UP UNTIL 1962, no area of Viking studies excited so much interest and produced so much material on the basis of so little evidence as the story of the Viking discovery of North America. The root of the problem was that the two relevant sagas, the Saga of Eirik the Red and the Saga of the Greenlanders,1 gave clear descriptions – albeit often contradictory – of a series of Viking voyages to North America and of an attempted colonisation of the continent, and yet, despite all their stirring tales of discovery, there was no undisputed archaeological evidence that the Vikings had ever been there at all.
The accounts of the two sagas – collectively known as The Vinland Sagas – differ significantly, but both indicate that sometime around the year 1000 (and probably in 1001 or 1002), Vikings from Greenland sailed west and came across a new coastline, which they explored and, in subsequent years, attempted to settle. Eirik’s Saga mentions only two voyages, while the Saga of the Greenlanders – believed to be the older of the two (and dating from before 1263)2 – gives a more complex account of six exploratory sailings. According to the latter version, the North American mainland was first sighted shortly after 9863 by Bjarni Herjólfsson who, like so many sea-captains in the sagas, was driven off-course while sailing from Iceland to Greenland, where his father had gone to join Eirik the Red’s new colony.4 Bjarni had been on a trading trip to Norway and was totally unaware that his father had emigrated. He decided to follow him, but, understandably a little hazy about the precise sailing route from Iceland to Greenland, he got lost in an extensive bank of fog and, when the mist cleared, found himself sailing along an unknown coastline, forested and punctuated by low hills.
Opting for caution, Bjarni decided against landing and pressed on northwards for two days until the landscape changed, turning flatter, but still tree-lined. His crew urged Bjarni to go ashore and take on fresh provisions of food and timber, but he refused and pushed on for three more days until the ship came to a much more forbidding land of soaring peaks and glaciers. Here, too, Bjarni would not make landfall, saying that the place was good for nothing and not even worth exploring. After a further four days’ sail, their progress assisted by stronger winds, the Icelanders came to a fourth land, which looked a lot more like the descriptions of Greenland that Bjarni had heard in Iceland. By an extraordinary coincidence, he had arrived at Herjolfsnes in Greenland, the site of his father’s farm, where he is said to have spent the rest of his days.
The Saga of the Greenlanders goes on to tell that a number of years later, Eirik the Red’s son Leif went to visit Bjarni and recruited thirty-five crewmen for a voyage in search of the land that Herjólfsson had seen fifteen or so years before. Originally, Leif Eiriksson had intended that his father should lead the expedition, but Eirik had a riding accident just before reaching the ship and returned home to his farm at Brattahlid, possibly a diplomatic manoeuvre to give his blessing to the venture, while allowing his son to act as its leader.
Having taken good note of Bjarni’s account, Leif decided to retrace his steps and so arrived first at the land which his predecessor had visited last. Just as Bjarni had described it, he encountered a landscape of glaciers, mountains and great flat slabs of rock, which he named Helluland (‘slab-land’). Leif then directed his ship southwards and soon came to the flat, forested land that Bjarni had visited second, and here he put a boat ashore, making the earliest known landing by Europeans in the Americas. After a very brief stay, Leif ordered his men back to the ship, and sailed on southwards from what he dubbed Markland (‘wood-land’). The vessel remained out of sight of land for two days and, when the Norsemen did catch a glimpse of a fresh shoreline, it was a far more inviting prospect, with just the right kind of fertile pasture-land to gladden the heart of any Viking intent on a new landnám. The rivers here were also full of salmon, and they decided to stay and build what the saga describes as buðir or ‘booths’, a term associated with the structures built to house chieftains’ followers during the annual things in Iceland (and which implies a temporary shelter).5
During the months Leif and his crew spent in the buðir they were amazed to find that the winter was frost-free and that, on the shortest day, the sun could still be seen from breakfast time to the middle of the afternoon.6 A scouting party was sent out to explore the surrounding countryside, and one of its members, a rather troublesome German named Tyrkir, went missing. When he eventually turned up, he was beside himself with excitement, prattling away in his native tongue, much to the bemusement of his Norse companions. When finally he became more coherent, he reverted to Norse and explained that he had found vines and grapes. From this discovery, Leif is said to have named the area Vinland (or ‘wine-land’).7 For reasons that are not immediately obvious, Leif then had his crew spend some considerable time gathering a cargo of grapes – or possibly raisins – for transportation back to Greenland.
Already at this point the narrative in Eirik’s Saga has departed from the timeline of the Saga of the Greenlanders, although many of the details are similar. Critically, several characters in the latter are left out entirely, so that there is no mention of Bjarni Herjólfsson’s prior sighting of the new land; Leif is said instead to have drifted accidentally onto the coast of Vinland on his return to Greenland from Norway, where he met Olaf Tryggvason and was commanded by the king to convert the Greenlanders to Christianity. Having discovered the three lands – but not, in this version, having named them – Leif returned to Greenland, saving the crew of a shipwrecked vessel on the way (for which he received his later nickname of Leif ‘the Lucky’), and then proceeded to deliver on his promise of Christianising his homeland.8
Eirik’s Saga also entirely misses out the second voyage to Vinland, this time by Thorvald Eiriksson, who borrowed his brother Leif’s ship and set out for Vinland with a crew of thirty. They are said to have stayed the winter in the booths that Leif and his men had built the previous year (which by now had become known as Leifsbuðir, ‘Leif’s booths’). In the spring, Thorvald sent a party out to the west to explore. This group came across signs of human habitation for the first time – a container that the Vikings believed was for drying corn. The following summer the Norsemen discovered a wooded fjord, which Thorvald, still keen to settle the land, decided was the ideal spot on which to build his farm. Unfortunately, it was also the place where the Vikings encountered their first Native Americans, whom they took by surprise while they were resting under their skin-boats. They killed all but one man, who escaped and later returned with reinforcements while the Norsemen, in turn, were asleep. In the ensuing skirmish Thorvald was fatally wounded by an arrow, despite a protective breastwork of branches, which he had ordered erected along the side of the ship. Erik’s Saga reports that the one responsible for Thorvald’s death was not a skraeling (as the Norsemen came to call the local inhabitants), but a ‘monopod’ – a man with a single foot – a character seemingly straight out of Herodotus’s Histories or some fantastic medieval traveller’s tale.9 Having buried their leader at a place that became known as Krossanes (‘Cross Head’), Thorvald’s disheartened followers returned home.
A new sortie to Vinland was planned by Thorstein, another of Erik’s sons, but he fell victim to an epidemic in Iceland and died before he could set out. In a particularly baroque touch, the saga writer has his shade rise from the dead to predict the future of his widow Guðrið, who soon thereafter remarried, this time to a wealthy Icelander named Thorfinn Karlsefni. It is his voyage to Vinland that forms the centrepiece of Erik’s Saga (and to an extent of the Saga of the Greenlanders) and the divergences between the two accounts have caused much subsequent puzzlement to historians trying to analyse the texts.
Guðrið and her new husband set off – the year was probably sometime around 101010 – and settled down at Leifsbuðir (which they had offered to buy, but which Leif, ever with an eye to commercial opportunities, refused to sell to them).11 They were accompanied by sixty men, five women and a quantity of livestock, clearly more of a colonising than a scouting expedition. All went well for a while, as the Norsemen were able to exploit the abundant local stocks of fish and hunting game and, when this proved insufficient, survived off the meat from the carcass of a rorqual (a type of baleen whale) that was fortuitously washed onto the beach. Even the first encounter with the skraelings went well – despite an early scare when they were terrified by the sound of the Vikings’ bull roaring – and the Norsemen received a pile of valuable furs in exchange for a supply of butter, which the skraelings greedily gobbled up.12
Shortly afterwards, Guðrið gave birth to a baby boy, whom she and Karlsefni named Snorri, and who has the honour of being the first recorded child of European descent born in the Americas.13 It was not long, however, before the skraelings became bolder and one of them was killed while trying to steal weapons, which the Vikings had steadfastly refused to trade with them. A revenge attack was inevitable and Karlsefni prepared for it, forcing the issue at a spot where a lake and a forest meant that the Norsemen could not easily be outflanked. Although the skraelings were beaten off and their chieftain killed, Karlsefni realised that, few as they were, the Vikings could not resist such attacks indefinitely. After staying a further winter – which the impossibility of sailing back home through the winter ice constrained them to do – he ordered the colony abandoned and a return to Greenland.
This was not quite the end for Viking Vinland (at least not according to the Saga of the Greenlanders) for there was a final, very ill-fated expedition led by yet another of Eirik the Red’s children, this time his daughter Freydís.14 The joint venture with Finnbogi and Helgi, both Icelanders, ended in disaster when Freydis – in the grand saga tradition of strong-minded women – provoked a quarrel and then engineered their murder. Under such circumstances it is hardly surprising that the survivors returned to Greenland the next spring, carrying another cargo of furs and bearing a secret they all swore never to tell (but, as it appears in the saga account, one of them at least clearly did).
Eirik’s Saga misses out Freydis’s voyage entirely (and has her instead playing a heroic role during Karlsefni’s expedition, when she is said to have carried a sword into battle and bared her breast, terrifying the skraelings so much that they fled). Instead, it assigns as much as possible of the Vinland glory to Karlsefni, attributing to him the naming of Helluland, Markland and Vinland. Its description of his initial explorations also adds a number of place-names to the mix for historians and archaeologists on the hunt for the location of Norse Vinland; having passed Markland – which conforms to the description of it in the Saga of the Greenlanders – they came to a shoreline with long beaches, which they called Furðustrandir (‘Marvel Beaches’, so-called for their extraordinary length). Two scouts, fleet-footed Scots named Haki and Hekja, were sent out to explore, and it is they who are said to have discovered the grapes and what the saga describes as ‘self-sown corn’.
In this version, Karlsefni initially settled down on the interior of a fjord he named Straumfjord, where the party suffered badly from lack of food during the winter. The following spring the Vikings – minus one ship under Thorhall the Hunter – sailed southwards in search of a more promising spot for the colony. After some days’ sail they found a place they named Hóp, where a river ran down to the sea from an inland lake, with fields of the same self-sown wheat that the Scots had found, grapes and a general abundance of fish and game. It all sounds so impossibly idyllic, and the similarities to the fabled Insulae Fortunatorum (‘the Blessed Isles’) described by the seventh-century Spanish encyclopaedist Isidore in his Etymologiae may indicate that the saga writer borrowed a convenient piece of literary topography for his own account.15
On their way to Hóp, the Vikings had encountered a group of skraelings sailing in nine skin-boats, armed with peculiar staves that made a rattling sound when waved. There was little contact between the two groups, but the skraelings returned the following spring in far greater numbers, again swinging their staves – the saga said they did so ‘sunwise’, a peculiarly precise touch – and they traded with the Vikings, eager to swap their furs for lengths of red cloth (as opposed to the butter of the Saga of the Greenlanders). Terrified by the roaring of Karlsefni’s bull, the skraelings fled and returned three weeks later, this time swinging their staves ‘anti-sunwise’, presumably a sign of displeasure or hostile intent. Despite hard fighting, in which the Vikings lost two men but slew a large number of skraelings – and during which Guðrið performed her bare-breasted rallying act – Karlsefni’s side eventually prevailed. The Vikings then retreated north from Hóp to Straumfjord, where they had left around 100 men.16 This would mean that Karlsefni had engaged in his southward exploration to Hóp with just forty men (nine having earlier departed with Thorhall the Hunter). Also, given the small size of the Greenland colony at the time, it shows what a large risk, in demographic terms, the Vikings were taking by their voyage to Vinland. Just as in the Saga of the Greenlanders account, Karlsefni realised that the numbers and hostility of the skraelings made it impossible to settle down in Vinland and he decided to return to Greenland. The departure, though, was delayed until the following spring and was not without incident, as one of the ships, captained by Bjarni Grimólfsson, became worm-eaten and sank soon after setting off. On the way back Karlsefni’s vessel picked up two skraeling boys in Markland, whom the Vikings took back to Greenland and had baptised.17
Yet for all the rich saga tradition of Viking settlement of North America, there was no concrete evidence on the ground. In the absence of an actual Viking site, interest in finding artefacts from Vinland began surprisingly early, fuelled by the publication by the Danish antiquary Carl Christian Rafn, the first secretary of the Royal Society of Northern Antiquaries, of the Antiquitates Americanæ, a compendious collection of all known archaeological and written sources relevant to Vinland.18 Rafn supposed the location of the Vinland of the sagas to lie on the coastline between Massachusetts and Rhode Island. A year before the issuing of his final volume, an exhibition at Christiansborg Palace in Copenhagen presented a range of finds from Greenland and Vinland to the Danish public, including the Kingitorssuaq Runestone, which had been discovered thirteen years previously.19
The thirst for Viking artefacts in the United States was insatiable, to the point where almost any ancient find that could not obviously be attributed to Native Americans was marked down instead as evidence of Norse activity. This included the Dighton Rock, a 40-ton sandstone boulder at Berkley on the Taunton River in Massachusetts, part of whose surface is covered in petroglyphs, which were seized on as evidence of either Phoenician or (according to C. C. Rafn) of Viking exploration. In the end, however, they proved after all to be of Native American provenance.
One curious contribution to the genre came from the poet Henry Wadsworth Longfellow, who embarked on a grand study tour of Europe in 1835–6,20 which took him to London, Stockholm and finally, in September 1835, to Copenhagen. With him he carried letters to various Danish academic luminaries, including Rafn, whom he met. Longfellow described him as ‘an Historian and publisher of Old Icelandic books, which he transcribes from the MSS of which the libraries are full . . . his eyes are always staring wide, so that he looks like the picture of a man who sees a ghost. He is, however, a very friendly, pleasant man, and gives me lessons in Icelandic.’21
Longfellow made good use of his tutorial in the Icelandic sagas and language, later penning several poems on Norse themes. The most famous of them at the time was ‘The Skeleton in Armor’, inspired by the discovery in 1832 of a corpse near Fall River, Massachusetts, which had been interred wearing a suit of crude chain-mail armour and a brass breastplate. Speculation raged as to the identity of the body, with some claiming it might be an ancient Egyptian or Phoenician traveller. Longfellow, however, was in no doubt that what had been unearthed was the burial place of a real Viking warrior, or, as he more poetically put it:
I was a Viking old!
My deeds, though manifold,
No Skald in song has told,
No Saga taught thee!
The skeleton and most of the artefacts associated with it perished in a fire at the Fall River Museum in 1843, although subsequent analysis of the metal tubes found with the body (which had survived the blaze) revealed them to be brass and of a type also found in other burials in Massachusetts that are definitely Native American.22
More durable than the Fall River skeleton – both physically and in the imaginations of those hungry for genuine American Viking artefacts – was another of the inspirations for ‘The Skeleton in Armor’, a round tower at Newport, Rhode Island. No lesser scholar than Rafn had written about it: ‘I am persuaded that all who are familiar with Old-Northern architecture will concur, that this building was erected at a period decidedly not later than the 12th century.’23
The tower, which has experienced later adaptations that make its close dating even more difficult, is first mentioned in documents in 1665, and so is clearly of seventeenth-century date or earlier. Some 28 feet high, constructed of dry stone, and held up by eight squat pillars at its base, it originally had two internal rooms, although the roof of the upper one has long since collapsed. Benedict Arnold, an early colonial governor of Rhode Island (who died in 1678), refers to the building in his will as his ‘stone-built wind mill’, and opponents of a Norse origin for the structure have taken this to mean that he had it built himself.24 Parallels have been drawn with a mill of not dissimilar design at Chesterton in Warwickshire, which Arnold is said to have used as a model. It is unclear, though, whether that building was actually in use as a mill (rather than an observatory) pre-1700, and so it is not a definitive proof of the Chesterton theory.25
There are elements about the Newport Tower’s design that do not entirely chime with the thesis that it was an early colonial mill. The walls are some 3 feet thick, and it has been estimated that it would have taken 450 tons of granite, the timber from four 60-foot-high trees and the labour of sixteen men working for a year to build it.26 Of course it similarly begs a number of questions to presume that a small Norse colony, which (given the size of Greenland’s population) can never have been more than a few hundred strong, could have devoted such a huge amount of effort to building a mill (or even a tower), when they have left no other stone structures in the Americas and nothing remotely similar in Greenland.27
Other seventeenth-century buildings have been excavated in the area surrounding the tower, and radiocarbon dates on the mortar of the tower are clustered around the period 1635–98,28 which would support the Benedict Arnold thesis. Attempts have been made to cite maps produced by the Italian explorer Giovanni da Verrazano as evidence for the existence of a stone structure at Newport at the time of his voyage in 1524. The ‘Norman villa’ on his map is, however, placed far to the north of Rhode Island, in the vicinity of Long Island, and, although it probably indicates some kind of settlement, it is not clear what. Furthermore, if da Verrazano really had found an imposing stone building such as the Tower, it seems strange that he mentioned it only in a cryptic reference on his map and, moreover, displaced it from its true position by 100 miles or more.
The Newport Tower theories, though, refuse to die. In addition to its attribution to the Vikings, it has been supposed variously to have been constructed by early medieval Irish monks, by a pair of fourteenth-century Venetian brothers,29 by sixteenth-century sailors as a lighthouse, by Earl Henry Sinclair of Orkney (imagined by eighteenth-century antiquaries to have ruled over an island kingdom in the Atlantic), or even by a fleet of Chinese explorers in 1421 led by the ‘three-jewelled eunuch’ Admiral Zheng He.30
Enthusiasm for Viking discoveries in America was at first rather tempered by the lack of availability of the sagas in English. The first recorded translation was of a fragment of Håkon’s Saga by the Scot James Johnstone in 1782. The Vinland Sagas themselves were not generally accessible (even in Old Norse) until their publication by Rafn in the Antiquitates Americanæ. They then came to the attention of the English-reading world with the publication in 1841 by North Ludlow Beamish of The Discovery of America by the Northmen, which drew heavily on Rafn’s work. As the nineteenth century went on, increasing numbers of Scandinavians migrated to the United States and Canada31 and as their numbers rose, so too did a feeling that Christopher Columbus had received rather too much credit for the European discovery of America. In 1874, Professor Rasmus Anderson of Wisconsin University encapsulated this feeling in his exploration of the Norse antecedents of the United States entitled America Not Discovered by Columbus.
Resentment at the general acceptance of Columbus as prime discoverer was heightened by the celebrations planned in 1892 to commemorate the 400th anniversary of his landing in the New World. Funds were raised by public subscription in Norway to build an exact replica of the Gokstad ship, in an attempt to upstage the event at Chicago’s Columbian Fair.32 The Viking set sail from Bergen on 29 April 1893 with a crew of eleven, and arrived in Chicago twenty-seven days later, having made its way via New York and the Erie Canal and Great Lakes. As an affirmation of the sea-worthiness of Viking vessels it was astonishing, and it did rather divert attention from the ‘Columbus’ themes of the expo, much as those behind the project had desired.33
With interest in things Norse reaching an unprecedented level, it is no coincidence that the next decade saw a rash of ‘discoveries’ in the United States. The one that has proved most stubbornly persistent is a runestone unearthed by Olaf Ohman, a migrant Swede, on his farm outside Kensington, Minnesota. According to Ohman’s story, which he somewhat modified as time went on, the stone – around 30 inches by 16 inches – was found underneath the root of an aspen tree. Sometime after the stone slab was dug up in 1898, Ohman’s young son noticed strange carvings on it, which the farmer then recognised as runic writing. And so began a century of claim and counter-claim. After a brief period of display in the window of a local bank, a copy of the inscription was sent to Professor O.J. Breda, a philologist at the University of Minnesota, and then in 1899 to another philologist, Professor George Curme at Northwestern University, both of whom dismissed it as a fake. Other eminent scholars, including Professors Gustav Storm and Sophus Bugge34 of Christiania (Oslo) University, also pronounced the runestone to be a fraud.
The stone, though, found its champions, including its most vociferous proponent, the Norwegian-American author Hjalmar Holand, who published the first translation of the runes in 1912. They were said to declare that ‘We are eight Swedes35 and 22 Norwegians on a journey of exploration from Vinland westward. We set traps by two shelters a day’s march north of this stone. One day some of us went fishing. On our return to camp we found 10 men dead, red with blood. May the Blessed Virgin Mary save us from evil. We left 10 men by the sea looking after the ships 14 days’ journey from this island. In the year 1362.’36
Holand sought to link the stone with the expedition of Poul Knudsson, which King Magnus of Denmark authorised in 1355, arguing that the Kensington Runestone was the record of an extension of this effort. Yet, quite apart from the fact that there is no independent evidence that Knudsson actually set out on his journey,37 there are a number of other significant obstacles to accepting the Runestone’s authenticity. The first is the location of the find – several weeks’ journey from the only currently accepted Viking site at L’Anse aux Meadows. It would have taken at least a fortnight to sail down the Red River to enter Lake Winnipeg, followed by a substantial overland trek to reach Kensington, all the while traversing territory occupied by large numbers of Native American tribes, for little discernible gain. Nor have there been any other archaeological traces of Norse activity so far to the south and inland – although absence of evidence does not mean that it might not be found in the future.
More telling is the linguistic data. The Runestone bears one of the longest runic inscriptions found to date anywhere in the Viking world. To have carved such a stone would have taken considerable time, and is not the sort of activity that a small party exploring in hostile territory many days’ journey from a safe haven might have been expected to indulge in. The contrast is instructive with the brevity of the inscriptions scratched on the interior of Maeshowe in Orkney by eleventh-century Vikings who were threatened by nothing more violent than the storm that caused them to take shelter there.38
The language and form of the Kensington Stone runes also indicate that it is either a fake or linguistically unique. The inscription includes an umlauted form of ‘o’, which did not begin to appear in Swedish until the sixteenth century, and runes for ‘j’ and a version of ‘n’, which had fallen into disuse several centuries before 1362.39 The text itself contains various anachronistic usages, such as dags rise for a ‘day’s journey’ instead of the expected form in Old Norse of daghs faerdh. Even more revealing is that several of the aberrant rune types have been linked to forms used in the central Swedish province of Dalarna in the nineteenth century – precisely the area where several of Ohman’s associates around Kensington had originally come from. Ohman himself admitted to having studied runes at school and he had in his possession a copy of a history of Sweden that discussed runic alphabets.40 Several others in Ohman’s circle may also have possessed the requisite knowledge to attempt such a forgery.
The Runestone, therefore, is almost certainly a fake, albeit a sophisticated one, which answered the need for Viking artefacts41 at a time of rising consciousness among Scandinavian-Americans of their position as the heirs of Leif Eiriksson. Even so, it continues to have its steadfast defenders; from February 1948 to February 1949 the Kensington Runestone even formed the centrepiece of an exhibition at the Smithsonian Institution in Washington. Today, it is the prize exhibit at the Runestone Museum in Alexandria, Minnesota, while a giant replica of the stone has been placed in the ‘Kensington Runestone Park’ that is centred on Ohman’s original farmstead.
New evidence appears from time to time on one side or other of the Kensington Runestone debate – such as the assertion in 2005 that other rune forms on the stone (such as a dotted ‘r’) are characteristic of Gotlandic script of the 1300s.42 However, the polemic surrounding the stone is as nothing compared to the fierce debate that has raged around the most famous Viking ‘artefact’ of them all, the Vinland Map. The map first came to light in 1957, when an antiquarian bookseller in New Haven, Connecticut, showed a fifteenth-century manuscript to Thomas Marston, the curator of Medieval and Renaissance Literature at Yale University Library, and his colleague Alexander Victor, the library’s maps curator. Bound up with the Tartar Relation (a hitherto unknown account of an embassy by the Franciscan friar Giovanni de Piano Carpini to the court of the Mongol ruler Ögodei Khan in 1245–47) was an apparently contemporary world map, which included a depiction of Greenland and the eastern seaboard of North America, including the place-name ‘Vinland’.
If genuine, this would be an astonishing find. The earliest known cartographic representation of North America had previously been the Canerio map, drawn by a Genoese map-maker around 1503. The first maps actually naming ‘Vinland’ (or Vinlandia) began to appear only in the late sixteenth century, starting with the rather crude Stefansson map of 1570, which includes labels for ‘Promontorium Vinlandia’ and ‘Skralinge Land’. There had been a number of fifteenth-century maps of the Atlantic that showed a variety of mythical islands bearing names such as Antilia, or the ‘Isle of Brazil’43 (the former appears on a Venetian nautical chart from 1424). There, is, however, no evidence that any of these referred to North America (although they may well display knowledge of Greenland, which appears as ‘Ilha Verda’ – literally meaning ‘green land’ – on a Catalan chart from around 1480).
Although the map that the Yale curators saw covers the entire world (or at least those parts of it supposed to have been known to its purported fifteenth-century creator), its principal interest lay in its depiction of an island in the middle of the Atlantic, which bears the label Viniland Insula. From this the map soon gained the nickname ‘The Vinland Map’, by which it has been known ever since. It also bears a series of inscriptions, which seem to support the idea that it contained information genuinely derived from the voyages to the Americas detailed in The Vinland Sagas. Part of one of them read:
By God’s will, after a long voyage from the island of Greenland to the south toward the most distant part of the western sea, sailing southwards amidst the ice, the companions Bjarni and Leif Eiriksson discovered a new land, extremely fertile, and even having vines, the which island they named Vinland. Eric, legate of the Apostolic See and Bishop of Greenland and the neighbouring regions, came to this vast and very rich land in the name of God the Almighty in the last year of our most blessed father Pascal, remained a long time through both summer and winter, and later returned north-eastward to Greenland.44
Apart from the obvious reference to the saga account of the discoverers of Vinland, the map’s annotation mentions Bishop Eirik Gnuppson Upsi, who is known from the Icelandic Annals to have gone to Vinland in 1121.45 The discrepancy between the year implied in the Vinland map inscription (which would be 1117, the last year of the reign of Pope Pascal) and the date given in the Icelandic records is a minor one compared to the concerns that soon began to be raised about the map’s authenticity. Its provenance was shrouded in uncertainty from the outset – it had been bought by the antiquarian bookseller Laurence Witten in Barcelona from the art dealer Enzo Ferrajoli de Ry, who maintained that it had come from a ‘private collection’, which he was not prepared to identify.46 There were also a number of holes in the manuscript caused by the action of bookworms, which did not match up between the Vinland Map and the Tartar Relation. This particular puzzle seemed to be solved by the serendipitous appearance in April 1958 of another medieval manuscript containing a section of the Speculum Historiale (‘The Mirror of History’) of Vincent of Beauvais. This new find also contained wormholes, which would match those on the Vinland Map and the Tartar Relation, if it were supposed that it had once been bound between them.
Victor bought the map but, although an initial examination at Yale declared it authentic and a book detailing those findings and introducing the map to the public was published in 1965,47 doubts lingered. An examination of the map at the British Museum in 1967 found anomalies in the script, which Marston had confidently declared to be identical to that in the Speculum and Tartar Relation manuscripts and to be a type of ‘bastard book cursive’ typical of Germany, France and the Low Countries around 1415–60.48 More seriously, analysis of the ink on the Vinland Map revealed it to be different in composition from that on the other two documents and, indeed, from most other medieval European manuscripts, having no trace of iron in it.
In 1974 the map was sent for more detailed examination to the McCrone Research Institute in Chicago. This found that the ink used on the manuscript contained traces of a milled-form anatase (a form of titanium dioxide) that indicated it to be of twentieth-century origin. Walter McCrone considered that the anatase was probably of a type patented in 1916 for use in the paint industry, leading him to conclude that the map had been drawn around 1920. Further analysis at University College London by researchers using advanced spectroscopic techniques confirmed the presence of anatase in the map’s ink, in contrast to its absence from the other two manuscripts.49
The map’s cartography also indicates that it is very probably a forgery. It depicts Greenland as an island, a fact that was not known for sure until the twentieth century. Even in 1965, when the map was published, no one had actually circumnavigated it.50 Furthermore, there are indications, such as the identical position of Greenland on the Vinland and Canerio maps, that the composer of the Vinland Map drew on sixteenth-century Portuguese models for his cartography.51
As to the identity of the forger, the pool of people with the necessary knowledge and skill to produce it must have been very small. One prime candidate is the Austrian Jesuit priest Josef Fischer, who wrote an introduction to a book on Norse discoveries in 1902. While researching this in the archive of Wolfegg Castle in Swabia, he had unearthed the Waldseemüller map, which dates to 1507 and bears the first published appearance of the name ‘America’. What the forger’s motive was in producing such a sophisticated fake is unclear, but the production of the Vinland Map might simply have been another attempt – more credible than the Kensington Runestone – to give The Vinland Sagas a grounding in fact.
Not all ‘Viking artefacts’ uncovered in the Americas before the discovery of L’Anse aux Meadows were fake. A genuine find was turned up in 1957 by two archaeologists digging at the Goddard Farm in Naskeag Point, central Maine. Although the Goddard site is a Native American one – and a rich one in material terms – it was a single silver coin embedded in the midden there that proved to be the most significant discovery. At first its importance was not recognised and it was misidentified as a twelfth-century English coin 1100–1135. It was not until 1978 that a British numismatic expert realised that the coin was not English at all, and a re-examination by the National Museum of Norway’s coin curator, Kolbjørn Skaare, found that it was a Norse penny, probably minted between 1065 and 1080. Analysis of the settlement pattern at the site indicated that it was occupied by Native Americans in the period 900–1500 and then abandoned and so, crucially, the coin must have been deposited before the next set of Europeans visited the area in the sixteenth century. The Goddard site was a hub in a series of trading networks that stretched as far north as Labrador, and clearly goods arrived there from some considerable distance. A particular set of thirty stone tools and hundreds of stone flake fragments from Goddard were made of Ramah chert, which is only found in northern Labrador, and these, together with a polished jade knife and a scraper from the Dorset culture, indicate that the settlement traded (though perhaps not directly) with Labrador. Although it is impossible to know exactly how the coin got to Maine, or how long it took to get there, it does show with some certainty that there were still Norsemen in the Americas some time after 1065, half a century after the most likely date of the voyages recounted in the sagas, and that one of them traded or lost the penny.
Almost every piece of information related in the sagas has been picked over numerous times in an attempt to establish the location of the elusive Vinland and, by analogy, Markland and Helluland, an activity that resulted in much feverish (and fruitless) speculation, before a genuine Viking-Age site was discovered in 1961. Most attempts to identify the area in which Vinland lay used incidental references in The Vinland Sagas to try and narrow down the location of the various places mentioned. The description in the Saga of the Greenlanders of the sun being visible between breakfast and the afternoon meal (dagmál) on the shortest day has been interpreted to mean that the sun was up by 9 a.m. and did not set until after 3 p.m., which would imply a latitude of somewhere between 40°N and 50°N, and a location for Vinland anywhere between New Jersey and the Gulf of St Lawrence.52 The references to grapes and self-sown wheat in the sagas have also provided fertile ground for scholarly speculation, with wild grapes no longer occurring to the north of southern Novia Scotia, while the reference to self-sown wheat, if it refers to a form of wild rye,53 would indicate roughly the same area. Conversely, the saga report that Leif’s expedition found salmon in Vinland suggests an area no further to the south than Maine.
The use of the term ‘skraeling’ in the sagas has also been much discussed as a means of narrowing down where the Vikings may have landed and the type of peoples they encountered there. It first appears in the Book of the Icelanders, composed by Ari the Learned, who died in 1158. His discussion of the colonisation of Greenland contains a reference to ‘the same kind of people had lived there as those who inhabit Vinland, and whom the Greenlanders called skraelings’. Some decades later, around 1170–5, the author of the Historia Norwegiae refers to ‘dwarves called skraelings’ who are said to live ‘beyonde the Greenlanders’ and who are said not to bleed when they are wounded.54
At first sight it might seem logical that the natives whom the Norse encountered in North America were Dorset Eskimo or Thule Inuit people. Although the Thule do not seem to have migrated as far south as Greenland’s Western Settlement in the mid-twelfth century, the Norse may have come across them in the Norðsetur, Greenland’s northern hunting grounds, or perhaps some remnants of the Dorset culture that preceded them.55
The Dorset lived in northern Labrador (one possible candidate for Helluland) at least until 1300 and so might have been encountered by the Norsemen there. Further south, too, there may have been Dorset Eskimos in Newfoundland at the time of Leif Eiriksson’s voyage, where they travelled in search of harp seal.56 Yet there are elements in the saga description of the skraelings that do not match what we know of Inuit or Dorset culture. The structure for drying corn, which Thorfinn Karlsefni and his men spotted, is more characteristic of those used by ancestors of Native American groups such as the Beothuk or Micmac than of the Inuit, while another saga reference to the nine men that Karlsefni’s expedition found sleeping under three boats does not sound as if it is referring to an Inuit kayak, as these were far too small to provide such shelter.57
The confirmation that the manufacturers of the Vinland Map, the Kensington Runestone and a host of other false Viking artefacts sought came when the Norwegian author Helge Ingstad announced that he had found a Viking settlement at L’Anse aux Meadows in Newfoundland. Initially Ingstad was greeted with a chorus of scepticism, in part because many scholars had placed Vinland elsewhere, mostly further to the south (with Christian Rafn plumping for Cape Cod, and Gustav Storm opting in 1887 for Nova Scotia).
L’Anse aux Meadows, the site of Ingstad’s discovery, lies in the far north of Newfoundland alongside the Strait of Belle Isle and facing Labrador. A series of test excavations in 1956 by the Danish archaeologist Jørgen Meldgaard at nearby Pistolet Bay had discovered nothing, but when Helge Ingstad arrived in Newfoundland in 1960 to prospect possible sites for Leif’s booths, a local man named George Decker took him to Épaves Bay, where a plain covered in dwarf willow and grass was cut through by the sparkling water of Black Duck Brook. Decker showed him a grassy terrace near the head of the bay from which some small heather-covered mounds protruded.
Ingstad was convinced this might represent the remains of the long-sought-after Norse settlement of Vinland, and in 1961 he and his wife, Anne Stine Ingstad, began excavating. The seven summers that the Ingstads spent digging at L’Anse aux Meadows yielded spectacular results, in sharp contrast to the century of disappointment and forgeries that had preceded it. Their initial digs, and subsequent excavations under the auspices of Parks Canada from 1972, uncovered eight or nine Norse-era buildings constructed of turf. The site was revealed to have been previously occupied by Dorset Eskimo, and in later centuries by Native American groups.58 However, it seems there were no native peoples living in the close vicinity at the time. This (given the bloody clashes between Vikings and skraelings which the sagas recount) would have greatly recommended it as a location for settlement. The two biggest structures – ‘House A’ (around 79 feet long and 15 wide), which lies closest to the brook, and ‘House F’ (around 65 feet long and 49 feet wide with six rooms), which was sited furthest east along the terrace – were both sufficiently large to indicate something rather more than temporary accommodation for a few passing Norsemen.
The most conclusive evidence that L’Anse aux Meadows was indeed a Viking site came from the 125 or so artefacts that the Ingstads found; around 100 of these were iron nails or nail fragments – not a material that was available to the native population, except in the form of meteorite iron from Greenland, which could not have been worked into nails. Another 650 items found by the Parks Canada excavations – many in the boggy ground that surrounded the houses – were made up largely of wooden debris left over from carpentry undertaken by the inhabitants of the site. More compelling yet was the remains of ‘House J’, a single-roomed structure in which the remnants of slag and bog-iron indicated that iron had been smelted there, with the remains of a charcoal kiln found next to it acting as further proof of the house’s role as a smithy. Other, smaller houses on the site appear to have been residential, with finds including iron rivets, parts of wooden vessels, a stone lamp, a ring-headed bronze pin of Viking-Age type and a spindle whorl made of soapstone.
What kind of settlement does the site at L’Anse aux Meadows represent? It is certainly tempting to equate it to Leifsbuðir, which would neatly resolve the question of where Viking Vinland is. Yet this would not answer the complicating question of where the Hóp and Straumfjord of the Saga of the Greenlanders should be supposed to lie. The two largest buildings, each with a private chamber at one end, are suggestive of someone with elite status and may have housed an entire ship’s crew together with its captain. The discovery of the spindle, indicating that weaving – an activity exclusively the domain of women – took place on the site, implies that whoever lived at L’Anse aux Meadows, they probably brought their families with them.
Although this suggests a colonising venture, other aspects of the site indicate a more temporary occupation. None of the houses show any signs of repair – although some of them were eventually destroyed by fire. The roofs of turf buildings generally need maintenance at least once every twenty years, and that the houses at L’Anse aux Meadows did not undergo any such remedial work indicates that their maximum period of occupation must have been around two decades. Further evidence of the short duration of the L’Anse aux Meadows settlement is the lack of a cemetery, which would surely have been needed after any prolonged period of habitation, and the small size of the site’s rubbish middens.
The amount of iron slag and bog-iron that has been found at the site amounts to a total of just 6 or 7 pounds, which is hardly indicative of constant large-scale production over a number of seasons, but is closer to the amount of smelting needed to effect spot repairs on a single vessel. The excavation of a large number of nails and iron rivets, which almost certainly came originally from a ship, is also supportive of this theory. Some of the nails have even been found to have split marks, indicating that they were removed with a chisel from the place where they were originally embedded (presumably in the ship’s timbers).59 In addition, even though some women at least were clearly present at L’Anse aux Meadows, no byres or barns have been found at the site. If the Norsemen came to Newfoundland with their families, then they did not, it seems, come to L’Anse aux Meadows with their livestock.
Tantalisingly, the remains of three butternuts (Juglans cinerea) were also found at the site, a fruit whose range extends only as far north as the St Lawrence Valley around New Brunswick (having more or less the same range as wild grapes), suggesting that the Vikings who occupied L’Anse aux Meadows must have journeyed at least that far south. Initial radiocarbon dates for wood taken from the site cluster around the tenth century, which, given that some of the material must have been driftwood or timber from trees cut down some decades before being deposited in Newfoundland, is not inconsistent with the dates derived from the sagas of 1000–20 for the Vinland voyages. Subsequent radiocarbon analysis has refined this to the period 980–1020,60 a more or less precise match with the saga material.
Instead of being a colonising venture, it seems more likely that L’Anse aux Meadows was a kind of staging-post or gateway to the rest of Vinland. Ice conditions in the seas around Greenland meant that it was not generally possible to sail from there until midsummer and so, in order to avoid being stranded, the return voyage from Vinland had to be completed by October. Considering that reaching landfall at L’Anse aux Meadows would have taken around two weeks from Greenland, then the wisdom of having a secure base to act as a platform from which to explore the rest of ‘Vinland’ is clear. Yet does this mean there might be other, possibly larger Norse sites yet to be discovered along the eastern seaboard of North America? The population that could be housed in the buildings at L’Anse aux Meadows has been estimated at about seventy61 and given that the entire population of the Greenland colony at the time may have been only 400–500,62 then a second such substantial settlement would have meant sparing one-third of the Greenlandic population for this colonising venture, which seems extremely unlikely. One possible candidate for a smaller waystation has, however, emerged. In October 2012, the Canadian archaeologist Patricia Sutherland announced that excavations at Tanfield Valley on the south-eastern coast of Baffin Island had yielded a whetstone with fragments of bronze and smelted iron embedded in it, both metals whose manufacture was unknown to local native peoples, but which could be produced by the Greenland Norse. Taken together with fragments of yarn found at Tanfield, which resembled that woven in Greenland at the time, this has suggested that the buildings found at Baffin may be the remains of a permanent Viking trading post. If this is the case, then Tanfield Valley might represent the second Viking site to be identified in North America, but quite how many Vikings lived there, over what period and whether they did so permanently, or as part of a seasonal trading pattern, is uncertain.
There are signs that at least some of the inhabitants of L’Anse aux Meadows may have come from Iceland. Five fire-starters (flint-like tools used to create sparks for starting fires) made from jasper originating in western Iceland were found by archaeologists, while four similar tools were made of Greenlandic jasper and one from stone taken from central Newfoundland, around 150 miles to the south-east (itself suggesting a range of inland exploration that would otherwise be unknown to us). But even if one or two ships’ crews came from Iceland, it still does not alter the fact that Vinland’s position made it dependent on Greenland, which in the early eleventh century could not possibly provide enough manpower to sustain it.
The occurrence of three different named sites in the sagas (Leifbuðir, Straumfjord and Hóp) does present complications, and should alert us that Vinland is in truth a region and not any particular place. Whether the Straumfjord of the Saga of the Greenlanders is the same as the Leifbuðir of Eirik’s Saga is unclear – a problem which, together with the search for the location of Hóp, has given rise to a substantial literature. Candidates for Straumfjord have included L’Anse aux Meadows itself and the Avalon peninsula of southern Newfoundland, while writers have sought Hóp in places as far afield as St Paul’s Bay on the Gros Morne peninsula, some 120 miles south of L’Anse aux Meadows, and the Hudson River.63
As to how and when the Viking colony in America died out, the picture is rather clearer. The evidence both from the sagas and from L’Anse aux Meadows suggests that the period from Leif’s first landing in Vinland to the final exploratory voyage was around twenty years, and that these were not followed up by further large-scale expeditions. This does not mean that after 1020 the Vikings never visited Vinland (or Helluland or Markland) again; there is evidence, both literary and archaeological, that they did so. An analysis of parts from ships unearthed at Norse ruins in Greenland showed that six out of ten samples were made from larch, which is not found in Greenland, but is native to North America.64 Furthermore, at the Farm Beneath the Sand in Greenland’s Western Settlement,65 fragments of bison hair and fibres from the fur of brown and black bears were found, which can only have come from North America. The Maine penny is also an indication of contacts between Vikings and native traders at least until the mid-eleventh century, although it is conceivable that it was traded all the way from Ellesmere in the sub-Arctic rather than through direct contacts in Vinland.
As far as documented contacts go, we have only the cryptic reference to Bishop Eirik Gnuppson’s journey to Vinland in 1121 and then the Greenlandic ship that arrived in the outer Straumfjord in Iceland in 1347. It was not large – the author of the Icelandic Annals stresses that even the smaller boats in Iceland were bigger – but it is recorded that the vessel was on a trip to Markland when it was blown off-course. The purpose of the expedition is not noted, but it is a fair guess that it was in search of timber. According to The Vinland Sagas, Markland had an abundance of timber, but without it the Greenland Norse otherwise had to rely on driftwood. There were doubtless other, unrecorded voyages from Greenland to North America – the 1347 expedition would have gone unremarked, had the ship not been driven to Iceland – and some of them may have occurred even later. But it is an intriguing thought that 350 years after the first Viking landings in North America and just 150 years before Columbus ‘discovered’ the Americas, it would still have been possible to encounter a Viking landing party engaged in the humdrum activity of chopping down trees for building wood.