Politics and Identity in Early Medieval Scotland

Alex Woolf 28th May 2010 Submitted as part of Ph.D. by Portfolio pacckage Politics and Identity in Early Medieval Scotland: Critical Discussion Introduction The regulations covering the submission of a portfolio of published works in lieu of a dissertation for the degree of Ph. D. require that I submit, in support of the published works, a „critical discussion‟ of them extending to at least ten thousand words. There are no clear guidelines as to what form this discussion should take and I have found no obvious templates upon which to model my text. What follows is therefore very tentative in both form and approach. I hope that the convener and examiners will appreciate the novel situation in which I find myself. The format I have chosen to follow divides the critical discussion into two portions of approximately the same length. In the first part I summarise each of the pieces of work submitted in turn describing the context of its production, its main points of argument and something of its reception and where appropriate its Nachleben. In the second part I take a more discursive approach and talk about how my work generally has fitted in to both the development of the discipline and my own personal development as a scholar and as a university teacher. I outline, in this part, both desiderata for the future of the subject area and some of my own sense of where my work will be taking me from here. I also use this opportunity to discuss some of the particular problems associated with the writing of early medieval Scottish history. Part One: summary of the works submitted. 1 The greater part of this portfolio is made up of my book From Pictland to Alba and, in terms of bulk and theme this might be considered enough to be comparable with a doctoral dissertation. 1 The constraints put on the format of the book by the publishers, however, did not allow for the full depth of referencing which one might expect from a thesis or a low-print-run research monograph and I have included in the portfolio five research papers, each cited in the volume, as exempla of the research which lies behind the production of the monograph.2 Four of these five papers appeared in peer-reviewed journals and the other appeared in the high-quality series of publications entitled Medieval Dublin edited by Sean Duffy,3 the recognised leader in the field of Irish Sea Studies in the central middle ages. Taken together, the work submitted in the portfolio charts my own career as a teacher and researcher within the field of Scottish History which began with my appointment to a lectureship in Celtic and Early Scottish History and Culture at the University of Edinburgh in autumn 1997. My interests up to that date had been on ethnic interaction and language change in the early Middle Ages and on the transition from Iron Age to medieval societies in parts of Europe where the Roman inheritance was comparatively insignificant. To some extent my long-term research goal has been, and continues to be, to attempt to demonstrate the linkage between these two areas of social evolution and ethnolinguistic identity.4 Early Scotland proved to be a perfect laboratory for studying such issues. Unlike the other „home nations‟, England, Ireland and Wales, which each comprised a dominant ethno-linguistic integrity, sharing language and custom, prior to the emergence of anything resembling a modern nation-state, Scotland was constructed from scratch, as it were, in this period. It emerged at the point of convergence of the various Insular ethnolinguistic continua and of other factors, indigenous and external. Because of this particular circumstance, linguistic, ethnic and political frontiers have always been at the centre of early Scottish studies rather than either being accepted as a given, or viewed as a marginal concern, as they have tended to be elsewhere. To what extent different elements of later medieval and early modern Scottish institutions, society and culture 1 From Pictland to Alba, 789-1070, Edinburgh (2007). After the fashion, perhaps, of Preparatory to Anglo-Saxon England: being the collected papers of Frank Merry Stenton, edited by Lady Doris Mary Stenton, Oxford (1970). 3 S. Duffy, ed., Medieval Dublin I-IX, Dublin (2000-2009). 4 For a re e t e a ple of ork i this field see Apartheid a d E o o i s i A glo-“a o E gla d , i N. J. Higham, ed., The Britons in Anglo-Axon England, Woodbridge (2007), pp. 115-129. 2 2 traced their descent from earlier Gaelic, British, English, Pictish or Norse elements has provoked continuous debate. The earliest of my Scottish pieces, included in this portfolio, is „Pictish matriliny reconsidered‟ which appeared in The Innes Review in 1998.5 This piece had actually had its genesis before I moved to Scotland when I was commissioned by Katherine Forsyth, whom I had met at the first Leeds International Medieval Congress, to critique the literature on Pictish succession for a small colloquium (known informally as Pictfest) which she was organising for summer 1995 in Rosemarkie. At this point I was very much an outsider on the Scottish scene with some familiarity with the material merely as a result of my commitment to comparative study. The paper presented at the colloquium, and the resulting article in The Innes Review, compared Pictish succession, as it can be reconstructed from later king-lists and from the obits of Pictish kings in Irish chronicles, to succession practice, explicit or observed, among the other early Insular peoples. The main thrust of the paper was a re-examination of the widely held view that Pictish kingship descended through matrilineal succession. This understanding arose from a brief passage in Bede‟s Historia Ecclesiastica and expanded versions of this narrative found in later Irish literary texts.6 The key phrase in Bede‟s text claims that when the case was in doubt kings would be chosen from the female line rather than the male.7 This phrase in itself leaves considerable doubt as to whether „matrilineal succession‟ was the norm or a desperate measure called for in desperate times. Proponents of the theory of matrilineal succession as the norm have thus been constrained to look elsewhere for supplementary indications that this was the practice. One of the mainstays of the argument in support of the hypothesis that Pictish royal succession passed through the female line had been the absence of evidence of father to son succession prior to the latter part of the eighth century.8 Alfred Smyth had suggested in 1984 that this might reflect not matrilineal succession but the fact that the king-lists represented succession to an Pi tish Matrili re o sidered , The Innes Review 49.2 (1998), 147-167. Bede, Historia Ecclesiastica Gentis Anglorum, I.1, ed. and trs. B. Colgrave and R. A. B. Mynors, Bede’s E lesiasti al History of the English People, Oxford (1969 and 1991), pp. 16-19. For a discussion of the Irish texts see G. Mac Eoi , O the Irish lege d of the origi of the Pi ts , Studia Hibernica 4 (1964), 138-54. 7 Ut ubi res venire in dubium, magis de feminea regum prosapia quam de masculine regem sibi eligerunt.... 8 For the most articulate expressions of the arguments in favour of matrilineal practice amongst the Picts see H. M. Chadwick, Early Scotland: the Picts, the Scots and the Welsh of Southern Scotland, Cambridge (1949), pp. 89-98; I. Henderson, The Picts, London (1967), pp. 31-32; M. O. Anderson, Kings and Kingship in Early Scotland, Edinburgh 1973, pp. 165- ; a d W. D. H. “ellar, Warlords, Hol Me a d atrili eal su essio , Innes Review 36 (1981), 2943. 5 6 3 over-kingship which was open to competing dynasties.9 In my piece I attempted to demonstrate that even in relatively localised provincial kingships across the Insular World, amongst AngloSaxons, Britons and Gaels, father to son succession was the exception rather than the norm in the earlier part of the period. This said I had to concede that the complete absence, so far as we are able to tell from the sources, of fathers and sons acceding to the kingship, even with intervals allowed for brothers, uncles and other collaterals to intervene, remains unusual.10 My paper was produced in parallel with, and is closely related to Thomas Owen Clancy‟s paper presented at the original Rosemarkie Pictfest which subsequently appeared in print under the title „Philosopher King: Nechtan mac Der-Ile‟.11 Both these papers form part of a trend in the scholarship since the 1990s to emphasise the similarity between the Picts and other Insular peoples in contrast to an earlier, and still popular, approach which promoted their „Otherness‟.12 Whilst there were doubtless singular aspects to Pictish culture and practice the assumption that the Picts were fundamentally different and mysterious had become a hindrance to moving forward our understanding of their society on a number of fronts. In my own view my own article is perhaps more important as a comparative study of succession practice in the early Insular World rather than as a definitive analysis of Pictish practices. Some of this material has been revisited recently by Nicholas Evans13 but major questions remain: what precisely is the relationship between the extant king-lists and the chronicle record (might the former have been constructed from the latter)? How should one understand the Latin phrase rex Pictorum, literally the ambiguous „king of Picts‟, which might be understood as either „the king of the Picts‟ or „a Pictish king‟?14 And, perhaps most significantly, why was the average reign-length of Pictish kings in this period, eleven-and-a half-years, so much shorter than that of kings in other parts of the early medieval world? 9 A. P. Smyth, Warlords and Holy Men: Scotland AD 80-1000, London (1984), pp. 36-84. Alasdair Ross, then a mature undergraduate at Aberdeen, was, unbeknownst to me and unknowing of me, orki g o a si ilar h pothesis at the ti e, to e pu lished as Pi tish Matrili ? , Northern Studies 34 (1999), 1122. 11 T. O. Cla , Philosopher Ki g: Ne hta a Der-Ile , Scottish Historical Review 83 (2004), 125-149. 12 I ha e tou hed o this issue to so e e te t i paper, O uist so of Uurguist: tyrannus carnifex or a David for the Pi ts? , i D. Hill a d M. Worthi gton, ed., Æthelbald and Offa: Two Eighth-Century Kings of Mercia, Oxford (2005), pp. 35- . “ee also Katheri e Fors th s Va Ha el le ture, Language in Pictland: the case against non-IndoEuropean Pictish, Utrecht (1997). 13 Ni holas E a s, ‘o al “u essio a d Ki gship a o g the Pi ts , The Innes Review 59.1 (2008), 1-48. 14 Irish chronicles covering the same period regularly use the terms rex Saxonum and rex Brittonum for the rulers of diverse provincial kingships. 10 4 The second paper in the portfolio „The “Moray Question” and the Kingship of Alba in the tenth and eleventh centuries‟, published in The Scottish Historical Review in 2000,15 continues the theme of succession practice and also had its genesis in the summer of 1995. At the International Celtic Congress, held in Edinburgh in that year, I gave a paper in which I tried to argue that the processes of Gaelicization north and south of the Mounth were significantly different because of the variant nature of the Gaelic conquest in the two halves of Pictavia. I was influenced at this time by Ben Hudson‟s analysis of the dynastic history, in which he argued for separated conquest of the two Pictish regions, north and south of the Mounth, by Gaelic conquests directed by the Cenél Loairn and Cenél nGabráin respectively.16 Perhaps as significantly, I was also following the broad-brush approach encouraged by the social theorists in the Department of Archaeology and Prehistory at the University of Sheffield where I was, at that time, a research student. The initial argument had presumed that more complex socio-economic structures had come into place south of the Mount by the later Pictish period which facilitated an elite dominance model of conquest whereas to the north a less complex more „tribal‟ society existed in which there was no room for cultural distance between elites and peasants.17 Whilst I would still stand by the theoretical under-pinning of the two models which I presented at the Congress, I fairly rapidly came to see that the empirical evidence which might have allowed me to apply them to the two halves of Pictland was neither qualitatively nor quantitively up to the job. The work that I subsequently put into the Pictish matriliny paper, discussed above, encouraged me to return to the material relating to the relationship between Moray and the southern portion of the kingdom of Alba (centred on the Tay basin) with new research questions. Hudson‟s argument, which represented the most sophisticated articulation of what had been an emerging consensus, was that the kingdom of Alba and the kingdom of Moray emerged as two separate, if broadly similar, Scottish kingdoms built, respectively, on the wreckage of the kingdoms of the southern and northern Picts. In my Celtic Congress paper I had argued that the relative absence of surviving Brittonic place-names and place-name elements in the region west of the Spey, together with the apparently less sophisticated nature of Moray in the twelfthThe Mora Questio a d the Ki gship of Al a i the te th a d ele e th e turies , Scottish Historical Review 79 (2000), 145-164. 16 B. T. Hudson, Kings of Celtic Scotland, Westport (1994), pp. 127-148. 17 For a e pressio of the ideas u derpi i g this disti tio see o paper Apartheid and Economics in Anglo-“a o E gla d i N. Higha , ed., Britons in Anglo-Saxon England, Woodbridge (2007), pp. 115-129. 15 5 century, pointed to a more complete „genocidal‟ or „barbarian‟ conquest of the North whilst greater evidence for toponymic and institutional continuity might imply something more akin to the elite conquest we associate with groups such as the Normans in the South. It became apparent to me, however, as I examined the data more closely, and came to understand better the processes which contributed to its preservation and recovery, that the evidence was not so clear cut as it had at first have appeared to me. The apparently greater institutional continuity south of the Mounth was largely the product of the chance survival of later medieval muniments and the toponymic evidence could similarly not be so precisely dated, nor were the patterns as clear cut as my original, somewhat cursory, examination had suggested.18 Comparing and contrasting the evidence relating to Moray in the tenth to twelfth centuries with that relating to the provinces to the south had, however, drawn my attention to material that caused me to question the emerging consensus about the relationship between the two regions. Only one individual, Máel Snechta mac Lulaig (†1085), is explicitly described as rí Muireb, „king of Moray‟, in the primary sources. A number of other people who can be identified as his agnatic kinsmen, on the basis of chronicle and pedigree evidence, are variously styled mormaer, „earl‟, or rí, „king‟, in the Irish chronicles but whilst the mormair are consistently linked to Moray those given the royal title, Máel Snechta excepted, are always styled rí Albann, „king of Alba‟. Combining this observation with recent, and then unpublished, work by Dauvit Broun which demonstrated that the surviving pedigree linking Máel Snechta‟s immediate male line with the Cenél Loairn dynasty of the seventh and eighth century Dál Riata was the product of either learned speculation or outright forgery in the late eleventh or, more likely, twelfth century, I was able to offer an alternative explanatory framework for the royal aspirations of Máel Snechta‟s forebears.19 The earliest member of the dynasty to be accorded the royal title in our surviving sources was Máel Snechta‟s great uncle, Findláech, whose obit appeared in the Annals of Ulster, where he is described as rí Albann, and in the Annals of Tigernach, where he is described as mormaer Muireb, under the year 1020. In the published paper I argued that it was unlikely to be a coincidence that the Moray dynasty seems to begin to pursue royal claims in the generation 18 My understanding of the complexity of the toponymic evidence has benefitted enormously from conversations over the years with a number of colleagues foremost amongst whom have been Simon Taylor and David N. Parsons. 19 Brou s ork o the ro al pedigrees e e tuall ade it i to pri t i his o ograph The Irish Identity of the Kingdom of the Scots in the Twelfth and Thirteenth Centuries, Woodbridge (1999). 6 following the disappearance of one of the two lines claiming descent from Cináed mac Ailpín between which the kingship of Alba had alternated throughout the tenth century. I argued that Findláech and his people represented the interests of the client base that had surrounded the line descended from Aed mac Cináeda to whom they may, but need not, have been connected in the female line. This suggestion also allowed the strict alternation of the kingship of Alba between the two Alpinid lines to be scrutinized and the suggestion to be made that by the latter part of the period the descendants of Clann Aeda had some sort of support base north of the Mounth. The paper thus combined an analysis of dynastic politics with a structural understanding of the geography of rulership and concluded by suggesting that the ultimate marginalisation of the lands around the Moray Firth was the product of the expansion of the kingdom south of the Firth of Forth. Prior to this expansion the distribution of resources between the Tay Basin and the Moray Firth was much more even. The research carried out for this paper underlay much of what was to become chapter 6 of From Pictland to Alba. The third paper included in this collection is from Medieval Dublin III and is the least „Scottish‟ of the pieces included here.20 The paper, entitled „Amlaíb Cuarán and the Gael, 941981‟, attempted an in-depth analysis of the career of a single viking leader from the mid-tenth century. It sought to demonstrate that this period sees the beginning of the integration of at least one of the Insular viking dynasties into the native political and cultural milieu. Whilst Amlaíb‟s recorded career is for the most part divided between Northumbria and the Irish midlands, the place of this paper within a portfolio concerned with Scottish History in this period seems justified on a number of grounds. Though not well attested, Amlaíb‟s political ambitions will have extended at least into parts of southern Scotland and perhaps, if his retirement to Iona is an indication, into the Hebrides. Perhaps more significantly his career and in particular his cultural alignment might be taken as exemplary, or at least indicative, of the wider processes of naturalisation and acculturation which were taking place amongst the communities of Scandinavian origin established in Britain and Ireland. Scandinavian Scotland is notoriously badly documented and traditionally narratives have tended to rely rather too heavily on material gleaned from the twelfth- to fourteenth-century Icelandic sagas. In order to speculate in a more informed way about the possible nature and outcomes of viking activity, Scottish historians are 20 A laí Cuará a d the Gael, - , i “. Duff , ed., Medieval Dublin III, Dublin (2002), 34-44. 7 well advised to pay some attention to the much better attested details of events and processes in the Irish Sea world and beyond. The main thrust of the paper was to argue that during the mid-tenth century competing branches of the dynasty of Ímar, which dominated Dublin and the Irish Sea region, were consistently allied with specific competing Irish dynasties and in particular that Amlaíb Cuarán had a close and enduring relationship with Congalach Cnogba of North Brega. The complexities of the relationship between Scandinavian and native leaders and between these men and major ecclesiastical institutions which the relative detail of the Irish chronicle record allows us to chart should caution us against over-simplistic analyses of potentially comparable situations in the less detailed Scottish and English record. Understanding the Irish and English context of the activities of the dynasty of Ímar was also essential for the writing of chapters 4 and 5, and to some extent 7, of From Pictland to Alba. This paper was notable for its stress on factionalism within the dynasty of Ímar, an area, which while not entirely ignored in previous scholarship had tended to be downplayed. The fourth paper on „The Origins and Ancestry of Somerled‟, published in Medieval Scandinavia 15, also examines the issue of Scandinavian Gaelic inter-action but from an altogether different perspective.21 The earliest version of the paper was prepared for the first Rannsachadh na Gàidhlig conference, held at the University of Aberdeen in summer 2000. One of the major problems within the historiography of early medieval Scotland is the extent to which literary histories produced in the fourteenth to seventeenth centuries have informed the construction of the narrative framework. Although the problematical nature of such sources has been well known for some considerable time, many of the factoids sown by their use in the past remain unidentified in the historiography. The present paper dealt with the figure of Gofraid son of Fergus a putatively ninth-century character only attested in pedigrees and chronicles after the mid-fourteenth century. Gofraid, who bears a Norse personal name but whose patronymic is Gaelic, has been viewed by some scholars as a credible figure who might be linked to the emergence of a hybrid Norse/Gaelic cultural grouping in Argyll and the Western Isles. 22 The The origi s a d a estr of “o erled: Gofraid a Fergusa a d the a als of the Four Masters , Medieval Scandinavia 15 (2005), 199-213. 22 For example; A. P. Smyth, Warlords and Holymen, London (1984), pp. 190- ; A. Je i gs, A Histori al “tud of the Gael a d the Norse i Wester “ otla d fro . to , U pu lished Ph.D. dissertatio , U i ersit of 21 8 purpose of my paper on the topic was to demonstrate the unlikeliness of this assumption and to excavate the stages by which Gofraid mac Fergusa had been intruded into the historiography. This endeavour led me into the unfamiliar territory of late medieval and early modern Gaelic genealogy. Methodologically this was a very educational exercise for me which has since stood me in good stead evaluating late sources for early history, an area in which I had previously been very timid. The work presented in this paper benefited greatly from previous scholarship, particularly works by David Sellar,23 Ben Hudson24 and Dauvit Broun25 and I was lucky enough to benefit from the direct help of all of these scholars, together with Dr Katharine Simms of Trinity College Dublin, in gathering materials for it. Working outside my comfort zone taught me the real value of collegiality and allowed me to learn to admit my limitations and ask for help without fear of exposing myself. I still feel less comfortable working with these kinds of evidence than with genuinely early medieval material but have certainly learned what questions to ask and of whom to ask them. The shadow of late medieval chroniclers such as Fordun and Bower lies over much early Scottish history and inevitably informs our reading of the meagre earlier sources just as the sixteenth and seventeenth century Clann Donald histories generate will „o the wisps on the west coast. Some understanding of how these texts were constructed has proved essential in constructing a narrative history of early Scotland. In an a paper written after this one, though published before, I suggested that the links between the house of Somerled and the Airgialla, from amongst whom the legendary Gofraid mac Fergusa was said to have sprung, go back no further than the mid-twelfth century when the Airgialla king Donnchad Ua Cerbaill extended his kingdom to the shores of the Irish Sea and when Somerled‟s family invited Donnchad‟s foundation of Mellifont to establish a daughter house at Saddel in Kintyre.26 The final paper in the portfolio, „Dún Nechtáin, Fortriu and the Geography of the Picts‟, is an example of teaching-led research. A final year undergraduate taking my Special Subject class on Conflict and Cohabitation: northern Britain c.550-750, Gary Stratton, who had a particular interest in military history, wrote an essay for me comparing the relative strengths of the views Edinburgh 1993), pp 148-79; R. A. McDonald, The Ki gdo of the Isles: S otla d’s Wester Sea oard, . -c. 1336, Phantassie (1997), pp. 30-1. 23 W. D. H “ellar, The origi s a d a estr of “o erled , Scottish Historical Review 45 (1966), 123-42. 24 B. T. Hudson, Kings of Celtic Scotland, Westport (1994), pp. 40-42. 25 D. Broun, The Irish identity of the Kingdom of the Scots. 26 The Age of “ea Ki gs: , pp. -109 in D. Omand, ed., The Argyll Book, Edinburgh (2004), at 105. 9 of Leslie Alcock and James Fraser on the context surrounding the battle of Nechtanesmere. 27 In the section of the essay in which he compared the two scholars variant views on precisely where the battle had been fought in the environs of Dunnichen, Angus, he questioned whether any reading of that particular landscape could be said to fit Bede‟s description of the battle as having been fought in “some narrow passes in the midst of inaccessible mountains”. Having recently visited Naughton, in north Fife, with Simon Taylor, I wondered whether the second element of place-name Dunnichen really was the Gaelic personal name „Nechtan‟, which elsewhere had come through Scots as „Naughton‟. It had occurred to me that it might in fact have developed from the personal name „Fechin‟ and have linked the place with the nearby ecclesiastical site of St Vigeans – both sites first appear, unambiguously, in the documentary record in the Arbroath abbey foundation charter issued by King William in 1178.28 Recourse to W. J. Watson‟s History of the Celtic Place-Names of Scotland demonstrated momentarily that this hypothesis was incorrect but it also drew my attention to the existence of a second settlement in Scotland with a name derived from an earlier Dún Nechtain, Dunachton in Badenoch.29 A swift enquiry of James Fraser,30 together with a rereading of Wainwright‟s seminal article on the battle,31 confirmed that this site had not been seriously considered as a possible location for the seventh-century battle in any of the earlier scholarship. Since the geography in the upper Spey valley better suited Gary‟s reading of Bede‟s description I felt that I should at least attempt to eliminate Dunachton as a possible alternative to Dunnichen if I could. The main stumbling block to identifying Dunachton with the site of the Battle of Nechtanesmere was the fact that the victor at the battle, Bridei son of Beli, was identified as king of Fortriu in the Irish chronicles.32 In the nineteenth century W. F. Skene had located the Pictish provincial kingdom of Fortriu in the extreme south of Pictland, in Strathearn and Menteith, and this L. Al o k, The “ite of the Battle of Du i he , Scottish Historical Review 75 (1996), 130-142, J. Fraser, The Battle of Dunnichen, 685, “troud . “ee also F. T. Wai right, Ne hta es ere , Antiquity 22 (1948) 82-97. 28 G. W. S. Barrow, (ed.), Regesta Regum Scottorum II, The Acts of William I, King of Scots, 1165-1214, Edinburgh (1971), no. 197, pp. 250-252. 29 W. J. Watson, The History of the Celtic Place-Names of Scotland, Edinburgh (1926 and 1993), p. 239. 30 Senior Lecturer in Celtic and Early Scottish History at the University of Edinburgh and author of The Battle of Dunnichen, 685, Stroud (2002). 31 F. T. Wai right, Ne hta es ere , Antiquity 22 (1948), 82-97. 32 AT 686 and AU 693.1. 27 10 identification had achieved something of the status of fact in the literature.33 The core of the article which I wrote as a result of these initial musings on Gary Stratton‟s essay comprised the collation and analysis of all the references to Fortriu in the annalistic record, which spread from the mid-seventh century to the early-tenth together with as many of the literary references as I was able to identify. The surprising outcome of this exercise was that the evidence accumulated seemed to indicate that Fortriu had in fact lain north of the Mounth and was to some extent, apparently the precursor of the province of Moray. Whilst I was unable to ascertain whether the battle of Nechtanesmere was more likely to have been fought at Dunachton than at Dunnichen the arguments concerning the location of Fortriu seem to have met relatively widespread acceptance in the wider scholarly community and have forced a re-evaluation of much of what has been understood about early Scottish history.34 Adomnán‟s narrative in Vita Columbae which located the powerful Pictish king Brude son of Maelchú in a fortress „by the mouth of the river Ness‟ can be seen to reflect the political situation in his own time rather than, necessarily, indicating a shift in the focus of Pictish power between the mid-sixth century and the later seventh. Similarly it allows us to postulate some geographical determinism lying behind the development of Dál Riata and Fortriu as the two most prominent kingdoms in northern Britain, lying, as they now appear to have done, at either end of the Great Glen, each, perhaps, controlling a bottleneck in the network of prestige-goods exchange.35 The apparent switch in dominance between the two kingdoms which occurred, arguably, in the decades around AD 700, from the rise of Bridei son of Beli to the reign of Onuist son of Wrguist, might thus be linked to wider changes in the European economy.36 This realignment of Pictish geography looks set to be 33 W. F. Skene, ed., Chronicles of the Picts and Scots and other Early Memorials of Scottish History, Edinburgh (1867), ciii. 34 The first seal of appro al i pri t a e ith the pu li atio of Bar ara Yorke s The Conversion of Britain, 600800, Harlow (2006), a few weeks before the appearance of my article (she had seen a copy of the final draft), at p. 47, and o the ap at p. i . Ni k E a s, i his ‘o al “u essio a d Ki gship a o g the Pi ts , also a epts this identification without question as does James Fraser in From Caledonia to Pictland, Edinburgh (2009). 35 The switch in relative dominance of the two kingdoms in the decades around 700 may reflect the replacement of a system of prestige-goods exchange focused on the Atlantic sea-ways by one based on the North Sea emporia. For these exchange networks in their wider context see the various works of Ewen Ca p ell, espe iall Trade i the Dark Age West: a peripheral a ti it ? , i B. Cra ford, ed., Scotland in Dark Age Britain, St Andrews (1996), pp.7992 and Continental and Mediterranean Imports to Atlantic Britain and Ireland, AD 400-800, York (2007). 36 For the shift in balance of power see now Fraser, From Caledonia to Pictland, pp. 287-319. For the changes to the wider European economic patterns see C. J. Wickham, Framing the Early Middle Ages: Europe and the Mediterranean, 400-800, Oxford (2005), pp. 681-92 and 805-19. 11 my most influential contribution to the historiography to date, and its full ramifications have yet to be taken on board by historians and archaeologists working in the field.37 This brings me to my monograph, From Pictland to Alba, commissioned shortly after my arrival in Edinburgh by John Davey, then of Edinburgh University Press, and by Roger Mason, who had himself been commissioned by John as the general editor of the proposed then volume New Edinburgh History of Scotland. From Pictland to Alba was to be the second volume in the sequence covering the Viking Age (indeed the working title proposed by John and Roger was Celts, Picts and Vikings). It was a departure from earlier comparable series in that the early medieval period was allowed two volumes, one covering the Early Christian phase (initially assigned to Katherine Forsyth but subsequently written by James Fraser when Katherine proved to be over-committed with more important projects)38 and my own volume taking the story forward from the late eighth century to the threshold of the Anglo-Norman era. While the remit of the series was explicitly one of political narrative, my approach differed from that of most of my predecessors in two main ways. Firstly, aware of how problematic any inferences or interpretations made on the basis of the extremely meagre and opaque sources for this period of Scottish history necessarily are, I tried, in every instance, to „show the workings‟ of my methodology by laying out the actual evidence before the reader, wherever possible, before presenting my own preferred reading of it. In many cases I gave the reader several plausible interpretations and left it at that. Secondly, but for the same reasons, I made my use of comparative material from other countries, mainly but not exclusively in early medieval Europe, much more explicit than my predecessors had generally done. Indeed, although it has not yet transpired, I foresaw that much of the negative criticism of my book might focus on the fact that there was too much in it that was not directly „about Scotland‟. Thankfully this approach has so far been vindicated in the published reviews and personal compliments which I have received though it has caused me some anxiety in anticipation. Another innovation in From Pictland to Alba was the decision to begin the book with a lengthy introduction, forty pages, largely comprising a composite image of early Insular social and economic structures constructed on the basis of a wide range of materials few of which derived 37 38 I must confess to some anxiety that this paper may mark the high point of my career altogether. J. Fraser, From Caledonia to Pictland, Edinburgh (2009). 12 from Scotland itself.39 This decision was informed by years of teaching early Scottish history. Much of the scholarship and secondary literature as a whole has focused on specific difficult points of textual interpretation in respect of king lists, annalistic chronicles, hagiography and pseudo-historical texts etc. Whilst engaging with these types of arguments and materials can help students to develop transferable skills in Quellenkritik, I had become aware that many of them were left at a complete loss as to the backdrop against which kings, clerics and chroniclers operated. Scotland, as is well known, has left us with no early law codes, rentals or charters, and certainly no tracts on estate management or the settlement of disputes, so in order to plug this gap I attempted to present a generalised picture of early Insular society, drawn from earlier Irish and English and later Welsh legal material along with some comparative material from Scandinavia and elsewhere. In some ways this part of the book best reflected my own research interests in understanding the social dynamics of northern barbarians (and I use that word unapologetically). In this introduction I tried to emphasise what anthropologists might call the „socially embedded‟ nature of political and cultural action, emphasising the ubiquity of mixed farming, the practical implications of royal itineration and the economic aspects of warfare. It was also my intention to emphasise the place of slavery in the early Insular World, something which many text-books seem to down play. The book concluded with a chapter, „From Pictavia to Albania‟, summarising the long term changes which had occurred in Scotland during the period and attempting to come to grips with some of the processual change which had come to pass and in particular with the apparent disappearance of Pictish identity and language. The interpretative material in this chapter was necessarily more speculative than the main body of the narrative and much of it had been held back until this final part of the book precisely so that it‟s more tendentious arguments did not discredit the narrative of the earlier chapters. There is something fundamentally problematic about having been commissioned to write a book about the political history of a country which had not yet really come into existence during the 39 I trodu tio : La d a d People: orther Britai i the Eighth Ce tur , pp. -40. 13 period covered but most readers seem to have understood the constraints under which the volume was written.40 Part Two: themes and reflections Choosing a title for a submission of this sort is inevitably a somewhat arbitrary decision. It would be dishonest to claim that there was a single over-arching theme that held such a portfolio together and my difficulty in reaching a point at which I felt comfortable submitting for a doctorate under these rubrics before now can largely be attributed to anxiety about whether a critical discussion of this sort could be produced to cover an appropriate collection of my published articles given the range of different subject areas and time periods they have covered. Indeed, I have sometimes been rash enough to admit that my general methodology in „producing outputs‟ is to write the first thing that comes into my head and keep writing until I dry up or run out of time. Nonetheless some recurring themes do seem to emerge from many of my published works and foremost amongst these, it might be argued, is the contention that social-historical processes are not independent of political action. In my previous attempt at gaining a Ph.D., by the more traditional method, I attempted to demonstrate that the „fact‟ that Old English came to be the most widely spoken language in much of early medieval Britain, at the expense of the Celtic and Romance dialects present at the end of the Roman period, must be, in itself, evidence about the nature of society and social interaction at the time. My goal in that piece of work was to try and identify what precisely it was evidence for. Some of that dissertation has now been published in its original or in slightly modified form.41 My two forays into Romano-British studies also touched on this theme and were, to some extent, „preparatory to‟ that dissertation.42 40 For the only negative review to have appeared to date see D. Rollason, in English Historical Review 125 (2010), 670-71. 41 The Brito s: fro ‘o a s to Bar aria s , i H.-W. Goetz, J. Jarnut and W. Pohl, ed., Regna and Gentes: The Relationship between Late Antique and Early Medieval Peoples and Kingdoms in the Transformation of the Roman World, Leiden (2003), 345; Apartheid a d E o o i s i A glo-“a o E gla d , i N. J. Higha , ed., The Britons in Anglo-Saxon England, Woodbridge (2007), 115-129 a d A dialogue of the deaf a d the du : ar haeolog , histor a d philolog , i ). L. De li a d C. N. J. Holas-Clark, ed., Approaching Interdisciplinarity: Archaeology, History and the Study of Early Medieval Britain, c.400-1100, Oxford (2009), 3-9. 42 ‘o a i g the Celts: a seg e tar approa h to a ulturatio , i ‘. Laure e a d J. Berr , ed., Cultural identity in the Roman Empire, London (1998), 111; Adventus, Patrocinium and the urban landscape in Later Roman Britai , i A. Leslie, ed., Theoretical Roman Archaeology and Architecture, Glasgow (1999), 33-47. 14 The period covered by From Pictland to Alba and the articles included in this portfolio is the period in which a polity recognisable as a precursor to modern Scotland first emerges. Questions concerning the inter-dependence of social identity and political action are thus paramount in understanding the processes operating in northern Britain at that time. Traditional narratives which presented the emergence of „Scotland‟ as the unproblematic result of a conquest of the Picts by the Scots of Dál Riata have necessarily come into question. An „elite‟ conquest might have been expected to have as little impact on the cultural identity of the country as that of the Normans did in England or indeed Ireland. On the other hand were we to conceptualise the transformation of Pictland to Alba as one which involved a near complete genocide over a relatively brief period we might have expected the result simply to be an extension of Ireland further eastwards. Much of the historiography of the age of Columba and Adomnán seems quite comfortable with the idea that Dál Riata was simply an extension of Ireland, and this approach might have been extended into later periods.43 For the later middle ages and early modern period there has been a tradition of treating the West Highlands and the Hebrides as a cultural extension of the Irish milieu, largely on the basis of the patronage displayed to Irish poets by kings and nobles in these areas, but this approach has rarely been extended into the core of the kingdom of Scotland east of the Highlands.44 Despite sharing many institutions with Ireland the kingdom of Alba displayed enough distinctive features to be afforded a separate identity. The problem then of the transformation of Pictavia into Albania can only be approached by investigating the specificity of the context of social interactions between groups claiming competing cultural identities and that specificity can only be identified by looking at the specifics of the Scottish context alongside a wider range of historical comparanda. Thus for example the cultural hegemony exercised by Irish literati amongst the dynasties of Norse origin in the Western Isles in the later middle ages might be compared with the spread of Sanskrit kavya amongst the Dravidian-speaking peoples of southern India in the first millennium of the Christian Era, which had relatively minimal impact on vernacular speech in most regions, and To ite o e e a ple, Col á Et hi gha s Church Organisation in Ireland, AD 650 to 1000, Naas (1999), unproblematically includes Iona within its discussions throughout but is otherwise notable for its lack of reference to other earlier Insular churches even when their institutional history might illuminate certain difficult aspects of the Irish evidence. 44 For a recent review and critique of this approach to West Highland history see Wilson M leod s Divided Gaels: Cultural identities in Scotland and Ireland, c.1200-1650, Oxford (2003). 43 15 thus may be quite independent of whatever factors led to the shift in local speech from Norse to Gaelic dialects at an as yet unidentified date.45 At a similar period court culture in eastern Scotland was capable of producing literature in Old French despite there being no evidence to suggest that that language was ever a widely spoken vernacular in the kingdom of Alba. 46 In this light, it can be argued that there may have been two separate processes at play which have become conflated under the catch-all label of Gaelicization. The first being the adoption of Middle Irish as a cosmopolitan language, comparable with Sanskrit or Old French in the examples noted above, and secondly the emergence of a vernacular in eastern Scotland that appears, from its very imperfect record, to have been a dialect of Gaelic. Each of these processes must have had their specific political, social and economic correlates and causes. Writing the early history of Scotland has, as noted above, particular problems. The historiographical inheritance of a grand narrative drawn from late and unreliable sources constrained, and compounded by the paucity of textual sources emanating from the period itself is further complicated by the mutually exclusive distribution of high quality archaeological data and contemporary textual material. The one area where this does not apply at present would seem to be in Early-Christian Argyll where excavation has taken place at Iona, the monastery at which most of our written sources were produced, and where a number of sites which appear in the chronicle record have been available for investigation, such as Dunadd and Dunollie.47 Even here in Argyll, however, our understanding of how such central sites mapped on to the wider settlement pattern and how these settlements, in their turn, exploited their environment is very poorly understood when compared to southern England or the Irish midlands in the same period. For “a skritizatio see “heldo Pollo k s The Language of the Gods in the World of Men: Sanskrit, Culture and Power in Pre-Modern India, Berkeley (2006). For Norse in the Western Isles see now Alan Macniven, The Norse in Islay: A settlement historical case-study for medieval Scandinavian activity in western maritime Scotland, Edinburgh (2006), unpublished PhD thesis. 46 D. D. R. Owen, William the Lion, 1143-1214: kingship and culture, East Linton (1997). 47 For an exemplary site report see A. Lane and E. Campbell, Dunadd: an Early Dalriadic Capital, Oxford (2000), unfortunately no comparable volume bringing together the various archaeological campaigns on Iona has yet been produ ed ut i the i teri Jerr O “ulli a s sur e Io a: ar haeologi al i estigatio s, , i D. Brou and T. O. Clancy, ed., Spes Scotorum: Hope of the Scots, Saint Columba, Iona and Scotland, Edinburgh (1999), pp. 215-244, taken together ith Fi ar M Cor i k s ore dis ursi e Io a: the ar haeolog of the earl o aster i C. Bourke, ed., Studies in the Cult of Saint Columba, Dublin (1997), serve to give a sense of the potential. 45 16 Evidence for activity at even these sites in Dál Riata seems to tail off in the eighth or ninth century.48 Elsewhere in Scotland the best understood archaeological landscapes of the early medieval period lie in the Northern Isles of Orkney and Shetland, from and about which there is almost no reliable textual information. This makes comparison of these island groups with those areas better attested textually but archaeologically comparatively poverty stricken, such as the Pictish heartlands around the Moray Firth and the Tay Basin and the British territories of the Southwest, extremely difficult. Leslie Alcock recognised this when he wrote in the Preface to his Kings and Warriors, Craftsmen and Priests “[t]he omission of the Outer Hebrides and the Northern Isles is the result of deliberation, not oversight. Visits going back to 1947 have convinced me of the rich variety of Early Medieval structures and artefacts in both areas; but it is a variety which (with rare exceptions mentioned in the text) differs in various degrees from the mainland and the Inner Hebrides: to have included them would have made a necessarily complicated account even more so.”49 The rich harvests which the islands have been able to ensure archaeologists has led to the concentration of field projects, and particularly university training digs, in these areas thus compounding the problem of variable rates of survival, between the islands and the mainland, with even more variable rates of recovery and recording. This situation of disjuncture between the archaeological record and the textual record, which comprises largely the view from Iona and Jarrow in the earlier part of the period and that from the Irish East Midlands and Durham in the later part, presents a major dilemma for the historian of Scotland. The complaint of the most recent reviewer of From Pictland to Alba, that the book “does convey the impression that the „new‟ history of Scotland is really about kings and dates” reflects a real problem.50 Geoffrey Barrow‟s innovative work on shires and thanages has raised interesting possibilities about the early social and economic history of the core parts of the Scottish kingdom but while he has clearly identified some interesting, and apparently wellentrenched, features of territorial organisation and dues present by the middle of the twelfth 48 Lane and Campbell, Dunadd, 93-97. Alcock, Kings and Warriors, p. xiv. 50 Rollason, English historical Review, 671. 49 17 century it is very hard, if not impossible to say how much older than that they are. 51 The claims for the antiquity and systematic nature of these features made by Alexander Grant are hard to substantiate or to delimit geographically.52 The recent reanalysis of medieval Irish land organisation undertaken by Paul MacCotter presents a possible opportunity for re-assessing the Scottish evidence alongside recent work on that evidence, particularly Dauvit Broun‟s contribution to the edited volume on the Book of Deer.53 Questions as to whether the administrative landscape of twelfth-century Scotland is a relic of Pictish times (and thus pretenth century) as opposed to a relatively recent innovation, in keeping with recent thoughts on the development of manorialism in England,54 remains an open question. Extensive archaeological work in the eastern lowlands combined with toponymic surveys may unearth new evidence relevant to such questions but such projects as exist in these areas at present are still in their early stages.55 My own training, in the Department of Archaeology at the University of Sheffield, encouraged by teachers such as John Collis, Richard Hodges and John Moreland and fellow postgraduates such as Mary-Ann Owoc, Ross Samson56 and Ron Ross, led me to engage with the big themes of social and economic history, the Fall of the Western Empire, World Trade Systems, the Rise of Feudalism, and with the work of scholars such as Klavs Randsborg and Chris Wickham. This approach was further fostered by the then very theoretical Department of Archaeology at the University of Wales in Lampeter where I held my first lectureship. On moving to Scotland and taking up my post at the University of Edinburgh, a post shared between the Departments of Celtic and of Scottish History, I rapidly learned that I had to learn a new G. W. “. Barro , Pre-feudal “ otla d: “hires a d Tha es , i his Kingdom of the Scots, London (1973), pp. 7-68. A. Gra t, Tha es a d Tha ages, fro the Ele e th to the Fourtee th Ce turies , i A. Gra t a d K. “tri ger, ed., Medieval Scotland: Crown, Lordship ad Community. Essays presented to G. W. S. Barrow, Edinburgh (1993), pp.3981, and idem The Co stru tio of the Earl “ ottish “tate , i J. ‘. Maddi ott a d D. M. Palliser, ed., The Medieval State; essays presented to James Campbell, London (2000), pp. 47-72. 53 P. MacCotter, Medieval Ireland: Territorial, Political and Economic Divisions, Du li a d D. Brou , The property records in the Book of Deer as a source for earl “ ottish so iet , i K. Fors th, ed., Studies in the Book of Deer, Du li . Alasdair ‘oss s lo g-awaited monograph on the dabhach may also be of interest when it appears. 54 See for example R. Faith, The English Peasantry and the Growth of Lordship, London (1999). 55 The SERF archaeological project in Strathearn and the AHRC funded History of Gaelic in Scotland project, both run under the auspices of the University of Glasgow, provide some hope that progress is finally being made in these areas. 56 Ross Samson was technically a Glasgow Ph.D. student working with Alcock but for personal reasons he moved to Sheffield during the period in which he was writing up his thesis and was a great influence upon my thinking at that time. 51 52 18 practice of discourse. The earliest of the papers submitted in this portfolio, „Pictish matriliny reconsidered‟, was peer-reviewed for The Innes Review by David Dumville and the typescript was returned to me with more red marks on it than anything I had ever received back from a teacher in my student days. My first reactions were mortification and self doubt, but after collecting myself together I applied myself to the task of interpreting Dumville‟s copious and at times quite acerbic comments and annotations. So began a third period of studentship in which I attempted to teach myself to think and write in the forensic manner typified by Dumville, his Cambridge-trained ASNaC57 colleagues, and other Early-Insular, and particularly Celtic, historians who had emerged in the wake of Kathleen Hughes and her peers in the course of the 1970s and early „80s. This brand of scholarship was very different from that in which I was originally trained and at times the two schools have seemed antithetical. In Sheffield I had been taught to test hypotheses, in lingua Asnacorum „hypothesis‟ seemed almost a dirty word. Dumville and his colleagues seemed to use Occam‟s Razor as an offensive weapon and any suggestion that one was applying „generally applicable rules‟ after the fashion of the New Archaeology was treated with the utmost contempt. „Plausible explanatory models‟ gave way to established readings and relationships between texts. One began to wonder if it was legitimate at all to speculate about a past world outside of the handful of surviving manuscripts (let alone the texts themselves, which seemed to be slipping into „hypothesistacy‟ [sic]). My Scottish years thus became a training period in which I had to attempt to impose an almost Zen-like discipline to my writing. I don‟t think I ever really achieved this, but am not certain if that is something to be ashamed of or to be proud of. When talking to students I tell them that the skill of writing and reading about history is to be able to distinguish the hard forensic evidence from the necessarily more imaginative and rhetorical discourse that surrounds it. An historian must be both the Crime Scene Investigator and the Counsel for the Prosecution. He or she must be able to marshal and present the evidence without contaminating it, yet at the same time be capable of producing a coherent and persuasive summing up. The ideal case-notes should allow subsequent re-investigation to separate the rhetoric from the forensic evidence. The difficult art of the procedure is producing a text that presents the detail of Quellenkritik while at the same time allowing itself to produce imaginative 57 The Department of Anglo-Saxon, Norse and Celtic and the University of Cambridge. 19 solutions to problems the sources alone cannot empirically solve. This can only be done through reference to comparative material from other historical and anthropological contexts. Successfully engaging with these two objectives in parallel is a challenge to which we have, perhaps, only recently reached the point from which we can rise to it in Early Insular Studies, though perhaps this is what Chris Wickham has been trying to do for the Mediterranean for some time.58 The pieces which I have presented in this portfolio represent distinct stages along my own path of training and learning. „Pictish matriliny‟ and „Gofraid mac Fergusa‟ are to some extent exercises in perfecting my grasp of lingua Asnacorum. „Dún Nechtain, Fortriu‟ and the book, From Pictland to Alba are my most successful attempts at wedding the two approaches and the others stand somewhere in between. In my future work I hope to be able to return to some of the bigger themes that interested me in my early career but to take back with me many of the skills and perspectives that I have learned in the course of my journey into early Scottish history. In particular I am interested in exploring the shifting balance of political and cultural power in the Insular World in the course of the early medieval period and to explore the methodologies of World Systems theory in this relatively microcosmic environment. What difference did the shift of international exotic imports from the western seaways to the North Sea make to factors other than simply their exchange? How and why did the Irish Church so outgrow its mother in Britain that by the eleventh century even Welsh hagiographers began to imagine that their founding fathers must have been trained in Ireland when in fact it was they who propped up and supported the infant Irish Church? Why was Bernicia the dominant force in Britain in the early to mid-seventh century when in almost all other periods of history it has been a relatively marginal part of the island? Were the causal factors behind this „anomaly‟ linked to the equally anomalous contemporary dominance in Ireland of the Donegal based Cenél Conaill? Why did coinage never spread beyond the English-speaking regions in the pre-Norman period?59 To move forward in these areas one has to abandon the paradigm of studying the Early Insular History within the straightjacket of the modern „home nations‟. From Pictland to Alba might have contributed more to our understanding of Scottish History had it not been a History 58 59 C. J. Wickham, Land and Power, London (1994). Dublin and Man provide a linked and very late exception to this rule. 20 of Scotland. Series such as the New Edinburgh History of Scotland are, however, an inevitability and I have tried to use that platform to prepare the reader for a wider scope. The unevenness of the data and the particular need in Early Scottish History to rely upon late sources and comparative examples presents a particular challenge which I hope I have risen to. Inevitably all the pieces which I have submitted for this portfolio are now in need of revision. New work, unavailable at the time of writing, has now appeared in print and my own understanding and experience continue to change and, I hope, expand. I can only hope that they have contributed both to the debate and to the dissemination of new work, that of others as much as my own, to a wider audience both among the general public and in university departments around the world.60 60 I would like to thank my colleagues at St Andrews for support in putting together this portfolio and application, particularly the current Head of School, Professor Andrew Pettigree, for smoothing the way through the bureaucracy and his predecessor, Professor John Hudson, for giving me lots of practical advice. My early medieval colleagues Dr Simon Maclean, Dr James Palmer, Dr Elina Screen and Dr Helena Carr have also been most supportive. Dr Fiona Edmonds of the Department of Anglo-Saxon, Norse and Celtic at the University of Cambridge read a earl draft of this riti al dis ussio a d pro ided er useful o e ts. I ould also like to tha k EUP s Esme Watson (now moved on) for providing me with copies of From Pictland to Alba, for the submission, at no charge (she has since left the employ Edinburgh University press but I do not believe that there is a connection between this fact and her generosity). My wider intellectual debts within the field of Early Scottish History are acknowledged in the front of From Pictland to Alba. 21