Article — "The Persistence of Discredited Ideas: Ariès, Medieval Childhood, and How Corrections Fail to Travel": Queen Eleanor of Provence fell sick with grief.
Her fifth child, Katherine, had been born disabled, never learned to speak, and died at three. The chronicler Matthew Paris reportedly recorded that the queen "could find no relief in medicine nor human consolation."
That sentence sat in the record for seven hundred years. It has never once slowed down the idea that medieval parents did not love their children.
I inherited that idea the way most people do. Growing up, I acquired a vague sense that medieval people didn't see childhood as we do — that they treated children as defective adults, and withheld love as a hedge against high infant mortality.
I could not have told you where it came from. It arrived already stripped of its author, its evidence, and its expiration date. It was simply a thing one knew.
The author, it turns out, was Philippe Ariès, whose Centuries of Childhood argued in 1960 that in medieval society "the idea of childhood did not exist." And the idea is wrong.
Not wrong in the way live scholarly debates are wrong — provisionally, with dissenters. Wrong in the way settled things are wrong. Nicholas Orme, whose Medieval Children is the leading study for England, put it as flatly as a historian can: "It cannot be overemphasised that there is nothing to be said for Ariès's view." Mistaken "not simply in detail but in substance. It is time to lay it to rest."
A wealth of later scholarship across Europe — Hanawalt, Finucane, Crawford, Shahar — reportedly emphasized parental care and related themes from every angle. Over the past thirty years, medieval historians have firmly put it to rest.
And yet here I am, in an essay, laying it to rest again. Which raises the real question. Not whether Ariès was right. Why the correction never arrives.
Consider the shape of what has to travel. The error is a single portable thesis: medieval people saw children as miniature adults and didn't love them. You can carry it in one sentence, and it flatters you as you carry it — it tells you the past was crueler and stupider than the present, that we have progressed away from casual barbarity into enlightenment.
The correction is not one sentence. It is 25 percent infant mortality and another 15 percent by age ten, and a median child growing up with one sibling and two parents in their thirties or forties — just like a median child today. It is 15 percent of average income spent feeding a child, and parents paying for their own children's apprenticeships and expecting little back.
It is a schoolboy in an Oxford textbook of the 1490s coming home for the holidays, where "we wept for joy, each to other." It is the Ancrene Wisse imagining Jesus playing hide-and-seek with a believer as a mother does with a child, then jumping out laughing "with outstretched arms."
None of that fits on a bumper sticker. The correction is a library; the myth is a matchbook. Attention allocates belief to whatever is cheap to hold, and the myth is very cheap to hold.
The correction is a library; the myth is a matchbook. Attention allocates belief to whatever is cheap to hold, and the myth is very cheap to hold.
There is also a problem of terrain. Orme did not fight Ariès where Ariès stood. Ariès wrote about medieval France; Orme wrote about medieval England, a completely separate data set.
As one reader noticed in Slate, Orme and Ariès "never really go toe-to-toe." Anyone opening Medieval Children for a Perry Mason moment — the evidence brandished, the Frenchman refuted — comes away disappointed. Orme demonstrates that English children existed as we understand children. He does not so much rebut Ariès as walk away from him.
Then there is the matter of voice. Ariès was oracular and tendentious — "Boys were the first specialized children," he announces. Orme is phlegmatic and common-sensical: "Boys, after all, will be boys, and girls will be girls."
One of these registers is quotable. The other is correct. They are not the same register, and in the economy of transmission, quotable beats correct, because quotable is what gets repeated at dinner.
This is why the myth keeps its phantom half-life outside the academy. The same Slate writer noted that "a surprising number of my lawyer friends still know that in the Middle Ages, kids were thought of as miniature adults." Educated non-specialists — the exact people who ought to have received the memo — still transmit the discredited version, decades after it died.
Now the sharpest twist, the one that should unsettle anyone who trusts corrections to do their work. Even the refuters keep the error alive, because they argue against a cartoon of what Ariès said.
Ariès never claimed parents were indifferent. His famous line runs: "In medieval society the idea of childhood did not exist; this is not to suggest that children were neglected, forsaken, or despised." He was making a claim about awareness of childhood as a distinct category — not about love. The claim that pre-modern parents were abusive toward their children is the hallmark of Lloyd DeMause, a different scholar entirely.
So the anti-Ariès literature attacks a misconstrual, and by attacking it, reprints it. Daniel Kline noticed that a whole refutation industry stays "firmly within Ariès' orbit," keeping his analysis wed to a putative refutation of a thing he didn't quite say. Even demolishing the myth requires stating the myth, and the statement is what people remember.
The correction, in other words, is doing the myth's advertising.
This mechanism is not unique to childhood. It is the standard fate of any idea that does cultural work the correction refuses to do.
Take al-Andalus. The Andalusian golden age — a tolerant convivencia — became load-bearing furniture for a certain progress narrative, and so did its inversion. When Darío Fernández-Morera tried to demolish the paradise in The Myth of the Andalusian Paradise, a reviewer noted he seemingly consulted no sources in the original Arabic, cherry-picked Qur'anic verses, and read them stripped of the interpretive tradition that governed how Andalusi Muslims actually understood them. The myth and its debunking were both cartoons, because the primary sources are hard and the thesis is easy.
Take Vasari, who in the quattrocento coined "gothic" to decry the "barbarousness" of pre-Renaissance art — a foil manufactured to flatter the Renaissance. Centuries later, William Morris made the same medieval art a foil pointing the other way, an emblem of artisanal collaboration against industrial squalor, while Pugin, per Art in America, made it a foil for monarchist restoration.
The Middle Ages, in each case, is not being examined. It is being used — as a golden age, as a dark age, as whatever the present needs behind it to feel like the present.
That is the pattern. Myths persist as furniture we need, not as claims we've checked. The load-bearing story survives; the load-bearing footnote gets discarded, because a footnote is exactly the kind of thing attention is built to shed.
Ariès's thesis stayed load-bearing because it held up a floor we wanted to stand on — the floor of progress, on which we are kinder to children than anyone has ever been. Kick out that thesis and you have to face the harder, flatter fact that Orme reached at the end of his book: "I believe they were ourselves, five hundred or a thousand years ago."
The durability of a discredited idea is not a failure of information. The information has been available since 1980, total and undisputed. The refutation exists; it simply doesn't travel, because it was never built to travel, and the error was.
So the myth stays, and the queen who could find no relief in medicine or human consolation stays in the archive, unread, a counter-example the myth survives precisely by never meeting.
We keep the story. We lose the footnote. And every so often someone writes another essay to lay Ariès to rest, knowing, as they write it, that they are also — one more time — passing him along.