Repository path: workshop/translations/pelsen/opportunity.md · rendered 2026-09-09
Page metadata (front matter)
| type | manifest |
|---|---|
| id | pelsen-opportunity |
| status | frozen |
| created | 2026-08-01 |
| updated | 2026-08-01 |
| links | workshop/regimes/R10-anchored-register.md, workshop/translations/pelsen/target-propositions.json, workshop/canon/pelsen/manifest.md, wiki/base/anchors/A-mchugh-presence/A-mchugh-presence.md |
R10 §6 opportunity list — «Pelsen» against A-mchugh-presence
Frozen before a word of the translation was written, and after the source was read once, which
is the order R10 §6 requires and the order the S074 run breached (wiki/backlog.md, discharged
S076). The target propositions were frozen earlier still — at S074, in
E-20260731f/claims.json, sha256 1e9430e4… — and are reused verbatim rather than re-decomposed.
What this list is for. So that a proposition found absent from the finished English is not confused with one the translator failed to install. A proposition with no material to act on cannot be a failure of the translation.
| # | opportunity | why |
|---|---|---|
| C01 no archaism | YES, heavy | 1898 Swedish in pre-1906 orthography, a häradshöfding, a horse-sleigh cab, an unheated tambur. Every one is an invitation to reach for period English. |
| C02 no period lexis | YES, heavy — and this is the hardest one | pels / pelsverk, drosksläde, mössa, hundra kronor, Faust at the Opera with Arnoldson. The realia are period. C02 and C11 (name the particulars) pull in opposite directions here and the pull is structural, not incidental. |
| C03 no era-marking slang or orthographic play | YES, one site | »Fy fan så du ser ut» — the story's only coarse exclamation. Every natural English rendering is era-marked in one direction or another. |
| C04 no dialect spelling | LOW | Nothing in the source is dialectal. The proposition can only be not violated; nothing pushes toward violating it. |
| C05 no learned or Latinate display | YES | ämbetsåligganden, instinktmässigt, lifförsäkringspremien, förmodan, missöde all have a ready Latinate English route. |
| C06 free indirect discourse without a signature | THIN, and the source is against it | Söderberg does his interiority as quoted monologue — sade han till sig själf, tänkte doktor Henck — not as FID. Two narratorial-judgment sites exist (Somliga människor bli feta af att gifta sig, andra bli magra; och så var allting bra igen, utom rocken). Installing more FID would mean converting quoted monologue, which the source does not license. |
| C07 third person limited, one consciousness | NONE — fixed by the source | The source is third person and stays with Henck. Not a translator's choice. |
| C08 say-attribution, no ornamental verbs | YES, and it is a direct conflict | The source attributes with sade ×7 but also svarade ×3, tänkte, hviskade, mumlade, tillade, fortsatte. Following C08 means flattening verbs the author chose. Expect X codes here. |
| C09 figuration sparse (four figures in 1,511 words) | PARTIAL | The sparseness clause has material: the source carries two similes in 1,273 words. The count clause is a measurement of another text and cannot hold of this one — recorded here, before translating, as inapplicable by construction. |
| C10 every figure brief, no conceit | YES, easily | Both source similes are one clause long. |
| C11 named particulars supply the texture | YES | Regeringsgatan, Hamngatan, Kungsträdgården, Faust, Arnoldson, a hundred kronor, ten thousand a year. |
| C12 particulars not glossed | YES, and it is the central conflict | Arnoldson is opaque to an English reader and häradshöfding has no English office. Glossing is the obvious move and C12 forbids it. |
| C13 very short paragraphs, one-sentence paragraphs structural | NONE — fixed by the source | The source already paragraphs this way (paragraphs of 4, 5, 6, 7 words). Re-paragraphing is outside R06/R10. |
| C14 forty-eight paragraphs in 1,511 words | INAPPLICABLE BY CONSTRUCTION | A count of presence-excerpt.txt. No translation of any source could satisfy it except by coincidence. |
| C15 obscenity flat, unattributed, uncommented | PARTIAL | One site (Fy fan), and it is attributed in the source — sade häradshöfding Richardt. The unattributed clause has no material; flat and uncommented does. |
| C16 contractions everywhere, exposition included | YES, and it conflicts with the period | Live at dozens of points, and every one of them trades against C01/C02's period setting. |
| C17 sentence length varies widely, unpatterned | MOSTLY FIXED | The source already runs from 4-word paragraphs to a 60-word sentence. |
| C18 narratorial presence, no pointable idiolect | YES | The translator can add mannerism or decline to. |
| C19 no sound play, no orthographic play, no typographic device beyond paragraphing | PARTIAL, and the source violates it | The printed text uses * scene dividers at three points. Satisfying C19 in full would mean deleting them, which is not a register decision. |
Summary, before translating
- Two propositions have no opportunity at all because the source fixes them: C07, C13.
- One is inapplicable by construction — it is a count of the anchor's own text: C14.
- Three cannot be fully satisfied on this material: C09 (count clause), C15 (unattributed), C19 (the scene dividers).
- Three are in direct conflict with the source or with each other: C02 vs C11/C12, and C08 against the author's own attributing verbs.
That is nine of nineteen propositions which a translator cannot simply execute, recorded before the first English word was written. It is not a criticism of the anchor — a feature catalogue is a description of a text, and descriptions do not have to be executable. It is the fact that makes the target codes meaningful.