Toward a triple-weak theory of phonology/phonetics

Abstract

Almost 20 years ago, Nearey proposed the Double-Weak theory of speech production and perception. He pointed to three dominant theories of speech production and perception. All begin by assuming the memory representations of language consist of phonemes made from segmental distinctive features. He called the linguistic approach “double-strong” since it assumed each phoneme has a simple set of articulatory correlates as well as acoustic correlates. The other two theories assume either that the articulatory definitions are simple and strong while the acoustics is messy. Or that acoustics plays the strong role while articulation is messy. However, Nearey's double-weak theory assumed no simple relation of either articulation or acoustics to phonemes was possible. In contrast, my theory agrees with Nearey but further argues there is no fixed, invariant set of discrete, non-overlapping phonemes for spelling words (Port, Lang Sci, 2010). This presentation will also discuss recent attempts to salvage the phoneme noting they include so many caveats and revisions that they basically cede the issue.