A study of attitudes of educated Nigerians towards varieties of World Englishes
Temitayo Olatoye (277262)
Master’s Degree Programme in Linguistic Sciences Philosophical faculty
University of Eastern Finland November 2018
ITÄ-SUOMEN YLIOPISTO – UNIVERSITY OF EASTERN FINLAND Tiedekunta – Faculty
Philosophical Faculty Osasto – School
School of Humanities Tekijät – Author
Temitayo Oluwayemisi Olatoye Työn nimi – Title
A Study of Attitudes of Educated Nigerians towards Varieties of World Englishes Pääaine – Main subject Työn laji – Level Päivämäärä –
Date Sivumäärä – Number of pages
General Linguistics Pro gradu -
tutkielma X 12.11.2018 71 pages+Appendix
Sivuainetutkielma Kandidaatin tutkielma Aineopintojen tutkielma Tiivistelmä – Abstract
This study investigates the attitudes of educated Nigerians towards Standard British English, Scottish English, American English, Acrolectal Nigerian English, Mesolectal Nigerian English and Ivorian English. Using an online questionnaire and the verbal guise technique, the attitudinal evaluations of 102 educated Nigerians were obtained via a six-point semantic differential scale with nine bipolar traits and subsequently analysed along status and solidarity dimensions.
The results indicate that American English, Acrolectal Nigerian English and Standard British English were rated most positively in terms of status, while American English was preferred to Standard British in terms of solidarity and status. The respondents were also able to identify the American, Standard British and mesolectal Nigerian varieties most correctly. Overall, the acrolectal Nigerian variety was evaluated most favourably in terms of status and solidarity, thus providing support for the adoption of an endonormative variety in Nigeria as a target norm for learners.
Avainsanat-Keywords
Language Attitudes, World Englishes, Verbal Guise Technique, Nigerian English
ITÄ-SUOMEN YLIOPISTO – UNIVERSITY OF EASTERN FINLAND Tiedekunta – Faculty
Philosophical Faculty Osasto – School
School of Humanities Tekijät – Author
Temitayo Oluwayemisi Olatoye Työn nimi – Title
A Study of Attitudes of Educated Nigerians towards Varieties of World Englishes Pääaine – Main subject Työn laji – Level Päivämäärä –
Date Sivumäärä – Number of pages
Yleinen Kielitiede Pro gradu -
tutkielma X 12.11.2018 71 siivua+liitteet Sivuainetutkielma
Kandidaatin tutkielma Aineopintojen tutkielma Tiivistelmä – Abstract
Tämän tutkielman aiheena oli tutkia koulutettujen nigerialaisten asenteita eri englannin kielen varieteetteja kohtaan. Tutkitut varieteetit olivat brittienglanti, skottienglanti, amerikanenglanti, niin kutsuttu akrolekti- ja mesolekti- Nigerian englanti sekä Norsunluurannikon englanti. Tutkimus suoritettiin verkkokyselynä käyttäen verbal guise- tekniikkaa, ja yhteensä 102 koulutetun Nigerialaisen asenteet selvitettiin yhdeksällä väitteellä, joihin vastattiin kuusiportaisella bipolaarisella semanttisen differentiaaliskaalan asteikolla. Vastaukset analysoitiin statuksen ja solidaarisuuden perusteella.
Tulokset osoittivat, että amerikanenglanti, akrolekti Nigerian englanti ja brittienglanti arvioitiin paremmaksi statuksen kannalta. Amerikanenglantia pidettiin brittienglantia parempana solidaarisuuden ja statuksen kannalta. Vastaajat pystyivät tunnistamaan parhaiten amerikanenglannin, brittienglannin ja mesolekti- Nigerian englannin piirteitä. Akrolekti- Nigerian englanti arvioitiin parhaaksi statuksen ja solidaarisuuden kannalta. Tämä tulos tukee endonormatiivisen Nigerian englannin valintaa normiksi oppijoille kouluympäristössä.
Avainsanat-Keywords
Kieliasenteet, Englanti maailmankielenä, Verbal guise- tekniikkaa, Nigerianenglanti
Table of Contents
1. Introduction ... 1
2. Varieties of English ... 5
2.1 English Worldwide ... 5
2.2 English in Nigeria ... 10
2.3 English as spoken in Nigeria………15
3. Investigating Language Attitudes ... 19
3.1 What Are Attitudes? ... 19
3.2 Components of Attitudes ... 20
3.3 Measuring Language Attitudes ... 21
3.4 Previous Research on Language Attitudes towards English……….24
4. Data and Methodology ... 30
4.1 Research Instrument ... 30
4.2 Participants ... 33
4.3 Speakers ... 35
4.4 Data Analysis……….………..36
4.4.1 Reliability Statistics ... 36
4.4.2 ANOVA ... 37
5. Results and Discussion ... 39
5.1 Results………39
5.1.1 Mean Ratings ... 39
5.1.2 Pairwise Comparisons... 42
5.2 Discussion………..47
5.2.1 Research Question 1 ... 47
5.2.2 Research Question 2 ... 56
5.2.3 Research Question 3 ... 58
6. Conclusion ... 60
References ... 63
Appendix I……….…….72
Appendix II……… 75
Appendix III………77
Appendix IV……….……...…78
List of Figures
Figure 1: Map of West Africa………..……….…..…..1
Figure 2: Kachru’s concentric circles……….…….……...6
Figure 3: A world system of Standard and Non-Standard Englishes…………...…...8
Figure 4: The Nigerian English continuum………...….…..17
Figure 5: The tripartite model of attitude structure……….……...…..…..20
Figure 6: Evaluation of the six varieties of English by 102 Nigerian respondents…...…..47
List of Tables
Table 1: SSCE in Nigeria showing performance in English……….……..…..…14Table 2: Statistical information about participants………...………....34
Table 3: Background information about speakers………...………...….35
Table 4: Cronbach Alpha coefficients for all speakers………...…….37
Table 5: Mean ratings (and standard deviation) of all attributes for 102 Nigerian respondents....40
Table 6: Test of within-subjects effects……….………...……...…41
Table 7: Post hoc test-pairwise comparisons of the varieties………….……….…...…..42
Table 8a: Pairwise comparisons - status dimension……….………...………43
Table 8b: Pairwise comparisons - solidarity dimension………...…………..45
Table 9: Ranking of the varieties………49
Table 10: Identification of the speaker’s country of origin by Nigerian respondents ...…58
Table 11: Categorising the varieties……….…….………...………….59
List of Abbreviations
AME American English AU Australian English AUE Australian English
ELC English Language Complex ELF English as Lingua Franca ELT English Language Teaching ENE Educated Nigerian English
GA General American
IVOR Ivorian English L1 First language
MGT Matched Guise Technique NIGacr Nigerian Acrolectal NIGmes Nigerian Mesolectal
NP Nigerian Pidgin
NZE New Zealand English
RP British Received Pronunciation SAE Standard American English SBE Standard British English SCOT Scottish English
SEE Southern English English SEI Speech Evaluation Instrument
SSCE Senior Secondary Certificate Examination VGT Verbal Guise Technique
WAEC West African Examination Council
WASSCE West African Senior Secondary Certificate Examination
WE World Englishes
Acknowledgement
I am grateful to my thesis supervisor, Professor Mikko Laitinen, for his guidance and support throughout my research.
I am also indebted to my colleagues, Kasia Wiśniewska and Deniz Özel, who provided valuable feedback and assistance during the writing process. I would also like to thank my sisters, Opeyemi Olatoye and Yetunde Fasunon, for their assistance with data collection.
1
1. Introduction
Africa is characterised by linguistic diversity and varied language contact situations within delimited geographical spaces. Of the 16 multilingual and multi-ethnic countries in West Africa, only Liberia, Sierra Leone, Gambia, Ghana and Nigeria have English as an official language.
Figure 1: Map of West Africa1 (adapted from Encyclopaedia Britannica)
In Nigeria, English is used for intranational and international communication, especially in the workplace. Business with English-speaking clients from other countries is conducted in the workplace and intercultural communication skills are often required to ensure local and international intelligibility. In urban areas, the axiomatic role of English language in both formal and less formal sectors cannot be overlooked. Given the numerous English medium schools, popular Nigerian movie industry “Nollywood”, music industry, radio and television broadcasts, social media, outdoor advertising and local businesses, it seems that a sizeable percentage of the urban population is exposed to English daily.
1 Chad, Cameroon and Equatorial Guinea are in Central Africa, not West Africa, however they are neighbouring countries to Nigeria.
2
Despite an increase in exposure towards local and foreign varieties of English (Igboanusi 2006, Awonusi 1994), a significant decline in the performance of Nigerian secondary school and university students in English language examinations has been reported over the years
(Akeredolu-Ale 2007). It is also not uncommon for companies to organise English language evaluation tests for prospective graduate employees before hiring. These tests are mostly based on the Standard British variety, which is the prescribed variety chiefly employed in the Nigerian educational system.
In the Nigerian society, proficiency in English is considered a symbol of good education.
Users of English are often judged by the way they speak. It is also not uncommon to hear jokes about different ethnic groups such as the Hausa, Igbo and Yoruba and their stereotyped accents.
Many Nigerians migrate from rural areas to urban areas to find employment or undertake courses of study; these rural dwellers are sometimes ridiculed for their pronunciation by urban dwellers. These occurrences suggest that certain varieties of English are perceived quite differently, in terms of status and prestige.
The purpose of this study is to investigate the attitudes of educated Nigerian adults towards local and foreign varieties of English. The target group is considered the core of the workforce, highly educated, as well as more exposed to popular media and language changes.
The target respondents in this study are graduates of tertiary institutions and they can be classified as competent users of standard English in Nigeria.
In sociolinguistics, numerous studies of World Englishes have been carried out with the aim of understanding language attitudes and linguistic preferences (Ladegaard 1998, Hiraga 2005, McKenzie 2008, Bernaisch 2012, Chien 2014, Chan 2016). To measure the degree of local acceptability and international intelligibility, attitudes towards a linguistic variety and its speakers can be evaluated using a direct or indirect approach.
Under the indirect approach, one of the methods used to obtain attitudinal evaluations is the matched guise technique introduced by Lambert et al. (1960). Using this method, the same
3
set of speakers are recorded multiple times speaking different languages, language varieties or accents. The respondents are subsequently asked to evaluate the speakers using a scale of personality traits, without being aware that the same speaker is being listened to under different guises. The verbal guise technique is slightly modified to accommodate the use of different speakers, especially when the accents or language varieties cannot be accurately produced by the same person(s). Moreover, with the global spread and increased heterogeneity of English (Mair 2013), it is often impossible to find one person that can produce authentic varieties of English from very different regions of the world.
In Nigeria, existing attitudinal research has mostly explored attitudes towards indigenous languages competing with one another in various domains, such as Ikwerre and Nigerian Pidgin (Ihemere 2007), Ogu and Yoruba (Onadipe-Shalom 2014). Igboanusi (2008) investigated the attitudes of Nigerians, towards mother tongue-based bilingual education in indigenous languages and English, while Adegbite (2003) studied the attitudes of university students towards the use of indigenous languages and English for different communicative functions.
However, not a lot of empirical research has been done on attitudes of Nigerians towards local and foreign varieties of English (Igboanusi 2006, 2003, Williams 1983).
Similar to previous research (McKenzie 2008, Chan 2016), the verbal guise technique was adopted in this present quantitative study for data collection. Data were obtained from 102 respondents during a three week-period. Audio recordings of speakers of different varieties of English were embedded in an online questionnaire. Participants in the study were required to listen to seven interviews of sport athletes and subsequently evaluate the speakers using a six- point semantic differential scale with nine bipolar traits.
To investigate the attitudes of educated Nigerians towards local and foreign varieties of English, the following research questions were formed:
4
1. What are the attitudes of educated Nigerian adults towards different varieties of English from the Inner Circle, Outer Circle and Expanding Circle?
2. Is there a significant difference between the attitudes of Nigerians towards standard and non-standard varieties from the Outer Circle (acrolectal and mesolectal varieties in Nigeria) and standard and non-standard varieties from the Inner Circle (Standard British and Scottish in the United Kingdom)?
3. Which of these varieties are correctly identified by the target group?
In chapter 2, the use of English in Nigeria and the spread of English worldwide is discussed. Chapter 3 presents an overview of previous research on language attitudes. In Chapters 4 and 5, the research method and results of the data analysis are presented and discussed, while recommendations are made in the concluding chapter.
5
2. Varieties of English
2.1 English worldwide
English like any other language is not spoken uniformly around the world. According to Mair (2013), “the more English spreads globally, the more heterogeneous it becomes internally”. As a world language with millions of speakers from different cultures, English has begotten numerous varieties. For example, English in Africa has been categorised into three regional varieties: East African English (EAfrE), Black Southern African English (BSAfrE) and West African English (WAfrE) (Mutonya 2008).
While a language variety refers to a shared linguistic system within a territorial speech community (Mair 2013), localised forms of English worldwide have been referred to by various terms such as ‘New Englishes’, International Englishes, and most popularly ‘World Englishes’.
The World Englishes paradigm (WE) has been increasingly studied in the last four decades and several researchers have attempted to describe the global use of English, using various models, based on how it is acquired or transplanted (Bolton 2012, Schneider 2003, Kachru 1992).
Quirk (1985) categorises English as a global language into three categories: English as a native language (ENL) situated primarily in countries, where English is the language used in all domains. Secondly, English as a Second language (ESL) situated primarily in countries, where English is not only acquired as a second language, but also in use in many domains. Thirdly, English as a foreign language (EFL) situated primarily in countries, where English is primarily a foreign language often used for international communication, but not for intranational
communication.
Kachru (1985,1992) proposed a model in which Englishes worldwide are divided into three circles. According to Kachru (1985), speakers of English worldwide can be divided into three concentric circles: the Inner Circle, the Outer Circle and the Expanding Circle. The Inner
6
Circle describes countries where English is chiefly spoken as a first language by a majority such as UK, USA and Australia. The Outer Circle describes countries, where English is
institutionalised as an official language and widely spoken as a second language (L2) such as Nigeria, Kenya, India, Singapore, Malaysia etc. The Expanding Circle describes countries, where English is a foreign language with no official functions such as China, Japan, Germany, Côte d’Ivoire etc.
Figure 2: Kachru’s concentric circles (Adapted from Kachru 1997, 1985)
These three circles have also been described as having norm-providing, norm-developing and norm dependent functions respectively. The Inner Circle varieties provide the widely
accepted standards such as SBE (Standard British English) also referred to as Received Pronunciation (RP) and SAE (Standard American English) which are often the designated models used for pedagogical purposes worldwide. The Outer Circle varieties, found mostly in former British colonies in Asia and Africa, are developing endonormative standards and the
The Expanding Circle e.g.
China Caribbean countries Egypt
Indonesia Israel Japan
Korea Saudi Arabia Nepal
South America Taiwan CIS
The Outer Circle e.g.
Bangladesh Ghana India Malaysia Nigeria Kenya Philippines Singapore Pakistan Sri Lanka Tanzania Zambia
The Inner Circle USA UK Canada Australia New Zealand
7
Expanding Circle is chiefly dependent on standards as prescribed by the Inner Circle varieties for international communication.
While the three circles model has been criticised as inadequate in explaining the often- complex linguistic variation and sociolinguistic reality present in many countries (Bruthiaux 2003, Mesthrie and Bhatt 2008), for many researchers it provides a simplistic model description of varieties of English worldwide, from which further analysis can be carried out.
Schneider’s (2003) dynamic model of world Englishes identifies five stages in the evolution of post-colonial varieties of English in the world. At the foundation stage, initial contact between the native speakers and indigenous community is established. This is followed by the exonormative stage, wherein English is acquired by an elite minority of the indigenous population. The nativization stage is considered most crucial as the substrate influence from the indigenous languages begins to shapen the distinctiveness of the new variety, often characterised by lexico-semantic differentiation and ownership of English. The fourth stage is the
endonormative stabilisation, whereby the new variety becomes increasingly accepted and codified. This stage is often followed by the differentiation stage, whereby distinct forms of the newly established variety become easily identifiable within the country’s borders.
McArthur (2003) suggests the English Language Complex (ELC) as a term which encompasses all varieties of English. According to Mesthrie and Bhatt (2008), the ELC consists of the metropolitan standards and colonial standards of English, regional and social dialects of English, Pidgin Englishes, Creole Englishes, English as a second language, English as a foreign language, immigrant Englishes, language-shift Englishes, jargon Englishes and hybrid Englishes.
Language-shift Englishes occur in speech communities, where the primary language has been replaced by English. This variety of English can evolve into a social dialect, associated with a particular social group. Jargon Englishes are basic, unstable varieties that can evolve into a Pidgin, while Hybrid Englishes develop as a result of language contact between an indigenous language and English, characterised by frequent code mixing.
8
To complement the ELC alongside previous models, Mair (2013) attempts to describe standard and non-standard varieties of English in a dynamic multilingual world via a model consisting of four hierarchical layers chiefly based on global influence in different spheres such as media, music, literature, education. Mair’s (2013) model depicts the power differentials between varieties of English in an increasingly globalised world, where English has become a pluricentric language. Thus, varieties with larger number of users and institutional support are ranked higher. For example, Standard British English (SBE) has strong institutional support and has been the norm provider for many post-colonial varieties, however, Standard American English (SAE) equally has strong institutional support and even more speakers, thus SAE is considered the hyper-central variety.
“World System of Standard and Non-Standard Englishes”
• Hyper-central variety / “hub” of the World System of Englishes: Standard American English
• Super-central varieties:
1. Standard: British English, Australian English, South African English, Nigerian English, Indian English, and a very small number of others
2. Non-Standard: AAVE, Jamaican Creole, popular London, and a very small number of others
(+ Domain-specific ELF uses: science, business, international law, etc.)
• Central varieties:
1. Standard: Irish English, Scottish (Standard) English, Jamaican English, Ghanaian English, Kenyan English, Sri Lankan English, Pakistani English, New Zealand English, and others
2. Non-Standard: Northern English urban koinés, US Southern, etc.
• Peripheral varieties:
1. Standard: Maltese English, St. Kitts English, Cameroonian English, Papua New Guinea English, and others.
2. Non-Standard: all traditionally rural based non-standard dialects, plus a large number of colonial varieties including pidgins and creoles.
Figure 3: A World System of Standard and Non-Standard Englishes (from Mair 2013)
9
At the hub of the model is Standard American English which has far reaching influence across continents in virtually all English-using domains. This is followed by standard and non- standard super-central varieties. The third layer has standard and non-standard central varieties, while the fourth layer has peripheral varieties.
Although Mair’s (2013) depiction of Standard American English as the hyper-central variety/ hub of world Englishes has not been proven to be untrue, theoretically, there is no standard accent for English. It is widely accepted that standard English can be spoken with any social or regional accent, although it was synonymous with British English spoken with the
‘Received Pronunciation’ (RP) accent in former British colonies such as Nigeria.
This present study is an attempt to shed more light on the acceptability of standard and non-standard varieties of English to educated Nigerians. In order to explore the attitudes of educated Nigerian adults towards standard and non-standard varieties of English in the Inner Circle and Outer Circle, it is important to understand the concept of standard and non-standard varieties of English.
The concept of standard vs non-standard varieties of English has been described in terms of mainly written forms vs mainly spoken forms. However, it can be argued that the relationship between standard and non-standard varieties is more complex. A standard variety is one which has undergone the process of standardisation. Trudgill (1999) states that
standardisation consists of language determination, codification and stabilisation. This can be interpreted as choosing a particular language variety for a specific purpose, creating literary works, grammar books and dictionaries for the variety and the wide-spread acceptance of this codified form within a society (Trudgill 1999).
Historically, Standard English was regarded as the widely accepted variety associated with the upper class in the society (Trudgill 1999). It was considered the prestigious variety and social dialect associated with the relatively well to do. Nowadays, Standard English can be described as the codified variety found in grammar books, dictionaries etc. which is prescribed as an ideal
10
model for pedagogical purposes and use in prestigious domains such as academic writing, broadcasting, official administration etc. as well as widely accepted by the ruling elite/highly educated (Trudgill and Hannah 2008). Thus, a non-standard variety of English is one which is substantially different from the accepted standard as described above.
Standard English is a social dialect that can be spoken with different accents. It is widely accepted that a particular accent may be stigmatised in a certain setting and considered
prestigious in another setting. Since a language variety cannot be considered inherently good or bad, favourable or unfavourable attitudes towards a dialect, an accent or speakers of certain varieties are actually a reflection of the social structure within the society (Trudgill 1974). For example, the domination of the USA on the international scene has led to the global recognition and wide acceptability of Standard American English.
Among the numerous varieties of English worldwide, two varieties have a privileged position in terms of standardisation. These native varieties are Standard British English and Standard American English which are widely recognised as the pedagogical models employed in English Language Teaching (ELT) curricula worldwide.
2.2 English in Nigeria
Nigeria is a multilingual country with an estimated population of 186 million (Central Intelligence Agency 2016). The inhabitants speak about 400–500 indigenous languages and numerous dialects, which are often mutually unintelligible.
Adegbija (2004, 1994) classifies the languages spoken in Nigeria into exoglossic,
endoglossic and Nigerian Pidgin. Exoglossic languages include Arabic (mainly in the North and in Islamic schools), French2 (declared in 1996, as Nigeria’s second official language) and most
2 In December 1996, the late Head of State Gen. Sani Abacha pronounced French as Nigeria’s second official language. According to the fourth edition of the National Policy on Education (2004:4e): “the
prescription of French Language in the primary and secondary school curriculum as a second official language”
is listed as one of the innovations and changes to be updated in the previous 1998 edition.
11
importantly English which is the de-facto official language used for interethnic and international communication. Nigerian Pidgin is a lingua franca used in both informal and formal domains, understood by more than 65% of the Nigerian population, but it has no official status in the country. Endoglossic languages consists of 400–500 indigenous languages with numerous dialects, including the three national languages Hausa, Igbo and Yoruba, which are recognised in the Nigerian constitution.
The 400–500 indigenous languages can also be further classified into three: large
(national level), medium (state level) and small languages (local level). Yoruba in the South-West, Hausa in the North and Igbo in the South-East, are the most developed indigenous languages having more than 20 million speakers each, and extensive use within and outside of their respective regions. Another set of 10–15 indigenous languages such as Tiv, Nupe, Idoma, Ijaw, Urhobo, Igala, Kanuri, Edo, Efik, Fulfulde have more than 1 million speakers each and extensive use in specific administrative states (Bamgbose 1994, Adegbija 2004). Some of the remaining indigenous languages have established orthographies, but many are primarily spoken languages with no written form.
During the British colonisation of Nigeria from 1861–1960, the English language was introduced as the language of administration and formal education. English remains the de facto official language in Nigeria with high prestige across public domains, business, education, science and technology, media, entertainment etc., while Yoruba, Hausa and Igbo are officially
recognised as national languages.
The acquisition of the English language in Nigeria can be traced to the establishment of formal education in the 19th century. During this period, English was acquired by a tiny fraction of the Nigerian population, who came into contact with native speakers of English in the Christian missionary schools or manumitted slaves, who had been taught the English language before returning to their homelands (Adetugbo 1978).
*However, scarcity of qualified French teachers and resources often hinders implementation.
12
From 1882 until today, English has retained its dominant position in the Nigerian educational system. Apart from a growing urban minority among the educated middle class and upper-class echelons of the Nigerian society, who have English as home language, an average Nigerian acquires English as a second language via formal education (Adegbite 2003). English is not only learnt as a subject, but it is also a medium of instruction in primary, secondary and tertiary institutions of learning. While a few elitist private schools have adopted the American curriculum, Standard British English (Received Pronunciation) continues to be the textbook variety taught across all levels (primary, secondary, tertiary) of the Nigerian educational system.
However, in reality, a nativized variety of English that is quite different from the target British variety is taught by Nigerian teachers in many schools (Udofot 2011).
Most of the Nigerian teachers, especially in government public schools have learnt English as a second language from non-native speakers and do not speak SBE themselves (Banjo 1993). In the educational system, there is a growing divergence between the prescribed norms in the school syllabus and actual usage by both learners and teachers. The variety of English in actual usage is a domesticated, nativized form with marked features, that is able to express Nigerian customs and ways of life, hereby facilitating intranational communication (Bamgbose 1998, Adegbija 1989).
According to Nigeria’s National Policy on Education drafted in 1977 and revised in 1981,1998, 2004 and 2013, the Nigerian education system is structured into:
• Early child care and development (0–4years)
• Basic education – Ages 5–15years (1 year of Pre-primary, 6 years of Primary and 3 years of Junior Secondary)
• Post-basic education of three years in Senior Secondary schools and technical colleges
• Tertiary education in colleges of education, monotechnics, polytechnics and universities.
The National Policy on Education (2013: 8) further states that:
13
“The medium of instruction in the Primary School shall be the language of immediate environment for the first three years in monolingual communities. During this period, English shall be taught as a subject.
From the fourth year, English shall progressively be used as a medium of instruction and the language of immediate environment and French and Arabic shall be taught as subjects”.
However, unlike public/government schools, this policy remains largely unimplemented in privately owned schools, as the default medium of instruction from the first year in private primary schools is English. Bamgbose, investigating African languages, reports as follows:
” It is amazing to find the large number of English medium private nursery and primary schools offering education to young children. Patronage of such schools comes not only from well-to-do parents but also from many lowly paid blue-collar workers, who believe that they are making a great sacrifice for their children’s future.” (2011:5)
In addition to this, the use of indigenous languages in urban English-medium private nursery and primary schools is discouraged and pupils are sometimes punished for speaking
“vernacular”3. Adegbite (2003:7) further argues that, “the straight use of a foreign language to teach children, who are not yet competent in their mother tongues may result in lack of competence in both the mother tongues and second language, especially when there is lack of adequate exposure to the second language”. This argument further supports the claim that subtractive bilingualism between the L1 and English is on the rise, especially in urban areas (Bamiro 2003). This is in contrast to additive bilingualism, in which English is being acquired via formal education, while the L1 continues to be developed simultaneously.
3 Indigenous languages are pejoratively called “vernacular”.
14
A key indicator of the standard of British English (SBE) in Nigeria is shown in the poor performances of secondary school students at the final year West African Senior Secondary Certificate Examination (WASSCE) in Nigeria. Between 2004 and 20124, the results show that over 600,000 students are automatically disqualified from the university admission process annually, based on their results in English language. Although a score of 40% is equivalent to a pass grade of E8, a minimum grade of C6 (50%)5 in English is required to gain admission into any tertiary institution in Nigeria. Students who score below 50% in English language have to rewrite the examination even if they score above 50% in other key subjects.
Table 1: Secondary School Certificate Examination in Nigeria showing performance in English
Year Number of Students Performance in English (Grades A1–C6 results)
2004 1,035,280 29.59%
2005 1,080,162 25.63%
2006 1,114,065 32.48%
2008 1,274,166 35.02%
2009 1,355,725 41.55%
2010 1,307,745 35.13%
2011 1,514,164 57.25%
2012 1,658,887 58.51%
Source: Adapted from WAEC 2008, Sakiyo and Badau 2015
From independence to today, Nigeria’s language policy has remained somewhat
ambiguous. The only references to language in the 1999 constitution of the Federal Republic of Nigeria are in sections 55 and 97:
“The business of the National Assembly shall be conducted in English, and in Hausa, Igbo and Yoruba when adequate arrangements have been made therefore.
4SSCE Result for year 2007 was not included in the report.
5 WAEC grading system: A1=Excellent (75-100%), B2=Very good (70-74%), B3=Good (65-69%), C4=Credit (60-64%), C5=Credit (55-59%), C6=Credit (50-54%), D7=Pass (45-49%), E8=Pass (40-44%), F9=Fail (0- 39%)
15
The business of a House of Assembly shall be conducted in English, but the House may in addition to English conduct the business of the House in one or more languages spoken in the State as the House may by resolution approve”.
It is important to note that tertiary education is not a requirement for political office in Nigeria. According to Section 65, 2a of the 1999 Nigerian constitution, “a person is qualified for election, if he has been educated up to at least School Certificate level or its equivalent”. This could be interpreted so that proficiency in English is not mandatory for lawmakers in the
National and State house of Assemblies, although English remains the de facto official language.
2.3 English as spoken in Nigeria
Contact with indigenous languages has left its mark on the phonological, syntactical and lexico- semantic structure of the English language in Nigeria. In the last six decades, English as spoken in Nigeria has been categorised into three or four varieties using sociolinguistic parameters such as ethnicity, geographical region, level of education and mother tongue interference (Brosnahan 1958, Adetugbo 1978, Banjo 1993, Udofot 2003, Igboanusi 2006). Bamgbose (1998) further states that, “Nigerian English is nothing more than an abstraction from individual first-language varieties such as Yoruba English, Igbo English, Hausa English, Edo English, etc.”
Banjo (1993) argued that the formal adoption of an indigenous variety of English for pedagogical purposes would be a more realistic target for learners in Nigeria. This standardising variety should be socially acceptable to Nigerians for everyday life and highly intelligible to the international English-speaking community. He further classified Nigerian English into a cline consisting of four varieties based on L1 influence. Speakers of variety one exhibit the highest number of L1 interference in speech, while speakers of variety four exhibit none.
In Banjo’s (1993) terminology, ‘variety one’ is characterised by numerous marked features such as high frequency of L1 interference in pronunciation, simplification of consonant clusters,
16
limited grammar, reduced vocabulary, semantic shifts. It is syllabic-timed instead of stress-timed, generally considered as unacceptable and unintelligible to users of standard varieties of English.
‘Variety two’ is the most widely spoken variety in Nigeria. It is a self-perpetuating variety with slightly fewer marked features than variety one. Speakers of this variety have spent more years in the educational system, possess wider vocabularies but vital phonemic distinctions are seldom made.
‘Variety three’ is mostly void of negative L1 transfers, more phonemic distinctions are made, and suprasegmental features are often constructively employed. Speakers of this variety have spent the longest periods in the educational system and include a growing urban minority, who have English as their home language. ‘Variety four’ is hardly any different from Standard British English (SBE) and describes a tiny fraction of Nigerians for whom SBE is the first language. This variety is typically acquired outside of Nigeria by virtue of birth or living abroad for extended periods of time.
Of the four varieties, Udofot (2003) approximated ‘variety one’ as the non-standard spoken Nigerian English, ‘variety two’ as standard spoken Nigerian English, variety three’ as sophisticated spoken Nigerian English and ‘variety four’ as too foreign to be applicable in the Nigerian sociolinguistic environment. A further examination of the stress patterns of these varieties indicated that the stress patterns in the sophisticated variety were not identical with those of a native speaker of SBE. Additionally, more syllabic-timed utterances, which are
characteristic of Nigerian indigenous languages, occurred in the non-standard variety at the lower end of the cline, than in the sophisticated variety at the higher end of the cline.
Nigerian English is not a homogenous variety of English, but rather a cluster of varieties and all educated Nigerians neither speak the same way nor have equal competence in English (Simo-Bobda 2007, Bamgbose 1998). Users of English in Nigeria have been classified into three groups based on communicative competence: basilectal, mesolectal and acrolectal. This
classification can be described as a continuum from a basilectal variety (least competent user), to
17
a mesolectal variety (competent user), to an acrolectal variety (most competent user) as illustrated in Figure 4 (Okoro 2004). Therefore, a basilectal speaker would have more “Nigerianisms”, local idioms, innovations and characteristic errors in his speech than the acrolectal speaker.
Nigerianisms, in this context, describe the characteristic features that have been transferred from various Nigerian indigenous languages to English, especially at the lexico-semantic and morpho- syntactic levels (Okunrinmeta 2014).
THE NIGERIAN ENGLISH CONTINUUM
COMMON-CORE FEATURES NIGERIANISMS
AND
LOCAL
IDIOMS CHARACTERISTIC ERRORS
LEAST COMPETENT USER --- MOST COMPETENT USER
Figure 4: The Nigerian English Continuum (adapted from Okoro 2004)
Igboanusi (2006) argued that the characteristic features of Nigerian English
pronunciation are more pronounced in the basilectal and mesolectal levels of competence than in the acrolectal level. Some of these features are listed as:
• The realization of the [əː] vowel as [ɔ] in words like ‘first’, ‘nurse’ or [a] in words like
‘sermon’, ‘thirteen’.
• The realization of the [ɪə] vowel as [ia] in words like ‘near’, ‘dear’.
• The realization of the [
ɛː
] vowel as [iɛ
, eɛ
, ia] in words like ‘square’, ‘pear’, ‘their’, ‘chair’• The deletion of plosives before syllable-final /s/ in words like ‘six’, ‘extra’.
• The deletion of post-vocalic /l/ in coda position in words like ‘talk’, ‘children’
18
• The pronunciation of “g” in words with final <ng> like ‘sing’, ‘hang’
• The occurrence of /u/ in final “cle” clusters in words like ‘people’, ‘pinnacle’, ‘bible’
• Monophthongisation of /ai, au/ to /a/ in words like ‘primary’, ‘round’
• Monophthongisation of /eɪ/ to /e/ and /əʊ/ to /o/ in words like ‘face’ and ‘goat’.
• The lack of distinction between long vowels and short vowels /ɪ/, /i:/ in words like bit and beat or /ɒ/, /ɔː/ in words like cot and caught.
It is important to note that there is considerable linguistic variation, even among the acrolectal users of English in Nigeria, as idiolects are influenced by numerous factors in the social environment. Nevertheless, this present study focuses on acrolectal users of English in Nigeria and their attitudes to different varieties of World Englishes.
19
3. Investigating Language Attitudes
3.1 What Are Attitudes?
Attitude is a central concept in social psychology and more recently also in sociolinguistics. The study of language attitudes has been mostly based on two approaches: the mentalist approach and the behaviourist approach (Agheyisi and Fishman 1984).
In the behaviourist approach, attitudes are not considered to be multi-dimensional. They are found in observed behaviours or overt responses to stimuli in a specific social environment.
While this approach may be a straightforward way of analysing language attitudes, it is considered somewhat limited, as one is unable to predict other behaviours by the same individual, outside of the initial setting (Agheyisi and Fishman 1984).
In the mentalist approach, which has been adopted by numerous researchers, attitude has been defined as: “a mental and neural state of readiness, organised through experience, exerting a directive or dynamic influence upon the individual’s response to all objects and situations with which it is related” (Allport 1935 as cited in Eagly and Chaiken 1993). Despite its relative widespread use, the mentalist approach has its limitations. Since the researcher infers attitudes from observable responses or reported data, they may not accurately portray a subject’s attitudes (Agheyisi and Fishman 1984).
As a hypothetical construct, attitudes are intangible entities that are seldom innate. They are learnt directly or indirectly, and the social environment plays a pivotal role in the shaping of individual attitudes towards attitude-objects, whether positive or negative (Garrett 2010). Based on the mentalist approach, the working definition of language attitudes in this study is adapted from Ajzen (2005), which simply states that, “an attitude is a disposition to respond favourably or unfavourably to an object, person, institution or event”. Thus, language attitudes are
considered as evaluative dispositions to a language, language variety or speakers of a language that are inferred from measurable responses.
20
3.2 Components of Attitudes
From the mentalist viewpoint, attitudes are described in terms of three components: affect, behaviour and cognition (Breckler 1984, Baker 1992, Ajzek 2005).
As illustrated in Figure 5, the affective aspect of attitudes involves feelings towards an attitude-object. Verbal statements of affect are often assessed in degrees of favourability or unfavourability. On the other hand, facial expressions and other bodily reactions of the
sympathetic nervous system are assumed to indicate non-verbal responses of affect. Second, the conative or behavioural aspect of attitudes involves both the verbal statements of behaviour and overt actions/behaviours carried out towards the attitude-object. Third, the cognitive aspect of attitudes involves both the verbal statements of beliefs and perceived responses towards the attitude-object.
Figure 5: The tripartite model of attitude structure (from Breckler,1984, after Rosenberg & Hovland,1960)
It is easier to evaluate these attitudinal components from verbal responses than from non-verbal responses. For example, in evaluating attitudes towards the use of Nigerian Pidgin (NP), a Nigerian student could believe that NP is easy to learn (cognitive component), could be enthusiastic about watching movies in NP (affective component) and could be an avid listener of Nigerian Pidgin radio stations (behavioural component). It would be subsequently assumed that this student has a positive attitude towards Nigerian Pidgin. Although these three components of an attitude may not always be in harmony, reactions to a particular speaker or message can be
21
affective or cognitive in nature (Cargile, A. and Giles, H. 1997) and not necessarily expressed in behaviour.
3.3 Measuring Language Attitudes
Several methods have been adopted in the measurement of language attitudes. These methods can be divided into three: societal treatment, direct and indirect methods.
The societal treatment method is seldom employed in large-scale quantitative studies. It involves the analysis of the societal treatment of a language or a language variety via a medium, i.e. linguistic evaluations are not elicited from respondents directly. The researcher infers the attitudes of the respondents from the way a language or language variety is treated in a society by employing ethnographic studies, observational studies, participant observation, discourse analysis amongst others (McKenzie 2010, Garrett 2010). It could involve the analysis of language
policies, media content or linguistic landscapes, which may not provide an accurate account of individual attitudes. Here, linguistic landscape refers to the language(s) visible in the public sphere; this includes public and private signs, billboards, posters, graffiti, advertisements, stickers etc.
The direct approach is used to elicit responses directly from the respondents. The researcher asks open-ended or closed-ended questions about language beliefs, feelings and knowledge via surveys, questionnaires or interviews. Open-ended questions are designed to elicit detailed responses about a particular subject, whereas close-ended questions are designed to elicit short responses or one-word answers. The direct approach has however been criticised for not being too reliable, as informants could be reluctant to reveal certain values for prestige reasons (Agheyisi and Fishman 1970).
The indirect method is employed to elicit attitudes from the respondents, such that the respondent is not overtly conscious of the process. The indirect method is chiefly carried out via
22
the Matched Guise Technique (MGT), which was designed by Wallace Lambert and his colleagues in the 1960s to elicit attitudes towards users of different language varieties (Fasold 1984). In this method, a bilingual speaker makes several recordings under different guises (could be two different languages or standard and non-standard varieties of a language), the
respondents listen to these speech recordings and subsequently evaluate the personal traits of the speakers. The recordings are often interposed such that they appear to belong to different
speakers.
In the MGT, respondents are unaware that the alternate recordings are by the same speaker, and the researcher is able to control both content and certain aspects of voice quality.
Ihemere (2006) investigated the attitudes of a bilingual community towards Nigerian Pidgin English (NPE) and Ikwerre in Southern Nigeria via closed ended questions and the matched guise technique. For the matched guise technique, two university graduates, male and female, read the same passage in both languages (NPE and Ikwerre). 76 participants were asked to evaluate the four varieties of speech using a semantic differential scale with ten traits, while being unaware that two of the guises were produced by the same individual.
Thus, differences in speaker evaluations can be ascribed to a predisposed bias or idea about those languages or language varieties on the listener’s part. However, it is not possible to fully control paralinguistic differences in the VGT. The Verbal Guise Technique (VGT) is a modified version of the Matched Guise Technique (MGT). It is often employed in situations, in which one speaker is unable to produce all language varieties convincingly. While both methods may appear similar initially, the MGT involves the same speaker, using at least two different language varieties or guises in the same recording, while different speakers are employed in the VGT. It is important to note that the MGT has its limitations. It has been criticised as unnatural and unauthentic as it requires a multilingual individual producing several guises, which may not be true representations of the language varieties being investigated (Garrett 2010).
23
In both the MGT and VGT, the measurement of dependent variables along varying dimensions is mostly done via semantic differential scales or Likert scales. Unlike the Likert scale, which is used to determine how strongly the listener agrees or disagrees with certain statements, the semantic differential scale involves the use of bipolar adjectives such as rich- poor, educated-uneducated, friendly-unfriendly etc. indicative of personality, character or physical traits.
Ladegaard (1998) employed the VGT in his study of language attitudes in Denmark. The findings from Ladegaard’s (1998) study suggest that Danish students have a favourable
disposition towards the American variety, but do not consider it an ideal model in terms of pronunciation, to replace the SBE in English as Lingua Franca (ELF) classrooms. His sample consisted of 96 Danish students, who evaluated the speech of five native speakers of Received Pronunciation, Cockney, Scottish, Australian and Standard American via fifteen traits on a semantic differential scale. These traits were classified into three dimensions: status and competence, social attractiveness and linguistic attractiveness.
However, other evaluative dimensions have been identified by various researchers. Zahn and Hopper (1985) designed the comprehensive Speech Evaluation Instrument (SEI). After carrying out an analysis of various traits and dimensions used in previous studies, they concluded that their original 150 questionnaire items could be reduced to 56 items and be chiefly assessed along three evaluative dimensions: social status, superiority and competence. The selection of evaluative dimensions and traits to be used in semantic differential scales can be done in several ways. A researcher could adopt some of the traits in Zahn and Hopper’s SEI, another could carry out pilot tests to determine traits that are most meaningful to the respondents, while another researcher could adopt previously used traits in similar studies alongside new traits derived from a pilot test study (Garrett 2010). To conclude, the use of contextually meaningful traits in the attitudinal evaluation instrument instils confidence in the results.
24
3.4 Previous research on language attitudes towards English
Multilingualism and multiethnicity are characteristic of most African countries. Compared to the Inner and Expanding Circles, relatively fewer studies have explored the multi-dimensionality of language attitudes in the Outer Circle. In his investigation of language attitudes in sub-Saharan Africa, Adegbija (1994) claims that, “deficiency in methodology appears to be the principle weakness of most language-attitude studies hither to carried out in sub-Saharan Africa”. This situation is slowly changing, as several empirical studies from sub-Saharan Africa have been published in academic journals in recent years. For instance, researchers such as (Ihemere 2006, Igboanusi 2008, Akande and Salami 2010, Onadipe-Shalom 2014) employed direct and indirect methods in investigating language attitudes and language shift in Nigeria.
Williams (1983) investigated the attitudes of Nigerian English-language teachers towards different accents of Nigerian English. Speech recordings of 11 educated Nigerians and 3 native speakers from UK, USA and Canada were employed in his study. The educated Nigerians were classified into three groups: heavily accented, mildly accented and near-native. Altogether, 81 respondents from teacher training programmes at two tertiary institutions listened to the recordings and rated the speakers via a 10-point scale with twenty items. The results indicated that the Canadian speaker was rated most favourably, followed by two of the near-native
Nigerian accents, subsequently followed by the British speaker and the American speaker in that order, while the heavily accented Nigerian speakers were rated least favourably. Although the respondents in this study were from about 36 ethnic groups, the Nigerian near-native accent and native accents from Canada, UK and USA were considered more intelligible than the other two varieties. This result suggests that an educated variety of Nigerian English, devoid of marked phonological features, will be internationally intelligible and socially accepted as a standardising norm suitable for pedagogical purposes in Nigeria.
25
While this study appears to be one of the earliest empirical attitudinal studies carried out in Nigeria, it lacked several details. The researcher neither listed the attributes nor provided details of the evaluative dimensions used in the study.
Muthwii and Kiokio (2010) explored the attitudes of 210 Kenyans towards ethnically marked Kenyan English, standard Kenyan English and Native Speaker English from the Inner Circle in the workplace, media and education. The participants were further classified based on their ethnicity, level of education (primary, secondary, university) and place of residence
(rural/urban). They reported that respondents were readily able to identify the different varieties of English. Ethnically marked Kenyan English was only preferred by a fraction of rural residents, standard Kenyan English (non-ethnically marked) was the widely preferred variety for use in the workplace, education and media, while only a minority of university graduates in urban areas considered the Native Speaker English to be ideal for both intranational and international
communication. Overall, the variety of English considered most beautiful/admirable by Kenyans was the Native Speaker variety. However, if audio recordings of these three varieties were not made available to the participants, it is rather doubtful that all 210 participants could clearly distinguish between these varieties and evaluate them via a paper questionnaire.
Igboanusi (2006) studied attitudes towards Ethnic English pronunciation in Nigeria. He reports the stigmatisation and public ridicule, which basilectal and mesolectal users of English in Nigeria encounter because of mother tongue interference. Among Igbo speakers, some of these features include the alternation of /l/ and /r/, such that words like ‘dangerous’, ‘clinic’ are pronounced as ‘dangelous’, ‘crinic’. Among Yoruba speakers, some of these features are the substitution of /f/ for /v/ or /ʃ/ for /ʧ/ such that words like ‘vendor’, ‘champion’ are
pronounced as ‘fendor’, ‘shampion’. Among Hausa speakers, some of these features include the alternation of /f/ and /p/ such that words like ‘people’, ‘difficult’ are pronounced as ‘feofle’, dippicult’. These stigmatised features have been categorised as shibboleths by which one can easily identify the user’s ethnicity.
26
Although this study makes a vital contribution to the existing body of research, the lack of empirical data makes it difficult to compare findings with similar studies.
Other researchers have asserted that the linguistic stereotyping of users of certain non- standard varieties makes them receive consistent negative evaluations, even in terms of social attractiveness (Bishop, Coupland and Garrett 2005). Apart from this, the speaker’s level of accentedness also plays a significant role in speech evaluation and language attitude studies as mild accents are rated more positively than heavy accents (Giles and Coupland 1991).
Chan (2016) investigated the attitudes of 386 Cantonese-speaking students (university, JSS and SSS) in Hong-Kong towards seven varieties of English selected from the Inner Circle (Britain, USA, Australia), Outer Circle (Hong Kong, India, Philippines) and Expanding Circle (China). The Verbal Guise Technique (VGT) was used to elicit the students’ attitudes along
‘status’ and ‘solidarity’ dimensions. Using voice recordings of different female speakers from the Speech Accent Archive, the students were asked to evaluate the speakers on a six-point
semantic-differential scale with eight bipolar adjectives (intelligent, rich, educated, successful and considerate, kind, friendly, honest) and subsequently guess where the speaker was from.
He reported that the students not only identified the speakers from Hong Kong, UK and USA more correctly than the others, thus indicating their familiarity with the local accent, RP and GA, but they also evaluated the Inner Circle varieties as more intelligible than others.
Inner Circle varieties were rated highest along the status (RP, GA, AUS) and solidarity
dimensions (AUS, RP, GA) in different orders, but the Hong Kong variety was rated surprisingly low along both dimensions. This rating suggests that Cantonese is perceived as the language of solidarity in Hong Kong, not the Hong Kong variety of English.
A similar attitudinal study (Ahmed et al. 2014) carried out in Malaysia concluded that Malaysian students preferred the English accents of non-native lecturers (Arabic, Chinese, Malay, Indian) to those of native lecturers from the USA and UK. The results thus support the social identity theory in which positive attitudes towards a variety of English or accent are an indicator
27
of an ingroup bias. In addition to this, the speaker identification task indicated that the Malaysian students were more familiar with American English than British English.
Bayard et al. (2001) reported the attitudinal evaluations of New Zealand English (NZE), Australian English (AUE), American English (AME) and Southern English English (SEE) by over 400 students from three countries (USA, Australia, New Zealand). In their multinational study, the participants listened to eight tape recordings of four male and female speakers of NZE, AUE, AME and SEE and subsequently evaluated the speakers using a six-point semantic differential scale with thirteen traits. These traits were evaluated along four dimensions: power, competence, status and solidarity and the study also included a speaker identification task. The results from the speaker identification task indicated that the American variety was the most accurately identified across the three groups of Australian, New Zealander and American participants. On the other hand, the male and female AME varieties received the highest ratings along the solidarity dimension, while the male and female SEE varieties received the second lowest ratings after the male NZE variety from the three groups. Bayard et al. (2001) further concluded that the global domination of American media is becoming more evident in attitudes to accents.
McKenzie (2008) argues that standard varieties of English tend to receive more favourable ratings along the status dimension, while non-standard varieties of English receive more favourable ratings along the solidarity dimension. In his investigation of Japanese students’
attitudes towards standard and non-standard varieties of English from the Inner and Expanding Circles (UK–Glasgow vernacular and Glasgow Standard English, USA–Southern US English Alabama and Midwest US English, Japan–Moderately accented and Heavily accented), he categorises eight bipolar traits into two dimensions: competence(status) and social attractiveness (solidarity).
His results demonstrate that the standard and non-standard American varieties were most positively rated in terms of competence followed by the UK varieties, while the Japanese
28
varieties received the lowest ratings. In terms of social attractiveness, the converse is the case.
The three non-standard varieties are rated most positively followed by the three standard
varieties in the following order: Japan, UK, USA. Further analysis also indicated that the varieties from the Inner Circle were more positively evaluated, when the speaker’s country of origin was correctly identified by the participants.
In his study of Americanisms in Nigerian English, Igboanusi (2003) claimed that despite the wide acceptance of American English in the Nigerian society, British English is still widely preferred most especially in the school curriculum. Despite the overlooked lexical differences between British and American English, educated Nigerians’ increased exposure to American English via popular media such as movies, pop music, TV shows etc. plays a significant role in the overall positive attitude towards American English. He further suggested that learners of English language in Nigeria should be adequately exposed to British English, American English and Nigerian English to ensure both local and international intelligibility.
Hartmann (2014) investigates the attitudes of Nigerians and Kenyans in the UK and Germany towards seven different varieties of English (UK, USA, Jamaica, Nigeria, Kenya, China and Germany) along two different levels of formality. Using online newscasts (formal) and broadcast interviews with female sport athletes (informal), Hartmann employs a 4-point Likert rating scale with nine traits alongside conversational interviews with participants. Preliminary results of the study indicated that overall attitudinal evaluations of the seven varieties on the formal level were positive; this suggests a higher tolerance/acceptance of different accents of English by Africans in diaspora. In addition to this, the American and British varieties were rated higher along the competence (status) dimension than along the solidarity dimension. However, responses showed a clear preference for British English over American English. The home country varieties (Kenya/Nigeria) were also rated high along both dimensions.
Previous research suggests that a native variety continues to be preferred over local varieties of English in Outer and Expanding Circle countries, especially along the
29
status/competence dimension. However, there seems to be an ongoing shift whereby the British variety of English is being replaced by the American variety in terms of status and solidarity.
Apart from Hartmann’s (2014) study, results from other researchers suggested that users of English are more familiar with American varieties than British varieties nowadays.
Results from several investigations support the hypothesis that speakers of standard varieties are associated with higher socioeconomic status and evaluated positively in terms of status and competence. On the other hand, speakers of non-standard varieties are associated with lower socioeconomic status and evaluated positively in terms of social attractiveness (McKenzie 2008).
In addition to this, users of English often have very different attitudes towards standard and non-standard varieties of English in their respective societies. Since very little research based on empirical data has been published about attitudes towards varieties of English in multilingual Nigeria, this study attempts to investigate the attitudes of acrolectal users of English towards acrolectal and mesolectal varieties of Nigerian English, as well as varieties from Inner Circle and Expanding Circle countries using online interviews of male athletes.
Unlike many attitudinal studies where university students are the target respondents, this study focuses on Nigerian adults, who have already graduated from tertiary institutions and currently use English language for communication in the workplace.
30
4. Data and Methodology
This study investigates the attitudes of educated Nigerian adults towards different varieties of English from the Inner, Outer and Expanding Circles, based on Kachru’s three circles model of world Englishes (Kachru 1985, 1992,1997). More specifically, it focuses on acrolectal users of English in Nigeria and their perceptions of six varieties of English from: England (Standard British English), Scotland, and USA (General American) in the Inner Circle, Nigeria (acrolectal and mesolectal varieties) in the Outer Circle and Côte d’Ivoire in the Expanding Circle. Since Nigeria is surrounded by French-speaking African countries, choosing a variety from a French speaking West African country like Côte d’Ivoire was considered apt.
4.1 Research Instrument
In this present study, data were collected via the verbal guise technique (see Appendix I), since it was impossible to find one speaker who could produce the selected varieties of English
convincingly. An online questionnaire was employed for data collection because it is an inexpensive means of obtaining both qualitative and quantitative data within a limited time frame. The link to the questionnaire, with the embedded audio files, could also be easily shared via social media as well as forwarded to friends and former colleagues in Nigeria. Moreover, it would have been extremely difficult to gather a sizeable number of the target respondents together in one place and have them listen to the audio recordings for the verbal guise task.
To answer the research questions, seven informal interviews of sport athletes were used for the verbal guise task. The seven athletes were from Jamaica, England, Nigeria, USA, Scotland and Côte d’Ivoire. The audio recordings were obtained from two YouTube channels: ‘Making of Champs’ and ‘Athletics Africa’. Written permission was obtained from both channels before downloading the video interviews, from which the audio recordings were extracted for use in
31
this study (see Appendix III and IV). The edited audio recordings were 33 seconds each except for the ‘SBE’ and ‘General American’ speakers, which were 31 seconds and 34 seconds
respectively.
The audio recordings were carefully selected to minimise variables such as background noise and interruptions, ensure clarity of speech, while ascertaining that certain phonological features of these varieties were distinguishable. Subsequently, the recordings were embedded in an online questionnaire hosted by www.soscisurvey.de 6and data was collected over a period of three weeks.
The online questionnaire had three sections (see appendix I). The first section, which had 12 questions, was used to obtain sociodemographic information about the participants such as age, sex, location, occupation, ethnicity, mother tongue and educational level.
The second section had the verbal guise task, in which participants had to listen to the seven varieties of English and evaluate them via nine attributes on a semantic differential scale with values ranging from 1(least Favourable) – 6(most favourable). A scale with an even number of values was chosen so that respondents would be more decisive and not merely opt for a middle value.
The attributes adopted in this study were adapted from Chan (2016), Westphal (2015) and Hartmann (2014). From these three studies, a total of 23 attributes were obtained: confident, authentic, clear, pleasant, cool, friendly, correct, arrogant, refined, professional, intelligent, rich, educated, successful, considerate, kind, honest, standard, proper, suitable, natural, modest and not twanging.
A mini pilot study with five Nigerians was carried out and participants were asked to describe the speakers, from the list of attributes. Two of the participants considered “pleasant”
and “friendly” synonymous, “not twanging” was replaced with “articulate” and certain attributes
6 www.soscisurvey.de is a website that hosts questionnaires and surveys for both academic and non- academic research.
