On the First Sentence: What Literary Titles Do, and What Translation Does to Them
There is a moment in the work of translation that nobody warns you about. It arrives before the first sentence, before the first word. You sit with the title of the book — those two or three or seven words the author chose, or fought for, or had imposed on them by an editor — and you realize you must decide what the book is called in a language the author never wrote in.
A title is not a label. It is the first act of interpretation a reader encounters. It compresses the entire work into a phrase that must serve as both door and threshold: an invitation, and a warning about what kind of attention the pages ahead will demand. When we translate a title, we are not merely carrying words across a border. We are performing literary criticism in two or three words, and we are doing it under conditions that allow no footnotes, no hedging, no “on the other hand.” The title must stand alone on the cover, in the catalogue, in the bookseller’s database, and it must do so without the translator present to explain what was lost.
I want to think about what happens to titles when they cross from Nordic and European languages into English, and what those transformations reveal about the way we read — or fail to read — the books behind them.
The Girl Who Was Not There
Consider the most famous example in modern translated fiction. Stieg Larsson’s first novel appeared in Sweden in 2005 as Män som hatar kvinnor — “Men Who Hate Women.” The English-language edition, published by Maclehose Press in the UK and Knopf in the US in 2008, became The Girl with the Dragon Tattoo.
The shift is not subtle. It is one of the most consequential retitlings in recent publishing history, and not only because it launched a naming formula — “The Girl Who/With” — that would be copied across dozens of crime novels that had nothing to do with Larsson’s work. The original title names a violence and a social critique. It tells you, before you open the book, that what follows is concerned with structural misogyny, with the hatred that moves through institutions and families and not merely through isolated criminals. The English title names a character and a symbol. It turns a systemic indictment into a character study, and it turns a character — Lisbeth Salander, whose complexity the novel itself is at pains to complicate — into a brand.
That same discipline applies to title and framing decisions: before publishing, editors need a way to test a heading promises the same thing the article actually delivers, which is where a book title generator that fits the project can function as a planning aid rather than a substitute for domain evidence.
This naming formula — the definite article plus “girl” plus a distinguishing feature — became a template that extended well beyond Larsson’s trilogy. It appeared on covers of books that shared nothing with his work except a market category. The formula worked because it promised a specific kind of protagonist: isolated, marked, exceptional. But the formula also erased the range of what Nordic crime fiction was doing — the social criticism, the institutional scrutiny, the attention to systemic violence — and replaced it with a character type that could be branded and sequenced indefinitely.
I am not arguing that The Girl with the Dragon Tattoo is a bad title. It is a title that works — devastatingly well — within the logic of the Anglophone thriller market. But it performs a different reading of the novel than Larsson’s title performs. The Swedish title is an accusation. The English title is an invitation to fascination. When translators and editors retitled this book, they were not being careless. They were making a critical judgment about what English-language readers would buy, and that judgment — accurate as it proved — displaced the author’s own interpretive frame.
This is what I mean when I say that titling is criticism. Every retitling is a reading, and every reading is an argument about what the book is.
The Understatement of Tove Jansson
Tove Jansson’s titles operate by a logic that is almost the inverse of Larsson’s. Where Larsson’s Swedish title is direct and accusatory, Jansson’s titles are quiet, descriptive, almost stubbornly literal. Sommarboken — “The Summer Book.” Den ärliga bedragaren — “The Honest Deceiver,” published in English as The True Deceiver.
Consider what Sommarboken does. It is a title without drama, without mystery, without a hook. It tells you only that this is a book about summer, and that it is, in some sense, a book — the Swedish definite form makes it feel almost generic, as though this were not a summer book but the summer book, the one that exists. Sort Of Books, the English publisher, kept it as The Summer Book, and Thomas Teal’s translation preserves the title’s plainness. What the title does is set a temperature. It tells you that the pages ahead will not escalate, will not twist, will not perform. It tells you that the drama, such as it is, lives in the texture of days on a small island in the Gulf of Finland, between a grandmother and a granddaughter, and that the book’s ambition is to hold that texture without ornament.
When you translate a title like Sommarboken, the temptation is to add. English-language publishing, particularly in the last two decades, has developed an appetite for titles that promise more than they describe — titles with a charge of mystery or emotional intensity. A more “market-ready” title for Jansson’s book might have been The Island Summer or The Light on the Rocks or something that signals a reading experience more dramatically. The decision to keep The Summer Book was a decision to trust the title’s plainness, and that trust is itself an interpretive act. It says: this book does not need to sell itself with a hook. Its plainness is its argument.
Den ärliga bedragaren is a more complicated case. The Swedish title holds a paradox in two words: the honest deceiver. The English title, The True Deceiver, published by NYRB Classics in Thomas Teal’s 2011 translation, shifts the paradox slightly. “True” in English carries connotations of fidelity and authenticity that “ärlig” — closer to “honest” or “upright” — does not quite have. The shift is small but real. “Honest” implies a relationship to truthfulness in conduct; “true” implies a relationship to truth as a quality of being. The English title deepens the philosophical resonance of the paradox while slightly narrowing its social one. Whether this is a loss or a gain depends on how you read the novel — and that, precisely, is the point. The title is already a reading.
Dag Solstad and the Deliberately Flat
Dag Solstad’s titles perform yet another function. T. Singer — simply the protagonist’s name and initial. Armand V. Fotnotene til en uutgitt bok — “Armand V.: Footnotes to an Unpublished Book.” Genkjennelsen — “The Recognition,” published in English as Sham.
Solstad’s titles are deliberately flat. They refuse to seduce. T. Singer gives you a name and nothing else — no predicament, no atmosphere, no hook. The English edition, published by New Directions, keeps it. Armand V. is more extreme still: the subtitle refers not to the content of the novel but to its own formal apparatus, as though the book is less interested in announcing itself than in describing its structure. These titles are anti-titles. They perform a refusal of the seductive function that titles are expected to serve, and in doing so, they make a claim about the kind of literature Solstad is writing: literature that does not court you, that does not announce its themes in a phrase, that demands you come to it rather than it coming to you.
When you translate Solstad’s titles, the difficulty is not linguistic but cultural. The English-language publishing market is not kind to anti-titles. A title like T. Singer signals to an algorithm nothing — no genre, no emotional register, no comparable title. It signals to a browser in a bookshop very little unless that browser already knows Solstad’s name. The decision to keep these titles in English is a decision to preserve the author’s refusal of seduction, even at the cost of marketability. It is a decision that says: the title’s flatness is not an accident but a feature, and to “improve” it would be to misread the book.
The case of Genkjennelsen becoming Sham is more complex. “Genkjennelse” in Norwegian means recognition — the act of recognizing, or of being recognized. Sham in English is a word about falseness, about pretense. The retitling shifts the novel’s center of gravity from an epistemological act (recognition) to a moral one (deception). Whether this is a loss or a gain depends on how you read the book — and that, precisely, is the point. The title is already a reading.
Finnish Compound Titles and the Problem of Compression
Finnish, like German and Dutch, builds compound words that can carry extraordinary semantic density in a single orthographic unit. This creates a particular problem for English translation, because English does not have the same compounding mechanism and tends to break such concepts into prepositional phrases or descriptive clauses — which immediately dilutes the compression.
Consider the word sielunmaisema, which I have written about elsewhere. It means, roughly, “landscape of the soul” — but in Finnish it is one word, and its one-wordness is part of its meaning. It suggests that the inner landscape and the outer landscape are not two things placed in relation but one thing, linguistically fused. You cannot translate sielunmaisema into a single English word. You must use three, and in using three, you undo the very fusion the word performs.
The same problem applies to titles. Finnish titles that use compound words carry a density that English must either approximate clumsily or abandon. When the title of Sofi Oksanen’s Puhdistus — a single word meaning “purification” or “cleansing” — was translated into English as Purge by Lola Rogers, published by Atlantic Books, the translation captured the word’s violence but lost some of its ritual connotation. Puhdistus in Finnish carries echoes of both ethnic cleansing and spiritual purification; Purge in English leans toward the violent and political, away from the spiritual. The retitling is not wrong — Purge is a strong, accurate word — but it is a reading, and like all readings, it emphasizes certain possibilities while letting others recede.
The Spectrum from Artisanal to Algorithmic
What I have been describing is a spectrum. At one end, there is the artisanal craft of translators and editors who spend weeks, sometimes months, deliberating over a title — testing it against the text, against the market, against the author’s intentions, against the sound of the words in the new language. At the other end, there is the mechanical generation of titles by algorithmic tools that calibrate output to genre conventions and market signals.
I do not want to pretend that the artisanal end is always noble and the algorithmic end is always crass. Tools exist on a spectrum of their own. Even a book title generator like Reedsy’s, which offers a “literary mode” alongside its “commercial mode,” at least acknowledges that different kinds of books require different naming logics. The distinction between commercial and literary calibration is itself a form of critical intelligence, even when the execution is mechanical.
And the Authors Guild has been clear in its guidance to writers that the preservation of human voice and judgment in creative work — including the naming of works — is not a romantic attachment but a professional standard, arguing that AI outputs are “generic mashups of pre-existing works” that cannot substitute for the unique thinking a writer brings to even the smallest decisions.
But the gap between these approaches matters most when the title is a translation. An algorithmic title generator, no matter how well calibrated, works within the conventions of a single linguistic and publishing ecosystem. Even a thoughtfully designed book title generator can reveal the Anglophone market’s naming expectations — but it cannot weigh what is lost when Puhdistus becomes Purge, because it cannot read the Finnish word’s resonance in the first place. It does not know what Män som hatar kvinnor means or why Sommarboken should not be called The Island Summer. It operates within what it knows, and what it knows is the Anglophone market’s conventions — the same conventions that turned an accusation into a brand.
The craft of retitling in translation requires a literary sensibility that holds two languages in mind simultaneously and asks, at every point, not merely “what does this mean?” but “what does this do?” — what temperature does this title set, what promise does it make, what refusal does it perform, what reading does it embed? This is not a task that can be compressed into a formula, because the relationship between a title and its book is never formulaic. It is particular. It is specific to the book, to the language, to the moment of publication, and to the reader who encounters it on a shelf or a screen and decides, in the space of a second, whether to pick it up.
The Title as Threshold
I keep a shelf of books whose titles I think about — not because the books are perfect, but because the titles do something I want to understand. The Summer Book is there. T. Singer is there. A worn paperback of Purge. A Finnish edition of a novel whose title is a single compound word so dense that I have never been able to settle on an English equivalent, and so it sits untranslated on my shelf, a small monument to the irreducibility of certain acts of language.
What I have learned from this shelf is that a title is never merely a name. It is a threshold between the book and the world, and the construction of that threshold — whether by the author, the translator, the editor, or an algorithm — is always an act of interpretation. When we cross that threshold, we are already reading. The only question is whose reading we are following, and whether we know it.
The best titles, like the best translations, are transparent in the sense that you cannot see the labor in them — but the labor is there, and it shapes everything that follows. The worst titles are the ones that make the book into something it is not, and the reader discovers this only fifty pages in, when the promise of the title begins to collide with the reality of the prose.
This is why I read titles slowly. This is why I tell my translation students to spend disproportionate time on the title — more time, often, than on any single sentence. The title is the first sentence the reader meets. It deserves the same care as the last.












