The Stakes of the Second Attempt
Every retranslation begins with a claim: that the existing version is wrong, dated, or insufficient. Sometimes the claim is justified by shifting literary norms — what read as natural English prose in 1901 can feel stilted or falsely elevated a century later. Sometimes it is justified by scholarship — a new critical edition of the source text, access to manuscripts that earlier translators lacked. Occasionally the claim is competitive, shaped by a publisher's confidence that a recognisable title will sell. Whatever the motivation, the moment a new translator opens a canonical Russian novel already available in English, the work becomes comparative. Readers will check. Critics will quote. The choices, made in silence at a desk, become public.
The Russian classical canon has attracted more sustained retranslation into English than almost any other literary tradition. Constance Garnett translated roughly seventy volumes from the 1890s onward, and her versions set the baseline against which every subsequent English translator has had to position themselves. Tolstoy, Dostoevsky, Turgenev, Gogol: Garnett rendered them all, and rendered them in a register that smoothed their idiosyncrasies into a consistent Edwardian flow. The charges against her — that she domesticated the foreignness of the originals, ironed out Dostoevsky's repetitions and Gogol's grotesque lurches — have shaped the agenda of retranslators ever since.
What the Passage Tests
Anna Karenina's opening sentence is the most retranslated line in Russian prose: Vse schastlivye sem'i pokhozhi drug na druga; kazhdaya neschastlivaya sem'ya neschastliva po-svoyemu. The grammar is declarative, the rhythm symmetrical, the tone aphoristic. Every English version has to decide the same question: how aphoristic, and in whose idiom?
Garnett's 1901 version reads: "Happy families are all alike; every unhappy family is unhappy in its own way." It has become so familiar that many readers take it for the original. The rhythm is clean, the syntax exact, the register elevated in an English key. What it does not reproduce is Tolstoy's slight awkwardness in the second clause — neschastliva po-svoyemu is plainer and more blunt in the original Russian than the English phrase "in its own way" now sounds, having been polished by a century of quotation.
The Pevear and Volokhonsky version, published by Penguin in 2000, reads: "All happy families are alike; each unhappy family is unhappy in its own way." The switch from "every" to "each" and the addition of "all" are calibrated attempts to restore a mild grammatical emphasis in the Russian — vse means "all" — and to separate the two clauses rhythmically. Critics who accuse Richard Pevear and Larissa Volokhonsky of register literalism find little target here; the opening is almost deliberately conservative. The argument they make is in the middle chapters, in dialogue, where Tolstoy's characters shift class register rapidly and Garnett's English tends to flatten those shifts.
Marian Schwartz's 2014 version, published by Yale University Press, reads: "All happy families resemble one another; each unhappy family is unhappy in its own way." "Resemble one another" rather than "are alike" is the operative choice. Pokhozhi drug na druga is literally closer to Schwartz's phrase — drug na druga means "one another" — and "resemble" is a less idiomatic, slightly more literal verb than "are alike." Whether that gain in precision costs anything in rhythm is the kind of question that passage-comparison makes answerable: Schwartz's sentence is a half-beat slower than Garnett's, which some readers will find more faithful to Tolstoy's measured opening and others will find laboured.
The Gogol Test
If the Tolstoy comparison rewards precision, Gogol rewards nerve. Myortvye dushi — Dead Souls, published 1842 — is a novel that changes its own register without warning, sliding between bureaucratic parody, lyric digression, grotesque comedy, and mock-epic apostrophe. The translator's problem is not which English word matches a Russian one but which English key matches Gogol's constant modulation.
Garnett's 1922 translation manages the surface competently and loses the register shifts. Her narration is consistently readable, which is precisely the difficulty: Gogol's narration is not consistently anything. The famous apostrophe to Russia — Rus'! Rus'! Vizhu tebya, iz moego chudnogo, prekrasnogo dali ya tebya vizhu — becomes in Garnett: "Russia! Russia! I see thee, from my lovely enchanted distance I see thee." The inversion "I see thee" is repeated, the exclamatory mode reproduced, but the slightly mad insistence of the original — the narrator looking back at his homeland from a distance that is simultaneously geographical and aesthetic — is smoothed into period rhetoric.
The grammar is declarative, the rhythm symmetrical, the tone aphoristic.
Robert Chandler has written extensively about the problem of translating writers who resist syntactic domestication, and his arguments about Andrei Platonov are directly applicable to Gogol: when a writer builds strangeness into the grammar itself, a translator who produces natural English sentences has already made a substantial interpretive decision. The question Gogol's retranslators face is whether to let the English become strange — allowing the mock-epic voice to sound genuinely odd — or to produce the controlled comic effect that a fluent English passage delivers. Pevear and Volokhonsky's 1996 Dead Souls (Pantheon) risks the first approach more consistently than Garnett. Their Gogol is noisier and more unstable, which some critics read as fidelity and others read as a different kind of distortion.
What Retranslation Proves
No retranslation is final, and none is wrong in a simple sense. Each version is a record of what a particular translator, in a particular decade, heard in the Russian — and what they judged their English-language reader could hear back. Garnett's versions are documents of Edwardian literary taste as much as they are records of Tolstoy or Gogol. Pevear and Volokhonsky's are documents of a late-twentieth-century appetite for literalism as a guarantor of authenticity. Marian Schwartz and Oliver Ready represent a more recent settlement: close reading of the source, a willingness to let the English be slightly strange, and a resistance to the idea that any single approach solves the problem.
What passage comparison shows — and what no general claim about method can show — is that translation choices are granular. The decision to write "resemble one another" rather than "are alike" is a decision about Tolstoy, not about English prose in the abstract. Retranslation does not resolve these decisions; it multiplies the record of them.