2. Ptolemy's Map
Ptolemy's map
Ptolemy’s map provides the earliest depiction of the geography of Scotland and contains information largely derived from the time of Agricola’s invasion (Richmond, 1922). It sets out the names of various Caledonian tribes and centres of population, as well as a list of the prominent coastal landmarks, river mouths and estuaries. North of the Forth (Boderiae Aest) Ptolemy lists twelve tribes and ten sites or polis mainly in Perthshire and the north east, possibly reflecting a pattern throughout his work that places ten towns on each page (Ferre, 1980). These ten polis seem to be a mixture of Roman forts and local tribal centres. The names of Lindum, Bannatia, Tamia and Devana form a line above which only two centres are referenced (Tuesis and Pinnata Castra) along with a conglomeration of tribal designations. It is very tempting to see in this the representation of the topographical dividing line which separates the highlands from the lowlands. This natural phenomenon would also account for the variance in detail between the two halves of Ptolemy's portrayal of northern Scotland.
South of this line, but north of the Forth and Clyde estuaries we encounter a place called Victoria. Victoria, or as she is better known Nike, is a winged Roman goddess associated with military conquest and the vanquishing of enemies. The Bridgeness distance stone from the Antonine Wall shows Victoria was honoured and venerated amongst the legions which invaded present day Scotland. Ptolemy's referencing of Victoria is probably also another example of this veneration. In the Roman world there were numerous cities called Nicopolis. Such a city was founded when Pompey vanquished Mithradates VI in Pontus, when Julius Caesar conquered the Ptolemies in Egypt, and most famously of all at Actium in 31BCE where Octavian, who was later known as Caesar Augustus, defeated Antony and Cleopatra (Nicoplois in Epirus). The honouring of Nike or Victoria at battle sites also persisted into the imperial age. For instance, we know the emperor Trajan founded a city called Nicopolis ad Istrum at the site of a critical battle in the Dacian Wars. All these Greco-Roman cities were not only dedicated to Nike (Victoria), but were situated at the site of an actual battle. The very ground on which a victory occurred came to be associated with and given over to the goddess. Therefore, this goddess was not only honoured after any significant conquest or victory, but often the sites of famous battles would be dedicated to her. This suggests the marking of Victoria on Ptolemy's map might be indicative of a significant military conquest (Maxwell 1990, 116-118).
This idea was widely advocated by Ogilive and Richmond (1967, 243-244) who believed Victoria denoted the fort at which IX Legion was rescued from imminent defeat by Agricola. They maintained after the repulsing of the Britons the fort was renamed in honour of the goddess. However, they never considered the possibility the designation might mark the site of a significant battle and conquest. Given the above precedents Victoria is usually acclaimed and honoured after a decisive battle in which Rome's enemies have been entirely vanquished and defeated, not some skirmish were the Roman army narrowly avoids defeat. It is invariably the final act of a war were thanks is given to the goddess for Rome's success over the course of a campaign (see Bridgeness Stone). Therefore, Ptolemy's Victoria is much more likely to mark the location of a significant and decisive Roman triumph and possibly the site of Mons Graupius. However, the locating of Victoria presents us with a number of challenges.
Locating 'Victoria'
The reading of Ptolemy's map is far from straight forward partly due to its orientation. Ptolemy has turned the lands north of the Forth-Clyde line through 90' degrees, effectively laying the north of Scotland on its side. However, the estuaries are still discernible with the Clyde (Clotae), Forth (Boderiae) and the Tay (Tauae) estuaries clearly marked. These markers place Ptolemy's Victoria north (west) of the Forth and south (east) of the Tay, in southern Perthshire or Fife. Beyond this identification there is a degree of supposition. However, if we plot the principal known Roman sites of northern Scotland there is a remarkable degree of congruence between the pattern of sites suggested by Ptolemy and what appears in our map. It would appear the internal logic of Ptolemy's map has survived the reorientation of Scotland and if we plot the sites according to the position of the rivers and estuaries, although there is a degree of geographical distortion, Ptolemy's representation is remarkably congruent with the orientation of each of these sites to one and other. This is especially true when it comes to those Roman forts known as 'the glen-blocking forts' or highland line forts. These forts constituted part of a Flavian defensive system which ran in almost a straight line from Malling (on the Lake of Menteith) to Stracathro beyond Brechin.
If we plot these Roman sites Ptolemy's Alauna, Lindum, Bannata, Tamia and Devana closely correspond with the Roman forts which have been identified at Ardroch, Dalginross, Fendoch, Inchtuthil, and Stracathro. It is also possible to suggest Tuesis and Pinnata Castra are proto-pictish indigenous sites which would correlate well with the later Pictish power centres at Tap o' Noth, near Rhynie and Burghead on the Moray coast. (Although we should note the proportionality of the native sites in the north-east is more distorted and truncated than that seen in relation to the alleged Roman forts. This probably reflects a greater awareness and control of the territory south of the line of the forts, which might also account for the lack of detail about locations beyond these forts). Ptolemy is providing a cartographic record and the interrelationships he is suggesting between poleis are essentially sound, with the exception of south east Scotland.
Roman Forts and Ptolemy's locations.
This affirmation is very much confirmed if we consider the actual names of some of the poleis. The names of Alauna, Lindum, Bannata, Tamia and Devana can all be read as indicative of Roman forts. According to the cartographic reconstruction above Tamia represents the legionary fortress at Inchtuthil. In 1926 the celebrated Scottish toponymist William Watson observed the connection between this name and the Tay river. He maintained the name of the Tay was originally derived form Tavia rather than Tava and meant 'silent and peaceful'. As V is often erroneously copied as M, the poleis Tamia reflects the name of the river. Therefore, the legionary fortress at Inchtuthil must have derived its name from the Tay, a pattern which corresponds with what is seen elsewhere in Roman Britain. As one commentator observes: ‘If we consider legionary forts in the west of Britain, each takes the name of the river that it stands on: Isca (Exeter), Isca (Caerleon) and Deva (Chester). Why should those in the east and north not do likewise?...What more appropriate name than Tava (Tacitus spelling Tauni) could be given to Inchtuthil, which was virtually surrounded by the river Tay’. Equally the name, Lindum mirrors the name of Lindum Colonia, the legionary fortress of Lincoln, while Devana carries echoes of another legionary fortress at Chester (Deva). There is also a suggestion that Bannatia might be a corruption of Britannia which again could be a link with one of the British legions. These names leave little room for doubt we are dealing with Roman forts. However, it would also be disingenuous not to recognise considerable debate has occurred in relation to the identification of these various places, especially in relation to Inchtuthill. Nevertheless, the patterns laid down by Ptolemy in the relationship of each site to the other closely correspondences with the location of the actual forts we have identified and the names would be consistent with this identification. Where then according to this analysis does Ptolemy place Victoria?
Victoria
Ptolemy’s locates Victoria between the Forth (Boderiae) and the Tay (Tavae). It is roughly in line with a river that is called Tina that is generally identified with the river Earn. On some maps Victoria even appears to be on this river. The site is also northeast of Alauna and southwest of a place called Orrea. The ordinance survey map of Roman Britain identifies Alauna with the Roman fort at Ardoch, near Braco, the water of Allan still preserving the name in the locality (Rivet and Smith, 1979). This places Victoria to the northeast of Ardoch. Orrea is more difficult to identify, but is usually assumed to be in the vicinity of Abernethy. The late seventh century Ravenna Cosmography speaks of a place called Poreoclassis which is thought to be the legionary base at Carpow, poreo probably coming from (H)orrea. However, Ptolemy locates Orrea a little ways from the coast and his location could suggest Glenearn Hill and the Iron Age settlement of Castle Law. It is possible this settlement provided a regional centre at the time of Agricola and may have been of sufficient significance to merit a reference by Ptolemy (Ferre, 1980). Such a location for Orrea means the orientation of Alauna, Victoria and Orrea on Ptolemy's map follows the line of the Ochil hills. This places Victoria somewhere between Ardoch and Castle Law in the Strathearn valley. If we follow the pattern of Ptolemy's map the most plausible location is a hill known as Rossie law, which is an outcrop from the very distinctive Craig Rossie. Rossie Law was one of the largest Iron Age hill forts in Scotland. When occupied it would have dominated and controlled most of Strathearn, regulating the movement of people from the north to the south and the east to the west. Again if we plot these locations on a map there is a remarkable coincidence between the pattern we see on Ptolemy's map and these sites. This pattern not only exists between Alauna, Victoria and Orrea, but also their relationship to the 'glen blocking' forts which we have previously identified. Ptolemy's map is entirely consistent.
However, it is unlikely Rossie Law was inhabited at the time of Agricola, although other Iron age forts sites in the area were. One in particular, Castle Craig, was excavated in 2011/12 by Glasgow University's Strathearn Environs and Royal Forteviot project (SERF) and showed definite signs of late first century CE habitation. On the summit of this hill a broch was uncovered with massive drystone walls over 5m thick and 1.5m high was uncovered. This discovery was exceptional for several reasons. As Stephen Driscoll, professor of historical archaeology at the University of Glasgow and director of the Serf project, explains: “The scale of the architecture is colossal and the tower-like structure would have visually dominated its surroundings. It’s not unreasonable to see this as a seat of a Celtic chieftain, who collected a wide range of luxury objects from the Roman world, perhaps through trading with the Romans or possibly even serving in the Roman army.” This find confirms the area around Rossie Law and the village of Dunning was very much an important power centre at the time of Agricola's campaign. However, the archeology also shows this monumental structure was destroyed and burnt to the ground at the end of the first or the beginning of the second century. Was this retribution by Caledonians for Roman collaboration or Rome exacting revenge for an insurrection? No definitive answer can be given, but what we can say is something highly significant which either directly or indirectly involved the Romans did occur near Rossie Law and Dunning. We have the destruction of one of the most important power centres in the district which coincides with the identification of a place called Victoria on Ptolemy's map and is contemporary with Agricola's campaign.
Yet the Glasgow University lecturer Duncan Campbell maintains 'it is hard to see how Ptolemy’s Geography can be used as evidence that the battle occurred in Perthshire' (Campbell 2015). This scepticism is based 'on the problems of plotting Ptolemy's co-ordinates on modern maps' and the questioning of any necessary link between Victoria and Mons Graupius. For instance, Rivet and Smith (1979) not only made a 90 degree correction to the overall orientation of the map's portrayal of northern Scotland, but deemed it necessary to add a further 70 degree correction for southern Scotland. This over correction meant the poleis of Orrea was identified by them with a site north of the Tay estuary. Similarly, Richmond (1967) tried to identify Pinnata Castra, which is clearly on the Moray coast, with Inchtuthil. Such identifications show an excessive mistrust of Ptolemy's map and creates a severe incongruity with his placing of rivers and estuaries. If we begin with these natural features and set the sites in relation to them in the way Ptolemy prescribes the cartographic expression of the map is found to have an internal logic. Ptolemy's poleis form patterns consistent with known sites which are not only congruent with the locating of the rivers and estuaries, but also other sites. Therefore, there is little doubt that Ptolemy's Victoria is located in Strathearn. Equally, while Victoria is not necessarily indicative of Mons Graupius, it could be. Given the way the goddess Victoria was often honoured after a significant victory and conclusion to a campaign the resistance to this proposition appears without any real foundation. The designation literarily means 'the place of victory'. Scholars like Iain Smith (1987) and James Fraser (2005) believe the placing of the battle of Mons Graupius in Strathearn is entirely plausible. However, Campbell is right in his assertion that this cannot be determined purely on the basis of Ptolemy alone. It requires a careful consideration of the archeological evidence of the Roman marching camps themselves.