Sailing the Viking Seas

I’ve finished the last of my beads (for now!), and I’m at the Viking Ship Museum in Roskilde. This museum, a short train ride outside of Copenhagen, should not be confused with the Viking Ship Museum in Oslo. The Norwegian Vikingskipshuset houses three of the most famous early Viking-Age ships—the Oseberg ship (ca. 800), the Gokstad ship (ca. 890), and the Tune ship (ca. 910). These ships were all buried in wealthy grave mounds during the first century of the Viking Age. The Danish Vikingeskibsmuseet, in contrast, houses five late Viking-Age ships that were found blocking a channel near the small town of Skuldelev.

Oseberg Ship, Viking Ship Museum, Oslo.
The Oseberg Ship, ca. 800. This ship was excavated in 1904 from a grave south of Oslo. It’s the most complete ship that survives today, but when curators reassembled it from loose fragments in 1909, they had to make some hard choices. Recent reanalysis suggests that the stem should lean further forward, which would prevent the body from narrowing too quickly at the front. The wider, more forward-leaning stem would actually help keep water out of the boat as she cut across the waves. (Viking Ship Museum, Oslo, NO.)

The five Skuldelev ships were scuttled in the 1070s as a defensive measure, restricting access to what was then a royal center at Roskilde. Historians think this barrier may have been built by Harald III ‘the Whetstone’ to discourage or even repel one of his rivals. The five ships that were sunk at Skuldelev represent a cross-section of late Viking-Age seafaring, and they stand in stark contrast to the early Viking-Age ships found in the Norwegian graves. The Norwegian ships were shallow and wide, with hulls that could accommodate troop transport or cargo shipment but weren’t specifically adapted to any particular purpose. The Skuldelev ships, however, break into a few distinct types: coastal ships for fishing and trading, cargo vessels for deep-sea commerce, and personnel carriers for rapid troop transport.

Skuldelev 1 at the Viking Ship Museum in Roskilde, DK.
Skuldelev 1. Built in Norway around 1030, this ship was scuttled to block access to Roskilde in the 1070s. Although perhaps less graceful than the Oseberg ship, Skuldelev 1’s bluff bow and wide beam made her the perfect sailing ship for taking cargo across the North Sea. (Viking Ship Museum, Roskilde, DK.)

It can be difficult to distinguish these ships in publications, and even seeing their remains in person doesn’t necessarily communicate what makes each ship unique. They have all been flattened by a thousand years in the sand, with varying degrees of disturbance, preservation, and recovery. The Skuldelev ships now survive merely as timbers in a cradle. They’re beautiful to look at and impressive to visit, but it’s hard to image what they originally looked like or how they originally sailed.

Fortunately, the Viking Ship Museum has reconstructed each of the Skuldelev ships. It’s taken a lot of guesswork. The original masts, sails, and rigging had all disappeared, and Viking-Age methods for navigating were never written down. So over the past forty years, a new breed of ‘experimental archaeologists’ have drawn on the archaeological record and living traditions of boatbuilding to understand how the peoples of the Viking Age built and sailed these ships.

The Sea Stallion of Glendalough sailing in Roskilde Fjord.
The Sea Stallion from Glendalough. Many visitors look to this replica of Skuldelev 2 for an impression of what a ‘real’ Viking-Age longship would look like. But only 25% of the original survives, so The Sea Stallion is in some ways realer-than-real—what Umberto Eco might describe as an example of hyperreality. Nevertheless, building and sailing this replica has taught researchers to ask new and better questions, as they tack between the traces of the past and the possibilities of the present. (Roskilde Fjord, DK.)

These efforts have produced a small but diverse fleet: Ottar, a reconstruction of an ocean-going merchant ship built in Norway around 1030 (Skuldelev 1); The Sea Stallion of Glendalough, a reconstruction of a 70-man troop transport built near Dublin in 1042 (Skuldelev 2); Roar Ege, reconstructing a Danish coastal trader from ca. 1040 (Skuldelev 3); Helge Ask, a small Danish longship from ca. 1030, which could accommodate about 30 warriors (Skuldelev 5); Kraka Fyr and Skjoldungen, both replicas of a single fishing vessel from western Norway, ca. 1030 (Skuldelev 6). A new reconstruction of Skuldelev 3 is currently underway to replace the recently retired Roar Ege, which will be moved to a dry display for visitors.

In the past few weeks, I’ve had the opportunity to sail Ottar (ocean trader), The Sea Stallion (troop transport), and Skjoldungen (coastal ship), as well as some of the other square-rigged ships that the museum owns. In a few hours on the water, I’ve learned more than static exhibitions or archaeological publications could ever tell me.

The stern of the Ottar, a reconstruction of Skuldelev 1.
Ottar. Sailing Ottar is easy work, when the wind is in your favor. An experienced skipper holds the braces in one hand to keep the sail at a perfect angle for catching wind. With his other hand, he directs a novice steersman toward port. (Roskilde Fjord, DK.)

I’ll start with the deep-sea merchant, known as a knarr from the Old Norse, named Ottar. This was my first experience with a Viking-Age replica. The crew was small, about ten people (more than enough), and the deck was large—but so was everything else! My initial instructions were to stay out of the way until I was yelled for, which didn’t take long. There was always something heavy to move or a rope that needed to be hauled. I spent most of my time at the front, pulling the leading edge of the sail from one side to the other so we could zigzag our way against the wind. The small size of the crew meant that everybody needed to know what was happening with the ship, which demands a certain style of leadership and fosters a certain type of camaraderie. I’m happy to say that my day with Ottar ended in a crew trip for ice cream followed by a cookout.

The Sea Stallion under sail.
The Sea Stallion from Glendalough. This massive square sail can drive The Sea Stallion forward at speeds up to 17 knots (20 mph / 30 kph). But it can also pull the ship over. It takes a large and attentive crew to balance the sail by moving from one side to the other. When the wind fails, these intelligent pieces of ballast transform into oarsmen. And when more care is needed, such as when maneuvering the ship in port, the crew turns into a human set of winches, hauling the boat from pier to pier. (Roskilde Fjord, DK.)

The next day I went sailing on the replica of the large troop transport, The Sea Stallion from Glendalough. This is the pride of the museum fleet and needs a crew of about 60 for safe sailing in open waters. The ship is sleek and fast, even in moderate wind. When the wind pushes the heavy woolen sail to one side and causes the ship to tilt after it, crewmembers who are not immediately engaged with sailing or steering the vessel instinctively move uphill. Their bodyweight helps keep the ship in balance. The basic sailing techniques are otherwise similar to the Ottar, but the extra length of The Sea Stallion means that one part of the crew isn’t always aware of what the other parts of the crew are doing. Chatter is kept at a minimum, and any crewmembers who aren’t involved in an active task keep low and out of the way, so the crewmembers who are managing the ship can see each other and coordinate their actions. The skipper at the stern relies on commands relayed through a caller at the mast and the eyes and judgements of an experienced lookout at the fore. This kind of ship could not be sailed without a sense of discipline, hierarchy, and trust.

The Skjoldungen under sail.
Skjoldungen. The coastal craft Skjoldungen proved the easiest ship to handle. The smaller and lighter sail is comfortably managed by three—one person at the bow to manage the sail’s leading edge or tack; another holding the braces reaching up to control the angle of the crossbeam or yard; and a third holding the bottom edge of the sheet, managing how much wind fills into the sail. A fourth person, usually the skipper, controls the rudder at the stern and commands the crew. A few extra hands are useful so that sailing (or rowing) may be done in shifts. (Roskilde Fjord, DK.)

My trip on the Skjoldungen was an interesting follow up. The Skjoldungen is a small coastal trader that fares well with a crew about the same size as the bulky cargo-carrier Ottar. But with its smaller size, the Skjoldungen handles with an agility that exceeds even the precise maneuvers of the well-drilled crews aboard The Sea Stallion. In 2016, the Skjoldungen was shipped to Greenland, where a small crew sailed and mostly rowed the ship 1000 km (645 mi / 560 nm) up the coast. Memories were still fresh, and the openness of the Greenland seascapes seems to have tightened the closeness of the crew. The veteran sailors rowed with an ease that looked like it was years and not weeks in the making—probably a fitting representation of the original Skuldelev 6, which was used for workaday tasks of fishing and trading and had to be moved regardless of the prevailing winds.

These experiences would not have been possible without the living boatyard at the Roskilde Viking Ship Museum or the active crews, who are always looking for recruits and supporters. For a sample of their activities (including better photos!), visit the Facebook pages for Ottar, The Sea Stallion, and Skjoldungen. Each ship in the boatyard generally does one multi-week trip during the summers, and the crews train regularly during warmer months. You can learn more about the various crews and join or support their efforts here. For more casual visitors, it’s possible to sail for a hands-on tour aboard an authentic clinker-built ship—some rowing required. Details may be found here.

Students from Suffolk County Community College at the Viking Ship Museum in Roskilde, DK.
A pier-side class at the Viking Ship Museum. The museum hosts not just tourists but also educational visits. I had the pleasure to spend several days sailing and traveling to Copenhagen with a group of students from Suffolk County Community College (Selden, NY), who were in Denmark as part of a uniquely immersive course in medieval history. (Viking Ship Museum, Roskilde, DK.)

The Viking Ship Museum is a unique hub for maritime archaeology and research, as well as working crafts and reconstructions, but individual replicas may also be found in other places. The Gokstad ship from Norway has probably been the most popular model for reconstruction. The earliest replica was the Viking, sailing from Norway for the World’s Columbian Exposition at Chicago in 1893 (currently in Geneva, IL). Midwesterners might also be interested in the Hjemkomst Viking Ship, collocated with the Hopperstad stave church replica in Moorhead, MN. Other recent descendants of the Gokstad ship include the Íslendingur, which sailed from Iceland to New York in 2000 (currently in Njarðvík, IS); the Lofotr and Vargfotr, which may both be visited and sailed at the Lofotr Vikingmuseum (Bøstad, NO); and the Gaia, in Sandefjord, NO, near the original Gokstad site.

Replica of a buried ship from Tuna i Badelunda.
Tuna i Badelunda Reconstruction. Although the big ships capture most of the attention, this simple boat immediately caught my eye. It’s a replica of the boat a woman was buried in at Tuna i Badelunda in central Sweden at the dawn of the Viking Age. It’s the kind of boat that helped knit local communities together, and its presence in the woman’s grave suggest that she too held a central role in her community. I’ve written more about a pair of graves from Tuna—and the stories they reveal—here. (Viking Ship Museum, Roskilde, DK.)

The Oseberg ship inspired the reconstruction Dronningen, which ultimately sank in a Mediterranean storm in 1992 along with Saga Siglar, a reconstruction of Skuldelev 1. A thorough reexamination of the Oseberg timbers led to a new and more seaworthy reconstruction, the Saga Oseberg, which is currently in Tønsberg, NO, near to the original Oseberg site. Replicas of both the Gokstad and the Oseberg ships are on display together at the Bergen Maritime Museum (NO). The Norwegian finds also helped inspire the Draken Harald Hårfagre, which began as an effort to imagine what a “great ship” (storskip) from the Norse sagas may have looked like, although it’s not based on any particular archaeological find. The Draken crossed the Atlantic in 2016, and plans are underway for an East Coast tour in 2018. Information on further replicas may be found at and (of course) Wikipedia.

Kaupang before the Coin

During the early Viking Age, a new settlement arose at Skiringsaal in Vestfold, just south of modern Oslo. It was a good place to be. Geologic forces had cut a route across the fjords, offering elites an opportunity to disperse among sheltered inlets while staying connected by land and sea. The trading place of Skiringsaal, now known as Kaupang after a later farmstead, stood at the heart of this region, linking it to the world.

The Gokstad ship.
The Gokstad Ship, ca. 890. This ship was excavated from a grave not far from Kaupang. Similar ships could carry people from Vestfold off on raids or between different ports of trade. Scholars now emphasize this double aspect of the Viking Age—raiding and trading—but how were these stories connected? (Viking Ship Museum, Oslo, NO.)

Tree rings from surviving wood indicate that the earliest structures were built around 803, while the minting dates from surviving coins show that things slowed down after 900, with a brief revival around 960. The beads help complete this picture. Glass imports boomed between 815 and 860, followed by the import and loss of a large number of Islamic coins between 860 and 890, and with a brief return of fancy glass and coin sometime after 900. But bead imports, overall, became sparse during the later Viking Age, both in Kaupang and in other settlements, at the same time that the silver economy was taking over the northern trade.

This picture poses a number of historical problems, such as the connection between Kaupang’s rise to prominence and the Danish royal interest in the region in 813, as well as the sudden flow of silver into Kaupang at the same time that viking activity was escalating in the West in the 860s. But as a scholar who works with beads, this chronology also poses a different set of problems. To state the question bluntly: If silver replaced beads as the dominant eastern import in the 860s, should we treat beads as we do silver, as a sort of currency for premodern trade? And if so, what does that mean about Kaupang and its place in the world?

Early Viking Age beads from Kaupang.
Some of the earliest beads from Kaupang. The black bead is a fragment of a ‘wasp’ bead, which was probably made in Scandinavia. Ribe or Sebbersund in Denmark are two likely places for production. The other two beads are long-distance imports and arrived via Russia. They were both made using a similar technique of piercing the glass—a piece of green mosaic on the left and a purple ball on the right. All three beads went out of style after 820, suggesting that Kaupang was active in regional and long-distance trade before that date. But they’re also relatively rare at Kaupang, indicating that the town’s heyday came somewhat later. (Kulturhistorisk museum, Oslo, NO.)

I think there are a few good reasons to see beads as a means of exchange that preceded silver. To explain why, I’d like to focus on two of the most common imports—segmented beads and cut beads. Both styles were made only in the Near East. The segmented beads were probably produced in the Eastern Mediterranean, and they began to travel north in the 790s, either up the Danube or through the Black Sea. The simple cut beads, however, have a later peak. They belong to the mid-800s when most long-distance imports can be traced from origins in Iran, through Russia, and out through the Baltic.

Because the segmented beads flourished earlier, it’s best to deal with them first. These beads came in many different colors, but they appeared most frequently in contrasting layers of glass that gave off the appearance of silver or gold, as well as in deep cobalt blue. These beads make up about 14% of the beads that I’ve seen from settlements (n=3852).

Beads from Kaupang.
Beads from Kaupang. Over 3000 beads have been recovered from this Viking Age town. These particular beads were excavated during rescue excavations to allow for modern construction. They form a fairly representative sample of the finds. Many of the segmented beads in this image lack a perforation but were still traded north. Most of the small blue, white, and yellow were cut from larger tubes and then reheated to round the edges. These are the most common beads from Kaupang. (Kulturhistorisk museum, Oslo, NO.)

But almost 14% of the segmented beads found in settlements—or 2% of all settlement bead finds—are considered ‘defective’. These beads lack a useable perforation because they came from the end of a glass tube that was blown and drawn, and one or both ends were left sealed. Archaeologists have been aware of this problem for years. They typically assume that segmented beads were imported in bulk, and then redistributors in Scandinavia tossed out the bad ones and sold off the good. If this were true, we should expect to find only a few ‘good’ beads in the settlements and a larger ratio on farmsteads or in graves. But segmented beads make up 14% of settlement finds against only 8% of grave finds (n=3067). So although merchants could sell the perforated beads as jewelry, they were more likely to keep them.

I admit that this might be a chronological problem. Perhaps my graves date from different periods than my settlements, which might explain the different ratios. However, we have a second test. If segmented beads were imported in bulk and then sorted for sale, we should expect the number of defect beads to get smaller at each stop they traveled into Scandinavia. But at Kaupang, which was at the far end of these routes, 18% of segmented beads are defective. This contrasts with 11% at Åhus and 4% at Hedeby, which were both intermediary towns. The surprisingly large number of ‘defects’ at Kaupang suggests that these beads retained their value as they moved from town to town during the Viking Age.

Beads from Viking-Age Norway
Beads from a ninth-century grave at Reine in Buskerud. These beads from the uplands above Kaupang show that this person was tied into the same networks supplying the town with segmented and cut beads—but they make up a much smaller portion of this collection. Many of the green beads in the middle may have been made locally—perhaps even from glass made in Kaupang or Hedeby—and the cylinder beads in the back seem to have come from a workshop at Birka in central Sweden. (Kulturhistorisk museum, Oslo, NO.)

Drawing on these observations, we can question the assumption that segmented beads were being measured by weight and traded as bulk goods. This assumption rests in part upon our knowledge of the later silver market. We have evidence that Scandinavian traders were hacking up silver coins by the 860s, measuring silver according to its weight rather than as a number of coins. Scholars suggest a logical progression—from people valuing commodities like glass for their weight, to people valuing just precious metals, to people valuing coins for the metals they represent.

In fact, the segmented beads don’t break into simple weight groups. Instead, the only thing they have in common is that their segments can easily be counted. Each segment could have been used as a unit of value, like dollars or cents. This would make a lot of sense in light of the subsequent period, when simple cut beads dominate the import market. Like segmented beads, the cut beads don’t break up into a few simple weight groups. Many were so small that they have almost no weight at all, frequently less than 0.01 g. Since the small weights of the later silver market measured about 2.0 g, it could take up to 2000 beads to register on a Viking-Age scale.

Cut glass beads from Kaupang.
Cut beads from Kaupang. This group of 52 cut beads weighs only 1.15 grams, which is less than most of the smallest Viking-Age weights. They weigh about the same as three small paperclips and less than half the weight of a US penny or a Euro cent. (Kulturhistorisk museum, Oslo, NO.)

So it seems to me that the segmented beads and the cut beads provided a way to measure value and make transactions during the early Viking Age. This basis for trade underwent a profound change in the 860s, when glass was abandoned for silver, and units of value were abandoned for weight. The flow of Islamic dirhams overtook the old trade routes used for glass, expanding all the way into England where Arabic coins have been found at viking sites.

To me, this indicates that Islamic imports should be connected to the viking activity of the West. And if silver imports were connected to viking violence, then beads might have been connected as well. So the beads of Kaupang leave us with a few lingering questions: What, if anything, could viking raiders have produced that the sellers of silver and glass and silver could have wanted? What did Islamic merchants want that viking raiders—and the middle men of Kaupang—could supply?

Silks from the Oseberg Ship, ca. 820.
Silk from the Oseberg Ship Burial, ca. 820. Silver and beads weren’t the only things imported into Scandinavia during the Viking Age. People from Vestfold also loved their silk, which they imported as finished products and perhaps also as raw materials for local textile work. These silks seem to have traveled the same routes through central Asia that carried the cut beads north. (Viking Ship Museum, Oslo, NO.)

The amount of materials excavated from Kaupang is vast. My research has been possible only through warm conversations and the generous help of many people here at the Kulturhistorisk museum in Oslo, as well as the solid foundations of previous research, much of which is available online. In particular, I’d point interested readers to:

Gotland during the Viking Age

It’s common wisdom that Gotland went it’s own way during the Viking Age—but that’s certainly not because Gotlanders weren’t connected. On the contrary, Gotlanders had a habit of collecting things that showed just how connected they were. From the thin soils of this rocky isle, archaeologists have uncovered more than 168,000 coins from the Viking Age, which is all the more remarkable since no one in Scandinavia was making coins at this time. Many of the Gotland coins still bear the marks showing when and where they were made, indicating that this idiosyncratic island was tied to trade routes spanning North Africa, Western Europe, the Middle East, and Central Asia.

Stavars skatt, a Viking Age hoard from Hemse in Gotland.
Stavars Skatt. This hoard from the mid-900s from southeast Gotland consisted of almost a thousand silver dirhams and two silver bracelets. Tens of thousands of dirhams have been discovered on Gotland, and although they have often been cut into smaller pieces to be used for their silver weight rather than as minted coins, numismatists have still been able to identify where and when many of these were made. It’s more difficult, however, to know how long it took for these coins to reach Gotland, and how long they circulated before they were buried. (Gotlands museum, Visby, SE.)

But when did Gotland become a hub for this trade? And did these routes exist before the ‘silver fever’ of the late 800s? This is where glass becomes important. During the early medieval period, almost all glass was produced solely in the Near East—Egypt, the Levant, Syria, and Iran. But it was used all over, including on Gotland. If Gotland glass looked different from the rest of Scandinavia prior to the 870s, then we have a sure indicator that Gotlanders were forging eastward connections before they developed their hunger for silver. To this end, I’ve examined some 2500 beads from Gotland (and still counting!). Here’s some of my initial observations.

Necklace beads of glass and fossil from Ire Grave 133B.
Ire Grave 133B, ca. 540-660. This is a typical Vendel Period necklace. The red, orange, and green glass beads at the top of the frame were common throughout the Baltic and appear in large numbers, for example, on Bornholm. These beads were probably imported in finished form. The white beads, however, were made from local fossils and rarely circulated beyond Gotland. (Gotlands museum, Visby, SE.)

To begin with, Gotland beads looked a lot like the rest of the Baltic during the Vendel Period preceding the Viking Age. This is an important baseline, because it shows that as the Iron Age transitioned into the Viking Age, people around the Baltic were largely consumers feeding off a single market—albeit one that was dispersed and had few if any major hubs. But Gotlanders did have local pride, and they fashioned beads made from some of the fossils found readily in the limestone bedrock that makes up much of the island. They made cylinders from the stems of ancient sea lilies, and they ground medium-sized round beads from the coral reefs that had been home to the world’s first vertebrates. These beads are relatively rare in other places, at least throughout the western Baltic which I know best, reinforcing the impression that Iron-Age Gotlanders were entering long-distance markets as consumers rather than as producers of exotic goods. They abandoned the cylinder beads perhaps before 700, although other fossil beads continued in use throughout the Viking Age.

Vendel Period fossil beads.
Vendel Period Beads from Gotland, ca. 540-660. The white cylinder beads were made from fossilized sea lilies (crinoids) common in the Gotland limestone. The unworked fossils at the front still have the appearance of plant stems, but the beads have been ground and polished into the smooth white appearance that Vendel Period Gotlanders seem to have preferred. The round white beads were mostly made from tabulate coral fossils, which often turn a buttery yellow when lit with a strong light. (Gotlands museum, Visby, SE.)

In the early 700s, new trading sites began to appear across southern Scandinavia at places like Ribe and Åhus. At first, these were mere trading camps, set up seasonally, but later they became more permanent towns for craftwork and exchange. Beadworkers lived in these towns, and their products traveled north among the elite communities living in what is now Sweden. But I’ve seen very few of these beads on Gotland. This tells us not only that Gotland was peripheral to the economic expansion that was tying the Baltic closer to Western Europe during the Merovingian/Carolingian transition. It tells us also that pre-Viking Age Gotlanders didn’t depend on these networks for access to glass—they must already have had some access via non-western routes, probably still making use of the dispersed networks that had been their basis for exchange throughout much of the Vendel Period.

Glass beads from Ribe, 725-760.
Beads from the Ribe Posthuset Excavation, 725-760. These particular beads were made (and lost) at a seasonal campground for craftsworkers in western Denmark. The blue beads decorated with red, white, and sometimes yellow were made primarily in Ribe and later at Åhus in southern Sweden. They occasionally spread to elite sites northward throughout Sweden, although I’ve seen very few in the collections on Gotland. Even as western craftsmen and merchants were increasingly sailing into the Baltic, Gotlanders looking for trade must have been seeking other routes. (Sydvestjyske museer, Ribe, DK.)

The only site where these new Scandinavian-made beads appeared in appreciable numbers seems to have been at Paviken, which was a trading site established on Gotland’s west coast perhaps as early as 750. Imports from 750 to 800 were primarily restricted to the generic colors of green, white, and blue. These colors appear not only dominant among the finds at Paviken, but also as the exclusive elements of bead assemblages in other places as well. For example, an elite grave at the old cemetery of Ire on Gotland’s east coast includes melted beads of green, white, and blue, and it should probably be dated to this period. Similarly, these colors make up the entire palette of the molten beads found near the Fröjel picture stone, indicating that this otherwise undatable monument was probably set up at about same time, in the years just prior to 800.

Glass beads from Paviken, Gotland.
Beads from Paviken, Gotland, ca. 750-850. This is a typical selection of beads from the site. The blue bead at the top left is a rare example that could be classified as being Ribe-style, suggesting that although traders may have been stopping at Paviken as early as the mid-700s, their visits were probably few and brief. The segmented beads at the top right show stronger connections to the early Viking Age towns of Hedeby and Åhus, where this style seems to have been a major import during the early 800s. The beads on the bottom with criss-crossed lines have long been associated with Birka, which was also growing at this time, although they appear in other places as well. But the red-and-black checkerboard at the right has few comparisons in other western collections, indicating that Gotlanders still tied into other trade routes pointing toward the centers of glass production in the Near East. (Gotlands museum, Visby, SE.)

Beads from the trading town of Paviken show increasing similarities to the west beginning in the early 800s. Beadmaking may already have been abandoned in northern Europe at that time, but the growing trading hubs were developing more-or-less direct connections with the Near East. Previously, most beads had been made from molten glass that had been wrapped around metal mandrels. But the new beads were made from glass that had been drawn or blown into tubes, and then formed into a desired shape like small discs or multiple ‘segments’ joined by narrow waists. Distribution patterns suggest that this technique might have been practiced solely near the primary production centers around the Eastern Mediterranean. Hedeby in southern Jutland was probably a main point of entry for these beads into the Baltic networks, with Birka in central Sweden and Truso in northern Poland as regional redistribution hubs. Paviken was able to shunt off some of this trade between Hedeby and Birka, with what appears to have been an increasing degree of success for a short period after about 830. Notably, similar beads have also been found around a pair of picture stones at Buttle, suggesting that these stones may also have been raised perhaps around 850.

Mandrel and Bead from Paviken, Gotland.
Mandrel and Bead from Paviken, Gotland, ca. 750-850. Despite the abundant evidence for beadmakers at sites like Ribe and Åhus in the 700s, most Viking Age glass beads seem to have been imported in finished forms. This unique find from Paviken, however, reveals just how Viking-Age beads were made. The beadmaker would have used an iron mandrel like this one (probably with a wooden handle that has decayed and disappeared) and wrap glass beads around it. They would usually have coated the mandrel with clay, so that they could slip the bead off when it was finished. Traces of this clay often survive fused to the glass interior of the bead, although it has dissolved from the mandrel. This bead also has a single depression, which is where the beadmaker added glass of a different color to create an eye. But since these different glasses had different chemical properties, they separated as the glass cooled or aged, which is a fairly common occurrence among archaeological finds. (Gotlands museum, Visby, SE.)

At the same time, however, new styles of beads begin to appear in the Gotland collections. Among the beads from Paviken, there are a few examples in turquoise glass with few if any parallels from the ninth-century West. Paviken also has a few examples of mosaic beads—made by a special technique of stacking glass so that it has a cross-section with a desired image or pattern, then placing tiles cut from these cross-sections together and wrapping them around a mandrel, so that they make a bead—which I have seen almost nowhere else. (Western sites also have mosaic beads, but not with these particular mosaic patterns.) Presumably, this means that Gotlanders were continuing to develop their own connections east, not mediated by the traders of Birka or Hedeby. Perhaps they had been inspired by their far-traveling neighbors, or perhaps they were driven by a desire to compete, but by the mid-800s, Gotlanders were surely seeking ways to cut out the middle man in their pursuit for eastern imports.

Beads of glass and cowrie shell from Ire Grave 218A.
Ire Grave 218A, ca. 850-900. This massive necklace of 217 beads was buried with a seven year-old girl, testifying to the hopes that Viking-Age families placed in their daughters and the sorrow of their loss. Most of the artifacts from this grave date to around 900, although at least one was in a style that didn’t become common until after 950. But judging by the beads, this burial either included a few old heirlooms or in fact occurred closer to 850. A date closer to 850 would suggest that the white seashell beads were coming north in large numbers before Gotlanders established their central position in the Viking-Age silver trade. (Gotlands museum, Visby, SE.)

One of most significant symbols of their success seems to be a sudden influx of cowrie shell beads, which appear in large numbers in the decades around 900, around the same time that silver began to flood north. Numismatic studies indicate that much of this silver was mined in the Hindu Kush between present-day Tajikistan, Afghanistan, and Pakistan, with a large portion reminted on its way via the Jewish Khazars living north of the Caspian Sea. But these cowrie shells came from even further afield, from the shores of the Arabian Sea. Interestingly, many of these shell beads seem to have been coated with some sort of a resin, perhaps to make them shine with a sparkly gloss or maybe even to give them the false appearance of thin but resilient beads made from amber. Regardless, these beads show that Gotlanders had achieved their own direct contacts east. More importantly, cowrie shells are almost entirely absent from the Paviken collections, suggesting not only that this trading town failed and folded before the silver tide began to flow, but also that Gotland’s trade was based not on urban merchants but rather on elite enterprise.

Beads from Kopparsvik Grave 189
Kopparsvik Grave 189, ca. 900-950. Artifacts in this grave allow us to date the burial as early as 900, but based on bead styles, I’d certainly place this collection later than Ire Grave 218A pictured above. The white beads are again cowrie shells, although many of them have been coated in a resin that at least now is a dirty brown. The lumpy dark beads with eyes sometimes mixed with lines became common across the northern world during the mid-900s, and these may in fact be some of the earliest examples in Scandinavia. Their later spread across the Baltic and into the North Sea helps illustrate how Gotlanders cemented control over long-distance networks and became trend-setters along the way. (Gotlands museum, Visby, SE.)

This story is, of course, not complete. I’m exploring ways to develop more precise dating for the beads, since currently I’m relying only on beads found in contexts with other datable objects. I also need to think carefully about what, if anything, these interpretations can tell me about the Viking Age slave trade, which is the primary subject of my research. In particular, it raises questions about who controlled the slave trade, when they might have controlled it, how far they could have trafficked their captives, and in what volume. Furthermore, my research methods have caused me to examine a lot of later materials that don’t bear directly on my research questions but may nevertheless lead to better analysis through comparison. For example, the contrasting beads from the merchants of Paviken and from the elite cemeteries of Ire and Barshalder suggest networks of exchange that diverged and ultimately conflicted. In contrast, the beads from the later town of Fröjel and the elite cemetery of Kopparsvik outside Visby suggest that elite and mercantile networks converged and ultimately reunited as the Viking Age drew to an end.

Fröjel Church, Gotland.
Fröjel Parish, Gotland. After Paviken declined in the mid-800s (an observation based solely on my interpretation of the beads excavated there), a new trading town subsequently sprung up at Fröjel, apparently in the late 900s. A small elite cemetery was excavated nearby, indicating that Gotland elites were learning to live as members of more urbanized trading communities. (Fröjel Parish, Gotland, SE.)

This post has been longer than most, but I write it with thanks to the staff of Gotlands Museum, whose hard work in supporting this research has been surpassed only by their hospitality in welcoming me as a guest. The researchers at the Uppsala University Gotland Campus have also generously offered me access to collections from their recent excavations, as well as fruitful conversation. It may take several years for this research to move from dissertation to publication, so I hope that during the interim, this brief summary may serve as a useful aid as they continue to develop their collections and support other students and researchers.

Making Love in an Iron Age

Sometimes it’s hard to talk about how we feel. Whether it’s at the doctor’s office or in a relationship, physical and emotional realities can be difficult to describe. Language never quite captures reality, and clichés often take the place of sincerity. But when words fail to serve us, our actions can capture the ways that we feel through the things that we do.

The gulf between unspeakable feeling and meaningful action is ever present in the archaeological record. Behind the plexiglas and placards of museum displays are artifacts taken from cemeteries and graves. These are traces of people who gathered together for a final opportunity to express the things that words could no longer say. As a scholar who works with these artifacts, it’s often hard to know exactly what they mean—but sometimes it’s all too clear.

Iron-Age burial mound at Tibble, near Tuna i Badelunda.
Most of the cemeteries I study have been destroyed by farming or modern construction—or else they wouldn’t have been excavated. This unexcavated burial mound sits atop a ridge near Badelunda in central Sweden, giving a sense for how other cemeteries in the area may have felt during their period of use. (Badelunda Tibble, Västerås, SE.)

I was recently working with artifacts from Tuna i Badelunda, a cemetery from the late Iron Age in central Sweden (ca. 600–1100 AD). No two Iron-Age burials are alike, but at Tuna, there was a clear preference for cremations, with only some of the bones being buried, often a mix of burnt and unburnt artifacts, sometimes even ships, and then surface monuments made of large stones, often laid out to resemble the ships beneath them. These graves were not for everyone, but for a rich elite who could command large numbers of mourners drawn from the dispersed settlements of the Mälaren valley. In a few places, the graves were laid in rows, suggesting that a person’s place in the community could be just as important as their individuality.

Glass beads from Grave 11 at Tuna i Badelunda.
The beads from Grave 11 at Tuna i Badelunda. The plain beads are common styles of the late Iron Age, with the bead in the back right falling out of fashion around the year 700. The decorated bead is a bit more unique, and may have been in fashion during the early 700s. (Västmanlands läns museum, Västerås, SE.)

Grave 11 is one such burial, added to a growing row of ship settings. This must have been an impressive and demanding funeral. It began amid animal sacrifices and a large cremation pyre. The pyre burned hot enough to render most of the bones beyond recognition. Once the fire cooled, someone went through the bones and selected 100 g of fragments representing the deceased human and the accompanying animals. These were then taken to the grave, where unburnt objects were added—slag that linked the deceased to objects of iron not in the grave, a single shard standing in for the whole of a pot, and three-and-a-half beads, presumably selected from a larger jewelry assemblage. Then the mourners built a mound on top of the grave, and the person who had been placed in the fire was now gathered into a community of stone.

Glass beads from Grave 14 at Tuna i Badelunda.
The beads from Grave 14 at Tuna i Badelunda. Three of these beads have been melted in fire, probably during cremation. The two fragments on the left with yellow and white lines make up one complete bead. The similar fragment with only white lines on the right has its matching fragment in nearby Grave 11. (Västmanlands läns museum, Västerås, SE.)

Grave 14 came later, wedged into the row right next to Grave 11. The funeral was similar and echoed earlier events: cremation, a careful selection of bones and artifacts, the construction of a monument. There were few artifacts, although the two ends of a belt suggest that the deceased was male. Excavators also found five glass beads, leading some archaeologists to suspect this may have been a female burial, but I think differently.

Whereas the beads in Grave 11 are a typical mix for Iron-Age women in their prime—a smallish group of simple beads with one or two unique accents—the beads in Grave 14 are something different. They all have some sort of decoration, and they include no plain beads at all. These were probably not part of a woman’s necklace. Moreover, some of the beads are burnt while others are not, reinforcing the impression that these were treated as individual objects rather than as part of a group. Most importantly, one of the unburnt fragments precisely matches the fragment laid in Grave 11.

Artifacts from Graves 11 and 14 at Tuna i Badelunda.
Together at last. Artifacts from Grave 11 on the left and Grave 14 on the right. In the center are matching fragments of a single bead, found with one half in each grave. (Västmanlands läns museum, Västerås, SE.)

This bead was broken along its center, and it could not have been restrung. The edges where the bead was broken are more worn on the piece from Grave 14. Between the burial of Grave 11 and the funeral for Grave 14, this fragment must have been carried and handled. It was a public token of grief, visible when Grave 11 was buried with one half and brought to a fitting end when the other half was laid nearby in Grave 14. Indeed, while the artifacts from Grave 11 point to things left out of the burial and a sense that something was missing—slag leftover from an absent artifact of iron, half of a broken bead, and a thin selection of bones—the things included with Grave 14 point toward a desire for fulfillment—a hefty 680 g of burnt bone, the matching ends of a belt, the missing half of the bead.

Two matching bead fragments from Graves 11 and 14 at Tuna i Badelunda.
Token of a broken heart? Two fragments of a single bead, buried in separate graves and placed back together again for the first time in a thousand years. (Västmanlands läns museum, Västerås, SE.)

It can be difficult to speak definitively about how people in the past experienced their lives and relationships. Indeed, it can be difficult to speak of our own experiences of love and life. But in this case, it seems the signs are clear. Even in an Age of Iron, it hurt to be separated from your other half, and love could be as fragile, as enduring, and as achingly beautiful as a broken bead of glass.

A special word of thanks is due to my own better half, who is celebrating her birthday today without me. We’re both looking forward to being together again!

Forging a New Elite for Viking-Age Funen

During the late 800s, a new elite established themselves on the island of Funen in the Danish archipelago. Perhaps they rose from the ranks of local farmers, or perhaps they migrated there from wealthier parts of Scandinavia. But whatever their origins, they saw a need to set themselves apart from previous generations. They did so in part by burying their dead in a new set of places.

Glass beads and amber beads from Viking Age Denmark.
Beads from nine graves at Kildehuse. Like other Viking-Age cemeteries found on Funen, Kildehuse produced only a small selection of beads. These mostly date to the 900s, giving the impression that Funen was a relatively quiet place for much of the 800s. (Odense Bys Museer, Odense, DK.)

Grave ACQ at Køstrup inaugurated one of these cemeteries. The attention that ACQ received indicates that she was either a powerful matriarch or that her survivors had hoped she would become one. They buried her in the height of fashion. She wore a simple linen underdress, and on top of it, another garment dyed blue with woad. Pleats ran down the center. Not only was this a conspicuous waste of fabric in a period when textile production was labor intensive. The vertical lines also drew the eye upward from the blue garment toward the cluster of jewelry at ACQ’s breast.

Brooches excavated from Køstrup Grave ACQ on Funen, DK.
The bronze brooches buried in Grave ACQ at Køstrup, DK. Archaeologists classify these as JP 51a, referring to an illustration in an early catalog of Viking-Age artifacts. Similar brooches are often found in contexts that date from 850-950, tending toward the early part of that range. (Illustration from Lindblom 1993: 153.)

Two bronze brooches provided support, pinned to the shoulder straps. These were in a garish style that archaeologists call JP 51a. They stood out like tortoise shells with intricate designs that could catch light from any direction. When ACQ was alive, these brooches would have sparkled as she moved. Now she was dead, but her brooches still caught fresh light each time the mourners moved around her, drawing their eyes back to the dazzling objects at the center of her grave.

Reconstruction of the dress and jewelry assemblage from Køstrup Grave ACQ.
The Viking-Age dress from Køstrup, as the woman buried in Grave ACQ may have worn it. Note that archaeologists aren’t sure whether she wore the beads at the top or the bottom of her brooches. Also, the patterns at the top of the dress were likely woven with colored thread, although these colors do not survive today. (Illustration from Rimstad 1998: cover.)

Between the brooches lay a string of eight beads. This is the largest Viking-Age necklace currently preserved in the museums of Funen. ACQ’s survivors must have known that women elsewhere could boast much more impressive displays. Perhaps to make up for this lack, ACQ’s dress had a thin strip of fabric running from brooch to brooch, woven with designs to help make her modest necklace seem like part of a larger, more colorful display.

Necklace beads from Køstrup Grave ACQ.
Necklace beads from Køstrup Grave ACQ. By studying where and when similar beads appeared, we can learn more about ACQ, her community, and the networks they were involved in. (Odense Bys Museer, Odense, DK.)

Each bead has its own story to tell. The darkest bead glows purple when held to the light. Similar beads have been found in towns like Ribe and Åhus, showing that ACQ had access to developing markets. Just as importantly, these other sites reveal how quickly this style went out of fashion after about 850, indicating that ACQ was not buried much later than 860. This was a period when viking activity was drastically increasing in the west, and it would be surprising if ACQ and the new elite were not somehow connected to the plunder and wealth of this so-called Great Heathen Army. This bead, then, offers tantalizing clues about ACQ’s place in the trading and raiding that defined the Viking Age.

Grave diagram for Køstrup Grave ACQ.
Beads were just part of ACQ’s grave, which is considered the richest burial in the Køstrup cemetery. Her brooches lay off-center, suggesting that she was buried on her side. At her hip lay a knife and a key, which presumably hung on a long cord from her brooches. At her head was a small casket made of maple. She was placed in a coffin, but this must have stayed open as all these things were arranged. A discoloration at the foot of her grave suggests that a pole or wooden stake marked the spot where she was buried, which became a focal point for the cemetery around her. (Illustration from Lindblom 1993: 153.)

The other colored beads would be more difficult to date if we didn’t have the short-lived purple bead as a point of reference. Several of these styles were especially popular during Funen’s glory days two centuries before, when the shrine of Gudme attracted pilgrims and may have inspired the earliest cycles of Norse mythology. Not only do these beads suggest deliberate connections to the past; so too does the burial site. Køstrup had already been used as a cemetery once before, and even though the people buried there must have been long forgotten, they left a distinctive burial mound in their place. The people who dug a grave for ACQ and started a new cemetery right next to this ancient mound were making strong claims about their abilities to dominate both the island landscape and its mythical past.

The clear beads also tell a story. These are made of rock crystal, a common variant of quartz. I’ve found this particular style—finely rounded, a bit larger than a centimeter, and almost perfectly clear—in a few specific spots: the boat burials of central Sweden, the merchant’s colony at Hedeby, and the aristocratic hoard of Lille Karleby, a site located suggestively close to homelands of the Danish monarchy. These diverse finds show that ACQ was linked into the major changes of the Viking Age: a thriving culture which treated warships as prestigious symbols, a network of trading towns promoting urban crafts and long-distance exchange, and growing kingdoms that would soon ally themselves with the Christian church, ushering in the end of the Viking Age and the dawn of the Middle Ages.

For reconstructions of ACQ’s dress, see:

For further information about Køstrup and Grave ACQ, see:

  • Charlotte Lindblom, “Køstrup – en nordvestfynsk vikingetidsgravplads,” Fynske Minder (1993): 143–168.
  • Charlotte Rimstad, “Vikinger i uld og guld,” Speciale (Copenhagen: Copenhagen University Forhistorisk Arkæologi, 1998).

For the reference catalog for Viking Age jewelry, see:

The Limfjord in the Viking Age: Centralization and Catastrophe

The Limfjord cuts across northern Denmark, offering a short and well-sheltered route from the North Sea to the Baltic. As maritime traffic picked up in the late 600s, ships began to ply the Limfjord more often. Traders coming from the commercial centers of the Frisian coast were taking their enterprise north, and by 705, they established a seasonal trading camp at Ribe in western Denmark. It gave them a final base before braving the North Sea into the Baltic. They carried beads with them wherever they went, and a few of their beads ended up in the young Limfjord settlement of Bejsebakken, indicating that they were using this route for their traffic by about 750.

Beads of orange, red, blue, and green glass from the Iron Age, from the cemetery at Lindholm Høje.
An elite circle of society made their home along the Limfjord as early as the 400s, rising to prominence just as Roman governance ebbed in the West. These beads were likely made in the Byzantine Empire and headed north into the Limfjord between 540 and 660 AD. (Lindholm Høje Museet, Nørresundby, DK.)

In the mid-700s, the seasonal camps of southern Scandinavia had just started to transition into a more permanent network of trading towns. Ribe might have led the way, perhaps around 750. Åhus, Ribe’s sister site in southern Sweden, made a similar transition before 790. And the Limfjord was likewise affected, with a new settlement picking up at Sebbersund at about the same time.

A “wasp” bead from Sebbersund, ca. 760–790. One of the earliest artifacts from Sebbersund, which would later develop into a major trading port for traffic taking the Limfjord between the east and west coasts of Jutland. (Aalborg Historiske Museum, Aalborg, DK.)

What these three sites show us—Ribe, Sebbersund, Åhus—is that on the very cusp of viking raids on England, Ireland, and France, a stable network of settlements had just recently been formed in Scandinavia. These settlements offered a reliable route for shipping goods out of the North Sea and into the Baltic, and the Limfjord was the linchpin that held this network together.

But the sites of the Limfjord share a common problem with many sites from the Viking Age. Although we have a large number of artifacts that can tell us about the early and late periods of these places, there’s not much to fill in our knowledge of the middle decades of their existence.

A model of the Lindholm Høje cemetery and an adjacent settlement, as they may have looked during the early Viking Age, ca. 800.
A model of the Lindholm Høje cemetery and its adjacent settlement, as they may have looked during the early Viking Age, ca. 800. Lindholm Høje sits atop a hill overlooking the Limfjord, and in windy weather it sounds just like the sea. Many of the graves have stones around them, laid out in the shapes of ships—a strong indicator of just how important sea travel was for the early Viking Age residents of the Limfjord. (Lindholm Høje Museet, Nørresundby, DK.)

The early phase ended sometime in the mid-800s, as the flow of glass beads into Sebbersund ground to a halt. A single coin minted by Louis the Pious between 822 and 840 was lost at the fledgling settlement of Aggersborg, indicating that the Limfjord was briefly but abortively linked into the coin economies of Western Europe. And the pagan cemetery at Lindholm Høje was first restructured and then subsequently abandoned after almost 500 years of continuous use. Local communities were reinventing who they were by redefining the ways they lived and died.

Blue and yellow drawn beads of early Islamic glass.
Tiny drawn beads are some of the last datable artifacts from the early phase of settlement at Sebbersund. These beads probably date to 800–850 and were made in the Eastern Mediterranean—probably in the early caliphate’s glass factories in Egypt or Syria. Similar beads have been found at places like Ribe and Hedeby, which also developed into important trading towns during the Viking Age. (Aalborg Historiske Museum, Aalborg, DK.)

When archaeologists regain clarity in the mid-900s, the Limfjord region looked completely different. Sand dunes had covered the pagan burials at Lindholm Høje, and the residents of Sebbersund had begun to bury their dead alongside one of Scandinaiva’s earliest churches. The town of Aggersborg was burned to the ground, and on top of its ashes, the Danish king Harald Bluetooth had built a huge Trelleborg-style fortress.

Viking Age combs from Sebbersund, Jutland, Denmark.
My research focuses on east-west routes during the early Viking Age, but the Limfjord was also an important north-south route. It seems that during this period, the Limfjord also had a channel leading north to Norway. Antlers taken from reindeer and caribou from northern Norway—perhaps traded from the nomadic Saami tribes—were crafted into fine combs at places like Sebbersund on the Limfjord. People proudly carried these items as badges of their good hygiene and their ability to acquire exotic goods. (Lindholm Høje Museet, Nørresundby, DK.)

Aggersborg dominated the maritime crossroads between England, Norway, and the Danish archipelago. Its walls stood 240 meters apart, enclosing 48 longhouses, each 32 meters in length. It was a remarkable sign of Danish royal power for anyone traveling through the Limfjord. Harald Bluetooth’s great-grandson Harthacnut installed a royal mint nearby at Aalborg. And one of his successors used the fort as a gathering point for his fleet—part of a failed attempt to invade England in 1085.

Ship stone setting from the Iron Age and Viking Age cemetery at Lindholm Høje, northern Jutland, Denmark.
A ship setting from Lindholm Høje. The weather in northern Jutland often mixes fog and wind, giving the impression of the spray of the sea. Many graves at Lindholm Høje have these ship-shaped stone settings, and it’s easy to imagine the dead were thought to sail off into the afterlife, much like the mythical Baldr, son of Odin. Sand dunes covered these graves during the 900s, until they were uncovered by archaeologists a thousand years later. They make an unearthly site to visit today. (Lindholm Høje, Nørresundby, DK.)

The Limfjord had evolved from a wayside for Frisian traders into the locus of Danish royal power and imperial ambitions. But sometime shortly before 1200, the shifting sands of the Jutland coast closed its access to the North Sea. This was no minor crisis. The Limfjord, once a proud commercial highway and the mustering place for armies, had turned suddenly into a brackish backwater. The suddenness of this change offers a grim reminder that even small environmental changes can rend a political and economic fabric that had been centuries in the making.

Glass beads from a cremation burial at Lindholm Høje, Denmark.
Lindholm Høje has 549 identified graves, spanning from the fifth to tenth centuries. It’s one of the most important archaeological sites in Scandinavia. Many of the dead were cremated together with their grave goods, so it’s difficult to identify a large number of the artifacts. Nevertheless, the number of blue beads in this group, and the complex mosaic patterns still visible on the big lumps of melted glass, allow us to assign this particular grave to the early 700s. (Lindholm Høje Museet, Nørresundby, DK.)

Consuming Beads: A Tentative Chronology of the Viking Age

This week I finish the second stage of my research, so I’d like to take a few moments to sum up what I’ve seen. In previous posts, I’ve documented highlights from my visits to Copenhagen, Lund (SE), Bornholm, and Schleswig (DE). Since then, I’ve been on whirlwind trips to Ribe, Aalborg, Odense, and Langeland. At this point, I’ve completed a survey of over 6,000 beads from the Viking-Age, a large number of which have yet to be published.

Has it all been worth it? I’ll let you decide. In the next few paragraphs, I’ll sketch a quick chronology of early Viking Age bead consumption as it now appears to me. I’ve seen pre-Viking Age beads from ca. 700 at central places like Uppåkra and Sorte Muld, as well as from the trading camps of Ribe and Åhus. And I’ve seen beads from the full flush of the Viking Age from the burgeoning emporia of Hedeby and Sebbersund, as well as from the small cemeteries of the Danish archipelago. These early and late beads look dramatically different, and not only do they come from different places, they also come from different kinds of places. Between 700 and 900, a whole new set of consumers gained access to necklace beads, and they were using them in a whole new set of ways.

Phase 1 (660-700). Scandinavian society revolved around central places during the early middle ages, otherwise referred to as the Germanic Iron Age. Elites who built their power at these sites distinguished themselves by showcasing exotic objects made from materials like glass and gold, which could not be obtained locally. They left glass and gold as votive deposits at places like Sorte Muld and Uppåkra, and they buried them with them when they died. Their societies stabilized with the rest of Europe as northern climates recovered from the ‘Late Antique Little Ice Age’, while the enduring strength of Eastern Mediterranean economies meant that access to exotic goods remained consistent throughout this period. This consistency contributed to a conservative sense of fashion, with styles of clothing and jewelry changing only slowly. Beads tended to be simple but made from high-quality glass. Favorite colors like blue, green, and white would have complemented the prominent blues of women’s dresses.

Beads attributed to Bækkegård Grave 109, ca. 630–660. Similar necklaces featuring white, blue, and green glass beads were common from 630–800. This necklace also included decorated beads, but these were all mixed up during shipment and can no longer be assigned to particular graves.  (National Museum of Denmark, Copenhagen, DK.)

Phase 2 (700-760). Western Europeans built on the improving climate with agricultural reforms and commercial enterprise. North Sea merchants carried this prosperity into Scandinavia by partnering with Danish elites to establish a trading camp at Ribe, a sheltered spot where coastal traders could exchange wares with the deep-sea merchants who traveled around Jutland into the Baltic. Soon the camp at Ribe had a companion market at Åhus in Sweden. The old elites must have watched these sites carefully, but craftspeople increasingly worked on their own terms, outside the patronage networks of central places like Uppåkra and Sorte Muld. They engaged in traditional work with local materials like amber and antler, and for the first time Scandinavian craftspeople also gained proficiency with glass. The new markets secured steady access to this exotic good, while aspiring elites were eager to consume the new fashions being made. These beads—typically a translucent blue glass decorated with red, white, and yellow rings—moved out from the fledgling markets and into the most prestigious circles of Scandinavian society.

Blue beads with decorated with complex lines of red and white, probably made in Ribe or Åhus and buried among elite families on Bornholm, 700–760. (National Museum of Denmark, Copenhagen, DK.)

Phase 3 (760-790). This seems to have been a period of retrenchment. The glass in Ribe and Åhus came from major production centers in the Near East, which prospered as the Islamic conquests put an end to the perennial conflicts between Byzantium and Persia. But the caliphate overextended, and in the 750s, it began to break apart. Distant provinces revolted and a major coup rocked the center. This interrupted the supply of new glass to Scandinavia, where glassworking faltered. The ubiquitous blue beads disappeared and were replaced by thin ‘wasp’ beads—a style that maximized length and minimized material. Many of these beads were black with yellow rings, but they appeared in other colors as well. Access to glass was the determining factor, and bead makers weren’t terribly concerned with color. From this perspective, the late 700s were bleak. Ribe’s trade restructured and Åhus may have been abandoned. Craftspeople and merchants dispersed to a looser but more robust network of smaller trading sites. Some of these would later flourish, but in the uncertain years of the late 700s, most remained ramshackle affairs that have left few archaeological traces.

Wasp bead from Sebbersund, ca. 760–790. One of the earliest artifacts from Sebbersund, which would later develop into a major trading port for traffic taking the Limfjord between the east and west coasts of Jutland. (Aalborg Historiske Museum, Aalborg, DK.)

Phase 4 (790-820). Glass imports renewed during this period, and the loose network of small sites began to consolidate around a few urban nodes. These sites show intensifying relationships with the Near East, no longer mediated through Francia and the Western Mediterranean. Islamic coins were circulating in Scandinavian markets by the 780s, and in the 790s, bead imports spiked. These beads came in a few standard styles made from drawn glass, which couldn’t be replicated in the north. Scandinavian glassworkers could make exquisite beads by heating glass and wrapping it around a mandrel, but they lacked the technology or expertise to blow glass, draw it into tubes, and form it into a desired shape. Bead imports proliferated at Ribe and the revived settlement of Åhus, but they’re curiously rare at elite sites and cemeteries. In part, this is because a large number of the beads lacked perforations, which raises questions about what exactly they were being used for. Overall, it seems that although craftspeople were still occupying traditional places, Scandinavian connections and consumption patterns were beginning to change.

Drawn beads found in a craftsworking site at Åhus, ca. 790–820. (Lund University Historical Museum, Lund, SE.)

Phase 5 (820-860). During this period, a new set of settlements left the old ones behind. Elite women stopped losing their beads at the central places of the Germanic Iron Age, which gives them a sense of abandonment. Ribe and Åhus also disappear from the archaeological record—if these communities persevered, they moved to new sites yet to be identified. Meanwhile, a different set of settlements began to take off. Places like Sebbersund and Hedeby had been among the trading posts that popped up in the late 700s, but only in the mid-800s did they became complex and densely populated sites. Their expanding trade included a new style of drawn bead—tiny rings of blue, yellow, white, and black. These beads rarely made it into elite graves, although hundreds were found in the so-called Hedeby harbor purse. This set of beads was found packaged with a handful of coins, suggesting that they might have served a monetary function as well. If so, they add a new dimension to our understanding of this period. Islamic coins were still rare, and most coins from this period came from Western Europe. But the Hedeby harbor purse suggests that Scandinavians were also forging connections east, well before the silver fever began in the 850s.

A selection of the 600 small drawn beads found in the Hedeby Harbor purse with seven silver coins, ca. 820–860. (Schloss Gottorf, Schleswig, DE.)

Phase 6 (860-900). This was another period of extreme disruption in the Islamic world, as short-lived caliphs struggled to control the Turkish slave-soldiers whom they had empowered. Silver imports slowed, as did glass. Meanwhile, Christianity was taking root in Scandinavian towns, and the new Christians quit burying their dead with grave goods. This led to a declining demand for beads in some places, even as a new demand sprang up in the Danish archipelago. A form of Norse paganism was taking shape there, building its mythology around the old cultic site of Gudme. People in the area started burying their dead with grave goods like necklaces, even as their Christian counterparts were giving it up. These cemeteries tend to be modest, suggesting limited material wealth, but several graves contain an extra body—presumably a slave sacrifice. These island burials contrast to the trading towns, which evidence economic distancing from the Islamic world and cultural convergence with the West. The appearance of glass beads in the Danish archipelago conversely suggests that not only did some Scandinavians maintain contact with the Islamic world, but that these Scandinavians also had access to extra human bodies—at the same time that vikings were reaping captives from the west and Islamic elites were seeking a new source of slaves for their harems and armies. This gives much food for thought about the role these island communities played and the potential extent of human trafficking in the Viking Age.

Stengade Grave BØ, ca. 950-1000. A woman aged 40–50 was buried on the island of Langeland with this set of beads, typical for the late Viking Age. There are some ‘retro’ styles, like the blue and white beads, but the quality of glass is different and they have corroded much more quickly than glass from the early Viking Age. There is also a pair of amber pendants, which are often found in cemeteries, like Stengade, where Thor’s hammers are also found. (Langelands Museum, Rudkøbing, DK.)

I must note that this chronology is only tentative—a working framework as I continue to analyze data and conduct new research. In particular, I am uncertain about the changes of the late 700s and whether this should be seen as a period of retrenchment, at least with regard to long-distance trade. Nevertheless, the glass evidence points to strong connections with the Islamic world beginning around 790 and intensifying in the early 800s. This indicates that these connections existed well before Scandinavians began to import Islamic silver in large quantities. Moreover, the ways in which the glass was being used gives us clues to what Scandinavians were doing to acquire it.