sign in sign up
<<
Home , Aballava, Aesica, Alauna (Maryport), Anderitum, Arbeia, Bannaventa

Proc Soc Antiq Scot, 128 (1998), 425-440

Recreatin possiblga e RomaFlaviaf o p nnma Britain with a detailed map for Alastair Strang*

ABSTRACT Since the rediscovery of Ptolemy's Geographia, many versions of a map of have been produced from datathe contains.it Former reconstructions, however, takenot accountdid the of variations in longitude scale which are discerned from Ptolemy's records. Two different scales had, in fact, been used by Ptolemy, in addition to other distortions, to accommodate Britain in a world which believedhe one-sixthbe to smaller than actual. reversingBy Ptolemy's procedures thatso his known places registered with their identified locations, possible,was it also, bringto unknownhis places into their optimum relative positions scalablea on map. becameIt clear during this analysis that Ptolemy must have extracted datahis from authoritativean rathermap than from disparate lists or itineraries and it is likely that this would have been of military origin and from the Flavian occupation period. possibleA arrangement beenhas deduced,such hypotheticalfor a map a and Flavian constructed.map This includes of all Ptolemy's data together with some obliged thatwas he or chose omit someto and recorded information periodthe for either priorcontemporaryor to with the Flavian period.

INTRODUCTION Despite being subjected to a small amount of corruption over the centuries, the data of Ptolemy's Geographia, for Britain, still provides a remarkable representation of the spatial relationship of the places recorded. Tha providin, is t recognize gw e latitud longitudd ean e scales appropriato t e a map generated from the data and are aware of individual inherent distortions that such a map displays evidens i t I . t from previously published reconstruction f Ptolemy'o ss wa thap e h tma s force emploo dt yvarieta distortionf y o Britis e th f o h creatin p n Islessi ma s . Recognitioghi a f no vita lGeographiae cluth n ei data (whic RichmonA hI d (1922, 288 soughd )ha t without success), namely Ptolemy's alignment ofEboracum, Isurium and directly north of , allowed Strang (199 19974& o comprehensivelt ) y presen n improvea t d map-projectior fo n Ptolemy's configuration of Britain and Ireland. This clue led to an understanding that Ptolemy had actually employed two different longitude scales in compiling his map and the resultant, improved map-projection is shown here in illus 1. Durin gseriea analysef so s (see Strang 1994, 1997, 199 forthcomingd 8an t becami ) e clear that Ptolemy had been manipulating data which was remarkably precise about spatial relationships and which could have been assembled only by persons with a good knowledge of

1* 9 Troutbeck Crescent, Bramcote, Beeston, NottinghaP 3B 9 mNG 426 | SOCIETY OF ANTIQUARIES OF SCOTLAND, 1998

Britain and Ireland. In the first century AD, there was only one agency capable of generating this quality of data for Britain and that was the Roman military who could co-ordinate legionary,

nava traded an l r informatio distancef no bearingd san s between these places. Becaus lace th k f 1 eo of awarenes f longitudo s latitudd ean e referencing, this extensive bod f spatiayo l information would only have been of military use if collated as a map (as opposed to being set out as numerous separate items of information or as itinerary representations) and this we know was ultimately conveye Alexandria.o dt 2 resolutioe Th probleme th f no s associated with Ptolemy's Roman Britain, allowin gdireca t comparison of Ptolemy's data with a modern, scalable map of Britain and Ireland, provided some clues for the projection and scalar-grid parameters to which a pre-Geographia map might have been drawn suitabla : e map-gri rectangulaa n do r projection with convenient horizontad an l vertical likelscalea d ysan referenc epurpose originth s i f thit I eo . s these papeus eo t rclue n si suggesting a hypothetical reconstruction. The data incorporated in such a map would most likely have been pre-Trajanic and to have originate Flaviae th n di n perio occupationf do . Unfortunately, Ptolemy concentrate placen do s defining coastal configuratio less swa interested nan interion di r chorography, other thaa r nfo few places require indicato dt dispositioe eth f tribano l territories orden I . produco t r morea e comprehensiv s beee ha Flavia nt i necessar p nma supplemeno yt e Geographiath t lisf placeo t s with others which Ptolemy may have chosen to omit or which were forced out by map distortions other and havnot s estil whicbeeof lmay nh he aware . This beelishas t n drawn from various complementary literary sources. Other known Flavian sitethosd san e whose Roman namee sar not yet known are included. Roman sites of known names, but without (apparently) Flavian- period historical or archaeological evidence are listed separately together with reasons for their exclusion from the reconstructe showe b t nye thay d ma tmap t somi s thesa ,f eo earlie d eha r use, In addition to this map of Britain a more detailed map of Scotland has been presented.

SOME CLUES CONCERNIN GPOSSIBLA E FLAVIA BRITAIF O P NNMA We know of no Flavian military map of Roman Britain existing today, but much of the information may have been available for example to Marinus of Tyre, prior to Ptolemy's Geographia. From our understanding of the distortions which Ptolemy introduced, it has been possibl reverso et procedures ehi thencd san speculato et e abou possibla t e representatioa f no sourc havthay p eetma ma bee contemporariess nhi r availablo m hi archaeologiste o eth t o T . f o , course, it is the locations of places on the map that are of critical importance and the recent analysis has made it possible to insert these with somewhat more confidence than before. Whilst analysin d synthesizinan g g Ptolemy' f Romao p nma s Britain a particularl, y intriguing point was the situation concerning two river mouths (Nabarus = Naver and Tamarus = Tamar) on the far north coast of Scotland and the south coast of , respectively. These specifio tw c locations, highlighte illun di , seeme1 s havo dt e escaped imposed displacements, although Ptolemy places to east and west of each appear to have been subjected to such adjustments. Now, on a modern map of Britain it is noted that their respective longitudes west of Greenwic e 4°14 4°10'd har an ' , indicating that thes (ano etw d probably well within Flavian tolerances) were regarded as directly north/south of each other at about 4°12', information most probably the result of Roman nautical determination. This then raised the question as to whether these locations could have defined some reference line from a previous map which Ptolemy wished retainee b o t d withi mucs nhi h distorted representation (illus Britaif 1)o Irelandd nan . STRANG: RECREATING A POSSIBLE FLAVIAN MAP OF ROMAN BRITAIN 427

27 28 29 30 31 64 (EANUS HYPERBOREUS

1/kTURACTONIUMQ OCAIAOUM K300UNUM

OCEANUS GERMANIC US

OCEANUS VERQIONUS WIROCCNIUM I OBRANNOGENIUM

PTOLEMrS GREAT BRITAIN AND IRELAND

SCALES LATITUDE =62.5 Roman Mile degrer spe e LONGITUDE ENGLAND = 41.67 Roman Mile degrer spe e SCOTLAND and IRELAND = 25.8 Roman MBes per degree

12 17 18 18 20 25 28 LONGITUDE (degrees) ILLU S1 Ptolemy' Romaf o p snma Britai Ireland nan d (after Strang 1994, 300) moderBritaiof The Irelannnand map presentedis illudin showins2, g this Tamar/Naver base line, which seems to conveniently bisect the east/west extent of the map. Clues as to the overall extent of the map come from the consideration of the two longitude grid scales employed in Ptolemy's map (illus 1): 25.8 Rm [Roman miles] per degree for Scotland and Ireland and 41.67 Rm per degree for England (Strang 1997, 20). The larger scale of the southern sectioe birectangulath f no r grid produce ratia s f 41.6o 7 -H 62. = 0.665 7 whics hi virtually identical to the value for Ptolemy's map of , giving England compatibility of longitude scale with this contiguous map to the south. 428 I SOCIETY OF ANTIQUARIES OF SCOTLAND, 1998

so

ILLUS 2 Modern map of Britain with rivers Tamar/Naver baseline STRANG: RECREATING A POSSIBLE FLAVIAN MAP OF ROMAN BRITAIN 429

valur degree pe Th 25.f eo m e8 R longitud e will coursef o , true b ,particulae e on onl r yfo r latitude fino T .d this latitude Ptolemy'r fo , s small-sized world necessars i t i , firso yt t fin cosins dit e by dividing the figure 25.8 by the constant latitude scale (62.5 Rm per degree) which he used. This yields a value of 0.413 giving a latitude angle of 65.62° north. The Ptolemy latitude of London is 54.02° so that the parallel of latitude that we are seeking identifo t y wouldsmals hi n l o ,world , have been (65.62 —54.02 Roma5 x 62.72 ) 5= n mileo st the north of London. On the full-sized world, for which there are 75 Rm per degree, this equates to 9.67° north of London or a true latitude of 51.42 (the actual latitude of London) + 9.67 = 61.09° north, just sufficient to encompass the Shetland Islands (Strang 1994, 95). Has Ptolemy therefore chosen his northern (rectangular) scale to suit the furthest extremity of such a map representing Britain wit hsimilara , erect north/south orientation?

A POSSIBLE FLAVIAN SCALE localizeA d militar woulp yma concernee db d with relative distanc directiod ean n (bearingf o ) locations and would not, at this early stage of cartographic development, be concerned with longitude/latitud worlr eo d context. However voyagea se , neighbourino st g territorie wels sa s a l land journeys woul recordee db regarded dan importants da . Henc erectangulara , uniform grid orientate cardinae th n di l directions woul adequate db conveniena d ean t t(a scalp sucr ema fo ha A4 size) might be 50 Roman miles (= 400 stades) per grid-space. The latitude scale, aligned with 61.09° nort modera n ho n map, could therefore progress southward 1n s8i interval Roma0 5 f so n miles eacbasa o ht e lin t 49.09ea ° north (require fullo dt y cove soute rth h coas England)f o t n O . an equivalent Ptolemy latitude scale (shown in Table 1) this progression would reach the southern limi Ptolemy'f o t Britaif o p t aboun sa ma Ptolem e nort° th 51 tn ho y scale ranga , somf eo 5 e92 Roman miles. For convenient practical comparison between Ptolemy and modern maps (see Strang 1994, 76 & 272), registration between distance and latitude (angular) scales occurs at (725 Roman south)milee th o st . This might well indicate tha grita d Romascal5 2 f eo n miles could have been used and also that, if significance attaches to London's location, the north/ south grid scale could possibly have originated there s showa , Tably nb (colum1 e . Thin4) s arrangement has therefore been adopted in illus 3, supplemented both by modern latitude scales and those of Ptolemy's smaller world. Similarly, longitude gridlines could progress outwards fro hypotheticae mth l (Naver/Tamar ) Roma25 r t intervala (o n )p 0 centrelin5 ma f o se th f eo miles to 300 Roman miles east and west. Then, using such a convenient and uniform scalar grid, Ptolemy's and other contemporary data can be incorporated to yield a plausible configuration for a possible Flavian map of Roman Britain latee th rf o ,firs t century.

GEOGRAPHIC INFORMATIO COMPLETE TH D NAN E FLAVIAP NMA Befor investigatioe eth n described earlier undertakes wa n (Strang 199 19974& ) ther beed eha n satisfactoro n y explanatio 'turnine th f n o f Scotland g o majorit e th d f Ptolemy'an 'yo s poleisn i Scotland remain unidentified. The geographic information included in the Flavian map (illus 3) s beeha n derived fro e Geographiamth located an d accordin thao gt t analysis, described dan supplemented, where possible, from other sources as explained below. 430 SOCIET ANTIQUARIEF YO SCOTLANDF SO , 1998

TABLE 1 Comparison betwee moderPtoleme e nth th d nan y longitude distanced san s

LATITUDE FLAVIAN MAP PTOLEMP YMA degrees Distance- Rom m'l's Scale Longitude ratio Actual Ptolemy re-map n/sL'nd'n Roman miles 'cosine (Lat. angle 61.00 9 65.62 725 25.8 0.413 60.75 2 6 65.22 700 60.59 65.02 37.5 687.5 SHETLAND centre 60.40 5 2 64.82 675 60.05 7 9 64.42 650 59.76 64.02 100 625 59.42 63.62 125 600 59.09 63.22 150 575 58.92 63.02 162.5 562.5 THULE media 58.76 63.22 175 550 58.42 62.42 200 525 58.09 62.02 225 500 57.76 61.62 250 475 NORTHERN SCALE 57.42 61.22 275 450 57.09 60.82 300 425 56.76 60.42 325 400 56.42 60.02 350 375 1 56.09 59.62 375 350 1 55.76 59.22 400 325 55.42 58.82 425 300 1 1 55.09 58.42 450 275 1 1 54.76 58.02 475 250 1 54.42 57.62 500 225 1 54.09 57.22 525 200 __ 1 53.76 56.82 550 175 53.42 56.42 575 150 (traditional Ptolemy ratio) 0.550 53.09 56.02 600 125 52.76 55.62 625 100 SOUTHERN SCALE 52.42 55.22 650 75 52.09 54.82 675 50 51.76 54.42 700 25 51.42 54.02 725 0 LONDINIUM 51.09 53.62 750 -25 50.76 53.22 775 -50 50.42 52.82 800 -75 50.09 52.42 825 -100 49.76 52.02 850 -125 49.42 51.62 875 -150 49.09 51.22 900 -175 50.82 925 -200 50.42 50.02 49.62 49.22 48.82 48.42 48.22 41.67 0.667

PTOLEMY'S GEOGRAPHY The majority of Ptolemy place-names are considered (Rivet & Smith 1979, Introduction) to be Romano-British, generall Celtif yo c derivation with some qualificatione ar onlw d fe an ,y a though wholle b o t t y Latin Romane Th . s also appea preferenc a hav o t rd eha strategicallr efo y locating their sites in proximity to established native .sites, often themselves of military or STRANG: RECREATING A POSSIBLE FLAVIAN MAP OF ROMAN BRITAIN | 431

defensive significance, and assuming names descriptive of these native sites. Ptolemy's poleis in Britai f Romao l appeae al nb no t rmilitar y significance (base n originao d l forr camo t p installations) wherea same sth e canno sai e Irelandr b t dfo , although defended native settlements may be appropriate. As in the Strang analysis (1997, Post-script), Votadini, when incorporated in the larger map-grid, is situated some 25 Roman miles south-east ofBremenium (High Rochester) and must certainly correspond wit earle hth y Roma Housesd n Re sit t a e , (Hind 1980, 165-71) largee . Th Ptolemy'f ro s longitude grids, therefore, incorporates all poleisVotadinie th f o tribe and the Boderiae aest (the Firth of Forth). As these poleis are associated with Ptolemy's map-scale for England, rather than with the rest of Scotland, are we to conclude (from this selective integration of data) that the boundary of the Votadini tribe (together with that of the , encompassing Birrens) somehow defined some early Flavian frontier s RiveA ?s ha t argued (1982, 322) 'Rome occupied tribal states rather than chunks of land'. Ptolemy uses mainly the Coritani title for this tribe but Coritavi is chosen to be more in keeping with evidence frocenturieo mtw s later which indicates Corieltauvi tribae th s la name (Tomlin 1983, 353-5). considee Ifw r legionar ywese fort th Britainf to n si , each take rivename e sth th f r eo tha t ti stands on: Isca (), Isca () and Deva (). Why should those in the east and north not do likewise? () stands on the river Ouse, whose estuary is described by Ptolem Aflov,s yt a Rive Smitbu d difficulttan d hha y reconciling this name other than just wite hth Celti 'river'r c fo roo - t.ab Could this hav facn ei t been transcribed from Ebou, providin rivee gth r name associated with (the later) Eboracum? norte th hn I also, what more appropriate name than Tava ( spelling Tauni) coul givee db Inchtuthilo nt , whic virtualls hwa y surroundee th y db unlese (i untirive d evidency y san ran lTa e should indicate otherwise). Ptolemy was obliged to omit the following two features. Tinea flumen (the Tyne) could not be included beside Vedra flumen (the Wear) in the angle between England and Scotland during Ptolemy's reinsertedrotatioe b w Scotlandf no n o n . ca Also t bu , , Cap omittee eb Wrato t d hha by Ptolemy to prevent its being interposed between Or cades insulae (Orkneys) and Orcas prom. (Duncansb configuratiow yne Heads hi n i ) f Scotlando n . However, this cape must have been such an important feature in Flavian navigation that it demands a title and reinsertion. In the bees ha nt i labelle p ma (Wrath)s da prom.

ARCHAEOLOGICA HISTORICAD LAN L INFORMATION Convenient list f pre-Flaviao s n fortfortressed an s d Flaviaan s n military site Britain i s e nar provided by Jones & Mattingly (1990, Tables 4.4,4.5 and maps 4.23,4.31) and Frere (1987, 56-7, 88-9 similarl) ,96 y lists military Julio-Claudiae siteth f so Flaviad nan n periods. These neee b o dt considered in conjunction with names of other Roman sites occupied in Britain during this period in assessing whethe have rw e authoritative evidenc able b ascrib o et o e t e name theso st e locations. Currently, only those with possible names have been inserted in the Flavian map. n ScotlandI , t includeno s Ptolemy wa b d y possibly becaus s sizeit f .o e According to Robertson (1975, 278), 'there is no evidence of any larger Flavian fort there, of the massive regular character witnessed, for example at Mailing, Bochastle, Fendoch, Cardean and Stracathro (E. Scotland)' (Robertson underlined those with a suspected Ptolemy name). In England, in addition to Ptolemaic and the few other foregoing Roman sites there are some 50 named pre-Flavian fortd fortressean s d militaran s y site f Flaviao s n date liste Joney b d & s Mattingly (1990). These have also been incorporated in illus 3. On a larger scale map further 432 I SOCIETY OF ANTIQUARIES OF SCOTLAND, 1998

ad occidentttn <——--- (millia pasium) -———> orienttmmi C CL . I t t'C'C 0 C C C 'CCCl. iL CCL CCC

W ILLU S3 Hypothetical Flaviaf Romao p nnma Britain STRANG: RECREATIN POSSIBLGA E FLAVIA ROMAF O P NMA BRITAIN 433

ad occidentem <———(millui passum)______> ad orientem C L 0 L C

59

58

57

56

I B ,

II

ILLUS 4 Hypothetical Flavian map of Roman Scotland 434 SOCIET ANTIQUARIEF YO SCOTLANDF SO , 1998

development would be possible. It is likely, for instance, that any known Roman river names would themselves have been established fro earln ma y datshould ean includede db . p Sucma ha coul made db e even more comprehensiv Flaviaf ei nr unname sitefa o s roadsg d(e ) were b o et added.

TABLE 2 Historical Additions. Source Feature Deduction Strabo64BC-Ao21 + Voyag stade 0 Gauo et 32 sl= = 40 Rm sailed between ebb tide in the evening and Sam hourr pe ,m sailinR 4 e i g speed approximateln i , y 10 hours. Diodorous Siculus fl 30 BC Beleriu day4 . mEuropp o s t e (300m R 05 stades37 = hour )pe distancr m R 4 t ea , AD 23-79 Oceanus Gallicus lies between Oceannus Britannicus and the Pyrenees Silina insulae identified as the Scilly Isles, Rivet & Smith 1979,457 - 9 Albion crossing from Morini shore (Gesoriacum/Boulogne)= 50 Rm Thyle ins. 6 days N of Britai (450m R 00 n st'des57 = ) N WCapf o e Wrath reaches Iceland Mela Pomponiu0 4 D A l sf Oceanus Atlanticuf o sW Europ Orcade0 3 g ee s Tacitus c AD 56-115 + Trisantona flumen identifie rivee th rs d a Trent , Rive Smitt& h 19798 47 , Cos'y c AD 650 Tinea fluraen identified as the , Rivet & Smith 1979,473 Anava flumen identifie rivee th s rd a Annan , Rivet, Brit 1982 1 1332 , t ItinerarAn Aycen6 c D2- t Blatobulgium identifie Birrenss da , Rive Smit& t h 1979, 268-9

Tabl (Appendix)liste3 Ptoleme th l sal y named Roman places include Flaviae th n di n map. Table 4 presents additional Roman place-names identified in England and , and cites by who r whermo e recorded, with actua r possiblo l e modern appropriatn namea d san e sourcr efo this identification. Liste Tabl n dseverae i ar e5 l more Romano-British names, know Britainr nfo , with reason they yswh canno includee b t illun di . Howevers3 , som f theseo e location alsy osma have had these names in Flavian or earlier periods. Complementary information that it is believed to have also been available to the Flavian military is contained in Table 2.

SPECULATIVE FLAVIAN MAP OF ROMAN SCOTLAND In Scotland a considerable number of Roman places were named by Ptolemy (with suggested identitie Strang)y b s . Ther mane ear y more sites, though, which show evidenc a Flavia f o e n presence but which have no Romano-British names associated with them. If these are combined, a more comprehensive Flavian map of Roman Scotland results (illus 4). For Scotland, Table 6 lists site knowf so r suspecteno d Flavian presenc t whosebu e Roma knownt nye namt r no Fo . s ei north-ease th Scotlandf to , possible Flavian camps (identifie faro ds ) have been include botn di h Romao n Tabld s A an n. e 6 p fort ma st beee havth ye ns ea discovere thin di s par f Scotlano t d thes e showe associateb b t e o campnye t y dma s with Ptolemy's Devana, Tuesis d Pinnataan .

ACKNOWLEDGEMENTS Gratitud expresses ei Maro dt k Hassal beinr fo lfirse suggesgo th t t concepe th t'Agricola n a f o t n map of Roman Britain' based on my analytical work on Ptolemy's Geographia of Roman Britain. doublm a I y gratefu r Davifo l d Shipstone's invaluable assistanc n presentationo e y An . STRANG: RECREATIN POSSIBLGA E FLAVIA ROMAF O P NNMA BRITAI5 43 | N

deficiencies, omissions and dubious or controversial statements are entirely my own responsibility. I trust tha resule th t t goes som towardy ewa preliminarsa y model from which stimulatiny b , g welcome improvement, Mark's vision migh tachievede somb y eda .

NOTES 1 Rivet & Smith (1979, 25), Frere (1980, 419) and Jones & Mattingly (1990, 18) also regard Ptolemy's geographical sources for Britain as primarily military. 2 Geographia would then result fro Agathodaimon?)n m(a fitting dat Ptolemy'o at s small world.

APPENDIX: TABLE6 S3-

SELECT BIBLIOGRAPHY FOR TABLES 3-6 Breez Breeze= e J 198, D Northern e 2Th Frontiers of Roman Britain. London. Bri .= t Societ Promotior yfo Romaf no n Studies. Crawford = Crawford, O G S 1949 Topography of Roman Scotland. Cambridge. Daniel Danielss= 198C , 6 'A Roman Cam Foths'e th t pa , Pop Archaeol, April (1986), 10-12. Frer Freree= 198S S , 7 Britannia. BCA. London. Hanso Hansonn= 198S W , 7 Agricola Conqueste th d an North.e oth f London. Hector = Hector, W 1880 Vanduara. Paisley. Howard = Howard,? 1969 Fort on 's Wall. Huddersfield. J & M = Jones & Mattingly 1991. Johnstone = Johnstone, D E 1977 The . CBA Res Rep 18. London. Maxwell 89 = Maxwell, G S 1989 The Romans in Scotland. Edinburgh. Maxwell 90 = Maxwell, G S 1990 A Battle Lost. Edinburgh. Nash W = Nash-Williams, V E 1954 The Roman Frontier in Wales. Cardiff. O.S. = Ordnance Survey 1991 Roman Britain Map. Southampton. Potte Potter= r 197W ,T 9 Romans North-Westn i England. . Richmond = Richmond 1956. Ri Taylox& = RixrTaylor & ,B 198S , 8 'Excavation t Bannaventaa s , 1970-1', Britannia , (1988)19 , 299-339. Rive= S Smitt& & R h 1979. Webste Webster= r 198G , 1 Rome Against Caratacus: 48-58.D A London. Wilso Wilsonn= 198A J Roman Guidee A 8R , th o t Remains of Britain. London. TABLE 3 Identification of places (in Geographia order) for Ptolemy's Britain & Ireland Ptolemy Preferred Location Ptolemy Preferred Ptolemy Preferred Localion Plolemy Preferred LO Place name Identification Type Place name Identification Type Place name Identification Type Place name Idenlificalion Type 8 BRITANNIA IRELAND& BRITAIN Islands ALBION ENGL'ND* Island VIRVEDRUM DUNCANSBY H'D Promontory EPIACUM Whilley Caslle? Fortif'd sile 5 SCOTL'ND H? IVERNIA IRELAND Island DUECALEDONIUS THEMINCHES Ocean VERUBIUM NOSSHEAD Promontory VINNOVIUM BINCHESTER Fortif'd sile HYPERBOREAN NORTH ATLANTIC Ocean NOVANTARUM RHINNSof Peninsula ILA HELMSDALE River mouth CATURACTONIUM CATTERICK Forlif 'd sile o GALLOWAY Tl BOREUM BLOODY HEAD Promontory NOVANTARUM MULL of Promontory RIPA ALTA HillofNigg High Bank CALAGUM BURROW- Fortif'd site > GALLOWAY LONSDALE z VENNICNIUM MALINHEAD Promontory RERIGONIUS LOCH RYAN Bay VARAR BEAULY FIRTH Estuary ISURIUM ALDBOROUGH Fortif'd sile H VIDUA FOYLE River mouth VINDOGARA IRVINE BAY Bay LOXA Findhorn?/Lossie? River moulh RIGODUNUM near Lancasler? Fortif'd site /O ARGITA BANN River mouth CLOTAE CLYDE Estuary TUESIS SPEY Esluary OLICANA Elslack Forlif'd sile Cl RHOBOGDIUM FAIR HEAD Promontory LEMANNONIUS Loch Fyne Bay CAELIS DEVERON River mouth EBORACUM YORK Fortif'd site ^. OCCIDENTALS To west of Ireland Ocean EPIDIUM MULLofKINTYRE Promontory TAEXALORUM KINNAIRD HEAD Promontory CAMUNLODUNUM SLACK Fortif'd site ^0 RAVIUS LONGUS Erne River mouth Firth of Lome River moulh DEVA E DE River moulh PETUAR1A BROUGH on Fortif'd site hrt Cfl MAGNATA Carmore City ITIS in Soun Sleef do t River moulh TAVA FIRTH ofTAY Esluary MEDIOLANIUM WHITCHURCH Fortif'd site LIBNIUS Moy River mouth VOLAS Loch Broom Bay TINA EDEN River moulh BRANNOGENIUM LEINTWARDINE Fortif'd sile y AUSOBA Clare River mouth (WRATH) (CAPE WRATH) Promontory BODERIA FIRTH of FORTH Esluary DEVA CHESTER Forlif'd sile C/3 SENUS SHANNON River mouth NABARUS NAVER River mouth ALAUNUS ALN River moulh VIROCONIUM Forlif'd sile DUR Maine River mouth TARVEDUM/ORCAS DUNNETHEAD Promontory (TINEA) (TYNE) River moulh LINDUM LINCOLN Fortif'd site o IERNUS Kenmare River mouth ABRAVANNUS Luce Water River mouth VEDRA WEAR River moulh RATE Fortif'd site NOTIUM Dursey Head Promontory IENUS Wigtowy nBa Estuary DUNUM TEESMOUTH Bay SALINAE near Skegness Fortif'd site L~ * VERGIONIUS To south of Ireland Ocean DEVA DEE River mouth GABRANTOVICUM Bridlingto y nBa Bay/h'rbour UROLANIUM STALBANS Fortif'd site DABRONA Lee River mouth NOVIUS NITH River mouth OCELUS SPURN HEAD Promontory VENTAICENORUM CAISTER SI Fortif'd site z EDMUND jo BIRGUS BARROW River mouth ITUNAE SOLWAY FIRTH Estuary A BUS HUMBER River moulh CAMUDOLANUM Fortif 'd site >_i SACRUM CARNSORE POINT Promontory MORICAMBE Wavermouth Bay Estuary METARIS THE WASH Estuary LUENTINUM Dolaucolhi/Pumsaint Fortif'd sile VO ^0 HIBERN1CUS IRISH SEA Ocean SETANTIORUM Morecamby eBa Harbour GARIENNUS YARE River mouth MARIDUNUM Forlif'd site MADONNUS Avoca River mouth BELISIMA RIBBLE Estuary Prominentia litoris Orfor projectioA d Nesn s BULLAEUM Fortif'd site MANAPIA Rathdrum Fortified site SETEIA MERSEY Esluary EIDUMANIS Blackwaler River moulh CORINIUM Fortif'd site OBOCA Uffey River mouth GANGANORUM BRAIC PWLHY L Promontory IAMESA THAMES Esluary CALEVA SILCHESTER Fortif'd site EBLANA Tara Fortified site TOESOBIS PRYSOR River mculh LUCOPIBIA Galehous Fleef eo t Forlified sile LONDINIUM LONDON Fortif'd site BUVINDAS BOYNE River mouth STUCCIA Ystwyth River mcuth RERIGONIUM Stranraer(s'theasl) Forlified sile DARVERNUM Fortif'd sile ISAMNIUM Cooley Point Promontory TUEROBIS Teifi River mcuth CARBANTORIGUM Raeburnfool Forlified sile RUTUPIAE RICHBOROUGH Fortif'd sile VINDERIS Dundruy mba River mouth OCTAPITARUM ST. DAVID'S HEAD Promontory UXELLUM Upper Tcviotdale Forlified sile NEOMAGUS Forlif'd sile LOGIA LAGAN River mouth TOBIUS TWYI River mculh COR DA Crawford Fortified site ISCHALIS Charterhouse Fortif'd site REGIA EMAIN MACHA Fortified site RATOSTHENES USK River mcuth NEWSTEAD Fortified site AQUAE CALIDAE BATH Fortif'd sile RHAEBA 'by Loch Ree, Athlone' Fortified site SABRINA SEVERN Estuary COLANICA Fortified sile VENTA BELGAR'M Eorlif 'd sile LABF.RUS Dun Ailin near Kildare Fortified site UXELLA Parrel Estuary VINDOGARA near Irvine Forlified sile DUNIUM Hod Hill Forlif'd site MACOLICUM MagHacha Fortified site HERCULIS HARTLAND POINT Promontory Slralhendrick Forlified sile VOLIBA near Liskeard Fortif'd sile REGIA altera MagT. Cunga Fortified site Antivestaeum.Bolerium LAND'S END Promontory Ardoch Forlified sile UXELLA near Barnstapie Forlif'd site DUNUM Gallion Du n Rathgall Fortified site Damnonium/Ocrinum LIZARE TH D Promontory LINDUM Mailing Forlified sile TAMARA near Tavistock Fortif'd sile HIBERNIS Cork Fortified site BRITANNICUS ENGLISH Ocean VICTORIA Fendoch Forlified sile ISCA EXETER Fortif'd site CHANNEL AEBUDA ISLAY Island CENIO KENWYN River mouth CURIA H'sd Re c Corbridge Forlified sile SCETIS SKYE Island item AEBUDA JURA Island TAMARUS TAMAR River mouth ALAUNA Learehild Forlified sile DUMNA LEWIS Island RHICINA RATHLIN Island ISCA EXE River mouth HIGH ROCHESTER Forlified sile ORCADES THE ORKNEYS Islands MALAEUS MULL Island ALAUNUS AXE River mouth BANNATIA CARDEAN Fortified sile THULE SHETLAND Island EPIDIUM ARRAN Island MAGNUS SOLENT Harbour TAMEIA STRACATHRO Fortified sile TOLIATIS Foulness? Island MONAOEDA MAN Island TRISANTONA ROTHER River mouth PINNATA CASTRA Thomshill? Forlified sile COUNNUS West ? Island MONA ANGLESEY Island NOVUS DOVER Harbour TUESIS Aberlour? Fortified sile VECTIS WIGHT Island ADRU deserla HOWTH Island CANTIUM South FORELAND Promontory ORREA Carpow Forlified sile LIMNU deserta LAMBAY Island GERMANICUS NORTA HSE Ocean DEVANA ? Forlified sile STRANG: RECREATING A POSSIBLE FLAVIAN MAP OF ROMAN BRITAIN 437

TABLE 4 Additional, significant, named pre-Flavian/Flavian sites/location Englann si Waled& s Site/Location Identity Appropriate Source Latin Name Modern Name Latin Name Date/Occupation Site type Sym. Nash W 54,77 Nas 54,7hW 7 FORT • Galava Ambleside R&S 79, 365 Wilso8 25 , n88 FORT • Vindocladia Badbury 0 50 , 79 S R& Brit, vii, 282 Min'r Set't O Combretorium Baylam House R&S 79, 313 J&M91.88 prob. FORT Q Clausentum? Bitterne/S'ton water R&S 79, 308 Frere 87, 276 prob. FORT Q Lavatris Bowes 4 38 , 79 S R& Wilson 88, 246 FORT Brancaster R&S 79, 274 Johnst'n 77, 28a FORT Cicucium? Brecon Gaer ( ) Nash W 54, 63 Wilson 88, 200 FORT Virosidum Brough by Bainbridge R&S 79, 506 J&M91.99 FORT Navio Brough on Noe R&S 79, 423 J&M91,99 FORT Verteris Brough u' Stainmore 79S , 49R& 6 Wilso0 25 , n88 FORT Brougham Wilson 88, 250 poss. 2nd c. FORT Aqu' Arnemetiae Buxton R&S 79, 254 Frer8 8 , e87 site C Canovium Caerhun Nash W 54,148 Wilso1 21 , n88 FORT Isca Caerleon Nas 54,hW 1 8 Wilson 88, 190 LEG. FORT • Caernarvon Nas 54hW , 28 Wilson 88, 212 FORT 79S , 49R& 3 Frer6 5 , e87 prob. FORT r_ Duroliponte Cambridge R&S 79, 351 Frere 87, 56 prob. FORC T Luguvallium 79S , 40R& 2 Wilso2 31 , n88 FORT Carvoran 79S , 40R& 7 Breeze 82, 68 FORT Lagentium Castleford 3 38 , 79 S R& J&M91.99 FORT Margidunum Castlehill R&S 79, 41 3 Frere 87, 56 FORT Caesaromagus 79S , 28R& 7 J&M91,88 FORT Chesterholm 79S , 50R& 2 Hanson87, 162 FORT Concangis Chester le Street R&S 79, 314 Brit 28, 1xv , FORT Danum R&S 79, 329 J&M91.99 FORT Dorchester 5 34 , 79 S R& Wilso0 7 , n88 prob. FORT L Salinae? Droitwich R'vennal06/31 J&M91,99 FORT Magiomnium Dropshot farm R&S 79, 406 J&M91,88 FORT Vindomora Ebchester 2 50 , 79 S R& Wilson 88, 245 FORT Levobrinta? Forden Gaer Nash W 54,48 Nash W 54,48 FORT Glo'cester/Kingshol9 36 , m79 S R& Wilson 88, 110 LEG. FORT • Durovigutum Godmanchester4 35 , 79 S R& J&M91,88 FORT Venonis High Cross 79S , 49R& 2 Brit, i, 184 prob. FORT C Lindinis R&S 79, 392 Brit, vii, 357 Maj.C Set't Verbeia? Ilkley 79S , 49R& 3 Wilson 88, 238 FORT Canonium Kelvedon R&S 79, 297 J&M91,88 Maj.C Set't Magnis Kenchester? 79S , 40R& 7 J&M91,99 prob. FORC T Bravoniacum Kirby Thore R&S 79, 276 J&M91.99 FORT Derventia Littlechester R&S 79, 334 Brit, vi, 244 LEG FORT • Alabum? Nas 54,6hW 7 Nash W 54,67 FORT Llanio Nas 54,4hW 0 Nash W 54,40 FORT Leucarum Loughor Nash W 54, 89 Nas 54hW , 89 FORT R&S 79, 387 2nd c. Wilson Harbour Derventio Malton 4 33 , 79 S R& Wilson 88, 237 FORT Manuessedum Mancetter R&S 79, 412 J&M91,88 VEX FORT Manucium 79S , 41R& 0 Wilso4 39 , n88 FORT Alauna R&S 79, 245 Wilson 88, 254 prob.fortlet Ardotalia? Melandra 79S , 25R& 7 Wilson 88, 165 FORT Nidum Neath Nash W 54, 91 Wilson 88, 203 FORT Condate North wich 79S , 31R& 6 Webster 8 1,1 02 FORT Voreda Old Penrith R&S 79, 508 J&M91.99 FORT Derventio ? R&S 79, 334 J&M91,99 FORT R&S 79, 446 Wilso5 3 , n88 FORTLET Uxacona Redhill R&S 79, 482 J&M91,88 FORT Bremetenacum Ribchester 79S , 27R& 7 Wilson 88, 239 FORT Dorobrivae Rochester 8 34 , 79 S R& Frer6 5 , e87 prob. FORT Q Abona Sea Mills 79S , 24R& 0 Wilso8 11 , n88 prob. FORT Q Vagniacis Springhead 79S , 48R& 5 1 Brit19 , ii , Min'rO Set't 438 SOCIETY OF ANTIQUARIES OF SCOTLAND, 1998

Pontes Staines? R&S 79, 441 J&M91.88 Maj. Set't O Ad Pontem Thorpe by N'ark R&S 79, 241 J&M91.88 FORT • Bovium Tilston? O.S.p 91ma , J&M91.99 Min'r Set't O R&S 79, 383 Frere 87, 56 prob. FORT Q Letocetum Wall 8 38 , 79 S R& Wilson 88, 181 VEX FORT T Alauna Watercrook 79S , 24R& 4 Potter 79, 358 FORT • Durobrivae Water Newton 8 34 , 79 S R& Wilson 88, 179 FORT • Whilton Lodge R&S 79, 265 Rix&Taylo7 33 , r88 Min'r Set't 0

TABLE 5 Roman named site Britain si n excluded from Flaviap nma Latin name Modern name Reference Exclusion Bibra Beckfoot R&S 79, 268 Bradwell Brit, xxiv, 302 centd 3r . Sulloniacis Brockley Hill R&S 79, 463 Garrianum Burgh Wilson 88, 155 c. AD 275 Magis Burrow Walls Brit, viii, 179 prob. 2nd cent. Caves Inn Brit, iv, 288 2nd cent. Noviomagus Crayford 79S , 42R& 8 Durocobrivis Dunstable . Brit17 , ,xi Mediobogdum Hardknott Wilson 88, 255 Hadrianic Bannovallum Horncastle Wilso6 18 , n88 third cent. AD Camboritum Lackford R&S 79, 163 Longovicium Lanchester Wilso4 24 , n88 Segelocum Littleborough R&S 79, 453 Salinae Middlewich Brit, vii1 ,32 Gabrosentum Moresby Frere 87, 120 after AD 128 Maglone Carlisled Ol ? 79S , 40R& 7 Sorviodunum Old Sarum 79S , 46R& 1 Wilson 88, 41 Sax.Shore centh 4t , Glannoventa Potter 79, 359 early Hadrianic Verlucio Sandy Lane R&S 79, 494 Villa Faustini Scole? R&S 79, 163 Portus Ardaoni? Portchester Brit, xxv, 31 3rd/4th cent. Causennis Saltersford Brit, xxv9 26 , 2n centd+ . R&S 79, 289 Tanatis Thanet R&S 79, 468 Pennocruccium Water Eaton R&S 79, 437 Westo ' Penyarnu d Brit, xxv, 271 HADRIAN'S WALL REGION Wilson 88, 277 Hadrianic Hadrianic Pons Aelii Newcastle Hadrianic Hadrianic Vindobala Rudchester Hadrianic Onnum Halton Hadrianic Cilurnum Chesters Hadrianic Brocolitia late Hadrianic Vercovicium Housesteads Hadrianic Great Chesters Hadrianic Birdoswald Howar9 d6 Trajanic sig.st'n Camboglanna Castlesteads Hadrianic Uxelodunum Hadrianic Hadrianic Concavata Hadrianic? Maia Bowness Hadrianic C'stra Explor'torum Netherby Hadrianic Fanum Cocidi Hadrianic Risingham Wilso5 31 , n88 prob. 2nd cent. SCOTLAND Veluniate Carriden Richmond 56/7 prob. Antonine Vanduara Paisley Hector 1880, xiii fictitious? STRANG: RECREATIN GPOSSIBLA E FLAVIA ROMAF O P NMA BRITAIN 439

TABLE 6 Un-named, probable Flavian site Roman si n Scotland Site MODERN RELEVANT SITE SITE SITE SITE No NAME SOURCE TYPE PERIOD STATUS SIZE 1 BLAKEHOPE Wilson 88, 316 fort Agricolan uncertain uncertain 2 Maxwel9 8 l fortlet Agricolan small 3 CAPPUCK Maxwell 89 fort Agricolan small 4 OXTON Maxwell 89 fortlet Agricolan possible large 5 ELGINHAUGH Maxwell 90 fort Agricolan standard 6 OAKWOOD Maxwel9 8 l fort Agricolan standard 7 EAST'R HAP'REW Maxwell 89 fort Agricolan standard 8 CASTLE GREG Maxwell 89 fortlet Agricolan small 9 BROOMHOLM Maxwell 89 fort Agricolan standard 10 TASSIEHOLM Maxwell 89 fort Agricolan standard 1 1 WARD LAW Maxwel9 8 l fort Agricolan possible standard 12 GLENLOCHAR Maxwell 89 fort Agricolan standard 13 DALSWINTON Maxwell 89 fort Agricolan large 14 DRUMLANRIG Maxwell 89 fort Agricolan small 15 LOUDONHILL Maxwell 89 fort Agricolan standard 16 CASTLEDYKES Maxwell 89 fort Agricolan large 17 BANKHEAD Maxwel9 8 l fortlet Agricolan small CRAMOND 2 40 Bri , 27 t fort Flavian? uncertain &R'berts'n 19831 ,42 18 Maxwell 90 fort Agricolan standard 19 Maxwell 90 fort Agricolan standard 20 MOLLINS Maxwell 90 fortlet Agricolan large Hanson87, 101 Agricola? possible 21 BAROCHAN Maxwell 90 fort Agricolan standard 22 DUMBARTON Maxwell 89 fort Agricolan possible standard 23 DRUMQUHASSLE Maxwel0 9 l fort Agricolan standard 24 DOUNE Maxwell 90 fort Agricolan large 25 GLENBANK Maxwel0 l9 fortlet Flavian small 26 KAIMS CASTLE Maxwel0 9 l fortlet Flavian small 27 STRAGEATH Maxwell 90 fort Agricolan standard 28 CASK RIDGE Maxwel0 l9 towers Flavian small(17) 29 BERTHA Maxwel0 l9 fort Agricolan possible large 30 CARGILL MAINS Maxwel0 9 l fortlet Flavian large 31 CARGILL Maxwel0 9 l fort Flavian standard 32 BOCHASTLE Maxwell 90 fort Agricolan standard 33 DALGINROSS Maxwel0 9 l fort Agricolan standard 34 INCHTUTHIL Maxwell 90 fortress Agricolan legionary 35 INVERQUHARITY Maxwel0 l9 fortlet Flavian large Temp Camp Marchins— g Inverquharitf etco N . y SStracathr= o type hectares a DUN Maxwel0 9 l Camp Flavian possible c. 3 b FINAVON Maxwel0 9 l Camp Flavian possible c. 15 c STRACATHRO Maxwell 90 Camp Agricolan possible Sc. 16 d Maxwell 90 Camp Agricolan possible c. 38 e NORMANDYKES? Maxwell 90 Camp Agricolan possible c.43 f KINTORE 1 Maxwel0 9 l Camp Agricolan possible c.46 g KINTORE 2 J&M 90, 86 Camp c. 8 h DURNO Maxwel0 9 l Camp Agricolan possible c. 58 i 1 Maxwell 90 Camp Agricolan possible c. 45 j YTHAN WELLS 2 Maxwel0 9 l Camp Agricolan possible Sc. 14 k BURNFIELD Maxwel0 l9 Camp Flavian possible + c8 . 1 MUIRYFOLD Maxwell 90 Camp Agricolan possible c. 44 m AUCHINHOVE Maxwell 90 Camp Agricolan possible Sc. 14 n BELLIE Crawford 49 Camp Agricolan possible c.4 o THOMSHILL? 2 Brit37 , 17 , Camp Agricolan possible 440 | SOCIETY OF ANTIQUARIES OF SCOTLAND, 1998

REFERENCES Frere 198S ,S 0 'Review articl Naminn ei g Roman Britain', Britannia, 11 (1980), 419-23. Hind F 198JG , 0 'The Romano-British name for Cotbridge', Britannia, (1980)1 1 , 165-71. Jones Mattingly& B , 199D , Atlasn 0A of Roman Britain. London. Richmond, IA 1922 'Ptolemaic Scotland', Proc Soc Antiq Scot, 56 (1922), 288-301. Richmond 195A I , 6 'Castellum Veluniate civiliand an Romaa n o s n frontier', Proc Antiqc So Scot,0 9 (1956-7), 1-6. RivetF 198 L 2,A 'The Brittones Anavionenses', Britannia, (1982)3 1 , Notes 321-2. Rivet, A L F & Smith, C 1979 The Place-Names of Roman Britain. London. Robertson, AS 1975 Birrens (Blatobulgium). Edinburgh. Robertson,AS 1983 'Roman coins found in Scotland, 1971-82', Proc Soc Antiq Scot, 113 (1983), 405-48. Strang, A 1994 Ptolemy's Geography Reappraised. Unpub PhD thesis. University of Nottingham. Strang 199,A 7 'Explaining Ptolemy's Roman Britain', Britannia, (1997)8 2 , 1-30. Strang 199A , 8 'The Analysi Ptolemy'f so s Geography', Cartographic (1998)) (1 5 3 , , J 27-47 . Strang forthcominA , g 'Ptolemy's Geograph lowee th f yro Danube region'. Tomlin, R S O 1983 'Non Coritani sed Corieltauvi', Antiq J, 63 (1983), 353-5.