Traprain Law is a massive volcanic intrusion, rising to 221m above the low surrounding landscape, which, as Fraser Hunter, FSA, has pointed out, was a major centre of the East Lothian plain throughout the Roman Iron Age.Footnote 1 The hill fort is the basis of any attempt to understand the relations between the peoples of southern Scotland and the Roman world. Alexander Curle and James Cree carried out key excavations on behalf of the Society of Antiquaries of Scotland in 1914–15 and 1919–23. On Monday 12 May 1919, Curle was summoned to the site by his foreman. When he arrived, he was astonished by the sight of about 22kg of fragments of ‘bowls, cups, spoons and a miscellaneous collection of pieces of plate, tarnished and soiled, but obviously of silver’.Footnote 2 In a very short time Curle published a model account of the treasure, discussing not only the individual pieces of plate and their use within the Empire but possible reasons for the treasure being broken up and the historical setting.Footnote 3 This paper, which arises out of the reassessment of the Traprain Law treasure by an international group of scholars at the invitation of the National Museums of Scotland, led by Fraser Hunter, FSA, is mainly concerned with the latter aspect of Curle's report.

The linking of archaeological discoveries to historical events is often attempted, but is full of risks. This is illustrated well by the stories that have been built around the Traprain Law treasure itself, and which assume that Hacksilber Footnote 4 was broken up by raiders or pirates, identified sometimes as Saxons but by others as Irish. In 1923 Curle suggested that the treasure was seized by Saxon ‘spoilers’ on a rich Roman site or sites on the continent, and was then taken back to Traprain Law.Footnote 5 His first argument was that the treasure as a whole could not have come from Britain and therefore ‘was the fruit of raids on the continent’ because: first, the assortment of ‘intermingled’ church plate, temple plate, table plate and personal ornaments did not make up a single service, like the Chaourse treasure, and that, in any case, services of plate ‘of magnitude and importance’ can hardly have belonged to settlements in Britain, ‘at least in the northern part of Britain under Roman occupation’; second, the ‘Teutonic ornaments’ had no connection of shape or style with ‘the Teutonic tribes that invaded Britain’; and third, given that two of the siliquae are issues of Honorius, ‘no coin of Honorius … has been found in the north of England’.Footnote 6

Curle concluded that all hoards of the period, whether containing complete objects or broken ones, demonstrate his thesis: ‘The sack of Rome by the Visigoths in 410, and the various inroads of the barbarians into the Roman Empire at this period, must have set in circulation, as bullion, masses of precious metal which had been seized as loot. The anticipation of pillage is shown in the concealment of such hoards as those of Corbridge and of the Esquiline Hill; while the realisation is vividly depicted by the condition of deposits like those of Traprain Law and Coleraine.’Footnote 7 Of these, the only other Hacksilber he mentioned was the treasure from Coleraine.Footnote 8

The hoard found in 1854 at Ballinrees, near Coleraine, in Northern Ireland, consisted of more than 1,500 silver coins, eleven ingots or bars and two fragments of ingots, and six fragments of vessels.Footnote 9 In the original publication, J Scott Porter concluded that the silver belonged to ‘a proconsul, a publicanus, an officer in the army’, who took the treasure to Gaul or Britain, and that it was then ‘broken up and sold for old silver’, after it was ‘injured by fire or other accident’, at which time it was sold into Ireland, where ‘the art of the silversmith was exercised … and old plate and disused or damaged coins were commercially imported as the materials of that trade’.Footnote 10 The date at which this occurred, he concluded, was after ‘the reign of Honorius, the latest emperor … whose inscription occurs on the coins, and probably before the erection of mints and the issuing of silver money by the Anglo-Saxons in Britain, the Franks in Gaul and the Visigoths in Spain … These limits would place the formation of this hoard between ad 423 and ad 600.’Footnote 11

Scott Porter's interpretation was not and is not convincing, and it is not surprising that more informed theories took its place. The first was by Sir Arthur Evans, whose short account in 1915 of the coins from Coleraine drew support for his dating of the deposition of the Coleraine treasure from a life of St Patrick published in 1905 by J B Bury, the great historian of the Late Roman Empire. Bury, adducing the events of Patrick's career from Patrick's own writings, from Irish annals and from later biographers, accorded these events absolute dates.Footnote 12 Key to this scheme of Patrick's life and, according to Bury, his birth in ad 389 (instead of the traditional ad 374) and death in ad 461, at the age of seventy-two (instead of the traditional ad 493, at the age of about 120), was a great raid on Britain by the Irish High King Niall Noígiallach, Niall of the Nine Hostages. Bury supposed the raid to have been that which is said to have occurred about ad 403–4, and that this was when Patrick, who tells us that he was fifteen at the time, was captured, just before Niall's death, according to the Annals of the Four Masters, in ad 405.Footnote 13 In 1915, Arthur Evans reviewed the coins of the Coleraine hoard and suggested that the date of the arrival of Patrick as a boy of sixteen in Ireland, as argued by Bury, about ad 403–4, ‘corresponds very nearly with that of the Coleraine hoard’, the deposition of which Evans placed ‘in or after ad 408’.Footnote 14 He concluded that: ‘It is by no means improbable, moreover, that the booty represented by the Coleraine hoard and the captivity of the boy Patrick were actually due to the same Irish raid, perhaps one of the latest enterprises of King Niall, who perished in “the Sea of Wight” about ad 405.’Footnote 15

Curle made no reference to Evans, to his dating of the deposition of the Coleraine treasure, and to his suggestions, based on Bury, of its origin; but in 1924, the year after the publication of Curle's volume on the Traprain Law treasure, William Ridgeway published a paper explicitly linking the treasures of Traprain Law and of Coleraine, their hacking, their ‘looting’ and their concealment. One of the purposes of his paper was to refute Evans's suggestion that the origin of the Coleraine treasure and its coins was Britain, and to show therefore that Curle's parallel statement about the origin of the coins from Traprain Law was correct.Footnote 16 Ridgeway's paper is in two parts. Using the evidence of coinage of about ad 400, as known in 1924, and invoking the numismatic work of R G Collingwood and advice from Harold Mattingly, he concluded that the silver coinage of Honorius, present in both treasures, was not to be found in the north of Britain, and that such coins which were found further south were not later than ad 398.Footnote 17 From this it followed, he maintained, first, that the coins – and accompanying objects – in the Coleraine (or Ballinrees) and Traprain Law treasures must have been acquired not in Britain but further afield, in Gaul, and, second, that the occasion of the looting and deposition of the Coleraine treasure was fixed by the latest coin in the Coleraine treasure, which he dated to ad 403.Footnote 18

In the second part of his paper Ridgeway attempted to establish ‘Who brought the booty to Traprain Law and Ballinrees?’Footnote 19 He argued that, in ad 395, on the death of Theodosius i and the succession of Honorius under the guardianship of Stilicho, Britain was invaded by Franks, Saxons, Picts, Scots [ie Irish] and Attacotti, and that the Roman poet Claudian, in his poem about the consulship of Stilicho, showed that ‘the most formidable onslaught had come from Ireland under one powerful leader acting in co-operation with the Picts and Saxons’.Footnote 20 Ridgeway aimed to ‘get further light upon this mobilising of all Ireland against the Roman dominions’.Footnote 21 He therefore cited Claudian, who referred to ‘Scotus’. Ridgeway interpreted ‘Scotus’ as his ‘one powerful [Irish] leader’, because he wished to identify him as Niall Noígiallach, Niall of the Nine Hostages.Footnote 22 Ridgeway believed that Niall was a historical figure, an Irish king and the eponymous ancestor of the Uí Néill who dominated Ireland from the sixth to the tenth centuries.Footnote 23 He commented that: ‘None of the historians of Roman Britain seem to have noticed the great part played at this time by Niall Naoighiallach.’Footnote 24 Either Ridgeway did not know about, or he ignored, J B Bury's 1905 book.

Ridgeway's own conclusion, after surveying the literary evidence for the existence of Niall, was that: ‘there can be no doubt (1) that there was such a king as Niall, (2) that he played a great part between 379 and 406 in the breaking down of the Roman power in Britain and Gaul, and (3) that [St] Patrick, whether captured in Scotland or Armorica, was brought to Ireland, probably about 387, on one of Niall's raids’.Footnote 25 In support, Ridgeway then compared the available numismatic evidence with the literary and concluded that: ‘As the coins found at Chester, Worcester, Caerleon, and Careen all point to a catastrophe having overtaken these towns in or about 395 – the very year in which “the Scot” mobilised all Ireland in combination with the Picts, the Saxons and the Franks against the Empire – there is very high probability, especially in view of the Irish evidence of Niall's continuous activities, that it was he who destroyed these towns in that year.’Footnote 26

From this Ridgeway came to a conclusion about the Coleraine and Traprain Law hoards:

Ridgeway's interpretation has since had a persistent following, appearing sometimes partially (eg referring to ‘Irish raiders’ but at a different date) and mostly in summary and referring more often to attacks by the Irish under Niall of the Nine Hostages and to St Patrick than to the evidence of the treasures themselves, in histories and archaeologies of Roman Britain and Ireland, histories of the early Church, and exhibition catalogues.Footnote 28 There have been, however, at least two strands of scholarship that undermine all this.

It is not surprising that Ridgeway should have thought of the deposition of coins as being always linked directly to danger. Coin evidence, in the form of a supposedly unusually large number of hoards, has been claimed for various raids and invasions across the Roman frontiers. Ridgeway does not mention it; but he must have known that, in 1900, Adrien Blanchet published a study of 871 coin hoards, deposited in Gaul and the Germanies between the first century bc and the early fifth century ad.Footnote 29 Blanchet's interpretation was based on the premise that hoards of coins were buried because of the imminent threat of an enemy, and that in Gaul and the Germanies the enemy was invariably invading Germans.Footnote 30 He surveyed the historical evidence for German incursions and then made an inventory of hoards of coins, by date and by find-place. He concluded that, like the historical sources, the coin evidence demonstrated that the German invasions caused the greatest trouble in Gaul in the second half of the third century and in the fourth and early fifth centuries, which accounted for the number of hoards buried between the reigns of Claudius ii (ad 268–70) and Honorius (ad 393–423) – almost half the hoards in his lists. Ridgeway would have been pleased to see that Blanchet concluded that the evidence from Britain was in line with the evidence from Gaul, and that Britain also has a bulge of coin hoards of the period ad 260–94, even though it did not suffer from Germanic incursions, a statement which is still true.Footnote 31

Modern numismatists, however, have put forward comprehensively different interpretations of third-century hoards. After forty years of debasement, the old coinage was replaced with coins of higher value by Aurelian, in ad 274, and by Diocletian, in ad 294. The old coins continued to circulate as small change, but their purchasing power continued to diminish. Most hoards of the period ad 260–94 consisted of the poor, low-value coins, but Richard Reece, FSA, has suggested that their owners could no longer be sure that they would have the same purchasing power when they were recovered from the ground.Footnote 32 Indeed, Reece has concluded, British hoards were probably put away because they had already lost their value, and a good reason for such a large number of hoards not being recovered was that the coins never regained it. The numismatists and archaeologists of Gaul and Germany have in the past been more reluctant to abandon a connection between hoards and Germanic invasions, but views have begun to change, and scholars are beginning to state that Blanchet's theory cannot be accepted as stated.Footnote 33

In the light of these reassessments it becomes clear that Blanchet's theory that all Gallic and German hoards were concealed solely because of Germanic invasions loses credibility.Footnote 34 The failure to recover so many, if not all, third-century hoards in Britain, Gaul and Germany was, rather, largely a consequence of an economic and monetary crisis.Footnote 35 If so, then Ridgeway's arguments, that the coin evidence demonstrates late fourth- and early fifth-century incursions, this time by Niall in Gaul and Britain, including the capture of Patrick, cannot stand.

The second part of Ridgeway's argument is that the ‘Scotus’ referred to by Claudian was Niall of the Nine Hostages. This depends on references in the Annals of the Four Masters, which date Niall's accession as king of Ireland to ad 379, and his death to ad 405, but which are not contemporary.Footnote 36 The activities and deaths of Niall's sons, however, are at the very end of the fifth and the beginning of the sixth. The annals suggest that his eldest son, Coirpre, was alive in 485, when he won a battle over the Leinstermen at Granard; they do not record the death of any of Niall's sons earlier than Maine (ad 440), whose death is followed by that of Loéguire (c 462), Eógan (c 465) and Conall Cremtheinne (c 480).Footnote 37 All this requires that Niall should have died not earlier than about ad 450.Footnote 38 It follows that St Patrick was not captured by Niall at the beginning of the fifth century.Footnote 39 More importantly, in the present context, the account of a specific raid on Britain, whether in ad 403–4, according to Bury, or in ad 405, according to Evans, or simply as ‘one of Niall's raids’, according to Ridgeway, is without foundation, and there is no justification for attaching to it the deposition either of the Coleraine treasure or of the Traprain Law treasure. The failure of attempts to do this illustrates the hazards of linking historical or pseudo-historical sources with archaeological discoveries. Hacksilber tells us nothing relevant to Niall, and vice versa, nor does Niall have anything to do with the last two decades of Roman rule in Britain.