A Review on Cross-Language Perception of Speakers from Different Language Families

Citation: Jamilu ABDULLAHI, PhD, Abdulmumini ISAH, PhD and Abubakar ZAKARI (2022). A Review on Cross-Language Perception of Speakers from Different Language Families. Yobe Journal of Language, Literature and Culture (YOJOLLAC), Vol. 10, Number 1. Department of African Languages and Linguistics, Yobe State University, Damaturu, Nigeria. ISSN 2449-0660

A REVIEW ON CROSS-LANGUAGE PERCEPTION OF SPEAKERS FROM DIFFERENT LANGUAGE FAMILIES

By

Jamilu ABDULLAHI, PhD

Abdulmumini ISAH, PhD

Abubakar ZAKARI

Abstract

The difficulties in the perception and production of non-native contrasts was due to dissimilarities or similarities between the first language (L1) and the second language (non-native language) phonological systems (Flege, 1995). The native language (L1) phonemes are likely to hinder the creation of new non-native (L2) phonological categories. Adult listeners of second language (L2) have significant difficulties in the perception of most phonetic dissimilarity that are not functional (non-existent) in their native languages (L1) Best et al, 1988; Flege, 1995). The aim of this review is to find out the level of discrimination of non-native phonemes across language families. This research reviewed fourteen (14) journal articles to examine the discrimination of native and non-native perception of speech sounds by speakers from different language backgrounds across the different language families. The result disclosed that Japanese speakers/listeners of Austronesian language family discriminate most and accurate than any other members of the same language family. The result shows that the discrimination by English native listeners of Indo-European language family was accurate, and the most among the other members. The results of this review equally reveals that Japanese, Thai, Cantonese and English speakers discriminate most of the non-native sounds when compared to other language family members used in this review.

Introduction

Early cross-language study of the discrimination by native speakers (L1) of vowel continua of American English and Swedish revealed a dissimilar perception pattern (Stevens et al. (1969). In the 1980s and early 1990s, theoretical debates about the nature of the mechanisms involved in the perception of speech continued to be pronounced from the three major perspective points of view: the (revised) theoretic motor account, the “general auditory” or “psychoacoustic” account, and the direct realist account (Best, 1995). In the 1980s and 1990s, cross-language studies of an expanded set of phonetic distinctions disclosed that the adult listeners’ degree of difficulty experienced in perceiving non-native (L2) phonemes differed over a considerable range. Werker & Tees (1983, 1984) stated that English speaking subjects have difficulty in the discrimination of a velar-uvular place different in voiceless ejective stops [k-q] of Nthlakapmx (Salish).

Second language (L2) speech perception and studies on cross-language speech perception mainly focuses on the discrimination of non-native contrasts as well as the identification of non-native dissimilarity that are missing (absent) in the listener’s native language (L1). Cross-language perception investigates how native languages experience influences how a non-native (L2) phoneme a listener identifies or maps the L2 sounds to their native language (Flege, 1995).

Cross-Language Study and Speech Perception

Perception refers to how stimuli are processed, sounds are heard, and how a concept about them is formed in the mind, consciously or not (Ellis, 1994). Second language (L2) speech perception and studies on cross-language speech perception mainly focuses on the discrimination of non-native contrasts as well as the identification of non-native dissimilarity that are missing (absent) in the listener’s native language (L1). Cross-language perception investigates how native languages experience influences how a non-native (L2) phoneme a listener identifies or maps the L2 sounds to their native language (Flege, 1995). In cross-language speech perception, it is not all about recognizing sounds correctly, but how comparable of different associate non-native (L2) phonemes with native sounds (L1) irrespective of categories. In a study carried out by Shimizu (1996) on Korean adults’ perception of consonants of English, the Korean listeners were given English phonemes and mixed together with Korean sounds (given as option) to identify.

The main aim of this review is to examine the level of discrimination of native and non-native perception of speech sounds by different speakers from different language background across the different language families, especially by reviewing the related research done in the subject matter from 1999 to date.

Reasons for Cross-Language Research

Understanding the processes of perception of speech, research in cross-language perception is a critical testing ground. Universal phenomena of speech perception can only be discovered through explorations across different language families and different languages. Cross-language research reduces our locality and insularity and lets us find out whether given perceptual experience are languages specific, or the speech processing of human beings in general is true. Studies of perceptual and that of production training disagree that units at a level phonetic or at a level of allophonic may be psychologically applicable. Other cases such as voicing, however, at the level of distinctive feature may be the basis of overview.

Research in cross-language has revitalized our studies of the natural change of speech perception in the first language (L1). Speech perception research on toddlers’ attained new importance and curiosity when it was shown that, not only that toddlers were sensitive to almost all phoneme contrasts that were examined, but that during the early first year of life, toddlers began to tune out, or pay no attention to the divisions in the languages around them that were not functional.

At the level of the adult learner of non-native language, there is an opportunity to study perceptual learning of significant and environmental valid sort. This contributes to our knowledge concerning cross-language training and more theories generally concerning learning of perception itself (Jenkins, 1979).

Language Family

A language family can be defined as a related group of languages from common antecedent which is referred as proto-language from that family. Comparative linguists establish language family membership. The languages that have shared genetic relationship or genealogical relationship, those languages are related. Membership in a family language group shared common features. Germanic languages for example share grammatical features and vocabularies as well as the sounds that are absent in the Pro-Indo-European languages. The language families are: Afro-Asiatic, Indo-European, and Austronesia (Steinhauer, 2013). Languages under Indo-European language family are: English, Swedish, German, Spanish, and Russian. Languages under Austronesian language family are: Thai, Cantonese, Korean and Vietnamese, Chinese, Japanese, Mandarine, and Taiwanese. Zulu is the only African language appeared in this review.

Methodology of the Review

In the process of conducting this review, an electronic database search was first executed in Science Direct Scholarly Journal database as well as Taylor & Francis, and Google Scholar. These databases were nominated to search for related journal articles for this review, as these databases provide wide-range of information from different sources across the globe. Taylor & Francis and Science Direct were among the databases subscribed by the University Putra Malaysia to provide adequate resources and enhance learning for the students. Google Scholar was also used because it is free to access everywhere without any subscription and it indexes effective literature across all range of disciplines. This research was also limited to a period of fifteen years back ranging from 1999-2015 in choosing the related literature for the purpose of reviewing. Cross-language, perception, non-native and native language, etc, were among the keywords used in searching for the related journal articles done on the subject matter. From Taylor & Francis, twenty-four (24) related journal articles were generated by the database, seven were selected among, and in Science Direct database, four (4) were selected among the eleven (11) articles generated by the database. Finally, three (3) journal articles were selected out of nine (9) generated by Google Scholar. Those articles were selected based on their relevance to the present study and reviewed out of the generated ones by the databases. The fourteen (14) articles selected were studied thoroughly and analyzed based on their themes. Discrimination of non-native phoneme was the first theme identified, as it is concern with how native speakers discriminate non-native phoneme. The second theme was concerned with perception of non-native sounds, and the third one was on speakers from diverse language families and their level of perception of non-native phonemes.

Native and Non-native Speakers of a Language

The first language an individual learns to speak is said to be a native language (Bloomfield, 1933). Native speakers of a language can be identified based on the language one knows best, acquires first, or uses most. Non-native speakers of a language are those who have additional native tongue other than their native language (mother tongue). Research on cross-language perception and production often compares the native (L1) and non-native (L2) speech. For instance, a research carried out by Tsukada et al (2007) explored how native Japanese, Cantonese, Korean and Vietnamese speakers discriminate non-native English /p/, /t/, /k/ sounds at final stage. Similarly, Schertz, et al (2015) examined the production and perception by Korean native speakers of stop contrasts in their native (L1) language and English (L2). Tsukada (2006) studied the perceptual ability of word-final stop by Thai English (TE) bilinguals and Australian English (AE) listeners in their native and non-native languages.

Discrimination of Non-native Sounds by Speakers from Different Language Families

Second language (L2) speech perception and studies on cross-language speech perception mainly focuses on the discrimination of non-native contrasts as well as the identification of non-native dissimilarity that are missing (absent) in the listener’s native language (L1).

Cross-language speech perception is not all about recognizing sounds correctly, but how rather comparable of different associate non-native (L2) phonemes are compared with native sounds (L1) irrespective of categories. In a similar study on perception by Tsukada, et al (2014), the perception of vowel length of Japanese by comparing it with four groups of listeners (American English, Italian, Japanese, and Thai). The results of the study show that discrimination by Japanese was more accurate than the other three non-native groups. Italian (Indo-European family) discrimination was not accurate. In discrimination accuracy, the three non-native groups did not differ despite the differences with contrasts in their native languages. This shows that Japanese (Japonic member of Astronesian family member) discriminate more accurately than Italian (Indo-European languages).

Another research explored the differences in the perception of lexical stress in Spanish by Italian and French native speakers. Perception by Italians was accurate as well as the discrimination. Francophone was less when compared with that of Italians (Alfano et al, 2010). This signifies that Italians discrimination of non-native sounds was accurate when compared to those other languages which hailed from same language family (Indo-European languages). Best & McRoberts (2003) examined the predictions of several theoretical accounts of developmental change in perception of infants’ non-native consonant contrasts. The result revealed that the discrimination was quite different in Zulu by American adult listeners, which the high level of accuracy of discrimination goes to Zulu (Bantu-African language). Tsukada, (2006) reported the perceptual ability of word-final stop by Australian English and Thai English bilingual listeners in their native and non-native languages. The two groups discriminate Thai /p/-/t/ most, particularly Australian English listeners. Listeners’ accuracy of discrimination influenced distinctively by type of stops they heard. The finding shows that first languages transfer alone cannot be sufficient to account for learners’ patterns of response in cross-language speech perception. This revealed that the discrimination by Australian English listeners (Pama-Nyungan family) was more accurate.

Ylinen et al (2005) explored the perceptual abilities between the second language (L2) Russian users of Finnish, and Russian L1 speakers. Discrimination was accurate by Russian L2 users of Finnish (Indo-European family) when compared to that of non-Finish-speaking Russians (Uralic member). Tsukada et al (2007) examined how native speakers of Japanese, Cantonese, Korean and Vietnamese discriminate English (p-k, p-t, t-k) at final stage. The results reveal that despite all the speakers are from the same language family, Japanese were able to discriminate the sounds accurately, but not Thai. Cantonese, Korean and Vietnamese listeners have experience with final unrealized stops in the native languages. Tsukada (2005) studied word-final contrasts discrimination of stops (p-k, p-t, t-k) of Thai and English language by Australian English, Thai and Japanese. The outcome shows that Japanese listeners were able to discriminate accurately. Thai listeners were not able to discriminate accurately as Japan. Australian English listeners despite their experience with unreleased stops in native language did not match the Thai listeners to discriminate the stops. So & Attina (2013) explored the effect of native language (L1) background on listeners perception of non-native (L2) vowels by Hong Kong Cantonese and Mandarine and Australian English. The findings of the study revealed that discrimination of Cantonese talkers’ vowels was easier. English listeners identification and that of Mandarin was similar to those of Cantonese, suggesting that English listeners have assimilated the Cantonese vowels as closely related to their native language vowels.

Werker & Tees (2002) explored the perception of a non-English (Salish) speech dissimilarity by adult Salish and English infants. The result shows that infant discriminate non-native (Salish) speech contrasts without any relevant experience of the non-native phonemes. Sun & Huang (2012) examined the tone perception by Taiwanese Min speakers and those of American English with AX discrimination task. Tone discrimination by Taiwanese listeners was more accurate than that of English. The results in lexical tone perception between tone and non-tone languages indicate qualitative differences. Willerman & Kuhl (1999) explored the linguistics experience effect on adult’s vowels by speakers of Swedish English and Spanish speakers. Discrimination of stimuli was observed across speakers of different languages. Swedish discrimination was more accurate than that of English and Spanish despite the fact that the languages are from the same language family. Linguistic experience plays a significant role in discrimination of vowels.

Conclusion

As mentioned earlier, the aim of this review is to find out the level of discrimination of non-native phonemes across language families. Based on the above discussion on discrimination by native (L1) and non-native (L2) speakers/learners of a language, it reveals that discrimination among speakers of different language families differs. The language families reviewed are: Afro-Asiatic, Indo-European, and Austronesia. Languages under Indo-European language family are: English, Swedish, German, Spanish, and Russian. Languages under Austronesian language family are: Thai, Cantonese, Korean and Vietnamese, Chinese, Japanese, Mandarin, and Taiwanese. Zulu is the only African language appeared in this review. The results of Austronesian language family, reveals that the Japanese speakers/listeners discriminate most and accurate than any other members of the same language family. But for the Indo-European language family members the result shows that English native listeners discrimination was accurate, and the most among the other members. Finally for African languages, it is observed that discrimination by American natives was quite impressive than that of Zulu speakers.

In conclusion, the findings of this review reveal that Japanese, Thai, Cantonese and English speakers discriminate most of the non-native sounds when compared to other language family members used in this review.

References:

Alfano, I. et al (2010). Cross-language speech perception: lexical stress in Spanish with Italian and Francophone subjects. La dimensione temporale del parlato. Atti del 5o convegno nazionale AISV - Associazione Italiana di Scienze della Voce. Università di Zurigo. 4-6 febbraio 2009. (pp. 455-74). Torriana: EDK Editore.

Best, C & McRoberts, G. W. (2003). Infant perception of non-native consonant contrasts in different ways. Language and Speech, 46 (2-3) 183-216.

Best, C. (1995). A direct realist view of cross-language speech. In W. Strange (ed.), Speech perception and linguistic experience: Issues in cross-language research, pp. 171-204. Timonium, MD: York Press.

Best, et al. (1988). The phonological basis of perceptual loss for non-native contrasts: Maintenance of discrimination among Zulu clicks by English-speaking adults and infants. Journal of Experimental Psychology: Human Perception and Performance 14:345-60.

Bloomfield, L. (1933). Language. Holt Rinehart Winston.

Ellis, R. (1994). Language transfer. In R. Ellis (Ed.) The study of second language acquisition (pp. 299-345). Oxford: University Press.

Flege, J. (1995). Second-language Speech learning: Findings and problems. In Strange, W. (Ed.), Speech Perception and Linguistic Experience: Issues in Cross-Language Research (pp. 233-277). Timonium, MD: York Press.

Freida, E. M. et al (1999). Adult’s perception of native and non-native vowels: Implications for the perceptual magnet effects. Perception and Psychophysics 61 (3) 561-577

Schertz, J. et al (2015). Individual differences in phonetic cue use in production and perception of a non-native sound contrast. Journal of Phonetics 52. 183–204.

Shimizu, K. (1996). A cross-language study of voicing contrasts of stop consonants in Asian languages. Tokyo: Seibido Publishing.

So, C. K. and Attina, V. (2013). Cross-language perception of Cantonese vowel spoken by native and non-natoive speakers. J. Psycholinguistics Res 43: 611-630

Steinhauer, M. H. (2013). Colonial history and language policy in insular Southeast Asia and Madagascar. In The Austronesian languages of Asia and Madagascar (pp. 87-108). Routledge.

Stevens, K. N. et al (1969). Cross-language study of vowel perception. Language and Speech 12: 1-23

Strange, W. (Ed.). (1995). Speech perception and linguistic experience: Issues in cross-language research. York Press.

Sun, K. and Huang, T. (2012). A cross-linguistic study of Taiwanese tone perception by Taiwanese and English listeners. J East Asian Linguist 21:305-327

Tsukada, K. (2005). Cross-language speech perception of final stops by Australian English, Japanese and Thai listeners. Proceedings of ISCA workshop on plasticity in speech perception (PSP 2005); London, UK. 15-17 June, 2005

Tsukada, K. (2006). Cross-language perception of word-final stops in Thai and English. Bilingualish: Language and Cognition 9 (3) 309-318. Cambridge University Press.

Tsukada, K. et al (2007). Cross-language perception of word-final stops: Comparison of Cantonese, Japanese, Korean and Vietnamese listeners. International Congress of Phonetic Sciences (16^th: 6-10 August, 2007, Germany

Welker, J. F. and Tees, R. C. (2002). Cross-language speech perception: Evidence for perceptual reorganization during the first year of life. Infant Behavior and Development. 25 121-133.

Werker, J. and Logan, J. (1985). Cross-language evidence for three factors in speech perception. Perception and Psychophysics 37, 35-44.

Werker, J. F. and Tees, R. C. (1983). Developmental changes across childhood in the perception of non-native speech sounds. Canadian Journal of Psychology 37: 278-86.

Werker, J. F. and Tees, R. C. (1984). Phonemic and phonetic factors in adult cross-language speech-perception. Journal of Acoustic Society of America 75: 1866-78

Willerman, R. and Kuhl, P. K. (1999). Cross-language speech perception: Swediah, English, and Spanish speakers’ perception of front rounded vowels. Journal of Experimental Psychology, 442-445.

Ylinen, S. (2005). The Perception of Phonological Quantity based on Durational Cues by Native Speakers, Second-language Users and Non speakers of Finnish. Language and Speech, 48 (3) 313-338.

APPENDIX

SUMMARY OF STUDIES ON CROSS-LANGUAGE PERCEPTION

AUTHOR	OBJECTIVE	DATA COLLECTION METHOD	DATA ANALYSIS METHOD	RESULT
Tsukada, K. et al (2007)	The study explored how native speakers of Japanese, Cantonese, Korean and Vietnamese discriminate English (p-k, p-t, t-k) at final stage.	Taylor & Francis	Experiment	Japanese were able to discriminate the sounds accurately, but not Thai. Cantonese, Korean and Vietnamese listeners have experience with final unrealized stops in the native languages.
Tsukada, K. (2005)	This research examines word-final contrasts discrimination of stops (p-k, p-t, t-k) of Thai and English language by Australian English, Thai and Japanese.	Science Direct	Experiment	Japanese listeners were able to discriminate accurately. Thai listeners were not able to discriminate accurately as Japan. Australian English listeners despite their experience with unreleased stops in native language did not match the Thai listeners to discriminate the stops.
Tsukada, K. et al (2014)	This study explores the perception of vowel length of Japanese by comparing it with four groups of listeners (American English, Italian, Japanese, and Thai).	Taylor & Francis	Experiment	Discrimination by Japanese was more accurate than the other three non-native groups. Italian discrimination was not accurate. In discrimination accuracy, the three non-native groups did not differ despite the differences with contrasts in their native languages.
Connie, K and Best, K. (2010)	This research examines the lexical tones perception by native Mandarine listeners group.	Science Direct	Experiment	All the groups, that is native Mandarine, Hong Kong, English listeners group and Japanese showed perceptual confusions with similar features. When compared with the other groups, listeners of English group sensitivity to tone identification was significantly lower.
So, C. & Attina, V. (2013)	This research explored the effect of native language (L1) background on listeners’ perception of non-native (L2) vowels by Hong Kong Cantonese and Mandarine and Australian English.	Taylor & Francis	Experiment	Discrimination of Cantonese talkers’ vowels was easier. English listeners identification and that of Mandarin was similar to those of Cantonese.
Tsukada, K. (2006)	This study determines the perceptual ability of word-final stop by Thai English (TE) bilinguals and Australian English (AE) listeners in their native and non-native languages.	Taylor & Francis	Experiment	The two groups discriminate Thai /p/-/t/ most. Listeners discrimination accuracy was influenced differentially by stop type consonants they heard.
Werker, J. and Tees, R. (2002)	This explored the perception of a non-English (Salish) speech dissimilarity by adult Salish and English infants.	Google Scholar	Experiment	Infant discriminate non-native speech contrasts without any relevant experience of the non-native phonemes.
Alfano, I. et al (2010)	This research explored the differences in the perception of lexical stress in Spanish by Italian and French native speakers.	Taylor & Francis	Experiment	Perception by Italians was accurate as well as the discrimination. Francophone was less when compared with that of Italians.
Freida, E. M. et al (1999)	This study was conducted to evaluate magnet effect by American English speakers ability to perceive native and non-native vowels that are in a diverse dialects.	Taylor & Francis	Experiment	Discrimination of stimuli close to subject was poor. Better discrimination was for variants in each condition.
Sun, K. and Huang, T. (2012)	This study examined the tone perception by Taiwanese Min speakers and those of American English with AX discrimination task.	Science Direct	Experiment	Tone discrimination by Taiwanese listeners was more accurate than that of English. The results in lexical tone perception between tone and nontone languages indicate qualitative differences.
Best, C. and McRoberts, G. (2003)	This study examined the predictions of several theoretical accounts of developmental change in perception of infants’ non-native consonant contrasts.	Google Scholar	Experiment	Discrimination was quite different in Zulu by American adult listeners.
Ylinen, S. et al (2005)	This study explored second language users of Russian and native Russian perception of phonological quantity on durational cues by native (L1) speakers.	Science Direct	Experiment	Discrimination result shows that native speakers disclose a phoneme boundary effect for duration of cued only for quantity.
Willerman, R. and Kuhl, P. (1999).	This paper examines the linguistics experience effect on adult’s vowels by speakers of Swedish English and Spanish speakers.	Taylor & Francis	Experiment	Discrimination of stimuli was observed across speakers of different languages. Linguistic experience plays a significant role in discrimination of vowels.