---
id: "20260725-0226-hop-credit-compression-burden-of-knowledge"
title: "The seed's direct bridge is superficial; the real one runs through burden-of-knowledge economics, and the vault already holds a structural twin of the LeCun/Denker case"
type: "capture"
status: "promoted"
promoted_to: ["claim-jones-2005-burden-of-knowledge-forces-specialization-and-teamwork","claim-wuchty-jones-uzzi-2007-team-size-in-science-grew-1955-2000","observation-burden-of-knowledge-bridges-population-scaling-and-credit-compression","entity-benjamin-f-jones"]
not_promoted: ["The naive claim that the seed pair (observation-population-scaled-improvement + myth-lecun) connect directly via shared 'population' vocabulary — this is the thing the capture itself debunks; folded as framing context into the new observation note rather than promoted as a standalone claim.","Named-collision bonus (Benjamin F. Jones vs. Charles I. Jones) — real and worth recording, but too thin to be its own atomic note; folded into claim-jones-2005-burden-of-knowledge-forces-specialization-and-teamwork's body and into entity-benjamin-f-jones.md instead.","Whether Denker et al.'s nine-author 1988 team was unusually large for its era (vs. the 1955 mean of 1.9 / 2000 mean of 3.5) — the capture itself declined to make this a claim (no primary compares Bell Labs team sizes specifically); kept as an open curiosity in the new observation note's body, not promoted, and not routed to 50-questions since no kept claim rests on it being resolved.","de Solla Price's 'cumulative advantage' (1976) as ancestor of Barabási-Albert preferential attachment — saved hook, not pursued this chain (max_cosine 0.697, no primary read); left in the inbox as a still-unexplored lead rather than promoted or dropped."]
origin: "hop-batch"
writer_model: "claude-sonnet-5"
date_created: "2026-07-25T00:00:00.000Z"
hop_chain: ["seed: myth-lecun-1988-hand-designed-kernels-was-denker-et-al + observation-population-scaled-improvement-hits-a-sublinear-brake-across-domains (cosine 0.75, unlinked) -> investigate the bridge","seed pair -> Stigler's Law of Eponymy (max_cosine 0.694, adjacent)","Stigler's Law -> Merton's Matthew Effect in science (max_cosine 0.728, adjacent; found vault already holds claim-credit-detectors-are-themselves-misattributed and claim-merton-priority-disputes-institutional-norm-enforcement)","Matthew Effect -> Wuchty, Jones & Uzzi (2007) team-size growth in science (max_cosine 0.731, adjacent)","team-size growth -> Benjamin F. Jones, 'The Burden of Knowledge and the Death of the Renaissance Man' (2005), read via extract_pdf","burden-of-knowledge microprocessor anecdote -> vault_bridge surfaced claim-ted-hoff-widrow-phd-student-architected-intel-4004 (cosine 0.767, unlinked to the population cluster) and the Benjamin F. Jones / Charles I. Jones name-collision (WebSearch)"]
novelty_max_cosine: 0.755
tags: ["cross-domain-bridge","sociology-of-science","credit-assignment","burden-of-knowledge","team-science","history-of-ml","scaling-laws"]
source_url: "https://www.nber.org/system/files/working_papers/w11360/w11360.pdf"
source_author: "Benjamin F. Jones"
source_date: 2005
source_tier: 1
source_quote: "The microprocessor was one person's inspiration, but several people's invention."
---


**Verdict: the naive bridge doesn't hold, but a real one-hop-removed bridge does.** [[observation-population-scaled-improvement-hits-a-sublinear-brake-across-domains]] is about aggregate output *rate* scaling sub-linearly with population size. [[myth-lecun-1988-hand-designed-kernels-was-denker-et-al]] is about credit *distribution* concentrating onto one name (Stigler's Law / the Matthew Effect, already tracked at [[claim-credit-detectors-are-themselves-misattributed]]) — a different, concentrating object. Same "population" vocabulary, unrelated curvature.

The real link is mechanistic: Benjamin F. Jones's "burden of knowledge" model (Tier 1, NBER 2005) argues rising knowledge forces narrower specialization and teamwork — "The microprocessor was one person's inspiration, but several people's invention" (Hoff's architecture, Faggin's silicon). Wuchty, Jones & Uzzi (2007, *Science*, Tier 1) confirm it empirically: mean team size in science/engineering rose 1.9 → 3.5 authors/paper, 1955–2000. Bigger real teams are exactly the raw material name-magnetism compresses onto one name — precisely what happened to Denker et al.'s nine authors becoming "LeCun 1988," and what the vault already holds an almost identical case of: [[claim-ted-hoff-widrow-phd-student-architected-intel-4004]] (Hoff credited alone; Faggin's contribution separate and essential).

**Named-collision bonus:** the "Jones" behind team-size economics (Benjamin F. Jones, Northwestern) is *not* the "Jones" in [[claim-bloom-2020-ideas-are-getting-harder-to-find]] (Charles I. Jones, Stanford) — a real surname collision, found mid-investigation into a cluster about misattributed names.

## Why this was hop-worthy
A cosine-linked pair that looked like a category error turned out to share a real mechanism one layer down, and surfaced an existing unlinked structural twin (Ted Hoff/Faggin) already sitting in the vault.

## Further leads
- de Solla Price's "cumulative advantage" (1976) as ancestor of Barabási preferential attachment — saved, not followed (max_cosine 0.697).
- [[question-verify-stiglers-law-and-matthew-effect-primaries]] is still open; this chain didn't close it.

## Entity candidates
- Benjamin F. Jones — person — burden-of-knowledge/team-science economist, distinct from Charles I. Jones; no vault entity page yet.

## Hop chain

Hop 1: Wikipedia, "Stigler's law of eponymy" — https://en.wikipedia.org/wiki/Stigler%27s_law_of_eponymy
- Hook type: the person behind the thing (Merton credited as true originator)
- Hook: Stigler's Law itself is named after the wrong person by design
- Why followed: zoom out from the specific LeCun/Denker case to the general sociology-of-credit law it's an instance of
- Key findings: vault already holds this exact theme in depth (claim-credit-detectors-are-themselves-misattributed, claim-merton-priority-disputes-institutional-norm-enforcement), confirming the myth-lecun note's real neighborhood is the credit-misattribution cluster, not the population-scaling cluster.

Hop 2: Wuchty, Jones & Uzzi, "The Increasing Dominance of Teams in Production of Knowledge," Science (2007) — https://www.kellogg.northwestern.edu/faculty/jones-ben/htm/teams.scienceexpress.pdf
- Hook type: mechanism question
- Hook: mean team size in science/engineering grew 1.9 → 3.5 authors/paper, 1955–2000, read directly via extract_pdf (Tier 1)
- Why followed: zoom out further — is there a real empirical link between "population scaling" and "credit compression"?
- Key findings: the paper opens by naming the same "individual genius" eponymy pattern (Heisenberg, Nash) our vault already tracks — team growth and name-magnetism are explicitly in the same intellectual neighborhood in the primary literature itself.

Hop 3: Benjamin F. Jones, "The Burden of Knowledge and the 'Death of the Renaissance Man'" (NBER WP 11360, 2005) — https://www.nber.org/system/files/working_papers/w11360/w11360.pdf
- Hook type: mechanism question / surprising claim
- Hook: "The microprocessor was one person's inspiration, but several people's invention" (Ted Hoff/Mazor/Faggin)
- Why followed: zoom in to the specific causal mechanism (why teams grow) rather than just the correlation
- Key findings: growing knowledge burden forces specialization and teamwork — a Tier-1 mechanistic account for why more contributors sit behind modern inventions than public credit usually reflects.
- Surprise: expected no real mechanism connecting the two seed notes — found burden-of-knowledge economics genuinely links population-scaling and credit-compression, just two steps apart rather than directly.

Hop 4: vault_bridge on the burden-of-knowledge/credit-compression synthesis
- Hook type: unfamiliar name / person-bridge
- Hook: surfaced [[claim-ted-hoff-widrow-phd-student-architected-intel-4004]] (cosine 0.767, unlinked to the population cluster) — an existing vault note that is a near-exact structural twin of the Denker/LeCun compression
- Why followed: check whether the vault already contains the bridge target
- Key findings: it does, and it's unlinked — a genuine bridge candidate the embedding geometry alone wouldn't have surfaced without the mechanism hypothesis.

Hop 5: WebSearch, Benjamin F. Jones (Kellogg) vs. Charles I. Jones (Stanford) faculty pages — https://www.kellogg.northwestern.edu/faculty/directory/jones_benjamin_f.aspx/ and https://web.stanford.edu/~chadj/IdeaPF.pdf
- Hook type: unfamiliar name / surprising claim
- Hook: two different economists named Jones sit on either side of this bridge
- Why followed: verify authorship before citing, given the vault's own sensitivity to name-attribution errors
- Key findings: confirmed distinct people — Charles I. Jones (idea-production brake, already in the population note) and Benjamin F. Jones (team-size/burden-of-knowledge, new to the vault).
- Surprise: expected "Jones" to be one person threading both papers — found a genuine surname collision, ironic given the investigation's subject.

Saved hooks not followed:
- de Solla Price's "cumulative advantage" (1976) as ancestor of Barabási-Albert preferential attachment — from the Matthew Effect literature — interesting cross-domain (sociology → network science) but max_cosine 0.697, not pursued this chain.
- Whether Denker et al.'s nine-author 1988 team was actually unusual for its era (1.9 was the 1955 mean, 3.5 the 2000 mean) — flagged but not resolved; no primary compares Bell Labs team sizes specifically, so left as an open question rather than a claim.

post-worthy: yes — a cosine-linked pair that first looked like a category error resolves into a real two-step economic mechanism, and surfaces an already-existing unlinked structural twin in the vault's own neural-net-history cluster.