Many scholars working in the field of language evolution interpret the Baldwin Effect (i.e. the hypothesis that learned behaviors may become inherited, thus affecting the direction of evolutionary change) as a powerful evolutionary mechanism. Baldwin's proposal, however, is highly controversial, in that the empirical support for it is far from conclusive. The aim of this article is to critically examine one of the main sources of evidence adduced in support of the Baldwin Effect, namely its alleged parity, as repeatedly assumed in Briscoe's (2000, 2002, 2003, 2005) approach to language phylogeny, with Waddington's genetic assimilation. It is argued here, however, that Baldwin's and Waddington's mechanisms are fundamentally different, and that this has important consequences for Briscoe's evolutionary model.