“Empty-handed on the cold wind to Valhalla…”

For all the violence the Vikings unleashed, their enemies and victims might find cold comfort in the torments Americans now inflict on them. We’ve twisted them into beloved ancestors, corny mascots, symbolic immigrants, religious touchstones, comic relief—and, this week, proponents of gender equity on the battlefield. The medieval past is grotesque, uninviting, and indifferent to our hopes. We wish so badly that it weren’t.

“Shieldmaidens are not a myth!” trumpeted a Tor.com blog post on Tuesday, sharing tidings of endless Éowyns in the EZ-Pass lane to the Bifröst:

“By studying osteological signs of gender within the bones themselves, researchers discovered that approximately half of the remains were actually female warriors, given a proper burial with their weapons . . . It’s been so difficult for people to envision women’s historical contributions as solely getting married and dying in childbirth, but you can’t argue with numbers—and fifty/fifty is pretty damn good.

Great Odin’s ophthalmologist! Holy hopping Hávamál! Half of all Viking warriors were women?

Alas, no. “Researchers discovered” nothing of the sort—but that didn’t stop wishful linkers from sharing the “news” hundreds of times via Twitter and countless times on Facebook.

So what’s going on here? Besides conflating “Viking” with “Norse,” the pseudonymous author of the Tor.com blog post misread a two-year-old USA Today summary of a 2011 article by scholar Shane McLeod, who most definitely has not delivered forsaken warrior maidens from their long-neglected graves. No, McLeod simply did the un-newsworthy work of reassessing burial evidence for the settlement of Norse women in eastern England in the late 800s, with nary a Brunhilde or Éowyn in sight.

You can find “Warriors and women: Norse migrants to eastern England up to 900 AD” in the August 2011 issue of the journal Early Medieval Europe. If you don’t have institutional access to scholarly databases, the article is imprisoned behind a $35 paywall, which is a shame, because although McLeod’s piece requires a slow, patient read, you don’t need expertise in ninth-century English history or modern osteology to understand it—just the ability to follow an argument about a couple dozen skeletons in a tiny corner of England at a very specific time in history, plus an openness to the possibility that McLeod hasn’t brought your “Game of Thrones” fantasies to life.

Here’s the gist of McLeod’s article, as concisely as I can retell it:

Focusing only on the area of eastern England occupied by the Norse in the 800s, he looks at one sample of six or seven burials from five locations dating from 865 to 878 A.D. where scholars had made assumptions about the sex of the dead based on the stuff buried with them. He compares them to a second sample: 14 burials from five sites (dating from 873 to the early 10th century) where osteologists determined the sex of the dead by examining their bones.

In the first group, only one person was tagged as female. In the second group, between four and six of the dead, perhaps half of the sample, were found to be female, even though based on grave goods, at least one of them might previously have been assumed to be male, because one of those women was buried with a sword. (Ah, but that woman was also interred with a child of indeterminate sex. What if the sword belonged to her young son? And look: someone in the first group who might have been a woman was buried with a sword, too…)

McLeod’s assessment is this: If we scientifically determine the sex of the dead based on their bones rather that assume their sex based on grave goods, we find more evidence (to pile atop existing evidence from jewelry finds) that Norse women came to England with Norse armies, earlier and in greater numbers than previously thought, rather than in a later wave of migration and settlement. Perhaps the men weren’t “a demobbed Norse army seeking Anglo-Saxon wives,” but intermarried with local women in smaller numbers than historians previously believed.

For the lay reader, that’s a disheartening hoard of unsexy conclusions—and a far cry from the Tor.com blogger’s claim, mindlessly brayed across social media, that “Half of the Warriors were Female.” It’s fantasy, not scholarship, and certainly not science, to interpret one woman buried with a sword, maybe two, as evidence for Norse women in combat.

Shane McLeod deserves better. Working with limited data pried out of ninth-century crevices, he recognizes that his sample size is tiny, that it’s tough to identify burials as “Norse” for sure, and that his findings are only “highly suggestive.” He’s precise, tentative, and conscious of counter-arguments, and he seems willing to go wherever the evidence takes him. His biggest accomplishment, however, is highlighting a major scholarly error. Experts who made assumptions about male versus female grave goods failed to reassess the biases they project backwards onto the Middle Ages—even though doing so is one of the traits even the most pop-minded academic medievalists will often claim distinguishes them from the duct-tape-sword-wielding masses.

Likewise, science-fiction fans are forever congratulating themselves for holding the right opinions on such subjects as evolution, but this time they lazily succumbed to fannish fantasies, failing to question a claim that deserved to be pummeled by doubt. I’ve done tons of social-media copywriting, so I get why that blogger just wanted to throw something out there after a holiday to beguile weekend-weary eyeballs—but come on.

Science doesn’t always tell us what we want to hear. Truth demands nuanced consideration of evidence, and reason demands skepticism, neither of which flourish on social media—so if you shared or re-tweeted the Tor article, congratulations! This week, in the name of medievalism, you made the world stupider.

[2019 update: Research into this subject has developed quite a bit since 2014, but I’m keeping this post online because it’s still a good example of how careful academic research gets turned into misleading clickbait. Feel free to leave links to updated scholarly research in the comments for future readers who find this post via Google.]