Then, it is well known, good men of the old sort, who could not abide to see new laws made and old laws undone, took to their ships and sailed away west. Some of them landed in Iceland; some went to Orkney, and others wandered about the coasts of the Irish Sea to find a home; and wherever they could find shelter and safety, there they settled.
W. G. COLLINGWOOD, Thorstein of the Mere (1895)1
In Thorstein of the Mere, Collingwood gives the account just quoted in the epigraph of this chapter describing how the eponymous hero’s grandfather arrived in Britain – one among the many Norwegian émigrés supposedly fleeing the autocratic tendencies of King Haraldr Hárfagri – Harald Finehair (r. c. 872–c. 932). In contemporary sources, the extent of Harald’s rule and the details of his life are murky and contradictory in the extreme.2 In the saga tradition of the twelfth century and later, by contrast, Harald plays a pivotal role. He is credited as being the first king to unify the Norwegian kingdom and thus providing the catalyst for an expansion of Norwegian emigration across the north Atlantic and into the Irish Sea. The result of this, so the founding mythology runs, was the establishment of independent colonies on Orkney, Shetland, Faroe and Iceland, and down the west coast of Scotland towards the Irish Sea – Viking statelets ruled by (depending on one’s view of the situation) freedom-loving pioneers or belligerent fugitives and outlaws, petty pirate kingdoms of the northern oceans.3 Indeed, Orkneyinga saga describes Harald mounting expeditions to the Northern and Western Isles and the Isle of Man ‘to teach a lesson to certain vikings he could no longer tolerate’, winning territory for the Norwegian crown in the process.
This version of events is most likely reflective (and supportive) of Norwegian territorial claims of c. 1200 when Orkneyinga saga was written, rather than an accurate account of what happened some 300 years earlier. The myth of Harald Finehair presents us with a classic example of how complex processes of emigration, cultural compromise and political consolidation could crystallize in legend around particular resonant figures at a much later date.4 Nevertheless, the extent of Scandinavian influence around the seaways of Britain cannot be disputed. Place-name evidence makes this plain, even if the apparently straightforward patterns we observe mask gaping holes in our understanding of the processes that gave rise to them. Consider, for example, the presence of Irish personal names, Gaelic place-name elements and characteristically Celtic word morphology mixed up with Old Norse in the place-names of north-west England.5 This could signify the presence of Gaelic- or Brittonic-speaking populations that were later overlaid by, or became subject to, or became influenced by Norse-speaking migrants – a possibility apparently strengthened by Old Norse place-names that seem to single out the presence of Gaelic-speakers in the landscape (for example, Ireby, ‘farm of the Irish’, in Cumbria). On the other hand it might indicate that some of the migrants themselves were of mixed Norse–Gaelic heritage. Such identities were indeed developing across this zone of cultural interchange.
Nor can the interconnectedness of this maritime world be denied – a world where the sea was a highway that led between strands and inlets, headlands and islets, storms and tides; a world where political authority could often only be provisional and where local chieftains and disparate peoples forged and broke alliances and moulded new identities that shifted and mutated with the ocean currents, or tore like clouds in the wide western sky. We can see this in the violence and insecurity implied by the Viking Age promontory forts of the Isle of Man or the defended settlement at Llanbedrgoch (Anglesey), where the skeletons of four men of Scandinavian origin and one woman were found buried in a ditch, a reminder of the brittleness of life lived on the margins of a world seething with violent profiteering, slave-trafficking and silver. Hoards of coins, arm-rings and bullion, stowed in the earth on Man and Anglesey, speak of the trade that flowed back and forth across the Irish Sea and up and down Britain’s western coasts, controlled by the people who built long-houses in the Hebrides at Udal on North Uist and Drimore Machair and Bornais on South Uist, at Braaid and Doarlish Cashen on Man. The same people, possessed of a buccaneering spirit of enterprise, bloody-handed entrepreneurs, ploughed the coastal waters naming the landmarks they passed in their own tongue, the essential vocabulary of a seafaring people; headlands and promontories, islands and bays: Aignish (ON egg-nes, ‘ridge headland’, Wester Ross); Skipnes (ON skip-nes, ‘ship headland’, Coll); Sandaig (sand-vík, ‘sandy bay’, Tiree); Gateholm, Grassholm, Priestholm (all Welsh coast), Steep Holm, Flat Holm (both Bristol Channel), all compounds with ON holmr (‘islet’); Anglesey, Bardsey, Ramsey – all with ON-ey (‘island’).
At the end of the eighth century, when the first Viking raids fell on the islands of the north, the territories that comprised northern Britain and the Irish Sea made up a diverse and heterogeneous world. There were the ethnically and linguistically British kingdoms of Wales, and the mysterious kingdom of Alt Clud in the region around the Clyde; there was the Gaelic kingdom of Dál Riata, centred on Argyll and the southern Hebrides, and the Pictish kingdom that dominated the highlands and Northern Isles. Of the Isles of Man and Anglesey very little is known, but both seem to have played an important role in trade and travel around the Irish Sea, with mixed Irish, British and perhaps even English populations, some of which were probably transient.6 Finally, there was the Northumbrian realm, extending to the Forth in the north and west across the Pennines to contest in power with the British-speaking peoples of Wales and Alt Clud.
By the early tenth century, however, after over a century of raiding, disruption, colonization and war, the Pictish kingdom was gone; so too was Alt Clud, at least in its original iteration, its capital of Dumbarton mouldering in ruin on the rock of the Clyde. Dál Riata was no more, its islands overrun, its identity lost. Northumbria was in the hands of Viking kings, and warlords of Scandinavian descent ruled the islands and sea-lanes from the Shetland strands to the Bristol Channel; even the relics of St Columba had been evacuated from Iona ‘to escape the foreigners [Gaill]’.7When, in 920, Edward the Elder raised his claim as overlord of the Scots, the Northumbrians (‘English, Norwegian and Danish’) and the Strathclyde Britons (who presumably included many of the former people of Alt Clud), the north had already become a graveyard of lost realms, the course of its history decisively rerouted.
But, as in England, it was out of this chaos and dissolution that new kingdoms and new identities would ultimately arise. Enduring Norse lordships in the Isle of Man and the Hebrides, and in Orkney, Shetland and Caithness, would last as political entities long into the Middle Ages, their distinctive cultural footprints still visible in the present day. Most enduring of all, however, was the reconfiguration of the broken polities of mainland Scotland, the mosaic reassembled from the jumble of tesserae left behind when the Viking tides relented. The colours were the same, but they were combined to make new shapes, and new pictures: what emerged was a kingdom of Alba – the kingdom of the Scots – a ‘Scot-land’ to mirror the ‘Angle-land’ coagulating to the south. Unlike England, however, the lack of a detailed historical record of the events that triggered and shaped these upheavals (even one as flawed as the Anglo-Saxon Chronicle and its derivatives) means that scholars are forced to rely more heavily on a patchwork of evidence, including the fragmentary written sources that do survive, and the much larger corpus of archaeological evidence which – though it can tell us an enormous amount about how, where and to what degree Scandinavian culture penetrated and altered the pre-existing communities it encountered – is often equivocal on matters of chronology and causation. Nevertheless, it is still possible to paint an impressionistic picture of what happened over the course of the ninth and early tenth centuries in these northern regions, even if it is painted with a fairly broad brush.
Raids around Britain’s northern shores had been occurring from the end of the eighth century onwards, but 839 is the earliest date that we have for an event with significant political ramifications. The Annals of Ulster state that the ‘heathen [gennti] won a battle against the men of Fortriu in which Euganan son of Óengus and Bran son of Óengus and Áed son of Boanta and an almost uncountable number of others fell’.8 Fortriu is the name given in Irish chronicles for the north-eastern part of what is now mainland Scotland, the region surrounding the Moray Firth – the great violent V-shaped gash that opens the Great Glen to the cold swells of the North Sea. It is a Gaelic rendering of the Roman tribal name Verturiones, and remained the heart of the Pictish kingdom.9 The ill-starred Euganan (or, to give him his Pictish spelling, ‘Wen’), was the Pictish king and Bran was his brother; Áed was king of Dál Riata, apparently at this point allied to the Picts and probably the junior partner in an unequal relationship.10 Nobody knows who these particular heathens were, nor where the battle took place. Its aftermath is a void of knowledge, the fate and identity of the victors opaque. We can infer, of course, that it was an unhappy day for the supporters of King Wen et al., but that is all. Such are the limits of the historical record in this period, however, that the fact that it was recorded at all argues strongly for its unusual significance. It may even have been, in the words of one of the principal modern historians of the early medieval north, ‘one of the most decisive and important battles in British history’.11
One reason for this significance is that, like the Viking assaults on Northumbria and East Anglia, the intervention of a raiding army had brought the curtain down on a native dynasty with total, irreversible finality. With the death of Wen and his brother, the line of Wrguist – the family that had risen to the Pictish throne with Onuist I in 732 – was broken. The next king to rule in Pictavia about whom anything is known was of a different ilk. Cinaed son of Alpín (better known to history as Kenneth McAlpin) is popularly imagined as a Gaelic Scot of Dál Riata who conquered the Picts and, in the process, founded the Scottish kingdom. This is a legend, however, with little to support it. Although it is true that he was king in Dál Riata ‘two years before he came to Pictavia’, Cinaed was probably a Pict.12 Nevertheless, his reign did mark the beginning of the increasing political and cultural amalgamation of Picts and Gaels. It also, however, coincided with devastating incursions from land and from sea, from foreigners and from neighbours.
At some point during Cinaed’s reign (839–58), the Britons of Alt Clud struck back at their erstwhile overlords, burning the Pictish settlement at Dunblane. This was not the end of Cinaed’s troubles: also during his reign Vikings ‘wasted Pictavia to Clunie and Dunkeld’ (probably striking from the mouth of the River Tay).13 It was in the west, however, that the most fundamental damage was being done. In 847, ‘the Northmen’, according to the Frankish Annals of St-Bertin, ‘got control of the islands all around Ireland, and stayed there without encountering any resistance from anyone’.14 These territories probably comprised (among others) the islands of Tiree, Mull, Islay and Arran and the Kintyre peninsula (as well as, perhaps, the Isle of Man), places which made up the entire seaward side of the Dál Riatan realm; their loss would have broken whatever connections the Gaelic kingdom still maintained across the Irish Sea, leaving the rump that remained little choice but to move towards ever-closer union with the Picts.
In 865 or 866 a Viking named Olaf (‘Amlaíb’, as rendered in Irish chronicles) arrived in the kingdom of the Picts. He is described in Irish sources as the son of the king of Laithlind, although since nobody knows what or where ‘Laithlind’ was supposed to be (other than being somewhere that Vikings emanated from) the information we have is – once again – not entirely helpful. Olaf had been causing chaos in Ireland since the early 850s, and in the reign of King Constantín (Cinaed’s nephew), he laid waste to the Pictish kingdom, occupying it for ten weeks ‘with his heathens’.15Worse, however, was in store for the resurgent Britons of Alt Clud. In 839 they had got stuck into the Picts by burning Dunbar. In 870, however, their own time had come: ‘Arx Alt Clud a gentilibus fracta est,’ the Welsh annals report, ‘the fortress of Alt Clud was broken by heathens.’16 The Annals of Ulster elaborate, recording that ‘Olaf and Ivar, two kings of the Northmen […] besieged that fortress [Alt Clud] and at the end of four months they destroyed and ransacked it.’17 This was the end for the fortress on Dumbarton Rock; it fell out of use and was mentioned no more. Two years later, Irish chronicles refer – for the first time – to the Britons of Strathclyde, a new political entity already beginning to pupate.
For Olaf’s friend Ivar, the breaking of ancient kingdoms had become something of a speciality. We have met this man before: the previous year he had been in East Anglia with his brother Ubbe, separating King Edmund’s talkative head from the rest of him. It is possible that he dragged a number of Anglian slaves with him when he sailed north, for in the following year Olaf and Ivar ‘returned to Dublin from Britain with two hundred ships, bringing away with them in captivity to Ireland a great prey of English, and Britons and Picts’.18 Olaf died at Constantín’s hands in, probably, 872.19 In the following year, the Irish chronicles record that ‘Ímar [Ivar], king of the Nordmanni [Northmen] of all Ireland and Britain, completed his life.’20 And what a life it had been. Ivar, over the course of his career, had presided over the collapse of Alt Clud, Northumbria and East Anglia, plied his bloody business across Mercia and the kingdoms of Ireland and – probably – all across the other northern realms as well. Whether he was really a ‘king’ in any sense we would now recognize is immaterial – he had done enough to match the achievements of any number of anointed monarchs – and the ‘dynasty of Ivar’, the ‘Uí Ímair’ as they came to be known in Irish sources, were major players across northern and western Britain and Ireland for generations.21
The former industrial shipyards of the Clyde are not the first place one might think to look for the remnants of an early medieval kingdom. Since the decline of Britain’s shipbuilding industry, starting in the 1950s, Govan has consistently been one of the poorest parts of the United Kingdom: as silence fell on the shipyards of Clydebank, social and economic problems multiplied in Glasgow’s brutalist housing estates, exacerbated by disastrous civic planning policies. Today Govan scores below the Glasgow average for employment, educational attainment and life expectancy; it scores more highly for alcohol-related deaths and the proportions (31.5 and 28 per cent respectively in 2016) of people claiming out-of-work benefits and those considered to be ‘income deprived’.22 To put this into perspective, the overall Glasgow figures for poverty and life expectancy are among the worst in the UK.23 ‘It is as if’, declared The Economist in 2012, ‘a malign vapour rises from the Clyde at night and settles in the lungs of sleeping Glaswegians.’24 But it was the river that made Govan the greatest shipbuilding powerhouse of the British Empire – in 1900 Govan built a fifth of all new global shipping25 – and it was the river that turned it into the centre of a durable, and powerful, kingdom of the early medieval world.
I have twice visited Govan Old Church. The first time, as a young PhD student, I was on a field trip organized by the Early Medieval Archaeology Student Symposium (EMASS). I was in Glasgow to give my first ever academic paper, an occasion I had made unnecessarily difficult for myself by consuming prodigious quantities of beer and deep-fried black pudding the night before. I have a dim, and faintly embarrassing, recollection of earnestly lecturing a fellow (Scottish) student on the defects of Scottish nationalism while swaying on a street corner outside a Glaswegian chip shop. My second opportunity to visit the Old Church was undertaken with a group of journalists and others from the British Museum – part of a press trip ahead of the Vikings exhibition of 2014, just before the referendum on Scottish independence.
The church is remarkable for both the antiquity of the setting and the continuity of worship (the current church, dedicated in 1888, is the fourth on this site – there has been a church here since the fifth century), but also for the remarkable collection of early medieval sculpture that is housed within. Such is the quality and quantity of the stonework produced at Govan Old Church that it is now believed likely that the churchyard here was a royal mausoleum for the kingdom of Strathclyde, the successor to the obsolete kingdom of Alt Clud. How this kingdom came into existence, and the ethnic and political dynamics that drove it in its earliest phases are deeply obscure, but there is one clue to what may have been a component of the make-up of this phoenix rising: a rare collection of distinctively Scandinavian-influenced monuments. We had come to see hogbacks.
Hogbacks are most probably gravestones.26 Not the familiar upright sort, those ubiquitous bedheads for the dead, but more like tomb covers – monuments that once lay lengthwise over the buried corpse below them, perhaps with a standing cross at head or feet.27 But, unlike recumbent grave slabs and rectangular tombs, hogbacks are, as their name suggests, curvilinear stone edifices, hump-backed and rounded. They are found across the north of England – mostly in Cumbria and Yorkshire – and in Scotland with a famous cluster (five in total) at Govan Old Church. They are curious objects. Examples from Yorkshire, like the ones at St Thomas’, Brompton (near Northallerton), are the most endearing, with three-dimensional sculpted bears that cuddle on to the ends of the stones. Most of them are decorated with ropes and cables of Viking-style knots and ring-work, chains of loops and braids incised into the stone. Most, too, have characteristic tegulations that pattern the upper part of the sculpture – overlapping tile-shapes that mean that hogbacks often resemble armadillos, nervous armoured beasts which might at any minute scuttle off into the shadows when you’re not looking (indeed, the animalistic qualities of these objects were not lost on one Govan sculptor, who added an eye and flipper-like feet to his carving, turning it into something that resembles some sort of antediluvian amphibian).
Hogbacks at St Thomas’, Brompton, drawn by W. G. Collingwood, 1927
It is the impression of a tiled roof that has led scholars to the understanding that these objects were intended to resemble houses. The Irish and the various peoples of Britain had long traditions of monumental carvings in stone – of crosses in particular – and house-shaped shrines of the type we have encountered before (as Viking loot) were reproduced in stone as tomb covers and as adornments for standing crosses like the early tenth-century Muiredach Cross in Co. Louth, Ireland. But the houses that the hogbacks are modelled on are manifestly unlike these prototypes; they are shingle-roofed and curving like an upturned keel, like the gently bowing roof-lines of the wooden Viking long-house. As we shall see, the idea of the hall as a home for the dead was a powerful concept in the pre-Christian thought-world of the Vikings, and it is certain that such imagery did not vanish from the imagination with any speed. It is, therefore, entirely understandable why Scandinavian settlers on the Clyde would choose this idea to commemorate the dead, even after their conversion to Christianity (which the presence of these stones at an ancient ecclesiastical centre implies). But the result of this impulse was the creation of something unique – a new form of artistic expression that shows how these new communities chose to communicate their sense of themselves to the world around them.
And so, here in Govan, the contradictions of Scottish nationalism feel magnified; few places – from a southerner’s perspective – are so unambiguously ‘Scottish’ as Glasgow. But in the incongruous setting of a strangely severe and massive Victorian neo-gothic church, one can find oneself surrounded by the stonecraft of a people who (if they included a substantial proportion of the surviving Britons of Alt Clud) probably spoke a language closer to Welsh than Gaelic, but among whom dwelt a number of others whose cultural affinities lay with a Scandinavian-infused world, a hybridized Norse–Gaelic culture that was establishing itself all around the Irish Sea. These were people whose own identities were in flux. Hogbacks show us these new identities at the point of their renegotiation. Here were people who saw and embraced traditions of stone carving and house-shaped reliquaries and grave-markers erected in churchyard enclosures and chose these traditions for themselves. But at the same time they were determined to do things in their own way, using their own art-styles, referencing the building styles that were familiar to them. Across northern Britain – from Yorkshire to Cumbria to Clydebank – they were crafting an identity that was distinct: distinct from the communities into which they had intruded, but also from the communities from which they had drifted. Nothing even resembling a hogback exists in Scandinavia.
A classic illustration of what this process meant in cultural terms is found in the transformations that overtook what had hitherto been conservative and ubiquitous accessories. Penannular brooches (open rings that hinge on a long pin, used to secure a cloak at the breast) had for centuries been a typical element of male and female dress in Ireland and the far north and west of Britain. These objects were to play a conspicuous role in defining and reframing identity in the Viking-infused maritime fringes of the west. As tastes and fashions changed with the influx of new people, these brooches swelled to grotesque size as their silver content increased, their bulbous terminals and flat planes chased with decorative interlace and the gaping maws of twisted Viking creatures, or elaborated into thistle-like knobs that announced a new identity taking root. A hoard of such objects, probably gathered together for their weight-value as silver bullion, was recovered from Flusco Pike on Newbiggin Moor near Penrith (Cumbria) over a period of almost two centuries; the first brooch was found in 1795, the second in 1830 and the rest in 1989.28 The largest of them – the brooch found in 1830 – has a pin that is c. 20½ inches in length. There is no way that this can have been a practical object (there are better ways, after all, of fastening a cloak than by using a silver spike weighing a pound and a half). Such objects were intended to communicate the status of their owners, as well as providing a means to carry portable wealth, a vulgar display of power in a society where silver was the crude measure of success.
But perhaps the most direct evidence for this sort of cultural transformation and synthesis can be found in a modification made to an object known, after the place of its finding, as the Hunterston Brooch.29 Hunterston is over the water from Great Cumbrae where the Firth of Clyde, having turned to the south, broadens out into wider waters around the Isle of Arran. It is not far from Govan, but it is closer to the southern Hebrides – the islands that the Annals of St-Bertin described as falling to Viking settlers in the middle of the ninth century. This pre-Viking penannular brooch is, by any standard, an exquisitely wrought piece of jewellery, a Dál Riatan work of art made in around 700, studded with amber gems and wriggling with golden filigree. Any owner of such an object would have fancied himself or herself as quite the business. We do not know for whom or by whom it was originally fashioned, but we do know the name of someone, two centuries after it was made, who owned it. And we know because the name of its owner was written on the back, scratched into the silver: ‘Melbrigda owns this brooch’. Melbrigda is an impeccably Gaelic name and thus not, in itself, surprising. What makes this truly significant is the way in which Melbrigda chose to express his ownership: for the language he used was not Irish, nor even Latin, but Old Norse, and the characters that he scratched were Viking runes.
This, then, was the birth of a new people, a new tick-box on the ethnicity forms of the early tenth century. Call it what you will, Anglo-Danish or Hiberno-Norse, Anglo-Scandinavian, even Cambro-Norse, the implication is the same: British (and Irish) Vikings had become distinct from their Scandinavian counterparts.30 And as the example of the hogbacks demonstrates, the way people buried their dead can tell us a lot about how they – or, more accurately, the people who cared for them in life – imagined their place in their world, their connections and their sense of self.
As the ninth century drew to a close, the successors of Cinaed son of Alpín continued to experience pressure brought to bear by Viking raiding armies. We see it only dimly, in the half-light cast by the sporadic notices of chroniclers who were often writing at some remove from the events they laconically describe. At some point during the eleven years after 899, it is recorded that Pictavia was ‘wasted’ by Vikings, and in 900 Domnall, son of Constantín I, was slain at Dunnottar ‘by the heathens’.31 The notice of his death is a major turning point in British history, for it marks the moment when the kingdom of the Picts slips for ever into the shadows: ‘Domnall son of Constantín,’ the Irish chronicles relay, ‘king of Alba, dies.’32 Thus a new kingdom was born, quietly and without fanfare. A century after the Viking Age had begun, not a single political entity of what is now Scotland remained radically unaltered: Alt Clud, Dál Riata, the kingdom of the Picts were all gone, to be replaced by an ill-defined kingdom of Strathclyde, a new kingdom of Alba and a wide coastal belt of more or less intensive Scandinavian settlement and political dominance. The new king of Alba, Constantín II, would reign for forty years. The political and cultural changes that occurred over this period would make permanent this burgeoning sense of nationhood, sharpening its identity, hardening its borders and bringing it – inevitably – into conflict with the Viking-infused powers to the south.33