Found on 2026-10-10 by a search the owner asked for: documented cases for CLAIM-091, against CLAIM-229’s objection that it grounds only on the fictional CASE-018.
The case
In Anna Karenina Part I ch. 4, Dolly confronts Stiva over his affair. She addresses him with вы, says one sentence with ты, and returns to вы (“скажите”). The narrator names the shift, and Stiva is grateful for it (Tolstoy, Анна Каренина, I.IV, https://ilibrary.ru/text/1099/p.4/index.html):
— Ты помнишь детей, чтоб играть с ними […] Она сказала ему «ты», и он с благодарностью взглянул на нее […] — […] Ну, скажите, после того… что было, разве возможно нам жить вместе?
The beat is a change of relation carried by a pronoun alone. English has one pronoun of address, so a word-for-word rendering (“You remember…”; “She said ‘you’ to him”) leaves the narrator’s sentence pointing at nothing. Three published translations handle it three ways.
- Dole, 1886: the word kept and glossed. “You think of the children,
because you like to play with them […]” / “She had used the familiar tui
(thou), and he looked at her with gratitude” (tr. Nathan Haskell Dole, New
York: T. Y. Crowell, 1886, p. 17; archive.org
annakarenina00tolsiala). - Garnett, 1901: a first-name vocative added. “You remember the children, Stiva, to play with them […]” / “She had called him ‘Stiva,’ and he glanced at her with gratitude” (tr. Constance Garnett, Part One ch. 4, Project Gutenberg #1399).
- The Maudes, 1918: an inclusive possessive added. “You think of our
children when you want to play with them […]” / “But she had spoken of
‘our children’, and looking gratefully at her he moved to take her hand”
(tr. Louise and Aylmer Maude, Oxford World’s Classics reissue 1998, Part I
ch. 4, p. 11; archive.org
anna-karenina-maude).
“Stiva” and “our” are not in Dolly’s Russian line. Each is a target-language marker of solidarity, and each translator rewrites the narrator’s sentence to point at it.
Companion passages. At I.20 Dolly’s return to ты marks reconciliation: «жена говорила с ним, называя его «ты», чего прежде не было» (https://ilibrary.ru/text/1099/p.20/index.html). Dole: “called him tui (thou), which had not been the case before” (pp. 80–81). Garnett: “addressed him as ‘Stiva,’ as she had not done before”. The Maudes: “addressed him familiarly in the second person singular” (p. 69/70). At I.5 Stiva is «на «ты»» with nearly everyone, and his low companions are his «постыдные «ты»» (https://ilibrary.ru/text/1099/p.5/index.html). Dole: “the familiar ‘tui’ (‘thou’)” (pp. 22–23). Garnett: “called almost all of them by their Christian names”. The Maudes: “his ‘disreputable pals’” (pp. 15–16).
A translator’s reasoning. Nicolas Pasternak Slater, whose translation with Maya Slater is announced from the Folio Society for 2026, reached Garnett’s I.5 solution independently and says why: “with someone like Stiva Oblonsky—who’s on easy terms with everyone and uses ty freely—I simply wrote ‘he was on first-name terms with everybody.’” Maya Slater calls ty/vy “the obvious challenge”: “Each time, you have to think of a solution that makes sense in English.” (Interview by Olga Kenton, Reading in Translation, 21 July 2025.) Nicolas’s remark that “an English reader is usually familiar with French tu and vous, whereas they’re not familiar with the Russian equivalents” is made of a different episode, the lovers’ move into French, not of I.4 or I.5.
The author on the dictionary equivalent. At III.14 Karenin writes to Anna in French, «употребляя местоимение «вы», не имеющее того характера холодности, который оно имеет на русском языке» (https://ilibrary.ru/text/1099/p.83/index.html). Garnett: “the plural ‘vous,’ which has not the same note of coldness as the corresponding Russian form.” At II.22 Vronsky speaks French to avoid «невозможно-холодного между ними вы и опасного ты по-русски» (https://ilibrary.ru/text/1099/p.56/index.html).
The social coordinate. Brown and Gilman 1960, p. 253: “The pronoun ‘thou’ is reserved, nowadays, to prayer and naive poetry”. So Dole’s “(thou)” points a modern English reader toward reverence or archaism rather than intimacy. Their p. 275 gives the reading the scene relies on: “The momentary shift of pronoun directly expresses a momentary shift of mood.”
What it can show
It is a documented instance of the mechanism in CLAIM-091. The address system differs between source and target, so the coordinate that changes is the code (k and g in TERM-031’s C = (u, s, a, k, g)), not the speaker. Two published translators, seventeen years apart, added words that are not in the source in order to carry a relation the source carries in a pronoun. For I.5, two translators 124 years apart reached the same compensation, and the later one gives his reasoning. The author attests the premise: the lexical equivalent, вы as vous, does not keep the relational value.
So it meets two of the three conditions the search set: (a) a social coordinate differs between source and target, and (b) a literal and a compensating rendering of the same passage are attested, with reasoning for one of the passages. That answers CLAIM-229 at the level of existence: CLAIM-091 no longer has only an invented case for its mechanism. This is the translation arm of A24, not the reattribution arm of CASE-018.
It bears weakly on the necessity half of CLAIM-061. The narrator’s sentence changes its proposition in Garnett (“She had called him ‘Stiva’” is false of the Russian scene) while, on the translators’ judgement, keeping what the sentence does. That is the act judged kept by the translator, not measured.
What it cannot show
Condition (c) is met only partly. Every judgement here about what the compensation conveys is a producer’s or the author’s: the translators’ choices, the Slaters’ stated reasoning, the Maudes’ footnote at IV.20 that explains “thou” when they do keep it, and Tolstoy’s statement about вы and vous. Brown and Gilman supply the general sociolinguistics of English “thou”. None of it is evidence of what readers recover. No reader, classroom or reception study was found, for these renderings or for T/V compensation into English at all. That readers recover the relation better from “Stiva” or “our children” than from “tui (thou)” is CLAIM-091’s defeat condition, and this case does not touch it.
Further limits:
- No bare literal is attested. No published translation found says “She said ‘you’ to him”. Dole’s borrowing with gloss is the closest literal, and it is itself an explicitation. The literal arm of the contrast is the record’s construction.
- Not the same relation. “Our children” turns a pronoun of intimacy into an appeal to shared parenthood, which may be a different act. “Stiva” is nearer, but English first-naming is a weaker and more ordinary mark than a marked ты among вы.
- Nothing about proportion. CLAIM-091’s “where it went” speaks of wording changing in proportion to the change of context. One beat and its companions say nothing about degree.
The strongest reply a defender would make. Translators are the competent bilingual readers the claim is about: four of them, from 1886 to 2025, independently judged that keeping the word loses the relation, and the author, a bilingual, judged the same of вы and vous. That is expert judgement converging, which is evidence. The reply is right that it is evidence that the literal fails for these producers. It is not evidence that the compensation succeeds for target readers, which is what the claim says; translators may compensate from habit or house style, and the Maudes’ footnote shows them unsure the literal would be understood, not sure the compensation would be.
What would close the gap. A reader test built from these attested texts. Give monolingual English readers I.4 in four versions: Dole’s; a bare literal (“She had said ‘you’ to him”), labelled as constructed; Garnett’s; the Maudes’. Ask whether Dolly’s attitude changed in that line, in which direction, and why Stiva is grateful. Use Russian readers of the source as the reference. CLAIM-091 predicts that the Garnett and Maude groups match the Russian readers better than the Dole and literal groups. The four versions could serve as items in the discriminating study, CASE-055, as attested renderings that change wording to keep the relation, alongside CASE-021’s factorial design.
Sources
Read directly on 2026-10-10, at the URL or scan given:
- L. N. Tolstoy, Анна Каренина, Russian text, Part I ch. IV, V, XX; Part II ch. XXII; Part III ch. XIV: https://ilibrary.ru/text/1099/p.4/index.html (also p.5, p.20, p.56, p.83).
- Anna Karénina, tr. Nathan Haskell Dole, New York: T. Y. Crowell, 1886: https://archive.org/details/annakarenina00tolsiala (OCR text).
- Anna Karenina, tr. Constance Garnett, 1901: Project Gutenberg #1399, https://www.gutenberg.org/ebooks/1399.
- Anna Karenina, tr. Louise and Aylmer Maude (1918), Oxford World’s Classics, intro. W. Gareth Jones, reissued 1998, ISBN 0-19-283381-2: https://archive.org/details/anna-karenina-maude (OCR text; page numbers as the scan gives them).
- Nicolas Pasternak Slater and Maya Slater, interview by Olga Kenton, Reading in Translation, 21 July 2025: https://readingintranslation.com/2025/07/21/nicolas-pasternak-slater-and-maya-slater-on-translating-anna-karenina/. Their translation itself is unseen here.
- R. Brown and A. Gilman, “The Pronouns of Power and Solidarity”, in T. A. Sebeok (ed.), Style in Language, MIT Press, 1960, pp. 253–276. No DOI; read in the transcription at https://pages.mtu.edu/~rlstrick/rsvtxt/pronoun.htm, whose page numbers are taken as the original’s (unverified against the printed volume).
Not read, and not leaned on: R. Brown and M. Ford, “Address in American English”, Journal of Abnormal and Social Psychology 62(2):375–385, 1961, doi:10.1037/h0042862. Bibliographic data checked; content unverified. It is the likely source for first name against title-and-surname as English’s address dimension, which is the dimension Garnett’s and the Slaters’ compensations use. The Pevear–Volokhonsky rendering of these passages was not seen lawfully and is not cited.