The hill fort of Dunadd, which dominates the landscape at the foot of Kilmartin Glen north of Lochgilphead in Argyll, is one of the best-known of the ancient sites in Scotland. Regarded for many years as the main citadel of the Scots, who had allegedly emigrated from Eire in the 5th-6th century AD, much archaeological investigation has been performed at this site in recent times, and here we examine some of the results, which are in many ways surprising. The archaeologists have unearthed how the kings who reigned during the so-called Dark Ages enjoyed rare dishes of exquisite foods, imported herbs and expensive clothing dyes. They have, through analysis, found traces of dill and coriander which were used in the flavouring of meat. Jewellery was produced locally using gold, silver and glass, in the form of beautifully crafted brooches and pins, incorporating Pictish, Celtic and Anglo-Saxon elements of design. The research also shows the variety of goods coming into Scotland along the western sea routes and provides solid evidence about the Scots who gradually superseded the Picts in eastern Scotland.

There were three early excavations at Dunadd, in 1904, 1929 and 1950. The 1904 dig showed evidence of an enclosed stone fort, and produced artefacts thought at that time to date mainly from the 5th-9th centuries AD. That of 1929 recovered more artefacts but contributed little else to the understanding of the site. RBK Stevenson's dig in the 50's defined Dunadd as a nuclear fort - having a central nucleus and looped lower enclosures, similar to that he had found at Dundurn in Perthshire, of a similar age and provenance. But it was Lane and Campbell's digs in the 1980's which produced some much more relevant and interesting information.

One such excavation produced evidence of the existence of a previously unknown metalworking area in the 7th century. More than 900 mould fragments and crucible parts were found, along with tools, raw materials and pieces of scrap metalwork. Many of these were very sophisticated, and the castings included pins, buckles and disks, but primarily what are known as penannular brooches, similar in form to the famous Hunterston brooch. Now on display in the National Museum of Scotland, this is one of the classic pieces of Celtic art, decorated in gold and amber. The moulds discovered are evidence that such pieces were produced in Scotland, certainly at Dunadd if not elsewhere, and as early as 650 AD rather than the previously thought 700's. It is entirely possible that the Hunterston brooch, once though to have been Irish in origin, was in fact made at Dunadd. Certainly many similar to this artefact were produced in this location.

Finds such as a trumpet-spiralled hanging bowl disc, and a pressblech (stamped animal decoration) which might have adorned something like a drinking horn, demonstrate that Dunadd and hence Dal Riata were central to the development of what became known as the Hiberno-Saxon style. Traces of the mineral colourant orpiment, a bright yellow originating in either Italy or Turkey, demonstrates not only the fact that ingredients for the manufacture of illuminated manuscripts, such as the Books of Durrow and Kells, were handled here, but also illustrates the amount of trade carried on by this western sea-port. Purple dyes of the type used to colour royal robes, generally believed to be from the south-west coast of France, have also been found. Continental pottery, finely crafted glassware, and raw materials from as far away as the Mediterranean reinforce the proof of this long-distance trade. The variety of finds shows us that the surplus wealth created by the manufacturing processes here was spent on luxury items from abroad as well as highly decorated personal jewellery. Dr Ewan Campbell has stated "Although we may think of early western Scotland as being peripheral and isolated, the archaeological evidence shows that this was not the case. Sea links enabled the early Scots to be closely involved in developments in Europe and their economy was at least as advanced as in England at the time".

Turning now to the previously long-held belief that the Scots arrived from Ireland, some new theories regarding this are now being expounded. According to Bede (8th century), the Irish Annals and the 'Senchus Fer n'Alban' (History of the Men of Scotland), of which both the latter contain 7th century elements but which were added to in the 10th century, this is the common belief. Since the 1970's several archaeologists noted the contrasts between this time period in Argyll and Ireland. Leslie Alcock stated that if there was such a large-scale population movement from Ireland to Dal Riata, "The Scotti came without luggage." Ewen Campbell has gone even further than this and argues that what manuscript 'evidence' still exists may be dismissed as early spin-doctoring, dynastic propaganda by later Scottish kings. He argues that the prevalence of Goidelic (the Irish form of the Celtic language) in early medieval Argyll owes more to local conservatism than as evidence of population movement from Ireland to Scotland. Brythonic Celtic may be regarded as a later development of Celtic than Goedelic.

Of course, these theories are difficult to prove, particularly in the case of language itself. There is little evidence found thus far for the 4th or 5th centuries in either Argyll or northern parts of Ireland. Most discoveries so far in Ireland date from the late 5th and 6th centuries, and little excavation has takeen place in other sites in Argyll. Only small parts of the sites available have been properly excavated using the most modern techniques, and no doubt much may yet be discovered. The fact remains that although there has obviously been traffic across the 12 miles between Kintyre and Antrim for at least 1500 years, probably from Neolithic times, definite links which prove mass migration into Dal Riata from Ireland have still to be uncovered, and the latest theories are gaining credence.

Dunadd, even after its decline in later times, remained an iconic feature of the landscape. The ogham inscription may have been cut as late as the 9th century, according to Katherine Forsyth, an ogham specialist at Glasgow University. If this proves to be the case, then Dunadd must have continued as a symbolic site after it ceased to be a stronghold and following the Viking settlement of this part of Western Scotland. As late as 1506 Dunadd was used for a series of royal proclamations, indicative of the continuing symbolic importance of this historic hilltop. As Lane points out, "The combination of a rockcut 'basin', two separate carved footprints, the incised outline of a boar or pig, and an inscription in the Irish linear script, known as ogham, cannot be paralleled anywhere in Britain or Ireland."

So the mystery continues to endure. Even the casual visitor to this site cannot fail to be moved by its splendid panoramic outlook, and to view the elements of kingship ceremonies outlined above is something which every Scot interested in their nation's history should take the trouble to experience.

With grateful acknowledgement to the December 2001 issue of the British Archaeology Journal, from whose pages the accumulated research of Alan Lane and Dr Ewan Campbell has been gleaned. Alan Lane is a specialist in Early Medieval Archaeology at the University of Cardiff. Dr Ewan Campbell is lecturer in Early Medieval Archaeology, Glasgow University.

Their collected research has now been published in a book:
Alan Lane & Ewan Campbell - 'Dunadd: An Early Dalriadic Capital', Oxbow, 2001 £45.00.