ABSTRACT
This chapter reconciles two opposing views in describing phonetic growth. The stochastic modeling approach of cognitive science does not preclude the view that infants impose meaning upon sounds on the basis of innate features or biases. Rather, ‘baby taking statistics’ is a prerequisite to baby imposing rules. Ontogenesis recapitulates phylogenesis: Children’s pattern-detecting pathways mature earlier than pathways that impose non-probabilistic hierarchical rules onto pattern detection. Perception of speech sounds changes to apperception (hearing what fits the rules rather than registering every nuance). The production of phonemes and allophones likewise shows an interaction of patterns (e.g. frequency) and rules (e.g. for substitutions and combinations). Children’s books can offer enriched soundscapes for perception with noisy books, sound play, rhyme, and even violations (such as zlock or thneed by Dr. Seuss). They can also cater to children’s pronunciations by curtailing their sound inventories age-appropriately, or else can challenge young speakers with tongue twisters. The chapter concludes with a warning that phonics books can pitch “Standard” orthoepy against how children actually align their pronunciations with their speech communities; instead, it places a premium on surprise (rather than drilling and conditioning) in triggering linguistic curiosity and exploration.
