Northmen, The Viking Saga 793-1241 AD

Chapter 1: Thule, Nydam and Gamla Uppsala. The origin of the Vikings

The Vikings did not spring into life fully formed at the end of the eighth century, even if it may have seemed that way to their startled and appalled victims. In reality, the breaking out of Viking raiding was the consequence of centuries of social and political evolution, which had created in Scandinavia a violent and predatory society. If these developments passed largely unnoticed in the rest of Europe it was only partly because of Scandinavia’s remoteness. In the literate Greco-Roman world of Classical antiquity, a deep cultural prejudice against the ‘barbarian’ meant that the peoples of northern Europe were little studied and rarely written about. This prejudice sur​vived into the Christian era, when Scandinavians were doubly damned for being pagans as well as barbarians. As Scandinavians them​selves did not develop a fully literate culture until after their conversion to Christianity at the end of the Viking Age, contemporary written evidence of Scandinavia’s historical development before the Viking Age is extremely scarce: Scandinavia’s prehistoric period was a long one.

Scandinavia’s earliest known literate visitor was the Greek explorer Pytheas of Massalia, who made a long voyage in the northern seas in the years around 320 BC. On his return home Pytheas wrote an account of his travels entitled On the Oceans. Unfortunately, this was lost in antiquity and is known today only from extracts preserved in the works of later Greek and Roman geographers. These show Pytheas to have been a scientifically minded traveller who estimated the latitude of the places he visited on his journey by measuring the height of the sun at noon and by the length of the days. In his own time, however, Pytheas was believed by many to have invented the whole story, so fantastic did it seem.

Pytheas’ home port of Massalia (now Marseilles), was founded in 600 BC by settlers from the Greek city of Phocea. The sheltered natural harbour was an obvious attraction and it was close to the valley of the river Rhône, which at that time was a major trade route bringing British tin and Baltic amber to the Mediterranean. The Phoceans had the reputation of being the most adventurous Greek seafarers. Soon after founding Massalia they had sailed through the fabled Pillars of Hercules – the Straits of Gibraltar – into the Atlantic Ocean to trade with the mineral-rich Iberian kingdom of Tartessos. One of them, Midacritus, was rumoured to have gone even further and brought back tin from Britain. However, around 500 BC the Phoceans were shut out of the Atlantic when the powerful North African city of Carthage gained control of the Pillars of Hercules. Carthage lived by trade and did not welcome foreign merchants in its sphere of influence. Pytheas’ expedition, therefore, was probably commercial, to seek out new trade routes for Massalia in areas not controlled by Carthage.

When he set out, Pytheas probably bypassed hostile Carthaginian territory by travelling overland from Massalia to the Bay of Biscay and there chartered a ship from one of the local Celtic tribes to take him on to Britain. The Veneti of Brittany were particularly well-known for building sturdy wooden sailing ships with which they carried on a brisk trade in tin with Britain. Pytheas landed at Belerion – Land’s End – and travelled the whole length of Britain. Everything the Greeks knew about Britain up until then was based on hearsay. For the first time Pytheas added some reliable facts. His estimate of Britain’s circumference as around 40,000 stades, approximately 4,500 miles, is remarkably close to the actual distance of around 4,700 miles. The next stage of Pytheas’ journey took him far beyond the edge of the known world. Setting out from an unidentified island off Britain’s north coast, Pytheas sailed north for six days until he reached the land he called Thule. Pytheas’ observation that the sun was below the horizon for only two or three hours at midsummer fixes Thule’s latitude at about 64° north. However, Pytheas had no means of calculating longitude.There is no doubt that Thule was a land in the far north but where exactly? The uncertainty of its location has made Thule more a symbol of ultimate hyperborean remoteness than a real place.

Iceland or even Greenland have been proposed as possible locations for Thule but, as this comment on Pytheas’ account by the Greek geographer Strabo (c. 63/64 BC–AD 24) makes clear, Thule was inhabited by farming peoples:

‘[Pytheas] might possibly seem to have made adequate use of the facts as regards the people who live close to the frozen zone, when he says that, the people live on millet and other herbs, and on fruits and roots; and where there are grain and honey, the people get their beverage, also, from them. As for the grain, he says, since they have no pure sunshine, that they pound it out in large storehouses, after first gathering in the ears thither; for the threshing floors become useless because of this lack of sunshine and because of the rains.’ The Geography of Strabo, bk IV 5.5 (Loeb Classics, 1917).

Greenland was inhabited only by early Inuit hunter-gatherers at this time, and Iceland by no one at all, so neither could have been Pytheas’ Thule. This means that Pytheas’ landfall must have been somewhere around Trondheim Fjord on Norway’s west coast. Despite its northerly latitude, the Norwegian coast has a relatively mild climate thanks to the influence of the warm Atlantic Gulf Stream current, which makes farming possible even north of the Arctic Circle. Trondheim Fjord’s sheltered south and east shores have some of Norway’s most fertile soils and farmers were settled on them as early as 2800 BC. Pytheas sailed still further north and his observations make it clear that he crossed the Arctic Circle. He also claimed that a day’s sail north of Thule was the Frozen Sea, though it is not clear if he actually saw this for himself or merely reported what other seafarers had told him.

Following his visit to Thule, Pytheas headed south to explore the Baltic, which he must have reached via the Skagerrak, the Kattegat and one of the passages through the Danish islands. Pytheas visited the unidentified island of Abalus from whose shores amber was collected. A translucent fossil resin with a fiery colour, amber had been prized in the Mediterranean world for thousands of years, not only because of its beauty but because of its seemingly magical electrostatic properties: called electrum by the Greeks, amber has given us the word ‘electricity’. The origins of amber were the subject of several myths but Pytheas was the first to establish its true source. Abalus has been identified as the Danish islands of Sjælland or Bornholm, the Samland peninsula near Kaliningrad (the richest source of amber today), and the North Sea island of Heligoland. Heligoland seems unlikely as Pytheas says that Abalus was a day’s sail from the lands of the Goths, who at that time lived on the Baltic coast. Pytheas explored the Baltic at least as far east as the Vistula, before returning to Massalia by a round-about route, following the River Tanais (Don) south to the Black Sea, where he would have had little difficulty finding a ship to take him home at one of the many Greek colonies there.

Brief though it is, Strabo’s extract from Pytheas, quoted above, is the earliest eyewitness account of the lives of the Vikings’ ancestors that we have, but beyond telling us that they enjoyed drinking mead and ale and had to dry their grain indoors, it doesn’t tell us much. If Pytheas did have more to say about the languages, customs and social institutions of the people of Thule, his readers did not think it worth preserving. To learn anything meaningful about the Vikings’ earliest ancestors we have to turn to archaeology.

The ancestors of the Vikings were most likely Stone Age farmers who began to colonise Scandinavia around 6,000 years ago, displacing or assimilating hunter-gatherers whose own ancestors had arrived at the end of the last Ice Age some 6,000 years earlier. These pioneer farmers belonged to the Corded Ware Culture (named for the way its pottery was decorated by pressing twisted cords into the wet clay), which originated on the north German plain. Although the connection will probably never be proven beyond doubt, this culture is associated with the early spread of the Germanic, Slavic and Baltic languages. If true, the settlers probably already spoke an early form of the modern Scandinavian languages, which all belong, with modern German, English, Dutch and Frisian, to the Germanic language family. The close genetic similarity between modern Danes, Norwegians and Swedes on the one hand, and modern north Germans on the other, strengthens rather than weakens this conclusion. No convincing evidence exists for any further substantial migration into Scandinavia before the later twentieth century. Scandinavia would make its mark on history as an exporter of population.

About 1800 BC bronze artefacts began to appear in Scandinavia. Bronze is an alloy of copper and tin, neither of which were available in Scandinavia at that time (Sweden’s rich copper reserves were not discovered until the Middle Ages). Scandinavians were, therefore, completely dependent on imported bronze. At first, finished bronze artefacts were imported, but after Scandinavian smiths mastered the skills of bronze casting they probably relied on imported bronze ingots, which were widely traded around Europe. This was the period when amber first began to be traded widely in Europe, so it was probably the commodity the early Scandinavians used to pay for their bronze. The high value placed on amber ensured that bronze was never in short supply in the north. The increase in long distance trade helped stimulate the development of a more hierarchical society, as demonstrated by the appearance of small numbers of richly furnished elite burials marked by earth barrows. Stone suitable for toolmaking is widespread but bronze’s exotic origins, and the specialised skills needed to make and cast it, allowed its distribution to be monopolised by a small elite whose power and status were thereby greatly enhanced. In the more fertile areas of southern Scandinavia, farms began to cluster in small villages. The typical dwelling was a longhouse – a long narrow building in which the family and its livestock lived under one roof, the people at one end, the animals in a byre at the other. The livestock helped keep the house warm in winter. The presence of a single large dwelling among otherwise smaller dwellings indicates that villages were dominated by a single headman or chief. In Norway and much of Sweden, dispersed settlement remained the norm until the end of the Viking Age.

Bronze tools were a great advance on stone tools but bronze was even more important for making status symbols, such as weapons, jewellery, razors, horned helmets, lurs (horns) and fittings for wheeled vehicles, and cult objects such as the magnificent ‘Sun Chariot’ from Trundholm in Denmark, a model of a four-wheeled horse-drawn wagon carrying a brilliantly gilded sun disc. The horned helmets, misinterpreted by antiquarians in the nineteenth century, helped give rise to the romantic, but mistaken, belief that Vikings wore horned helmets. Sadly, Vikings never wore horned helmets. The Bronze Age elite probably also achieved close control over the use and distribution of amber. Amber beads and other ornaments are common offerings in Stone Age graves in Scandinavia, but they are virtually absent from those of the Bronze Age. Amber is so light that it floats in salt water – another property that made it remarkable to the ancients (it also burns) – and is washed up on beaches around the North Sea and the Baltic for anyone to pick up. However, it appears that the elite claimed ownership of all amber washed up in their territories and could prevent others using it so they could prioritise its use for export.

It is during the Bronze Age (c. 1800 BC – c. 500 BC), that the importance of seafaring in Scandinavia first becomes obvious. No Bronze Age ships have yet been found in Scandinavia but representations of them are everywhere, carved on rocks and etched into bronze vessels and tools such as razors, and most prominently as stone ship-settings. The latter are groups of large stones arranged to form the outline shape of a ship that were used to mark graves. Sometimes taller stones are placed at the ends of the settings to give the impression of raised prows and, more rarely, there are raised stones in the position where, in a real ship, a mast would have been. Most ship-settings range in length from around 6 feet (1.8 m) to 50 feet (15.25 m) but the longest, the now largely destroyed setting at Jelling in Jutland, is about 1,100 feet (335 m) long. Over 2,000 settings survive, with a major concentration on the Swedish island of Gotland, but these are probably only a fraction of those originally built. Many of the survivors are now incomplete as a result of farmers removing stones to build walls or clear land for the plough, and it is likely that many more have been completely destroyed in this way. The first ship-settings were built in the second half of the Bronze Age and they continued to be built almost until the end of the Viking Age, nearly 2,000 years later. It is impossible to be certain what beliefs were associated with these symbolic ships or, for that matter, that those beliefs remained the same throughout the long period in which the settings were built, but they were probably intended in some way to transport the soul of the deceased to the afterlife. The use of real ships in burials, which began in the centuries immediately before the Viking Age, was probably a development of these beliefs.

Even more numerous than ship settings are petroglyphs showing large canoe-like boats crewed by warriors armed with spears and axes, as well as wheeled vehicles, animals and sun discs. The boats are always shown in silhouette and have distinctive double beaked prows at each end. No other details of the boats’ construction are shown on the petroglyphs, however. The boat petroglyphs are usually carefully sited in natural channels on the rocks, along which rainwater and melted snow would flow to create a lifelike scene. It is unlikely that the petroglyphs were carved simply because Bronze Age people liked to see pictures of boats. They probably depict mythological scenes or had some ritual purpose. The ships are often associated with petroglyphs of sun discs which, with artefacts like the Trundholm Sun Chariot, should probably be interpreted as evidence of a solar cult. Solar cults were widespread in later Bronze Age Europe and are indicative of an increasing importance of sky gods, which were, of course, the dominant gods of the Norse pantheon in Viking times. Another religious change that affected much of Europe in this period was the adoption of cremation as the normal way to dispose of the dead. This was accompanied by a decline in the practice of placing grave goods in burials. Clearly these developments must reflect a major change in attitudes to afterlife. The valuable metalwork that would have gone into graves was now buried as votive hoards in bogs. As places where the separate realms of earth, water and air mingled, bogs were seen as particularly numinous places. However, votive hoards were not merely a way of appeasing the gods; they helped maintain the status of the elite by creating an artificial shortage of metals.

Because of environmental changes most Bronze Age petroglyphs cannot now be appreciated in their original context. A good example is the UNESCO World Heritage site of Tanumshede in Bohuslän on Sweden’s west coast, where there are around 600 petroglyphs spread over a 126 acre (51 hectare) site. When originally carved the Tanumshede petroglyphs were on the shore of a shallow fjord, but they are now well inland and surrounded by pine forest. During the Ice Age, the enormous weight of the Scandinavian ice sheet depressed the land surface by over 2,000 feet (610 m). When the ice sheets melted, sea levels rose and this vast depression flooded, forming the Baltic Sea. Relieved of its burden, the land, more slowly, began to rebound and will continue to do so for thousands of years to come. This process, which is known to geologists as isostatic uplift, means that Scandinavia’s coastline has been constantly changing throughout human history. Fishing and trading communities that depended on access to the sea have often been forced to relocate themselves as the uplift has left them high and dry. The Baltic Sea is steadily shrinking and in about 2,000 years time its northern arm, the Gulf of Bothnia, will be mostly dry land.

During the Iron Age (500 BC – AD 800), Scandinavian society grad​ually acquired the characteristics that directly caused the Viking expansion. The Scandinavian Iron Age is conventionally divided into three periods, the early or pre-Roman Iron Age (500 BC – AD 1), the Roman Iron Age (AD 1 – 400), and the Germanic Iron Age (400 – 800). The introduction of iron had an immediate and dramatic impact in Scandinavia. Scandinavians had been totally reliant on imported bronze to make tools and other artefacts but bog iron, a low grade, easily worked, iron ore that accumulates in bogs and marshes, is abundant throughout Scandinavia. This new-found self-sufficiency caused the decline of the long-distance trade systems that had sustained the Bronze Age elites. With their control over the distribution of metals broken, their status and power collapsed, and it is five centuries before there is evidence for the re-emergence of a social elite.

The wider availability of metal tools contributed to agricultural expansion and a rising population, and an increase in conflict. In the late second century BC population pressure led directly to the first of what would be many migrations out of Scandinavia. Faced with a critical shortage of farmland, around 120 BC two tribes from northern Jutland, the Cimbri and the Teutones, set out on a migration in search of new homelands. Their search took them on a destructive rampage across much of central and western Europe before they invaded Italy in 102 BC and were finally annihilated by the Romans. Although it ended in disaster, this migration was just a foretaste of what was to come. Many of the Germanic tribes who invaded the Roman Empire in the fifth century traced their legendary origins to Scandinavia. The Goths believed they had originated in Götaland in southern Sweden; the Burgundians from the island of Borgundarholm, now Bornholm in Denmark; and the Vandals from Jutland. The Angles and Jutes, who joined the Saxons in settling Britain, both certainly came from Jutland. Writing around 550, the Gothic historian Jordanes described Scandza as ‘the womb of peoples’ because it had given birth to so many tribes. The Viking expansion was really just the last phase of an extended period of migrations out of the north.

Conflict is the essence of what is possibly the single most important archaeological find of the early Iron Age: a near-complete ‘war canoe’ and a hoard of weapons that were buried together in a bog at Hjortspring on the Danish island of Als around the time of Pytheas’ travels. The boat itself is the oldest plank-built boat yet found in Scandinavia and has enormous significance as the earliest known ancestor of the Viking longship. What is immediately most striking about the Hjortspring boat is its distinctive double beaked prows, which closely resemble those of the boats depicted in the Bronze Age petroglyphs, so it is likely to represent a well-established tradition of boat building. The boat was 56 feet (17 m) long by 6 feet (1.8 m) broad and was built from just five lime wood planks: a broad bottom plank with two overlapping planks on each side. This method of building a hull from overlapped planks, known variously as clinker, lapstrake or Nordic construction, is what marks the Hjortspring boat out as the earliest known ancestor of the Viking longships, whose hulls were built in the same way. The ends of the Hjortspring boat were closed with two carved wooden blocks that served as stemposts. The function, if any, of the projecting beaks is unknown. They may have been intended to ride up over the gunwales of an enemy boat and capsize it, or they may simply be a hangover from an earlier stage in the development of the Nordic boat-building tradition that were retained for cosmetic reasons: they do give the boat a racy appearance. No metal was used in the boat’s construction: the planks were sewn together and fastened to internal strengthening ribs with ropes made of lime tree bast. The boat was paddled by a crew of twenty – a useful number for a raiding party – who sat on thwarts set at the level of the gunwale. There was a steering oar at both ends, so the boat could be sailed in either direction. This would have been a great advantage for raiding because the boat could run right up a beach and the crew would not have had to turn it around if they needed to make a quick getaway. The boat was skilfully built to be as light as possible and sea trials with a replica have shown that it was fast, stable and relatively seaworthy. The boat was sunk in the bog with enough weapons to equip a small army: 138 iron tipped spears, thirty-one bone- or antler-tipped spears, eleven iron swords, sixty to eighty shields, and around twenty coats of mail, all but one of which survived only as rust prints in the peat. Plates from a bronze cauldron and the bones of a horse, a dog and a puppy, a lamb and calf were also found. Both ship and weapons survived thanks to the acidic and anaerobic (oxygen deficient) conditions found in peat bogs, which preserve organic materials like wood, textiles and leather by pickling them, while the lack of oxygen retards the rusting of iron.

The Hjortspring find is one of the earliest examples of a practice of sacrificing the spoils of war that became widespread in Scandinavia and adjacent areas of north Germany during the early Iron Age. No other known sacrifice approaches the Hjortspring find in scale, however, and its deposition must commemorate a major battle. The most likely scenario is that the ship and weapons belonged to a large army that invaded Als only to be defeated by the local inhabitants, who offered their plunder to their gods as a thank-offering for victory. There are enough weapons to equip at least eighty warriors, so the invaders would have needed a fleet of at least four Hjortspring-type boats and, of course, we don’t known how many of them escaped. It is clear, at least, that raiding by sea was already a serious business in early Iron Age Scandinavia.

It was not only weapons and boats that were sacrificed in bogs, people were too. No bog bodies have been found in Norway or Sweden but over 200 have been found in Denmark and neighbouring areas of northern Germany. Though the acidic conditions in the bogs have often completely dissolved the victims’ bones, in many cases their hair, skin and internal organs are so well-preserved that post mortem examinations have revealed much about their health, diet and causes of death. One surprise is that seafood appears not to have been an important part of the Danish diet in early Iron Age times. Most bog bodies show signs of a violent death, like Tollund Man, killed c. 400 BC by hanging, and Grauballe Man, whose throat was cut from ear to ear around 100 years later. Some of the victims were found pinned down in the bogs by heavy branches. In Germania, a treatise on the Germanic peoples written in AD 98, the Roman historian Tacitus says that this was one of the methods used by the German tribes to execute criminals.

The warlike character of Scandinavian society intensified in the course of the Roman Iron Age. Many Roman weapons have been found in votive hoards, especially in Denmark, suggesting that Scandinavians frequently fought with their German neighbours to the south, who had direct access to Roman weaponry. The increasing importance of war in society is indicated by the appearance of warrior graves furnished with weapons, evidence that a warrior elite now dominated Scandinavian society. A small number of these graves are furnished with imported luxury goods, such as Roman silverware, jewellery and glass, indicating the rise within this elite of a class of chieftains or petty kings. Everyday Roman goods, like pottery and coins have also been found in some quantity in Scandinavia, show that trade with the Roman Empire was not confined to luxuries. There may have been direct trade with the empire by sea, but it is perhaps more likely that Roman goods reached Scandinavia through intermediaries in Germany. Not surprisingly, Roman artefacts are most common in Denmark, but they are not evenly distributed over the country. One remarkable concentration of Roman goods is found in the Stevns area of the island of Sjælland, suggesting that this was the centre of a powerful chiefdom or small kingdom, which could control trade over a wide area. Another striking site from the later part of the period is Gudme on the island of Fyn, where evidence for a 154-foot (47 m) long hall has been found: the largest known in Scandinavia from this period, it has been called ‘the King’s Hall’ and, certainly, a hall of such size implies the existence of a strong central authority. Over 1,000 Roman coins, including twenty gold denarii, have also been found at the site. Gudme means ‘god’s home’, so the place may have been a cult centre. Closely associated with Gudme is a seasonal port and trading place at Lundeborg, where Roman coins and other imports have been found, along with evidence of shipbuilding. This close association between religion and trade is also seen at the Iron Age trading place at Uppåkra near Lund in southern Sweden, where the remains of a wooden temple have been found. It is likely that trade fairs were held during religious festivals when plenty of visitors could be expected.

No single place has provided more spectacular evidence of the warlike character of Scandinavia in the Roman Iron Age than Nydam Moss in southern Jutland. Now just north of the Danish-German border, in the Iron Age Nydam was probably in the territory of the Angles, the Germanic tribe from whom the English get their name. The moss is now a rather soggy meadow but in Roman times it was a reed-fringed lake. In the 1830s, local farmers digging peat from the by then silted up lake began to find old iron weapons and shields. These discoveries eventually caught the attention of antiquarians and between 1859 and 1863 the moss was excavated by the Danish archaeologist Conrad Engelhardt, who discovered large quantities of weapons, two intact clinker-built ships, one built of oak and one of pine, and another oak ship which had been deliberately broken up before its deposition. The excavations were brought to an end by the outbreak of war between Denmark and Prussia in 1864, after which the area remained under German rule until 1920. During the war the pine ship was chopped up for firewood by German soldiers and burned. Systematic re-excavation of the site in 1984–97 produced thousands more artefacts.

The modern science of dendrochronology, the analysis of the pattern of tree rings preserved in ancient timbers, has dated the oak ship’s construction very precisely to 310–320. The ship was not new when it was sacrificed, so it was probably sunk in the bog around 350. The larger of the two ships, the oak ship, was around 70 feet (921.3 m)long by 12 feet (3.65 m) broad and was propelled by a crew of thirty oarsmen. The ship was double-ended, with long raking prows and was steered by a side rudder, which was only loosely attached to the hull. Like the Hjortspring boat, which was found only a few miles away, the oak ship was built of overlapping planks, but instead of being sewn together they are fastened using iron clench nails. Internal strengthening frames were lashed to the hull planks using lime-bast rope as on the Hjortspring boat. No fittings for a mast were found so the oak ship did not have a sail. Drawings made of the pine boat before its destruction show that it was about 61 feet (18.6 m) long by 10 feet (3 m) broad, had a crew of about twenty two oarsmen and was built in a generally similar way to the oak ship. There was no evidence that the ship had a mast. The modern re-excavation of the site discovered many fragments of the pine ship, the most important of which was a side rudder, which had been attached firmly to the side of the ship on a wooden boss. This type of side rudder continued to be used on longships until after the end of the Viking Age. Rudders were always fitted to the right-hand side of the ship, hence ‘starboard’ (from Old Norse styri/steer and borð/side of the ship). A shield found in the ship was made of timber felled in 296, so the ship was probably sacrificed in the early fourth century. Most of the third ship is thought still to be in the bog, but it was certainly rowed rather than paddled and its planks were fastened with iron clench nails. This ship was built of wood felled in AD 190, so it was probably sacrificed in the early third century.

The change since the early Iron Age from paddling to rowing is significant. For raiding, paddling has the advantage that all the crew can see where they are going, can keep a look-out for the enemy, and can disembark and re-embark more quickly than a crew of oarsmen. On the other hand, rowing is much more energy efficient than paddling so its adoption made it possible to raid further afield. The pine ship is evidence for this as it was probably built in Sweden. Pines large enough for shipbuilding did not grow in southern Scandinavia at this time and the ship’s timbers were decorated with patterns that are also found on contemporary inscribed stones in Sweden. The timing of the transition is uncertain but the earliest evidence for the use of oars is a rowlock found in a bog in Hordaland in Norway, which dates to c. 30 BC – AD 250. The question of the timing of the adoption of the sail in Scandinavia is a controversial one because the evidence is inconclusive. The Celtic peoples of Gaul, Britain and Ireland certainly used sailing ships in pre-Roman times and in his Histories, Tacitus describes the German tribes of the North Sea coast using sailing ships in their wars with Rome in the first century AD. However, in Germania, he also says that the Suiones (the Swedes) used neither sails nor oars on their ships. The two Nydam ships, of course, also used oars but not sails. At the time the ships were sacrificed, the Angles’ southern neighbours, the Saxons, were using sailing ships for pirate raids on Roman Britain and earning notoriety for their practice of sacrificing Roman prisoners to obtain a fair wind home. The Scandinavians cannot, therefore, have been ignorant of the sail in the fourth century. Many Scandinavian mercenaries and merchants must also have been familiar with Roman sailing ships. Despite this, the earliest clear evidence for the use of sails in the region is a seventh-century inscribed stone from Karlby on Jutland’s east coast showing a Nydam-type ship under sail.

The slow adoption of the sail in Scandinavia is hard to explain, especially as the technology itself is not complex: a woollen blanket or leather cloak, two wooden poles and some rope are all that would have been needed to make a rudimentary sail. The most commonly advanced theory, that the keels of ships like the Nydam ships were too weak to support the stresses of sailing, has never been tested experimentally and seems unconvincing given that, globally, sails have been fitted to all manner of watercraft, a great many of which have been technologically far less sophisticated than either of the Nydam ships. The argument usually advanced is that if the sail was not adopted it was because there was no perceived need for it. Warships needed large crews anyway and a sail would simply make a raiding ship more conspicuous (Vikings sometimes lowered their sails when approaching a hostile coast to increase their chances of landing unobserved), so it may not have seemed so advantageous for short range raiding in sheltered fjords and coastal waters. Chiefs and kings may also have seen commanding a crew of oarsmen as an expression of their own power. However, rowing long distances is hard work even for those accustomed to it, so these arguments are not really convincing. Perhaps it was only when Scandinavians began setting out on raiding and trading voyages beyond Scandinavian waters in the fifth century that the benefits of the sail become obvious to these technologically conservative seafarers?

The ships were only part of the Nydam find. Excavations have uncovered thousands of weapons, or parts of weapons, including swords, spears, lances, axes, and bows and arrows, elaborately decorated wooden scabbards, silver fittings from scabbards and belts, silver bars, and other personal items like combs and wooden storage boxes. The largest number of weapons were found in and around the ships but there were also many other weapon sacrifices in the bog. Most consisted of only a few spear or lance heads but one, which was surrounded by a fence of thirty-six swords thrust down into the bog, contained over 1,000 objects. Deposited c. 450 – 475, this was the last known weapon sacrifice at Nydam and one of the last in Scandinavia. Beliefs were changing again, bogs lost their significance as sacred places and the custom of bog sacrifices died out.

The finds from Nydam Moss illustrate another change in the north, the beginnings of literacy. The early Germans and Scandinavians wrote using runes, an alphabet of twig-like characters known as the futhark after the names of its first three characters. Though often inscribed on stone and metalwork, runes were originally designed to be carved on wood because the characters avoid horizontal lines, which would not have been clearly distinguishable from the grain. The oldest known runic inscription reads harja, a man’s name, and was found on a comb from Vimose bog on the Danish island of Fyn, which was made c. AD 150. The largest concentration of early runic inscriptions has been found in southern Scandinavia but it is not certain that this was the area where they were invented as runes were used by all the Germanic peoples. The origin of runes is surrounded by myth. In the Viking Age Scandinavians believed that runes were a gift of Odin, who had hanged himself, impaled on a spear, from the World Tree Yggdrasil for nine days to learn their secret. They are now more prosaically thought to be derived from Latin letters, which early Germans could easily have become familiar with through contacts with Roman merchants or during mercenary service in the Roman army.

Runes were certainly not for everyone’s use in Iron Age Scandinavia. Of the thousands of artefacts recovered from Nydam Moss only ten carry inscriptions in runes. Most are on war gear, arrow and lance shafts, a lance head, a decorative bead from a sword, a scabbard, a silver belt fitting is the only inscribed artefact without an exclusively military function. This suggests that runes were associated with high social status. The inscriptions are all very short, only one or two words, and the majority simply record names, either of the artefacts’ presumed owners, the craftsmen who made them, or the runemasters who carved the runes. For example, the scabbard – found in the pine ship, carries the inscription harkilaR ahti. The meaning of ahti is unknown but harkilaR is a man’s name. Swords were rare and expensive at that time, so was harkilaR the defeated commander of the pine ship? One runic inscription, on a lance shaft, does not spell out a word but is just a sequence of runes and rune-like symbols. This hints at how runes were seen, that the individual characters were at least as significant as the words they spelled.

In most civilizations, the main impetus to the development of writing was the need to keep records when society became too large and complex for unaided human memory to keep all the information needed for good government. Utilitarian lists and tax records came first. Memorials, literature, historical, religious and philosophical texts all came later. The Germanic-Scandinavian world was nowhere near this level of complexity in the Roman Iron Age, so writing fulfilled a different function. Rune means ‘secret’ or ‘something hidden’, so there was something esoteric about them. In the Viking Age, runes were believed to have magic properties. Each rune had its own name, embodying gods, ideas and powers. The act of writing a rune harnessed that power. Carving one of the runes named after gods, such as the ‘T’-shaped Tiwaz rune associated with the war god Tiwaz or Tyr, was an invocation for the god’s protection. In this way the act of carving a runic charm on an object turned it into a protective amulet. However, not just anyone could carve protective runes, they had to be carved by a trained runemaster if they were to be effective. Errors would make them impotent or even harmful. In Scandinavia, the use of runes remained limited to names and charms until the Viking Age, when longer commemorative inscriptions began to be made. Viking graffiti of the ‘Halfdan was here’ variety has been found across the Viking world, from Greenland to Greece, suggesting that by that time literacy in runes had become widespread. Because they had pagan overtones, most of the Germanic peoples gave up using runes and adopted the Latin alphabet soon after they converted to Christianity. However, they continued to be used in medieval Scandinavia, when even law codes and other texts were written in runes. In the Dalarna district of central Sweden, a tradition of writing runic charms survived into the twentieth century.

The indirect cause of Scandinavia’s changing society was Roman influence on the German tribes to the south. By the end of the first century BC the Germanic tribes and the Roman Empire shared a common frontier along the Rhine and Danube rivers. Despite incursions by both sides on each other’s territory, this frontier remained stable for 400 years. Contact with the Roman Empire had a great impact on those tribes closest to the border. Plunder from raids, Roman subsidies to friendly tribes, trade, and wages for mercenary service enriched the border tribes. Tribes further north raided and traded with the border tribes, so becoming enriched in turn. Those who led successful raids or gained control of the distribution of trade goods were soon set apart from the rest of society by their greater wealth and status. Roman writers, such as Tacitus, attest that the comitatus or war band became the central institution of Germanic society in this period. Known in Viking Age Scandinavia as the lið or hirð, the comitatus was made up of young warriors who entered the service of a chief or king. In return for their loyalty and military service the warriors of the comitatus expected to receive food and lodging, gifts of weapons and jewellery, and a share of war booty. The warriors swore loyalty to their chief for life but their loyalty was conditional on the chief fulfilling his side of the bargain. A chief who did not, or could not, reward his warriors would not have a comitatus for long. Chiefs who were poor warriors fell by the wayside, those who were good warriors consolidated their power because their success attracted more warriors, and a stronger comitatus led to more success in war. This dynamic created a violent and predatory society in which war was the surest route to wealth, status and power. Another effect was to concentrate power in fewer and fewer hands, increasing competition between ambitious men within a tribe, and to encourage the merging of tribes. In some cases this was because a stronger tribe conquered and absorbed a weaker one, but just as often it was done voluntarily. Many tribes allied, forming coalitions to wage war more effectively. When their unity was cemented by success in war, these coalitions became the basis for new ethnic identities. The Saxons and the Franks, for example, both developed from tribal coalitions in this way.

The Germanic Iron Age (400 – 800) was Scandinavia’s heroic age, a proto-historical period that was half remembered in legendary traditions of dragon-slaying warriors and great battles. At the beginning of the period the process of centralisation that had transformed the Germanic world had still not progressed far in Scandinavia. Jordanes listed more than twenty tribes living in the ‘island of Scandza’ in his history of the Goths, and this doesn’t include the Angles and Jutes who lived in Jutland, which he didn’t count as part of Scandza. Jordanes’ list is based ultimately on the testimony of Rodulf, the exiled king of a Norwegian tribe called the Rani. According to Jordanes, two tribes had already become pre-eminent, however: the Swedes or, as they called themselves, the Svear, and the Danes, whose territory then included Skåne and Blekinge in the far south of modern Sweden. Also prominent were the Götar, who lived between the Swedes and the Danes in Sweden’s densely forested Southern Uplands. Around eight tribes lived in Norway; their homelands can be identified with some certainty because they are etymologically related to the names of regions of modern Norway. The Raumarici most likely lived in Romerike, the Alogi in Hålogaland north of the Arctic Circle, the Rugi in Rogaland, and so on. Rodulf’s Rani probably lived in Romsdal, the valley of the River Rauma, in the west of the country. Thanks to its rugged geography, Norway remained a land of local tribes even at the beginning of the Viking Age. Elsewhere, most of the tribes named by Jordanes had vanished by this time. The Danes had absorbed the Angles and Jutes and another tribe mentioned by Jordanes called the Heruls, who lived between the Götar and the Danes. The Swedes and Götar had absorbed the rest. This was certainly not a peaceful process. Fortresses proliferated across Scandinavia – over 1,500 are known from this period. On the 80-mile-long island of Öland, nineteen stone ring-forts were built around this time so no one would have been more than two or three miles from a refuge. At the same time there was a general movement of settlement away from the coast, a sure sign that piracy was endemic. The Viking Age may not have started in western Europe until 793 but something like it was already well under way in the Baltic Sea.

The first half of the Germanic Iron Age is known as the Migration Period (400 – 500), after the series of Germanic migrations that resulted in the complete collapse of the Western Roman Empire in 476. The ultimate cause of the Germanic migrations was the arrival c. 370 in eastern Europe of the Huns, a ferocious Turkic nomad people from Central Asia. Those tribes who could took flight in a desperate search for safer homelands, displacing other tribes and setting almost the whole Germanic world in motion. Some tribes were broken up and absorbed by others, and new ethnic identities were forged from ad hoc coalitions. Many tribes, including the Goths, Vandals, Burgundians and Suevi, sought refuge in the Roman Empire, overwhelming its border defences and founding new kingdoms on its territory. The Huns never reached Scandinavia but the political chaos of the age created opportunities for the enterprising. Britain slipped out of Roman control in 410 and was left exposed to the Saxons, who seized land and began to settle the rich lands of the south-east and the Midlands. Saxons also took advantage of the chaos the invasions caused in Roman Gaul, settling in the Pas de Calais, Normandy, and on the River Loire. At the same time they raided as far north as the Orkney Islands, as far west as Ireland, and as far south as Aquitaine. The Angles soon joined the Saxons in Britain, settling along the east coast from East Anglia north to the Firth of Forth. So too did the Jutes, whose main settlements were probably in Kent. Another tribe from southern Scandinavia, the Heruls, launched pirate raids as far afield as Aquitaine and northern Spain but they made no known settlements. A branch of this well-travelled people had already migrated to Ukraine in the third century, and from there launched pirate raids around the Black Sea and the eastern Mediterranean. Those Heruls who remained in Scandinavia were conquered by the Danes in the sixth century. This is most likely the period that sailing ships began to be used in Scandinavia as it is scarcely credible that the Angles, Jutes and Heruls should have undertaken such long voyages of settlement and piracy in rowing ships, taking weeks or months, when their Saxon neighbours were crossing the same seas, for the same purposes, in much swifter sailing ships.

Scandinavian raiders were also busy much closer to home, raiding Frisia, a region on the North Sea coast now divided between Germany and the Netherlands. In c. 528, Frisia was raided by the Scandinavian king Hygelac, who went on to sail down the Rhine as far as Nijmegen before he was defeated and killed by the Franks. It is a sign that Scandinavia was now truly beginning to emerge from prehistory that Hygelac’s raid was recorded in four independent literary sources, including Gregory of Tours’ near contemporary History of the Franks and the eighth-century Anglo-Saxon epic poem ‘Beowulf’. Unfortunately, the sources don’t agree whose king Hygelac was. Gregory of Tours, and two other Frankish sources, describe Hygelac as a king of the Danes – their earliest appearance in history – but in ‘Beowulf’ he is called king of the Geats, that is the Götar, from southern Sweden, or even the Jutes of Jutland. In the poem the hero Beowulf is said to have taken part in the raid, swimming home after his king’s defeat, in full armour, underwater. Beowulf goes on to save the Danish king Hrothgar from the man-eating troll-like monster Grendel and his equally awful mother, become king of the Geats, and finally die slaying a dragon that was ravaging his lands. ‘Beowulf’ also describes another Danish raid on Frisia as does another early Anglo-Saxon poem, the fragmentary ‘Finnsburg’. A Frankish poem, composed c. 570, records another major raid by Danes but this was also driven off by the Franks. No further Danish raids on Frisia are recorded until the Viking Age, so this defeat appears to have deterred them from interfering in what the Franks regarded as their sphere of influence for 200 years.

The Migration Period was a quite literal golden age for Scandinavia. In the course of their migrations, the Germans and Huns relieved the Romans of enormous amounts of gold and silver, either as plunder or payments of tribute. Much of this gold eventually found its way to Scandinavia, whether by trade or plundering raids across the Baltic, or in the pockets of homeward-bound mercenaries. One of the routes by which much of this gold reached Scandinavia was through Eastern Europe and across the Baltic to the islands of Bornholm, Öland and Gotland, where several treasure hoards dating to this period have been found. The richest hoard of the period, however, was found in the eighteenth century at Tureholm in Södermanland in central Sweden and contained 26.5 pounds (12 kg) of gold. Treasures may be buried for two reasons: ritual offerings to the gods or, in the days before banks, for security. However, in the second case, the owner’s intention was eventually to recover the treasure, not leave it in the ground as an expensive time capsule for modern archaeologists or metal-detectorists to discover. There is no evidence that most of these treasures were buried for ritual reasons so the failure of the owners to recover so many hoards is best seen as yet another sign of the pervasive insecurity of the period. These islands would have been particularly exposed to piracy and the owners of the unrecovered hoards may well have been killed in raids or captured and carried off for the slave markets.

Most of the imported Roman gold was melted down and turned into spectacular jewellery and other prestige objects for the aristocracy. It was in the early part of the period that goldsmiths and silversmiths in southern Scandinavia developed the Scandinavian-Germanic animal art style, which used the stylised and enormously elongated bodies of real and imaginary animals to create interlaced patterns of astonishing complexity. The new art was probably a response to the turbulent times, creating a new language of symbols that were full of meaning to those who had the knowledge to read them. Unfortunately, that knowledge is now lost. Taken to Britain by the Anglo-Saxons, the animal style merged with indigenous Celtic art styles to create the hybrid Hiberno-Saxon style, whose finest expressions are found in illuminated manuscripts such as the Lindisfarne Gospels and the Book of Kells. In Scandinavia, animal art developed through a succession of styles until it was replaced by the imported Christian Romanesque style at the end of the Viking Age.

One of the characteristic items of Migration Period jewellery are bracteates, gold medallions modelled loosely on Roman medallions, which were worn as pendants. Bracteates frequently have the motif of a man’s head and a horse – thought to represent Odin and his steed Sleipnir – and sometimes also runic inscriptions, most of which have defied interpretation. Few artefacts, however, could have displayed the wealth of their owner more impressively than the two ornate gold drinking horns found at Gallehus in Jutland, the larger of which was 30 inches (75.8 cm) long and weighed over 7.7 pounds (3.5 kg). Horns like this, together with other precious tableware, fine jewellery and weapons, would have been displayed at the lavish warrior feasts that were, after wars, a chief’s or king’s most important opportunities to enhance their reputations by feeding their followers heroic portions of meat, filling them with ale or mead, and showering them with valuable gifts. Another example of Roman influence in this period are guldgubber (‘old men of gold’). These are tiny gold foil votive plaques impressed with figures of men or, more rarely, women or couples, which are thought to be inspired by Roman temple money. Around 75 per cent of the 3,000 guldgubber found so far come from Sorte Muld, a trade and cult centre on Bornholm. Guldgubber were mass produced as many were clearly stamped with the same moulds.

The turbulent times shaped Norse legend as well as metalwork. It was in this period that the Volsunga Saga, the most important of the Norse legendary sagas, began to take shape. The saga centres on the deeds of the legendary hero Sigurd, the forging of his magical sword Gram, his slaying of the dragon Fafnir and his acquisition of its cursed treasure hoard, and his eventual murder at the instigation of his spurned Valkyrie lover Brynhild. While a plot like that is unlikely to have any basis in historical fact, several of the saga’s leading characters are historically identifiable figures from the Migration Period. Brynhild’s husband Gunnar is based on the Burgundian king Gundahar, who was killed in battle with the Huns in 437; King Atli, who kills Gunnar is a barely disguised Attila the Hun (d. 453); and Jormunrek, the husband of Sigurd’s daughter Svanhild, is inspired by Ermanaric, a king of the Goths who committed suicide after being defeated by the Huns in 375. Sigurd himself is thought by some to be based on the Frankish king Sigibert (d. 575), who was murdered as a result of family feuding between his wife, his brother and his brother’s lover. If so, Sigibert’s Burgundian wife Brunhilda may, then, be the inspiration for Brynhild.

The Late Germanic Iron Age (550 – 800) saw the emergence of powerful regional kingdoms in Denmark and Sweden. Scandinavia was still largely beyond the horizons of literate Europeans so these kingdoms’ existence is deduced primarily from archaeological evidence such as major defence works, planned settlements, richly furnished burials and feasting halls, all of which point to the presence of strong centralised authorities that controlled considerable material and human resources. One of these kingdoms was probably centred in southern Jutland where a 19-mile long (30 km) earth and timber rampart, known today as the Danevirke, was built across the neck of the peninsula between Hollingstedt and Schleswig. Although the Danevirke is now in Germany, in the early Middle Ages, this sparsely populated area’s marshlands and infertile heaths made it a natural frontier between the Danes and the Saxon tribes to the south. The Danevirke began as a simple earth bank, built around the middle of the seventh century. About eighty years later the height of the rampart was raised and a timber palisade was built on top, turning it into a much more effective obstacle. Thanks to the science of dendrochronology the date of the palisade’s construction can be fixed precisely – the timbers used to strengthen the rampart were felled in 737. The Danevirke, which still survives to a height of nearly 20 feet (6.1 m) in places, was strengthened several more times during the Middle Ages before it fell out of use in the fourteenth century. The construction of the Danevirke was probably overseen from a recently discovered high-status settlement at Flüsing, near Schleswig. An eighth-century feasting hall, roughly 100 feet (30.5 m) long by 30 feet (9.1 m) broad, excavated here was surrounded by up to 200 smaller buildings, which could together have accommodated up to 1,000 warriors on a temporary basis.

Another construction work possibly commissioned by the same king is a canal across an isthmus on the island of Samsø, off Jutland’s east coast. This has been dated by dendrochronology to exactly 726. It was probably built to make it easier for warships to control the sea routes on both sides of the island. The foundation of Scandinavia’s oldest town, Ribe on Jutland’s west coast, can also be dated to this period. A site about 220 yards (201 m) long and 70 yards (64 m) wide was drained, levelled with a layer of sand over 2 feet thick, and divided up into rectangular plots. Oak planks from a timber-lined well date the event to between 704 and 710. Around 720, a central street was laid out and this was paved with planks around 730. No traces of permanent buildings have been found on the site but there are signs of temporary huts and craft workshops so Ribe functioned at first as a seasonal market place. The market place was surrounded by a ditch and fence. These were too small to be for defence so were probably intended to make it easier for Ribe’s ruler to manage access and collect tolls.

Ribe was linked in to extensive trade networks, extending to Italy, Byzantium and Norway, but the most common imported artefacts originated in the Frankish kingdom: lava quernstones from the Eifel Mountains, and glass and pottery from the Rhineland. Large amounts of unworked amber have been found on the site so this was presumably an important export. There is evidence that large quantities of cattle were brought to the market, so perishable goods like hides were probably also exported. Ribe’s foundation demonstrates the existence of a ruler who could control where and when trade was conducted in his territory and presumably also guarantee traders’ security when visiting the site. Scandinavia’s earliest coins, imitations of a Frisian coin type known as a scaetta, were produced at Ribe c. 720, so this ruler could also to some extent control the means of exchange. A permanently inhabited site about 250 yards (229 m) south-east of the market may have been the ruler’s compound. The identity of the ruler cannot be ascertained with certainty but there is a good chance that it was Angantyr, a Danish king who was visited by the Anglo-Saxon monk Willibrord on the first Christian mission to Scandinavia in c. 725. Willibrord’s biographer described Angantyr as ‘crueller than a wild animal and harder than a stone’, but he greeted the missionary politely enough even though he showed no interest in converting to Christianity.

Angantyr was not the only king in Denmark. In ‘Beowulf’, the hero’s Danish host, Hrothgar, is described as a member of the Scylding dynasty. The same dynasty appears in semi-legendary saga traditions and in the Gesta Danorum (‘Deeds of the Danes’) by the twelfth century Danish historian Saxo Grammaticus as the Skjöldungs. Opinion about the historical reality of the Skjöldungs has wavered over the years: the consensus today is that the dynasty really did exist but that the stories that have come down to us belong more to the realm of legend than fact. Traditionally, the Skjöldungs were associated with the village of Gammel Lejre (‘Old Lejre’) on the island of Sjælland. An extraordinary concentration of impressive prehistoric barrows, dating from the Neolithic through to the late Viking Age, surround the village, marking it as a place that once possessed intense and enduring spiritual significance. In the last thirty years, archaeological excavations at Gammel Lejre have revealed that a succession of great timber feasting halls were built there between the sixth and tenth centuries, confirming that it was also a royal power centre in the period when the first Danish kingdoms were being forged. The attraction of the site for the kings of Sjælland must have been its many ancient monuments: they will have hoped to strengthen their authority by associating themselves with a place of such obvious ancient power.

The earliest hall at Gammel Lejre was built at Fredshøj, close to a prominent Bronze Age burial mound on a low ridge overlooking the marshy valley of the Lejre river. The bow-sided hall was around 150 feet (45 m) long and 20 feet (7 m) wide and has been dated to around the second quarter of the sixth century. If Hrothgar was a real historical figure, this would likely have been his hall. Nearby was a hørg, a sacrificial offering place or altar made of a pile of stones. Pits surrounding the hørg contained the remains of broken pots and thousands of sacrificial animals. In the early seventh century, Fredshøj was abandoned in favour of Mysselhøjgård about 550 yards (500 m) to the south. Like the Fredshøj site, this was also on a ridge overlooking the river and was 160 feet (48.5 m) long by 38 feet (11.5 m) wide, covered an area of 600 square yards (500 sqm), and was subdivided into a central hall and storerooms and residential rooms. Several large houses around the hall were probably built to accommodate household warriors and guests. A timber palisade surrounded the hall and houses so that access to them could be easily controlled.

Outside the royal compound there was a small colony of craftworkers who supplied the royal family with the prestige metalwork they needed to display their own status and to hand out as gifts to their warriors. A large farm about 550 yards (500 m) to the north would have supplied the community’s food. As at Fredshøj, there was a hørg, close to the feasting hall. The German Thietmar of Merseburg, writing around 1016, described a religious festival involving the sacrifice of ninety-nine humans, the same number of horses and an unspecified number of dogs and cocks, which was held at Gammel Lejre every ninth year on 6 January. So far, no evidence of human sacrifice has been found at the site. During the period when Mysselhøjgård was occupied the largest of all the religious monuments at Gammel Lejre was built. This is a now incomplete 282-foot (86 m) long stone ship that was used for burials and religious ceremonies. The Mysselhøjgård compound remained in use for over 350 years. During this time the feasting hall and its secondary buildings were completely rebuilt many times, replaced by a new hall of approximately the same size and plan. This too was eventually pulled down and replaced by a new hall sited a few yards to the north. Halls continued to be pulled down and rebuilt at Gammel Lejre until around 1000, when the site was abandoned, probably because of its pagan associations, in favour of the new Christian centre of Roskilde, five miles to the north.

Prehistoric monuments frequently become associated in folklore with historical or legendary figures. Until it was proven to date to the Neolithic, one of the barrows at Gammel Lejre was believed locally to be the burial place of the most famous of the Skjöldung kings, Harald Hildetand (‘Wartooth’). Harald most likely lived around the same time as Angantyr of Ribe, so the barrow could not have been raised for him, and, in any case, the legendary traditions agree that he is buried in Sweden on the site of the Battle of Bråvalla, in which he was killed fighting against the Swedish king Sigurd Ring. The location of Bråvalla, the greatest battle of Scandinavia’s proto-historical period, is unknown but was traditionally thought to have been near Bråviken Fjord, in Östergötland. Harald had no political motives for invading Sweden. He had enjoyed a long and successful career of raiding, conquest and plundering, but having reached the ripe and improbably old age of 150, he was becoming seriously worried that he would die in bed and so forfeit the chance to go to Valhalla. Harald’s sole motive was, therefore, to seek an opportunity to die fighting in battle. In some versions of the story, Harald was felled by the hand of Odin himself, who battered him to death with a club. The victors burned Harald on a funeral pyre, bidding him ride straight to Valhalla, together with the fifteen kings and 30,000 other warriors who had fallen in the battle. Plenty of mead would have been drunk that night in Valhalla.

In Sweden the last period of the Germanic Iron Age is known as the Vendel Period after a remarkable cemetery of fourteen high status graves at the modern village of Vendel, a few miles north of Uppsala in Uppland in central Sweden. This fertile region was the homeland of the Swedes. The burials at Vendel are the richest in Sweden for this period and most probably belong to members of a royal dynasty. The bodies were interred, without cremation, in large boats of up to 32 feet (10 m) in length, perhaps to transport them to the realm of the dead, and were surrounded with valuable grave goods, food and cooking gear for the journey, glass, superb weapons and armour, including a bronze-decorated iron helmet, hunting dogs, horses and saddles and, in one grave, a falcon. The distinctive animal-interlace ornament found on much of the metalwork gave its name to the Vendel art-style. Also near Vendel is a 16 feet (3 m) high barrow known as Ottarshögen (‘Ottar’s barrow’) after King Ottar of the Swedish Yngling dynasty who, like the Danish Skjöldungs, belong to the shadowlands between legend and history. The barrow contained the remains of a man and a woman. A well-worn Roman gold solidus minted in 477 was found in the burial, probably dating it to the sixth century. Ottar is the Swedish King Ohthere who is mentioned in ‘Beowulf’ as a contemporary of the eponymous hero, so it is not impossible that the barrow could really be his grave.

In the semi-legendary Icelandic saga traditions the Ynglings were descended from the fertility god Freyr and his consort, the giantess Gerðr. The dynasty’s name comes from Freyr’s alternative name, Yngvi (Yngling means ‘descendent of Yngvi’). The site most closely associated with the Ynglings is Gamla Uppsala (‘Old Uppsala’), which, in the Viking Age, was a major cult centre dedicated to Freyr, Thor and Odin. The site lies a few miles north of the modern university town of Uppsala on the fertile plain of the River Fyris. In the Viking Age, Gamla Uppsala was easily accessible by ship from Lake Mälaren to the south, which was at that time a long, shallow fjord penetrating deep inland from the Baltic Sea – the gradual post-glacial rebound of the land turned Mälaren into a lake around 1200. The most striking monuments at Gamla Uppsala are three enormous burial barrows traditionally associated with the Yngling kings Aun, Egil and Adils. The two mounds that have been excavated, those associated with Aun and Egil, each contained the cremated remains of a high ranking male and warrior gear including poorly preserved helmets decorated in the Vendel style. Hundreds of smaller burial mounds surround the three great mounds. These are just the few survivors of centuries of farmland improvements: originally there were as many as 3,000 burial mounds around Gamla Uppsala. Those burials that have been investigated date consistently to the sixth century or later. A low flat topped mound to the east of the great barrows, known as the Tingshögar (‘thing-mound’), is probably where the Disting was held in historical times. This was the annual thing (‘assembly’) of the Swedes, which got its name because it was held at the same time as the late-winter Dísablót, a sacrifice in honour of the Dísir, a group of female fertility spirits. Underneath Gamla Uppsala’s twelfth century church are the remains of an earlier wooden structure. These are thought to be of a wooden temple that contained idols of Freyr, Thor and Odin: the temple was said to covered with gold in the late Viking Age. According to the German ecclesiastical writer Adam of Bremen (d. c. 1080), a festival in honour of the three gods was still being celebrated in this temple in the 1070s. The festival was held once every nine years around the time of the spring equinox and lasted for nine days. On each day one human male was sacrificed, together with other male animals, including horses and dogs, so that all together the gods were offered seventy-two living creatures. The bodies were hanged in a sacred grove near the temple and left to putrefy. On one occasion, according to saga traditions, the Yngling king Domalde was sacrificed to appease the gods after he had presided over two years of failed harvests. Recent excavations have discovered the traces of two rows of wooden poles, the longest of which is around 1,000 yards (915 m) long. Probably erected in the fifth century, the rows’ purpose is as yet unknown but some of the post holes contained animal bones, possibly from sacrifices.

In the sixth century a large feasting hall, around 164 feet (50 m) long, was built on an artificial platform south of the temple. The hall had a bow-sided plan, similar to the halls at Gammel Lejre, and may have looked rather like an upturned boat. The hall had several grand entrances, one of which was decorated with wrought iron spiral ornaments. The hall burned down in the eighth century and was not rebuilt. The area around the hall was densely populated by craftworkers. There is evidence of gold, silver, lead, bronze, glass and garnet working.

The kingdom of the Swedes was connected to an extensive network of trade routes through the island trading post of Helgö in Lake Mälaren, which developed in the fifth century. Helgö means ‘holy island’ and gold foil plaques decorated with gods and monsters similar to others found at religious sites in Denmark, suggest that the island was a pagan cult centre where markets were held at festival times. No other site in Sweden of the period has produced so much evidence of trade and manufacture. Jewellery making was a particularly important activity: thousands of broken moulds used for casting bronze brooches have been found. Iron working was also carried on. A hoard of seventy-six sixth-century Byzantine gold coins suggests that Helgö had trade links with the Mediterranean, possibly it was a market for furs. In the seventh and eighth centuries, more exotic objects turned up, including a bronze crosier from Ireland, baptismal spoons from Egypt, and a bronze statuette of the Buddha from India. These objects may have passed through many hands on their way to Helgö, so they are not evidence that Swedish merchants were ranging as far as India. However, Swedish merchants had already begun to establish bases east of the Baltic and explore the Russian river systems, which in the Viking Age they would follow to the Black and Caspian Seas. Grobina in Latvia, where three Scandinavian cemeteries dating to 650 – 800 have been found, was one of the earliest Swedish colonies in the east. One of the cemeteries contained warrior burials with artefacts similar to those found in the Vendel cemetery, while artefacts from the other two shows links to Gotland. There is evidence too that a Scandinavian colony was established at Elblag in Poland as early as c. 650.

Swedish penetration of the Russian river system began early in the eighth century and by 750, Scandinavian merchants were living side by side with Finns and Slavs at the fur trading centre of Staraja Ladoga, on the Volkhov River near where it flows into Lake Ladoga. Scandinavians also headed east to plunder as a remarkable double ship burial found recently at Salme in Estonia spectacularly demonstrates. All previously discovered ship burials contained the remains of only one or two people – a high status individual and sometimes a sacrificed slave to accompany them in the afterlife. These two ships between them contained around forty individuals, all well-built mature males, many of them with obvious battle injuries. Weapons and jewellery decorated in the Vendel style identify the dead as Swedes and date the burial to c. 750. The ships were poorly preserved but enough evidence survived confidently to identify the larger of the two as a sailing ship, the oldest so far found in the Baltic region.

Norway’s rugged geography proved to be a major obstacle to state formation. Overland travel through the mountains was all but impossible for much of the year because of snow so the main links between regions was by sea. Thousands of islands and skerries created a sheltered inshore passage along the coast, the ‘North Way’ from which the country got its name, but despite this seafaring came to an end every October and did not resume until the end of March. Ships were hauled onshore to be stored for the winter in boatsheds and nausts (sheltered hollows). The relative isolation of communities bred local independence and the impressive archaeological evidence for state formation in the immediate pre-Viking period that is seen in Denmark and Sweden is absent from Norway. However, here too political power was gradually being centralised. The best evidence comes from Borre in the Vestfold, the sheltered region on the west side of Oslo Fjord, where there is a cemetery of seven large barrows and twenty-five smaller ones (the Borrehaugene). There probably were once many more as some are known to have been destroyed by quarrying for road stone in the nineteenth century. One of those destroyed contained a warrior’s ship burial but it was not scientifically excavated. In the same area, a great timber hall was built at Huseby in the mid-eighth century. In the Viking Age the area around Huseby was known as Skiringssal, a power centre associated in the saga traditions with a Norwegian branch of the Yngling kings. Less than 2 miles from Huseby at Kaupang (‘trading place’) on Viksfjord, a semi-urban trading place and craft centre had developed by 800, no doubt due to the stimulus provided by the nearby royal centre. Evidence for state formation is more limited but a major barrow cemetery with several burned ship burials, now largely destroyed, at Myklebust on Nordfjorden, points to another power centre in the west of Norway, and other impressive barrows are found at Raknehaugen in Romerike, north of Oslo, and at Svenshaug in Hedmark, in east-central Norway.

At the end of the eighth century, the Scandinavian kingdoms were all still highly unstable. Scandinavia had a relatively numerous class of men who could aspire to kingship. In theory, Scandinavian kingship was elective and any man possessed of royal blood, whether from his father’s side or his mother’s side, was eligible for kingship. Illegitimacy was no bar. However, as power became ever more concentrated, as chiefdoms were subordinated to kingdoms and lesser kingdoms were subordinated by larger kingdoms, the opportunities to rule were becoming ever fewer. With many potential claimants for a throne, succession disputes were common. Joint kingship was a common solution where two rival claimants enjoyed equal support and were willing to compromise, but disputed successions often led to destabilising civil wars. If they were fortunate enough to survive, the losers of these conflicts would be forced into exile but, being possessed of the charisma of royal blood, all was not lost to them. Early Scandinavian rulers were primarily rulers of men rather than territory so any man of royal blood who could attract a warrior following might be recognised as a king by his men even if he did not actually have a kingdom. These ‘sea kings’ could turn pirate and, with luck, might win a fortune, a reputation as a great warrior, and loyal armed following with which to make a new bid for power at home. Or, as the Viking Age progressed, win a new kingdom for himself abroad. A reigning king might also find it expedient to go on Viking raids, to bolster his own reputation and to gain extra wealth to reward his own warriors and keep them loyal so that he could fight off challenges to his authority. Members of the chieftain class, the jarls (regional lords) and hersar (local chiefs) were faced with the same necessities as the growth of royal power began to encroach on their traditional independence. Any man rich enough to own a longship and raise a crew to man it had a strong incentive to go on Viking raids. At the same time western Europe was becoming an attractive target for Viking raids. The long economic recession that followed the collapse of the Western Roman Empire in the fifth century was coming to an end as political stability began to return. As trade with the north increased, Scandinavian merchants had ample opportunities to learn about western Europe’s rich, and largely unguarded, ports and monasteries. The potential spoils of raiding the west would amply repay the increased risks of sailing further afield. The violence that for so long had characterised Scandinavian society was about to spill over into the rest of Europe.