Who actually wrote "All happy families are alike"?

Tolstoy wrote the idea, in Russian, and the exact English words you are quoting were written by Richard Pevear and Larissa Volokhonsky in 2000. Their translation of the opening of Anna Karenina reads "All happy families are alike; each unhappy family is unhappy in its own way," and that is the version the internet settled on. It matches neither of the classic public-domain translations, though it reads like a graft of the two.

The version going around

"All happy families are alike; each unhappy family is unhappy in its own way."

You have seen it on posters, in essays, in a thousand social posts, always credited flatly to Leo Tolstoy, Anna Karenina, 1877. The attribution to Tolstoy is fair. The attribution of those exact English words to Tolstoy is not, because Tolstoy never wrote a word of English fiction in his life.

The verdict, with the receipt

Tolstoy's Russian is not in dispute. The novel opens: "Все счастливые семьи похожи друг на друга, каждая несчастливая семья несчастлива по-своему." A 1918 Russian-language edition printed in New York carries the same sentence in the old orthography. Note the punctuation. Tolstoy used a comma between the two halves. The semicolon everyone shares is an English translator's choice.

Every English version of that sentence is therefore a decision made by a translator, and the two great public-domain translators decided differently.

Constance Garnett, in 1901, wrote "Happy families are all alike; every unhappy family is unhappy in its own way." Not "All happy families." Not "each." We checked this against the Project Gutenberg text character by character.

Louise and Aylmer Maude, in 1918, wrote "All happy families resemble one another, each unhappy family is unhappy in its own way." Not "are alike." A comma, not a semicolon. We checked this against two independent scans of the Maude edition, and they agree.

The circulating line takes Garnett's semicolon and her "are alike," takes Maude's "All" opening and her "each," and matches neither sentence. For years, quote checkers (including our own kill log) filed it as a composite, a Frankenstein of the two old translations. The bytes tell a cleaner story. That exact wording is the Pevear and Volokhonsky translation, published by Penguin in 2000, which won the PEN Translation Prize and became the modern standard. It reads like a blend of Garnett and Maude because Pevear and Volokhonsky landed between them. So the quote is real, the attribution is incomplete, and there is a practical wrinkle: the version everyone shares is under copyright until well into this century, while the two versions almost nobody shares are free for anyone to print.

The actual openings, side by side

Constance Garnett, 1901 (Project Gutenberg ebook #1399, public domain):

"Happy families are all alike; every unhappy family is unhappy in its own way."

Louise and Aylmer Maude, 1918 (Oxford World's Classics text, Internet Archive scans, public domain):

"All happy families resemble one another, each unhappy family is unhappy in its own way."

The circulating version (Richard Pevear and Larissa Volokhonsky, 2000, under copyright):

"All happy families are alike; each unhappy family is unhappy in its own way."

Three sentences, one Russian original, three sets of choices. If you want to quote Tolstoy freely, Garnett and Maude are yours. If you quote the version on the poster, you are quoting two living translators, and they deserve the credit line.

How Lumoro handles this

Every quote in Lumoro's morning texts is verified against a named translation before it sends. Not "attributed to," not "as quoted in." A specific translator, a specific edition, checked against the printed page. If we cannot name the person who wrote the English words, the line does not go out. It is slower. It is also the only way a quote stays trustworthy after a hundred years of retelling.