The Feno program - Fenocosm record v59: card 07: Levitin, Chordia & Menon 2012, "Musical rhythm spectra from Bach to Joplin obey a 1/f power law" — the home-domain anchor (P-11.10)

Record v59 — card 07: Levitin, Chordia & Menon 2012, "Musical rhythm spectra from Bach to Joplin obey a 1/f power law" — the home-domain anchor (P-11.10). A sealed, append-only results record of the Feno program (repository arm: Fenocosm). The audit campaign's seventh executed card (audit/PROTOCOL.md; audit/cards/07-levitin-2012-music.md) and its home-domain anchor: the canonical "rhythm is 1/f" citation (PNAS 109:3716, DOI 10.1073/pnas.1113828109), whose score-side claim the constellation's own performed-music narrative leans against. Two claims, scored separately (the house idiom of records v56 and v58), each audited as the paper reads (amendment 1). Claim A — the power-law form, quoted from the abstract: "We found that an overwhelming majority of rhythms obeyed a 1/f^β power law across 16 subgenres and 40 composers, with β ranging from ∼0.5–1" (Results: "the median β (0.78, SD = 0.35) was not significantly different from 1"). Claim B — the global-structure inference and the composer claim, quoted from the Results: "the 1/f rhythm spectrum reflects the global structure of correlations across the entire piece, and ... this structure is a consequence of the specific ordering of the rhythms, not their mere presence in the piece at random locations"; "Beethoven's rhythms exhibited the largest βs ..., Mozart's βs were among the smallest ..., whereas those of Haydn were intermediate ... and significantly different from either". Access and data, stated first (amendment 1). The text is not open access; the publisher's PDF and SI Appendices 1 (per-genre) and 2 (per-composer) were read during execution through the Internet Archive's copies, and the SI Materials-and-Methods / Tables S1–S2 file is not archived (not read). Every pipeline convention carries a locator in the card's P0 block and protocol.py's MUS_* block, and every convention the text does not state is marked unstated and pinned. The card's initialisation is corrected: the paper did run a null — the note onsets shuffled globally with the durations kept, read per genre by a Wilcoxon signed-rank test — and states its estimator and fit range; what it never ran is a null that keeps the linear correlation structure, a null that keeps the local musical structure, or a goodness-of-fit of the power law. The corpus is the paper's own database, KernScores (kern.humdrum.org, which answered HTTP 503 at execution), read from its file-for-file GitHub mirrors: 24 collections, 8,176 .krn files, 1,520 movements with more than 200 merged onsets (the paper's own filter), of which 1,143 lie in the eight genres this corpus shares with the paper's sixteen (chorale, fugue, mazurka, prelude, quartet, ragtime, sonata, symphony) and nine composers with ten or more movements on both sides share published means (Bach, Beethoven, Chopin, Corelli, Haydn, Joplin, Mozart, Scarlatti, Vivaldi). A named subset-equivalent of the paper's 1,788 movements — the same database, a different host, a different mix (the paper's 723 quartets and 586 sonatas dominate its median; this corpus's shared-genre means are reweighted to the paper's own per-genre counts for the pooled bearing) — and every word is scoped to it. Every convention was pushed before any score file was fetched (commit 05b6e6d; amendment 6), with one dated compute-only pre-run amendment (4ea71ba; the card's P0 item 15). Register ALLIED, at maximum strength. No personal material; the audit is of a pipeline. The paper's own null is a null of the right kind against the "mere presence" reading and is credited as such; where a claim survives this record says so first; where a word is carried by the letter of a pinned function, the record says what carries it and what the other reading would be. Regeneration ```bash python3 -m venv venv && venv/bin/pip install -r requirements-dev.txt ./verify.sh # apparatus gate (incl. smoke_music) venv/bin/python -m fenocosm.p59_music --workers 4 # -> out/p59_music.* (~30 min at 4 workers; --resume restarts from the stage checkpoints) venv/bin/python -m fenocosm.p59_music --sensitivity out/p59_music.json # the dated post-hoc block (~1 min), appended to the JSON venv/bin/python -m fenocosm.p59_music --refigure out/p59_music.json # the figure from the sealed JSON cp out/p59_music.{json,csv,png} out/v59/ ``` The external inputs, all fetched at run time into data/music/ (git-ignored) and sealed in the JSON's provenance block: the 24 KernScores collections shallow-cloned from their GitHub mirrors (every clone's HEAD commit recorded; one SHA-256 manifest over the 8,176 .krn files read). The KernScores host itself (kern.humdrum.org) answered HTTP 503 at execution. Everything else is synthetic and seeded. Deposit note. All seven Feno-program repositories are private and remain private, so this deposit is standalone: it carries its own reproduction materials, its metadata contains no repository link, and sibling work is referenced by DOI rather than by repository path. Part of the Feno program. The source repository is private; the reproduction materials for this record are contained in the record itself.