Corpus boundary — Wattles v1

Fixed before any extraction (same discipline as the Schopenhauer / Tolstoy builds).

Included

  • The Science of Getting Rich (1910) — PROOF WORK, corpus assembled
    (sources/sgr_en_units.json: preface + chapters I–XVII, 497 paragraph units,
    Gutenberg #59844; publisher back-matter — the NAUTILUS advertisement and
    transcriber notes after the Summary — is cut in the build, not in the pool). English is the ORIGINAL text, so it is the evidence pool;
    no translation is needed for evidence. RU on the pages is a reading layer,
    marked «перевод наш».

Included separately, marked

  • The Preface (section_type: preface) — author self-characterization and the
    named philosophical lineage (Descartes, Spinoza, Leibniz, Schopenhauer, Hegel,
    Emerson). Cited as context, not as soul core.
  • The publisher's back-matter (the NAUTILUS advertisement after the Summary) —
    not Wattles' argument; excluded from the evidence pool.

Excluded

  • Wattles' other books — The Science of Being Well (Gutenberg #33917) and
    The Science of Being Great — verified PD, not yet fetched. SGR references
    them as "allied sciences"; they get their own corpus when ingested.
  • The Nautilus magazine articles ("What is Truth?", 1909) — named in the
    Preface, not in the ingested corpus.
  • Later New-Thought reformulations and abridgements (e.g. the 1976 / modern
    self-help repackagings, "condensed" editions) — meme-distorted, no clean
    edition trace; kept out of the evidence pool.
  • Quotes attributed to Wattles online without a page in SGR — see
    QUOTE_AUTHENTICITY.md.

Reason

SGR is a single compact, complete published work; it is a stable primary source
and is fully in the public domain (Wattles died 1911). The New-Thought reception
around him is heavily paraphrased and re-titled, so everything without an SGR
edition trace stays out of the evidence pool.