The objection
What the designs use. The stimulus sets the record’s designs name are these.
- The modernized refrains of Henley’s Villon (CASE-012), English rewritings of an English text. CASE-021’s study takes “our Henley translations as stimuli”, the exchange’s own English renderings of Henley.
- The spending sentence, which CASE-018’s note calls “this case with its literary source removed”, crossed with topic in CASE-040: English, and, as CASE-040 says, “every cell is an example label, not a stimulus”.
- The language-model experiments (CASE-005). Experiment 2 chooses among “translations” but names no source text and no language pair.
- Text → image → caption → image chains (CASE-009), whose design is about modality (CLAIM-087).
- Recasts of trope instances from TV Tropes (CASE-038), an English catalogue.
The only interlingual cases in the record, Villon→Henley (CASE-010) and Marot→Hofstadter (CASE-031), are put into no design. The nearest thing to an interlingual design is CASE-040’s cross-corpus version (A210 §5), which fits topic models “to a source corpus and to its translation, paraphrase or adaptation”, and names no language pair. No entry in claims, cases or questions uses the word “interlingual”.
Why it matters. The defeat conditions are disjunctive. CLAIM-115’s ranges over “translations, retellings or cross-modal renderings”, and CLAIM-139’s over “translations, recastings and media”. A result on English paraphrase alone can therefore defeat, or support, a thesis whose subject is what survives translation. CLAIM-139’s second disjunct, failure “to transfer across languages or media”, is the one cross-language test in the line, and it is a condition, not a design; the “or media” lets a cross-modal result meet it.
What carries across paraphrase need not carry across languages. Languages encode footing and politeness in grammar to different degrees: T/V pronouns, honorifics, evidentials. A rendering into a language that marks the speaker’s relation to the addressee must choose a footing that an English paraphrase can leave open. CLAIM-033 already holds, after Jakobson, that “Languages differ in what they must convey”. Footing and stance, the observables CLAIM-115 names, are among the things languages force differently, so evidence from paraphrase within English does not transfer to them by default.
What it does not say
- It does not say intralingual results are irrelevant. Paraphrase, retelling and cross-modal rendering are in the thesis’s scope.
- It does not say any interlingual result would differ. It says the programme cannot tell.
What would answer it
- An interlingual design: at least one language pair in which footing or politeness is grammatically marked on one side only, with judges on both sides.
- Or a restatement of the thesis’s scope, so that the title’s question is not settled by intralingual evidence alone.