The title of a translated novel is the first sentence the reader meets, and it is the only sentence many readers will ever meet. It appears on rights lists, in grant applications, in catalogue spreads, and on the spines of subscription parcels before a single page has been read. And yet, of all the translator’s decisions, it is the one least often discussed in public. The prose gets its afterwords; the title gets a shrug. This essay is an attempt to treat the title with the seriousness of any other line of translation — as an interpretive act made under real constraints, and one that can be tested against the book it names.
The title as a promise made in advance
A translated title is a promise made before the reader has any other evidence. It should be judged not by literal fidelity to the source phrase but by whether the book keeps the promise the English words make. Selma Lagerlöf’s Memoirs of a Child, published by Norvik Press in 2022 as the second volume of the Mårbacka trilogy, is a useful case precisely because its English name does quiet work that the Swedish original does not specify in the same way. Norvik describes the book as “the second part of a (notionally!) autobiographical trilogy,” in which “the mature author reveals the roots of the young bibliophile’s growing skill in deploying fiction to manipulate and embellish reality, producing a wryly charming, tongue-in-cheek account.” An English title that foregrounds memoir and child simultaneously signals the genre’s slipperiness — a child’s memoir is already a contradiction, and the contradiction is the book. The title is not a transcription; it is a reading.
What makes the Lagerlöf case valuable for thinking about titles is that the translation decisions around it are documented in print. Norvik’s Lagerlöf in English series, which began in June 2011 with Lord Arne’s Silver, The Phantom Carriage and The Löwensköld Ring, has grown to thirteen titles, translated from Swedish by Linda Schenck, Peter Graves and Sarah Death. As Norvik records, “Schenck, Graves and Death have also contributed with their own translators’ afterwords in their respective translations. These chapters make for an intriguing read about different aspects of translating each particular book.” The afterword is where a translator defends choices in public. The title deserves the same defence, and rarely receives it.
Series identity as a second author
A title in a series is never only a name for one book. It must also work as an entry in a catalogue, a continuation of a trilogy, a line on a subscription list. The Mårbacka trilogy — Mårbacka, Memoirs of a Child, and the third volume — must name three books as one project while remaining individually intelligible. Memoirs of a Child carries the word Child where the Swedish carries the place-name tradition of the first volume; the English titles must signal kinship without repeating each other. This is a different job from translating a phrase. It is closer to naming a set.
The same pressure operates at the level of the press. Norvik, a non-profit publisher that “publishes high-quality translations of Nordic and Baltic fiction” and also produces Swedish Book Review, launched in 1983 to present “Swedish-language literature to the English-speaking world, through feature articles, translated extracts, interviews, book reviews and more,” builds lists whose titles must cohere. A title that works for one book may not work for a shelf. The translator proposing an English title is therefore never the only author of it: the series, the list, and the press’s identity all have a hand in the naming.
The infrastructure that circulates names early
Before a translation exists, an English title often already does. FILI — the Finnish Literature Exchange, which “supports and promotes the publication of Finnish literature in translation around the world” — runs translation and printing grants for publishers, maintains a translation database of Finnish literature since 1853, and publicises rights deals using English titles, such as Max Seeck’s The Faithful Reader. NORLA, its Norwegian counterpart, offers application-based subsidies for translators and publishers, and its Books from Norway database provides English-language information on over 2,000 Norwegian titles, “utarbeidet av rettighetshavere og NORLA” — prepared by rights holders and NORLA itself. The point is not that these bodies dictate titles; it is that English names circulate through rights listings, databases and grant paperwork long before a translator sits down with page one. Naming happens inside a promotional pipeline, not only in a translator’s study, and the eventual published title must contend with whatever English name the book has already been travelling under.
The independent press adds a further constraint. Peirene Press, which publishes at least six translated books a year and sends them directly to subscribers, describes its model plainly: “We rely on our subscribers to sustain our work. Their commitment empowers us to make bold editorial choices and plans for the future.” A subscription list is a promise of a certain kind of book, delivered sight unseen. In that economy, a title is not decoration; it is the subscriber’s only advance evidence of what the parcel will hold. A title that misrepresents the book’s register does damage beyond the individual sale — it erodes the trust the list runs on.
A practitioner’s test for whether a title earns its place
What follows is a proposed method, not a documented industry standard — a set of questions I have found useful when a title must be argued for rather than merely chosen. First: read the candidate title aloud. Titles live in the mouth — at bookshop counters, festival introductions, reading-group recommendations — and a phrase that stumbles when spoken will stumble forever. Second: ask what register the title promises, and whether the book keeps that promise. A title that signals domestic warmth for a book of cold irony has lied, however elegant the lie. Third: ask what the source title was doing — not what it meant, but what work it performed in its own literary culture — and whether the English does something comparable. A Finnish title may carry a proverb’s cadence, a Swedish title a place-name’s weight; the English need not reproduce the material, but it should reproduce the function. Fourth, and hardest: state the loss in one sentence. If the translator cannot say, in a single sentence, what the English title gives up relative to the original, the decision has not yet been made — it has only been deferred to the publisher.
Consider how this test would treat Memoirs of a Child. Read aloud, it moves easily. Its register — wry, slightly arch, a memoir that admits its own artifice — matches Norvik’s description of a “tongue-in-cheek account — one that we should beware of taking at face value.” Its function, signalling a self-aware pseudo-autobiography, is comparable to what the Swedish title does within its trilogy. And the loss is statable: whatever place-rooted resonance the Swedish carries, the English trades for genre-slipperiness. That is a defensible trade, articulated. Most titles never get that articulation.
On lists, tools, and where the decision actually happens
Generating candidates is the easy part of title work; the discipline is in the testing. Long lists help precisely because the first instinct is usually the most literal option, and literalness is rarely what a title needs. Some translators now widen their candidate lists with generative brainstorming tools — the Unsloppy AI book title generator is one such option among many — and there is nothing to object to in that, provided the tool is treated as a source of raw material and not of judgement. A generator can produce twenty candidates in the time it takes to write one; it cannot know that the book undercuts its own narration, or that the trilogy’s third volume must not collide with the second. The decision point remains where it has always been: the translator’s own lists, tested against the text, and the conversation with the publisher about what the list and the series need the name to do. No tool should be credited with a published title, and none needs to be; the value is in escaping the first instinct, not in outsourcing the choice.
Closing: the sentence before all others
The title is the translator’s most public decision and the least examined one. It is read by thousands who will never open the book, shaped by series identity and subscription economics, and circulated by grant and rights infrastructure before the translation exists. The remedy is not more fidelity but more articulation: afterwords defend the prose, and the title deserves the same defence. Read it aloud, check its promise against the book’s register, ask what the source title was doing, and state the loss in one sentence. A title that survives those four questions has earned its place on the spine. A title that has never been asked them has merely been allowed to stand.