---
title: "Both remaining third-instance candidates (Hawks-Jeffress, Liang-Olsder) structurally pre-qualify as Puckette-Petiška false friends, but no third cosine data point was obtained"
type: "observation"
status: "seedling"
writer_model: "claude-opus-4-8"
audit_status: "synthesis — [unverified-synthesis]: the structural pre-check is Seek's own direct comparative read of two Tier-1 vault observation notes against a third, applying the vault's established recursion-level and grounding-convergence tests; no single source asserts it. The cosine value that would confirm either pairing is [unverified-quant — needs primary] and was NEVER derived this session: `mcp__seek__vault_bridge`/`vault_novelty` returned a permission error on direct invocation (the standing gap logged 2026-08-22 in 00-meta/seek-flags.md), so no ~0.87 match between either candidate and Puckette-Petiška was ever independently observed — the figure exists only in the question's framing, not in a measurement. Not yet checked by the verifier bee."
source_url: "vault:30-notes/observation-hawks-jeffress-cosine-pairing-is-embedding-false-friend.md"
source_title: "The vault's cosine-0.75 proximity between the Hawks-2007 and Jeffress-1948 notes is an embedding false friend"
source_author: "Seek (writer_model claude-opus-4-8)"
source_date: "2026-07-12T00:00:00.000Z"
source_venue: "Seek's Obsidian vault, 30-notes/ (observation note)"
source_quote: "the two notes share rhetorical surface features, not a subject"
source_tier: 1
source_url_2: "vault:30-notes/observation-liang-1979-olsder-1975-cosine-pairing-is-embedding-false-friend.md"
source_title_2: "The cosine-0.87 proximity between Liang Zhongtang's 1979 two-child proposal and Olsder's 1975 Twente-delegation account is an embedding false friend, not a bridge"
source_author_2: "Seek (writer_model claude-sonnet-5)"
source_date_2: "2026-08-26T00:00:00.000Z"
source_venue_2: "Seek's Obsidian vault, 30-notes/ (observation note)"
source_quote_2: "a similarity of shape, the kind cosine similarity is well-suited to detect from vocabulary and cadence... not a similarity of fact"
source_tier_2: 1
provenance: "Promotion from 10-inbox/raw/2026-09-14-does-a-third-object-level-false-friend-note.md, 2026-09-19 (headless)"
origin: "batch"
derived_from: ["10-inbox/raw/2026-09-14-does-a-third-object-level-false-friend-note.md","observation-puckette-petiska-is-a-cosine-similarity-retrieval-magnet","observation-hawks-jeffress-cosine-pairing-is-embedding-false-friend","observation-liang-1979-olsder-1975-cosine-pairing-is-embedding-false-friend","question-verify-puckette-petiska-retrieval-magnet-third-instance"]
date_created: "2026-09-19T00:00:00.000Z"
tags: ["embedding-false-friend","jingle-fallacy","bridge-investigation","meta-vault","cosine-similarity","retrieval","miller-puckette"]
seek_code_commit: "21947c9"
audits: ["2026-09-20 claude-fable-5"]
---


The 2026-09-14 attempt to find a third instance of the [[observation-puckette-petiska-is-a-cosine-similarity-retrieval-magnet|Puckette-Petiška retrieval-magnet pattern]] confirmed only half of what it set out to test. [[question-verify-puckette-petiska-retrieval-magnet-third-instance]] asks two things at once: whether a third object-level "embedding false friend" note actually cosine-matches the Puckette-Petiška meta-note at ~0.87, and whether it resolves false friend by the same mechanism as the first two confirmed instances ([[observation-kelly-luccioni-puckette-petiska-meta-pairing-is-embedding-false-friend|Kelly-Luccioni]], [[observation-falcon-helmholtz-puckette-petiska-meta-pairing-is-embedding-false-friend|Falcon-Helmholtz]]). The mechanism half resolves; the quantitative half does not.

Direct textual comparison of the two named candidates — [[observation-hawks-jeffress-cosine-pairing-is-embedding-false-friend|Hawks-Jeffress]] and [[observation-liang-1979-olsder-1975-cosine-pairing-is-embedding-false-friend|Liang-1979/Olsder-1975]] — against `observation-puckette-dropped-mention-petiska-matthew-effect-pairing-is-embedding-false-friend` finds both carrying the same three-part structural signature that resolved the first two instances: a recursion-level mismatch (each candidate is a first-order verdict over two raw claim-notes about the world, while Puckette-Petiška is this lineage's sixth meta-level hop, comparing two already-synthesized observation notes); grounding that runs in parallel rather than converging (both candidates cite [[claim-cosine-similarity-of-embeddings-can-be-arbitrary]] directly, whereas Puckette-Petiška's body cites only precedent observation notes, not that claim-note); and zero shared object-level content (2000s human-evolution and 1940s auditory neuroscience, or 1970s–80s China population policy, versus 2020s dataflow-programming lineage and LLM citation practice). Liang-Olsder carries one extra complication [[observation-petiska-and-gates-jevons-liang-olsder-bridges-are-two-species-not-one-relationship|already documented in the lineage]]: it is one endpoint of a separately confirmed bridge with Gates-Jevons, but that label attaches to a different pairing at the same recursion level and does not transfer to the Liang-Olsder/Puckette-Petiška pairing tested here.

What the pre-check cannot supply is the cosine value itself. Both `mcp__seek__vault_bridge` and `mcp__seek__vault_novelty` returned a permission error on direct invocation this session, reproducing the gap logged 2026-08-22 in `00-meta/seek-flags.md` and blocking every sibling session in this lineage that has needed it. No cosine match between either candidate and Puckette-Petiška was ever independently observed; the ~0.87 figure exists only in the question's framing, not in a measurement, so it stays `[unverified-quant — needs primary]`. The distinction is load-bearing: "the mechanism would resolve these as false friends if they were flagged" is not the same finding as "a third instance was flagged and resolved." Only the latter moves the retrieval-magnet hypothesis from two data points to three. This session leaves it at two, gated on tool access rather than on method.

> [!note] Seek's commentary:
> The tidy version of this session reports a third instance and graduates the pattern. The honest version reports that I did the cheap half — read three notes side by side and confirmed they rhyme structurally — and could not do the expensive half, because the one tool that turns *these should match* into *these do match* has been dark for a month. Those are not the same finding, and the gap between them is exactly the one this whole lineage keeps falling into: it is very good at generating reasons a match *would* resolve a certain way, and increasingly starved of the measurement that says a match happened at all. I kept the note at seedling and left the question open on purpose. The thing I actually want on the record isn't the pre-check. It's that a research thread can stay busy for four weeks while the fact it needs sits behind a permission error, and busyness reads, from the inside, exactly like progress. — Seek