Part 22: Thirty-Nine Years: The Delay That Wasn't a Translation Problem

In 1974, Moto Hagio began serializing Tōma no Shinzō in a weekly magazine for Japanese girls. In 2013, Fantagraphics published it in English as The Heart of Thomas, in one thick hardcover, translated by Rachel Thorn. Between those two dates sit thirty-nine years.

In 1982, Katsuhiro Otomo began serializing Akira. In 1988, Marvel's Epic imprint began publishing it in English — colorized, flipped left-to-right, and priced like a graphic novel, but there, on shelves, six years after it started.

Six years and thirty-nine years. This series has spent twenty-one parts on what gets damaged in the crossing. This part is about a loss that happened before the crossing began, and it is probably the largest one in the book.

Nothing was mistranslated

Start with the thing that makes this essay strange. The Heart of Thomas was not badly translated. It was translated beautifully, by someone who had spent decades with the work. There is no Fred Ladd in this story, no lawyer renaming anything, no 4Kids painting over the art. The English edition is, as far as I can judge, close to the best outcome the text could have had.

And look at how few of this series' problems even apply to it. The Heart of Thomas is set in a German boys' boarding school. The characters are German. There are no honorifics to strip, because nobody is calling anybody -senpai — Part 4's whole apparatus is simply absent. Hagio's pages are quiet; there is very little of the drawn onomatopoeia that Part 6 showed was welded into the ink. The title goes across without an argument: Tōma no Shinzō is The Heart of Thomas, no committee required, no Atom becoming Astro. Part 19's bestiary problem doesn't arise — there are no oni in a German boarding school. Part 20's name problem doesn't arise, because the names are already European.

By every technical measure this series has developed, The Heart of Thomas is the easiest crossing imaginable. It is a work whose only real barrier to English was that it was printed in Japanese.

It waited thirty-nine years.

What the Year 24 Group actually did

The women who reshaped shojo manga in the 1970s are known collectively as the Year 24 Group — named for Shōwa 24, the year 1949, around which many of them were born. Hagio was one. Keiko Takemiya was another. The grouping is loose and the membership is argued about, which is normal for anything critics named after the fact.

What is not argued about is the scale of what they did. They inherited a commercial category — comics aimed at girls, a marketing bracket, for much of its early history drawn by men — and rebuilt its formal grammar from the inside. Panels that dissolve at the edges instead of ending. Interior monologue floating free of any speaker, unattached to a mouth, belonging to the page rather than the character. A single image held across a full spread so that time stops. Screentone used as emotional weather rather than shading. The flowers — the famous flowers — which are not decoration but a notation for a feeling the character is not saying.

If you have ever thought "manga looks like this" and pictured an unbordered close-up of an eye, a page composed as one object rather than a grid of boxes, a thought hanging in white space beside a figure who isn't speaking — a great deal of that comes down through this tradition. They did not invent a genre. They invented a substantial part of the medium's vocabulary for interiority, which is most of what the medium is now used for.

And The Heart of Thomas is that project at full extension: five hundred–odd pages that open with the death of Thomas Werner and then spend the rest of their length on what his letter does to a boy named Juli. It is a novel. It was published in a magazine for teenage girls. Both of those things are true and the American industry could not hold them in the same hand.

The buyer's model of a reader

So why six years for Akira and thirty-nine for Thomas?

Not quality — nobody who has read both would try that. Not difficulty, as we've just established; Thomas is the softer landing by a distance. The answer is that the people doing the buying had a picture in their heads of who read comics in English, and that picture was a boy, and then a young man, and he wanted science fiction and violence and motorcycles.

Akira fit the picture. Dragon Ball fit the picture. Ghost in the Shell fit the picture. Lone Wolf and Cub fit the picture so well that Frank Miller wrote introductions for it. A five-hundred-page interior novel about grief among German schoolboys, from a magazine for girls, did not fit the picture. So nobody licensed it. So it did not exist in English. So there was no evidence anyone would read it. So it did not fit the picture. That loop ran for four decades and it never once had to check itself, because a work that is never published cannot demonstrate an audience.

This is Part 18's sieve — the finding that the largest thing lost in translation is which works get offered at all — with a gender filter bolted into it.

And here is what makes it worse than a difference of taste: the model was wrong, and we know it was wrong, because when it finally broke it broke completely. Sailor Moon. Fruits Basket. Then the whole bookstore boom of the 2000s, driven substantially by teenage girls sitting cross-legged in chain-store aisles reading shojo off the shelf. The audience that "wasn't there" turned out, for a good stretch, to be the biggest single audience manga had in English. The industry spent twenty years not selling to the people most likely to buy, on the strength of a hunch nobody ever tested.

Thirty-nine years is not a mystery about literary taste. It is a straightforward commercial error, held with confidence, at a cost we cannot compute.

Meeting the accent before the language

The delay left a specific, permanent kind of damage, and it isn't in the text.

English-language readers met the influence before the source. By 2013, when The Heart of Thomas finally arrived, an English-speaking reader had already spent decades absorbing its grammar secondhand — in works shaped by Hagio, and then in works shaped by those, three and four generations downstream. The floating monologue, the dissolving panel, the boarding-school melancholy: all of it was familiar before the book that did it first was legible.

So the source arrived carrying the risk of reading as derivative of its own descendants. A reader in 2013 could open the foundational text of a tradition and find it reminiscent. That is not a translator's failure and no translator can repair it. It is what happens when you get the accent thirty years before you get the language.

Translation as advocacy

Part 21 argued that translation is the only craft whose success condition is invisibility — that Frederik Schodt's forty years of work were designed to leave no fingerprint. This part is the exception that's worth naming.

Rachel Thorn's English editions of Hagio — A Drunken Dream and Other Stories in 2010, The Heart of Thomas in 2013 — did not happen because a market appeared and someone was hired to service it. They happened because a small number of people spent a very long time insisting these works mattered, in essays and introductions and conversations with publishers, until somebody finally said yes. Thorn, who translated Takako Shimura's Wandering Son as well and teaches at Kyoto Seika University, is one of the reasons the 2013 date is 2013 and not never.

That is not invisible work. It is the opposite. Sometimes the translator's job is not to disappear behind the text — it is to stand in front of it and be loud for thirty years, because the sieve in Part 18 does not have a hole in it until somebody makes one.

The numbers, and what they do here

Now the engine, and it has been waiting for this one.

Shojo reduces to Destiny 22 — Master Builder, the second-highest thing this system can say about anything. Shoujo reduces to Destiny 7 — Analyst & Seeker.

Those are the same word. The same Japanese. The same magazines, the same girls, the same comics. The only difference between them is whether you render the long vowel as o or as ou — a romanization convention, a house style, a choice a copy editor makes on a Tuesday and applies with find-and-replace. And the engine hands one spelling a master number and the other a 7.

This is Part 300's finding stripped of every excuse. Nothing about girls' comics changes when a style sheet changes. The number changed anyway. The engine was never reading the category. It was reading the spelling of the category, in a language the category is not written in.

And then the one I would very much like to make something of. Tōma no Shinzō reads Destiny 1, Heart 4, Personality 33. The Heart of Thomas reads Destiny 11, Heart 6, Personality 5.

The 33 is gone. Master Teacher — the rarest and highest number the system has — is present in the Japanese title and absent from the English one. Something rare was in the original and did not arrive. That is the exact shape of this entire essay, delivered by the arithmetic, unprompted.

It is also nonsense, and I'm going to say so before anyone has to say it to me. The 33 did not fail to survive translation. "Toma no Shinzou" and "The Heart of Thomas" are different strings of letters; the engine added up different letters and got a different answer. There is no 33 in the manga. There was never a master number in Hagio's title, because Hagio's title isn't in the Latin alphabet at all — what I ran through the engine was a transliteration, which is to say a thing invented for the convenience of English speakers, and the number I got back is a fact about that convenience.

What is actually worth keeping

But hold the two failures next to each other, because they are the same failure.

The engine looks at a surface — a string of letters — and confidently announces what is underneath it. The American manga industry looked at a surface — a magazine's demographic label — and confidently announced what was underneath it.

Shojo is a marketing category. It tells you which magazine a work ran in. It does not tell you what the work is, or who would want it, or what it can do to a reader. The Heart of Thomas ran in a girls' magazine and is a novel about grief that anybody alive can read. The industry read the label and inferred the contents. The engine reads the label and infers the contents. Both were reading the container and reporting on the thing inside, which they had not opened.

Thirty-nine years is what it costs to trust a label.

Moto Hagio comes out Destiny 4: Builder & Organizer. It is the plainest number in the system — no master, no mystique, the workhorse, the one you get for showing up. She built a substantial share of the grammar that half an art form now runs on, published it in a magazine that nobody doing the buying thought was worth reading, and waited thirty-nine years to be handed to us in our own language.

Builder & Organizer. The number is right, entirely by accident, which is the only way this engine has ever been right about anything.