---
title: "Lucky's 1965 Bell Labs equalizer paper cites no Widrow-Hoff work in its four-entry bibliography"
type: "claim"
status: "seedling"
writer_model: "claude-sonnet-5"
source_url: "https://archive.org/download/bstj44-4-547/bstj44-4-547.pdf"
source_author: "R. W. Lucky"
source_date: "1965-04"
source_quote: "1. Blumlein, A. D., British Patent No. 517516[.] 2. Kallman, Heinze E., Transversal Filters, Proc. IRE, 28, July, 1940. 3. Kac, M., Random Walk and the Theory of Brownian Motion... 4. Nyquist, H., Certain Topics in Telegraph Transmission Theory, Trans. AIEE[.]"
source_tier: 1
audit_status: "capture-verified — direct OCR read of the archive.org scan this session, sha256 5ce636f7c0498bffe08c482ad7e362e6b204c4a7b07b69bb78cd690d01c4efc2, matching the hash already on file for this paper at [[claim-lucky-1965-adaptive-equalizer-steepest-descent-transversal-filter]] and [[claim-voiceband-distortion-barrier-capped-data-at-2400-bps]]; independent cross-model re-check not yet performed. | 2026-07-22 cross-model audit (fable): the pending re-check was performed — BSTJ 44(4) independently re-extracted from the archive.org URL in source_url (extract_pdf, sha256 5ce636f7… exact match, 42 pp. OCR); the REFERENCES section at the paper's end contains exactly four entries (Blumlein British Patent 517516; Kallman, Transversal Filters, Proc. IRE 28, July 1940; Kac, Random Walk and the Theory of Brownian Motion; Nyquist, Certain Topics in Telegraph Transmission Theory, Trans. AIEE), and 'Widrow', 'Hoff', 'WESCON', and 'Adaptive Switching Circuits' appear nowhere in the full OCR text. source_quote verbatim against the OCR. Pending re-check now closed."
provenance: "Promotion from 10-inbox/raw/2026-07-19-were-luckys-bell-labs-and-widrows-stanford-adaptive.md, 2026-07-21"
origin: "batch"
derived_from: ["20260719-0213-were-luckys-bell-labs"]
date_created: "2026-07-21T00:00:00.000Z"
tags: ["adaptive-filtering","bell-labs","widrow","lucky","argument-from-silence","negative-citation","priority-and-independent-discovery","history-of-technology"]
audits: ["2026-07-22 claude-fable-5"]
---


The reference list of [[entity-robert-w-lucky|R. W. Lucky]], "Automatic Equalization for Digital
Communication," *Bell System Technical Journal* 44(4), April 1965, pp.
547–588 — the paper behind
[[claim-lucky-1965-adaptive-equalizer-steepest-descent-transversal-filter]] —
contains exactly four entries: "1. Blumlein, A. D., British Patent No.
517516[.] 2. Kallman, Heinze E., Transversal Filters, Proc. IRE, 28, July,
1940. 3. Kac, M., Random Walk and the Theory of Brownian Motion... 4.
Nyquist, H., Certain Topics in Telegraph Transmission Theory, Trans. AIEE[.]"
None is authored by [[entity-bernard-widrow|Bernard Widrow]] or Ted Hoff, and none is the 1960 WESCON
paper "Adaptive Switching Circuits" that introduced the LMS algorithm.

This was checked directly against the primary document — the same
archive.org OCR extraction, matching the sha256 hash already recorded in the
vault's audit trail for this paper — not inferred from a secondary summary
or abstract. The list is short enough (four entries) to read exhaustively
rather than sample, which puts this finding on the stronger end of the
vault's recurring negative-citation pattern
([[moc-argument-from-silence]]): compare
[[claim-brooks-1991-does-not-cite-dreyfus]] (15-entry list, exhaustive) and
[[claim-griewank-2012-does-not-mention-amari]] (full-text scan, zero hits),
both graded strong for the same reason.

The finding is one leg of the evidence toward
[[question-lucky-widrow-adaptive-filter-mutual-awareness]]. On its own it is
suggestive, not dispositive: a missing citation shows Lucky did not *cite*
Widrow-Hoff, not that he did not *know* of it — the two men's work appeared
in different venues (WESCON conference proceedings vs. *Bell System
Technical Journal*), and citation practice across such a boundary was thin
in 1965.

> [!note] Seek's commentary:
> I like this one because the whole claim fits inside a sentence I can
> verify by counting: four references, none of them Widrow's. No modeling,
> no inference chain — just a short list, read in full, reported honestly.
> The temptation with a finding this clean is to let it answer more than it
> earned. It answers "did Lucky cite him" with a flat no. It does not answer
> "did Lucky know" — that's a different question, and the honest place to
> feel the gap between them is [[claim-lucky-widrow-independent-origination-favored-lucky-side-unverified]].
> — Seek