Elsevier

Journal of Phonetics

Volume 88, September 2021, 101084
Journal of Phonetics

Research Article
Sociophonetic variation in English /l/ in the child-directed speech of English-Malay bilinguals

https://doi.org/10.1016/j.wocn.2021.101084Get rights and content

Highlights

  • Phonetic variation of English coda /l/ in spontaneous child-directed speech was examined.

  • Variants of /l/ used and darkness of /l/ varied according to situational context.

  • Less variability in /l/ was observed in fathers’ production.

  • Choices may be influenced by the socio-indexical meanings of the variants.

  • Social and cultural factors may account for differences between mothers and fathers.

Abstract

Three realisations of syllable-final /l/ have been described in previous work on Singapore English: vocalised-l (or deleted-l in some phonetic contexts; the local norms), dark-l (a form associated with the exonormative standards), and clear-l (a Malay-derived phonetic trait observed in the speech of some English-Malay bilinguals). This study examined whether, how and why Singaporean English-Malay bilinguals vary their English /l/ in their child-directed speech, and whether the phonetic variation, if any, could be socially-conditioned. The laterals in the English child-directed speech of ten father-mother dyads with their preschoolers were analysed using auditory and acoustic methods. All participants were simultaneous or early English-Malay bilinguals. The findings revealed that in informal contexts, both mothers and fathers used a relatively clearer /l/ in all syllable positions. Contrastingly, in formal contexts that involved teaching and learning, the coda laterals of mothers were significantly darker, thereby exhibiting positional contrast between onset and coda laterals. They also produced significantly more vocalised-l in these contexts. Fathers, however, did not show differentiation in the darkness of the laterals, nor did their laterals show significant positional differences in formal contexts, although some fathers of younger children did produce more vocalised-l than they did in informal contexts. The variation observed was discussed by exploring the potential socio-indexical meanings of these variants of /l/ within the context of variationist accounts of Singapore English and by drawing parallels with socially-conditioned variation in bilectal monolinguals and ethnolect speakers. Differences between maternal and paternal CDS patterns could be attributed to gender roles and cultural expectations of mothers’ dominant role in child-rearing, and may also be a result of and enabled by Malay women’s potentially more complex repertoire range.

Introduction

Adults often modify their speech when interacting with very young children. In contrast with adult-directed speech (ADS), child-directed speech (CDS) is generally characterised as having shorter, syntactically simpler utterances, with many repetitions and isolated words and phrases. Speaking rate is also reduced, and there are also more prosodic repetitions, longer pauses, and a higher average pitch and wider pitch range (see Saint-Georges et al. [2013] for a review). One of the functions of CDS is to engage the attention of the child and convey emotional affect through acoustic exaggerations (e.g. Singh, Morgan, & Best, 2002). CDS also facilitates language learning as it conveys language-specific phonological information, and caregivers enhance phonetic contrasts to provide more canonical input and reduce variability in their production (e.g. Kuhl et al., 1997, Werker et al., 2007, Cristià, 2010). Modifications in CDS may also be socially-conditioned and involve the use of alternative phonetic forms, thereby encoding indexical information (Foulkes & Hay, 2015, Nardy et al., 2013). This study examines whether, how, and why English-Malay bilingual caregivers in Singapore vary their use of variants of /l/ in their English CDS towards their preschoolers.

Segmental modifications in CDS have been found to vary with the age and gender of the child, and communicative context. Foulkes, Docherty, & Watt (2005) examined the use of standard versus other less prestigious and stigmatised local variants of (t) by mothers of children aged 2;0–4;0 living in Tyneside, England. They found that, not only did mothers in general use more standard [t] in CDS than in ADS, but more standard [t] was also used by mothers of girls and with younger children. Some evidence, however, showed that men made fewer modifications in their CDS. In their investigation into the acquisition of sociolinguistic norms in preschool children, Smith and colleagues (Smith, Durham, & Fortune, 2007, Smith, Durham, & Richards, 2013) examined the use of several variables in Buckie, Scotland in adults and children aged 2;6–4;2. One of the features studied was the lexically-conditioned hoose variable, which involves the alternation between standard diphthong [ʌʉ] and the monophthong [uː] in the MOUTH lexical set of words like house, down. The latter variant is stereotypical of Scots or northern varieties of English and used most by working-class males in spontaneous informal speech. They reported that, not only was there more use of the standard variant in CDS than ADS and in CDS towards younger children, but there were also stylistic constraints on use. According to one of Labov’s (2006) principles of transmission that linguistic variation is transmitted to children as stylistic differentiation on the formal-informal dimension (p.437), they found that caregivers used more of the local variant in contexts of play/routine than in those of teaching/discipline. However, they found that the same effects were not observed for most of other variables; some variables mirrored community norms very quickly while others remained variable in the early stages of language acquisition. This led Smith and colleagues to conclude that variables have different ‘sociolinguistic value’ in CDS. Roberts, 1999, Roberts, 2013, who investigated mothers’ variable use of monophthongal long (ay) variable, as in [kaːt] for kite in Southern American English, also found that mothers used more diphthongal (ay) when talking to their children (aged 1;6–1;7) than when talking to an interviewer. One mother also emphasised and exaggerated the diphthongal glide when teaching new vocabulary to her child. Roberts explained that the use of the more standard variant was in part due to their role as teachers of language. As Foulkes, Docherty, & Watt (2005) pointed out, segmental choice in CDS must be “viewed with one eye on the social-indexical values of the alternatives” (p.198); caregivers in these studies used both standard and nonstandard forms in CDS according to the norms of the community, and this was argued to be important in helping children construct a full sociolinguistic repertoire.

CDS in bilingual contexts involves even more variability. Compared to monolinguals, bilinguals vary greatly in their language experiences and background, and so do the specific phonetic and phonological properties of their CDS, which can differ from one bilingual to another, and from their monolingual counterparts, to varying degrees. Differential features in CDS may be due to caregivers being non-native speakers or late learners of the L2 (e.g. Fish, García-Sierra, Ramírez-Esparza, & Kuhl, 2017). Khattab (2002, see also Khattab, 2011), for example, reported that the Lebanese caregivers in her study who had lived in Yorkshire for over 10 years used clear-l syllable-finally instead of dark-l in their English CDS, possibly due to the influence of their Arabic L1. In some communities, distinctive features that emerge from language contact and acquisition are transmitted to and retained by later generations to become associated with particular socio-demographic groups, and further become reallocated with social functions (e.g. Sharma & Sankaran, 2011). The social-indexical meanings of these features allow them to be strategically used as part of one’s ethnolinguistic repertoire, such as to index their ethnic identities or cultural affiliations (Benor, 2010, Eckert, 2008, Hoffman and Walker, 2010), even if they are not dominant in or no longer bilinguals of the substrate or ethnic community language (e.g. Kirkham, 2017). Sharma (2011), for example, examined the use of ethnically-marked variants in the production of /t/, coda /l/, and the FACE and GOAT vowels in second generation British-born Asians (younger and older males and females) towards different interlocutors. She found that the older men and younger women were more strategic and differentiated than others in their use of the different variants; they were generally more ethnic in their use of variants with Asian speakers and with their direct family, and more mainstream with Anglo interlocutors. She argued that the differences in the diversity of the social networks of the participants, the socio-political context that the speakers grew up in, and their cultural orientation could explain why some speakers commanded a more complex repertoire range.

In the same way that bilectal monolinguals and ethnolect speakers vary their speech styles, Singaporeans may choose from their English repertoire features belonging to established standards (the prescriptive norms) and local forms, some of which more ethnically marked than others (see Leimgruber, 2013, pp. 26-63, for a discussion). Recent descriptions of variation in Singapore English (SgE) that are aligned with third-wave variationist sociolinguistics examined language use based on the socio-indexical meanings of these linguistic resources (e.g. Alsagoff, 2007, Leimgruber, 2013). Depending on the context of use, variants that are associated with standard varieties of English may index formality, authority, and educational attainment. Contrastingly, local features, which include ‘Singlish’ and ethnic markers, embody sociocultural capital and may index informality, camaraderie, and group membership. In terms of segmental modifications, Moorthy & Deterding (2000), for example, found that the Singaporean undergraduates in their study used more dental fricatives in a formal conversation with a British lecturer compared to speaking with a Singaporean student that they were familiar with, where th-stopping was more frequent. Leimgruber (2013, p. 66) also described the release or aspiration of coda stops, which are usually not released in SgE, to index a pretentious or pedantic stance in some contexts. In formal styles, Singaporeans were also found to be less ethnically accented (e.g. Deterding & Poedjosoedarmo, 2000, Sim, 2019). As a result of significant language shifts since the 1960's, Bolton & Ng (2014) described the various ethnic groups in Singapore to be in a similar situation to immigrant groups elsewhere in the world, in that the third generation of Singaporeans is increasingly more competent in English than their ethnic languages. Like the second-generation speakers in Sharma (2011) and Sharma and Sankaran (2011), therefore, language choices that the current generation of Singaporeans make, including the use of ethnically distinct features, are less likely to be related to English proficiency, imperfect learning or cognitive constraints, but more so to be as a result of and motivated by socio-cultural factors.

This study aims to find out whether and how Singaporean English-Malay bilingual caregivers make segmental modifications in their CDS towards their young children, and the possible socio-indexical factors that modulate its variation. The feature of focus is syllable-final /l/. This presents an interesting case as there are potentially three forms that have been described in previous SgE studies that may be used by these caregivers: vocalised-l (or deleted-l in some phonetic contexts; the predominant local forms), dark-l (the variant associated with exonormative standards), and clear-l (a Malay-derived variant used by some English-Malay bilinguals).

Cross-linguistically, alveolar laterals differ with regard to their degree of velarisation and/or pharyngealisation, with some languages having a darker (more velarised/pharyngealised) variant than others. Articulatorily, darker /l/ is characterised by a greater degree of tongue predorsum lowering and of postdorsum retraction towards the uvular area or upper pharyngeal wall, and the alveolar closure may also be more anterior (see Recasens & Espinosa, 2005). While the darkness of /l/ is a scalar phonetic property, language varieties can be categorised according to whether they exhibit a clearer or a darker /l/ variant (Recasens, 2012). In addition, some languages exhibit a clear or dark variant in all syllable positions, while others exhibit both that are syllabically conditioned (Recasens, 2004, Recasens, 2012, Recasens and Espinosa, 2005). Southern varieties of British English and American English, for instance, are typically described to have a clearer lateral in the syllable onset and a darker lateral in coda position (Sproat and Fujimura, 1993, Wells, 1992). Coda laterals may also be vocalised in some language varieties, a process by which the tongue tip contact with the alveolar ridge is lost, and is replaced by either a (labial-)velar approximant or a back vowel or semivowel. Further, for some varieties of English (e.g. Hong Kong English [Wee, 2008] and African American English [Thomas, 2007]), coda /l/ is argued to be possibly deleted in certain phonetic environments, such as after a back, rounded vowel.

Syllable-final /l/ in SgE tends to be vocalised. Deterding (2007) added that coda /l/ may also be deleted after back vowels (e.g. ball [bɔː], pull [puː]) or when it follows a schwa (e.g. little [lɪtə]; syllabic [l] does not typically occur in SgE). Using a generative approach, Wee (2008) argued that the underlying representation for lateral-final words in SgE is similar to Standard English, and the surface forms are derived from L-vocalisation rule and not L-deletion. He further explained that laterals that are preceded by back vowels also undergo the vocalisation rule, but the vocalised /l/ may assimilate to the respective preceding back vowel due to ease of articulation, thereby lengthening the vowel. As with past descriptions of and studies on coda /l/ in SgE, syllable-final /l/ vocalisation and deletion are treated as forms of one dialectal feature in this study, which is referred to here as L-lessness, following studies on African American English (see Thomas, 2007). Tan (2005) examined the production of syllable-final /l/ in conversational speech and read speech of educated Chinese Singaporeans. Based on listening judgement tests by ten Chinese Singaporeans and four British listeners, he found that while no speakers consistently used dark-l or vocalised-l in all their speech, the percentage of vocalised-l varied significantly between speakers, ranging from 39% to 89%, but reported no significant gender effects. There were also significantly more incidences of vocalised-l in faster read speech, though no effect of style between read speech and conversational data was observed. However, as he pointed out, the conversational speech and read speech data were not matched, and therefore linguistic factors such as phonetic environment could not be controlled.

Some studies have found that there are ethnic differences in the speech of Singaporeans, such that their ethnicity could be identified from their speech alone (e.g. Deterding & Poedjosoedarmo, 2000). Sim, 2015, Sim, 2019 found differences in the production of /l/ by Singaporean English-Malay bilinguals. Malay /l/ is typically realised as a voiced alveolar lateral, and laterals are always clear, in all word positions (Clynes & Deterding, 2011, Yunus, 1980). The distribution of Malay /l/ is also similar to English /l/: it occurs word-initially (e.g. lima ‘five’), word-finally (e.g. muncul ‘appear’), syllable-finally (usually forming a consonant cluster across morpheme boundaries before suffixes; e.g. meninggalkan ‘to leave behind’), and intervocalically (e.g. tilam ‘mattress’). Sim measured the production of /l/ by ten Singaporean English-Malay early sequential bilinguals1 between the ages of 19 and 28 (M = 23.1, SD = 2.51) in spontaneous speech using F1 and F2 as acoustic cues. He found that the Malay subjects preserved 54.8% of all absolute word-final /l/, and the rest were vocalised or dropped. He also noted that the coda laterals of English-dominant subjects were darker, whereas almost all produced by the Malay-dominant subjects were much clearer, with a statistically significant difference in the F2 but not F1. All participants were early or simultaneous bilinguals, however, and should have formed separate phonetic categories for their two languages or at least show distinct production patterns for the two languages (Barlow, Branson, & Nip, 2013, Khattab, 2002, Khattab, 2011, Kirkham & McCarthy, 2021). Sim posited that, rather than this being an effect of cross-linguistic influence, clear-l could have been learned through the input, similar to how British Asians acquired ethnically-marked features (e.g. Kirkham, 2017, Sharma, 2011). The retention and use of coda clear-l could also have been motivated by socio-indexical reasons; based on the results from the language background survey, his Malay-dominant subjects were associated with more Malay-dominant families and social circles, and identified more with a Malay-speaking culture.

Several studies show how the use of allophones of the alveolar lateral can be socially conditioned. British Asian English, for instance, is often characterised as having clearer allophones of coda /l/, due to likely effects of languages with clearer /l/ variants such as Panjabi, Urdu or Arabic, and is used variably to signal group membership or to index social distinctions among peer groups (e.g. Khattab, 2002, Kirkham, 2017, Sharma, 2011). The use or avoidance of distinctive features can also be attributed to other social meanings that emerged from various sociohistorical processes. One such example is Simonet, 2010a, Simonet, 2010b study of the alveolar laterals of Catalan-Spanish adult bilinguals. Majorcan Catalan has dark-l in all positions, while Spanish has clear-l in all positions. Simonet revealed that, especially in Majorca, dark laterals seemed to index local and rural origin of a speaker and used stereotypically by native Spanish speakers and Spanish-dominant bilinguals when joking about Catalan-accented Spanish. He further explained that this was perhaps so because Spanish monolingual speakers settled mostly in the main Majorcan metropolitan areas during the mass migratory waves in the 1950's and 1960's, when Majorcan Catalan had a low level of social prestige for socio-political reasons. This led Simonet to posit that a reason why his Spanish-dominant female subjects had a merged L1 + L2 lateral category could be because they may have distanced themselves from what they might have perceived as Catalan-accented Spanish, which could also explain why they also produced clearer laterals than older females of similar linguistic background. A few studies have also reported gender effects. Mackenzie, Decker, & Pierson (2015), for instance, studied the English speech of speakers in Irish-settled areas of Newfoundland, Canada, which was reported to exhibit clear-l in all positions. They found that, like the pattern in standard North American English, darker /l/ was used word-finally. However, they observed acoustic differences between women and men, where women made a significantly greater difference in terms of lateral darkness between initial and final /l/. They interpreted this as indicating that men were preserving more traditional variants than women. In another study, Clothier (2019) compared the production of /l/ between Australians with Lebanese ethnic identities that had parents and/or grandparents who were born in Lebanon, and Australia English speakers of Anglo-Celtic Australian heritage. He found that Lebanese Australian women with stronger, denser ties with the Lebanese community made a sharper distinction between dark-l in final position and clear-l in initial position, showing no substratum transfer, illustrating how men and women can be socialised into their ethnicities differently.

The above studies have shown how social factors modulate the linguistic choices of bilectal monolinguals in their CDS, and also described how alternative speech forms, in particular the variants of /l/, can be used strategically by bilinguals or speakers of ethnolects based on their socio-indexical meanings. Many of the same social factors influence the linguistic choices that Singaporeans make, as they choose from their repertoire alternative forms belonging to standard varieties and local dialect features, the latter including features that are ethnically distinct, based on their communicative needs. This study aims to find out whether and how Singaporean English-Malay bilingual caregivers vary their use of variants of /l/ in their English CDS towards their preschoolers, and the possible social factors that modulate its variation. To this end, it aims to answer these research questions:

  • 1.

    What syllable-final /l/ variant(s) do English-Malay bilingual caregivers use in their CDS?

  • 2.

    Do the variants of /l/ used in CDS vary according to situational context?

  • 3.

    Are there differences in the production patterns between mothers and fathers?

  • 4.

    Is the phonetic variation, if any, socially conditioned?

Section snippets

Participants

The corpus used in this study comprises ten Singaporean English-Malay bilingual families that included the father, mother and their firstborn of ages 3;1 to 6;4 (M = 55.8 months, SD = 12.43). The child participants had not started attending primary school; children in Singapore only enter primary school upon the year they turn seven. The children were all simultaneous bilinguals, having been exposed to both languages by the age of three (Genesee & Nicoladis, 2007). The families were recruited

L-less versus retained coda laterals

The proportions of coda /l/ tokens that were l-less (i.e. vocalised/deleted) and retained according to parent and formality of situational context are shown in Fig. 3. By visual inspection of the figure, overall, both fathers and mothers share the same production patterns: the proportions of retained /l/ were greater in informal contexts, but in formal contexts, more /l/ tokens were l-less. Mixed-effects generalised linear regression was run to model the binary outcome of a coda lateral being

Discussion

This study set out to find out whether and how Singaporean English-Malay bilingual caregivers vary their use of variants of /l/ in their CDS towards their preschoolers according to situational context, and the possible socio-indexical reasons that could explain the phonetic variability. To remind the reader, there are three forms of syllable-final /l/ that have been described in previous SgE studies: l-lessness (vocalised-l or deleted-l, the predominant local forms), dark-l (the variant

Data statement

As the corpus contains speech data of very young children, the parents were assured raw data would remain confidential and would not be shared.

Sociophonetic variation in English /l/ in the child-directed speech of English-Malay bilinguals.

Declaration of Competing Interest

The author declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The author would like to thank the families who participated in this study. He would also like to thank his PhD supervisor, Brechtje Post for her comments on this paper, Katrina Li Kechun for her help with the statistical analyses in an earlier version of this paper, and Hurul Ain and Dr Alicia Izharuddin for their insights into the Malay community. This work was supported by the Faculty of Modern and Medieval Languages and Linguistics and Jesus College, University of Cambridge; and the

References (94)

  • R. Sproat et al.

    Allophonic variation in English /l/ and its implications for phonetic implementation

    Journal of Phonetics

    (1993)
  • J.F. Werker et al.

    Infant-directed speech supports phonetic category learning in English and Japanese

    Cognition

    (2007)
  • L. Alsagoff

    Singlish: Negotiating culture, capital and identity

  • K. Bolton et al.

    The dynamics of multilingualism in contemporary Singapore

    World Englishes

    (2014)
  • Z. Bao

    The making of vernacular Singapore English: System, transfer and filter

    (2015)
  • J. Barlow et al.

    Phonetic equivalence in the acquisition of /l/ by Spanish-English bilingual children

    Bilingualism: Language and Cognition

    (2013)
  • D. Bates et al.

    Fitting linear mixed-effects models using lme4

    Journal of Statistical Software

    (2015)
  • S.B. Benor

    Ethnolinguistic repertoire: Shifting the analytic focus in language and ethnicity

    Journal of Sociolinguistics

    (2010)
  • D. Birdsong et al.

    Bilingual language profile: An easy-to-use instrument to assess bilingualism

    (2012)
  • Boersma, P., & Weenink, D. (2019). Praat: Doing phonetics by computer (6.0.45) [Computer software]....
  • T. Broesch et al.

    Fathers’ infant-directed speech in a small-scale society

    Child Development

    (2018)
  • F. Cavallaro et al.

    Singapore Colloquial English: Issues of prestige and identity

    World Englishes

    (2014)
  • F. Cavallaro et al.

    Language use and language shift among the Malays in Singapore

    Applied Linguistics Review

    (2010)
  • P. Carter et al.

    F2 variation in Newcastle and Leeds English liquid systems

    Journal of the International Phonetic Association

    (2007)
  • E.L.J. Chong et al.

    Young people, Malay and English in multilingual Singapore

    World Englishes

    (2014)
  • Clothier, J. (2019). A sociophonetic analysis of /l/ darkness and Lebanese Australian ethnic identity in Australian...
  • A. Clynes et al.

    Standard Malay (Brunei)

    Journal of the International Phonetic Association

    (2011)
  • A. Cristià

    Phonetic enhancement of sibilants in infant-directed speech

    The Journal of the Acoustical Society of America

    (2010)
  • J. Davidson

    Phonetic interference of Catalan in Barcelonan Spanish: A sociolinguistic approach to lateral velarization

  • Deterding, D. (2007). Singapore English. Edinburgh University Press....
  • D. Deterding et al.

    To what extent can the ethnic group of young Singaporeans be identified from their speech?

  • P. Eckert

    Where do ethnolects stop?

    International Journal of Bilingualism

    (2008)
  • L.G.W. En et al.

    A preliminary report on the English phonology of typically developing English-Mandarin bilingual preschool Singaporean children: English phonology of typically developing English-Mandarin Singaporean children

    International Journal of Language & Communication Disorders

    (2014)
  • P. Foulkes et al.

    Phonological variation in child-directed speech

    Language

    (2005)
  • P. Foulkes et al.

    The emergence of sociophonetic structure

  • F. Genesee et al.

    Bilingual first language acquisition

  • F. Grosjean

    The bilingual’s language modes

  • T. Hills

    Duck and Goose (T. Hills, Illus.)

    (2006)
  • M.F. Hoffman et al.

    Ethnolects and the city: Ethnic orientation and linguistic variation in Toronto English

    Language Variation and Change

    (2010)
  • S. Holmes-Elliott et al.

    Dressing down up north: DRESS-lowering and /l/ allophony in a Scottish dialect

    Language Variation and Change

    (2018)
  • A.M. Kassim

    Malay language as a foreign language and the Singapore’s education system

    Gema Online Journal of Language Studies

    (2008)
  • M. Kehoe

    Cross-linguistic interaction: A retrospective and prospective view

  • G. Khattab

    /L/ production in English-Arabic bilingual speakers

    International Journal of Bilingualism

    (2002)
  • G. Khattab

    Acquisition of Lebanese Arabic and Yorkshire English /l/ by bilingual and monolingual children: A comparative spectrographic study

  • S. Kirkham

    Ethnicity and phonetic variation in Sheffield English liquids

    Journal of the International Phonetic Association

    (2017)
  • S. Kirkham et al.

    Acquiring allophonic structure and phonetic detail in a bilingual community: The production of laterals by Sylheti-English bilingual children

    International Journal of Bilingualism

    (2021)
  • P.K. Kuhl et al.

    Cross-language analysis of phonetic units in language addressed to infants

    Science

    (1997)
  • View full text