The  Boresti tribe and Mons Graupius


1  The purpose of this page is to identify the Boresti tribe mentioned by Tacitus in his Agricola, 38  and then to discuss the consequences of that identification regarding the location of the battle of Mons Graupius, as well as the significance and consequences of the battle.

2   The term ‘British Celts’ is used at various places on this page simply as a label to identify the Celts who came over to Britain from the Continent in the few hundred years leading up to about 100BC, and their descendants, and to distinguish them from the pre-Celtic population of Britain and from the Gaelic-speaking Scots who came to Britain from Ireland in the 5th century AD. Note that those British Celts living north of the Antonine Wall were referred to as ‘Picts’ by some late Roman writers and have been so called by many other writers since then. These northern British Celts are discussed in detail in ‘The Celtic “Picts”’, accessible from the Main menu above.

3   The explanation of the identification of the Boresti will be difficult to understand unless one has some knowledge of the structure of Romano-British Celtic place-names. An explanation of the building-blocks used in Celtic place-names is given in Chapter 1 of the Home menu, whereas the structure of compound names is discussed in Chapter 2. A brief recapitulation might be useful here. Different waves of Celtic settlers arriving from the Continent used different letters of the alphabet to signify ‘hill’. These letters are n, s, m, r and l, though each of the letters n and l appears to have been used by two different groups of settlers arriving at different times. The normal chronological order of the hill-letters appears to have been n1, s, m, r, l1, n2, l2, though there is some regional variation, for example in Lincolnshire the hill-letter n2 appears to have arrived before l1, this being seen in Croconcalana (apparently the Celtic form of the Crococalana of the Antonine Itinerary) and Banvobalum (the Celtic form of the Bannovalum of Ravenna). There are three different types of Celtic place-name – old-style names, transitional names and inversion-type names. In old-style place-names a hill-letter is used as generic term and a qualifier is placed before it. The qualifier may be the letter b signifying ‘high’, as in the br of Bremia, or a c/g signifying ‘steep’, as in the Cer of Cerma. In transitional place-names the hill-letter has two qualifiers, one before it and the other after. The qualifier ‘high’ may come before the hill-letter, in which case it is represented by the letter b. Alternatively the qualifier ‘high’ may come after the hill-letter, in which case it is represented by the letter t. The qualifier c/g meaning ‘steep’ may appear before or after the hill-letter. Examples of transitional place-names or name-elements include the broc of Brocara, the cunet of Cunetio, the cent of Gabrocentio and the bulg of Bladobulgio (the Blatobulgio of the Antonine Itinerary). In an inversion-type place-name the hill-letter is again used as generic term but the qualifier comes after it. The qualifier may be a t signifying ‘high’ as in Litana or a c/g signifying ‘steep’, as in Alicuna and Isca. The letter d is used in some place-names to represent ‘summit’, ‘top of hill’.  This d is used as a generic term and is qualified by one of the hill-letters. The hill-letter comes before the d in old-style place-names and after it in inversion-type names. And the qualifying hill-letters may themselves be qualified by letters signifying ‘high’ or ‘steep’ as set out above for old-style and inversion-type names. Bereda and Corda are examples of old-style place-names of this kind, the dert of Omiretedertis an example of an inversion-type name-element. The writer is not aware of any transitional name or name-element employing the letter d signifying ‘summit’. The letter v is used in some place-names to represent ‘side of hill’, ‘slope’. This v is used in the same way as the d meaning ‘summit’. We thus see old-style place-names such as Banva (Ravenna’s Banna) and Glevon, and inversion-type names such as Venta and inversion-type name-elements such as the Ver of Verulamium.

4   Like most tribes north of the line of the Antonine Wall the Boresti appear to take their tribal name from a topographical place-name,  presumably the name of  the tribal centre. The table below indicates the tribal name in the case of each of ten other tribes, the hill-letter(s) used in the name, the topographical elements employed (shown bold in the third column) and some clarifying comments.


Tribal name Hill-letters used Topographical elements Comments
Caereni r Caber + eni Caber = ‘steep high hill’; b dropped
Caledoni l1 Caled + oni


Caled = ‘steep hill summit’


Carnonacae r, n2 Car + n + o + nac + ae


Car = ‘steep hill’, nac = ‘hill steep’


Cerones r Cer + ones

Cer = ‘steep hill’

Cornavi r, n2 Cor + n + avi


Cor  = ‘steep hill’


Decantae (r), n1 Derc + a + nd +ae

r between d = ‘summit’ and c = ‘steep’ because the Decantae used the hill-letter r

nd = ‘hill summit’
Epidi (n2)

E + bind + i



bind = ‘high hill summit’

b → p; n dropped


Smertae s, m, r S + m + e + rd +ae


rd = ‘hill-summit’


Vacomagi (n2), m Vanc + o + mag + i


n between v = ’slope’ and c = ‘steep’ because the Vacomagi used the hill-letter n2

mag = ‘hill steep’


Verturiones r, l Vert + u + l + iones

Name originally Vertuliones

Vert = ‘slope of hill high’


The topographical place-names of the tribal centres will then be obtained by employing a Celtic ending such as ion or onion with the topographical elements shown in column 3. The tribal centres would thus have names such as Caberenion, Caledonion, Carnonacion, Ceronion, Cornavion, Dercandion, Ebindion, Smerdion, Vancomagion and Vertulion. By simple changes well-known in Romano-British place-names  the tribal name CabereniCaereni (omission of b), Dercandae → Decantae (omission of r; dt), EbindiEpidi (bp; omission of n), SmerdaeSmertae (dt), VancomagiVacomagi (omission of n) and VertulionesVerturiones (l/r interchange).

5   Boresti is a name of exactly the same kind as those given in the table in paragraph 4. But the name isn’t quite right as it stands. Bor is an old style element and st an inversion-type element, so the elements are in the wrong order. The Celtic name must therefore have been Voresti, comprising the inversion-type element vor in the hill-letter r and meaning ‘slope, side of hill’, and st meaning ‘hill high’. The form of the name in Tacitus merely shows the common v/b interchange. The tribal centre would then have a name somewhat of the form Vorestion.

6   Note that the place-name Vorestion  would be entirely appropriate for the topography of Forfar. This town has been the county town for several centuries and  given its fairly central location in Strathmore it may well have been a tribal centre two thousand years ago.  There is a hill immediately south of the town, rising to some 170 metres, and indeed part of the modern town stands on the lower slopes of the hill. The two elements st meaning ‘hill high’ and Vor meaning ‘slope’, ‘side of hill’ thus seem entirely appropriate. Perhaps the settlement of the s-people was at the foot of the hill and then when the r-people took over they added a new part to the settlement, that new part standing on the slope, hence the Vor element in the name.

7   At some stage the British Celtic place-name may have acquired an additional element, perhaps a Gaelic or even Anglian element of the form (????), so that the place-name became Vorest(????) which was then shortened by deletion of internal letters (as happened in a number of other cases – see Alphabetical List/Changes in names over time, 11.1 for examples) to a form Vorfar, this changing later to Forfar.

8   Ptolemy gives us the names of various tribes who occupied territory in northeast Scotland. He mentions the Venicones, who occupied a swathe of land extending eastwards from Braco (Ardoch) in the direction of the river Leven, the Damnoni, who occupied Strathearn to the west of Abernethy, the Vacomagi, who occupied Perthshire north of the confluence of the Almond and Tay, including the upper Tay valley, the Taxali, who occupied part at least of Aberdeenshire, and the Decantae, who occupied the Inverness area. In addition Ammianus Marcellinus mentions the Verturiones, who occupied Strathspey. Leaving aside the high Grampians, the only sizeable region of northeast Scotland for which we have no tribal name is in fact Strathmore. Now, other pages of this website have demonstrated that Strathmore was occupied by a people who used the hill-letter r,  this being seen in the names Rascatonion (Ravenna’s Ravatonium) at Stracathro and Rugulentum (Ravenna’s Ugrulentum) at Cardean, at the confluence of the Dean Water and river Isla.  And Vorestion is a topographical place-name in which the latest hill-letter is in fact r. The name is correct for the topography of Forfar and the Vor of Vorestion  appears to have survived in the For of Forfar. It thus seems clear beyond reasonable doubt  that the Voresti, the Boresti of Tacitus, were indeed a British Celtic tribe living in Strathmore, and in the absence of evidence to the contrary one assumes their territory  extended from the river North Esk (perhaps a little further north) down to the river Tay and its estuary.

9   As a direct consequence of identifying the Boresti as a tribe living in Strathmore we can rule out any location north of the river North Esk as the site of the battle of Mons Graupius. For if the battlefield had been north of the North Esk then after the battle, following the actual text of the Agricola 38, the Romans would have had to march south to reach Boresti territory at the northern end of Strathmore and would there have taken hostages from the Boresti. Agricola could then have got word to the commander of the navy to tell him to sail round Britain. But then Agricola would have had to march south through Strathmore, through the territory of the Boresti, to reach the winter quarters of his troops, since these winter quarters were all in the south. But by any reasonable interpretation of the Agricola 38, Agricola did not march his troops slowly through the territory of the Boresti, but through the territory of other tribes. Given this inconsistency, it follows logically that the battle of Mons Graupius cannot have been fought north of the river North Esk.   

10   Furthermore, since after the battle of Mons Graupius Agricola marched his army to the frontier of the Boresti, or into their territory, we can conclude that the battle did not actually take place in the territory of the Boresti, i.e. between the North Esk and the river Tay and its estuary.

11   It is thus clear from paragraphs 9 and 10 above that the battle of Mons Graupius took place west or south of the river Tay. This is fully consistent with the view expressed in Home/Chapter 16: Roman place-names in Scotland, 6 and 6.1 that the name Victorie  in Ravenna identifies the site of the battle of Mons Graupius as a location in Strathearn, between the Water of May and the Ruthven Water, apparently at the foot of Craig Rossie, near the village of Dunning. The information in the Agricola 38 then makes sense. The day after the battle Agricola sent out scouts and they reported that they saw no sign of the enemy regrouping. Agricola then marched his army east along Strathearn to the river Tay, apparently the frontier of the Boresti. He took hostages from that tribe and then ordered his navy, presumably at anchor in the Firth of Tay, to sail round Britain. He then marched his army slowly through the territory of the tribes he had defeated, i.e. through Fife, Kinross and Clackmannanshire, and then sent his troops to their winter quarters.

12   As a consequence of the above it seems quite clear that the Boresti did not themselves participate in the battle of Mons Graupius. And if they did not participate in the battle it is likely that the tribes living further north did not participate either. What this means is that the battle of Mons Graupius was not the knock-out blow which finally brought British Celtic resistance to the Roman occupation to an end, forty years after the invasion of AD43. It was a battle which only broke the resistance of the tribes living in Fife, Kinross and Clackmannanshire. These were an l2-people living in northeast Fife, the Venicones living between Braco and the river Leven, and the rump of the Damnoni who had not been brought under Roman control in AD82. It was no doubt a useful victory for the Romans, but it was not a great triumph as has often been claimed.

13   Agricola was apparently recalled to Rome after the battle of Mons Graupius, but it is clear that there was still a great deal for his successor as governor to do. He appears to have annexed the territory of the Boresti in AD84, the year after the battle of Mons Graupius, and built several forts in their territory, namely at Cargill, Cardean and Stracathro. He also built a fort further north, somewhere on the Bervie Water. Presumably the fortlet at Inverquharity was built at the same time. It is likely that in AD84 the Romans also annexed at least part of the territory of the Vacomagi, since they probably built Tuessis at Bertha, at the confluence of the Almond and Tay, in that year. It is likely also that the new governor sent a sizeable reconnaissance party north of the Mounth in AD84 (as evidenced by the line of Stracathro-type marching camps between Stonehaven and the mouth of the river Spey) to gather information which might help the Romans estimate the size and strength of any army that could perhaps be mustered against them if they were to attempt to annexe the rest of Britain, and also perhaps to assess the suitability of roads and tracks for the movement of a large Roman army. The report of that reconnaissance party and that of the naval commander who sailed round Britain after the battle of Mons Graupius were presumably considered by the emperor and his advisers during the winter of AD84/85 and a decision was taken to annexe the rest of Britain. As a consequence of that decision work started on the construction of the legionary fortress at Inchtuthil and a fort called Tamia (mentioned in the Geography of Ptolemy) was built further upstream, perhaps near the confluence of the Tay and Tummel.  

14   But the plans came to nothing. Events far away on the Continent led to a legion and some auxiliary troops being withdrawn from Britain, whereupon the Roman high command apparently decided they did not have the manpower they would need to complete the conquest of Britain. Those plans were therefore put on ice and Inchtuthil and Stracathro were dismantled. The same presumably happened to some or all of the other forts which had been built after the battle of Mons Graupius. The Romans withdrew south. How far south is not clear but in the early years of the new century the emperor Trajan established a new frontier from the Tyne to the Solway. There were further expeditions into the region north of the Forth-Clyde line in the Antonine period, perhaps as far as the Tay, and in the Severan period, right up to the Moray Firth, but the Romans were never to conquer the whole of Britain. And in the  4th century the northern Britsh Celts were to launch raids on Roman Britain, these culminating in the so-called Barbarian Conspiracy of AD367 in which the British Celts played an active part. To what extent the activities of the British Celts contributed to the downfall of Roman Britain is not clear, but 43 years after the Barbarian Conspiracy, around AD410, Roman Britain just faded away into history.



[This page was last modified on 19 February 2021]