Skip to content

Editorial corrections: characterization scope (GR, glossary), Breuer–Laine–Piilo citation and backflow direction, ch01 citations, README - #927

Merged
amaybaum merged 13 commits into
mainfrom
draft/editorial-corrections-20261025
Oct 26, 2026
Merged

amaybaum merged 13 commits into
mainfrom
draft/editorial-corrections-20261025

Conversation

@amaybaum

@amaybaum amaybaum commented Oct 25, 2026 •

Copy link
Copy Markdown
Owner

Independent review: ACCEPT at b7cda8d2. The reconciliation d6fde588 was checked: ACCEPT. Merge: authorized by the owner at exact head d6fde588, as a normal merge commit.

This PR makes the owner-approved editorial and bibliographic corrections recorded by the novelty audit (#918, novelty-v5/AUDIT.md §4, F21–F25 and F27). It adds no theorem, assumption or novelty claim, and changes no scientific classification.

Corrections

The characterization's proved scope. Accessible non-Markovianity holds exactly when every finite accessible horizon containing a memory witness admits a deterministic realization with C1, C3 and C4.

  • Each direction has its own proof (§A.34):
    • the readback lemma [Main §2.3];
    • the finite-horizon process dilation [Main §3.4].
  • Both are proved in prose.
  • C2 lies outside the equivalence. Its necessity for physical hidden sectors is conditional on the mixing hypothesis.
  • Sites:
    • GR:20 and GR:567;
    • Main:680;
    • glossary:39 / FULL:6659;
    • ch04:180 / FULL:1297;
    • Theorem B.7.6 and appendix-b:22, :406 (+FULL);
    • Substratum:111 and :126;
    • Explainer:358, :378 and :1026.

Breuer–Laine–Piilo.

  • Backflow is stated as a one-way witness of P-indivisibility, at Main:149, Explainer:148 and ch01:101 / FULL:445.
  • Main:332 cites BLP itself as [63], and Rivas–Huelga–Plenio [4] for the ancilla-assisted witness.
  • BLP is added to both bibliographies.

ch01 citations. The numeric citations become author-year: Bao–Carroll–Singh 2017, and Abac et al. 2025 (PRL 135, 111403).

book/README.md:64. The Page curve summary is now conditional on H-scramble.

Voice. "redefined" is removed at Theorem B.7.6.

Review

Head Verdict
bb14ab0c BLOCK (R1: "accessible" horizons; R2: GR:567)
b7cda8d2 ACCEPT. The propagation is complete.
d6fde588 (reconciliation with #926) ACCEPT

The reconciliation check of d6fde588 established:

CI at d6fde588: 25 pass, 1 skipped.

Recorded, not changed (outside the approved scope)

  • GR:567's later sentence. "The mixing hypothesis holds … so the equivalence applies in our universe" conflicts with GR:20, GR:643 and Main:708.
  • Backflow measure. It is equated with P-indivisibility at ch18:74, ch13, Medicine:423 and ch17:168.
  • "Characterizes CP-divisibility". It is stated without its scope at Main:332 and appendix-b:96.
  • Smaller items:
    • glossary:27 on C2 necessity;
    • README:64's H-slope conditions;
    • ch01:73's "current";
    • Main:614's "(C1)–(C4)" form;
    • F14–F20, F26 and Main:92.

Related: #918, #925 (merged), #926 (merged).

🤖 Generated with Claude Code

https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb

claude added 12 commits October 25, 2026 19:32
The prerequisites paragraph listed C1–C4 as "necessary as well as
sufficient for accessible non-Markovianity". Main's characterization
theorem (§3.4) is (ii) ⟺ (iii) with (iii) a per-witness-horizon
realization carrying C1, C3 and C4; C2 is deliberately absent from the
equivalence, and its necessity for physical hidden sectors is the
conditional physical-memory theorem, conditional on the hidden-sector
mixing hypothesis (ETH its motivation). The clause now states that
equivalence with each direction's witness (readback lemma, §2.3;
finite-horizon process dilation, §3.4; both proved in prose, Main:608,
:715–717), and the closing sentence drops the "C1–C4" label from "the
memory-bearing equivalence".

Item (ii) of the same paragraph is untouched (draft #926 edits it).
GR.tex regenerated; GR.pdf follows in the final build commit.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
The "Characterization theorem" entry called it "the framework's central
result" and stated it as non-Markovianity iff a realization satisfies
"the four conditions C1–C4". Main names S ⟺ D and the universal
hidden-memory statement as the core (Main:99), and the characterization
is (ii) ⟺ (iii) with (iii) a per-witness-horizon realization carrying
C1, C3 and C4, C2 deliberately absent (Main:516; book anchor appendix-b
Theorem B.7.6, appendix-c §C.5). The entry now states that equivalence,
names what each direction rests on (C4 readback; finite-horizon process
dilation), and keeps C2's physical necessity conditional on the mixing
hypothesis. It agrees with the glossary's own C1 entry (glossary:25).

Mirrored in FULL.md (§A.14); FULL.tex regenerated; FULL.pdf follows in
the final build commit.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
…(F21, F22)

Attribution (F21). Main:332 attributed "the Breuer-Laine-Piilo
criterion" to [3, 4] (Breuer–Petruccione 2002; Rivas–Huelga–Plenio
2010); BLP 2009 had no entry in Main or in the book bibliography. Added
as Main [63] (appended; nothing renumbered) and to book/bibliography.md
(and its FULL.md copy), from the arXiv record and Crossref:
H.-P. Breuer, E.-M. Laine, J. Piilo, Phys. Rev. Lett. 103, 210401
(2009), doi:10.1103/PhysRevLett.103.210401, arXiv:0908.0238. Main:332
now cites [63]; on the same line the caps emphasis "ANCILLA-ASSISTED"
becomes italic (§A.33). ch01:101 / FULL:445 cited BLP as "[9]", which
under Main's numbering is Bylicka–Johansson–Acín; it is now the
author-year "Breuer, Laine, and Piilo (2009)".

Direction (F22). Main:149, Explainer:148 and ch01:101 / FULL:445 stated
that divisibility failure (P-indivisibility) is equivalent to
information backflow. The corpus establishes one direction: a
P-divisible process contracts total-variation distance, so backflow
certifies indivisibility (Main:121), and the converse is false — there
are P-indivisible pairs with every distance non-increasing (Main:121,
translation_probes.py), and system-only backflow is "sufficient
evidence of indivisibility, never necessary" (Main:332). Each site now
states that direction and that the converse fails. BLP's own claim is
described as their abstract states it: backflow as the signature of
non-Markovian dynamics. Explainer (superseded) is corrected at :148 only.

Other BLP sites checked, unchanged: Explainer:624 (its "cited elsewhere
in this framework" now holds), ch18:78 / FULL:4896 and appendix-b:96 /
FULL:5883 (consistent with Main:332).

Main.tex, Explainer.tex and FULL.tex regenerated; PDFs follow in the
final build commit.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
ch01 carried numeric citations, [2] at :32 and [8] at :85 (with [9] at
:101, resolved in the preceding commit), against the author-year
book/bibliography.md, where they resolved to nothing. Both date from
the chapter's import (334eef0) and follow no current numbering:

- [2], "holographic entropy bounds": Main's [2], cited by Main:684 for
  the same statement — Bao, Carroll and Singh, "The Hilbert space of
  quantum gravity is locally finite-dimensional", whose argument bounds
  a region's degrees of freedom by the finite black-hole entropy. It
  had no book entry; added from the arXiv record and Crossref:
  Int. J. Mod. Phys. D 26(12), 1743013 (2017),
  doi:10.1142/S0218271817430131, arXiv:1704.00066.
- [8]: introduced (334eef0) as the source of "gravitational-wave
  measurements at 99.999% agreement", i.e. the GW250114 paper (Main's
  [8] is Strasberg–Esposito, which does not fit). It is now
  "(Abac et al. 2025)", the existing book entry, placed at the
  measurement it supports.

Mirrored in FULL.md, including its bibliography copy; FULL.tex
regenerated; FULL.pdf follows in the final build commit.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
The ch07 summary listed "the Page curve from nested trace-out" without
the scrambling condition that ch07:12, ch07 §7.6, ch09:127, glossary:113
and appendix-a G11 attach to it (GR Theorem A.9 is conditional on A.8,
hence on H-scramble). It now reads "conditional on the scrambling
condition H-scramble", in ch07:12's wording. book/README.md has no
mirror in FULL.md and no build product.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
Built with sh ./build.sh GR Main Explainer and sh ./build.sh --book
(pandoc 3.1.3, xelatex, tools/unicode-fix.tex); no dropped glyphs. The
build reproduced the committed .tex byte for byte. Page counts are
unchanged (GR 86, Main 90, Explainer 68, book 552).

Kept as a separate commit so that a conflict with another PR's
generated files is resolved by rebuilding from the merged sources,
never by hand-merging PDFs.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
…e at GR:567, Main:680, ch04:180

R1. The added equivalence quantified over every finite horizon containing
a memory witness; Main's (iii) and the process dilation are stated for
finite accessible horizons (Main:496, :516, :520). GR:20 and
glossary:39 / FULL:6659 now read "every finite accessible horizon",
and the glossary's reverse direction "each accessible witness horizon".

R2 and N4 (§A.25). Three more sites stated the characterization as
"(ii) ⟺ (iii) under (C1)–(C4)" or as an identification with
"embedded observation under C1–C4", putting C2 inside the equivalence
against Main:516:
- GR:567, in the review's wording: (iii) carries (C1), (C3) and (C4);
  (C2) lies outside the equivalence, its necessity for physical hidden
  sectors conditional on the mixing hypothesis (ETH its motivation);
- Main:680: "(ii) ⟺ (iii), with (iii) carrying (C1), (C3) and (C4) and
  (C2) outside it";
- ch04:180 / FULL mirror: "embedded observation under C1, C3 and C4 (…;
  C2's necessity for physical hidden sectors conditional on the mixing
  hypothesis)". The paragraph's equilibrium argument (ch04:178) already
  routes C2's failure through C4's, so it still follows.
GR:567's later sentence, that the mixing hypothesis holds and the
equivalence applies in our universe, is a separate status claim and is
unchanged.

N1, N2. GR:20 closing: "the ETH condition for C2 necessity" → "the
eigenstate thermalization hypothesis, which motivates the mixing
hypothesis behind C2 necessity"; and "the memory-bearing sector those
conditions characterize" → "that C1, C3 and C4 characterize".

N3. Main:332: the ancilla-assisted witness cites [4] again. Rivas,
Huelga and Plenio, PRL 105, 050403 (2010; arXiv:0911.4270), introduce
the witness monitoring a system's entanglement with a shielded ancilla
(checked in the arXiv text).

N5. Abac et al. 2025, bibliography.md and its FULL copy: "Physical
Review Letters 135: 111403", checked against Crossref for
doi:10.1103/kw5g-d732 (vol. 135, issue 11, art. 111403, 2025-09-10).

GR.tex, Main.tex and FULL.tex regenerated; PDFs follow in the next
commit.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
Built with sh ./build.sh GR Main and sh ./build.sh --book (pandoc
3.1.3, xelatex, tools/unicode-fix.tex); no dropped glyphs; the build
reproduced the committed .tex byte for byte. Page counts unchanged (GR
86, Main 90, book 552). Explainer untouched by the preceding commit.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
B.7.6 stated the characterization as holding "on each finite horizon"
and displayed it "per finite horizon". Main's (iii) quantifies over
every finite accessible horizon long enough to contain a memory witness
(Main:520), and the process dilation behind (ii) ⟹ (iii) is stated for
a finite accessible horizon (Main:496); nothing else in B.7.6 restricts
the horizon ("accessible" there qualifies only non-Markovianity). Both
the statement and the display now say "finite accessible horizon",
as GR:20 and glossary:39 do after the R1 fix.

Mirrored in FULL.md; FULL.tex regenerated; FULL.pdf follows in the next
commit.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
Built with sh ./build.sh --book (pandoc 3.1.3, xelatex,
tools/unicode-fix.tex); no dropped glyphs; the build reproduced the
committed FULL.tex byte for byte. 552 pages, unchanged.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
…efined"

The same equivalence as R1 (GR:20, glossary:39, B.7.6) was quantified
over every finite horizon at its shorthand sites. Main's (iii) is over
finite accessible horizons (Main:520), and the process dilation is
stated for a finite accessible horizon (Main:496). None of these sites
restricts the horizon in its own sentence. In their own register:
- Substratum:111: "⟺ C1/C3/C4 per finite horizon" → "per finite
  accessible horizon";
- Substratum:126; appendix-b:22 and :406 with their FULL mirrors:
  "realization per horizon" → "realization per accessible horizon"
  (Main:608's own wording);
- Explainer:378: "on each finite horizon" → "on each finite accessible
  horizon"; the paragraph's other wording is unchanged;
- Explainer:358: "(per finite horizon)" → "(per finite accessible
  horizon)"; Explainer:1026: "per horizon" → "per accessible horizon".
  These two state the same equivalence and were found by the re-grep.

appendix-b:454 / FULL:6241: "In the redefined form of [Main §3.4]" →
"In the form of [Main §3.4]" (§A.30: no self-narration).

Substratum.tex, Explainer.tex and FULL.tex regenerated; PDFs follow in
the next commit.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
Built with sh ./build.sh Substratum Explainer and sh ./build.sh --book
(pandoc 3.1.3, xelatex, tools/unicode-fix.tex); no dropped glyphs; the
build reproduced the committed .tex byte for byte. Page counts
unchanged (Substratum 52, Explainer 68, book 552).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
Reconciliation of #927 after #926 landed. First parent b7cda8d (#927
head), second parent c1cade8 (main).

Conflicts and resolution:
- papers/GR.md:20 (text): both PRs edit this one line, in disjoint
  spans. Resolved by keeping both: #927's item (i) clause, its R1/N1/N2
  edits and its closing-sentence edits, and #926's canonical-measure
  insertion in item (ii). Checked mechanically: the merged line minus
  #926's insertion equals #927's line, and the merged line minus
  #927's spans equals main's line.
- papers/GR.tex, GR.pdf, Explainer.tex, Explainer.pdf,
  book/The-Incompleteness-of-Observation-FULL.tex, FULL.pdf (generated):
  not hand-merged; regenerated from the merged sources with
  sh ./build.sh GR Main Explainer Substratum and sh ./build.sh --book.
  Main and Substratum rebuilt to identical .tex and PDF text; their
  committed PDFs are kept, since only build metadata differed.

Auto-merged without conflict, checked: book/README.md, FULL.md,
appendix-c-objections.md, ch01-observation.md, ch07-gravity.md,
papers/Explainer.md.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
@amaybaum
amaybaum marked this pull request as ready for review October 26, 2026 04:31
@amaybaum
amaybaum merged commit 03eb139 into main Oct 26, 2026
26 checks passed
@amaybaum
amaybaum deleted the draft/editorial-corrections-20261025 branch October 26, 2026 04:31
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants