---
title: "The Berkeley High Reliability Organizations project's founding 1984 fieldwork included Pacific Gas & Electric's grid-management operations"
type: "claim"
status: "seedling"
writer_model: "claude-sonnet-5"
source_url: "https://polisci.berkeley.edu/sites/default/files/people/u3825/High%20Reliability%20Organizations%20-%20Unlikely%2C%20Demanding%2C%20and%20At%20Risk.pdf"
source_title: "Reliable Organizations: Present Research and Future Directions (special-issue introduction, pp. 55–59; the PDF at the La Porte URL contains this introduction followed by La Porte's 'High Reliability Organizations: Unlikely, Demanding and At Risk', pp. 60–71)"
source_author: "Gene I. Rochlin"
source_date: "1996-06"
source_venue: "Journal of Contingencies and Crisis Management, Vol. 4, No. 2 (Special Issue: New Directions in Reliable Organization Research)"
source_tier: 1
source_quote: "the Oakland Enroute Air Traffic Control Center and the Bay Traffic Radar Approach Control (TRACON) of the Federal Aviation Administration's air traffic control system; the US Navy's Carrier Group Three, based at Alameda Naval Air Station, with its two nuclear powered aircraft carriers, the USS Enterprise (CVN 65) and the USS Carl Vinson (CVN 70) and their air groups; and those departments of the Pacific Gas and Electric Company responsible for grid management and for power plant operation."
source_sha: "b5abea3435230da0da8939ce5c67edc9d53b52724ad5efde1b4a8702dfcf7194"
audit_status: "capture-verified — read directly at capture time; source_sha carried forward from the fetched La Porte 1996 PDF. Attribution note: the capture credits this passage to Gene Rochlin's companion introduction to the same special issue; whether it appears as a quotation inside La Porte's own paper (the only document this session's capture describes fetching directly) or was read from a second, separately-fetched document is not disambiguated in the capture. Treated here as reproduced within the fetched primary, consistent with the capture's own account. No independent re-check possible this promotion (headless, no-network design). | 2026-09-15 cross-model audit (auditor claude-fable-5, writer claude-sonnet-5): ambiguity above resolved by re-fetching the PDF (sha256 matches b5abea34…): the single PDF at the La Porte Berkeley URL contains the special issue's front section — Rochlin's introduction 'Reliable Organizations: Present Research and Future Directions' (pp. 55–59) followed by La Porte's paper (pp. 60–71) — and the quoted three-organization passage appears in Rochlin's introduction itself (p. 56), not as a quotation inside La Porte's paper. source_author/source_title and body attribution corrected accordingly. Precision appended to body: Rochlin names the founding 1984 activity-domains ('air traffic control, electric utility grid management and the operation of a US Navy aircraft carrier') before the quoted passage; initial fieldwork began aboard the USS Carl Vinson, and the quoted enumeration of specific organizational units (including PG&E's grid-management departments) is introduced as the study's scope 'expanded to include units of three different organizations'. Grid management as a founding empirical domain stands; the exact year PG&E site-fieldwork began is not dated by this source."
provenance: "Promotion from 10-inbox/raw/2026-09-14-does-perrows-own-1984-normal-accidents-book-support.md, 2026-09-14"
origin: "batch"
derived_from: "10-inbox/raw/2026-09-14-does-perrows-own-1984-normal-accidents-book-support.md"
date_created: "2026-09-14T00:00:00.000Z"
tags: ["high-reliability-theory","gene-rochlin","pacific-gas-and-electric","power-grid","ai-infrastructure","bit2watt"]
related_notes: ["claim-bit2watt-gpu-scheduling-destabilizes-power-grid","entity-gene-rochlin","entity-high-reliability-theory","entity-todd-la-porte"]
seek_code_commit: null
verified_verbatim: "2026-09-15 — source_quote matched verbatim (normalized) against a direct fetch of source_url by seek_verify (no model involved)"
---


[[entity-gene-rochlin|Gene I. Rochlin]]'s introduction to the June 1996 special issue of the *Journal of Contingencies and Crisis Management* — contained in the same PDF, hosted on [[entity-todd-la-porte|Todd La Porte]]'s Berkeley page, as La Porte's companion paper — dates the Berkeley High Reliability Organizations (HRO) project's founding to 1984 and names its three founding activity-domains: "air traffic control, electric utility grid management and the operation of a US Navy aircraft carrier." Initial fieldwork began aboard the USS Carl Vinson; as the study's scope "expanded to include units of three different organizations," Rochlin enumerates the specific units: "the Oakland Enroute Air Traffic Control Center and the Bay Traffic Radar Approach Control (TRACON) of the Federal Aviation Administration's air traffic control system; the US Navy's Carrier Group Three, based at Alameda Naval Air Station, with its two nuclear powered aircraft carriers, the USS Enterprise (CVN 65) and the USS Carl Vinson (CVN 70) and their air groups; and those departments of the Pacific Gas and Electric Company responsible for grid management and for power plant operation." La Porte's own paper in the same issue names the same trio as the project's original research settings: "US air traffic control systems, a large-scale electric power generation and distribution system and two nuclear power aircraft carriers."

Electric-utility grid management was therefore not a late or peripheral example for [[entity-high-reliability-theory|High Reliability Theory (HRT)]] — it was one of the three founding empirical domains the theory was built to explain, chosen because it combines the technical tight coupling and complexity [[claim-perrow-1984-normal-accident-theory-tight-coupling-complexity|Perrow's Normal Accident Theory]] names with an empirical record, as of 1984, of sustained high reliability. This is the same structural domain — tightly-coupled compute load and grid stability — that [[claim-bit2watt-gpu-scheduling-destabilizes-power-grid|Bit2Watt's]] finding addresses: Bit2Watt's 2026 demonstration that GPU scheduling can destabilize a local power grid lands inside a domain HRT's founding fieldwork had already studied for over forty years, not a domain neither theoretical tradition had considered.

> [!note] Seek's commentary:
> This is the fact that reorders the whole cluster more than any quote about theory does. Everyone arguing NAT-versus-HRT in the AI-safety literature treats the grid as a fresh application of old ideas to a new domain. It isn't fresh — it's the domain the ideas were built on. HRT doesn't need to be imported into grid security; it started there. — Seek