This paper aims at studying the interchangeability among the Arabic /m/, /n/, /l/, and /r/ sounds. It presents copious data from Arabic dialects, Modern Standard Arabic (MSA) and Classical Arabic (CA). The phonological process responsible for most changes is assimilation. Most changes are conditional or explained by some sound change processes like neutralization, dissimilation, analogy, haplology, and the phenomenon of gemination. Phonological rules of most sound changes are stated and then drawn in notation. The change is illustrated in all permutations and combinations among the four except from m to l, m to r, and r to m. The study is narrowly descriptive.

    This work aims at determining some reliable acoustic cues for the Automatic Identification of Arabic dialects by geographical zone. It represents an attempt to integrate Arabic dialectology into the recent and promising field of Automatic Language Identification. We used perceptual experiments to determine a set of efficient acoustic cues for the discrimination of Western vs. Eastern Arabic dialects. Among the entire set of discriminating criteria that emerged from our perceptual studies, two were, a priori, relevant for our ultimate goal, since they occur exclusively in one area or the other. Specifically, these two criteria are vocalic distribution and the realization of distinctive vowel length contrast in spontaneous speech. Our acoustic studies reveal, on the one hand, that Western dialects develop central vocalic positions whereas their Eastern counterparts tend to prefer peripheral ones. On the other hand, we show that the ratios long vowel./short vowel are comparable for dialects in the same area. On the contrary, there are significant differences between Western and Eastern dialects, the latter attesting a higher long/short ratio. In order to evaluate the robustness of the two criteria we studied for Automatic Language Identification, we performed a set of experiments using a model based on an automatic vowel detection algorithm and statistical vowel system modeling. Results show that Western and Middle-Eastern dialects may be discriminated using spectral features combined with vowel duration cues since the rate of correct identification ranges from 70% to 90% depending on the number of cues retained.

    La configuration mélodique de la question totale à réponse oui/non obéit à un patron montant-descendant. La partie montante s'étend sur la totalité de_______te énoncé jusqu'à la fin de la syllabe pénultième. En revanche, la syllabe finale est porteuse d'un glissando négatif d'une grande ampleur. La présente étude examine par des procédés de synthèse la pertinence de cette chute mélodique finale. Les résultats dégagés confirment que cette dernière est la clé de voûte de la construction interrogative.

    The theoretical analysis of the data reveals that the differences between the predicted sequences and the data-generated sequences are mainly differences of order of structures within stages rather than difference between orders of acquisition stages. Although the developmental sequence for ASL agreement morphology is slightly different from the PT-predicted sequence, it is, nevertheless, consistent with the basic principles that relate to (a) the effect of different patterns of grammatical information exchange in the TL on language development and (b) the processing resources and processing strategies required to handle sound patterns. Overall, the patterns of language development in the context of ASL reflect the importance of the notion of feature unification and saliency, as described within Lexical Functional Grammar (LFG) in the acquisition of agreement morphology.

    This research is based on the assumption that attitudinal factors are at the origin of spelling, capitalization and punctuation errors which may result from students' indifference, which in turn, may be due a problem of first language (Ll) interference: namely Arabic interference. Moreover, the students' indifference may derive from the teachers' lack of focus on these mechanics when teaching or giving feedback in class. This may be related to the course designers' lack of focus on these elements in the handbooks used in the writing courses. It is assumed that content-based courses teachers, and even writing teachers, devote more time to the teaching and correction of content, lexical, semantic, grammatical items, and text organization than to spelling, punctuation and capitalization. The aim of this research is to highlight the importance of these discourse mechanics in EFL writing in general and academic writing more specifically.

    Cet article montre que les élèves n'utilisent pas les schèmes générés par L1 dans l'acquisition de L3 et qu'ils n'utilisent que les schèmes générés par L2. Il y a deux raisons fondamentales à cette situation. Premièrement, l'anglais est enseigné essentiellement à partir de textes écrits. L1 n'étant pas enseignée à l'école, dans la plupart des cas, les élèves n'ont pas à propos de L1 des schèmes basés sur la lecture. La deuxième raison est d'ordre psychosociologique. L1 est considérée dans un rapport de dominé à dominant vis-à-vis de L3. Ceci crée un blocage psychologique chez les élèves dans l'activation des schèmes de L1 quand ils abordent L3. Ils sont plus aptes à utiliser les schèmes de L2 considérée de même valeur que L3. La communication prend fin par la présentation de quelques suggestions pédagogiques pour l'enseignement de l'anglais.