 1990, 176-78 E3 Aquae Sulis RL Bath Burnham 1990, 165-76 H2 Arbury Road RL Scott 1993, 33-34 I4 Ardres FRA See Map 11 E3 Ariconium RL Weston-under-Penyard TIR Condate 105 H2 Ashill RL Scott 1993, 129 G2 Ashton RL Burnham 1990, 279-81 F3 Asthall RL TIR Condate 21; Smith 1987, 236-38 F3 Atrebates R S England Rivet 1979, 259-60; TIR Condate 21 Augusta = Londinium I4 [Augusta] Ambianorum FRA See Map 11 D3 *Aventius fl. L Ewenni Rivet 1979, 260-61; TIR Condate 49 G3 Baldock RL Burnham 1990, 281-88 G2 Bancroft RL Williams 1994 F2 Bannaventa RL Whilton Lodge Smith 1987, 222-23 G1 Bannovallum RL Horncastle Burnham 1990, 240-45 F3 Barnsley Park RL Scott 1993, 68 D3 Barry R Evans 1988 H2 Bartlow RL Scott 1993, 32-33 F3 Barton Court Farm RL Scott 1993, 161 E3 Barton Farm RL Scott 1993, 70 F2 Barton-in-Fabis RL Scott 1993, 153-54 G4 Batten Hanger RL Scott 1993, 192 G4 Beddingham RL Scott 1993, 58 G3 Beddington RL Scott 1993, 78 F3 Bedwyn RL Scott 1993, 202 I4 Béhen FRA See Map 11 E3 Belgae R S England Rivet 1979, 267; TIR Condate 25 E1 Belisama fl. R Ribble Rivet 1979, 267-68; TIR BritSep 65 G4 Bignor RL Scott 1993, 182 A2 Birgos? fl. R Barrow IRE TIR BritSep 5 E2 Bishopstone RL Scott 1993, 89 F4 Bitterne RL Cotton 1958; TIR Condate 27 E2 Blackwardine RL TIR BritSep 8 E3 Blestium RL Monmouth TIR Condate 72 Bolerion Akron = Antiouestaion Akron D3 ‘Bomio’ RL Cowbridge Burnham 1990 Bononia = Gesoriacum A1 Bououinda fl.