Hopkins reports that Perrow in 1994 had 'no difficulty' with Sagan's list of NAT mechanisms, which includes the inevitability proposition

charles-perrownormal-accident-theoryinevitabilitysaganhigh-reliability-theoryhopkins

Hopkins lists the seven "causal mechanisms" of Normal Accident Theory (NAT) from Scott Sagan's 1993 The Limits of Safety, the first being that accidents are inevitable in complex, tightly coupled systems, and reports that "Perrow (1994) has no difficulty with this characterisation of his work." On that account Perrow endorsed the inevitability proposition in 1994. That is a secondhand attestation of the wording claim-collins-2017-review-leaves-perrow-inevitability-wording-unconfirmed could not confirm from the 1984 book, and it is Hopkins's report, not Perrow's verbatim wording.

Hopkins ties the 1984 inevitability thesis to a centralisation/decentralisation contradiction: "This conflict makes it impossible to devise an authority style which will reduce the risk of accidents in complex, tightly coupled systems. This explains just why it is that accidents are inevitable in complex, tightly coupled systems." HRT's "nested authority structures" (La Porte & Consolini 1991, pp. 31–32) contest exactly that step; Hopkins: "It follows that there is no theoretical reason why accidents in complex, tightly coupled systems should be inevitable." This is the mechanism-level disagreement underneath claim-la-porte-1996-hrt-accepts-perrow-preconditions-denies-inevitability.

Caveats: Hopkins's page cites to the 1984 book (p. 70 definition, p. 97 four-cell table, p. 175 marine 5–10% figure) are secondary, and the figure is [unverified-quant — needs primary]. The primary text remains blocked (observation-perrow-1984-primary-text-blocked-across-every-route-attempted); see claim-perrow-1984-normal-accident-theory-tight-coupling-complexity.