# Side discussion — the numerology of risk in space flight (the shuttle's moving round number)

*Three sources, pulled 2026-06-13, for a Reckoning-Science side panel on "acceptable risk" in human spaceflight —
the same round-number-standing-in-for-ignorance device, in the most vivid possible setting (a number attached to
an astronaut's life). Register **H39–H41**.*

## The sources
- **Feynman, "Personal Observations on the Reliability of the Shuttle," Appendix F, *Report of the Presidential
  Commission on the Space Shuttle Challenger Accident* (the Rogers Report), Vol. 2, 1986** — NASA, public domain.
  Local: `sources/Feynman_RogersReport_AppendixF_shuttle-reliability.html/.txt`. **[CONFIRMED-primary]**, register **H39**.
- **Elisabeth Paté-Cornell & Robin Dillon, "Probabilistic risk analysis for the NASA space shuttle: a brief
  history and current work," *Reliability Engineering and System Safety* 74(3):345–352 (2001)** (DOI
  10.1016/S0951-8320(01)00081-3). Local: `papers/lateral/patecornell-dillon2001_shuttle-PRA-history.pdf/.txt`.
  **[CONFIRMED-primary]**, register **H40**. *(Operator Taildrop.)*
- **William J. Broad, "NASA Puts Shuttle Mission's Risk at 1 in 100," *New York Times*, 26 Jul 2005** (p. A14).
  Local: `sources/NYT_2005-07-26_shuttle-risk-1-in-100.txt`. **[secondary]** (journalism; quotes NASA directly),
  register **H41**.

## The number that would not hold still
NASA's stated probability of catastrophic shuttle loss swung across **three orders of magnitude**, tracking
institutional confidence and disaster rather than any fixed underlying reality (NYT, Broad 2005; Paté-Cornell &
Dillon 2001):
- **Pre-Challenger (management):** ~**1 in 100,000** per flight — "much closer to that of commercial jets" (the
  Air Transport Association then put an airline disaster at 1 in 2,000,000). Paté-Cornell: NASA experts
  "estimated the probability of a **Solid Rocket Booster (SRB) failure at 10⁻⁵ per flight without any formal
  systems analysis** needed to support such an estimate," giving an LOVC [loss of vehicle and crew] risk "in the
  order of one in several thousand per flight."
- **Feynman / working engineers (1986):** ~**1 in 100**.
- **Post-Challenger, missions resume (1988):** **1 in 50**.
- **Optimism creep (1990s):** 1 in 145 → 1 in 161 → **1 in 254 (1998)**; then new studies "brought the figure
  back down to **1 in 123**."
- **Post-Columbia (2005):** "roughly **1 flight in 100**" (NASA spokesman Allard Beutel) — "because we have a
  better understanding of the craft's workings."
- **The actual record, meanwhile:** "the actual rate of catastrophic failure… now stands at **2 flights in 113,
  or 1 in 57**" (NYT 2005). The aspiration: "'You'd like to go to 1 in 1,000,' he said. 'But you're never goin[g
  to].'"

## Feynman, verbatim (Appendix F) — the management round number as "fantasy"
- "The estimates range from roughly **1 in 100 to 1 in 100,000**. The **higher figures come from the working
  engineers, and the very low figures from management**."
- "A risk of **1 part in 100,000** would imply that one could put a Shuttle up each day for **300 years**
  expecting to lose only one."
- "it would appear that, for whatever purpose… **the management of NASA exaggerates the reliability of its
  product, to the point of fantasy**."
- The closing line: "**For a successful technology, reality must take precedence over public relations, for
  nature cannot be fooled.**"

## Why it belongs in the series — three hooks
1. **The same device, naked.** A round number stands in for knowledge no one has — and here the number is
   pinned to a human life. The managerial **10⁻⁵** ("without any formal systems analysis," Paté-Cornell) is the
   aerospace twin of FDA's **10⁻⁶** (Hutt/SOM) and Mantel's **10⁻⁸** — all chosen, not derived. Feynman's
   "to the point of fantasy" is Kelly's "Myth of 10⁻⁶" (H12) and Mantel's "we just pulled it out of a hat."
2. **The same dedeterminism-vs-probability fault line as the chemical story.** Paté-Cornell: probabilistic
   failure risk "**clashed (and still does) with the engineering culture, primarily based on safety factors**…
   the problem with safety factors is that they are **not directly linked to the probability of system
   failure**." That is *exactly* the NOAEL-safety-factor (Lehman-Fitzhugh, essay one) vs quantitative-risk
   (Mantel-Bryan→10⁻⁶) split this dossier tracks — playing out in rocketry on the same timeline.
3. **"Who bears the risk," taken to the top.** The acceptable per-exposure risk the institutions actually
   tolerate scales with the perceived benefit and who volunteers:
   - **Astronaut:** ~**10⁻² per flight** (1 in 100) — the most permissive of all.
   - **Deployed soldier:** **10⁻⁴** (per 1-yr deployment; Army RD-230, H36).
   - **Food-additive consumer:** **10⁻⁶ lifetime** (FDA, H25/H32).
   - **Radiation public ("negligible"):** **10⁻⁷/yr** (NCRP, H34).
   Six orders of magnitude separate the astronaut from the eater — not because the biology differs, but because
   the *benefit*, the *volition*, and the *politics* do. The round number is a benefit-weighted political choice
   wearing the costume of a measurement — the series' thesis, from a rocket.