# Excerpt — GRAS self-certification: the mechanism Frawley was actually working

*The 1958 Food Additives Amendment that forced 2-year studies on Hercules's rosins also created the off-ramp
Frawley invoked: a substance "generally recognized as safe" by qualified experts is **not** a food additive and
needs no FDA pre-approval. His "twenty-four other toxicologists … in writing to the FDA" was a GRAS
self-certification — manufacture the expert consensus, declare it safe, skip the test. Here is the modern
endpoint of that move. Sources held in `papers/lateral/`.*

## U.S. GAO (2010), GAO-10-246 — *FDA Should Strengthen Its Oversight of … GRAS* — **[CONFIRMED-primary]**
> Companies may determine a substance is GRAS … **"without FDA's approval or knowledge."**

> FDA "does not have information about other GRAS determinations companies have made because companies are
> **not required to inform FDA of them.**"

## FDA (2016), 81 FR 54960 — *Substances Generally Recognized as Safe* (final rule) — **[CONFIRMED-primary]**
The GRAS-notification procedure is **voluntary**: a company "is permitted, but is not required" to tell FDA of a
GRAS conclusion. (Notices accepted from 1998; finalized 2016. Docket FDA-1997-N-0020.)

## NRDC (2014), *Generally Recognized as Secret* (Neltner & Maffini) — **[secondary — advocacy NGO]**
> "275 chemicals … from 56 companies that appear to be marketed for use in food based on **undisclosed** GRAS
> safety determinations."

> "[An] estimate that there have been **1,000 such secret GRAS determinations**."

(Of the 56 companies: 21 "never responded to NRDC's requests" for the safety basis of 218 chemicals.)

## The through-line
Frawley invoked the GRAS **concept** in 1968 (the bureaucratic self-GRAS pathway came later — say "concept" to
avoid anachronism). The COI rhyme is exact: Frawley's "consensus" was his own employer's lobby; the modern
critique is that the consensus is the manufacturer's own staff and paid panels. The precise figures —
of **451** GRAS notifications (1997–2012): **22.4%** of safety assessments made by an additive manufacturer's
employee, **13.3%** by an employee of a consulting firm selected by the manufacturer, **64.3%** by an expert
panel selected by the manufacturer or its firm, and **0%** by an independent third party — are in **Neltner et
al., "Conflicts of Interest in Approvals of Additives to Food Determined to Be GRAS: Out of Balance," *JAMA
Intern. Med.* 173(22):2032–2036 (2013)**, DOI 10.1001/jamainternmed.2013.10559 — **NOW HELD** (operator Taildrop
2026-06-11): `papers/lateral/neltner2013_jama_gras-conflicts-of-interest.pdf`, register **H15**,
**[CONFIRMED-primary]**. Ties to
`reckoningscience/sections.md` §3/§4 (the conference) and §6 (the conflict pattern).