Repository path: workshop/translations/osso-di-morto/R07-v2/translation.md · rendered 2026-09-09

T-osso-di-morto-R07-v2 — Tarchetti, «Un osso di morto», opening, under R07 v1.0

What this is and why it exists. A second, independent R07 rendering of the same 357-word Italian unit that T-osso-di-morto-R07-v1 renders. It is the X′ of pair O in E-20260801e-lead-carryover: a first arm written by the same agent under the same frozen rule set, in a different session and a different context, so that the archived second arm (T-osso-di-morto-R08-v1) can be compared against a first arm it never saw.

Blindness, stated plainly. T-osso-di-morto-R07-v1 and T-osso-di-morto-R08-v1 were not opened — not their translations, not their logs — before this file was committed. Neither was any published English of Tarchetti. The only things read were source-it.txt, R07, and this project's own record that its five earlier R07 runs all declared American English (R07 §Measured defects 9), which is how F5's dialect was chosen without consulting the archived arm.

Dialect declared under F5: American English.

Priming: none known. The lead has read E-20260728e-venuti-specifiability/design.md §Procedure, which states that the R07 arm was written before the R08 arm and quotes no English.

Translation

I leave it to my readers to judge the inexplicable event I am about to describe.

In 1855 I settled in Pavia and took up the study of drawing at a private school there. After a few months I had become friendly with a man named Frederick M., who taught pathology and clinical medicine at the university and who died of a sudden stroke a few months after I met him. He loved the sciences, and his own above all, and he had unusual gifts of mind. But like all anatomists, and like clinicians in general, he was a deep and incurable skeptic. He was one on principle, and I could never bring him around to my own beliefs, however hard I tried in the heated, passionate arguments we had about it every day. Even so — and I am glad to do his memory this justice — he was always tolerant of convictions that were not his own, and everyone who knew him, myself included, has kept the fondest memory of him. A few days before he died he advised me to attend his anatomy lectures, arguing that I would learn a good deal from them that would help my drawing. I agreed, though the idea repelled me; and out of vanity, to seem braver to him than I was, I asked him for some human bones. He gave them to me, and I set them on the mantelpiece in my room. When he died I stopped going to the anatomy course, and later I gave up drawing as well. I kept the bones for years all the same, since the habit of seeing them had made me almost indifferent to them; and it is only a few months since a sudden fear took hold of me and I decided to bury them, keeping nothing but a single kneecap. This small round smooth bone called up no frightening ideas in me, and from the first moment I had it I had meant it, given its shape and its size, to serve as a paperweight. It had been sitting on my writing table for eleven years when I lost it in the inexplicable way I am about to describe.

Translator's log

Written as the draft was written. Both coverage tests are recorded at the moment of decision, as R07 §Measured defects 1 and 5 require: excl is the exclusion test under which every published cov(R07) before S067 was computed (a rule excludes at least one live option), and strict is §5's own text (a rule bears and exactly one live option survives). Rates below are over contested sites only, and under-count total choices by construction.

Coverage over 18 contested sites: cov(R07) = 12/18 = 66.7% by the exclusion test, 5/18 = 27.8% strict. Both are printed because R07 §Measured defects 1 forbids reporting the exclusion figure alone, and the gap between them here (38.9 points) is the same artifact that section measured across four earlier runs.

Missing rules recorded: M1 (typography and layout) at site 11. No new one.

What was hard. Sites 12 and 13 are the same difficulty twice: Tarchetti's narrator is describing his own squeamishness in a register that keeps its distance from it, and every fluent English option either names the feeling more plainly than the Italian does or drops it. F8 orders a choice and supplies no basis for choosing, which is cov(R07)'s P code doing exactly what it is defined to do.

Contamination

suspected, and unmeasured, with the basis stated rather than assumed. The only English translation of «Un osso di morto» this project knows of is Venuti's own, in Fantastic Tales (Mercury House, 1992) — in copyright and not freely reachable, so CLAUDE.md's standing measurement rule cannot be executed on this work at all. The declaration is therefore a placeholder of the kind CLAUDE.md names as such, and this artifact joins the five existing unmeasured declarations as a sixth.

What can be said without a comparator: the run is R07, whose rules push toward the most ordinary available English, and a rendering pushed toward ordinariness is exactly the kind whose overlap with an unseen published version would be hardest to interpret. Nothing here clears it.