Ulysses: a first reading, and the apparatus built around it

Four Odysseys

Dimension 16 · Type IV, Comparative (the book against other versions of itself) · Build status: built · Panel score: not yet scored

Built to every register it budgets: the two-axis chart (diagram), the audience deck (presentation), and — 19 July 2026 — the document PDF. It was marked back from built to building earlier the same day (C-26), because its PDF register was being closed by a linear document made outside this project; that document is its SOURCE MATERIAL, filed under Reading, and the register is this one. The Wilson half was verified against public sources (C-30) and its markings are final; the Nolan half is second-hand permanently, its four questions closed as unresolved rather than pending (C-31). Supplies dimension 9's analytic frame — refusing plot and refusing register are different acts. Re-founded 4 August 2026 (joint session 09; chair's resolution of dim16-fourth-case, OPEN until the author rules it): three plotted works — Homer, Wilson, Joyce — carry the two-axis finding and the supply to dimension 9; the reported film holds no axis position in any register, and every report about it is held, with status and date, in data/nolan-reports.json — a separately-denominated record on the dimension's existing surfaces, not a fifth register class (base document 00's architecture is the author's). The ~70% marker is demoted from a number to a sentence; the ADD/REMOVE contrast is withdrawn as a finding and restated as a question the evidence cannot close. Published under A10a: no second-hand term may occupy a position on any measured axis in any register. Saramago's dissent is published beside the register at full strength, with the tripwire that drift back toward layout parity proves him right; his second dissent — that the three-way comparison should not be published in any form — travels beside the four-instruments finding (what-is-owed-the-twelve, clause c). Corrected 4 August 2026: the unit, claim and shows fields previously counted Nolan 2026 among the plotted works and stated the ADD/REMOVE contrast as the dimension's finding; the artifacts carry the same dated corrections beside their own prose.

The claim
Fidelity to Homer's plot and interior orientation are two separable axes, usually collapsed into one. Collapsed, the three plotted Odysseys line up on a single scale of faithfulness; separated, they stop lining up at all — a three-point finding that never rested on the reported film.
The unit
the work, measured on two axes — fidelity to Homer's plot and genre, and interior orientation (Homer · Wilson 2017 · Joyce 1922) — with reports of Nolan 2026 held to a separately-denominated register, off both axes
What it shows
Wilson and Joyce share an anti-mystification impulse — both remove something (archaism; heroic distance) rather than add. Wilson de-mystifies the sentence; Joyce discards the myth's events and re-mystifies the ordinary. Two different kinds of empathy on the same two axes, established from the works themselves — and whether a third kind, empathy by ADDITION, stands against them is a question the evidence cannot close: the reported film was its sole would-be occupant, and every claim about that film is second-hand, permanently (C-31). The sharpest asymmetry is held at the register rather than argued as a symmetry: at the hanging of the slave girls, Wilson refuses to look away and renders it flatly — checkable in her own book — while the film is reported to remove the scene altogether, displacing it onto a single betrayer; only one half of that comparison can be checked, and what it supports is the four-instruments finding published on the register of reports, with Saramago's dissent beside it.

The artifact

Registers

The rule that built it

The inclusion rule
The comparison admits "the work, measured on two axes — fidelity to Homer's plot and genre, and interior orientation (Homer · Wilson 2017 · Joyce 1922)" — the reported film is admitted only to the register of reports, never to an axis; its evidence rule: "Every claim about Nolan's film in this document is second-hand", graded High/Medium/Partial/Unconfirmed (A5), plus the published rule: no second-hand term may occupy a position on any measured axis in any register. Stated in artifacts/four-odysseys/four-odysseys.print.html (the evidence-asymmetry admission box, including "Nothing here has been promoted on the grounds that two documents agree"; the register rule in the register-of-reports headnote). Cited by region, not line — the line offsets that stood here were typed by the run that shifted them (L-40, corrected by the closing audit of 5 August 2026)..
What was rejected under it
No "What was left out" section. What is kept are refusal records inside the caveats: two figures "struck as untraceable" — the 12,110 line count and the "173 versus 18" word count (four-odysseys.print.html (the #strong Wilson paragraph and the caveats reproduction); entry 16's confidence field); the words "strict" and the 12,110 withdrawn, each by one load-bearing word; the admitted commentary's "interpretations are not treated as evidence" (entry 16's sources field, L-12); and the four open film questions closed as permanently unresolved but "kept here with their last known state rather than deleted" under anchor A4 (same file). And the film's own axis position, withdrawn 4 August 2026 — the register of reports, data/nolan-reports.json.
What kind of thing it claims
An ATTACKABLE CLAIM — under joint session 09's discriminating test (an inventory can only be attacked at its inclusion rule; a reading at its claim): Its claim field (this record's own claim field) asserts a refutable proposition about orderings: "Collapsed, the three plotted Odysseys line up on a single scale of faithfulness; separated, they stop lining up at all" — and the print register stakes it independently of the weak half: the central finding "does not depend on the Nolan reporting at all", Wilson's placement "established from the translation itself" (four-odysseys.print.html (the #strong section)). Session 09 attacked it at the claim's warrant — the ~70% Nolan marker as "furniture" (docs/panel-session-09.md) — though the session's own binary named dim 9 the only attackable claim among those it graded (same file); decided here from this record's own claim field, as instructed.

In the reading

Sources

Confidence

Every claim about Nolan's film is second-hand, and the later commentary does NOT corroborate the earlier handoff: both rest on day-one journalism about the same film, so agreement between them is not two witnesses. One correction to how this was first stated (C-30): the outlets named — Vanity Fair, GQ, TODAY.com — are the commentary's own; the handoff cites a different list, so what is demonstrated is an overlap in KIND of source rather than a shared citation. Nothing has been promoted on the grounds that two documents agree. Two further limits on the admitted commentary: its bibliography is a block list, so no individual claim in it is traceable to a particular source; and it presents nine departures where its own cited Vanity Fair piece names ten, so its selection is editorial. The one claim in it that would matter most — that Wilson has confirmed she was not consulted by Nolan — cannot be supported from its own citations, whose only Wilson interview predates the film by nine years; it is held at Medium permanently — the interview would have to exist and be found, and neither is this project's to pursue. The Wilson half of this dimension was checked against public web sources on 19 July 2026, after catch C-28 found it asserting a verification nobody had performed. The check was adversarial — a second pass instructed to break the first — and it changed the markings in both directions. Nothing found is false; all three claims were overstated, each by one load-bearing word. (1) THE LINE COUNT. That the translation has the same number of lines as the Greek is now High: Wilson's Translator's Note says so in her own words, W. W. Norton's book page repeats it, and independent readers of the printed book — Roberts in The Spenser Review, Burrow in the LRB, Miller in the Washington Post — confirm it. But the figure 12,110 is withdrawn as Unconfirmed: no primary source, publisher, catalogue or review gives a number for the English. It is arithmetic inherited from a Wikipedia sentence, and the Greek total is itself edition-dependent — the Perseus text numbers to 12,110 but prints 12,107 verses. And 'line-for-line' is Wilson's own phrase for her working constraint, not a description of a strict mapping: a reader comparing the texts finds her rendering of the tenth Greek line beginning on her ninth and ending on her eleventh. (2) THE METER. That it is unrhymed iambic pentameter, chosen as the English analogue to dactylic hexameter, is High and primary. 'Strict' is withdrawn — it appears in no source of any class; Wilson's invariable word is regular, and the one critic who examines the meter closely (Armstrong, LARB 2018) describes her lines as flexible, notes eleven-syllable feminine endings, and contrasts her explicitly with Pope. Her own most-quoted line opens on a trochaic inversion. (3) 'SLAVE'. This is the best-evidenced claim and the one that matters most here. That the choice is deliberate is High and primary: the Translator's Note states that rendering a dmoe as 'maid' or 'domestic servant' 'would imply that she was free,' so she has 'often used slave.' That earlier translators avoided the word is High and verified against the text — the only place in this chain where printed evidence was checked rather than relayed: in Fagles's Books 21–22 'slave' appears zero times, against 'handmaids,' 'serving-women,' 'maids,' and Telemachus's 'You sluts — the suitors' whores!' But the implied uniform substitution is Partial: Wilson writes 'often,' concedes the word is less specific than the Greek's several terms, and — the sharpest finding — in the Book 22 hanging, the very scene this dimension cites, she writes 'these girls' where the Greek has dmoai. Anyone citing Book 22 as the proof text is citing the wrong passage; what it does support is the other half of this dimension's claim, that she refuses to look away — the line is 'I refuse to grant these girls a clean death.' Two figures are struck as untraceable: the 12,110 above, and a '173 versus 18' word-count that lives only in a single Quora answer. Independence: there are two ancestors, not one — Wilson's Translator's Note for everything about intent, and Wyatt Mason's November 2017 NYT Magazine profile for the numbers and the 'line for line' phrasing; Wilson restated across six venues is one witness speaking six times. These markings are final. The author has ruled that this project will not consult primary sources it does not hold: there is no library visit, no printed Norton edition, no screening of the film. That is a standing limit on the project, not a task deferred, and it is stated here so no future session records it as outstanding work. What the open web bears is what this dimension claims, and the markings above say exactly how much that is. Recorded as catches C-30 and C-31. The four open questions about the film are CLOSED AS PERMANENTLY UNRESOLVED (author's ruling, 19 July 2026, catch C-31): each would need the film itself, and this project does not consult sources it does not hold. They are kept here with their last known state rather than deleted, because an abandoned path is shown, not hidden (anchor A4). That state was: Calypso's arc reported as erasure then release — she gives the lotus so he forgets, and lets him go once he remembers what he is returning to — which narrowed the question without closing it; the Cattle of the Sun resolution and the Laestrygonian child's gender both thin; and whether Nolan has ever mentioned Joyce still unevidenced, the admitted commentary's assertion that he reads Homer through Joyce being the commentator's interpretation rather than a source, which closes nothing. Three departures reported late — Sinon imported from Virgil's Aeneid in place of Achilles, the elided hanging of the slave girls, and an ending in which Odysseus is exiled rather than celebrated — all push Nolan's fidelity toward the low edge of the stated ~70%; the marker was not moved, because second-hand evidence does not earn that precision, and it will not be moved now. Corrected 4 August 2026 (joint session 09, cross-examination; chair's resolution of the author's item dim16-fourth-case under the 4 August delegation — the item stays open until the author rules it): the forward-binding sentence above held — the marker was never moved. It has now been removed: the dimension is re-founded on three plotted works, Homer, Wilson and Joyce, and the reported film holds no position on either axis in any register. The stated ~70% is demoted from a number to a sentence — a percentage asserts that a measurement was made, and none was — and the honest form is: journalism reports substantial fidelity with named departures; no measurement was made and none can be. The ADD/REMOVE contrast is withdrawn as a finding and restated as a question the evidence cannot close: the reported film was the sole occupant of the ADD pole. Every reported claim, its status wording as held, and its date of entry are in the separately-denominated register of reports, data/nolan-reports.json, struck 4 August 2026 with the count of reports standing carried by its entries rather than by a figure; the register is not a fifth register class, and the four closed questions keep their last known states there. One rule is published with it: no second-hand term may occupy a position on any measured axis in any register — HTML, PDF, deck or diagram. Saramago's dissent is published beside the register at full strength — a separately-denominated register is 'the same evasion at one remove'; strip the film of any position until the project holds it — with its tripwire recorded: if the register drifts back toward parity of layout with the plotted works, he was right and the finding was a compromise. And one correction inside the Wilson half, as the verifier required: the line 'I refuse to grant these girls a clean death' is, in her rendering, TELEMACHUS's sentence — a killer's stated cruelty, not a translator's comment on her own practice; the passage still attests that she does not soften the scene.

All dimensions · by episode · the working document