Pendant l'acquisition, l’enfant francophone reçoit comme input une langue orale où les frontières entre les mots sont masquées à cause de trois phénomènes de resyllabation très communs : la liaison (ex. les ours [lezuʁs]), l’élision (ex. l'avion [la.vjɔ̃]) et l’enchaînement (ex. une araignée [y.na.ʁɛ.ɲe]). Ces phénomènes sont à l'origine de l'apparition de mauvaises segmentations dans les productions des enfants à partir de l'âge de 2 ans, classifiées sous les noms de « remplacements » ou « consonnes erronées » (ex. de(s)[n]éléphants [de.ne.le.fɑ̃] pour des[z]éléphants [de.ze.le.fɑ̃]) et de « omissions » ou « non-réalisations » (u(n)[Ø]avion [ɛ̃.a.vjɔ̃] pour un[n]avion [ɛ.na.vjɔ̃]). Les études développementales (longitudinales ou transversales) observent une diminution de ces erreurs entre 4 et 5 ans. Si dans la langue orale, les phénomènes de resyllabation causent un non-alignement entre les frontières syllabiques et les frontières lexicales, dans la langue écrite, les frontières lexicales des mots sont délimitées par des marques graphiques (<les ours>, <une ambulance>, <l'éléphant>). Au moyen d'une étude longitudinale conduite sur 43 enfants francophones à partir du CP (6;3) jusqu'au CE1 (7;6), nous montrons que l’apprentissage de l'écriture et de la lecture, dénommés literacy, aide les enfants à fixer les formes lexicales et à mieux produire à l'oral et que cette variable influence aussi le traitement cognitif des séquences resyllabifiées. À partir de ces résultats, nous proposons également un modèle phonologique pour l'acquisition de la liaison après 5 ans qui tient compte des résultats phonétiques enregistrés et de l'apprentissage de la langue écrite. / Early word segmentation in French is complicated by three phenomena of re-syllabification: liaison (word1 les [le], ʽthe’ + word2 ours [uʁs], ʽbears’ → les[z]ours [le.zuʁs], ʽthe bears’), elision (word1 le [le], ʽthe’ + ours [uʁs], word2 ʽbear’→ l'ours [luʁs], ʽthe bear’) and enchainement (word1 une [yn], ʽa’ + word2 autruche [ot.ʁyʃ], ʽostrich’ → une autruche [y.not.ʁyʃ], ʽan otrich’). Theses processes of re-syllabification entail children's errors in oral productions such as replacements (le(s)[n]éléphants [le.ne.le.fɑ̃] for les[z]éléphants [le.ze.le.fɑ̃], ʽthe elephants’) and non-realizations (u(n)[Ø]ours [ɛ̃.a.vjɔ̃] for un[n]ours [ɛ̃.nuʁs], ʽa bear’). Even though the re-syllabified sequences of word1+word2 are perceived blended, in writing forms they appear separated by graphic markers, as blanks (les ours, une autruche) or apostrophes (l'ours). In our longitudinal study we evaluate the role of literacy in French word segmentation of re-syllabified sequences comparing the results of oral tasks and writing-reading tasks in the same group of children in three points according to literacy skills. Data analysis shows that oral productions of re-syllabfied sequences are influenced by literacy: the complete acquisition of written lexical forms and the strengthen reading skills help to correctly perform oral productions and phenomena of re-syllabification. We also notice that re-syllabified sequences entail a processing cost in word recognition. The thesis ends with the proposal of a phonological model that explains word boundaries acquisition after 5 years of age considering literacy as an important factor.

In spoken French, the phonological processes of liaison and resyllabification can render word and syllable boundaries ambiguous. In the case of liaison, for example, the final /n/ of the masculine indefinite article un [œ̃] is latent in isolation or before word beginning with a consonant (un stylo [œ̃.sti.lo] 'a pen'); however, when followed by a vowel-initial word the /n/ surfaces and is resyllabified as the onset of that word (un ami [œ̃.na.mi] 'a pen'). Thus, the phrases un air 'a melody' and un nerf 'a nerve' are produced with identical phonemic content and syllable boundaries [œ̃.nɛʁ]. Some research has suggested that speakers of French give listeners acoustic cues to word boundaries by varying the duration of consonants that surface in liaison environments relative to consonant produced word-initially. Production studies (e.g. Wauquier-Gravelines 1996; Spinelli et al. 2003) have demonstrated that liaison consonants (e.g. /n/ in un air) are significantly shorter than the same consonant in initial position (e.g. /n/ in un nerf). Studies on the perception of spoken French have suggested that listeners exploit these durational differences in the segmentation of running speech (e.g. Gaskell et al. 2002; Spinelli et al. 2003), though no study to date has tested this hypothesis directly. The current study employs a direct test of the exploitation of duration as a segmentation cue by manipulating this single acoustic factor while holding all other factors in the signal constant. Thirty-six native speakers of French and 54 adult learners of French as a second language (L2) were tested on both an AX discrimination task and a forced-choice identification task which employed stimuli in which the durations of pivotal consonants (e.g. /n/ in [œ̃.nɛʁ]) were instrumentally shortened and lengthened. The results suggest that duration alone can indeed modulate the lexical interpretation of ambiguous sequences in spoken French. Shortened stimuli elicited a significantly larger proportion of vowel-initial (liaison) responses, while lengthened stimuli elicited a significantly larger proportion of consonant-initial responses, indicating that both native and (advanced) non-native speakers are indeed sensitive to this acoustic cue. These results add to a growing body of work demonstrating that listeners use extremely fined-grained acoustic detail to modulate lexical access (e.g. Salverda et al. 2003; Shatzman & McQueen 2006). In addition, the current results have manifest ramifications for study of the upper limits of L2 acquisition and the plasticity of the adult perceptual system in that several advanced learners of French showed evidence nativelike perceptual sensitivity to non-contrastive phonological variation.