In finis Borestorum exercitum deducit
Agr. 38.2

He led his army down into the territory of the Boresti

At the end of the summer of AD83 Agricola followed up his success at Mons Graupius with a show of strength both on land and sea. I have already explained my views on his naval ambitions. But where do the Boresti fit into the plans of the victorious army? In 1986 Ian Keillar produced for popular consumption an article entitled In Fines Borestorum - To the Land of the Boresti. In eager anticipation I sought to find answers to my questions. Who were the Boresti? Where were they located? Alas! No answers. They did not appear in any of the maps and, apart from the Mattingly/Handford translation (with all its flaws) of the sentence, there was nothing about the Boresti whom I assume were to be located in Moray. Who were these people? Two questions need to be answered. Firstly, why is the tribe of the Boresti unattested in Ptolemy? Are we really to suppose, as Burn 364 and Rivet /Smith 365 suggest, that they were a sub-division of one of the tribes which do appear on Ptolemy's map, or, as Mann/Breeze 366 propose, that they certainly seem to have escaped his notice or, as Breeze367 puts it so dramatically, were the lost (?) tribe of Ptolemy? Such a solution is merely an argumentum ex silentio, but at least is better than to suppose that Tacitus mentioned them because they were the furthest known tribe on the mainland. This leads to my second question. Why would Tacitus be bothered to mention an insignificant tribe in a Scottish context where he normally generalises 368? No other tribe of Scotland is mentioned from the moment Agricola crossed the Forth/Clyde isthmus. Are the Boresti as much a myth as Trucculensis portus? Hanson 369 noted that the location of this tribe remains a mystery and talks of a more northerly location, perhaps in Angus or in Moray where Keillar and Jones /Mattingly 370 also locate it, while Breeze371 , half-correct, proposes that their name may indicate that they were the people who lived at the end of the island. Henderson's suggestion 372 that Boresti might be a made-up geographical appellation and might fall out of use quickly cannot be supported. Till 373 paradoxically sited them in southern Scotland, north of the Forth/Clyde line. But no one is prepared to offer the obvious solution, that Tacitus never wrote anything of the sort. We need to examine the possibility of textual corruption in a codex where proper names are shown, on more than one occasion, to be suspect. I believe that there is a transmission error here, and perhaps one of long-standing. Had the correct form appeared in earlier MSS, it was likely to have been recorded by E2m who specialises in producing good quality doublets of dubious proper names 374. Ogilvie 375 noted that E was copied from an exemplar which was either in 'scriptura continua' or did not have the division of words satisfactorily resolved. This I believe to be the source of textual contamination and I suggest that Borestorum comprises two words and that we should read,

in finis boreos totum exercitum deducit

He led his entire army down into the northern extremities

This is palaeographically sound and the corruption from -t to -r and the fusion of words is exactly paralleled at Agr. 27.2, where E's meaningless locatura is corrected by E2 to loca tuta.

Just as Agricola split his army into three divisions in Caledonia376 in AD82, so it was reunited the following year for what was virtually a triumphal procession northwards to the point where Caledonia ended. No more imposing or more demoralising sight could have greeted the enemy than the full might of Agricola's army, with the fleet dominating the coastal waters. The total 377 conquest of Britain only makes sense if the Romans took hostages from the tribes living on the northernmost coast of Scotland where the island actually ended, thus meeting the demands of the army, inveniendum Britanniae terminum 378. This interpretation would invalidate Hanson's observation 379 that Tacitus makes no claim for Agricola reaching the end of the island, while a march northwards would negate Keppie's suggestion 380 that Agricola's descent into the territory of the Boresti could refer to a southwards march towards Aberdeen or Stonehaven. It is in reaching the northern limits of the island, as much as in its circumnavigation, that we find the final expression of total conquest381

I believe that Tacitus' northern extremities lie in Caithness. What we have is a Tacitean generalisation. After all, what was the point of introducing an insignificant tribe which, like Trucculensis portus, would be of no historical relevance to an audience accustomed to discourse and lacking maps 382? The Agricola is a monograph whose focal point is a personality 383, unlike the Germania, where tribes form the basis of the work. If we cannot locate the Boresti, can we expect Tacitus' readership to fare better?

Now it may be argued that the standard word for northern in Tacitus is septentrionalis and that boreus is a rare usage. But only two examples of the former 384 should not militate against the acceptability of the latter. Meridianus and australis are equally acceptable among Latin writers for southern. It is not impossible for boreus to be used by an historian to provide relative bearings. Ammianus, like Tacitus, uses septentrionalis twice 385, but in a geographical digression describing the Persidis extremitates (the extremities of Persia) he refers to boreum latus (the north side) 386 . Classical sources distinguish between septentrionalis Oceanus (North Sea)387 and Oceanus Hyperboreus (Ocean beyond the north)388. Greek is much more flexible than Latin in defining geographical bearings. Strabo uses such words as more northerly (boreioteroi)389 and refers to Thule as the most northerly (boreiotaten) 390 of the British isles, far more precise than the un-geographical ultima (furthest) of Virgil, Pliny and Solinus. Had such adjectival forms existed in Latin they would have been used. Dionysius Periegetes, writing in the second century, refers to the Mysi as the more northerly of the Thracians (boreioteroi Threikon)391, and it is interesting to note that Priscian, writing his Latin paraphrase of Dionysius in the fifth century, produces Mysi Thracum boreis in finibus orti 392. What Priscian has done is to emphasise the noun fines (limits, ends, extremities) rather than the adjective. The problem caused by the absence of superlative directional adjectives in Latin is solved by switching the emphasis from the adjective to the noun and selecting a noun containing a superlative sense (furthest point, extremity) and applying to it the word northern. The Greek word for extremity is akron. Polybius in a geographical digression 393 uses the plural form akra in reference to the extremities of Greece, and Strabo 394 employs the same word for the extremities of the inhabited world (cf. Tacitus' priorum exercituum terminos, finem Britanniae 395 and nos terrarum ...extremos) 396. Consequently the farthest points north (i.e. northern extremities) would be boreia akra (fines borei). Tacitus has converted a Greek expression into a latinised form. In reference to Ptolemy's map Ogilvie 397 stressed the Roman origin of the entire body of information and noted the failure to transpose into Greek the Latin terminal endings of the three northern capes (Dunnet Head, Duncansby Head, Noss Head). I grant that Ptolemy has transliterated many Latin names. But natural features which had existed long before the presence of the Romans in the area may have originally been given Greek names from earlier explorations and surveys (such as those in which Pytheas and Demetrius of Tarsus 398 had participated). Ogilvie's choice of words, northern capes, is a direct, if unconscious, rendering of boreia akra, which may have been the collective name given by Greek explorers to the Caithness coastline which stretches from Wick to Thurso. This area could well be Tacitus' fines borei. It is worth noting that Ptolemy 399 uses boreion akron for Bloody Foreland in Donegal, clearly not evidence for transliteration from Latin.

There is no doubt that deducit suggests that Agricola was taking his troops down from higher ground onto the coastal plains (the likeliest location would be the plains beyond Wick and Thurso) to link up with his fleet, perhaps in Sinclair's Bay, a convenient point for the fleet to begin its circumnavigation 400 . The tribes who gave hostages would, on the basis of Strang's co-ordinates 401, be Ptolemy's broch-dwelling Smertae and Lugi, doubtless known to Agricola, but of no significance to Tacitus. The taking of hostages suggests a virtual capitulation in the face of a land and sea cordon. This need not be seen as a major operation. The thousands who had fled from Mons Graupius posed no further threat, and were out of the reckoning (neque usquam conglobari hostes compertum 402). The opportunity for Agricola to reach the terminus Britanniae without interference was too good to be missed.

Possible locations for the site of Mons Graupius are well documented 403 and a march towards Caithness might well follow a battle somewhere in Sutherland. A.R. Birley 404 was right to support Henderson's argument 405 in favour of the far north as the site of the battle, and equally correct to draw attention to the words used by Calgacus 406, nulla iam ultra gens, nisi fluctus et saxa and by Agricola 407, in ipso terrarum ac naturae fine. The rhetoric put into the mouth of Calgacus comprises factual detail which Tacitus derived from his father-in-law 408. Agricola knew exactly where he was, and his remark that finem Britanniae castris et armis tenemus 409 is surely an indication that marching camps should be sought much further north. The northern extremities of Britannia lie in Caithness, not in Moray. Tacitus knew this because Agricola's fleet had proved so. Keillar's identification of forts at Thomshill and Cawdor, if they are genuinely Roman 410, would testify to a permanent or semi-permanent occupation in Moray much further north than Stracathro, the long considered limit of Flavian occupation. If forts can be located further north, then so can marching camps. If the battle of Mons Graupius and Agricola's subsequent descent to the plain occurred where I suggest, it gives support to Tacitus' reference that Agricola's forces had reached the finem Britanniae and that total conquest had been achieved 411. It is somewhat ironic that Agricola, the paragon of modesty, is being pushed by Tacitus to rival the achievements of the egoistic Alexander who was seized by cupido visendi Oceanum adeundique terminos mundi 412. No Roman army had ever before reached the northern extremities of Britain, just as no Roman navy had ever before reached Thule. In terms of exploration and conquest it would be reasonable to claim that Agricola and Alexander had reached the furthest limits in their respective parts of the world. 413

Send email to the author:

INTRODUCTION | PART ONE: Shetland: The Classical Geographical Context
PART TWO: The Manuscripts of Tacitus' Agricola | PART THREE: Reassessment of Vocabulary and Sense
PART FOUR: The Search for a Harbour | PART FIVE: Thule in Contemporary Latin Literature
CONCLUSION | APPENDIX The Boresti: The Creation of a Myth

Copyright © 2002 by Stan Wolfson. All rights reserved.