The Perception and Production of Arabic Lexical Stress by Learners of Arabic: A Usage-Based Account

Abstract

Studies in second language (L2) stress perception and production over the past few decades have focused on the role of the native language (L1) of the L2 learner and how it systematically influences their performance in stress perception and production. However, these studies have not adequately explored and incorporated an important factor: input frequency of stress patterns (henceforth frequency), a factor that has been widely explored in other disciplines and has been found to be crucially relevant to language processing and learning. To bridge this gap in the literature, this study examines the effect of frequency, in addition to the role played by the learners’ L1, on the perception and production of primary lexical stress in Arabic by L2 learners of Arabic in an experimental environment. To this end, a stress perception and production experiment was conducted on first- and second-year L1 English and L1 Chinese learners of Arabic as well as L1 Arabic speakers as controls. In the experiment, the participants completed a stress production task, a stress identification task, and a lexical decision task, where they were asked to produce stimuli that were nonsense words with frequency-biased stress patterns, to listen then identify the position of the stressed syllable in these frequency-biased stimuli, and to determine whether the stimuli were real Arabic words or not, respectively. The results indicate a more evident frequency effect in the stress production task, where it had a local effect on learners’ performance on stimuli. Specifically, they were significantly quicker and more accurate in producing the stimuli when the stimuli had a frequent stress pattern whereas they were slower and less accurate when the stimuli had infrequent stress patterns. In contrast, the results show a more global effect of participants’ L1 on their perception and production of stress, as typological differences were found in the performance in the stress perception and production tasks. L1 speakers of Arabic consistently had slower and less accurate performance than the L2 learners in the stress identification task. The L1 Chinese participants had systematically more fluent and accurate production than their L1 English counterparts, which is argued to be contributed by the L2 Chinese participants’ better utilization of correlates for stress. These findings are taken to be in partial support for the role of frequency in stress perception and production, as significant differences were found in contexts with larger frequency contrast but not ones with moderate-to-small frequency contrast, and the fact that the performance of the participants was, to a large extent, conditioned by the preferences for acoustic cues and prosodic characteristics of their L1. However, frequency of input should not be disregarded, since it did capture aspects of learners’ performance that were not conditioned by their L1. Future studies should build upon the method implemented in the present study to further explore the role of frequency in L2 learners in higher proficiency as well. Pedagogically, the findings from the present study provided several implications for current Arabic curricular development and teaching practices, including raising the awareness of Arabic instructors of lexical stress and the importance of lexical stress for L2 teaching, developing teaching materials, and reflecting on current curricula practices that simultaneously engage multiple varieties exhibiting different stress systems