Repository path: workshop/translations/iliad-sarpedon/R07-v1/translation.md · rendered 2026-09-09

T-iliad-sarpedon-R07-v1 — Iliad XII.310–327, Sarpedon to Glaucus, under R07 (fluency)

• Source: Homer, Iliad XII.310–327 (Greek: Monro–Allen OCT via the Perseus canonical text tlg0012.tlg001.perseus-grc2; stored at workshop/experiments/E-20260802b-pole-uniformity/materials/source-grc.json). Public domain.
• Regime: R07 v1.0 (frozen), with the S073 binding condition discharged in E-20260802b-pole-uniformity §4 before the source was opened for translation: IR1 taken literally (F10 requires the plain rendering), IR2 taken textually (F6 forbids nothing).
• Declared dialect (F5): American English.
• Translated by: the lead agent, as a labeled subject. Cost $0; not ledgered (charter §3, A4).
• Pair: T-iliad-sarpedon-R08-v1, same locus, same session, R08.
• Contamination: none, on a measurement rather than an assertion — see §Contamination below.
• Evaluation: none yet. The lead never judges its own translation (charter §5).

The translation

Glaucus, why is it that the two of us get the greatest honors at home in Lycia — the best seat, the best cut of the meat, a cup that is always full — and why does everybody there look at us the way they look at gods? Why do we hold that big estate on the banks of the Xanthus, good land, orchard and wheat field both? Because of all that, we have to stand now in the Lycian front line and go into the worst of the fighting, so that the Lycians in their heavy armor can say of us: "Our kings rule Lycia with real honor. They eat the best mutton and drink the best sweet wine, but they are strong men as well, because they fight in the front line with the rest of us." My friend, if getting clear of this war meant that the two of us would then live forever and never grow old, I would not be fighting in the front line myself and I would not be sending you into the battle where a man makes his name. But that is not how it is. Death is waiting for us in more forms than anyone can count, and nobody alive gets away from it.

184 English words from 18 Greek lines.

Contamination

none, measured before the locus was translated, per the standing rule in CLAUDE.md. The lead rendered a different span of the same book (XII.331–341) from the Greek alone, and tools/dependence_check.py compared it with Butler 1898 and Lang–Leaf–Myers 1883:

Zero shared 7-grams against both comparators, with the lead's longest runs at the same order as the two published translators' own. Both quantities are reported because run length alone has failed as a proxy three times (CLAUDE.md).

One exposure, declared. While the gate span was being extracted from Butler and Lang, the extraction window opened on line 328, the closing line of Sarpedon's speech, in both renderings. Line 328 is therefore outside this translation's locus and outside every comparison this translation takes part in. Nothing of XII.310–327 was seen in any English rendering before this text and its log were written.

Translator's log

Written as the draft was written. Decisions and alternatives, not quality claims. Frozen by commit before any evaluation was designed. Coverage codes are R07 §5's, under the strict test (a rule names the feature and exactly one live option survives), and the exclusion-test rate is not computed (S073 binding condition).

20 logged decisions. Tallies are computed by analysis/logparse.py, which parses this table; the figures are in RS-20260802b-pole-uniformity and are not restated here by hand (binding condition, workshop/translations/README.md).

Missing rules that fired

• M6 (proper-noun knowledge, not convention) — fired at sites 1 and 6. No rule bears on whether a reader knows what the Xanthus is or what holding a temenos meant; F4 settled the spelling and nothing settled the sense.
• M4 (holding a lexical choice steady) — the front line is named three times and nothing in F1–F10 requires the same phrase each time.
• M2 (sequence of tenses) — the counterfactual at 322–325 has no rule bearing on it.

One observation, flagged

internal-judgment-only. Every rule that decided a site here decided it by naming a property of English, never by naming a property of the source — F4 spelling, F2 currency, F7 syntax, F8 disambiguation, F10 plainness. That is RS-20260801 §11 arriving on a sixth source language, and it is a statement about where the rule set looks, not about whether the rendering is good.