Everybody must have heard the phrase to go berserk, but not everybody is aware of the fact how little is known about berserks and how obscure the word berserk is. Berserks were mentioned for the first time in a poem commemorating King Harald Fairhair’s victory in a battle that occurred around the year 872. The language of the poem is, consequently, Old Norwegian. For that period, Old Norwegian means the same as Old Icelandic. All we learn from the relevant lines is that “the berserks roared, the battle was in full swing, the wolfskins howled and shook the irons.” It is hard to decide whether wolfskins is a synonym of berserks or whether there were two groups of warriors (one roared, the other howled?) and on whose side the berserks made noises. Be that as it may, the information on the original berserks is admittedly scanty. Perhaps the poet (Old Scandinavian court poets were called skalds) coined the word berserk himself, but it may have existed in the language before him. Contrary to expectation, it occurs most rarely in later poetry, and, when it does, it means “warrior,” without any specification, and only with reference to the heroes of old. Once we hear that the great god Thor fought berserks’ brides. Since Thor’s main opponents were giants, berserks’ brides probably meant “giantesses.” Female monsters were feared more than superhuman males (thus Beowulf overpowered Grendel, a mighty “troll,” but nearly perished by the hand of Grendel’s vengeful mother), so that Thor cannot be accused of attacking defenseless girls.
The greatest Old Icelandic historian was Snorri Sturluson. He lived in the 13th century, and we owe several priceless books to him. One of them treats the history of the kings of Norway. As was common in those days, Snorri began his work with a mythological introduction, for royalty needs divine origins, and in a short chapter he said that Odin (the Old Norse form is Othin, rather than Odin), the main god of the Scandinavian pantheon, had a retinue of fearful warriors who “fought without armor and acted like mad dogs or wolves. They bit their shields and were strong as bears or bulls. They killed people, and neither fire nor iron did them any harm.” This he adds, “is called berserk rage.” In English we say going berserk (like going amuck), but we too know what rage is, though more often on the road than in battle.
Snorri’s description comes as a great surprise. In addition to his magnificent history of the kings of Norway, he wrote a book called the Edda, a collection of ancient Scandinavian myths. Odin figures prominently in it, but his wild retinue is not mentioned a single time. He is usually depicted as traveling alone or accompanied by two other gods at most. Nor was the word berserk of any importance to Snorri. The source of this passage is a mystery, and no one can tell why berserks failed to appear in the Edda. In the absence of facts theories purporting to explain the role of Odin’s berserks are many. I also have a theory, but it runs counter to those proposed by many eminent scholars, for which reason it found little support. Yet, like a true berserk, I roar and howl and stick to my guns (or should it be spears, slings, and arrows for the sake of preserving the local coloring?).
Berserks reemerged in Icelandic sagas (prose narratives), recorded mainly in the 13th century, when Snorri was active. But there they are gangs of vagrant marauders, intimidating farmers, raping women, and killing everybody who dares oppose them. It is in the sagas that they bite shields, fall to the ground, with their mouths foaming and frenzy making them allegedly invulnerable to fire and iron (they cannot be killed with a sword, but a cudgel does fine), and practice other stage effects. I suspect that, while writing an introduction to The History of the Kings of Norway, Snorri borrowed the portraits of berserks from the literary clichés flourishing in his lifetime. Real, not epic, berserks certainly existed, though they were exterminated in both Norway and Iceland before Snorri’s birth. Nobler berserks, the choicest warriors of kings, are mentioned in the so-called legendary sagas, and it seems that a vague memory of such bodyguards went back to at least the 8th century. Later bandits may have called themselves berserks, to aggrandize themselves, or perhaps the population called them this. It matters little who gave them such a name, for they did not resemble their predecessors of King Harald’s epoch. If Snorri had heard or read myths about Odin’s berserks, he would have retold them in the Edda. Apparently, he did not. So I assume that he knew none and, in his history, modernized the god’s image under the influence of literary tradition.
The problem is complicated by our ignorance of the etymology of the word berserk. We remember that Snorri mentioned berserks’ custom of fighting without armor and roaring like bears. The second part of the noun berserk (-serk) means “shirt,” but the first is ambiguous: it may mean “bear” (which accords well with roaring) or “bare” (in reference to throwing off armor in battle; however, being without armor is not the same as being naked), for in Old Norse the words for bare and for some forms of bear are as close as they are in Modern English. (Has anyone seen a pin I saw in Cambridge, Massachusetts, in the mid-seventies: “Bare with me”? It was worn by a grinning female. No one seemed to be paying attention.)
Bears play an outstanding role in the history of Germanic cults. On the other hand, medieval sources, both Scandinavian and Irish, describe scenes of heroes fleeing in a panic when women expose themselves to them. No superstitions are connected with male nudity. Thus, either interpretation (“bareshirt” and “bearshirt”) makes some sense. Until the middle of the 19th century Icelanders had no doubt that “bareshirt” is correct. Then an influential Icelandic scholar opted for “bearshirt,” but seventy years later the original theory again found an excellent supporter. I think he was right. Recapitulating his arguments here would take me too far afield. The main of them is that berr “bear” did not exist in this form in Old Norse, and other compounds with ber- “bear” as the first element have not been recorded either (a single exception is dubious). It is also unclear whether serk- was current as a technical term for “skin” or “shirt” as early as the 8th century.
Those who will delve into the berserk problem will find numerous things, intriguing but largely irrelevant. Did berserks form unions? If so, did those unions have a religious character? Did berserks consume poisonous mushrooms and, intoxicated like hashish eaters, attack their enemies? Were berserks akin to wervolves? Both agony and ecstasy fill the pages of the works devoted to those semimythological creatures. Little is known, a lot has been surmised. Some medieval Scandinavian warriors were certainly called berserks. They started as kings’ bodyguards. Theirs was a dignified name. With the dissolution of early feudal retinues like King Harald’s, those groups degenerated into plundering riffraff, their members turned into brigands, and the word acquired negative connotations. (The same happened to the word Viking.) Odin was hardly surrounded by berserks, Snorri’s evidence to the contrary notwithstanding. It is more likely that berserk first meant “bareshirt” (that is, someone who fights with nothing but a shirt on) even if berserks roared like bears in battle. Anyone who would try to go to battle with a bearskin on will find himself easily overheated and incapacitated. A few of my pivotal statements can be and have been contested, and herein lies the beauty of scholarship. Some people, as Snorri put it, make mistakes and others correct them, so that everybody has something to do.
Anatoly Liberman is the author of Word Origins…And How We Know Them as well as An Analytic Dictionary of English Etymology: An Introduction. His column on word origins, The Oxford Etymologist, appears here, each Wednesday. Send your etymology question to email@example.com; he’ll do his best to avoid responding with “origin unknown.”