Urdu Consonants Acquisition by Children

Abstract

This study investigates the acquisition of speech segments of Urdu. The research is conducted by listening to spoken consonants by children of various age groups, between 8-30 months. It also attempts at reporting the differences and developments of consonants in a specific sequence in a specific age group and speech errors. Six children from each gender have been selected. The findings of the paper are that girls acquire some of the consonants earlier than boys. Secondly, voiced consonants develop first as compared to devoiced. Order of their acquisition is from front place of articulation to back. Stops are acquired first in manner of articulation. Whereby, aspirated consonants of Urdu language are acquired in the end. Moreover, stops tend to replace most of the consonants before their acquisition. To a large extent, the findings of this paper are similar to some other research conducted on many other languages. Although, this research has some limitations regarding methodology yet, it is very helpful for further research in Urdu language acquisition, especially in the domain of language disabilities.

Headnote

ABSTRACT

This study investigates the acquisition of speech segments of Urdu. The research is conducted by listening to spoken consonants by children of various age groups, between 8-30 months. It also attempts at reporting the differences and developments of consonants in a specific sequence in a specific age group and speech errors. Six children from each gender have been selected. The findings of the paper are that girls acquire some of the consonants earlier than boys. Secondly, voiced consonants develop first as compared to devoiced. Order of their acquisition is from front place of articulation to back. Stops are acquired first in manner of articulation. Whereby, aspirated consonants of Urdu language are acquired in the end. Moreover, stops tend to replace most of the consonants before their acquisition. To a large extent, the findings of this paper are similar to some other research conducted on many other languages. Although, this research has some limitations regarding methodology yet, it is very helpful for further research in Urdu language acquisition, especially in the domain of language disabilities.

Keywords: speech acquisition, Urdu, listen and spoken constraints, children, language difficulties.

1. INTRODUCTION

Language acquisition is one of the basic human traits. It is the process of acquiring the ability to comprehend and more important is to speak the language. The word 'acquisition', fundamentally, refers to 'First Language Acquisition'. Whereby, to develop the ability to speak second language is known as 'learning processes'. Some unique characteristics of language acquisition are that it is developed without any explicit teaching, in a limited amount of time and over varying time periods. It also depends on what is heard. A normal human being acquires the language successfully across all its elements in a proper sequence. The order is as follows: phonology, morphology, syntax, semantics and pragmatics. Moreover, all languages have more or less identical ways to be acquired across the globe.

A significant focus of the study is time span of language acquisition. Miller et al. (2009) describe the process of acquisition that a physically and mentally normal child acquires particular aspects of language in specific time period. They further describe that the speed, with which a child completes this initial period, is a noteworthy feature of language acquisition. A normal child masters this complex process with astounding speed. This speed is not halted or conditioned by socioeconomic, cultural or historical factors. Instead, any given linguistics environment, all children from all cultural background arrive at the semantics and syntax stage of language acquisition within broadly identical span of time with the same speed.

Apart from the time span, this process of acquisition is highly structured. Words, syllables and even phonemes are not uttered randomly, but rather are acquired systematically. As described by Andriana, Brooks and Hedge (2000) only certain phonemes are acquired in a specific period of time. Hence, this highly regular pattern leads to hierarchical structure which is elaborated in its place and manner of articulation of consonants. This process starts off with vowels and as children approach to combine consonants with vowels, and the syllable pattern becomes the center of attention as explained by O'Grady, Dobrovolsky and Katamba (1996).

Many of the consonants are difficult to acquire in the very first attempt. These consonants are replaced by already acquired consonants. So, the study of speech errors becomes a core focus for study, including the type of error which is made when a child produces one or more component sounds of a word with a different sound or sounds.

Furthermore, gender plays a role in the speed of language acquisition process. A general notion is that girls acquire language earlier than boys. Thus, gender becomes the significant issue of language acquisition researches.

All in all, language acquisition researches have a substantial role to play in the understanding of developmental disorders of language and specific language impairment. Understanding speech pathology, it helps in the treatment of communication disorders. It also contributes to an appreciation and explanation of exceptional language abilities. Once the language acquisition process is known, many other corridors of research are opened such as role of parental speech and interrelationship between language and nonverbal development.

2. LITERATURE REVIE

Language acquisition is a systematic process. It is acquired in the pattern of stages. Hamann (1996) in "Language Acquisition" sums up this pattern into seven stages as following:

1. From birth to 6 months (pre-linguistic stage i.e. cooing)

2. From 6-8 months (onset of babbling i.e. the manifestation of phonology)

3. From 10-12 months

4. From 12-18 months

5. From 18-24 months

6. From 24- 30 months

7. From 30-onward onset of morphology, syntax, semantic and pragmatics.

These stages mark the development of child language from single vowel and consonant to syntax and semantics.

O'Grady, Dobrovolsky, and Katamba (1996) have documented that vowels are generally acquired before consonants. As soon as child starts acquiring consonants, the development of CV (consonant vowel) syllables begins.

According to Mahler and Dupoux (1990) at around 6-8 months all children produce repetitive syllables like 'dadada'. Hamann presents two universally accepted babbling types: canonical babbling and variegated babbling. The former is the repetition of same syllable (babababa) and the later one is when in this repetition, one of the syllables is substituted by the other acquired syllable such as 'ba' by 'da' in babadaba.

Though, babbling is language specific, there are significant similarities in babblings cross-linguistically. According to O'Grady et al. (1996), in the babblings of babies from 15 languages including English, Thai, Japanese, Arabic, Hindi and Mayan frequently found consonants are [p, b, m, t, d, m, k, g, s, h, w, j] and infrequently found consonants are [f, v, t, d, s, z, 1, r, n]. Hamann in "Language Acquisition" notes that the consonants which are very frequent in the words of a language tend to be equally frequent in the babbling of a baby exposed to that language. For example labials are more frequent in French than in English and occur more often in the babbling of French children than in the babbling production of English children. Similarly dentals are found in the babbling of Japanese children.

O'Grady et al. (1996) present typical inventory for two year old (24 months) in English i.e. stops (p, b, t, d, k, g) nasals (m, n), and fricatives (f, s, h). However, Bosma and Smif s study (1993) notes that two to three years old (24-36 months) children exhibit speech errors in uttering the consonants which are not yet acquired.

A hierarchical structure is recognized in the language acquisition across the globe. In terms of the place of articulation, labials are often acquired first, followed (with some variation) by alveolar, velars, and alveo-palatals. Interdentals are acquired last as described by O Grady et al (1996).

About the sequence in manner of articulation, O'Grady et al. (1996) in Contemporary Linguistics (1996) describes that stops have tendency to occur first. These authors describe the substitution of liquid by glides. Eimas (2010) evidences that /la / and /ra/ sounds are not distinguished in production until quite late in acquisition sequence.

Furthermore, when we talk about manner articulation, voiced consonants develop first in this sequence of stops. Richtsmeier (2010) notes firstly, that sometimes children voice all initial devoiced consonants. Secondly, he explains that final voiced consonants may also be devoiced even though it is part of normal speech. For example, /d/ is changed in to /t/ in washed, baked. Smit (1979) also supports this process of devoicing. Thirdly, Richtsmeier explains that children devoice velars more than alveolar and labials.

The above mentioned process of voicing and devoicing marks the existence of speech errors. Speech errors are of interest because they provide an important metric of the child's phonological development (Peña, Flegde 2000; Smit 1979). One type of errors is speech errors is stopping. Stopping refers to the substitution of a fricative by a stop (Richtsmeier; (2010).

O' Grady, Aronoff, Miller (2009) describe three types of substitution errors:

(1) Fricatives and affricates are replaced by corresponding stops (/s/ of Sing by /t/ [tiN]. The process is called as stopping. Stopping refers to the substitution of a fricative by a stop (Richtsmeier; 2010). Cross-linguistically, Locke (1983) concludes that voiced fricatives are most likely to be stopped. The two- to three-year-old children in Smit's (1993) study exhibit pervasive stopping, but it becomes less frequent at later ages and primarily affects the fricatives /v ö/. The word-initial affricate AI3/ is also commonly stopped, and voiceless and word-final fricatives are occasionally stopped, as well. Smit's (1993) review also finds occasional but relatively rare stopping of the /w/ and /I/. According to Srivastava (1974) evidence of stopping in other languages includes studies of Hindi.

(2) Sound's place of articulation is moved forward (// / in ship is moved forward to /s/ [sip] and the process is known as fronting.

(3) Liquids are replaces by glides. Gliding of liquids is also marked by Bernhardt and Stemberger.

Finally, Xiong Xin (2010) in, a teacher at the Hunan University of Arts and Science, instigates the researcher to explore if gender makes much difference in language acquisition. His study "Language acquisition" finds that girls often have a greater internal motivation for language acquisition than boys, making language acquisition easier for girls.

Unfortunately, the researcher could not find any valuable work done on Urdu language acquisition. So, the study aims to explore all above mentioned processes in Urdu language.

3. METHODOLOGY

3.1 OBJECTIVES

* To explore the Urdu language acquisition process by children.

* To investigate the speech errors made in Urdu language acquisition.

* To study the effects of gender on acquisition process.

3.2 RESEARCH QUESTIONS

1. What is the order of acquisition process in Urdu language regarding manner and place of articulation?

2. What is the speech errors made in Urdu language?

3. Does gender affects the Urdu language acquisition?

3.3 SUBJECTS

To gather data regarding child acquisition of Urdu consonants, children between the ages of eight months to thirty months were surveyed. To check the development precisely, this age group was further divided into five subgroups: 8-10 months, 10-12 months, 12-18 months, 18-24 months and 24-30 months. Twelve children from each age group were selected.

Six children from each gender combined to form total twelve of each age group. As a result of clues from the literature review gender seems important to be considered in this kind of research in order to investigate whether gender makes difference in language acquisition or not.

3.4 DATA COLLECTION

For data collection these five groups were further grouped into two: A and B. Group A included children from 8-24 months and group B consisted of children from 24-30 months.

Data collection was based on a corpus of Urdu consonants. Children have more tendency to use CV in preference to more complex syllables by children (Hodson and Paden (1991); Ingram, (1989). So, mainly four different corpuses are selected for group A and group B i.e. two for each group. One corpus was to check the initial consonant development and the other for the final consonant acquisition, amongst each age group. Furthermore, the template for the initial consonant of Group A was CV and for the final consonant was of VC template. Similarly, to survey initial consonant template for group B was CV and CVC for final consonant. / \J and /D/, /DrV are removed from CV lists as no word is found starting from these consonants. For the same reason, aspirated [mfi, nfi] are removed from both initial and final lists. Similarly [3, v] is not included in the VC and CVC list. Please find the complete word list in the list of appendix A.

3.5 EXPERIMENTAL CONDITIONS

All children were encouraged to repeat the respective list of words after the observer. It was quite difficult to get children of Group A to repeat the words easily, so; visual aids were used to facilitate the data collection. The use of visual aids is also found in the study of "Phonological Error Distributions in the Iowa- Nebraska Articulation Norms" by Ann, 1993.

For this purpose they were presented with pictures and models. Resultantly, they tried to repeat the name of the pictures while playing with them.

The children were silently observed and were listened to carefully. Recording of the produced vocalic sounds by using electronic devices was impossible as it diverted the children's attention. Neither was it possible to record by hiding a device because the voices were then too low to record.

3.6 RESULTS

The key findings are presented in the following tables. The following points will aid in their interpretation.

* The darkest area shows fully developed consonants.

* Consonants with white space exhibit substitution by all of the children.

* Consonants in lightest shaded area present that all uttered the consonant wrong except three.

* The dark grey shade shows four are right and two are wrong.

* The light grey represents that two could not utter the consonant and those who uttered, were wrong.

* All the blank box show unuttered consonants by all.

DISCUSSION

As mentioned in the literature review, 6-8 months is the onset stage for vocal babblinglbut that continues to the age of 10 months. Delaney explains that [p, b, m, t, d, n, k, g, s, h, w, j] are frequently found in babbling of children from 15 languages. In Urdu consonant acquisition only [b, d] is found. To record the babbling of Urdu natives, children of 68 8 were just observed but not included in the tables. It is also evident in Tables a & B that until the age of 8-10 months /b/and / d/ are fully acquired. So, in the age of 6-8 months both canonical and variegated babbling is found i.e. babababa and babada or dadada and dadaba respectively as cited in literature review from Hamman (1996).

As mentioned in section 2, a two year old (24 months) English child acquires stops (p, b, t, d, k, g) nasals (m, n), and fricatives (f, s, h). In Urdu consonants acquired by 24 months are stops (p, b, t, d, and k), lateral (1) and nasals (m, n) as it is evident in Tables a & B.

There was evident from the study that children from age 24-30 months exhibit errors in acquiring the rest of the consonants.

As O'Grady explains (regarding the place of articulation) consonants are acquired in order of (with some variation) labials, alveolar, velars, and alveo-palatals (retroflex). In Urdu the same order is found for consonant acquisition. It is evident from Tables a & B that by the age of 8-10 months labial /b/ comes first in language followed by dental /d/. By the advent of 10-12 months other consonants are also included i.e. m, n. Again /m/ is labial and /n/ is alveolar. Moreover, by the age of 18-30 months acquisition includes velars /k and g/. Retroflex are acquired in the last amongst stops.

The question arises why all the labials are not acquired first and follow the sequence mentioned by O'Grady et al (1996). As investigated by Richtsmeier (2010) that children voice all initial devoiced consonants. It shows voiced consonants are acquired before devoiced. The order of their place of articulation is same as worked out by O' Grady. For example, labial voiced /b/ is acquired first and before the acquisition of labial devoiced /p/, dental /d/ is acquired and then voiceless labial /p/ shows its presence. Perhaps for this reason dental /d/ is acquired before labial /p/ as shown in tables A &B. Furthermore, as Richtsmeier (2010) describes that children devoice velars more than alveolar and labials. In Urdu it is also found that child acquisition starts by voiced labials and dentals but as this moves towards velar consonantal acquisition, voiceless /k/ is developed first.

In case of the manner of articulation, Grady et al. (2009) describe that stops occur first in language acquisition. This has been found in the study to be very much true for the Urdu language both at initial and final position with stops followed by laterals. Tables A & B also mark the evidence that affricates are acquired before fricatives as it is clear that sometimes /s/ and / J7 are replaced by / tJ7. Other than stops and lateral, all the rest of the consonants are subjected to errors before they are fully developed.

Richtsmeier (2010) and O' Grady, Aronoff, Archibald, Miller (2009) describe that fricatives are substituted by their corresponding stops i.e. voiceless fricative is substituted by its voiceless stop and so is the matter with voiced. It also accounts for Urdu consonants at initial and final position which is evident in table A and B such as /s/, / J7 by /t/ (voiceless) and /z/, / 3/ by / d/ (voiced) and so on. Similarly in affricates, it is clear in tables as / tJ7 is being replaced by /t/ and AI3/ by / d/. Conversely, it is also obvious in Tables A& B that fricative / J7 is replaced by another fricative /s/ and non- aspirated affricates /tJ7 and AI3/ are substituted by their own aspirated affricates / tp / and / d3fi/. This is evidence that stopping is frequent but not always.

The next substitution Miller et al. (2009) describes, occurs by moving forward of a sound's place of articulation. This again is represented in the Urdu language as it is clear in Tables A & B that alveolar /s/ and post alveolar / tJ7 is moved forward and is substituted by dental /t/. Similarly post alveolar / J7 and / d3/ are moved forward to be substituted by alveolar /s/ and dental /d/ respectively. In Urdu somewhere different substitution is seen by few children regarding alveolar /s/, /J7 and /z/ are being replaced by post-alveolar affricates / tJ7 and / d3/ respectively. Hence, we can conclude that in case of fricatives children recognize voice but moving forward place of articulation is not always true.

Moreover, in case of affricates, substitution by moving forward is not always true rather they are sometimes replaced by their own aspirated form instead of stops. This is also found in case of velars but voiced are devoiced as /g/ and /y/ by ñd unlike affricates where voice is recognized.

Thirdly, Miller et al. (2009) describe that laterals are replaced by glides. There is no evidence that in case of Urdu it is true. Urdu does not have /w/ but sound / j/ is present but it does not replace /l/ rather, iM is developed earlier. As Smit (1993) explores rare stopping of glide /w/, in Urdu glide /j/ is substituted by stop / d/. Moreover, as Eimas (2010) describes that children are unable to differentiate both /r/ and /l/, it is also evident in the Tables A &B as by the age of 30 months children substitute /r/ by /l/. Furthermore, place of l\l is moved forward to be substituted by alveolar /l/.

In Urdu specific consonants i.e. aspirated sounds are substituted by their corresponding non-aspirated sounds.

Succinctly, from the above discussion, we can summarize that consonants which are specific only to the Urdu language are developed later. These are velars /X/ and / y/, and all aspirated sounds. Amongst retroflex /{/ and /c|/ are acquired earlier but as compared to other stops they develop later. / \J is not developed even by the age of 30 months.

It is clear from the Tables A & B that speech errors are a significant part of language acquisition. Smit (1993) evidenced that stopping continues by the age of three years. Similar evidences has been found in this study of Urdu acquisition. In tables A & B, by the age of 30 months stops are exhibiting its presence. In Urdu, although substitution moves from stops to other consonants as soon as children develop consonants other than stops yet stops are more frequent substitution. Tables A & B clearly manifest that children always recognize voice but not place of articulation. Somewhere if they recognize place, they are unable to recognize aspiration. All in all, it is also exhibiting the fact that even by the age of 30 months children are unable to acquire all of the consonants of their native language.

Hodson and Paden, (1991) and; Ingram, (1989) describe that children tend to pronounce CV syllable first as compared to more complex syllabic forms. From a comparison of table C and D , it is marked that CV template is acquired by children earlier than VC.

Finally, the most important finding of the paper is that two genders (boys and girls) acquire language differently. As described by Xiong Xin, girls appear to possess greater internal motivation in acquisition of language than boys. Certainly, there is evidence from data presented in Tables C & D that girls develop certain consonants earlier than boys

In conclusion, we can summarize Urdu consonant acquisition as following:

* /b/ and /d/ are more frequent in babbling.

* Voiced consonants are developed first.

* In place of articulation, acquisition occurs in order of labials, dentals, alveolar, post alveolar, velars and retroflex.

* In the manner of articulation it is in order of stops, nasals, lateral, affricates and fricatives.

* Trill, flap and glides are not acquired until the age of 30 months.

* All the consonants that are specific to Urdu language are acquired last.

* Even, by the age of 30 months, all the consonants are not developed.

* Miller's et al (2009) first two rules of substitution errors (stopping and moving forward place of articulation) are also frequent in Urdu.

* Stopping is evident until the age of 30 months but as soon as affricates are acquired, stopping is interfered with substitution by affricates.

* The CV syllable is acquired before VC or CVC.

* Girls seem to acquire more consonants than boys in the same duration.

References

References

Cornelia, H. (1996). Language Acquisition. Kluwer: Dordrecht.

Differences in Language Development Available at: http://www.ehow.com/info_8080973_differences-languagedevelopment.html#ixzz2GWlaKfOJ (last accessed - 10-10-2012).

Eimas, D. (2010). Perspective on the study of speech. New Jersy: Lawrence Erlbaum Associates.

Ingram, David, (1989). Child language acquisition: Method, description and explanation. Cambridge: Cambridge University Press.

Miller, J, O'Grady, Aronoff (2009). A constraint on the discrimination of speech by infants. Language and speech. Vol. 34. P 251-163.

Mehler, J. and Dupoux, E. (1990): Acquisition of language: Infants and Adult data."Naitrehumain. Paris: Editions Odile Jacob.

O' Grady, Dobrovolsky, M, Katamba, and F. (1996), Contemprary Linguistics: an introduction to linguistics. Macmillan higher education.

Peña, B, Adriana, and M. N. Hegde. 2000. Assessment and Treatment of Articulation and Phonological Disorders in Children. PRO-ED, incorporated.

Richtsmeier, P. (2010). Child phoneme errors are not substitutions. Toronto working papers in Linguistics. Vol. 33.

Smit, B. (1979). A phonetic analysis of consonantal devoicing in children's speech. Journal of children language. Cambridge university press.

Smit, B. 1993, Phonologic error distributions in the Iowa-Nebraska Articulation Norms Project: Constant Singletons. Journal of speech and hearing Research, Volume 36.

Srivastava, G. P. (1974). A child acquisition of Hindi consonants. Indian Linguistics, 35; 112-8. P. 268.