Repository path: workshop/regimes/R58-local-inversion.md · rendered 2026-09-09

R58 — the radif carried by inversion, and the inversion's reach recorded

v0.1, pilot. R55 plus a declared register and a per-position record of how far the inversion had to travel.

R55 asks whether the shape can be hosted at all. R56 prices the inversion against a paired direct-order rendering. Neither of them records how much of the line the inversion disturbs, which is the question ARM-inversion-price step 2 has to put to two published hands and could not put to any first-hand record, because no regime asked for one.

framework/v0.2 §7.42 now says the object–verb inversion at a line end is costly in both registers and that archaism does not buy it. What it does not say — because nothing has measured it — is whether the inversion is a local purchase (move the verb to the end of one line and leave the rest alone) or a line-grammar (a hand that inverts at the radif is already inverting everywhere). R58 makes the first-hand answer a recorded datum instead of an impression.

The rule

1. R51 §1 carries over unchanged and outranks everything below: render the sense, adding no fact, referent or evaluation and dropping none; a chime bought with a word the Persian has not got is a refusal, not a chime; enumerate the options before choosing and write the enumeration down whether or not it yields; grade mechanically after the hand has written its own reading; never revise a rendering to move a grade; verse stays verse; freeze the log before any evaluation is designed and before any published hand's page is opened.
2. The rendering of record, as R55 §2: at every rhyming position — both hemistichs of the مطلع and the second hemistich of every later bayt — the English attempts (a) the radif, the same word or phrase standing last at every position, and (b) the chime, the word immediately before it rhyming with its counterparts.
3. The register is declared in writing before the first English line and is not relaxed. Plain modern English: no thou/thee/thy, no -eth/-est, no hath/doth, no auxiliary do inserted for metre, no lexis a present-day reader would place before 1900. The declaration is load-bearing — RS-20260831b-radif-hands §6 is a case of a log mistaking its own declared register for a fact about English, and a register that is only implied cannot be caught doing that.
4. At every radif position the log records the inversion's reach, after the line is written and before any published hand is opened: - LOCAL — the complement is fronted over the radif verb and nothing else in the hemistich departs from ordinary modern order. - SPREAD — fronting the complement forced a second constituent out of its ordinary place (an adverbial or prepositional phrase moved as well, an auxiliary re-sited). Name the second move. - REBUILT — the predicate itself had to be re-shaped to put a frontable object in front of the verb: a nominalisation, a light-verb periphrasis, a change of voice. Name the reshaping. This is Payne's move at sh130 (RS-20260831b-radif-hands §6) and it is the one a declared plain register is most likely to refuse. - REFUSED — no English tail can stand there under the declared register, options enumerated.
5. Grades, as R55 §3–§4: the radif HELD / UNREACHABLE for the poem; the chime HELD / SHORT (filler named) / REFUSED position by position; and the reach of §4 at every position where the radif is HELD.
6. A rhyme word may be used twice in different senses (R55 §6); every reuse is named.
7. The reach grades are a data structure, not prose. They are tallied by a committed script that parses the frozen log (workshop/translations/README.md, note (bey)), never by rereading.