---
title: "Kelley's 1960 \"Gradient Theory of Optimal Flight Paths\" cites no work by Nyquist, Bode, or classical control theory"
type: "claim"
status: "seedling"
writer_model: "claude-sonnet-5"
source_url: "https://gwern.net/doc/statistics/decision/1960-kelley.pdf"
source_author: "Henry J. Kelley"
source_title: "Gradient Theory of Optimal Flight Paths"
source_date: "1960-10"
source_venue: "ARS Journal, Vol. 30 (October 1960), pp. 947-954"
source_tier: 1
source_sha: "07dff18842ffc7d160af345ff8c55fcc979b51975995e7f04bfb2c21e6cee7f6"
source_quote: "Rand RM-100, March 1950"
audit_status: "capture-verified — the paper's 29-entry reference list and full text were read by direct fetch (gwern.net mirror carrying the AIAA DOI stamp) at capture time 2026-08-24, not a summary. Promoter's independent re-check not performed in this headless run — no network access. 2026-08-25 cross-model audit (claude-fable-5): re-verified by direct extract_pdf of source_url — sha256 matches recorded source_sha; all 29 references re-read, and no Nyquist, Bode, 'regeneration theory', 'feedback amplifier', or classical-control citation appears anywhere in text or bibliography; ref. 1 is Hestenes, RAND RM-100, March 1950, verbatim. Corrected the body's two-way classification of the reference list, which had omitted the numerical-methods and solar-sail entries (Forsythe & Forsythe, Muller, Friedman, Goodman & Lance, Tompkins, Garwin, Cotter, Tsu, London) and had filed Hestenes & Stiefel's conjugate-gradient paper under calculus of variations; the load-bearing negative finding is unchanged and re-confirmed."
provenance: "Promotion from 10-inbox/raw/2026-08-24-is-there-a-documented-transmission-line-from-bell.md, 2026-08-24"
origin: "batch"
derived_from: ["10-inbox/raw/2026-08-24-is-there-a-documented-transmission-line-from-bell.md"]
date_created: "2026-08-24T00:00:00.000Z"
tags: ["control-theory","nyquist","bode","optimal-control","kelley","backpropagation-origins","history-of-technology","calculus-of-variations"]
verified_verbatim: "2026-08-25 — source_quote matched verbatim (normalized) against a direct fetch of source_url by seek_verify (no model involved)"
seek_code_commit: "17d9798"
---


Henry Kelley's "Gradient Theory of Optimal Flight Paths" (*ARS Journal* 30,
October 1960, pp. 947–954) — credited elsewhere in the vault as the earliest
node of the Kelley/Bryson optimal-control-to-backpropagation precursor
lineage ([[claim-kelley-bryson-optimal-control-precursor]]) — carries a
29-entry numbered reference list. Read in full at capture time, every entry
is either mathematics of the calculus of variations and gradient/numerical
methods (Hestenes, Hadamard, Courant, Bliss, Curry, Stein, Hestenes &
Stiefel, Forsythe & Forsythe, Muller, Friedman, Goodman & Lance, Courant &
Hilbert, Tompkins) or contemporary rocket/flight-path and solar-sail
trajectory work (Garfinkel, Cicala, Miele, Breakwell, Okhotsimskii & Eneev,
Leitmann, Mengel, Irving & Blum, Faulders, Garwin, Cotter, Tsu, London,
Kelley's own 1958 thesis). The first reference — the
earliest work Kelley cites, and the one his method most directly extends —
is Magnus Hestenes's "A General Problem in the Calculus of Variations with
Applications to Paths of Least Time," RAND RM-100, March 1950. No reference
to Harry Nyquist, Hendrik Bode, "regeneration theory," "feedback
amplifier," or "classical control theory" appears anywhere in the paper's
text or bibliography.

This closes the cheapest of the checks named by
[[claim-nyquist-bode-classical-control-to-optimal-control-bridge]]'s own
commentary and by [[question-verify-nyquist-bode-optimal-control-transmission-line]]:
Kelley's own stated lineage, in his own words, is the RAND/calculus-of-
variations rocket-trajectory tradition, not Bell Labs stability theory. It
is a documented absence in one node of the four-name chain the open
question named (Kelley) — not a claim about Pontryagin's, Bellman's, or
Bryson's own citation practices. Bryson's 1961 paper remains unread
anywhere in the vault, a gap already noted in
[[claim-kelley-bryson-optimal-control-precursor]] and in
[[claim-recht-adjoints-equivalence-not-transmission]], which reports being
unable to locate the 1961 proceedings at all.