The 'Chaum rule', which is that basically every interesting construct and protocol in applied cryptography goes back to an obscure typewritten manuscript by David Chaum from the early 80s, strikes again, kinda:

The idea of a distributed key generation without a trusted third party seems to first get addressed in a 1990 paper by Ingmarsson and Simmons (quickly followed up with the much more famous Pedersen paper in 91). In that '90 paper they bemoan that no one had previously taken seriously the idea of avoiding a trusted third party, *except* an '85 paper by Meadows which i can't find on the internet. They then note that Meadows

"..attributes the question of whether a shared secret scheme can be set up without the assistance of a trusted key distribution center to Chaum, however the paper of his that she cites, 'Some Open Questions' did not appear in the proceedings of Crypto '84 where she references it."