# *De Minimis Risk* (Whipple ed., 1987) — close-read of ch. 6 (Travis & Richter) and ch. 8 (Flamm/Rulis)

*Two chapters of the held volume (register **H34**; `papers/lateral/whipple1987_de-minimis-risk_vol2_plenum.txt`),
close-read 2026-06-13. Both are **[CONFIRMED-primary]**.*

## Ch. 6 — Curtis C. Travis & Samantha A. Richter (Oak Ridge), "On Defining a De Minimis Risk Level for Carcinogens" (pp. 61–73)
The de-minimis-*level* companion to Travis et al. 1987 (register **H24**, the 132-decisions paper). Two methods,
both yielding a number **between 10⁻⁴ and 10⁻⁶** — and neither is a safety threshold:
- **Method 1 — natural-background standard deviation** (extending Adler & Weinberg 1978): "since no correlations
  have been detected between variations in natural background radiation and adverse health effects, small
  additions to natural exposure should be acceptable. **The difficulty lies in defining 'small.'**" Using the SD
  of background radiation/chemical-carcinogen concentrations (benzene, formaldehyde, chloroform) → a lifetime
  cancer risk "**on the order of 10⁻⁴**" (Table 1: 20 mrem/yr SD → 1.5 × 10⁻⁴; benzene/chloroform SDs → ~2 × 10⁻⁴).
  (ll. 2884–95, 3033–35, 3529–31)
- **Method 2 — revealed regulatory practice** (their own 132-decisions review): "regulatory practices cannot be
  explained on the basis of lifetime risk alone… The key… is the relationship between lifetime risk and
  **population risk**." For small exposed populations, agencies seldom acted below **~10⁻⁴**; "as population risk
  approaches 250 cancer deaths per year (… the total U.S. population), the de minimis level drops to **10⁻⁶**."
  (ll. 3453–61)
- **Conclusion:** "the de minimis level **varies from 10⁻⁴ to 10⁻⁶, depending on the size of the population
  impact**… the regulatory process combines many economic, social, and political factors into an implicit
  expression of society's willingness to accept risk." (ll. 3529–36) → **10⁻⁶ is the floor reserved for
  whole-population exposures; it is a function of population size and politics, not toxicology.**

## Ch. 8 — Flamm, Lake, Lorentzen, **Rulis**, Schwartz & Troxell (FDA), "Carcinogenic Potencies and Establishment of a Threshold of Regulation for Food Contact Substances" (pp. 87–92)
The **chemical-side machinery** — the 1987 precursor to Rulis 1992 (H13) and the 1995 ToR rule (D6):
- The no-threshold premise: "the prevailing view is that **most carcinogens do not have 'no effect' levels**."
  (ll. 4343–44)
- The method: a **Gaussian probability distribution of carcinogen potencies** (343 carcinogens from Gold et
  al.'s Carcinogenic Potency Database), transformed to a "risk-equivalent exposure distribution" via **low-dose
  linearity** ("low-dose risk equals potency times exposure"). (ll. 4354–70)
- The result: "**1 ppt** [part per trillion in the diet] corresponds to a dose of between 10⁻⁷ and 10⁻⁸ mg/kg
  per day… the same 1-ppt cutoff would **exclude almost all risks of 10⁻⁶ or greater**." (ll. 4428–33) — the
  concentration-as-proxy device behind 0.5 ppb.
- **The risk standard is openly arbitrary:** "by establishing a given risk standard, **say 10⁻⁸ or 10⁻⁶ per
  lifetime**… **Clearly, any risk level could be chosen** and a similar analysis undertaken." (ll. 4366–67, 4433–34)
- **Grounded on the de minimis doctrine:** its **Reference 1 is *Monsanto v. Kennedy*, 613 F.2d 947 (D.C. Cir.
  1979)** (l. 4468) — the chemical 10⁻⁶ rests explicitly on the judicial de minimis authority in `10` §2.

## Why these matter
- **Ch. 6 quantifies the thesis:** even when you try to *derive* a de minimis cancer level scientifically (from
  background variation), you get ~10⁻⁴; **10⁻⁶ only appears as the politically-set floor for whole-population
  exposures.** The number tracks population size, not biology — the same finding as Travis et al. 1987 (H24).
- **Ch. 8 shows the chemical apparatus** (potency distribution + linearity → a ppt/ppb cutoff) and concedes the
  cutoff's risk standard is a free choice ("any risk level could be chosen") — and ties it to *Monsanto*. It is
  Rulis's own earlier statement of the calculation that became 21 CFR 170.39.
- Together with ch. 11 (NCRP radiation NIRL, 10⁻⁷/yr), the volume is the **one place the chemical and radiation
  de minimis traditions are set side by side** — and all three numbers (10⁻⁴, 10⁻⁶, 10⁻⁷/yr) are presented as
  *chosen*, not derived.