Abstract
This study evaluates 39 different phonetic inventories of 16 Spanish‐speaking children (ages 0;11 to 5;1) in terms of hierarchical complexity. Phonetic featural differences are considered in order to evaluate the proposed implicational hierarchy of Dinnsen et al.'s phonetic inventory typology for English. The children's phonetic inventories are examined independently and in relation to one another. Five hierarchical complexity levels are proposed, similar to those of English and other languages, although with some language‐specific differences. These findings have implications for theoretical assumptions about the universality of phonetic inventory development, and for remediation of Spanish‐speaking children with phonological impairments.
Notes
Notes
1. Chomsky and Halle's (Citation1968) feature system is generally used to distinguish among phonemes and phoneme classes, thereby serving a distinctive (contrastive) function. In Dinnsen et al.'s (Citation1990) original study and in the present one, the feature system was used to describe sound classes of phonetic inventories; the phonemic contrastiveness of each sound in each child's inventory was not addressed. Indeed, a two‐time occurrence is hardly sufficient to indicate any functional role of a sound. Nevertheless, the same feature system has also been used to indicate phonetic (or surface) properties of sound classes, such as when vowels are nasalized in English. Although [±nasal] is not contrastive for English vowel phonemes, when a vowel precedes a tautosyllabic nasal, it becomes [+nasal]. In this sense, then, the [+nasal] feature, which is not a distinctive property of the language, is used to characterize a particular class of sounds when in a particular phonetic context. The feature system was used by Dinnsen et al. in this same sense for the characterization of children's phonetic inventories.
2. A fourth child discussed by Montes Giraldo is not considered here due to insufficient amounts of data for reliable determination of inventories at any given point in time.
3. To clarify, in the case of [bolθa], this would not be a correct production in any dialect of Spanish; in the case of [θiŋko], this would be a correct production in Peninsular Spanish, but not in any of the Spanish dialects considered in the present study (Hammond, Citation2001).
4. Recall that this was also the case for Cantonese‐speaking children's inventories (Stokes, Citation2002; Stokes and To, Citation2002).