Pinker and Prince vs. Rumelhart and McClelland

Steven Pinker and Jacques Mehler (1988:1) from the introduction to the special issue of Cognition (pp. 1-2) on Connections and Symbols

During the past two years there has been more discussion of the foundations of cognitive science than in the 25 years preceding. The impetus for this reexamination has been a new approach to studying the mind, called “Connectionism”, “Parallel Distributed Processing”, or “Neural Networks”. The assumptions behind this approach differ in substantial ways from the “central dogma” of cognitive science, that intelligence is the result of the manipulation of structured symbolic expressions. Instead, connectionists suggest that intelligence is to be understood as the result of the transmission of activation levels in large networks of densely interconnected simple units.

Rumelhart and McClelland (1986) On Learning the Past Tenses of English Verbs (NB many of the papers from this volume are available here).

Pinker and Prince (1988) On Language and Connectionism:Analysis of a Parallel Distributed Processing Model of Language Acquisition (the actual Cognition paper is available on Pinker’s page)

Seidenberg and Plaut’s overview of the past tense debate, from a festschrift for Jay McClelland. Pinker (2006) provides an overview from the other side, in a festschrift for Alan Prince.

Rogers and McClelland’s introduction to the PDP at 25 volume includes some discussion of the past tense debate, and provides the broader context. Section 5.2 and 5.3 are particularly useful on questions about integrative vs. purely associative models.

Steven Pinker on his collaboration with Alan Prince, including extremely interesting archival materials. It contains this passage (cf. p. 218 of Rumelhart and McClelland 1986):

McClelland and his collaborator David Rumelhart had claimed that their model made generative linguistics, of the kind practiced by Alan, obsolete.

From Marcus et al. (1992:2):