The longitudinal development of clusters in French

Abstract

A B S T R A C TStudies of English and German find that children tend to acquire word-final consonant clusters before word-initial consonant clusters. This order of acquisition is generally attributed to articulatory, frequency and/or morphological factors. This contrasts with recent experimental findings from French, where two-year-olds were better at producing word-initial than word-final clusters (Demuth & Kehoe, 2006). The purpose of the present study was to examine Frenchspeaking children's longitudinal acquisition of clusters to determine if these results replicate developmentally. Analysis of spontaneous speech productions from two French-speaking children between one and three years confirmed the earlier acquisition of initial clusters, even when sonority factors were controlled. The findings suggest that French-speaking children acquire complexity at the beginnings of words before complexity appears word-finally. The role of frequency, morphological, structural and input factors is discussed. I N T R O D U C T I O NMuch of the research on the children's acquisition of consonant clusters has focused on word-initial position in English (e.g. Templin, 1957 ; see Kirk & Demuth (2005) for review). In contrast, relatively little attention has been paid to children's developing abilities with word-final clusters. Studies that have addressed this issue report that in German and English, wordfinal clusters are typically produced before word-initial clusters (e.g.[*] We thank Harriet Jisa and other members of Dynamique du

Teachability and Learnability of English Pronunciation Features for Vietnamese-Speaking Learners

“…Studies of the L1 acquisition of English, where there is a morphological advantage for early acquisition of final clusters show that final clusters are acquired before initial clusters (Kirk and Demuth 2005). Initial clusters were, however, observed to be acquired before final clusters by French-speaking infants by Demuth and McCullough (Demuth and McCullough 2009). They attribute the fact that complexity appears at the beginning of words before at the ends of words to French not having heavy word-final morphophonology, unlike English.…”

“…by Taylor (1993) and Jenkins (2000), Pienemann's model can be applied to pronunciation, it would suggest that while some undesirable features of L1 accents can be remedied informally, others are not affected by explicit teaching Further, this approach implies that there is an order of acquisition of pronunciation features for second language learners, just as has been found for child language acquisition (Jakobson's 1942 order of acquisition steps). Examples of this are Demuth and McCullough (2009) and Kirk and Demuth (2005) working on the acquisition of clusters in French and English speaking children and Shillcock and Westermann (1997) on phonotactics, though Menn and Stoel-Gammon (1995) found inconsistencies in the order of acquisition of phonemes and allophones, and Dutch children have been found to vary in whether they acquire word-initial or word-final clusters first.…”

Input frequency and the acquisition of syllable structure in Polish

“…For example, multiple studies have found that cluster reduction patterns in child productions of onsets follow universal preferences, retaining either the least sonorous consonants (Gnanadesikan 2004;Pater 1997;Pater & Barlow 2003;Ohala 1999) or A c c e p t e d M a n u s c r i p t 9 those that form structural heads (Goad & Rose 2004). Other studies have found that development of clusters, codas, and syllabification is sensitive to universal sonority principles (Stites, Demuth & Kirk 2004;Demuth & McCullough 2009;Łukaszewicz 2006;Ohala 1999).…”

“…Of particular relevance is the fact that the vast A c c e p t e d M a n u s c r i p t 12 majority of studies on phonotactic probability have relied on segment level statistics to study segment level production or perception. However, a number of studies have found that statistics over coarser-grained representations play a role in development at higher levels of organization such as syllable structure (Vihman 1993;Levelt et al 2000;Demuth & McCullough 2009;Roark & Demuth 2000). As a result, it is difficult to estimate the utility of any particular frequency model for phonological learning more generally, and it is unclear how input statistics calculated at lower levels may interact with development at higher levels, and vice-versa.…”

“…Some prior studies focused on particular aspects of phonological development such as clusters (Kirk & Demuth 2005;Lleó & Prinz 1996), fricatives (McAllister Byun 2011), and codas (Stites, Demuth & Kirk 2004) have examined corpora smaller than 2000 words. Corpora of two to five thousand word tokens have been used to study development of basic syllable structure (Rose 2000;Salidis & Johnson 1997) and to analyze sonority effects in cluster acquisition (Demuth & McCullough 2009). Finally, larger corpora (greater than 20,000 utterances) have been used to develop more comprehensive analyses of children's phonological systems and to study variation across children (Compton & Streeter 1977;Fikkert 1994;Levelt, Schiller & Levelt 2000;Pater 1997).…”