[NB. The reader will probably find it easier to understand this page if he or she has already acquainted himself or herself with the contents of Chapters 1 and 2. Chapter 1 concerns the basic building-blocks used in Romano-British place-names and Chapter 2 explains how those building blocks were assembled by the Celts to form compound names. The numbers accompanying some names are those provided by Richmond and Crawford in Richmond and Crawford 1949.]
Deva Victrix to Bannovalum
Sandon and Leicester, nodal points of the Midlands
1 Ravenna has come from Mediolano (84) at Whitchurch to Sandonio (85) at Sandon, and then with a minute change of alignment it goes on to Ratecorion (92) at Leicester. But before going on to Leicester it lists places arranged in a circle around Sandon. This list starts with Deva Victrix (86) at Chester and then gives Veratino (87). This is presumably the same place as the Rutunio of the AI Iter II, and so lay 12 Roman miles south of Chester and 11 Roman miles north of Wroxeter on Margary road 6a, which would tend to place it at the point where the road crossed the river Roden. Rutunio is a river name, and Veratino will be the same name with the river-letter b, changed to v, added at the beginning (cf. river-names Vedra and Veromo). The Romans will merely have adopted the river-name for a fort which they built close to the river. Then come Lutudaron (88) at Wall and, further round the circle, Derbentione (89) at Littlechester. The lut of Lutudaron is an inversion-type element meaning ‘hill high’ and the dar an inversion-type element meaning ‘summit of hill’, so the name Lutudaron is a perfect topographical description of the site of the early fort on the summit of the hill at Wall. Derbentione is of course a river-name transferred by the Romans to a fort which they built close to the river, now called the Derwent. Finally, Ravenna moves further anticlockwise round the circle and lists Salinis (90) at Middlewich and Condate (91) at Northwich. Note that Wall, Sandon, Middlewich and Northwich do in fact lie on a straight line, so the forts there were presumably early forts built before the road-builders arrived.
2 And now Ravenna moves on to Ratecorion and proceeds to use it as a node, for it lists places on four alignments radiating from Leicester. The first place listed is Eltanori (93) at Lutterworth and then come Lectoceto (94) at Towcester, Iaciodulma (95) apparently at Little Brickhill, Virolanium (96) at St. Albans and Londinium Augusti (97) at London. Inserted after London are Cesaromago (98) at Chelmsford and Manulodulo Colonia (99) at Colchester. We will skip the next two names and mention Durobrisin (102) at Water Newton and Ventacenonum (103) at Caistor St. Edmund. Then come Lindum Colonia (104) at Lincoln and Bannovalum (105), apparently at Caistor to the northeast of Lincoln. One notes that the first group of names lie on an alignment to the south of Leicester, the third group (Durobrisin and Ventacenonum) on a west-east alignment out of Leicester and the final two names on a northeast alignment out of Leicester. It seems quite clear that the remaining two names, Durcinate (100) and Duroviguto (101) lay on an alignment generally to the southeast out of Leicester. Now there is of course a Roman road on precisely that alignment – it is Margary road 57a, and almost immediately after crossing the Welland the road starts to climb a steep, high hill. This is the ideal location for a place called Durcinate, where cinat is an element of the cant, cent, cunet and canat family. It is a transitional element in the hill-letter n and means ‘steep hill high’. Cottingham is thus to be identified as Ravenna’s Durcinate. Godmanchester is normally identified as Duroviguto. It also lies on Margary road 57a, but about half way from Leicester to Colchester, so if Godmanchester had been Duroviguto one might have expected Ravenna to list Colchester after Godmanchester instead of fitting it in after Chelmsford. It would thus appear that Duroviguto was indeed on road 57a, but not so far east as Godmanchester. The most likely location is at the crossing of the Nene, at the junction of Margary road 57a with road 570 coming down the east bank of the Nene from Water Newton. It might even be the case that the Durov of Duroviguto changed to Drob and eventually this changed to the Thrap of modern Thrapston. As to the name itself, the second element, viguto, looks like a topographical element which has lost its hill-letter. Perhaps the Celtic name had been somewhat like virguto meaning ‘on the side of a hill steep high’. The Crow Hill hillfort (at SP 957 715, just north of Irthlingborough, Northamptonshire) stands on a steep, high slope overlooking the river Nene. The name Virguto would be entirely appropriate for this hillfort. There would appear to be no other hillfort or Iron Age settlement near Thrapston, or anywhere near Godmanchester, for which the name Virguto would be appropriate. Presumably, then, the Crow Hill hillfort was indeed Celtic Virguto and the Romans simply transferred the name to a new town which they built some eight or nine kilometres away, close to the Nene at a point just north of Thrapston, the Duro prefix of the Roman name indicating that it was a new town. And in the case of Durcinate, Celtic Cinate may have been at Cottingham and Roman Durcinate on the banks of the Welland in the vicinity of Cottingham, most probably at the point where Margary road 57a crossed the river.
3 Ptolemy mentions a few names in the region covered by the names on alignments out of Sandon and Leicester. He mentions Deva at Chester, Ratae at Leicester, Urolanium at St. Albans, Salinae (unknown, but said by Ptolemy to be in the territory of the Catuvellauni), Londinium at London, Camulodunum at Colchester and Lindum at Lincoln.
4 A word about Ratecorion and Eltanori. Richmond and Crawford give the names Ratecorion and Eltanori but indicate that two of the surviving Ravenna texts give the spelling Eltavori (Richmond and Crawford 1949, 18). Now, Hassall and Tomlin mention a tegula found at Caves Inn, just south of Lutterworth, the tegula bearing an inscription including the name Ci) vitatis Corieltavvorum (Hassall and Tomlin 1983). One will remember that in Ravenna the names simply appear one after another with a dot between adjacent names. The above inscription suggests that the original names in Ravenna were Rate.Corieltavori, that at some copying stage the dot has moved to yield Ratecori.Eltavori, that an on has been added to Ratecori at some stage to make the name look more complete, and the spelling of Eltavori has changed to Eltanori. But the correct original forms of the names would be Rate and Corieltavori. Ekwall indicates that Lutterworth was Lutresurde in 1086 and suggests that this form might include a river-name Hlutre derived from Old English hluttor, meaning ‘clean, pure’ (Ekwall 1960). It seems just as likely, however, that the lutre element of Lutresurde is derived from Eltavori, the development being perhaps Eltavori → Eltauori → Eltauri → Lutri → Lutre.
5 Ravenna gives Lectoceto as the name of Romano-British Towcester, whereas Iter II and Iter VI of the Antonine Itinerary give Lactodoro. Iter II gives Etoceto for Romano-British Wall. Now, Ravenna’s Lutudaron is a precise and correct topographical description of the location of the first fort at Wall, the fort on the summit of the hill. And Ravenna’s Lectoceto is a precise and correct topographical description of the location of Towcester. The original form will have been Lectoseto, where lect is an inversion-type element meaning ‘hill steep high’ and set is an inversion-type element meaning ‘hill high’, and there can be no doubt whatsoever that Towcester is located close to steep, high hills. Note, too, that the river-letters corresponding to the hill-letters l and s of Lectoseto are t and b, and both of these are present, with the b changed to v, in the river name Tove, the modern town-name Towcester being a back formation from the river-name. There can thus be no doubt that Lutudaron and Lectoceto were respectively the names of Romano-British Wall and Towcester. The Iter II names illustrate confusion of Lutudaron and Lectoceto by taking the first part of each with the second part of the other, to yield Lutuceto and Lectodaron, though the forms actually given in Iter II are Etoceto and Lactodoro. Lutuceto lost its initial L and Utuceto became Etoceto, and Lectodaron changed to Lactodoro. But it is quite clear that the names Etoceto and Lactodoro are the result of confusion on the part of the compiler of Iter II and that the names did not actually exist in those forms. It is thus also quite clear that Towcester never was a walled town full of dairymen, as is commonly believed.
6 Of the other names discussed in this chapter, the vir of Virolanium at St. Albans is an inversion-type element meaning ‘slope’ or ‘side of hill’, indicating that the Roman town was located on a slope, as indeed it was, and Londinium is just a compound in the hill-letters l and n, the nd element indicating that Londinium was located on the summit of a hill, or at least on the top of raised ground. Manulodulo looks a very odd name, perhaps just Camuloduno wrongly spellt. But it is possible that the Celtic name had been Camanulodulo and that the initial Ca was lost or dropped at some stage, as happened in the case of some other names. Camanulodulo would be essentially the same name as Ptolemy's Camunlodunum in the territory of the Brigantes. The letter n in the middle of both forms would just be the name-ending of the original name Cam meaning 'steep hill'. The dulo of Camanulodulo may be correct, or may simply be a mistake for duno. The name Camuloduno is traditionally thought to be based on the name Camulos of a Celtic war god, though there is no evidence that there was ever an s in the name. And coins have been found bearing the text CAM, CAMU, CAMUL and even CAMULODUNO, some of them having been minted as much as 48 years before the Roman invasion (Rivet and Smith 1979, 294). It thus seems best to regard Camuloduno as a straightforward topographical compound in the hill-letters m, l and n, where the initial Cam means 'steep hill' and duno means 'summit of hill', the complete name thus referring to a location on top of the ridge at Colchester.
7 The spelling of Bannovalum is unlikely to be correct. Bann will originally have been the old-style element banv meaning ‘high hill slope’ and val is an inversion-type element meaning ‘slope of hill’. But the elements are in the wrong order, so presumably val was originally bal meaning ‘high hill’, Banvobalum then being a straightforward old-style compound in the hill-letters n and l. There has long been uncertainty as to whether Bannovalum was at Horncastle or Caistor-on-the-Wolds. The name is not entirely inappropriate for Horncastle since the town stands on the lower slopes of a hill rising to over 120 metres. But the topography at Caistor is much more striking. The town actually stands on an escarpment, the top of the modern town being some 80 metres higher than the bottom, so a name comprising two elements of which the first means ‘high hill slope’ and the second ‘high hill’ is clearly very appropriate for the location. Of the two places, then, Caistor is the more likely to have been Banvobalum. One reason why past writers have associated the name with Horncastle is that one of the two rivers at Horncastle is the Bain and it was thought that there might be some connection between the river-name Bain and the Ban part of Bannovalum. But there need not be any connection between the two. The river-name Bain may just be the river-letter b and a name-ending, as appears to be case with the river Bain at Bainbridge in Yorkshire. But, taking into account all that is said above, whilst Caistor is the more likely to have been Bannovalum, one cannot rule out Horncastle completely.
8 It is convenient to refer to Manduesedo (Mancetter) at this point. This name does not appear in Ravenna or Ptolemy but in Iter II of the Antonine Itinerary. However, the name was mentioned in the Introduction to this study and so it is fitting that the meaning of the name be explained at some point. The name Manduesedo has nothing to do with chariots, as has been assumed in the past. The hill-letter m appears to be used in an inversion-type manner, so that the name means 'hill called (or of) Anduesedo', where Anduesedo will have been the name of Oldbury hill-fort (and = sed = 'hill summit'). Manduesedo was thus most probably a new Celtic settlement (replacing the hill-fort) on the low ground at the foot of the hill. The Roman fort built on the west bank of the river Anker will then have taken its name from that new settlement, and later the name was transferred to the new civilian settlement on the other side of the river, on Watling Street. But it is of course possible that the full name Manduesedo had been applied to the hill-fort and was simply transferred by the Romans to their new fort down near the river Anker.
9 The locations of the forts or settlements discussed in this chapter are shown on the maps reproduced below. Note that Bannovalum is not shown on the maps – it was most probably at Caistor on the Wolds, to the northeast of Lindum Colonia at Lincoln. Likewise Iaciodulma, apparently at Little Brickhill, near Milton Keynes, is not shown on the maps. Apart from Rate and Corieltavori the forms of the names are those given by Richmond and Crawford, as is the numbering of the names. The identifications are as follows:
85 Sandonio Sandon on the river Trent
86 Deva Victrix Chester
87 Veratino at or close to the point where Margary road 6a crosses the river Roden
88 Lutudaron Wall
89 Derbentione Littlechester
90 Salinis Middlewich
91 Condate Northwich
92 Rate Leicester
93 Corieltavori Lutterworth
94 Lectoceto Towcester
96 Virolanium St. Albans
97 Londinium Augusti London
98 Cesaromago Chelmsford
99 Manulodulo Colonia Colchester
100 Durcinate Cottingham
101 Duroviguto Thrapston
102 Durobrisin Water Newton
103 Ventacenomum Caistor St. Edmund
104 Lindum Colonia Lincoln
[This page was last modified on 02 April 2022]