Why this Article is not about the Acquisition of Phonology

0. Introduction In order to understand what is going on when phonology is acquired we clearly need to understand what is being acquired. We need to understand what phonology is, and what aspects of it have to be learned. have deepened our understanding of the phonological component of the human linguistic system, its interfaces with other components and the extent and nature of cross-linguistic variation. This latter point is central to the issue of acquisition. I follow the general view that the human linguistic system (including its phonological component) consists of a fixed innate template (UG) which underdetermines a given linguistic system. This system is " fleshed out " by a vector of parameter settings which, along with UG, defines the linguistic system in question. In this view acquisition consists of parameter setting upon exposure to primary linguistic data. The computational model discussed in Dresher & Kaye, 1990 was based on these assumptions. One of the purposes of that study was to show that this model was at least plausible. At the time our understanding of the parameters of stress systems (" metrical phonology ") was the most developed and it was for that reason that we chose to model them in our study. Today much more of the observed phonological diversity has been reduced to a small set of parameters and so much more ambitious models of phonological acquisition are possible. I will take it to be an obvious point that to the extent that phonological diversity can be shown to be rather small, the acquisition model can be correspondingly simplified. One of the key moves for gaining an understanding of the acquisition of phonology (indeed of phonology itself) is the liberation of phonology from any notion of " phonetics ". Here I take " phonetics " to mean considerations involving concepts of articulatory organs, air stream mechanisms and anything else involved in the production of speech sounds by humans. Aside from placing certain anatomical limitations on how human speech sounds can be produced (e.g. we can place the tip of our tongue behind our upper teeth but not on our left earlobe) knowledge of these anatomical facts has played no role in our knowledge of phonology nor in the functioning of phonology itself. Failure to grasp this point has muddied the waters considerably and led to the general lack of theoretical progress in mainstream phonological approaches that we …