Mutual Intelligibility and Acceptability of Regional Indian English (RIE). Accents, Attitudes and Intelligibility of RIE

Master's Thesis, 2021

118 Pages, Grade: 65.00


Table of Content

A quantitative analysis of mutual intelligibility and acceptability of regional Indian English

List of figures

List of tables

Chapter 1 - Overview
1.1 Introduction
1.2 Background and context
1.2.1 Indian English and World Englishes (WEs)
1.2.2 The multi-lingual context
1.2.3 The educational context
1.2.4 Standard IE (SIE) and Regional IE (RIE)
1.3 Rationale of the study
1.3.1 On mutual intelligibility
1.3.2 On mutual perceived acceptability
1.4 Aim of the study
1.4.1 The research questions

Chapter 2 – Literature review
2.1 Introduction
2.2 Intelligibility
2.2.1 Understanding’ and ‘intelligibility’ in World Englishes
2.2.2 Theoretical frameworks and approaches – WEs approach
2.2.3 Research on intelligibility Testing intelligibility measures – an empirical study Mixed studies on Intelligibility and attitudes
2.2.5 Conclusion
2.3 Language attitudes and the ‘status’ and ‘solidarity’ dimensions – theoretical framework
2.3.1 Introduction
2.3.2 Definitions and constructs
2.3.3 Accent recognition and stereotyping
2.3.4 Methods and techniques
2.3.5 Research studies that used status and solidarity – World Englishes paradigm
2.3.6 Language attitude studies on IE

Chapter 3 – Methodology
3.1 Introduction
3.2 The rationale for selecting the scalar rating method
3.1 The selection process of participants
3.1.1 The rationale for the selection of speaker group
3.1.2 The rationale for the selection of respondent group
3.3 Data collection process
3.3.1 The online questionnaire / survey
3.3.2 The Verbal-Guise Technique (VGT) and the speech sample used
3.3.3 The scales used
3.4 Ethics
3.4 Data analysis
3.4.1 Statistical methods
3.4.2 On Likert scales, T-tests and ANOVAs

4. Results
4.1 Introduction
4.2. Some methodological aspects
4.3. Accent recognition
4.3.1 Introduction
4.3.2 Accent recognition by speakers’ region
4.3.3 Accent recognition by respondents region
4.3.4 Conclusion
4.4 Intelligibility
4.4.1 Introduction
4.4.2 Question 3: Perceived intelligibility Intelligibility by speakers’ region Intelligibility by respondents’ region
4.4.4 Question 14: Expected communication problems Communication problems by speakers’ region
4.5 Attitudes
4.5.1 Introduction
4.5.3. Status and solidarity by speakers’ region (1503)
4.5.4 Status and solidarity by respondents’ region
4.5.5 Principal Component Analysis (PCA)
4.6 English Medium (EM) versus Vernacular Medium (VM) education
4.6.1 Introduction
4.6.2 Recognition of EM educated speakers
4.6.3 EM/VM and accent recognition
4.6.4 EM/VM and intelligibility Perceived intelligibility Expected communication problems
4.6.5 EM/VM and attitudes

5. Discussion
5.1 Introduction
5.2 On preferences and bias
5.3 Accent recognition – main findings and implication
5.3 Intelligibility
5.3.1 Perceived intelligibility
5.3.2 Regional bias in intelligibility
5.3.3 Conclusion
5.4 Language attitudes - Status and solidarity evaluations
5.4.1 Status and solidarity among speakers’ and respondents’ groups
5.4.2. Regional bias in status and solidarity
5.4.2 Principal Component Analysis
5.4.3 Conclusion
5.5 Medium of education
5.5.1 Accent recognition and EM/VM
5.5.2 Intelligibility and EM/VM
5.5.3 Status, solidarity and EM/VM
5.5.4 Educational bias
5.5.4 Conclusion

6. Conclusions
6.1 General findings
6.2 Limitations and drawbacks of the study
6.2 Implications and recommendations


Appendix 1 – Questionnaire

Appendix 2 – Informed Consent Form

Appendix 3 – Personal Details Form

Appendix 4 – Participant invitation letter and Information Sheet

Appendix 5 – Research data

Appendix 6. Additional tables Chapter 4

List of figures

Figure 1. Distribution of major language groups in India and the four regions that were chosen for this study

Figure 2. Frequency distribution of ratings of Question 3: “How well do you understand the speaker?”

List of tables

Table 1. Accent recognition, aggregated by speakers’ region and respondents’ region. Absolute numbers of correct recognition are given, both including speakers and respondents from the same region (incl. ‘self’, N=256), as well as without these (excl. ‘self’, (N=192)

Table 2. Two-way ANOVA on accent recognition scores of speakers and respondents from the four different regions, speakers averaged per region (N=128)

Table 3. Intelligibility by speakers’ and respondents’ region, averaged including speakers and respondents from the same region (incl. “self”, N=128), as well as without these (excl. “self”) (N=96)

Table 4. Two-way ANOVA on perceived intelligibility ratings of speakers (averaged per region) and respondents from the four different regions (N=128)

Table 5. Expected communication problems (Q14) by speakers’ region, averaged including speakers and respondents from the same region (inc. ‘self’), as well as without these (exc. ‘self’)

Table 6. Two-way ANOVA on ratings on expected communication problems (Q14) of speakers and respondents from the four different regions (N=128)

Table 7. Status and solidarity by speakers’ and respondents’ region. Values for status are averaged values of five ‘status traits,’ values for solidarity averaged values of five ‘solidarity traits.’ Minimum and maximum values are indicated in bold (N=256)

Table 8. Two-way ANOVAs on status and solidarity ratings of speakers and respondents from the four different regions, including speakers and respondents from the same region

Table 9. Component loadings on two primary components of PCA with Varimax rotation on 5 status and 5 solidarity traits. Traits with loadings ≤ .15 were omitted for clarity

Table 10. Component loadings on primary components of PCA with Varimax and Oblimin rotations on 13 traits, including status and solidarity traits (each 5 traits), intelligibility (2 traits) and perceived English Medium-education (1 trait). Traits with loadings ≤ .15 have were omitted for clarity

Table 11. Accent recognition by respondents, including speakers and respondents from the same region, for speakers with English medium-education (N=256)

Table 12. Perceived intelligibility (Q3), per speakers’ region and medium of education (EM, VM), with speakers and respondents from the same region included (inc. ‘self’, N=256), as well as without these (excl. ‘self’, N=192)

Table 13. Status, per speakers’ region and medium of education (EM, VM), with speakers and respondents from the same region included (inc. ‘self’, N=256), as well as without these (excl. ‘self’, N=192)

Table 14. Solidarity, per speakers’ region and medium of education (EM, VM), with speakers and respondents from the same region included (inc. ‘self’, N=256), as well as without these (excl. ‘self’, N=192)

Table 15. Average scores on perceived intelligibility for alle speakers (N=256), speakers whose accent was correctly identified (N=115), and those whose accent was incorrectly identified (N=141), including speakers and respondents from the same region

Table 16. Average ratings on status for alle speakers (N=256), speakers whose accent was correctly identified (N=115), and those whose accent was incorrectly identified (N=141), including speakers and respondents from the same region

Table 17. Average ratings on status for VM speakers who were correctly identified as such, incorrectly identified as EM speakers and all VM speakers (N=128)

Chapter 1 - Overview

1.1 Introduction

This paper examines the complex relationship between accents, attitudes and intelligibility and their covert but conspicuous impact on the varieties of Regional Indian English (henceforth RIE1 ) in India.

Accents tell “the story of who you are” in terms of being the most revealing predicators of a speaker’s geographical origins and social background, including perceived educational levels, professional status, social class and identity (Matsuda 1991, 1329; Beinhoff 2014, 59).

These “social and acoustic” aspects of an accent often give rise to stereotypes and value judgements of the speakers and/or groups, especially if it differs in noticeable ways from established pronunciation patterns (Levis and Zhou 2018, 2). This leads to negative reactions or language attitudes that are reflected in systemic discrimination in areas of education and employment (Munro and Derwing 1995, 290).

It can be argued that to eschew this discriminatory practice of stereotyping non-native or foreign accents, the intelligibility construct emerged as early as 1900 (Henry Sweet) as an alternative, egalitarian L2 pronunciation target option for L2 learners (Munro and Derwing 2015, 378). Hence, intelligibility of L2 speech pattern found itself to be one of the main concerns of accents (Beinhoff 2014, 59).

In the study of accents, the object of study is not so much the speaker’s speech style but the listener’s hearing and understanding (Lippi-Green 2012, 45). Munro makes a further salient point by noting that the effect of accents on communication can be best judged by an “unsophisticated listener” to see how understandable L2 speakers are within their community (Munro 2008, 200). This kind of judgement data of listeners’ perceptions are considered to be “gold standard”, (Derwing and Munro 2009, 478).

However, it is also possible that listeners’ perceptions are a result of stereotyping and negative value judgements. This being so, the present paper is concerned with two issues primarily: a) the mutual intelligibility of L2 speech (IE speech) and b) acceptability in terms of language attitudes towards RIE accents/speakers. In other words, the present study is a listeners’ judgement study that attempts to see how mutually intelligible and acceptable IE speakers are within their own regions.

Following Munro and Derwing’s (1995, 1997) model’s example, this study refers to the intelligibility dimension as ‘perceived’ intelligibility as it examines the listeners’ subjective judgements.

1.2 Background and context

The next few paragraphs touch upon the crucial socio-linguistic context to this study by looking at Indian English and its place in the World Englishes paradigm, regional (multilingual) context and the education context.

1.2.1 Indian English and World Englishes (WEs)

Many New Varieties of English (NVEs) have emerged in the Outer Circle2 countries such as Singapore, India, and Nigeria where people possess “an excellent command of the English language” with their own well-established pronunciation patterns that vary “quite substantially” from the pronunciation patterns of native speakers from the Inner Circle countries such as Britain, the USA and Australia (Kachru 1985; Kachru 2005, 14; Deterding 2010, 364).

Among the NVEs, IE is the oldest and the “largest institutionalised second language variety” with the highest range of functions in “institutionalised” settings such as “administration, education, media, judiciary, legislature, etc.” (Sridhar 2020, 241; Hoffman et al. 2011, 258; Mukherjee 2010, 167; Pingali 2009, 5).

Today, IE is probably used by at least 300 million people, all with varying accents because of different regional, social and educational backgrounds (Sridhar 2020, 242; Pandey 2015, 301).

1.2.2 The multi-lingual context

According to Sridhar (2020) the nature of IE can be best explained by its “function in the complex multilingual ecology of India” (Sridhar 2020, 241). India has twenty-nine states and seven union territories, most of which are “organised and determined” linguistically (Pingali 2009).1 The primary languages spoken in these states are known as regional languages or mother tongue languages, i.e., the native language of that region.

These regional/mother tongue languages belong to four major language groups or families. The Indo-Aryan group of over 50 languages includes Kashmiri, Punjabi, Sindhi, Marathi, Assamese Bengali, Gujarati, Hindi, Oriya, among others. It is spoken by about 70 percent of the population from northern, central and western India (Sridhar 2020, 243). The Tibeto-Burman group includes 84 languages such as Bothi (Ladakhi), Angami, Ao, Bodo, etc., and is spoken by only 1 percent of the population living in the northern and eastern border areas (ibid.; Pingali 2009, 2). The Austro-Asiatic group, found only in East India, has 20 languages such as Munda, Santhali, Khasi, etc., and is spoken by about 1 percent (possibly less). The Dravidian group, spoken by about 20 percent of the population, has 26 languages that includes Tamil, Telugu, Malayalam and Kannada (Sridhar 2020, 243; Pingali 2009, 2).

Most IE speakers therefore have at least three languages in their linguistic repertoire: Mother-Tongue (MT) and/or the Regional Language (RL) (depending on migration), Hindi and English. As a result, first language (L1) influences of MT and/or RL affect L2 pronunciation of IE quite significantly, resulting in the wide ranging regional variations found in the IE speech pattern. The four major regions selected in this study are from the four corners of the country that correspond to three major language groups of India (see figure 1): Indo Dravidian (Karnataka/South), Tibeto-Burman (Ladakh/ North) and Indo-Aryan (Maharashtra/West and West Bengal/East). The reason that two regions were chosen from the Indo-Aryan language group is that this is the most populous, geographical widespread and diverse group of languages.

Abbildung in dieser Leseprobe nicht enthalten

Figure 1. Distribution of major language groups in India and the four regions that were chosen for this study.

1.2.3 The educational context

The Indian education policy follows the “three language policy” (TLF), wherein each child learns the regional language, Hindi and English (Wiltshire 2005, 280). However, because of the deeply ingrained colonial legacy of the idea that “good education equals knowledge of English” (based on the Educational Dispatch of Charles Wood, 1854), the private, expensive schools and colleges with English as the Medium of Instruction (MI) are still considered to be the best (Sridhar 2020, 245). The subsidised government and public schools with Hindi or regional state languages as MI are considered inferior, and indeed are “of poor quality” in reality (Mohanty 2006, 268). The private, expensive, English Medium (EM) schools can only be afforded by the well-to-do, leaving the Vernacular Medium (VM) government schools for the socially disadvantaged (Mohanty 2010, 144-45).

Thus, educational inequality exacerbates an existing socio-economic class dichotomy (Sridhar 2020, 245-6). This EM/VM divide, also called the “linguistic double divide” is a crucial element in this study (Mohanty 2010, 144). The EM/VM divide is further aggravated by the EM educated speaking the Standard variety of IE (henceforth SIE3 ) and the VM-educated speaking the non-standard Regional variety of IE (RIE) .

1.2.4 Standard IE (SIE) and Regional IE (RIE)

Broadly speaking, a Standard language variety subscribes to “codified norms” that define “correct spoken and written” forms in terms of grammar, lexicon and pronunciation, while non-standard language varieties are those that deviate from the “codified”, correct norms in certain ways (Dragojevic and Giles 2016, 3).

However, Lippi-Green (1997) considers “standard language” to be a “myth”, at any rate, a “hypothetical social construct” ordained by the educated of that land (Lippi Green 2012, 44, 57). Corroborating Lippi- Green’s observation, the Standard variety of Indian English is indeed the “language of the educated” used by a relatiely “small proportion of proficient IE speakers, accepted as the Standard IE that the rest of India “aspires to acquire” (Lippi-Green 2012, 57; Bayley and Villareal 2018, 1; Pingali 2009, 18).

Hence, a vast majority of the population – who cannot access EM education –speak a Regional variety of IE (RIE) that deviates from the Standard Variety of IE (SIE) because of the speakers’ regional/mother tongue L1 influence and other socio-economic factors.

According to Pandey (2015, 301), EM education and higher education have "helped reduce the variation" in the pronunciation and accent of SIE, the general variety of acceptable standard across the nation. Pingali too concurs that “schools and peer groups influence accent more than MT/RL influence” (2009, 18). Consequently, the SIE accent is generally found to be free of regional pronunciation patterns irrespective of the regional background of the speaker (Kachru 1976, 233).

1.3 Rationale of the study

1.3.1 On mutual intelligibility

Decades ago, Kachru posed the critical question of whether “Indian English (henceforth IE) is intelligible to Indians all over the sub-continent, in spite of regional and language bound variations” (1976, 29). Kachru’s own response to it was that only “Standard’ or ‘educated’ IE has pan-Indian intelligibility”. To be noted is that this response was a personal observation, not the result of an empirical study. Since then and now, no study has actually tested this claim, and certainly not whether Regional Indian English has pan-Indian intelligibility. Hence, this study aims to address this glaring gap in research, by examining the perceived ‘mutual intelligibility’ of both SIE and RIE accents to see if they are understood by each other within the same regional lingua-groups.

1.3.2 On mutual perceived acceptability

The Standard IE accent has always been linked with higher status, as well as higher levels of wealth and education, while the regional accent is not (Pingali 2009, 18). This systemic discrimination can have a powerful impact on the minds of the well-educated youth in India who speak with the SIE accent to subconsciously, if not consciously, stereotype those with RIE accents. Language attitude studies reveal that perceptions of listeners towards speakers “who use certain linguistic features” different from their own, can “cause bias” and impair communications quite seriously (Preston and Robinson 2003, 133).

So, when Indian English speakers from the four corners of the country communicate in English, lack of mutual intelligibility and acceptability is potentially a critical issue, especially in high-stake communications.

1.4 Aim of the study

Hence, as “communicating successfully” and “getting the message across” is a critical consideration (Schneider 2011, 217), this study aims to determine the perceived mutual intelligibility of RIE accents from the SIE speakers’ perspective; find out if there is attitudinal bias among young SIE speakers towards RIE speakers based on the EM/VM divide; and if the bias has any effect on the perceived intelligibility of RIE accents.

1.4.1 The research questions

1. Do IE speakers from different regions in India recognise the regional accents of IE speakers of the same and other regions?
2. Are the regional IE accents in India mutually intelligible for listeners from the different regions?
3. Are the 4 regional accents mutually acceptable to listeners from the same 4 regions?
a. How do speakers from different regions rate the IE speech of speakers of the same and other regions with respect to a number of attitudinal traits?
b. Do the data confirm that ‘status’ and ‘solidarity’ can be recognised as overarching attitudinal dimensions?
4. Does English medium-education and higher (university) level education improve ‘perceived intelligibility’ and ‘acceptability’ as compared with vernacular medium-education?

This study aims to look into the phenomenon of bias as well. However, as bias is an implicit characteristic that is generally kept veiled from public scrutiny, it could not be incorporated in the questionnaire for analysis. As a result, it was not included overtly as a research question either. Primarily, also because determining the ‘acceptability’ of an accent (a main research question) implies looking for instances of stereotyping. And stereotyping creates the very conditions for bias to grow and fester (Deutschmann and Steinvall 2020, 651). Hence, as bias is a result of a process, this study will attempt to statistically shed light on this subject through data interpretation in Chapter 5 (discussion).

Chapter 2 – Literature review

2.1 Introduction

This chapter is divided into two parts: intelligibility and acceptability. The first part begins with definitions and constructs, followed by theoretical approaches and frameworks that have used the term ‘understanding’ in original, nuanced ways to (de)construct the notion of intelligibility in the context of L2 pronunciation of World Englishes. The second part will be on acceptability that will follow the same format.

2.2 Intelligibility

Serious research studies on either intelligibility or comprehensibility of IE have not been conducted in India (Gargesh 2018, 465). Apparently, the “question of intelligibility doesn’t arise”, as all regional or sub-varieties of IE are presumed to have a broad homogenous base (ibid.). Most studies on IE have therefore focussed on comparing the regional varieties of IE with the RP model, phonologically mostly (Gargesh 2008, 231-2; see Balasubramanium 1972; Kelkar 1957). Bansal (1969) is the only “classic work on the intelligibility of IE” (Rajagopalan 2010, 466). In the absence of relevant precedent studies within India, I look at intelligibility more conceptually than empirically in this chapter.

2.2.1 Understanding’ and ‘intelligibility’ in World Englishes

So far, there is no “satisfactory definition” of intelligibility nor a “consensus” on the best method to measure it (Munro and Derwing 2015, 377). The general definition of intelligibility that receives a grudging nod from the research community at large, and the present study, is Derwing and Munro’s description, which is, “the extent to which a speaker’s message is understood by the listener” (1995a, 76; 1999, 289). In time, researchers’ interest in ‘intelligibility’ of L2 speech led to the notion of ‘comprehensibility’ (Derwing 2018, 1).

Both intelligibility and comprehensibility are underpinned by the core concept of ‘understanding’. However, to best explain the term ‘understanding’, it has to be construed in the context of the situation it is being used in. Hence, when used in the context of differences across English varieties, as this study does, the term ‘understanding’ has generally to do with intelligibility issues, especially, if an utterance is perceived “to make sense” (Kachru and Nelson 2011, 66; Gargesh 2018, 465).

2.2.2 Theoretical frameworks and approaches – WEs approach

Catford (1950) underpins ‘understanding’ as the primary component of the intelligibility construct. Catford puts the onus of intelligibility squarely on the hearer’s perception of the speaker’s utterance by noting that “speech is intelligible if the hearer’s response is appropriate to the linguistic forms of the utterance” (1950, 8). Even though Catford’s concepts are not an outcome of any empirical study, he was the first to underscore the importance of “familiarity with a variety of speech lowering the threshold of intelligibility” (1950, 14).

Smith and Nelson (1985) introduced a tripartite framework with “three levels of understanding that include 1) “Intelligibility: word/utterance recognition; 2) comprehensibility: word/utterance meaning and 3) interpretability: meaning behind word/utterance” (Munro and Derwing 2015, 378; Berns 1990, 33; Nelson 2011, 23; Smith and Nelson 2006, 429; Gargesh 2018, 465). Smith (1992) broadened its scope by explaining it as “degrees of understanding on a continuum, with intelligibility being lowest and interpretability being highest” (76). This hierarchical character incurred criticism as “comprehension of speech is not a linear process” meaning that if some utterances take more time to process then a heirarchy with intelligibility being lowest does not work (Munro and Derwing 2015, 378).

In 1982, Varonis and Gass defined comprehensibility as “ease of understanding” from the listener’s judgement, using a “perceptual rating scale” such as the Likert-type scale (Derwing 2018, 1; Munro and Derwing 2015, 378; Munro and Derwing 1999, 291). The most important outcome of this study was that ‘understanding’ is possible only from considering the “linguistic input as a whole” (Varonis and Gass 1982, 131). Varonis and Gass’s (1982, 1984) research was later criticised for the way they operationalised ‘comprehensibility’ from the native speaker’s perspective of non-native speech (Browne 2019, 86).

Derwing and Munro’s (1997) and Munro et al. ’s (2006) model of intelligibility is based on Munro and Derwing’s (1995, 73) study that investigated the “inter-relationships between accentedness, perceived comprehensibility and intelligibility”. They too employ a tripartite distinction (non-hierarchical) based on understanding. So, in this model, accent underscores the “difference”, comprehensibility is the “listener’s effort”, and intelligibility can be said to be the final outcome of “how much the listener actually understands” (Derwing and Munro 2009, 480). However, Munro and Derwing’s approach too incurs criticism for a vagueness in their application. For instance, in Munro et al. ’s (2006) study, it is not clear to the reader how listeners made their judgments on accentedness in all three dimensions (Nelson 2011, 72).

Lastly, I touch upon Kachru’s arbitrary scale of bilingualism, also referred to as the cline of intelligibility as it is based on the intelligibility criterion (1965, 393). The cline has three measuring points: “zero point, the central point, and the ambilingual point” (ibid.). The zero point implies negligible proficiency in English, which strongly impacts intelligibility. The central point implies “adequate” proficiency, impacting intelligibility moderately while the ambilingual point refers to the highly proficient level of the Standard (educated) IE bilingual with high intelligibility in social and professional contexts (Kachru 1965, 394). This cline assumes “mutual intelligibility with all speakers of the language at one end” (Berns 1990, 40). This assumption of mutual intelligibility is practically improbable as it implies that interlocutors will always have “identical background and education” (Kachru 1976, 228). From the point of view of this study, the SIE (EM educated) speaker would be at the ambigual point and the RIE (VM educated) speaker at the central point of this continuum scale.

2.2.3 Research on intelligibility Testing intelligibility measures – an empirical study

Traditionally, intelligibility testing measures are of two types: functional testing and listener judgement measures. Although dictation and cloze-passage tests are the most “commonly used devices” for testing intelligibility of L2 speech, “neither is [apparently] a perfectly reliable sort of test” (Kachru and Nelson 2011, 74; Nelson 2011, 32). However, the general consensus among L2 pronunciation researchers is that the listener judgement measure, known as the scalar rating method is best for measuring comprehensibility, not intelligibility (Derwing and Munro 2009, 478; Kang et al. 2018, 118).

This being so, Kang et al. ’s study proposed to determine the most reliable intelligibility measure from five measures – 1) true/false statements; 2) scalar ratings; 3) nonsense sentences; 4) filtered sentences; 5) and transcription of speech (2018, 115). What is most significant and valid for our study is that Kang et al. also decided to test the hypothesis that “scalar ratings of ‘understanding’ are less valid for measuring intelligibility” (2018, 118).

The results contradicted the researchers’ own hypothesis and showed that the scalar rating method was in fact the only approach that was significantly predicted by all five categories of phonological variables (vowel/consonant errors, impeding fluency, enhancing fluency, impeding prosody and enhancing prosody) as opposed to the other intelligibility measures that were predicted by only few of the phonological variables. In other words, in making scalar judgements of intelligibility, listeners tended to process all phonological characteristics, making it the most all-encompassing of the intelligibility measures tested. It is for this reason that Kang et al. describes the scalar rating method as a “holistic, global approach” to measure intelligibility (2018, 138). These findings provide this study a solid basis to adopt the scalar rating methodology to test the intelligibility measure. Mixed studies on Intelligibility and attitudes

Even though the scalar rating method has been proved to be an all-encompassing measure to test intelligibility, intelligibility studies generally do not use it exclusively to test intelligibility but combine it with language attitudes. Most of these language attitude studies use the scalar rating method with intelligibility as subtext of the main attitudinal study. In contrast, the present study addresses the issue of intelligibility as one of the two major components of the study, the other being language attitudes. As a result, relevant studies were hard to find. So, I briefly touch upon the seminal studies of Smith and Rafiqzad (1979) and Smith (1992) that informed Smith’s frameworks. This will be followed by Chan’s (2016) and Al Dosari’s (2011) studies that are a combination of language attitudes and intelligibility. Munro and Derwing’s (1995) study will be briefly discussed for its findings of correlations between accent and intelligibility followed by Bansal’s (1969) study.

Smith and Rafiqzad’s (1979) premier study stunned the conservative ELT world by unequivocally eschewing the “primacy of the native speaker” with its striking finding that the native speaker (US) was the least intelligible and non native speakers (India, Japan and Malaysia) were more intelligiblet. Smith’s (1992) study too found a similar pattern with native speakers from Britain and United States, who were found to be neither the most “easily understood” nor the best listeners to understand the different varieties of English.


1 Regional Indian English (RIE) is the variety of Indian English with regional L1 influences that is spoken by the vast majority of Indians.

2 In the Kachruvian model of 3 circles the ‘Inner Circle’ is represented by countries such as Australia, New Zealand, UK and USA, where English functions primarily as a first language. The ‘Outer Circle’ is represented by nations such as India, Singapore, and the Philippines, where English is used as an institutionalised additional language; and the ‘Expanding Circle’ is represented by countries such as China, Thailand, Taiwan, and Korea, where English is used primarily as a foreign language (Kachru, 1998, 93).

3 Standard Indian English (SIE) is the educated variety of Indian English spoken by the upper and middle classes that is accepted as the Standard variety of IE (Pingali 2009, 18).

Excerpt out of 118 pages


Mutual Intelligibility and Acceptability of Regional Indian English (RIE). Accents, Attitudes and Intelligibility of RIE
University of Leicester
Applied Linguistics and TESOL
Catalog Number
ISBN (Book)
Ist marker remarks: "This is a very interesting study. You demonstrate very good knowledge of relevant theories and concepts, and you apply these competently to the analysis and discussion of your findings." 2nd Marker's comments: "Well done! This is a mammoth work which, at 20000 words, is well over the word limit. You have chosen an interesting topic and set up an investigation which is well founded in previous research".
Regional Indian English, Indian English, mutual intelligibility of Regional Indian English, Indian accents acceptability
Quote paper
Suchie Dutta (Author), 2021, Mutual Intelligibility and Acceptability of Regional Indian English (RIE). Accents, Attitudes and Intelligibility of RIE, Munich, GRIN Verlag,


  • No comments yet.
Read the ebook
Title: Mutual Intelligibility and Acceptability of Regional Indian English (RIE). Accents, Attitudes and Intelligibility of RIE

Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free