عنوان مقاله [English]
Background and aim: One of the main factors that differentiate child speech from adults’ is the phonological processes that happen in child language, including voicing and devoicing. The present study aims to consider and compare the quality of voicing/devoicing errors occurring in Farsi-speaking children with protracted and typical phonological development. Moreover, this study compares voicing/devoicing errors of typical Farsi-speaking children with the voicing/devoicing errors of children speaking other languages.
Method: Five children with typical phonological development (2;6-4;0) and 5 children with protracted phonological development (4;6-6) participated in this qualitative cross-sectional study. The typical children were collected from nursery schools and the protracted children were recruited from speech therapy clinics in Karaj. They were chosen via convenience sampling method. Their voicing/devoicing errors were investigated using Naming-picture task (Farsi phonology test with pictures) wherein each child asked to produce 133 single words.
Results: Devoicing errors were distributed more widely than voicing errors in typical and atypical phonological development, especially in word-final position. The typical children participating in this study rarely displayed word-medial devoicing; however, atypical children illustrated several cases of it. Word-initial and medial voicing were observed in both groups of children; however, word-final voicing was merely detected in atypical children.
Conclusion: Results revealed that the differences observed in voicing/devoicing processes between children with protracted and typical phonological development is due to the different hierarchy of markedness and faithfulness constrains in their internalized grammar. Furthermore, there are major similarities in voicing process between typical Farsi-speaking children and typical children speaking other languages.
The notion of systematicity in erroneous productions, which, at first, was presented to the field of linguistics by Stampe (1969), was later adapted and named phonological processes by the authors like Ingram (1976). Voicing and Devoicing are two known phonological processes which are from substitution category in which certain sounds or sound classes that exist in the target phonological system do not appear in the children’s system or may merely appear in certain contexts. This more restricted inventory results in some specific substitution patterns including voicing and devoicing. According to Grunwell (1985) and Stole-Gammon and Dunn (1985), the study of phonological processes highlights the correlation between the adult and child production, and presents a framework to describe patterns of typical and protracted phonological acquisition. In view of that, the present study aims to consider Voicing and Devoicing phonological processes in Farsi-speaking children with typical phonological development (TD) and protracted phonological development (PD) that have not been studied so far. To examine the errors produced through voicing/devoicing processes by the above mentioned groups of Persian-speaking children, this study, first, considers the types of voicing/devoicing errors found in Persian children with typical and protracted phonological development and then compares the error patterns found in the two groups with each other. Moreover, the results will be examined within the framework of Optimality Theory (OT) (Prince and Smolensky, 1993; McCarty and Prince, 1994, 1995) to infer the underlying internal grammar that creates such voicing/devoicing patterns in children with typical and protracted phonological development. The results of this study also will be compared with the results found in the children speaking languages other than Farsi wherever it is applicable.
The participants in this cross-sectional study are five children diagnosed with functional (non-organic) phonological disorder (PD) (Se, Ti, Me, Ze, and He) who range in age from 4:6 years to 6 years, and five children with typical phonological development (TD) (El, Al, Sa, Ma, Ro) ranging in age from 2:6 years to 4 years. All the children were selected after considering their physical and mental health and after receiving written consent from their parents. In order to collect the necessary data from Farsi speaking children, a specific production test, called, Naming-Picture Task (NPT) has been applied.The words used in this test were selected through four criteria: a) all of them are related to the objects and animals that children generally are familiar with; b) they are all concrete names and let the researcher present them in pictures to the participants; c) the selected target words cover all phonemes and syllables in Farsi; d) It has also been attempted not to make the test too lengthy because it may make the child tired and reduces the accuracy of the produced data. Using these criteria and the results of a pilot study, 132 words were elected for NPT. The pictures used to represent the words in NPT were mainly provided from a book named “painted children dictionary” (Rotako, 2004). The data was collected by NPT from both groups of children during two-three separate sessions, depending on the age of the child and his/her cooperation in answering the questions. All the answers were recorded by means of a high quality solid state sound recorder (Samsung Voice Recorder model YP-VP1).
Results showed that both voicing and devoicing errors occurred in TD and PD children; however, the occurrence of devoicing errors was more than voicing errors in both groups. Furthermore, results showed that the place of target segments in the words, i.e. word- initial, medial, and final positions, had an effect on the quality and quantity of voicing/devoicing errors. One of the most frequent cases of devoicing was word-final devoicing that was observed in both PD and TD groups. Devoicing was also detected in word-initial position in both groups; however, these processes had a higher frequency in PD compared to TD children. Voicing errors occurred in early stages of the phonological development in typically developing children and the frequency of these errors decreases from the youngest child (El, 2;9) to the eldest child (Ro 4;0). However, these errors were observed in all PD children irrespective of their age and though they are all elder than the oldest TD child. This study also examined the number of voicing/devoicing errors in codas and onsets. Though, onsets are not always in the beginning of words and codas in the end of words, but the findings related to these places have been similar to the results observed in segments that have been in word-initial and word-final positions.
The data collected from the present study investigating voicing/devoicing processes in Farsi-speaking children with typical (TD) and protracted phonological development (PD) was primarily examined in OT frame work and it is concluded that the differences observed in voicing/devoicing processes between PD and TD groups is rooted in the differences in the hierarchy of markedness and faithfulness constrains in their internalized grammar. Moreover, the comparison of the results of this study with the results of similar studies on children speaking languages other than Farsi illustrated similarities in voicing process between them. Similar to children speaking other languages, the TD and PD children demonstrated more devoicing errors than voicing errors which supports the claim that voiceless segments serve as defaults for most children. In addition, this study illustrated that there is a general tendency for voicing in onsets and devoicing in codas which supports the onset-coda asymmetry suggested by Dinnsen (1996). The age-wise decrease in the frequency of voicing errors found in the TD children’s data confirmed the previous finding that voicing errors normally occur in early stages of phonological development and by 3;6 years most children are correctly producing the voicing contrast (Richtsmeier, 2010). However, the existence of voicing errors in PD children, irrespective of their age, displayed that the inverse relation between the frequency of voicing errors and age does not probably apply to atypical phonological development.