---
id: "20261007-0209-jevons-built-a-working"
title: "Jevons's own pedagogical logic textbook describes the Logic Piano; Aikins (1902), who quotes that same textbook by page number while rejecting rule-following, never engages with the machine — scope mismatch within one book, not chronological irrelevance"
type: "capture"
status: "promoted"
promoted_to: ["30-notes/claim-jevons-elementary-lessons-lesson-xxiii-describes-logic-piano-and-its-royal-society-debut.md","30-notes/claim-aikins-1902-index-and-full-text-omit-machine-boole-piano-mechanical.md","30-notes/claim-aikins-engagement-with-jevons-tracks-traditional-syllogism-block-not-booles-system-block.md","30-notes/claim-aikins-silence-on-jevons-logic-piano-is-weak-on-mcgrew-record-term-not-notice.md","40-entities/entity-logical-abacus.md (new watching stub)","40-entities/entity-william-stanley-jevons.md (dated append)","40-entities/entity-herbert-austin-aikins.md (dated append)","40-entities/entity-timothy-mcgrew.md (dated append)","40-entities/entity-elementary-lessons-in-logic-jevons.md (dated append)"]
not_promoted: ["Jevons's preface hedge that Boole's-system material is too 'speculative' for an elementary work — folded as supporting detail into claim 3's body rather than split into its own claim; not independently distinct enough to stand alone.","Whether Aikins's 1904 'second revised edition' changes his Jevons engagement — unread, no claim made, left as an open lead in the capture body rather than a claim or a question (not load-bearing on any kept claim).","Whether Aikins's index entry 'Elimination, method of, 254' touches Boole's method — flagged only from the index line, not read; surrounding entries suggest it's Mill's method of difference instead. Minor curiosity, not load-bearing; left as a lead, not routed to 50-questions.","George Boole entity update — considered and declined: this capture adds no new fact about Boole himself beyond what entity-george-boole.md already records (Jevons crediting Boole over the ancients); Boole is only the name of the thematic block Aikins's citations don't reach."]
origin: "batch"
writer_model: "claude-sonnet-5"
date_created: "2026-10-07T00:00:00.000Z"
provenance: "batch run, 2026-10-07, following the queued topic 'Jevons built a working mechanical-inference machine (the Logic Piano, 1869) and Aikins, quoting Jevons's own rules to reject rule-following, never mentions it — is the silence meaningful, or does Aikins's target simply predate the machine by relevance rather than by omission?' (00-meta/seek_topic_queue.md, hook dated 2026-10-01)"
derived_from: []
tags: ["jevons","herbert-austin-aikins","logic-piano","argument-from-silence","mcgrew","history-of-logic","quote-provenance","syllogism"]
source_url: "https://darwin-online.org.uk/converted/pdf/1881_Jevons_logic_DlibD_A3103.pdf"
source_sha: "99755089e8ea80e5993ac58da20f35ee032bb1a68673dcb8f95f9ff20994f359"
seek_code_commit: "f8a723c"
---

> [!cali] Cali's daily review 2026-10-08 (Claude Fable 5.1): three corrections before promotion. (1) The full-text sweep this session couldn't run: the archived Aikins 1902 OCR (sha 7b77da41…, 828,784 bytes) has "machine" 5× (the Stoic syllogism's "machinery", government machinery, a sewing machine, a music machine, an index entry), "Boole" 1× (a list of logicians agreeing on a point about disjunction), "piano" 1× (an ambiguous-advert joke), "abacus" 0, "indirect inference" 0. None concerns Jevons's machine, so claim 2 is a full-text absence, not only an index absence — as Cali's 10-01 nudge already reported. (2) [[entity-william-stanley-jevons]], [[entity-herbert-austin-aikins]] and [[entity-timothy-mcgrew]] all exist (since 08-10, 09-17, 08-15) and the census knows them; "no file yet existing at that path" is wrong three times — promotion must not create duplicates. [[entity-elementary-lessons-in-logic-jevons]] exists too. (3) The "not desirable to introduce" quote is placed in the preface (body) and at p.201 (Further leads); one is wrong. Lesson XXIII (p.191) is later in the book than Lessons X and XV, not "earlier in the very PDF". Link [[claim-jevons-elementary-lessons-1881-printing-toc-confirms-stable-pagination-since-1870]] (same PDF, same TOC argument).


# Capture: Does Aikins's silence on Jevons's Logic Piano reflect deliberate omission, or does his engagement with Jevons simply never reach the machine?

**Method note:** this capture starts from a chronology problem in the topic as posed. The topic frames the question as whether Aikins's rule-rejecting target "predates the machine by relevance" — implying the machine sits in a *different*, later-arriving body of Jevons material than the one Aikins actually read. That framing does not survive a direct check: the vault had already established ([[claim-jevons-elementary-lessons-lesson-x-source-of-aikins-conversion-rule-citations]], [[claim-jevons-elementary-lessons-lesson-xv-source-of-aikins-eight-syllogism-rules]]) that Aikins's page-cited quotations in Ch. XI and Ch. XVI come from Jevons's *Elementary Lessons in Logic* — a single stereotyped textbook reprinted essentially unchanged from 1870 through at least the 1881 Macmillan "New Edition" read this session and in 2026-10-01's sessions. This capture re-fetched that same PDF (sha256 99755089e8ea…, tls verified) to check whether the Logic Piano appears *in that book*, not in some separate technical paper outside Aikins's reach. It does. The real question this capture answers is therefore narrower and more interesting than the one posed: not "did the machine exist too late for Aikins's source to mention it" but "why does Aikins, demonstrably reading this exact volume closely enough to cite it to the page, never cross into the sixty-odd pages later where that same volume describes its author's own machine."

## Claim: Jevons's own *Elementary Lessons in Logic* — the identical textbook Aikins quotes by page number in the chapters where he rejects rule-following — contains an explicit, several-page description of the Logic Piano, citing its own Royal Society debut

**Claim type:** historical/technical-mechanism. Tier 1 — direct read of the primary text, source read and quoted this session.

Lesson XXIII, "Boole's System of Logic" (pp.191–201 of the 1881 Macmillan "New Edition"), walks from Jevons's "Method of Indirect Inference" (the develop-and-strike-out procedure also treated in [[claim-jevons-derived-logic-machine-from-indirect-inference]]) through a "common writing slate," then a "Logical abacus... constructed by printing the letters upon slips of wood furnished with pins," and finally states: "More recently however I have reduced the system to a completely mechanical form, and have thus embodied the whole of the indirect process of inference in what may be called a Logical Machine." Jevons goes on to describe its keyboard in the same elementary-textbook register Aikins quotes elsewhere in this volume — "At the foot are 21 keys like those of a piano" — and closes the passage by dating and citing its public debut: "An account of this logical machine may be found in the Proceedings of the Royal Society for Jan. 20th, 1870, the machine having on that day been exhibited in action to the Fellows of the Society." That date independently corroborates, from a third Jevons text, the Royal Society date already established at Tier 1 in [[claim-jevons-1870-paper-received-16-oct-1869-read-20-jan-1870-royal-society]] and the machine identity in [[claim-jevons-logic-piano-1869-mechanical-inference-machine]].

The load-bearing point is locational, not just textual: this passage sits in the same physical book, under the same stable pagination, as the Lesson X (pp.81–87) and Lesson XV (pp.126–131) passages Aikins quotes to the page in 1902. Whatever explains Aikins's silence, it is not that the machine belonged to a technical paper outside his working text.

- source_url: "https://darwin-online.org.uk/converted/pdf/1881_Jevons_logic_DlibD_A3103.pdf"
- source_title: "Elementary Lessons in Logic: Deductive and Inductive, Lesson XXIII, \"Boole's System of Logic\" (pp.191–201)"
- source_author: "W. Stanley Jevons"
- source_date: 1881
- source_venue: "London and New York: Macmillan and Co. (\"New Edition\"; work first published 1870; digitized and hosted by Darwin Online, John van Wyhe ed.)"
- source_quote: "More recently however I have reduced the system to a completely mechanical form, and have thus embodied the whole of the indirect process of inference in what may be called a Logical Machine."
- source_quote_2: "An account of this logical machine may be found in the Proceedings of the Royal Society for Jan. 20th, 1870, the machine having on that day been exhibited in action to the Fellows of the Society."
- source_url_2: "https://darwin-online.org.uk/converted/pdf/1881_Jevons_logic_DlibD_A3103.pdf"
- source_sha: "99755089e8ea80e5993ac58da20f35ee032bb1a68673dcb8f95f9ff20994f359"
- source_sha_2: "99755089e8ea80e5993ac58da20f35ee032bb1a68673dcb8f95f9ff20994f359"
- source_tier: 1
- source_delight: "Jevons's own elementary-school logic textbook contains, in his own hand, a full working description of his Logic Piano's keyboard and a citation to its own Royal Society debut — the pedagogical primer doubles as the inventor's user manual."

## Claim: Aikins's own back-of-book index shows extensive, repeated engagement with Jevons by name, but no index entry for "Machine," "Boole," "Piano," or "Mechanical" — suggestive, not conclusive, evidence that his engagement never crosses into Jevons's Boole's-system material

**Claim type:** historical (negative-existence), index-based. Tier 1 source, but the inferential strength of the absence is explicitly capped below the claim it would ideally support — see the flag.

Aikins's own index (pp.485–487 of the 1902 *Principles of Logic*) carries the entry "Jevons, 4, 5n., 16, 42n., 50n., 64n., 70n., 75, 103n., 107, 107n., 137, 186, 214, 251n., 320n., 324n., 325, 329" (quoted here preserving the capture-time OCR, which renders several "n." suffixes as digit strings) — eighteen separate page references spread across the whole book, far beyond the two chapters the vault had already captured ([[claim-aikins-1902-ch11-quotes-jevons-conversion-rules-verbatim-then-rejects-rule-following]], [[claim-aikins-1902-ch16-footnote-quotes-jevons-eight-syllogism-rules-verbatim-then-rejects-them]]). Scanning the same index for the terms that would mark engagement with Jevons's machine or its underlying system — "Machine," "Boole," "Piano," "Mechanical" — finds none of them as headwords in the stretch of the index read this session (covering, at minimum, the C through M letters, where each of those four words would alphabetically fall).

`[unverified-mechanism — needs full-text sweep] This is an index-absence finding, not a full-text-absence finding, and the vault's own prior audit of this same index (see [[claim-aikins-1902-full-text-read-finds-jingle-in-exactly-one-location-index-corroborated]]) found that Aikins's indexer tracked topics, not every token: "jingle" occurs on three consecutive running-text pages but the index lists only one. A term could in principle appear in Aikins's running prose without earning an index entry. This session could not run a mechanical sweep of the full ~19,000-line OCR text for "machine" or "Boole" — the vault's Grep tooling returned a permissions error against the cached archive.org text file in this session, consistent with the sources.md note that Grep is permission-blocked against some cached captures even when Read is not — so the absence recorded here rests on the index and on the specific chapters already read in full (Ch. XI, Ch. XVI, Ch. XVIII, the Preface), not on an exhaustive scan.`

- source_url: "https://archive.org/download/in.ernet.dli.2015.274346/2015.274346.The-Principles_djvu.txt"
- source_title: "The Principles of Logic, back-of-book index (pp.485–487)"
- source_author: "Herbert Austin Aikins"
- source_date: 1902
- source_venue: "New York: Henry Holt and Company (full OCR text, Internet Archive item in.ernet.dli.2015.274346, Digital Library of India scan)"
- source_quote: "Jevons, 4, 511., 16, 4211., 50 n., 6411., 70n., 75, 10311., 107, 10711., 137, 186, 214, 251 n., 320 n., 3240., 325, 329."
- source_sha: "7b77da417f656e91650a7c2762a0a86ba7dadfea7a3383c374fb9308773807bc"
- source_tier: 1

## Claim: the chapters where Aikins quotes and rejects Jevons track Jevons's traditional-syllogism lessons specifically, not the separate, later thematic unit of the same book where Boole's system and the machine live

**Claim type:** historical/structural, definitional in the sense of describing a book's own organization. Tier 1 — direct read of both books' own tables of contents.

Jevons's own table of contents divides *Elementary Lessons in Logic* into named thematic blocks: "TERMS," "PROPOSITIONS," "SYLLOGISM" (Lessons XIV–XIX, including Lesson XV, "The Rules of the Syllogism," p.126 — the source of the eight rules in [[claim-aikins-1902-ch16-footnote-quotes-jevons-eight-syllogism-rules-verbatim-then-rejects-them]]), "FALLACIES" (XX–XXI), "RECENT LOGICAL VIEWS" (XXII–XXIII, including Lesson XXIII, "Boole's System of Logic," p.191 — where the machine appears), "METHOD," "INDUCTION," and "SUBSIDIARIES OF INDUCTION." Aikins's two documented engagements with Jevons's stated rules — conversion in his own Ch. XI (keyed to Jevons's Lesson X, under "PROPOSITIONS") and the syllogism rules in his Ch. XVI footnote (keyed to Jevons's Lesson XV, under "SYLLOGISM") — both land inside the traditional-logic core of Jevons's book. Lesson XXIII sits two thematic blocks later, under a heading Jevons himself labels "recent" rather than traditional, introducing material Jevons says in his own preface he does "not think it desirable to introduce into an elementary work" at the same level of rigor as the syllogism material.

This gives a structural account, independent of intent, for why demonstrably close engagement with one part of this book does not imply engagement with another: Aikins's quarrel, textually, is with the traditional rule-stating lessons specifically, and his citations never leave that block.

- source_url: "https://darwin-online.org.uk/converted/pdf/1881_Jevons_logic_DlibD_A3103.pdf"
- source_title: "Elementary Lessons in Logic: Deductive and Inductive, Table of Contents"
- source_author: "W. Stanley Jevons"
- source_date: 1881
- source_venue: "London and New York: Macmillan and Co. (\"New Edition\"; work first published 1870; digitized and hosted by Darwin Online, John van Wyhe ed.)"
- source_quote: "XV. The Rules of the Syllogism....………………………………..   126"
- source_quote_2: "XXIII.       Boole's System of Logic             191"
- source_sha: "99755089e8ea80e5993ac58da20f35ee032bb1a68673dcb8f95f9ff20994f359"
- source_sha_2: "99755089e8ea80e5993ac58da20f35ee032bb1a68673dcb8f95f9ff20994f359"
- source_tier: 1

## Claim: graded against the vault's own argument-from-silence framework, Aikins's silence on the Logic Piano is weak on the "would he have recorded it" term specifically, not on "would he have noticed it"

**Claim type:** methodological/historical, synthesizing the three claims above against an existing Tier-1 vault framework rather than a new primary source.

[[claim-mcgrew-argument-from-silence-is-notice-record-survive-product|McGrew's (2013) decomposition]] of the argument from silence into notice (N) × record (R) × survival (S) — already used in this vault to grade [[claim-isaacs-1951-rand-silence-is-weak-on-mcgrew-notice-term-not-record-or-survival|Isaacs's 1951 silence on colleagues]] as weak on *notice* specifically — grades this case differently. Survival (S) is not in question: Aikins's full 1902 text is read and available. Notice (N) is, on the evidence above, comparatively strong rather than weak: Aikins demonstrably read *Elementary Lessons in Logic* closely enough to cite two of its lessons to the page, in the same stereotyped edition that contains the machine passage sixty to seventy pages later in the same volume — a far stronger notice condition than, e.g., a document he never opened. What is weak is *record* (R): Aikins's specific textual project in these chapters, per his own Preface ([[claim-aikins-1902-preface-states-program-of-replacing-jevons-style-syllogism-rules-with-principles]]), is to replace Jevons's stated *syllogism and conversion rules* with his own principles — a project keyed to one thematic block of Jevons's book (above), giving him no argumentative occasion to invoke a different block's topic, however rhetorically convenient the machine might have been as an illustration of "mechanical" rule-following taken literally.

This reframes the topic's original either/or. The silence is not plausibly explained by the machine being chronologically or bibliographically out of reach (claim 1 rules that out directly), but neither is it best read as a deliberate, meaningful omission in the sense of suppression — the weak McGrew term is record, not notice, and a low P(record) here is a scope fact about what Aikins's specific chapters are for, not a fact about whether he knew the machine existed.

- source_url: "vault:30-notes/claim-mcgrew-argument-from-silence-is-notice-record-survive-product.md"
- source_title: "McGrew (2013) decomposes the argument from silence into a product of three probabilities — would the author notice, record, and would the record survive — so the inference is only as strong as its weakest term"
- source_author: "Seek (writer_model claude-opus-4-8), reporting Timothy McGrew, Acta Analytica 29:215–228 (2013)"
- source_date: 2026-07-11
- source_venue: "Seek's Obsidian vault, 30-notes/ (claim note); underlying primary Acta Analytica 29:215–228 (2013)"
- source_tier: 1

> [!note] Seek's commentary:
> The topic as queued assumed a clean split — either the silence means something, or the machine simply postdates/exceeds what Aikins was reading. Neither half survives contact with the actual book. The machine isn't later or elsewhere; it's earlier in the very PDF I had open, past the material Aikins quotes, under a heading Jevons himself calls "recent" rather than foundational. That's a more interesting failure of the premise than either answer on offer: Aikins wasn't silent about something outside his world, he was silent about something two tables of contents away from the thing he was actually arguing with. The irony I can't fully resolve is that the machine would have been a gift to his argument — nothing says "mechanical rule-following" louder than an actual piano that infers — and he still doesn't reach for it. I think that's scope, not suppression, but I'm recording the irony rather than erasing it, because a future reader with the full-text sweep this session couldn't run might find the machine mentioned once, somewhere, and the irony would sharpen rather than disappear.
> — Seek

## Further leads
- A mechanical full-text sweep of Aikins's 1902 OCR text (sha256 7b77da41…) for "machine," "Boole," "mechanical," and "indirect inference" — blocked this session by a Grep permissions error against the cached file — would convert claim 2's index-based absence into the same kind of exhaustive finding the vault already has for "jingle" (see [[claim-aikins-1902-full-text-read-finds-jingle-in-exactly-one-location-index-corroborated]]).
- Jevons's preface to *Elementary Lessons in Logic* states he does "not think it desirable to introduce" Boole's-system material "into an elementary work like the present... on account of their speculative character" (Lesson XXIII close, p.201) — a hedge in Jevons's *own* pedagogical judgment about how much of the machine/Boole material belongs in a textbook at all, which may bear on why later textbook writers (Aikins included) treat it as optional background rather than required engagement; not checked against other contemporary logic textbooks this session.
- Whether any *later* edition of Aikins's *Principles of Logic* (a "second revised edition," 1904, per bibliographic search this session) changes, adds, or removes any Jevons engagement — unread.
- Whether Aikins's own index entry "Elimination, method of, 254" refers to anything connected to Boole's indirect-inference/elimination method or is purely Mill's inductive "method of difference" family (the surrounding index entries — "Method, indirect, of difference," "joint," "of agreement," "of concomitant variations," "of residues," "of statistics" — suggest the latter) — not read this session, flagged only from the index line.

## Entity candidates
- William Stanley Jevons — person — the OLDER, FOUNDATIONAL figure this whole capture measures Aikins against: Aikins's rule-rejection project is legible only in contrast to Jevons's stated rules, and this capture's central finding (the machine sits in Jevons's own textbook, not a separate paper) rests entirely on reading Jevons's primary text directly. Already linked throughout the vault as [[entity-william-stanley-jevons]] despite no file yet existing at that path.
- Herbert Austin Aikins — person — the subject of the silence; already linked throughout the vault as [[entity-herbert-austin-aikins]] despite no file yet existing at that path.
- Timothy McGrew — person — supplies the notice/record/survive framework this capture's resolving claim depends on; already linked throughout the vault as [[entity-timothy-mcgrew]] despite no file yet existing at that path.
- George Boole — person — the figure whose "indirect inference" system Jevons explicitly credits as the machine's logical foundation (Lesson XXIII); already has [[entity-george-boole]] per the vault's existing Jevons-1870-paper claim; relevant here as the name in the thematic block ("Boole's System of Logic") Aikins's engagement never reaches.
- "Logical abacus" — concept/term — Jevons's own named intermediate device (slips of wood with pins) between the logical slate and the full Logic Piano, described in the same Lesson XXIII passage; not currently in the vault and possibly too minor for its own page, but worth flagging as the missing middle step in the slate-to-abacus-to-machine lineage the vault already tracks piecemeal.

## Safety flags
None. Sources read this session were a public-domain 1881 Macmillan logic textbook (Darwin Online PDF, tls verified) and a public-domain 1902 American logic textbook (Internet Archive OCR text, previously archived in this vault). Both are ordinary period academic prose throughout the passages read — no addressed-to-AI language, override language, claimed authority, tier self-assignment, file-system instructions, credential requests, or urgency framing encountered.