In the spring of the year 536, somewhere in the Mediterranean world, a Byzantine official named Procopius (a historian of the eastern Roman empire, writing in Greek in the 540s, and a careful eyewitness to his own century) noticed that the sun had stopped behaving like the sun.

He wrote it down because it frightened him. "The sun gave forth its light without brightness, like the moon, during this whole year," he recorded, "and it seemed exceedingly like the sun in eclipse, for the beams it shed were not clear." A man called Cassiodorus (a senior official serving the Gothic kings who then ruled Italy, writing in the same years) described a blue sun at midday, a moon without splendor, fruit that would not ripen, and frosts in the middle of summer. In China, the court historians who kept the official dynastic annals logged a year when stars were invisible and snow fell in August and the harvests failed. From the Mediterranean to the Yellow River, on three continents at once, the great civilizations of the sixth century all wrote down the same impossible thing: the sky had dimmed, and it did not clear.

This book begins there, with a dimmed sky over a cold country, because the society it is about, the Vendel Age (pronounced "VEN-del," the late Iron Age of Scandinavia, roughly 536 to 793 CE), was very nearly killed in its cradle by what happened next. And it begins with a confession, because the confession shapes the whole book.

The people this book is about left almost no words of their own.

For two and a half centuries, the Scandinavians of the Vendel Age wrote nothing that historians would recognize as a record. They kept no chronicles, no annals, no king-lists, no letters, no law codes, no tax rolls. They had a script (the runes, which Chapter 7 returns to), but they used it the way you might use a luggage tag, to mark a name or a maker or a curse on an object, not to keep an archive. They built no stone temples for archaeologists to dig up. The handful of outsiders who mention them at all were far away, writing in other languages, mostly long afterward. There is no Tacitus for the Vendel Age, no contemporary describing it from the inside or even from the next valley over.

So almost everything in this book is reconstructed from three kinds of evidence, and it is worth naming them up front because they will recur on every page. First, the graves: the Vendel-Age elite buried their dead with extraordinary, deliberate display, and a rich grave is a sentence the dead meant the living to read. Second, the things: workshops, coins, ships, beads, the physical residue of a working economy. Third, and most dangerously, much later writing: Norse sagas, poems, and one German churchman, all composed three to seven centuries after the events, by Christians, about a pagan world their own ancestors had stopped being part of. That third category is radioactive. It is vivid, it is quotable, and it is the single biggest way a history like this one can go wrong, by treating a thirteenth-century Icelandic story as if it were a sixth-century Swedish fact. This book will use the late sources, because they are sometimes all there is, but it will flag them every single time, and it will tell you when a thing is grave-evidence (firm), inference (reasonable), or saga (a story about the past, not a report from it).

One more rule, stated once so it does not have to be repeated. When this book says "the first" (the first Scandinavian town, the first northern coin, the first recorded raid), it means the first that has survived in the evidence we can dig up or read, not the first that ever was. Absence of evidence in a thinly excavated, barely literate world is not evidence of absence. Archaeological luck decides what we get to claim. Hold that caveat under everything that follows; it is doing quiet work in every chapter.

With that on the table, we can go back to the dimming sky, because the Scandinavians felt that sky harder than almost anyone on earth.

The cause is now known, and it is one of the genuine triumphs of modern science reaching into a period with no written history. Ice cores drilled from Greenland and Antarctica trap, year by year, the chemical fallout of the atmosphere, including the sulfur that big volcanic eruptions blast into the stratosphere. The cores show an enormous sulfur spike in 536, a second around 540, and a third around 547. A massive volcanic eruption (the source is debated, with a high-latitude Northern Hemisphere volcano the usual candidate and an Icelandic eruption among the leading suggestions) threw a veil of sulfate aerosol high into the atmosphere, where it scattered sunlight back into space and refused to rain out for months. The result was a volcanic winter: a year and more of dimmed sun, depressed temperatures, and shortened growing seasons across the entire Northern Hemisphere. Tree rings from Scandinavia, Siberia, and the Alps all record the same thing the texts do, a cold so severe that by several reconstructions the 530s and 540s were among the coldest decades of the last two thousand years. Then it happened again in 540, and a third time, and the cold of what scientists now call the Late Antique Little Ice Age dragged on, by some reconstructions, for the better part of a century. The continental story of 536, the plague of Justinian that swept the Mediterranean a few years later, and what it all did to the late Roman world as Germanic peoples carved kingdoms out of its western half is its own catastrophe, the long upheaval historians call the Migration Period; the dimmed sky fell, among much else, on the middle of Justinian's long war to take Italy back from The Goths. This chapter is about what it did to the far, cold north, where there was the least margin to lose.

That margin matters, and it is the whole reason the disaster lands so hard here. Scandinavia is a high-latitude country, much of it as far north as Alaska, where grain agriculture in the sixth century was already a near-run thing. Barley and rye ripened in a short summer with no slack in it; a single cold, sunless season meant a failed harvest, and two or three in a row meant mass death by starvation in the winters that followed. The surviving evidence for what then happened is not a chronicle but a landscape. Across large parts of southern and central Sweden and Norway, pollen records show farmland reverting to forest, and the archaeology shows farms abandoned, villages emptied, whole districts going quiet. The scale is debated and the regional pattern is uneven, but the mainstream reading is stark: some districts may have lost half their population or more in the dust-veil decades. This was not a migration. It was a die-off. The country that would, two and a half centuries later, launch the longships against Europe came within a few bad winters of simply ceasing to be a populated place.

We have one more category of evidence for the catastrophe, and it is the eeriest, because it is the dead north remembering its own near-death in a story.

Centuries later, when Norse myth was finally written down, it included a prophecy of the end of the world, Ragnarök (pronounced "RAG-na-rurk," the doom of the gods). The signs of Ragnarök begin with the Fimbulwinter (pronounced "FIM-bul-vin-ter," roughly "the mighty winter"): three terrible winters in a row with no summer between them, the sun devoured, the world starving in darkness and cold while brother kills brother for food. Many scholars now think this is not coincidence. The Fimbulwinter looks very much like a cultural memory of 536–540, a real climate trauma so deep it was woven into the myth of how the world ends and folded into the inherited story-shape these peoples shared with the wider Germanic world (the broad family of related northern European peoples and languages, from the Scandinavians to the continental Germans to the Anglo-Saxons of England, all descended from a common Iron Age root) they had come from. This is exactly the kind of late-source claim this book promised to flag, so here is the flag: the Fimbulwinter is recorded only in poems written down some six or seven centuries after 536, and "myth preserving a real disaster" is an attractive, plausible, but not provable hypothesis. Treat it as a haunting possibility, not a documented fact. What is documented is the ice, the tree rings, the abandoned farms, and one more thing: the gold. The decades on either side of 536 saw an extraordinary quantity of it buried in Scandinavian ground and never dug back up. Some of it was Roman. The late Roman empire, its frontiers under constant pressure from the Germanic peoples beyond them, had spent the fifth and early sixth centuries buying peace and soldiers the only way a rich, overstretched state can: it paid out staggering amounts of gold to barbarian leaders as tribute, as wages for hired warriors, and as outright bribes to keep them quiet or fighting on Rome's side. A river of that gold had flowed north as solidi (pronounced "SOL-ih-dee," the standard late Roman gold coin) into Scandinavian hands. Much of it was then melted and restruck into the north's own status objects, the bracteates (pronounced "BRAK-tee-ats," thin one-sided stamped gold medallions worn as amulets, their imagery showing a horned or horse-borne figure many read as an early form of the god Odin). Bracteate production seems to have peaked in the early 500s and then fallen away sharply. And in hoard after hoard, especially in the south Swedish and Danish landscape, this wealth went into the earth right around the crisis and stayed there, the owners never returning. People bury treasure when the world is ending and dig it up when it is safe again. The buried gold of the 530s is, read this way, the physical signature of a society in terror, and the fact that so much was never recovered is the signature of owners who did not survive.

It is worth pausing on the art these dying people made, because it is the thread that survives the catastrophe and runs straight into everything after.

The Migration-Period Scandinavians were brilliant metalworkers, and their signature was the animal style: ornament built entirely out of stylized beasts, their bodies elongated into ribbons, looping and biting and interlacing until a brooch or a sword-mount becomes a single writhing knot of creatures, with no blank surface left anywhere. The earliest Vendel-Age version archaeologists call Style II, in which the animals dissolve almost completely into endless interlace. Picture a gilded sword-hilt three inches across on which, if you follow one ribbon-bodied creature with your eye, its jaw bites the hindquarters of the next, whose foreleg becomes the spine of a third, the whole surface a closed loop of beasts with no beginning and no end. This was not decoration in the modern sense of "making things pretty." In a world where the line between the wild and the human was a religious frontier (Chapter 4), covering a warrior's gear with a controlled tangle of supernatural beasts almost certainly invoked protective power, the way a later culture might cover a shield with saints. The animal style is the longest continuous thread in this whole book. It survives the die-off, mutates through the boat-graves, and is still going, recognizably the same tradition, on the prows and weather-vanes of the ships that sail in Chapter 8.

So this is the prelude: a thinly populated, barely literate, gold-rich, beast-art-making society at the agricultural edge of the world, hit by the worst climate shock in two millennia, depopulated, terrified, burying its wealth and its dead. And then, slowly, not dying. By the 560s the pollen and the farms suggest the bleeding had stopped. The survivors who held land now held the land of the dead too, and out of that brutal consolidation, on a base of empty farms and concentrated wealth, a new and harder kind of elite began to climb. We meet them, as we meet almost everyone in this book, in their graves.

The Vendel Age

The Year the Sun Failed

The Men in the Boats

Old Uppsala and the Kings Who May Not Have Existed

The Gods Before the Vikings

The Reach of a Cold Island

Towns and Silver

The Ship and the Script

The Longships Sail