The Realisations of English Vowels by Native Speakers of Kanuri and Ngizim Languages: Implication for EFL Pronunciation

Cite this article as: Yusuf, S. A., & Muhammad-Gombe, U. (2026). The realisations of English vowels by native speakers of Kanuri and Ngizim languages: Implication for EFL pronunciation. Sokoto Journal of Linguistics and Communication Studies (SOJOLICS), 2(1), 535-548. https://doi.org/10.36349/sojolics.2026.v02i01.044

By

SalisuAlhaji Yusuf

Department of Linguistics& African Languages

University of Abuja, Nigeria

Salisu.yusuf@uniabuja.edu.ng

+2348066124911

&

Umar Muhammad-Gombe, PhD

Department of English& Literary Studies, Yobe State University

Damaturu, Nigeria

umarmgombe@gmail.com

Abstract

This study examines the realisation of English vowel sounds by native speakers of Kanuri and Ngizim, with emphasis on short and long monophthongs and selected diphthongs. Data were collected from fifteen native speakers of Kanuri in Bama, Borno State, and another set of fifteen native speakers of Ngizim in Potiskum, Yobe State,through controlled speech tasks involving wordlists and reading passages. The respondents’ pronunciations were analysed in comparison with Received Pronunciation (RP). The findings show systematic vowel substitutions in both varieties, largely influenced by first language phonological structure and English orthography. The English KIT vowel /ɪ/ was realised as [iː], [e], and [ə] in Kanuri-English, while Ngizim speakers produced mainly [iː] and [e]. Respondents from both groups showed difficulty indistinguishing vowel length contrasts, frequently shortening long vowels such as /iː/ to [i]. The schwa /ə/ was highly variable, with Kanuri speakers producing [ɔ], [e], [a], [aː], and [u], and Ngizim speakers producing [e], [a], [aː], [ɔ], and [u], often depending on spelling. The STRUT vowel /ʌ/ was consistently realised as [ɔ] in both varieties. Long vowels such as /ɜː/ and /ɔː/ were also replaced with more familiar vowel qualities, while /uː/ was frequently shortened. Diphthongs were identified as the most problematic category, with both groups reducing /eɪ/ and /əʊ/ to monophthongs, particularly in Ngizim-English where diphthong inventory is limited. Overall, the study demonstrates that vowel deviations in Kanuri and Ngizim English are shaped by phonological transfer, vowel length neutralisation and orthographic influence.

Keywords: Kanuri, Ngizim, L1 influence, Spelling influence, L2 learning, Vowels

1.0       Introduction

The study of second-language (L2) vowel production has long been a central concern in applied linguistics, phonetics and sociophonetics, particularly in multilingual contexts where first-language (L1) phonological systems influence L2 acquisition. English, with its large and phonetically complex vowel system comprising up to twenty vowel phonemes—including monophthongs, diphthongs and contrastive vowel length (Roach, 2003)—presents significant challenges for speakers of languages with smaller vowel inventories. Among such languages are KanuriandNgizim, both spoken in northeastern Nigeria. Kanuri is characterised by a six-vowel system (/i, e, a, u, o, ə/) with limited contrastive length in everyday speech (Bulakarima&Shettima, 2012), whereas Ngizim exhibits ten vowel phonemes, including four short, four long and two diphthongal vowels, with length contrasts playing a phonemic role (Schuh, 1981). Unlike the English vowel system, most Nigerian languages have seven vowels or even fewer vowels (Adewole 2016). Prior research indicates that speakers of languages with smaller vowel inventories tend to substitute or merge L2 vowel contrasts based on the closest L1 equivalents, resulting in systematic deviations in vowel quality, length and diphthong realisation (Flege, 1995). While a considerable body of work exists on English vowel acquisition by speakers of major African languages such as Yoruba and Hausa, there is limited empirical evidence on the production of English vowels by Kanuri and Ngizim speakers. This gap is particularly significant in forensic linguistics, sociophonetics and ESL pedagogy, where accurate vowel production has implications for intelligibility, accent analysis and language teaching.

The present study seeks to address this gap by providing a comparative analysis of English vowel production by native speakers of Kanuri and Ngizim, with particular attention to L1 transfer effects, vowel length and diphthong articulation. The study investigates how the size and structure of L1 vowel inventoriesinfluence L2 English vowel realisation, contributing to broader theories of cross-linguistic influence and L2 phonological acquisition.The inconsistency between English phonemic representations and orthographic forms presents a significant challenge for second language (L2) learners, particularly those whose first languages have highly transparent or phonemic orthographies, such as Kanuri and Ngizim. Unlike languages with relatively consistent grapheme–phoneme correspondence, English orthography is characterised by considerable irregularity, where identical graphemes may represent multiple sounds and identical sounds may be represented by different graphemes (Cook, 2004; Crystal, 2003). For learners from phonemically transparent language backgrounds, this mismatch can lead to orthography-driven pronunciation strategies, whereby written forms are interpreted through L1 sound–symbol expectations. Such reliance on grapheme-based decoding often results in systematic mispronunciations, especially in vowel production (Kenworthy, 1987).

The relationship between sound and spelling has long been debated in second language pronunciation research. Some scholars argue that English orthography demonstrates greater regularity than is commonly assumed and that grapheme–phoneme correspondences can assist learners in predicting pronunciation. For example, Lado (1982) maintains that systematic spelling patterns may facilitate phonological learning, while Oswalt (1973) suggests that the phonemic value of graphemes is often predictable from their phonological environment. From this perspective, written input can enhance learners’ awareness of sound–symbol relationships and support pronunciation development. Brown (1987) further emphasises that learners must develop the ability to interpret orthographic patterns in order to anticipate correct sound production.

1.1.      The Vowel Systems of English

English has a rich vowel inventory comprising up to twenty vowel phonemes, as summarised in Table 1 (Roach, 2003). The system includes monophthongs, both short (/ɪ, e, æ, ʌ, ə, ɒ, ʊ/) and long (/iː, uː, ɑː, ɔː, ɜː/), alongside eight diphthongs (/eɪ, aɪ, ɔɪ, aʊ, əʊ, ɪə, eə, ʊə/).

1.1.1.   The English pure vowel system (monophthongs)

The English pure vowel system contains twelve monophthongs, comprising seven short vowels /ɪ, e, æ, ʌ, ə ɒ, ʊ/ and five long vowels /iː, uː, ɑː, ɔː, ɜː/ (Gimson,1962).

1.1.2.   The English diphthongs

There are diphthong sounds in English /eɪ, aɪ, ɔɪ, aʊ, əʊ, ɪə, eə, ʊə/ (Roach, 2009).

Table 1. Summary of the English vowels

Short monophthongs	æ ʊ ə ɪ ʌ ɒ e
Long monophthongs	ɪ: ɑ: ɜ: ʊ: ɒ:
Diphthongs	ɪə  eə  ʊə  eɪ  aɪ  ɒɪ aʊəʊ

 

1.2.      The Kanuri Vowel System

1.2.1.   The Kanuri pure vowel system

Kanuri contains a set of six vowel phonemes: /i, e, ə, a, u, o/ (Bulakarima&Shettima, 2012), but the languagedoes not contrast vowel length in normal speech. However, early studies (Cyffer& Hutchison, 1979) describe Kanuri as having a set of six short vowels /i, e, a, u, o, ə/, with no contrastive long vowels. Jarrett (1978) argued that Kanuri historically operated with two parallel series of short and long vowels (e.g., /i ~ ii, e ~ ee, a ~ aa, u ~ uu, o ~ oo/), although this distinction has been lost in modern spoken usage. More recent descriptions such as (Bulakarima&Shettima, 2012) confirm that vowel length is no longer phonemically contrastive in everyday speech. Long vowels and diphthongs remain prosodically significant in poetic registers, idiophonesand Arabic loanwords, often marked by the semi-vowels ʾalif, wāw, and yāʾ (Modu, 2012). This indicates that while Kanuri exhibits a smaller inventory in normal speech, vowel length and diphthongs are preserved in specialised contexts.

1.2.2.   The Kanuri diphthongs

Kanuri researchers such as Cyffer& Hutchison (1979)had argued that there are no diphthongs in Kanuri, and the combination of two vowels in some Kanuri words is a result of intervocalic consonant elision in the language. However, recently, poetic traditions showed that long vowels and diphthongs remain active in Kanuri. Modu (2012) demonstrates that vowels are lengthened in rhyme positions, especially line-final open syllables, and Arabic loanwords introduce systematic lengthening through the semi-vowels ʾalif, wāw, and yāʾ. Diphthongs also appear in recitation through syllable reduction, idiophones and umlaut (Guda). Thus, while everyday Kanuri is best described with short monophthongs, the poetic register preserves long and diphthongal vowels as prosodically significant features.

1.3.      The Ngizim Vowel System

According to Schuh (1981), Ngizim has ten vowel phonemes, consisting of eight monophthongs and two diphthongs, as shown in Table 2. The monophthongs are further categorised into four short vowels (/a, i, o, u/) and four long vowels (/aa, ii, ee, uu/). Long and short vowels are contrastive in all positions, as illustrated by kaawa ‘playing’ versus kawa ‘fondness’ and adan ‘crying’ versus aadan ‘south’. Schuh (1979) earlier identified the two diphthongs /ai/ and /au/ in Ngizim.  Long vowels in Ngizim have positional constraints: /ii/ and /uu/ occur only in medial positions, /aa/ is rare in final position, and long vowels do not occur in closed syllables, with a few exceptions such as kaaktulu ‘measure’ (Schuh 1978a: 255). With the exception of two words beginning in long /ii/, long /ii/ and /uu/ are found only in medial position. Neither occurs in word-final position, and long /aa/ is also rare in this position (Schuh 1978a, p. 255).Ngizim’s two diphthongs, /ai/ and /au/, involve a glide from one vowel quality to another, and are limited in number (Lawani, 2009).

Table 2. The Ngizim vowels

Ngizimmonophthongs (short vowel)	/a, i, o, u/ (Schuh, 1981)
Ngizimmonophthongs (long vowels)	/ii, ee, aa, uu/ (Schuh, 1981)
Ngizim diphthongs	/ai, au/ (Schuh, 1981)

 

1.4.      Comparison of English, Kanuri and Ngizim Vowels

Vowel systems in languages vary widely in terms of inventory size, vowel length contrast anddiphthong presence. English, Kanuri and Ngizim exhibit significant differences in these aspects, reflecting typological variation between a European language and West African languages. While English possesses a large and phonetically complex vowel inventory, Ngizim and Kanuri have smaller inventories, with Kanuri notably lacking contrastive vowel length in modern speech. Unlike Ngizim, English long vowels have fewer positional restrictions, and diphthongs are abundant, contributing to its phonetic complexity. A cross-linguistic comparison reveals marked differences in vowel inventory size, diphthongpresence and vowel length contrast, as summarised in Table 3.

 

Table 3. Comparative table of vowel inventory of English, Kanuri and Ngizim languages

Feature	English	Kanuri	Ngizim
Number of vowel phonemes	20	6	10
Monophthongs	12	6	8
Diphthongs	8	Limited; poetic/loanwords	2 (/ai, au/)
Vowel length contrast	Yes	No (except poetic/loanwords)	Yes
Short vowels	/ɪ, e, æ, ʌ, ə ɒ, ʊ/	/i, e, a, u, o, ə/	/a, i, o, u/
Long vowels	/iː, uː, ɑː, ɔː, ɜː/	Mostly lost in modern speech	/aa, ii, ee, uu/
Positional constraints	Few	Not contrastive	Long vowels restricted to medial/open syllables
Diphthong quality	/eɪ, aɪ, ɔɪ, aʊ, əʊ, ɪə, eə,ʊə/	Limited to poetry	/ai, au/ only

1.5.      Research Questions

The research is guided by the following three questions: (i) do native Kanuri and Ngizim speakers realize Englishmonophthong vowels in terms of vowel quality and length? (ii) How do native Kanuri and Ngizim speakers produce English diphthongs, and what patterns of L1 transfer are observable?(iii)To what extent do the differences in L1 vowel inventories (Kanuri vs. Ngizim) affect the accuracy of English vowel production among speakers of these languages?

2.         Theoretical framework/model of approach

This study is anchored in Flege's Speech Learning Model (SLM) (Flege, 1995; Flege& Bohn, 2021) and Best's Perceptual Assimilation Model (PAM) (Best, 1995; Best & Tyler, 2007). These complementary models provide a robust framework for explaining how native speakers of Kanuri and Ngizim perceive and realise English vowel sounds.Flege's Speech Learning Model (SLM) proposes that second-language (L2) speech sounds are interpreted through the phonological categories of the learner's first language (L1). According to the model, learners compare incoming L2 sounds with existing L1 categories and may assimilate unfamiliar sounds to the closest native equivalent. When an L2 vowel is perceived as sufficiently similar to an L1 vowel, learners are less likely to establish a new phonetic category, and instead produce the L2 sound using the articulatory properties of the existing L1 category. This process often results in vowel substitution, merger or approximation. Conversely, L2 sounds that are perceived as significantly different from any L1 category are more likely to be acquired as new phonetic categories and produced more accurately (Flege, 1995; Flege& Bohn, 2021).

Complementing SLM, Best's Perceptual Assimilation Model emphasises the perceptual dimension of second-language speech learning. PAM argues that listeners perceive unfamiliar L2 sounds by assimilating them to the closest native-language categories. The model predicts that the degree of difficulty in distinguishing and producing L2 sounds depends on how they are mapped onto L1 phonological categories. Where two English vowels are assimilated to a single Kanuri or Ngizim vowel category, learners are likely to have difficulty distinguishing and accurately producing the contrast. In contrast, English vowels that are perceived as distinct from existing L1 categories are expected to be more readily differentiated and acquired (Best, 1995; Best & Tyler, 2007).

The combined application of SLM and PAM is particularly relevant to the present study because English possesses a considerably larger and more complex vowel inventory than either Kanuri or Ngizim. English contrasts involving tense and lax vowels, central vowels and several diphthongs have no direct equivalents in the vowel systems of Kanuri and Ngizim. Consequently, speakers of these languages may perceive and realise such vowels through the filter of their native phonological systems. Similar English vowels may be merged into a single L1 category, while unfamiliar vowels may be replaced by the nearest available native vowel. The integration of SLM and PAM therefore provides a comprehensive framework for analysing the production of English vowels by Kanuri and Ngizim speakers. The models enable the study to explain systematic patterns of vowel substitution, approximation and transfer, as well as differences in pronunciation accuracy arising from the structural relationship between the English, Kanuri and Ngizim vowel systems.

3.         Methodology

The study adopted a descriptive phonological approach to examine the production of English vowels by native speakers of Kanuri and Ngizim. Data were collectedfrom fifteen native speakers of Kanuri in Bama, Borno State, and another set of fifteen native speakers of Ngizim in Potiskum, Yobe State,through controlled pronunciation tasks consisting of a wordlist and reading materials designed to contain key English monophthongs and diphthongs. Participants were asked to pronounce the selected lexical items, and their vowel realisations were recorded and transcribed using the International Phonetic Alphabet (IPA). The recorded forms were then compared with Received Pronunciation (RP) targets in order to identify systematic deviations, substitution patterns and phonological processes influenced by first language vowel inventories. Quantitative frequency counts were also used to determine the most dominant realisations across speakers and vowel categories.

3.1.      Research Design

This study employs a comparative cross-sectional design to investigate English vowel production among native speakers of Kanuri and Ngizim. The design allows for the systematic comparison of vowel realisations between the two L1 groups and with Standard English norms. Aphonetic-contrastive approachis adopted, focusing onmonophthongs,diphthongs and vowel length, to identify patterns of L1 transfer, substitution or merging.

3.2.      Participants

The study involved 30 adult participants, aged 18–35, equally divided between the two language groups: 15 Kanuri speakers (8 males, 7 females) and 15 Ngizim speakers (7 males, 8 females). All participants met the following inclusion criteria: (1) Native speaker of Kanuri and Ngizim, with L1 exposure from birth (2) Intermediate proficiency in English (verified via school or university transcripts) (3) No reported speech or hearingimpairments. Participants were recruited from secondary schools and tertiaryinstitutionsin Borno and Yobe States, northeastern Nigeria, ensuring comparable sociolinguistic backgrounds.

3.3.      Materials

The following materials were used: Fourty English words representing all majormonophthongs (short and long) and all eight Standard English diphthongs (Roach, 2003). Words were selected to cover varied syllable structures and minimise lexical bias. Participants read the word list aloud from a printed sheet in a quiet environment.

3.4.      Recording Equipment

Audio was recorded using a high-quality digital recorder (44.1 kHz sampling rate, 16-bit resolution) with an external microphone to ensure clarity of vowel articulation.

4.         Results

4.1.      The realisation of English short monophthongs by native speakers of Kanuri and Ngizim

4.1.1    The English KIT vowel

This vowel is realised as three varied sounds in Kanuri-English: [i:], [e] and [ə]. The second and third realisations are influenced by the spelling of English words containing vowel letter <e>. Whereas Ngizim native speakers have only two realisations [i:] and [e] for the English vowel phoneme, with the first realisation being more prominent, as shown in Table 4. The majority of the Ngizim respondents pronounced /ii/ as /i/ in the word seat or /i/ with /ii/ in the word example, and /i/ instead of /ii/ in the word people. From the above statement, it shows that the majority of the respondents could not differentiate between the long and short monophthongs, e.g., seat /si:t/ as sit /sit/, shirt /ʃɜ:t/ as /ʃet/ etc.

Table 4. The Realisation of KIT Vowel

Phoneme	Kanuri realisation	Examples	Number of attempts/ % of realisation
/i/	[i]		100%
	[e]	college [kɔledʒ]	100%
	[ə]	citizen  [sɪtɪzən]	100%
	Ngizim realisation
	[i]	did [did]	100%
	[e]	started [statet]	100%

 

4.1.2.   TheEnglish SCHWA

This vowel is variably realised by the native speakers of Kanuri and Ngizim languages, some of the realisations were dependent on the spelling of English words, as shown in Table 5. The realisations are heavily dependent on the spelling of the affected words. The SCHWA vowel is variably represented in the orthography by more than five graphemes or letters of the English alphabet, and each of them is likely to influence the articulation of the vowel, providing strong evidence of spelling interference.

Table 5. The Realisation of SCHWA Vowel

Phoneme	Kanuri realisation	Examples	No. of attempts/ % of realisation
/ə/	[ͻ]	nation [neɪʃɔn]	100%
	[e]	development [debelɔpment]	100%
	[a]	critical [krɪtɪkal]	100%
	[a:]	dollars [dɒlɑ:s]
	[u]	people [pɪpul] sustain [sustain]
	Ngizim realisation
	[e]	Competition [kᴐmpetiʃan]	(49/60) 82%
	[a]	heart [hat]	(51/60) 85%
	[a:]	perhaps [pa:haps]	(31/60) 53%
	[ͻ]	polite [pͻlait] police [pͻlis]	60/60) 100% (60/60) 100%
	[u]	saturday [satude]	(58/60) 99%

 

4.1.3.   TheEnglish STRUT

This vowel is invariably realised as [ͻ] by both native speakers of Kanuri and Ngizim, as evident in Table 6.

Table 6. The Realisation of STRUT Vowel

Phoneme	Kanuri realisation	Examples	No. of attempts/ % of realisation
/ʌ/	[ͻ]	but [bɔt]	100%
	Ngizim realisation
	[ͻ]	must [mͻs] monday [mͻnde]	(58/60) 96% (52/60) 87%

 

4.2.      The realisation of the English long monophthongs by Kanuri and Ngizim speakers

4.2.1.   The English FLEECE

Thisvowel is shortened and realised as [i] in both Kanuri- and Ngizim-English varieties, as shown in Table 7.

Table 7. The Realisation of FLEECE vowel

Phoneme	Kanuri realisation	Examples	No. of attempts/ % of realisation
/i:/	[i]	seat   [sit]	100%
	Ngizim realisation
	[i]	seat   [sit] week [wik]	(56/60) 93.3% (44/60) 73.3%

 

4.2.2.   The English NURSE

This vowel is realised in three varied forms by the native speakers of Kanuri. For Ngizim, the vowel is realised in two forms. Each ofthe realisationsis based on the spelling differences, as shown in Table 8.

Table 8. The Realisation of NURSE Vowel

Phoneme	Kanuri realisation	Examples	No. of attempts/ % of realisation
	[ͻ]	working [wͻkin]	100%
	[e]	first [fest]	100%
	[a]	surge [sadʒ]	100%
	Ngizim realisation
	[ͻ]	church [ʧͻʧ]	(55/60) 91%
	[e]	girl [gel] first [fest]	(55/60) 91% (54/60) 90%

 

4.2.3.   The English GOOSE and THOUGHT

Thesevowels are invariably shortened as [ͻ] and [u] in both Kanuri and Ngizim English varieties, as evident in Table 9.

Table 9. The Realisation of GOOSE and THOUGHT Vowels

Phoneme	Kanuri realisation	Examples	No. of attempts/ % of realisation
/ʊ:/	[u]	lose [lus]	100%
	Ngizim realisation
	[u]	move [mub]	100%
	Kanuri realisation
/ͻ:/	[ͻ]	door [dͻ]
	Ngizim realisation
	[ͻ]	ball [bͻl]	63%

 

4.3.      The realisation of the English diphthongs by native speakers of Kanuri and Ngizim

4.3.1.   Diphthongs

This is the most problematic area for most Chadic languages, including Ngizim speaking Englishlearners. Most of the diphthong sounds could not be formed in Ngizim because only two diphthongs were identified in the language. Diphthong sounds are clearly identified in the dictionary of Ngizim published by Schuh, (1979), where Ngizim diphthong sounds were clearly stated. The majority of the respondents pronounced /eɪ/ as /e/ in the word ‘‘late /leɪt/’’ which was realised as [let]; ‘‘main’’ /meɪn/ as [mein] and ‘‘gate’’ /geɪt/ as [geɪt], respectively. The[let] realisation results in another independent word ‘let’ /let/ in English. Similarly, the realisations of the words ‘‘main’’ as [men] and ‘‘gate’’ as [get] equal to the native English realisations of other two independent English words: ‘‘get and men’’.

4.1.7.   The English FACE vowel

The FACE vowel has two realisations in Kanuri-English, while Ngizim maintains only one realisation, as indicated in Table 10 below.

Table 10. The Realisation of FACE Vowel

Phoneme	Kanuri realisation	Examples	No. of attempts/ % of realisation
/ei/	[e]	name [nem]	100%
	[e:]	eight [e:t]	100%
	Ngizim realisation
	[e]	eight [et] main [men]	(60/60) 100% (51/60) 85%

 

4.1.8.   The English GOAT vowel

This vowel is variably realised in two forms in Kanuri andNgizim, as evident in Table 11. It is mispronounced in the word “go’’ /gəu/ and “load’’ /ləud/as [ͻ] and [ͻ:], respectively, by most of theNgizim native speakers.

Table 11. The Realisation of GOAT vowel

Phoneme	Kanuri realisation	Examples	No. of attempts/ % of realisation
/əu/	[ͻ]	home[hͻm]	(17/17) 100%
	[ͻ:]	loan [lͻ:n]	(2/17) 12%
	Ngizim realisation
	[ͻ]	go [gͻ]	(60/60) 100%
	[ͻ:]	load [lͻ:d]	(49/60) 82%

 

5.         Discussion

The study examined the production of English monophthongs and a few diphthongs by native speakers of Kanuri and Ngizim, with particular attention to vowel quality, vowel length, and the influence of first language (L1) vowel inventories on second language (L2) pronunciation. The findings reveal systematic patterns of vowel substitution, shortening and diphthong reduction across both groups, confirming that L1 phonological structure and English orthographic interference plays a major role in shaping English vowel production.

5.1       Realisation of English monophthongs

In response to the first research question, the results show that both Kanuri and Ngizimspeakers demonstrate noticeable deviations in the production of English monophthongs.This is particularly truein the areas of vowel quality and vowel length. A major pattern observed is the frequent failure to maintain the long–short vowel contrast in English. For example, the English FLEECE vowel /iː/ was commonly realised as [i] by both groups, and Ngizim respondents often pronounced long vowels such as /iː/ in ‘seat’as [i], producing sit instead of seat. This confirms that vowel length contrast in English poses a major challenge, even for speakers whose L1 contains long vowels. The GOOSE vowel /uː/ and the THOUGHT vowel /ɔː/ were also shortened, reflecting the tendency of both speaker groups to reduce long vowels to short equivalents.

Ngizim speakers showed relatively greater consistency in their vowel productions compared to Kanuri speakers, particularly in havingtheir fewer realisation variants for certain vowels. For instance, the KIT vowel showed three different realisations in Kanuri-English [i:], [e], [ə], while Ngizim speakers mainly produced two: [i] and [e]. This indicates that Kanuri speakers exhibit greater variability and greater susceptibility to orthographic influence. Although Ngizim speakers displayed greater stability in vowel quality, both groups still struggled considerably with English length distinctions, suggesting that the presence of vowel length in Ngizim does not automatically guarantee accurate transfer to English phonological contrasts.

The SCHWA vowel /ə/ emerged as another problematic area, as it was realised variably in both groups and often replaced with full vowels such as [a], [e], [ɔ], or [u]. These substitutions appear to reflect the absence of a central reduced vowel system in the speakers’ L1 phonologies and also show the strong effect of English spelling on pronunciation. This supports the view that reduced vowels in English are difficult for learners whose L1 vowel systems lack schwa-like reduction patterns. The STRUT vowel /ʌ/ was consistently realised as [ɔ] by both Kanuri and Ngizim speakers (e.g., but → [bɔt]), confirming strong L1 transfer effects, as both languages favour back rounded vowel qualities rather than the mid-central English STRUT target vowel.

5.2       Realisation of English diphthongs

Regarding the second research question, the findings demonstrate that diphthongs constitute one of the most challenging areas of English vowel production for both Kanuri and Ngizim speakers. Across the data, English diphthongs were frequently reduced to monophthongs or replaced with shorter vowel sequences. For example, the FACE diphthong /eɪ/ was commonly realised as [e] in both groups. This reduction not only affects pronunciation accuracy but may also lead to lexical ambiguity because the resulting outputs resemble other English words. Similarly, the GOAT diphthong /əʊ/ which was regularly realised as [ɔ] or [ɔː] indicates a systematic preference for back rounded vowels and reflect the limited diphthong inventory available in both languages.Furthermore, diphthong simplification appears to be influenced by the phonological constraints of the L1 as well as the learners’ tendency to map English vowel sequences onto the closest available vowel categories in their native phonological systems. The evidence suggests that speakers employ a consistent strategy of monophthongisation, especially when the English diphthong does not correspond to any familiar L1 diphthong pattern.

5.3       Effect of L1 vowel inventory differences and accuracy

In addressing the third research question, the study found that differences in vowel inventory size and complexity between Kanuri and Ngizim do influence English vowel production, particularly in the degree of variability and stability observed across vowel categories. Ngizim speakers generally produced fewer alternative variants for the same English phoneme and showed more consistent patterns of substitution. Kanuri speakers, in contrast, demonstrated more variation, with multiple realisations for a single English vowel phoneme (e.g., KIT and NURSE vowels), suggesting a less stable mapping between English targets and L1 vowel categories.The findings suggest that Ngizim speakers may benefit from a vowel system that includes clearer length contrasts, which potentially enables more stable categorisation of English vowels. Although Ngizim speakers displayed comparatively greater consistency, they still exhibited major deviations from English targets, particularly in the area of vowel length and diphthong production. This indicates that inventory similarity alone is not sufficient; other factors such as orthographic interference, educational exposure, and limited contact with native-like English input may also play important roles. In addition, the findings show that orthographic influence affects both groups but appears stronger among Kanuri speakers, as seen in the KIT vowel and SCHWA realisations where spelling triggered alternative vowel choices. This suggests that learners may rely heavily on grapheme-to-phoneme mapping strategies shaped by their prior literacy practices, which may not align with English orthographic irregularities. Such spelling-driven pronunciation patterns may contribute to persistent pronunciation deviations and may eventually lead to fossilised forms, especially when learners repeatedly use these patterns in communication without corrective feedback.

6.         Conclusion    

This study investigated the realisation of English vowels by native speakers of Kanuri and Ngizim, focusing on short monophthongs, long monophthongs and diphthongs. The findings reveal that both groups exhibit systematic deviations from Received Pronunciation, especially through vowel substitution, shortening of long vowels and monophthongisation of diphthongs. Both groups demonstrated difficulty in distinguishing between long and short vowel contrasts, resulting in frequent shortening of long vowels. The SCHWA /ə/ was highly variable and strongly influenced by English spelling. The STRUT vowel /ʌ/ was consistently replaced by [ɔ] in both varieties, reflecting phonemic absence and transfer effects. Diphthongs emerged as the most challenging category, particularly in Ngizim-English, where diphthongs were frequently reduced to monophthongs, causing potential semantic ambiguity and reduced intelligibility. The study concludes that the vowel deviations found in Kanuri and Ngizim English are shaped mainly by first language L1 transfer effects, phonological constraints and orthographic interference. These findings highlight the need for explicit pronunciation teaching strategies, as detailed below.

7.         Implication for teaching and recommendation

The findings of this study have important implications for the teaching of English pronunciation among Kanuri and Ngizim speakers, as well as for broader descriptions of Nigerian English varieties. First, the systematic shortening of English long vowels and the frequent neutralisation of vowel length contrasts suggest that vowel duration is not a salient phonemic feature for many learners from these linguistic backgrounds. This indicates that pronunciation instruction should explicitly emphasise the functional role of vowel length in English meaning distinctions (e.g., seat vs. sit), using minimal pairs and auditory discrimination tasks. Second, the high variability in the realisation of the SCHWAvowel demonstrates that learners struggle with vowel reduction patterns, largely because reduced vowels are inconsistently represented in English spelling.

Consequently, pedagogical approaches should focus on stress-based pronunciation training, encouraging learners to recognise that unstressed syllables in English often contain centralised reduced vowels rather than full vowel qualities. Third, the consistent substitution of central vowels such as /ʌ/ with back vowels like [ɔ] reflects the absence of central vowel contrasts in the learners’ first languages. Teachers should therefore incorporate articulatory explanations, including tongue height and centrality, supported by visual vowel charts to improve learners’ awareness of vowel space differences. For diphthongs, pronunciation training should include step-by-step glide production exercises and listening-based reinforcement, since diphthong reduction may lead to unintelligibility and lexical confusion.

Overall, the study recommends targeted pronunciation interventions that prioritise vowel length, vowel reduction and diphthong articulation, while also addressing orthographic interference through phonemic awareness activities. Such findings can support the development of learner-specific pronunciation materials and improve intelligibility-based instruction in Nigerian ESL contexts.

8.         Limitation of the study

This study is limited to the analysis of selected English monophthongs and a few diphthongs produced by a relatively small sample of Kanuri and Ngizim speakers, and other variables such as age, educational background, English exposure and frequency of interaction with proficient English speakers were not systematically investigated. The findings may therefore not represent all speakers of the two languages, particularly across different age groups, educational levels, degree of exposure to the language, frequency of interaction with proficient English speakers or the full range of English vowel contrasts. In addition, the study focused solely on production data and did not examine learners' perception of English vowels. Future research should therefore employ larger, more, age-, education-, and exposure-diverse samples, and incorporate both perception and production tasks.

References:

Adewole, A. (2016). Vowel inventories in Nigerian languages. Lagos: Language Press.

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

Best, C. T., & Tyler, M. D. (2007). Nonnative and Second-language Speech Perception: Commonalities and complementarities. In O.-S. Bohn & M. J. Munro (Eds.), Language experience in second language speech learning: In honor of James Emil Flege (pp. 13–34). John Benjamins.

Brown, H. D. (1987). Principles of language learning and teaching (2nd ed.). Prentice Hall.

Bulakarima, A., &Shettima, S. (2012). A sketch of Kanuri phonology and tone. Maiduguri: University of Maiduguri Press.

Cook, V. (2004). The English writing system. Hodder Education.

Crystal, D. (2003). The Cambridge Encyclopaedia of the English language (2nd ed.). Cambridge University Press.

Cyffer, N., & Hutchison, J. P. (1979). The Standard Kanuri Orthography. Lagos: Thomas Nelson.

Flege, J. E. (1995). Second language speech learning: Theory, findings, and problems. In W. Strange (Ed.), Speech perception and linguistic experience: Issues in cross-language research (pp. 233–277). York Press.

Flege, J. E., & Bohn, O.-S. (2021). The revised Speech Learning Model (SLM-r). Second Language Speech, 1(1), 3–83. https://doi.org/10.1525/sls.2021.1.1.3

Gimson, A. C. (1962). An Introduction to the Pronunciation of English Language: London Retrieve:             http://www.celt.stir.ac.uk/satff/higdox/stephen/phono/phonolg.htm.

Jarrett, K. A. (1978). Vowel length in Kanuri. HarsunanNijeriya, VIII, 22–50.

Kenworthy (1987). Simple Codes and the Source of the Second Language Learner’s: Initial Heuristic Hypothesis. Studies in Second Language Acquisition. London

Lado, R. (1957). Linguistics across Culture: Applied Linguistics Teachers. Ann Arbor:  University of Michingan Press.

Lawani, A. (2009). English phonetics and phonology. Ibadan: Spectrum.

Modu, M. A. (2012). An Analysis of Quantitative Arabic Feet Patterns into Kanuri Poetry. In Literature, Muslim Societies and Contemporary: A Festschrift in Honour of Professor Tijani El-Miskin, 1, 37–52. Berlin: Scholars’ Press.

Oswalt D. and Bowen J. (1973). The Oxford Handbook of the History of Linguistics: retrieved from: www.scholink.org/ojs/index.php/selt

Roach, P. (2003). English Phonetics and Phonology (2^nded.). Cambridge: Cambridge        University Press.

Roach, P. (2009). English Phonetics and Phonology (4th ed.). Cambridge University Press.

Schuh, Russell G. (1978a). “Bade/Ngizim vowels and syllable structure.” Studies inAfrican Linguistics,9, 247-283.

Schuh R. G.  (1979). Ngizim English Dictionary:Retrive: www.http//books.goog/e.com.ng.

Schuh, R. G. (1981). Ngizim grammar and phonetics. Dallas: SIL Publications.