PHONOTACTICS OF AUTISTIC CHILDREN IN LAGOS STATE, NIGERIA 

 

 

 

 

 

 

BY; 

 

 

 

Esther Eyitosho ABE 

Matric. No: 211971 

B.A. (Ed.) English Studies (Ife), M.A. English Language (Lagos) 

 

 

 

 

 

 

 

A Thesis in the Department of English 

Submitted to the Faculty of Arts 

in partial fulfilment of the requirements for the Degree of 

 

 

 

DOCTOR OF PHILOSOPHY 

 

of the 

 

UNIVERSITY OF IBADAN 

 

 

 

 

 

 

AUGUST, 2023 

 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

ii 

 

 

 

CERTIFICATION 

 

I certify that this work was carried out by Esther Eyitosho ABE (Matric. No.: 211971) 

in the Department of English, University of Ibadan, Nigeria under my supervision. 

 

 

 

 

 

 

 

____________________________ 

Supervision 

A.B. Sunday 

B.A., M.A Ph.D. (Ibadan) 

Department of English, 

University of Ibadan, Ibadan, Nigeria 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

iii 

 

 

 

 

DEDICATION  

 

To the Glory of God and for the service of humanity! 

 

Dad, I wish I could see the pride on your face one more time. This is for you. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

iv 

 

 

ACKNOWLEDGEMENTS 

First to God, the author of life, for the strength and grace to commence and complete 

this work; Father, without Him, I am nothing. All that I am, all that I own, and all that I 

will ever be, I surrender everything at His feet. A billion tongues are not enough to give 

Him thanks. 

Before I begin my train of thoughts on this journey, I am indebted to Dr A.B. Sunday, 

my supervisor whose excellence and expertise have imparted on me much more than 

what he was assigned to do. He has gone ahead to become my mentor whose 

thoroughness and counsel have not only imprinted indelible marks in my life but have 

also helped to birth an entire career path and a passion that I never conceived before I 

met him. God used him in ways he would never know and my story would never be 

complete without him. I am grateful and God bless him. His wife, Dr Yewande Sunday 

has also been a silent support in more ways than I can count. I thank her for been there 

during this programme. 

I acknowledge the Head of Department, Professor Olutoyin B. Jegede, all lecturers, and 

administrative staff of the Department of English, for their professional support, 

assistance and encouragement at every stage of this work. 

The role my husband, Commander Gbenga Abe, played at the very beginning of this 

Ph.D. journey is a crucial one that cannot be mentioned later but now. While courting 

many years ago, he asked what my life goals were, one of which was to become a Ph.D. 

holder. He wrote it down and has since been running with it! As a new mother of our 

three-month old baby, he got my Masters’ form and again, while I was pregnant with 

our second child, he went ahead to get the form for Ph.D., reminding me of my goals at 

a time I was not even thinking about them. I thank him for always believing in me and 

my life’s vision more than I do at times. He is my propelling force and I cherish doing 

life with him every single day. 

I appreciate Dr Ayo Osisanwo, my lecturer, friend and mentor. He knew what was best 

for me and did not deny me the right to be supervised by the best, at a time I was not so 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

v 

sure. Now, I see that all he really wanted for me was the best! Through him, I came in 

contact with the very best supervisor. I have no regrets. God bless and continually lift 

him up. 

Professor A. Atolagbe, my supervisor during my Master’s programme  bmakes this list. 

Her admonishment not to waste time but do what I had to do in good time has been 

heeded and it has paid off today. I am so thankful for her prayers and words of 

encouragement. 

Dr Ronke E. Okhuosi became so many things to me in one sweeping moment – teacher, 

friend, host, accountability partner and more. I will not forget how she and her husband 

nursed and cared for me when I had to undergo a procedure at University College 

Hospital. She and her family have touched me in ways I cannot even recount and sweet 

Shalom too. I am blessed to have them in my life. I am grateful. 

The rate at which my father-in-law, Group Captain (Rtd.) K.R. Abe, took interest in this 

research assured me that I am truly loved. He called and asked questions about my 

progress at every point. In fact, he helped to contact some of the facilities visited and 

took me there himself. I am grateful for his support, prayers and encouragement. He did 

not make me feel the loss of my dear father, Elder E.S. Omokore, when he left to glory. 

His paternal love scaled me through! In the same breath, I appreciate the maternal love 

that surrounded me all through with their prayers, presence and counsel, watching over 

the children whenever I was away: Dcns. N.E. Omokore, Dcns. M.B. Abe; Mr and Mrs 

E.B. Majekodunmi, e jeun omo ati omomo pe l’oruko Jesu, Amin. My sister, Mrs D.T. 

Pacheco, all my siblings and siblings-in-law, I appreciate their love.  

I acknowledge the following families for their love and support directly and indirectly: 

the Adeleyes, Adeolus, Jayeolas, Babalolas, Lawals, Ajides, Ogunsolas and my friends 

– Mrs. J. Aruomaren, Mrs. E. Neesae and Mrs. V. Igbokah. Special thanks also to the 

coordinators of the facilities used in the research: Mrs. J. Ahmedu, Mrs. D. Akande, Mrs. 

U. Emefoh and Dr D. Aliu. 

I have met wonderful people during the course of this work and these relationships were 

spun due to the leadership style of our supervisor who took us in as one family. Indeed, 

beyond becoming academic siblings, we have a lasting relationship. Dr David 

Olorunsogo, who helped to connect me with those who had direct impact on my work. 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

vi 

Mr A. Babayemi and Ms Grace Olarewaju, who became my Optimality Theory tutor 

even before she began the programme herself. Dr Mathias Chukwu – my proposal 

buddy, Mrs Sabainah Akanbi – my birthday twinny, Mrs Ibijoke Omole, Drs Matthew 

Ilori, A. Atanda and Mr Eyitayo Folorunso, I thank them for their help and counsel. 

There are a host of others that time and space would not permit me to mention. I deeply 

appreciate them. 

Last but certainly not the least are my wonderful princesses, Faith and Favour Abe, 

whose love, cooperation, contribution and prayers have encouraged me along the way. 

Faith, my defence, had to grow up quickly and become an assistant mom whenever I 

could not leave the table; she helped out in several ways that made life easy for me 

during this work, I am touched by her sacrifices. Favour was ever obedient and brought 

smiles to my tired face seamlessly, thanks my delight. They would rejoice at my 

triumphs, bear the exasperations with me and always prayed for God’s inspiration and 

speed so I could finish my work on time. God has indeed answered their prayers and this 

is the evidence. I pray they will fulfill their glorious destinies in Christ. Amen. 

 

Esther Abe 

June, 2023 

 

 

 

  

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

vii 

     ABSTRACT 

Phonotactics, which accounts for sound patterning, is affected by language regression in 

children with autism. Existing linguistic studies on autism in Nigeria focused mainly on 

general description of the disorder and developmental challenges. However, little 

attention was devoted to systematic description of their phonological patterns. 

Therefore, this study was designed to examine the speech production of autistic children 

in Lagos State, Nigeria, with a view to describing the phonological processes involved 

in the production of phonemes and syllables and the constraint ranking responsible for 

them. 

Alan Prince and Paul Smolensky’s Optimality Theory served as the framework, while 

the descriptive design was adopted. The convenience sampling technique was employed 

to select four special-needs facilities in three local government areas (Amuwo-odofin, 

Badagry, and Ikeja) of Lagos State. Availability of and accessibility to relevant data 

necessitated the selection of the facilities. Twenty-four children with autism under the 

age of 17 years (Amuwo-odofin: 17, Badagry: 2, and Ikeja: 5) accessing the facilities 

were purposively selected. Speeches of the participants were audio-recorded twice a 

week for two years. The data were subjected to phonological and acoustic analyses. 

Three phonological processes were dominant: deletion, substitution and epenthesis. 

Vowels were generally modified. Deletion was dominant in the rendition of consonants, 

with the lateral sound /l/ mostly affected. Substitution and epenthesis affected mainly 

vowels. Central vowels /ᴧ, ɜ, ə/ were strengthened and /ɔ/, /e/, /æ/ were used for each, 

respectively. Epenthesis was prominent in the realisation of vocalic phonemes, with /ɪ/ 

and /ə/ substituted as /i:/ and /æ/, respectively. It was also dominant in the articulation 

of triphthongs /eɪə/ and /ɔɪə/. Monophthongs were either retained or substituted with 

long or strong vowels. Diphthongs were monophthongised or substituted with stronger 

versions.  Triphthongs were either substituted with diphthongs or epenthesised with an 

intrusive /j/, creating disyllabic sounds. With regard to syllables, simple onsets were 

produced comfortably more than complex onsets, while both simple and complex codas 

were deleted, reduced or produced with great difficulty. Constraint ranking favoured 

markedness over faithfulness. The constraints responsible for vowels were *SCHWA, 

NOCODA, NODIPH and *HIATUS. *COMPLEX ONSET and *COMPLEX CODA 

were the markedness constraints accountable for the simplified outcome of complex 

onsets, complex codas, open and closed syllables. *HIATUS was ranked above MAX-

V for the emergence of minimum syllables. Consonant clusters, especially at coda 

positions, were reduced or rendered open, as *COMPLEX CODA was preferred above 

MAX. There were inconsistencies in the intensity values. The values were either 

abnormally high (73.94dB) or low (52. 89dB). Against a threshold of 63.42dB, their 

pitch values were either extremely high or low.  

The phonotactics of autistic children in Lagos State, Nigeria is characterised by 

simplification of difficult phonemes and syllables through deployment of deletion, 

substitution and epenthesis. Therefore, multidisciplinary therapists should be employed 

to improve the treatment and rehabilitation of autistic children. 

Keywords: Phonotactics, Phonological processes, Syllabic structure, Nigerian 

autistic children  

Word count: 465 

 

 

 

 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

viii 

 

 

 

 

TABLE OF CONTENTS 

 

Title                   page 

Title page          i 

Certification          ii 

Dedication          iii 

Acknowledgements         iv 

Abstract          vii 

Table of contents         viii 

List of tables          xii 

List of figures          xiii 

List of tableaux         xiv 

 

CHAPTER ONE: INTRODUCTION 

 

1.1 Background to the study        1 

1.2 Statement of the problem        3 

1.3 Aim and objectives        4 

1.4 Scope of the study        4 

1.5 Significance of the study       4 

1.6 Summary         5 

CHAPTER TWO: LITERATURE REVIEW AND THEORETICAL  

FRAMEWORK 

2.0 Preamble          6 

2.1 Review of relevant studies       6 

2.1.1 Autism          6 

2.1.1.1 Brief history of autism spectrum disorder     7 

2.1.1.2 Prevalence of autism        7 

2.1.1.3 Characteristics and types of autism      8 

2.1.1.4 Facial features of autistic people      10 

2.1.1.5 Challenges of autism        11 

2.1.1.6 Causes of autism        11 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

ix 

2.1.17 Language in autism         12 

2.1.1.8 Language behaviour among subtypes      13 

2.1.1.9 Autism in Nigeria        14 

2.1.1.10 Intervention in autism        15 

2.1.2 Phonotactics         17 

2.1.2.1  Sonority sequencing principle       20 

2.1.3 Syllable         20 

2.1.3.1 The syllabic structure        22 

2.1.4 Studies on autistic phonology       27 

2.1.5 Studies on phonotactics       29 

2.1.6 Phonological processes       33 

2.1.7 Appraisal of literature review       36 

2. 2 Theoretical framework       36 

2. 2.1    Optimality Theory        36 

2. 2.2 Basic tenets and assumptions of Optimality Theory    37 

2. 2.3 Basic components of Optimality Theory     37 

2. 2.3.1 Generator         38 

2. 2.3.2 Evaluator         38 

2. 2.3.3 Constraints         39 

2.2.4 Conflicts between markedness and faithfulness constraints   41 

2.2.5 Faithful and unfaithful mappings       42 

2.2.6 Richness of the base        42 

2. 2.7   Presentation of the tableau       43 

2.2.8 Optimality Theory and prosody      46 

2.2.9  Optimality Theory and syllable       46 

2.2.10  Optimality theory and phonotactics      48 

2.2.11 Optimality theory and language disorder     51 

2.2.12 Constraints relevant to this study      52 

2.3 Summary         52 

CHAPTER THREE:  METHODOLOGY 

3.0  Preamble         53 

3.1 Research design        53 

3.2 Study area         53 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

x 

3.3  Study population        54 

3.4 Sampling procedure        54 

3.5 Research instrument        55 

3.6 Methods of data collection       55 

3.7 Methods of data analysis       56 

3.8 Ethical consideration        57 

3.9 Summary         58 

CHAPTER FOUR: ANALYSIS AND DISCUSSION 

4.0 Preamble         59 

4.1 Demographic characteristics of research participants    59 

4.2 Phoneme sequencing by the autistic children     61 

4.2.1 Sequence of front vowels       62 

4.2.2 Sequence of back vowels       77 

4.2.3 Sequence of central vowels       81 

4.2.4 Sequence of diphthongs       103 

4.2.5 Sequence of triphthongs       127 

4.2.6 Consonant phoneme sequence in autistic children’s onsets   144 

4.2.7 Consonant phoneme sequence in autistic children’s codas   150 

4.3 Patterns of consonant clusters in autistic children    157 

4.3.1 Patterns of complex onsets in autistic children    157 

4.3.2 Patterns of complex codas in autistic children    169 

4.4 Pattern of syllable structures by the participants    187 

4.5 Impact of intervention on autistics’ phonotactics    224 

4.5.1 Early intervention         225 

4.5.2 Intermediate intervention       228 

4.5.3 Late intervention        231 

4.5.4 Factors that affect intervention       233 

4.5.4.1 Background of affected child        234 

4.5.4.2 Effects of diagnosis on affected children      235 

4.5.4.3 Effects of intervention and management on autism    236 

4.5.4.4 Autism in Nigeria today       237 

4.6  Discussion of findings       239 

4.6.1 Phoneme sequences in autistic utterances      239 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

xi 

4.6.2 Patterns of autistic’s interactions with cluster    241 

4.6.3 Autistics syllable structure        242 

4.6.4 Constraints responsible for the emerging patterns    243 

4.6.5 Impact of intervention on autistics’ phonotactics    244 

4.6.5.1 Factors that affected intervention       244 

4.7 Summary         246 

 

CHAPTER FIVE: SUMMARY AND CONCLUSIONS 

5.0 Preamble          247 

5.1  Summary of the study        247 

5.1.1  Phoneme sequences in Nigerian autistic utterances    247 

5.1.2  The pattern of autistic children’s consonant clusters    248 

5.1.3  The syllable structure of autistic children     248  

5.1.4 Constraints responsible for all the outcomes     249 

5.1.5 Impact of intervention on autistics’ phonotactics    249 

5.2 Conclusion          250 

5.3 Limitations of the study        250 

5.4  Recommendations        251 

5.5  Contributions to knowledge        253 

5.6 Suggestions for further studies       253 

REFERENCES         255  

APPENDICES         265 

 

 

 

 

 

 

 

 

 

 

 

  

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

xii 

LIST OF TABLES 

Title                     page 

Table 2.1 The syllable structure        26 

Table 4.1 Demographic characteristics of research participants   60 

Table 4.2: Variants of front vowels produced by the participants   62 

Table 4.3: Variants of back vowels produced by the participants   77 

Table 4.4: Variants of central vowels produced by the participants   81 

Table 4.5: Variants of centring diphthongs produced by the participants   104 

Table 4.6: Variants of closing diphthongs produced by the participants  113 

Table 4.7: Variation of the triphthongs produced by the participants  128 

Table 4.8: Variants of the participants’ stops      145 

Table 4.9: Variants of the participants’ fricatives     147 

Table 4.10: Variants of the participants’ laterals     149 

Table 4.11: Variants of participants’ stops in simple codas    151 

Table 4.12: Variants of participants’ fricatives in simple coda    153 

Table 4.13: Variants of participants’ lateral in simple coda     155 

Table 4.14: Variants of participants’ sequence in C2 onset    158 

Table 4.15: Variants of the participants’ sequence in C3 onset   164 

Table 4.16:  Variants of participants’ C2 coda clusters    170 

Table 4.17:  Variants of the participants’ C3 coda clusters    179 

Table 4.18:  Variants of the participants’ C4 coda clusters    181 

Table 4.19:  Variants of participants’ minimum syllable structures   188 

Table 4.20: Open syllables in the autistic children’s utterances   201 

Table 4.21:  Closed syllables in the autistic children’s utterances   210 

Table 4.22: An overview of autistics’ productions among subtypes   223 

Table 4.23: First and second performances for the mildly affected autistics  226 

Table 4.24: First and second performances of moderately affected autistics  229 

Table 4.25: First and second performances of severely affected autistics  232 

 

 

 

 

 

 

 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

xiii 

 

 

 

LIST OF FIGURES 

Title                               page 

 

Figure 2.1 The syllable schema        21 

Figure 2.2 The syllable structure        23 

Figure 4.1: Acoustic representation of colleague     74 

Figure 4.2: Acoustic representation of plunge     86 

Figure 4.3 Representation of /ɔ/ on acoustic vowel chart    88 

Figure 4.4 Acoustic representation of navigate     118 

Figure 4.5: Acoustic representation of fire      139 

Figure 4.6 Representation of /æ/ on acoustic vowel chart    141 

Figure 4.7 Acoustic representation of screw      168 

Figure 4.8 Acoustic representation of sprint      177 

Figure 4.9 Acoustic representation of sculpts      185 

Figure 4.10 Acoustic representation of air      193 

Figure 4.11 Representation of /e/ on an acoustic vowel chart   195 

Figure 4.12 Acoustic representation of our      199 

Figure 4.13  Representation of the acoustic analysis for boys   217 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

xiv 

 

LIST OF TABLEAUX 

Title                    page 

Tableau 2.1: Illustration of an OT tableau      45 

Tableau 4.1: The emergence of /spi:nt/      64 

Tableau 4.2: The emergence of /kɔ:ji:n/      66 

Tableau 4.3: the emergence of /pu:/       68 

Tableau 4.4: The emergence of /skʊ/       70 

Tableau 4.5: The emergence of /kɔlɪ/       72 

Tableau 4.6: the emergence of /kɔ/       79 

Tableau 4.7: The emergence of /plɔnʤ/      84 

Tableau 4.8: The emergence of /im.pe.di.men/     90 

Tableau 4.9: The emergence of /ɑ.bɑ/      92 

Tableau 4.10: The emergence of /ɑ.rɑ/      94 

Tableau 4.11: The emergence of /pɑ:/      96 

Tableau 4.12: The emergence of /im.pi.di.ment/     98 

Tableau 4.13: the emergence of /kæ/       100 

Tableau 4.14 the emergence of /eji:/ and /eri:/      102 

Tableau 4.15: the emergence of /pie/       107 

Tableau 4.16: The emergence of /bes/      109 

Tableau 4.17: the emergence of /pjo/       111 

Tableau 4.18: The emergence of /nɑviɡe/      116 

Tableau 4:19: The emergence of /sprɑ:t/      120 

Tableau 4.20: the emergence of /bɔs/       122 

Tableau 4.22: the emergence of /ɡo/       124 

Tableau 4.23: The emergence of /prɑ:d/       126 

Tableau 4.24: The emergence of /rojæ/      131 

Tableau 4.25: the emergence of /fævæ/      133 

Tableau 4.26: The emergence of /pejæ/       135 

Tableau 4.27: The emergence of /pæjæ/      137 

Tableau 4.28: The emergence of /owæ/      143 

Tableau 4.29: The emergence of /pejæ/      160 

Tableau 4.30: The emergence of /fæwæ/      162 

Tableau 4.31: The emergence of /du:/      166 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

xv 

Tableau 4.32: The emergence of /plɔʤ/      173 

Tableau 4.33: The emergence of /spi:/      175 

Tableau 4.34: The emergence of /skɔ:/       183 

Tableau 4.35: The emergence of /e/       191 

Tableau 4.34: The emergence of /æwæ/      197 

Tableau 4.35: The emergence of /CVCV/      204 

Tableau 4.36: The emergence of /CCV/      206 

Tableau 4.37: The emergence of /CCVCV/      208 

Tableau 4.38: The emergence of /CV/      215 

Tableau 4.39: The emergence of /CCCV/      219 

Tableau 4.40: The emergence of /CCVCC/      221 

 

 

 

 

 

 

 

 

 

 

  

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

1 

 

 

 

CHAPTER ONE 

INTRODUCTION 

1.1 Background to the study 

 

Communication between typical and atypical language users can be challenging because 

of the speech impairments that characterise the latter. These impairments characterise 

atypical conditions such as dyslexia, Down syndrome, and autisms, and help to detect 

the conditions. In developing countries, however, this linguistic basis for diagnosis is 

underexplored, especially for autism. The possibility that communication challenges, 

particularly phonological impairments, suggest the earliest indicators of autism, as well 

as other atypical conditions, is ignored in these parts of the world (Tager-Flusberg, 

2005). These language indices are a major criterion for autism detection in children 

because unlike other conditions, autism is not associated with any physical deformity or 

feature detectable by the eyes (Loftus, 2021). 

Autism spectrum disorder (ASD) can be detected in children through their 

communication regression because the level of language competence attained before 

eighteen months is often suddenly lost or greatly reduced. This language regression is 

also often accompanied by social, developmental, and behavioural uncertainties (Lord, 

Shulman & DiLavore, 2004). Bhandari (2019) states that ASD is a lifelong condition 

which is usually diagnosed in younger children between 18 months and 3 years in most 

developed countries and can be managed by medication, education, therapy, and 

alternate medication. Otherwise, language development and competence in autism 

would be marked with difficulty, unlike that of a typically developing child (TDC) which 

it can be measured against (Kwok, Brown, Smyth & Cardy, 2015; Nordgren, 2020).  

Therefore, language is central to identifying and understanding autism and, although this 

knowledge has not prompted enough related research in the field of linguistics, 

numerous studies have been carried out in several other fields to understand this 

condition better. Thus, there is a need for more language-oriented investigations of this 

phenomenon so as to engender more wholesome studies.  

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

2 

Moreover, there is a relatively low knowledge and awareness level about autism among 

medical personnel, health workers, parents and guardians in developing countries, unlike 

the advanced information, sophisticated medication, treatments, therapies, and case 

management techniques that have been developed for autistic and other atypically 

developing patients over time in developed nations (Bakare, Ebigbo, Agomoh & 

Menkiti, 2008; Igwe, Ahanotu, Bakare, Anchor & Igwe, 2011; Bakare, Tunde-

Ayindmode, Adewuya, Bello-Mojeed, Sale, James, Yunusa, Obindo, Igwe, Odinka, 

Okafo, Oshodi, Okonoda, Munir & Orovwigho, 2015; Adeosun, Enye, Ayorinde, 

Ogundele & Adewole, 2017). In these advanced nations, professionals like medical 

doctors, psychologists, speech therapists, physiotherapists, and special educators are 

readily available to assess, diagnose, and develop an individualised educational plan for 

each child with ASD (Nwokolo, 2010; Alli, 2019). Apart from the fact that this level of 

advancement is unavailable in Nigeria, Nwokolo (2010) also notes that there are many 

undiagnosed children affected by autism in Nigeria but who are labelled imbeciles, fools, 

and even witches. As a result, these children do not benefit from the treatment available 

in hospitals.  

Wolk, Edward, and Brenan (2016) suggest that there are about 600,000 autistic people 

in Nigeria, and Guaranty Trust Bank Orange Ribbon Initiative (2015) report that the 

number has increased to over 1 million known cases. World Health Organisation (2018) 

report that there may be 1 case of autism in every 125 children born in Nigeria and this 

brings the estimated number of autistic children to 662,400 children in a population of 

180 million, with 46% below fifteen years (UNICEF- United Nations Children’s Fund, 

2021). The figure is not commensurate to the number of autistic children registered in 

special schools around the country, where customised education is being received. 

Considering the above, there is a need for a higher rate of autism diagnosis among 

Nigerian children, as well as an application of the resources of language to benefit 

autistic children as early as possible. Specifically, phonological studies can assist autistic 

detection because it covers the intricacies of speech sound production, profiling of 

speech sounds and syllabic structure which are central to autism diagnosis. This research 

investigates the sequence of sounds in the utterances of autistic children. 

 

 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

3 

1. 2  Statement of the problem 

Studies in language and autism are so limited that Boucher and Anns (2018) assert that 

significant clinical language impairments in ASD are under-researched. Being a 

relatively common developmental disorder, lack of thorough research in this area 

exposes it to speculative interpretation which may be misleading. Hence, there is an 

urgent need to focus on expanding the literature on ASD diagnosis and intervention in 

developing countries, particularly, Nigeria.  

It is worthy of note that language regression is a major factor for detecting and 

understanding autism in children (Lord, Shulman & DiLavore, 2004) and there is a 

relatively low knowledge of, and response to autism in Nigeria (Bakare et al., 2008; 

Igwe et al., 2011; Bakare et al., 2015; Adeosun et al., 2017). Although remarkable 

language studies have been carried out on language use by autistic people (Tager-

Flusberg, 2006; Schoen, Paul & Chawarska, 2011; Nwanze, 2013; Akinmurele, 2019), 

only a few focused on phonology (Ojo, 2012; Esan, 2018). Most studies on phonological 

descriptions of atypical language users have focused mainly on monolingual speakers of 

English or other foreign languages from advanced countries. Examples of such studies 

include Luce and Large (2001), Leonard, Davis and Deevy (2007), Yu, Abrego-Collier, 

and Sondergger (2013), An, Brizan, Morales, Syed and Rosenberg (2015), Nordgren 

(2015), Boucher and Anns (2018), and Synder, Cohn and Zellou (2019). The few studies 

that focused on autistic phonology in Nigeria (Ojo, 2012; Nwanze, 2013; Esan, 2018) 

carried out these investigations mostly from the perspective of adults and did not focus 

on the word construction level of the participants, where their phonotactics can be 

generated.   

Also, a few existing studies on phonotactics are on educated users of English in Nigeria 

(Soneye and Faleye, 2015); others exist on second languages (Mbah & David, 2013; 

Unubi & Ikani, 2018). The phonological aspect is considered to be relevant because it 

covers speech sounds production and the way these sounds are strung together to make 

syllables and words. Thus, the phonotactics of language as spoken by autistic children 

can help to establish a phonological system. Therefore, there is a need to carry out a 

phonological study of Nigerian autistic children’s English, particularly those who are 

still undergoing the language acquisition process. This is for the purpose of establishing 

the phonotactics of English as used by autistic children in Nigeria, thus providing a 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

4 

phonological profile of their speech and aiding higher rates of diagnosis and intervention 

in Nigeria. 

1.3 Aim and objectives 

The aim of this study was to establish the sequence of sounds in the utterances of 

English-speaking autistic children in Lagos State, Nigeria. The objectives were to: 

i. describe the phoneme sequences employed in autistics’ utterances, 

ii. establish the patterns of consonant clusters, 

iii. identify the structure of their syllable,  

iv. determine the constraints responsible for these patterns, and 

v. investigate the impact of intervention on their phonotactics. 

1.4  Scope of the study 

 

The focus of this research is autistic children’s spoken English, with specific attention 

to their phonotactics. Twenty-four English-speaking autistic Nigerian children with 

varying degrees of language competence and at different levels of intervention 

constituted the sample population. The study covers their vowels and consonant 

renditions at initial, medial, and final positions. It also considers consonant clusters at 

onset and coda positions, and syllabic structures including the English syllabic structure 

C0-3 V C0-4 and monosyllabic and polysyllabic words. 

1.5 Significance of the study 

By investigating the phonology of this population, this study helps to provide 

enlightenment on the description of the sound patterns and syllabic structures of 

autistics’ phonology. Also, the study contributes to the literature on the phonology of 

special needs population in Nigeria which is also useful for medical personnel, clinicians 

and special educators, as well as speech pathologists and therapists concerned with ASD. 

Additionally, specialists like neurologists, psychologists and others concerned with 

diagnosing the disorder will be more informed about the language peculiarities of 

autistic children and diagnosed people; thus, this study will improve diagnoses and 

intervention. 

To a greater degree, the findings of this study will arm special educators with the relevant 

information to develop a much more comprehensive lesson plan with achievable aims 

and objectives that have tangible and measurable results. In addition to this, having 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

5 

established the scarcity of literature in the phonotactics of special needs people in 

Nigeria, this study will indeed be a springboard for upcoming researchers in this field.  

The social implications of this study are that it will aid intelligibility between ASD and 

typical language users in the society where they are found, thus, bridging the 

communication gap and breakdown between typical language users and ASD speakers. 

This will further aid the integration of ASD patients into different spheres of the society 

without side-lining or stigmatising them as the norm still is today. Moreover, this will 

aid the standardisation of Nigerian English status which will have an aspect of ASD 

phonology catered for.  

Finally, this research will contribute to the existing literature on autistic children’s 

spoken English phonology. It will aid the language profiling of autistic children’s 

phonology of English, thus, creating a template of what to expect in the language of an 

autistic Nigerian child. Also, this study will be a guide to medical practitioners, parents 

and teachers for detecting autism much earlier in children. 

 

1.6 Summary 

This chapter presented the introductory aspects of the study. It contained the objectives, 

the scope and the significance of the study. The next chapter focuses on the review of 

relevant conceptual and empirical literature to the study. 

 

 

 

  

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

6 

 

 

 

CHAPTER TWO 

LITERATURE REVIEW AND THEORETICAL FRAMEWORK 

2.0 Preamble  

This chapter explores relevant conceptual and empirical literature available on autism, 

syllabic structure, phonology of atypical people and phonotactics studies. The chapter 

also covers related topics to the present study as well as the theoretical framework 

engaged in this study. 

2.1  Review of relevant studies  

This section examines relevant literature to the study and they are presented under the 

following sub-headings below.   

2.1.1  Autism  

Autism spectrum disorder (ASD) is a neurological condition which mostly interferes 

with communication, language, social and developmental behaviour. It is a psychiatric 

disorder which is characterised by trouble with social interaction, restricted interests, 

repetitive behaviour and others as particular to each autistic person. Autism is a lifelong 

condition which is usually diagnosed in younger children within the ages of 18 months 

and 3 years in most developed countries and can be managed by medication, education, 

therapy, and alternate medication (Bhandari, 2019). 

Associated with medical and mental health issues, ASD can have a wide-ranging impact 

on the body. More than half of an estimated 30-61% of autistic children are affected with 

one or more chronic sleep challenges (Mazurek & Petroski, 2015). Anxiety disorders are 

thought to affect approximately 11-40% of ASD children, while depression affects an 

estimated 7% of autistic children. Up to 26% of adults with ASD are likely to suffer 8 

times more from one or more chronic gastrointestinal disorders than other children. 

Furthermore, about one-third of ASD children suffer from epilepsy (Alli, 2019). Autism-

associated health problems extend across a person’s life span, from childhood to old age. 

There are medications to help manage autism, however, Risperidone and Aripiprazole 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

7 

are the only medications approved by Food and Drug Administration (FDA) for assisting 

agitation and irritation in ASD people.  

2.1.1.1 Brief history of autism spectrum disorder 

Autism was first referred to as a range of neuro-psychological conditions in the early 

1900s. The term was got from a Greek word, “autos” which means “self” and which 

describes a condition in which a person is removed from social interaction and becomes 

isolated. Researchers from the United States of America began to use the term “autism” 

in the 1940s to describe emotional or social problems in children though the term itself 

was first used by Eugene Bleuler, a German psychiatrist in 1906. Asperger’s syndrome 

was also adopted around the same period to describe similar conditions in Germany. In 

the 1960s, however, professionals began to separate the understanding of autism from 

schizophrenia. Before then, early researchers had linked autism to schizophrenia (Wolff, 

2004; Evans, 2013; Zeldovich, 2018). 

The 1960s and 1970s witnessed the search for cure, treatments, and medications; as such, 

some experimental and sometimes extreme measures were taken to cause a behavioural 

change, these measures relied on pain and punishment. In the 1980s and 1990s, therapy 

was introduced and the use of higher controlled learning environments emerged as the 

primary treatment of autism and related conditions. Today, behavioural therapy and 

educational interventions remain the approved FDA mainstays of treatment for autism. 

As of June, 2021, FDA has approved a device for diagnosing. Other treatments are added 

as needed (Evans, 2013; FDA, 2021; Pathak, 2021). 

2.1.1.2 Prevalence of autism 

According to Centers for Disease Control and Prevention (2018), autism is said to affect 

an estimate of 1 in 59 children in the United States, and 4 in every 5 affected, are said to 

be boys. In 2018, it was determined that 1 in every 59 children was diagnosed with ASD 

and among every 151 autistics, 1 is a girl, while 37 are boys; indicating that boys are 

more likely to be diagnosed with ASD than girls. The reasons for this are still unknown. 

Contrarily, in a study conducted by Chinawa, Manyike, Aniwada, Chinawa, Obu, 

Odetunde, Nwokocha and Ibekwe (2016), they discovered that there are no gender 

differences in the cases studied. Ahmed, Ahmed, Baba, Legbo, Nauzo, Omar, and Tahir 

(2019) contradicted earlier research, stating that more females were affected by the 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

8 

disorder than their male counterparts, however, they stated that a possible reason for this 

finding could be the setting of the research. 

According to GTBORI (2015), there are about a million cases of autism in Nigeria today. 

This number is still uncertified because UNICEF (2021) claims that there is at least 1 

case of autism in every 125 births in Nigeria. Many are yet to be diagnosed as a result 

of ignorance and some other reasons. Nwokolo (2010) submits that there are also some 

misgivings about ASD, for instance, it was misconstrued to be a “Whiteman” disease, 

and later, it was perceived to be the sickness of the rich. Nwokolo (2010) and some other 

studies refuted these claims, correcting the misgivings. Regarding the misconception of 

ASD being a “Whiteman” disease, this was as a result of advanced research in advanced 

countries, whereas none whatsoever existed in developing countries, and about ASD 

being a sickness of the rich, the high cost of diagnosis and intervention was such that 

only the rich could afford it. Thus, these reasons birthed the misconceptions, however, 

studies have shown that ASD can affect all ethnic and socioeconomic groups and early 

intervention provides the best opportunity to promote a healthy development and 

discover benefits that can last a lifetime. Also, early intervention is very important 

because there is no cure yet, the condition can only be managed. 

2.1.1.3 Characteristics and types of autism 

 

Autism spectrum disorder (ASD) is a broad range of disorders characterised by 

difficulties with social skills, repetitive behaviours, speech and nonverbal 

communication. As a spectrum disorder, each autistic challenge varies from highly 

skilled, mildly to severely challenged, and affected people may require from significant 

to lesser support in their daily lives, while for some others, in a few cases, live entirely 

independently.  

Because of the broad spectrum disorders that variously differ from one autistic to the 

other, all ranges of autism now fall under the term autism spectrum disorders (ASD). It 

is worthy of note that irrespective of the social type of autism, communication 

impairment is common to them. In 2013, the American Psychiatric Association (APA) 

combined four distinct Autism diagnoses under ASD. These diagnoses are autistic 

disorder, childhood disintegrative disorder, pervasive developmental disorder and 

Asperger’s syndrome according to Diagnostic by Statistical Manual of Mental Disorders 

IV (DSM-4). 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

9 

1. Autistic Disorder: This is the most common and is characterised by social 

interactions, communication and imaginative play in children younger than 3 

years. It is the severe kind along the spectrum disorder. 

2. Childhood Disintegrative Disorder: These children develop normally for at least 

2 years and then lose some or most of their communication and social skills. This 

is an extremely rare disorder and its existence as a separate condition is a matter 

of debate among many mental health professionals.  

3. Pervasive Developmental Disorder (PPD) is also known as typical autism. This 

is a type of catch-all category for children who have some autistic behaviour but 

who do not fit into other categories. PDD is not as severe as autistic disorder, 

neither is it as mild as Asperger’s, it is therefore tagged Moderate along the 

spectrum. 

4. Asperger’s syndrome: Children with Asperger’s syndrome do not have language 

impairments; in fact, they typically perform averagely or above the average limit 

on their intelligent test results. Nonetheless, they share the same social issues and 

narrow interests as children with autistic disorder. Their typical to strong verbal 

language skills and intellectual ability distinguish them from the other forms of 

autism and they are considered Mild on the spectrum. 

Recently, however, autistic disorder, pervasive developmental disorder, and Asperger’s 

syndrome have become popular as some researchers have shown that delayed language 

and loss of language characterise other autistic types and so should not be the 

characteristic of a subtype. Some special schools in Nigeria still identify affected 

children by these subtypes, while others have adjusted to grouping them according to 

their level of severity, that is, mild, moderate or severe autism. 

In accordance to Diagnostic by Statistical Manual of Mental Disorders V (DSM-5) 

classification of autism, it is safer to identify affected population in levels – levels 1 – 3, 

where 1 is mildly impaired; 2 is moderately impaired and 3 is severely impaired. These 

different levels need assistance accordingly. This is because when early intervention is 

introduced to autistics, the severity of the disorder tends to become moderate and even 

mild, such that they could be adopted into mainstream education and compete with 

typically developing children as they grow. The DSM-5 classification of autism has been 

adopted for this work. 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

10 

As stated by Wing (1981), three main diagnostic features of ASD are identifiable based 

strictly on behaviour known as Wing’s triad: lack of social responsiveness, compulsive, 

ritualistic behaviour and impaired communication, and language skill. The diagnostic 

criteria related to language are lack of, or late development of language without any 

attempt to compensate with gestures, impairment in the ability to initiate or sustain 

conversation and stereotypical, repetitive, and idiosyncratic language. 

However, it is worthy of note that the severity of the disorder can be categorised based 

on the various diagnostic features of Wing’s triad. It is possible for a child to be severe 

in one feature and be moderate or even mild, in the others. The level of severity adopted 

for this work is strictly in relation to language and communication. Hence, a child 

categorised as mild in this study (language), may be moderate or even severe in social 

or behavioural development. 

2.1.1.4 Facial features of autistic people  

Some arguments about autism being identifiable by some facial features have mandated 

this subject to be re-examined. There might be physical facial features similar to autistics 

but these are not palpable to the human eye without the use of advanced technology with 

specialised equipment to detect these features. Aldridge, George, Cole, Austin, 

Takahachi, Duan, and Miles (2011) applied the use of a high technology, 

stereophotogrammetric, to mark the facial characteristics in 65 pre-pubertal boys from 

different subgroups of autism. They found that the distance between the eyes and the 

nose or the nose to the mouth is slightly shorter than those of 45 typically developing 

children. This information cannot be detected by merely looking at those affected with 

the disorder and there is no indication if it is the same in pre-pubertal girls with ASD as 

well. The findings were sustained in another study by Gilani, Tan, Russell-Smith, 

Maybery, Mian, Eastwood, Shafait, Goonewardene, and Whitehouse (2017).  

In a follow-up study by Tan, Gilani, Maybery, Mark and Whitehouse (2017), it was 

discovered that autistics had lesser muscular features in their facial makeup. Both the 

boys and the girls tend to have plumper facial features than their typically developing 

counterparts. Still, these findings are not very satisfactory to parents and guardians of 

affected children because these features are not distinct enough to eradicate doubts 

(Loftus, 2021). These parents believe that people would have been easier on their 

children if their conditions and struggles were immediately visible. 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

11 

2.1.1.5 Challenges of autism  

As part of the challenges encountered by the population of autistic people, it has been 

estimated that one-third of them are non-verbal, while 31% of them have intellectual 

disability with significant challenges in their daily functions (Tager-Flusberg & Kasari 

2013); they are faced with bullying and about 28% are reputed to be self-injurious. In 

consonance with a report submitted by National Autism Association (2014), drowning 

remains the leading cause of death among them, as 80% of deaths are associated with 

wanderings and bolting by those younger than 14 years. They are mostly marked by 

difficulty to interact socially, restricted interests, sameness of activity, and they are 

hypersensitive to lights, sounds, tastes, and so on. Also, they exhibit difficulty in 

corresponding conversation and nonverbal conversation skills such as distance, 

loudness, tone, and so on (Edward-Elmhurst, 2022). Their movements are uncoordinated 

and clumsy, and they suffer anxiety and depression. With early diagnosis and 

intervention, many are able to overcome some or most of these challenges. 

2.1.1.6 Causes of autism 

Although there are no known causes for ASD, some researchers have identified some of 

the contributing factors that could influence the presence of the disorder in a carrier.  

i. Some genetic factors that contribute to the disorder are largely present in some 

families which increase the risk that a child born in that family may develop 

autism. These genes in themselves do not cause autism but contribute to the 

chances of the disorder (Chaste & Leboyer, 2012).   

ii.  Children born to older parents may have increased risks of developing ASD 

(Deweerdt, 2020). 

iii. Parents who already have a child with ASD have a 2 to 18% chance of having a 

second child who is also affected (Meyer, 2013).  

iv. Studies have shown that among identical twins, the chances of the other 

developing the disorder when one has autism is about 36 to 95% and there is a 

31% chance in non-identical twins (Frazier, Thompson, Youngstrom, Law, 

Hardan, Eng & Morris, 2014).  

v. The notion of childhood vaccination being a contributing factor to ASD is false 

(Nwokolo, 2010).  

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

12 

vi.  Other contributing factors may include advanced age of either or both parents, 

complications during pregnancy or at birth, such as extreme prematurity, low 

birth weight, multiple pregnancies, and pregnancies spaced less than a year apart.  

2.1.1.7 Language in autism 

Siegal and Blades (2003) submit that autism is characterised by varying degrees of 

disorders in language, communication, and imagination. It has disproportionate 

language impairment that requires further investigation. Tager-Flusberg and Joseph 

(2003) aver that low intelligence quotient is consistent with the evidence that the level 

of language ability which is linked to the severity of autism correlates with lack of 

communication and reciprocal social interaction that is symptomatic of autism. Many 

ASD carriers share characteristics of children without ASD but have specific language 

impairment (SLI), in that although both have similar grammatical errors, ASD 

communication has more interest in objects than in people.  

According to American Psychological Association (1994), issues affecting autism are 

abnormal language, atypical intonation, vocal quality, idiosyncratic use of words and 

stereotypical phrases, echolalia and pronoun reversal. Because of the versatility of 

language behavioural display in autistics, Tager-Flusberg and Cooper (1999) suggest 

that autism cannot be defined by any individual language phenotype. Tager-Flusberg 

and Joseph (2003) further submit that language investigation can be more homogenous 

within a particular subtype population of ASD.  

Generally, from year 1, young children with autism are less responsive to their names or 

their mother's voice. They usually acquire some words at 12-18 months, and then lose 

them in what Lord et al. (2004) describe as 'language regression'. During this time, the 

child does not learn new words and they fail to engage in communicative routines which 

they may have been participating in before. They also experience loss of social skills. 

At age 2, they have the expressive and receptive competence of a typical 9-month old 

baby and exhibit severe delay in language development. This delay continues till age 5. 

From then, Tager-Flusberg (2001) notes that there is often a significant correlation 

between intelligence quotient and language outcomes, although higher levels of 

nonverbal intelligence quotient IQ are not always associated with a higher level of 

language skills. Their level of language improvement can be related to their social 

relations and background in late adolescence. Paul and Cohen (1984) submit that, all 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

13 

through adolescence and adulthood, comprehensive and expressive abilities continue to 

improve in ASD carriers and the latter shows greater improvement rates. Cantwell and 

Baker (1989) say that ASD adult carriers show substantial improvements in formal 

aspects of language but are deficit in receptive language skill which becomes more 

severe than it was at adolescence or in early childhood. Tager-Flusberg (2005) concludes 

that there is some optimism for children who receive early diagnosis and access to 

intensive intervention, especially in language and communication skills. 

2.1.1.8 Language behaviours among subtypes 

Pervasive developmental disorders profile children with a milder form of ASD and 

might reveal lesser impairments because carriers tend to have better developed language 

skills and lower levels of developmental delays. Chawarska and Shic (2009) and Streiner 

et al. (2016) report that the social identification of ASD appears from the age of 2; this 

social identification ranges from limited to declining interest in the social world. Carriers 

have a limited interest in social face-to-face interactions and they seek physical comfort 

in their parents. They respond infrequently to being called for social interaction, and do 

not return a social smile. They exhibit poor facial gestures including eye contact, and 

their affective expressions and gaze cues are unrelated to the activities of others. 

Asperger's syndrome is not mostly characterised by lack of language delays and 

cognition but by marked social deficits. Carriers may possess a slower ability to interpret 

or understand verbal and non-verbal language, such as gestures, tone of voice and lack 

of social skills themselves (Tager-Flusberg & Kasari, 2013). Its existence suggests that 

even though abnormalities in communication are a core feature of pervasive 

developmental disorders, slower language acquisition is not sufficient for a conclusive 

diagnosis within the spectrum disorders associated with autism, however, other aspects 

of language such as comprehension may be indicative.  

Paul, Shriberg, Black and Santen (2011) note that delayed vocalisations appear very 

early in ASD and may be the first indication of ASD. By the end of the second year, 

deficit in understanding other people’s gestures and verbal communication or the 

inability to initiate communicative acts become very pronounced. The ability to 

understand and respond to language is often more impaired than the ability to produce 

speech sounds (Chawarska & Shic, 2009; Ellis, Venker, McDuffie & Abbeduto, 2012). 

Toddlers with ASD direct their vocalisations to others less frequently and show atypical 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

14 

intonation in their speech (Brian, Bryson, Zwaigenbaum, McDermott & Rombough, 

2008). Streiner (2016) continues that, gestures such as pointing, showing or giving 

objects for the purpose of sharing interests or making requests are absent; hence, these 

characteristics differentiate ASD from other language delays without disability. The 

unusual linguistic features in ASD include echolalia and scripting, that is verbatim 

repetition. Other features include repetition of utterances and phrases previously heard 

from books, movies, and other sources and abnormal pitch or intonation becoming 

apparent (Schoen et al. 2016). 

2.1.1.9 Autism in Nigeria  

According to Nwokolo (2010), the first mention of autism in Nigeria can be traced to 

1943 when children with ASD were thought to be mentally retarded, deprived of 

maternal affection and the condition was thought to be spiritual. Unfortunately, many 

years after, this is still the norm in most rural communities where there are poor 

healthcare systems. People living with autism are thought to be possessed or evil. The 

level of awareness as to the nature of the condition is pathetically low, even within the 

medical community and it affects the social interaction and language development of 

autistic people in Nigeria (Bakare et al., 2008; Igwe et al., 2011; Ojo, 2012; Bakare et 

al., 2015; Adeosun et al., 2017). 

Many children are undiagnosed or misdiagnosed, hidden at home or labelled as mentally 

ill, deaf and dumb, or in other cases, they end up on the streets tagged as insane. 

However, this is not the case in advanced countries, where a multidisciplinary team is 

available to assess, diagnose and develop an individualised educational plan for each 

child with autism. Exorcism seems to be the most common treatment in Nigeria. The 

prevalent ignorance about ASD in the Nigerian society affects the country to such an 

extent that mothers are blamed for their children’s autism. They are stigmatised and 

discriminated against even within the medical community, where such has gone for help. 

Other issues are the lack of welfare programmes to cater for carriers, the absence of 

governmental funding for special needs and very poor educational plan. Mostly, parents 

are left to cater to their affected children on their own. Only a few affluent ones are able 

to afford the best management and remedy for their wards (Falodun, 2021).  

Currently, a handful of special needs schools have been established in some parts of the 

country to cater for autistics and other atypical children at a price that is unaffordable by 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

15 

the poor. Some of the coordinators of such places are not fully trained, while some of 

them are self-motivated parents, educators and medical personnel who started the school 

as a means to cater for their own atypical children. Falodun (2021) submits that the 

problems that still characterise the autism situation in Nigeria include delayed diagnosis 

due to parents and medical personnel’s lack of knowledge about ASD, inadequately 

trained personnel, dearth of adequately trained personnel, lack of health facilities, and 

financial obstacles. 

Another major problem of autism in Nigeria is the lack of information about the actual 

number of autistic population because there is yet to be any exercise to discover the 

actual number of cases across the states. What exists are sectional studies of prevalence 

in some parts of Nigeria such as Ojo (2012) and Adeosun et al. (2017) from south-

western Nigeria; Chinwa et al. (2016) from south-east; and Ahmed et al. (2018) from 

north-east Nigeria. These do not give a holistic view of what the autistic population in 

Nigeria is, as the studies were not homogenously carried out. There is a need for a larger 

research in which every state in the country will be represented in order to get a better 

idea of the ASD population in Nigeria. Nwokolo (2010) concludes that there are still so 

many cases of undiagnosed children with ASD in Nigeria today and even those that have 

been diagnosed do not enjoy adequate care. There is a need for the government to include 

the welfare and funding of atypical children in their policies as many still rely on aids 

and funds from international communities like USAID, UNICEF, and UN which are 

barely adequate and mostly misappropriated. 

2.1.1.10 Intervention in Autism 

Intervention is the action taken to improve a medical condition or disorder. With so 

much emphasis laid on the benefits of intervention, there has been much optimism and 

advocacy for early diagnosis and intensive intervention as being the basic means of 

improving and managing ASD over the years; thus, this has made it necessary to 

examine intervention in this study. The main goal for early diagnosis is to provide 

optimal strategies for facilitating the child’s development and optimise short and long 

term outcomes. There are numerous studies that report the effectiveness of early 

intervention on preschool children (Golan and Baron-Cohen et al. 2006). Current Best 

Practices (CBP) of the National Research Council (2001) suggest that programmes 

should begin early, as soon as the child is diagnosed and that these programmes should 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

16 

be individualised to each child’s needs. Additionally, they must be intensive and 

comprehensive, with a minimum of 25 hours per week, addressing various areas of 

developmental needs, such as parent education and training. Furthermore, there should 

be a database for keeping the records of the child’s progress (Woods & Weatherby, 

2003).  

Goldstein (2002) avers that non-speakers within the autistic population are decreasing 

with the prevalence of early intervention. This is reiterated by Tager-Flusberg (2005) 

who says that the proportion of children with ASD who fail to acquire functional 

language is diminishing; the increase in diagnosis equals an increase in improvement 

through intervention. Myles (2002) submits that written materials have been shown to 

provide a helpful medium of intervention for ASD which helps to increase 

communicative and social behaviour. Bakare et al. (2008) and Bakare et al. (2011) 

further state that early diagnosis and intervention have been known to improve prognosis 

in children with ASD. Therefore, late presentation for diagnosis leads to late 

intervention.  

In Nigeria, children averaged eight years old are diagnosed with ASD due to low levels 

of knowledge and awareness about the disorder, and this in turn affects intelligibility 

because early diagnosis and intervention aid speech and language development. 

Therefore, it is of utmost importance to diagnose and intervene early for an improved 

quality of life. Most popular and available special needs intervention services in Nigeria 

are accessed at prices which are mostly affordable only by the rich in the country. Most 

popular among the services rendered in the country are Behavioural, Speech and 

Educational therapies/services. However, the following services are proposed by the 

Autism Speakers Resource Guide (2019) among others.   

1. Cognitive behavioural therapy can help to address anxiety and other unwanted 

personal behaviours that challenge affected children.  

2. Social skills training classes can help to understand conversational skills and 

social cues in order to aid better social interaction in public.  

3. Speech therapy can help with voice control to effectively improve 

communication and interaction skills. This will encourage improved social skills 

in communication and language development.  

4. Occupational therapy can improve coordination. The therapy is useful for 

building up coordination skills which should match their levels of need. 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

17 

5.  Physical therapy helps to create activities that will encourage exercises. These 

exercises will aid good posture and balance, and will develop motor control in 

games or sporting activities.  

6. Psychoactive medicines help to manage associated anxiety, depression and 

attention deficit and hyperactivity disorder (ADHD). 

7. Training for parents, guardians and service providers is a crucial aspect of 

management that cannot be overlooked. Everyone who wants to handle an 

autistic child must be armed with the necessary skills to assist people living with 

autism and other special needs.   

2.1.2 Phonotactics 

Phonotactics is the study of the rules governing the possible phoneme sequences in a 

language. The word is formed from a Greek word ‘phone’ which means sound or voice, 

and ‘tacticos’ which has to do with arrangement. It is the part of phonology that restricts 

the possible sound sequences and syllable structures in a language. Syllables cover 

onsets, codas and nucleus. The syllabic structure of English is (C0-3) V (C0-4), where C0-

3 and C0-4 are the optional elements and V is mandatory (Crystal, 2003; Roach, 2010; 

Yule, 2010; Osisanwo, 2012).  

Furthermore, phonotactics is concerned with the rules and constraints that govern a 

language. In English, some sounds do not begin a word or syllable and some other 

sounds do not end a word or syllable either. For example, /ʒ, ŋ/ do not begin a word and 

/h/ is not accepted at the final position of a word. The rare example of genre that begins 

with /ʒ/ in English is a loan word. Even though English allows C0-3 and C0-4 to cluster at 

initial and final positions, respectively, it restricts the choice of phonemes allowed to 

occupy those positions and sonority constraints must be applied on the choice of 

phonemes combined. For instance, at the onset position, mainly stops, fricatives or 

glides, liquids or voiceless stops, nasal + voiceless stops + liquid, glide can occupy this 

position, while at the coda position, nasals may precede voiceless plosives, but only if 

they share the same place of articulation as we have in words like jump /ʤʌmp/, stunt 

/stʌnt/, and stink /stiŋk/ (Cystal, 2003; Bailey & Hahn, 2005). 

According to Gaygen (1997), prohibited sound sequences may occur for different 

reasons including borrowing from other languages. For example, Tsunami is realised as 

/sunæmi/ (for non-Japanese speakers) or /tsunæmi/ for Japanese speakers. Also, putting 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

18 

prefixes and words together may bring about prohibited sequences. For example, ‘sts’ 

in posts, and ‘stst’ in textstore. The following phonological processes have been found 

to be employed in treating sequence overlaps in English. 

i. Deletion: a sound is deleted to make the process acceptable. For example, it 

is possible for textstore /tekststɔ/ to become /tesstɔ/, deleting the voiceless 

velar stop /k/ and the voiceless alveolar stop /t/, and allowing the merge of 

the voiceless alveolar fricatives /s/. while Roach (2010) sees this as a form 

of mastery of language in a connected speech, Soneye and Faleye (2015) 

attributed it to a feature of Educated Nigerian Speaker of English (ENSE) 

because voiceless velar stop /k/ in a triple consonant cluster at coda segment 

is always deleted. Also, alveolar clusters are not permitted in word medial 

position of an ENSE.  

ii. Insertion: sometimes, a sound is inserted to make the sound sequence in a 

word obey the phonotactic rules in a language. For instance, among the 

uneducated Yoruba English speakers, some words with clusters like bread 

may experience insertion and become ‘buredi’, inserting /u, i/ into the word 

(Osisanwo, 2012). This claim was also made by Adegite and Akindele 

(1999), however, this assertion is due for review because ‘buredi’ is a Yoruba 

rendition of the English word bread and is only used within Yoruba sentences 

and structures, not English ones. Thus, ‘buredi’ in a Yoruba construction is 

linguistically equivalent to bread in an English construction. 

iii. Assimilation: a situation where the place of articulation tends to change to 

the place of articulation of the following phoneme is known as assimilation. 

In phonotactics, this process disallows adjacent stops if they differ in voicing. 

For example, talk + ed – [d] becomes /tɔ:kt/ 

Orthography and pronunciation are a major issue in phonotactics. Phonotactics guides 

the number of syllables permissible in words; it guides what phonemes are permissible 

in what positions; it dictates the types of consonants and vowels that can combine within 

a syllable. Furthermore, it determines the presence or absence of consonant clusters at 

the beginning or ending of a syllable, it determines the number, type and sequence of 

consonants that constitute a consonant cluster. Phonotactics is the area of phonology 

concerned with the analysis and descriptions of permitted sound sequences in a 

language. 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

19 

Word phonotactics means that only clusters that can begin or end a syllable can also 

begin or end a word. In multisyllabic words, clusters consist of syllable-final and initial 

sequences. Phonotactics is highly language-specific. Glottal sounds are not permitted to 

come before nasal sounds as we have in ‘kn’ and ‘gn’ in English words – this accounts 

for the silencing of such glottal sounds.  

The internal structures of syllables are as follows: onset, nucleus and coda. The onset 

and coda are the optional elements of a syllable, while the nucleus is the only obligatory 

item. The syllable is further divided into two parts – the onset, which is the optional 

element, and rhyme, the obligatory element comprising nucleus and coda. Nucleus is the 

only obligatory element in a syllable, thus, it is able to form a vowel-only syllable, while 

coda, just like the onset, is optional. Syllabic consonants, represented by semi-vowel 

glides and liquids, can also stand in for a nucleus in a syllable (Hammond, 2004; Roach, 

2010). 

Phonotactics has been known to have an effect on the process of second language 

vocabulary acquisition. According to Harley (2003), constraints that affect the English 

phonotactics are 

1. Syllables must have a nucleus. 

2. Consonant doubling is prohibited, that is, gemination. 

3. The velar nasal sound /ŋ/ cannot occupy an onset position. 

4. The glottal fricative /h/ cannot assume the coda position. 

5. Affricates or glottal fricative /h/ cannot be in a complex onset 

6. An obstruent (plosives, affricates and fricatives) must be the initial consonant in 

a complex onset for example, a blend of obstruent sounds such as *ntat or *rkoop 

are not permitted. 

7. It is impossible for the alveolar fricative /s/ to come after an initial obstruent in 

a complex onset. It has to be a liquid /l, r/ or a glide /w, j/. 

8. All relevant phonotactic rules must be obeyed when there is a subsequence 

within a sequence of consonants (the substring principal rule). 

9. Glides are not permitted in the coda of a syllable, except the off-glides in 

diphthongs. 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

20 

10. These approximant, nasal and fricatives /r, ŋ, ʒ or ð/ cannot occupy the second 

consonant in a complex coda, for example, asthma is mostly pronounced as 

/ӕzmə/ or /ӕsmə/ but rarely as /ӕzðmə/. 

11. There is an assimilation interplay in a complex coda between the second 

consonant which may become voiced if the first consonant is voiced. 

12. Any stop coming after the nasals /m/ or /ŋ/ in the coda must assume the same 

place of articulation with them and become homorganic with the nasals. 

13. It insists on voice sharing for any two obstruents in the same coda position. For 

example, kids /kɪdz/ and kits /kɪts/. 

2.1.2.1 Sonority sequencing principle 

The elements of a syllable are not just assumed, they must be universally distributed 

following a set of rules known as the sonority sequencing principle (SSP). The SSP 

posits that the nucleus has the highest or maximal sonority in any syllable and that the 

farther a phoneme is from a nucleus, the less sonorous it is. The ranking of sonority in 

the hierarchy of any speech sound is language-specific but it is hardly different from one 

to the other. This is because almost all languages of the world create their syllables 

approximately in similar ways when it comes to sonority (Jany, Gordon, Nash & Takara, 

2007). 

The voiceless alveolar fricative, /s/, has a lower sonority hierarchy in comparison to the 

alveolar lateral approximant /l/, thus, permitting /sl/ to co-exist at the onset and /ls/ at 

the coda is also allowed, but not the other way around. Therefore, /slɪps/ and /pʌls/ are 

permitted but not /lsɪps/ or /pʌsl/. The SSP has a cross-lingual proneness but refrains 

from accounting for patterns with complex syllables which may violate the rule (Robert, 

2001).  

2.1.3 Syllable 

A syllable is the smallest unit of sound which can be produced (Egbokhare, 1994). It 

mediates between the individual segment units: consonants and vowels, and combines 

them into words. Roach (2002: 277) and Yule (2010: 46) describe the syllable as a beat 

in an utterance, which must contain a vowel or a vowel-like sound. It may or may not 

be accompanied by a consonant sound(s) preceding or following it. Esan (2018:7) 

simply describes it as the minimum feature of prosody. 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

21 

 

 

 

 

 

 

 

       

 

 

        

 

 

 

Figure 2.1 The syllable schema   (Adapted from Yule 2010: 46) 

 

 

 

 

 

 

 

 

 

 

 

 

Syllable 

Onset         Rhyme 

Nucleus          Coda 

    Consonant(s)  Vowel          Consonant(s) 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

22 

 

Roach (2010: 56) describes four types of syllables as follows: 

a. Minimum syllable containing only a single vowel in isolation, ‘are’ /ɑ:/, ‘or’ /ɔ:/, 

‘err’ /з:/ 

b. Syllable with onset, for example, ‘bar’ /bɑ:/, ‘key’ /ki:/, ‘more’ /mɔ:/ 

c. Syllable with coda, for example, ‘arm’ /a:m/, ‘ought’ /ɔ:t/, ‘ease’ /i:z/ 

d. Syllables with both onsets and codas, for example, ‘ran’ /rӕn/, ‘sat’ /sӕt/, ‘fill’ 

/fɪl/ 

 

2.1.3.1 The syllabic structure 

 

The syllabic structure suggests the presence of consonant clusters in English. Consonant 

clusters do not exist in some Nigerian languages, especially the Yoruba language. The 

nucleus is the important part of the syllable which is mostly represented by a vowel or a 

vowel-like sound. It may, or may not have an onset or a coda, that is, consonants before 

or after the vowel. Syllables like these with an onset like ‘no’, ‘me’, ‘to’, and so on are 

open syllables because they have no codas. They are also referred to as zero coda (Roach, 

2010); words like ‘up’, ‘at’, ‘cup’, ‘at’, and so on are closed syllables whether they have 

onsets or not. Roach (2010) expresses this in the following schemata:  

 

 

 

 

 

 

 

 

 

 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

23 

 

 

 

 

 

 

Vowel 

 

 

 

 

 

Figure 2.2 The syllable structure (Roach 2010) 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Pre-Initial Initial     Post-initial 

 

           Onset 

Syllable 

Pre- Final    Post-         Post- 

Final  Final (1)   Final (2) 

  

  Coda 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

24 

 

1. Onsets and codas 

Onsets and codas can have more than one consonant – an onset can have up to three 

consonants at the onset position and up to four for coda. However, the appearance of /ŋ, 

ʒ/ as onset is rare. Combinations that are acceptable in lesser clusters, that is, a CC or C2 

structure, are stops and fricatives at the pre-initial onset position and liquids /l, r/ or 

glides /w/ at initial position. However, clusters in a CCC or C3onset structure must be 

an ‘s’ at the pre-initial, followed by a voiceless stop /p, t, k/ at the initial position and 

then at the post initial position is a liquid /l, r/ or a glide /w/. For example, C = ‘to’, ‘see’, 

‘me’; CC = ‘free’, ‘tree’, ‘dew’; CCC = ‘splash’, ‘strong’, ‘scream’, ‘square’. 

In English, no words end with more than four consonants. Nevertheless, /h, w/ at the 

coda position are also rare. At the coda position, the pre-final position is mostly 

represented by /ŋ, n, m, s/, while the final position constitutes /p, t, k, d, Ө/; at the post 

final position, (1), /s, z, t, d, Ө/ are usually found, and at post final position (2), /s, t/ are 

found. For example, C = ‘sees’, ‘boys’, ‘kissed’, ‘bored’, ‘teeth’, ‘sing’, ‘corn’, ‘come’; 

CC = ‘bump’, ‘bumped’, ‘bank’, ‘bald’, ‘eighth’; CCC = ‘helped’, ‘banks’, ‘fifths’, 

‘bonds’, ‘lapsed’; CCCC = ‘twelfths’, ‘prompts’, ‘attempts’, ‘strengths’, ‘texts’, etc. 

2. Nucleus/Peak 

The only obligatory segment of a syllable is the nucleus, represented by vowels or 

syllabic consonants. Roach (2010) describes the syllable represented by vowels as a 

minimum syllable and submits that any vowel can represent a minimum syllable. 

However, /ʊ/ is rarely a minimum syllable. Mohamed (2018) analysed the nucleus as V, 

V:, VV, VVV, and C, quoting Harris (1994). Expressed as monophthongs: V = short 

vowel, V: = long vowel, VV = diphthongs, VVV = triphthongs mostly at word final, 

further realised as V3 = V1 V2 V3 (author’s concept). 

Simple nucleus + coda = hut, bit 

Simple nucleus + complex coda= felt, hands, tempt 

Complex nucleus + complex coda = feels, joint, fields, points. 

The rhyme template is presented as V, VCC, VCCC, VCCCC; V:, V:C, V:CC, V:CCC, 

VV, VVC, VVCC, VVCCC. 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

25 

Mohamed (2018) presents 20 possible syllabic/monosyllabic structures as: V, CV, CCV, 

CCCV, VC, CVC, CCVC, CCCVC, VCC, CVCC, CCVCC, CCCVCC, VCCC, 

CVCCC, CCVCCC, CCCVCCC, VCCCC, CVCCCC, CCVCCCC, CCCVCCCC.  

From the foregoing, it is worthy of note that the syllabic structure can also be written as 

C0-3V1-3C0-4, which can further be expressed as C1C2C3V1V2V3C1C2C3C4 in the 

expanded form.  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

26 

 

 

Table 2.1 The syllable structure 

Onset  Nucleus  Coda  Rhyme  

C V C V 

CC V: CC V: 

CCC VV CCC VC, VCC, VCC 

 VVV CCCC V:C, V:CC, V:CCC 

 C  VV, VVC, VVCC, VVCCC 

   VVV 

       

(Adapted from Mohamed, 2018: 107) 

 

 

 

 

 

 

 

 

 

 

 

 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

27 

The above shows that monosyllables are words made up of one syllable or a segment in 

an utterance as seen above. Disyllables and poly/multi-syllable words imply words with 

two or more syllables that have varying phonotactic composition. However, the 

explanation above can account for syllabic segments in such multi-syllabic words. 

2.1.4 Studies on autistic phonology  

Schoen, Paul and Chawarska (2011) study the phonology and vocal behaviour of 30 

autistic toddlers between the ages of 18 and 36 months. They were paired with 11 age-

matched and 23 language-matched controls. The result showed that toddlers with ASD 

produced speech-like vocalisations similar to those of the language-matched peers, but 

produced significantly more atypical non-speech vocalisations when compared to both 

groups. Also, the speech-like sound productions made by the autistic toddlers can be 

linked to the level of their language acquisition as seen in the language of typically 

developing toddlers. However, the study is silent about the children’s intervention, 

whether it was before or during the period of data collection, this would have impacted 

the progress that the children made, and should have been presented in the study. Also, 

phonology as referred to in the study is merely an observation of the vocal noises that 

the children made and not really their speech sounds. Thus, the phonological perspective 

which the study claims is quite questionable.  

Ojo (2012) explores the phonological issues in twelve autistic people’s speech. The 

phonological disorders discovered in their utterances were consonant elision at 

the initial and final positions, substitutions between sounds that have the same place or 

manner of articulation, and inconsistencies in English stress placement. The 

study emphasises that autistics have limited issues with vowels but are largely 

divergent with consonants. However, the study’s attention on vowel and consonant 

phonemes is limited to the segmental level, while the peculiar combination of these 

segments is not explored. Furthermore, the focus of the study is on autistic people’s 

English phonology for the purpose of description only, and not to create a template 

for identifying their phonology.   

Nwanze’s (2013) study investigates the arguable notion that language skills in autistic 

African children are inferior to those of the autistic children in the global west. This 

investigation engaged the Nigerian children to investigate the veracity of this claim. The 

spoken language of Nigerian children who manifested features of autism was used to 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

28 

carry out this investigation. The study sampled 145 children with ASD, aged 3-10 years 

with a ratio of 5 boys to 2 girls. The study reveals that 54% of children with ASD are 

non-verbal, while the verbal ones have a mean length utterance (MLU) of over three 

morphemes; also, the study concludes that there is minimal difference in the 

performance between boys and girls. However, the study is limited to the length and 

pace of the children’s utterances, without in-depth consideration of any core area of 

phonology or the language acquisition process. There was also no clear conclusion to 

the research question that the study set out to answer, that is, whether the language skills 

of autistic children in Africa were inferiror to those in the global west. 

Nordgren’s (2015) longitudinal study of one autistic subject concludes that consistent 

and early intervention in children with ASD could improve their sound production 

qualitatively and quantitatively. Although this submission is widely corroborated in 

available literature on autism, Nordgren’s other submissions are quite arguable because 

they are generated from the study of only one child and these submissions are unlikely 

to be generalisable because of such factors like idiosyncrasies. However, this assertion 

about a consistent and early intervention is corroborated by Wolk, Edwards and Brennan 

(2016) in their investigation of the phonological difficulties faced by children with 

autism.   

In another study, Brennan, Wagley, Kovelman, Bowyer and Lajiness-O’Nell, (2016) use 

magnetoencephalography to test the cascading effects of speech sound processing in 

children with ASD and age-matched controls of 8-12-year-olds. Results show that 

autistic participants produced non-problematic sequences around 330ms after onset of 

critical phoneme which suggests that ASD affects phonological processing in autistics’ 

speeches. Majority of the results were within the core of neurological studies, thus, 

sidelining linguists in the presentation of the findings. There is a need to discuss the 

results in relation to linguistics since it concerns language, that way, actual progress in 

autistic utterances can be measured. However, the study successfully reveals the extent 

to which advanced countries have gone in language studies.   

Kjellmer, Fernell, Gillberg, and Norrelgen (2018) examine the speech and language 

profiles in 4 – 6 year-old intellectually capable children who have been diagnosed early 

with ASD. The result shows that almost 60% have moderate to severe language 

problem, and nearly half (46%) exhibited combined expressive and receptive language 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

29 

problems. Phonological speech problems were found in 21% of the total group 

and almost 1 in 4 show expressive but not receptive problems. These suggest that the 

level of developmental language disorder may be more damaging than expected. The 

study reveals that audible phonological speech problems tend to decrease with age. 

These findings also show that language challenges in autistic people is not as a result 

of intellectual disability because this study established that even intellectually capable 

children still experience language challenges because of autism.   

Umera-Okeke and Iroegbu (2016) identify some features such as substitution, deletion, 

and epenthesis in children with autism and Down syndrome among others. Esan (2018) 

corroborates these findings in her investigation of the prosodic features of people with 

developmental and acquired language disorders which covered most special needs 

group including autism. In addition to the above findings, she observes that simpler 

syllabic structures are preferred to complex ones. However, only adults were mainly 

sampled in this study, also, only one autistic was sampled and there was no background 

information provided about this autistic participant, whether the subject had enjoyed 

some sort of intervention and for how long, or not at all.  Also, one subject is not at all 

representative of the autistic population.  

A few of the studies above have established that language is the first step to diagnose a 

child with ASD (Schoen et al., 2011; Baron-Cohen, Lambardo & Lai, 2013); some have 

examined the phonological disorders and difficulties that autistic children encounter as 

they grow older (Schoen et al., 2011; Ojo, 2012; Nwanze, 2013; Umera-Okeke, 2016; 

Esan, 2018); and others have investigated the influence of intervention on improving the 

speech outcome in autistic children (Nordgren, 2015; Wolk et al., 2016; Kjellmer et al., 

2018). The most central to the current study is Esan (2018) which examines the syllabic 

structure of one autistic adult; thus, there is still much to do regarding the investigation 

of the phonotactics of autistic children’s spoken English in Nigeria. This area of 

phonology is necessary because it creates an avenue to examine how this population 

connect segmental sounds to form syllables and words, thus, showing their language 

progression.  

2.1.5 Studies on phonotactics   

Phonotactics as a subject matter has not enjoyed so much attention from scholars, thus, 

there are sparse scholarly materials on phonotactics which constitutes a major focus of 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

30 

the study. The review of the available materials on this subject helps to establish the 

extent of phonotactic studies in Nigeria.   

Luce and Large (2001) examine phonotactics, density and entropy in spoken word 

recognition to extract the combined effects of probabilistic phonotactics and lexical 

competition. This is done by generating words and non-words that are varied 

orthographically and in phonotactics. Forty-five English-speaking adults were give 45 

CVC words to produce in a sophisticated technology laboratory. Each adult had about 3 

seconds to respond to a pre-pronounced word, after which the computer automatically 

recorded correct or incorrect against each token after the allotted time limit. Result 

confirmed and extended previous studies, it also uncovered the effect that cannot be 

predicted by a simple combination of activation and combination at sub-lexical and 

lexical levels of representation under specific circumstances. Neighbourhood density 

and probabilistic phonotactics may combine to produce non-addictive or synergistic 

effects of lexical competition on processing time. This study, though very contributory, 

is not grounded in any phonological or linguistic theory, making the study less 

linguistically or phonologically inclined, since results were only presented and discussed 

statistically.  

Btoosh (2006) used Optimality Theory to examine constraint interactions in Jordanian 

Arabic phonotactics. Karak-Arab (KA), which is the language of Btoosh’s interest, 

treated the syllable using two major phonological processes: re-syllabification and 

epenthesis. The study compares past Arabian studies in phonotactics to the current study 

and then goes on to problematic words that were previously ignored by those studies. Its 

analysis revealed the ideologies in previous studies, the assignment of unlicensed 

segments to semi-syllables, the representation of geminates and the reduction of 

unlicensed segments in ultra-heavy syllable by morphology. The findings show an 

extremely tiny difference between substitution of ALIGN (R) and ALIGN (W) 

dominating Arabic. Syllables in KA, a CV language, are full of compliance to default 

onset constraint and without an onset, the language resorts to either an onset-motivated 

epenthesis or an onset-motivated re-syllabification process to repair such a syllable. As 

rich as the contributions of this study are, the paper lacked proper structuring and this 

downplayed the various aspects of the work. Also, the constraints relevant to the study 

are not presented systematically, making the constraints difficult to follow. The 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

31 

researcher is also an informant in the study; thus, the objectivity of the work is quite 

questionable. Finally, no justification was provided for the choice of data used.    

In another study where a group of pre-school-aged children with specific language 

impairment (SLI) were presented for study along with some typically growing peers, 

Leonard, Davis, and Deevy (2007) found out that items of low phonotactic probability 

distinguished the children with SLI and Typically Developing children matched for Age 

(TD-A) from Typically Developing children matched for Mean Length of Utterance 

(TD-MLU). The study examined phonotactics and the use of regular past tense 

morphemes and inflections in novel verbs, this differed from their phonotactic 

probabilities. These children were less likely to use the past tense inflection ‘–ed’ with 

novel verbs of low phonotactic probability than with novel verbs of high phonotactic 

probability. As insightful as the study is, it lacks the proper use of basic terminologies 

in presenting the processes involved in the children’s interactions with data. Result was 

only discussed statistically; it did not employ any linguistic or phonological theory in 

the study.  

In another study, Mbah and David (2013) examine the constraint-based study of Tiv 

learners of English phonotactics in order to note the possible constraints the Tiv learners 

of English as L2 encounter. The objectives of the study are to analyse, ascertain and 

unveil the challenges that Tivs encounter in sonority hierarchical arrangement of 

phonemes in the syllable structure of the English language. The study reveals that Tiv 

learners of English violate the phonotactic rules of English in their realisation of 

consonant assimilation and vowel harmony and that these learners experience 

difficulties in realising English syllables with consonant clusters and this causes transfer 

of features from L1. The result also shows that Tivs have onset restrictions for syllables 

that is, no vowel can begin a syllable except a word. Aside updating the literature, this 

study contributes no new knowledge to what has been presented by earlier studies as 

reviewed in the work except the adoption of Optimality Theory in the analysis which 

gave a new perspective, although it was not explore in-depth. Analysis did not really 

match the objectives set. In other words, the objective of the study should have indicated 

that the work would rank the constraints responsible for the Tiv phonotactics; hence, 

these weakened the findings of the study. Also, no methodology was presented to 

provide information about how the study was carried out, if the participants were 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

32 

educated, if their socio-cultural background affected the way they spoke or, if all Tiv 

phonotactics were expected to be such.  

Soneye and Faleye (2015), in their study of syllable phonotactics in Educated Nigerian 

Spoken English (ENSE), aimed to bridge the gap aiding the dearth of corpus-driven 

research in the area of focus. Using 31 postgraduate respondents, with a higher 

percentage of respondents from the southwestern region of the country, the study 

revealed that whenever C1 is /k/ in a triple-consonant-clustered coda environment, /k/ 

was always deleted. Substitution was rarely employed in the data, it occurred only when 

voiced alveolar plosive /d/ replaced its voiceless counterpart and voiced alveolar 

fricative /z/ is devoiced. The lateral /l/ is usually deleted in post-vocalic pre-consonantal 

positions.  Alveolar clusters [str] at medial position is not permissible in Educated 

Nigerian Spoken English (ENSE), examples are ‘registration’, ‘restriction’, 

‘infrastructure’, and so on. These were pronounced without the /t/. Deletion of 

consonants in 3-consonant-cluster English words was a prominent feature in ENSE 

and may be attributed to other social factors since most respondents were from the south-

west. Reduction and simplification could lead to different productions which could also 

lead to wrong perceptions and serious implications. As insightful as these contributions 

are, they are incomplete without the onset. The authors admit that participants from the 

south-west were more than the other respondents and this may have influenced the 

results to some extent. A reconfirmation of data may be needed from the northern and 

eastern regions of the country to ensure that the findings are substantial. 

Young children’s sensitivity to probabilistic phonotactics in their developing lexicon 

was the focus of Coady and Aslin (2017) who examined 24 children - 12 two and a half 

and 12 three-and-a-half year olds - to test their sensitivity to probabilistic phonotactic 

structure. The high-phonotactic frequency contained non-words with phonemes that 

were always frequent in syllable segments while the low-phonotactic frequency group 

contained least frequent sounds in a syllable segment. The result showed that these 

children responded mostly to the high phonemes than the low ones, indicating sensitivity 

to phoneme frequency. Also, additional sensitivity to individual segments increased with 

age. Older children were more responsive to frequency of larger units. Young children 

are reactive to fine-grained acoustic-phonetic information in the developing lexicon, 

while sensitivity to all aspects of the sound structure increased over development and 

were mostly responded to by the older children, not the younger ones. These are relevant 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

33 

to the current study beacause the findings in Coady and Aslin (2017) can be measured 

against those in the present study to see how autistics perform in contrast to typically 

developing children, using actual words. A longitudinal study to observe how these 

children acquire phonotactics from simple to complex as they advance in age would 

have made this study a well-rounded one. 

Unubi and Ikani (2018) studied Igala syllable and phonotactics using secondary data. 

The result showed that consonants are not allowed or permitted in word final positions 

in Igala; the pattern of monosyllabic and polysyllabic syllables were V, VC, and CV 

syllables, consonants could occur word-finally in idiophonic words and few others in 

the language but there are strict limitations to labial and velar nasal sounds /m, ŋ/. When 

consonants appear in word initial positions, the words are verbs and if in word medial 

positions, those words are nouns. Despite being an addendum to readership, the study 

seemed to lack depth as phonological theories could have been employed to give more 

authenticity to the analyses and discussions.   

2.1.6 Phonological processes 

It is expedient to review some relevant studies in phonological processes in relation to 

the phonotactics of autistic children. Phonological sounds can be explained in isolation 

but when these segmental sounds are found in the association of other sounds, some 

phonological processes are bound to take place. Among several processes identified by 

scholars, (Yule, 2010, 1996; Roach, 2010, 1991; Oyebade, 2008; Cruttenden, 2001; 

Gimson, 2001; Jones 2003, 1972), only three of these processes would be considered in 

this study: substitution, deletion and epenthesis. This is because most of the other 

processes can be subsumed under these three. 

a. Substitution: This process involves replacing a sound with another. This can be 

done in various ways such as reducing the sound quality, changing the manner 

or place of articulation or out replacing a sound with another that shares nothing 

in common with the replaced sound. Atolagbe (2001) posits that substitution is 

the replacement of a sound with another sound or features of a sound in a specific 

environment. Alabi (2007) submits that it is a major phonological interference 

triggered by language contact phenomenon. Alabi’s suggestion interprets the 

situation in which substitution occurs and that the unavailability of an L2 sound 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

34 

in a speaker’s first language (L1) is responsible for a substitution with an 

available sound in the speaker’s L1. Most of the other processes such as 

assimilation, fronting, devoicing, stopping, and so on can be grouped under 

substitution. Dollaghan and Campbell (1998) describe substitution as an error. 

b. Deletion/elision: This term is interchangeably replaced with elision (Yule, 

2010:47; Roach, 2010:114). According to Yule (2010), elision is a show of 

efficiency in the language using illustrations as ‘you and me’, /ju:ənmi:/ and 

‘friendship’, /frenʃɪp/. Roach (2010) identified four reasons for elision: (i) loss 

of weak sounds, especially the schwa sound /ə/, which occurs after aspirated 

voiceless plosives. This is not the case when these voiceless plosives /p, t, k/ 

come after the voiceless alveolar fricative, /s/, Roach (2010) employed the 

following examples: ‘potato’, /pʰteɪtəʊ/; ‘tomato’, /tʰmɑ:təʊ/; ‘canary’, /kʰneəri/; 

other reasons provided for elision were: (ii) weak vowels with syllabic 

consonants, (iii) avoidance of complex consonant clusters, and (iv) loss of final 

/v/ in ‘of’ before consonants. The author expressed doubt about whether all other 

cases of contractions of grammatical words should be categorised as elision or 

not. 

Osisanwo (2012: 179) describes deletion as the disappearance of one or more 

sounds in rapid speech and identified two cases of deletion -- vowel and 

consonant deletions. The explanations on vowel deletion tallies with Roach’s 

(2010) explanation provided in the previous paragraph. Consonant deletion 

mostly occurs in cases where consonant clusters or simplification of clusters 

exist. This often involves the loss of alveolars /t, d/, especially when in 

combination with other consonants (Jones, 2003). Weak syllable deletion, cluster 

reduction and final consonant deletion are features of deletion which are also 

described as an error by Dollaghan and Campell (1998).  

c. Epenthesis: This is a case of insertion of additional sound in a sequence of 

phonemes. According to Osisanwo (2012: 185), it occurs mostly when a word of 

one language is adopted in another language whose rules of phonotactics do not 

permit a particular sequence of sounds. Alabi (2007) describes epenthesis as a 

superfluous insertion of a vowel segment as a means of alleviating the 

complexity of consonant clusters -- a common feature of English. For example, 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

35 

the insertion of /i/ in the following words, /miliki/ or /milki/, for ‘milk’ and 

/girama/, ‘grammar’ by mostly Yoruba speakers of English explicates this. 

Palatal fronting and insertion are other examples of epenthesis which Dollaghan 

and Campell (1998) do not see as an error but as a process of language acquisition 

in children. 

Sometimes, these processes overlap in an output. For example, ‘water’ becomes 

/wawa/; this output has undergone series of processes which overlap. Deletion, 

substitution and epenthesis can be deduced. First, ‘ter’ has been deleted from the 

output and substituted with /wa/. Also, the insertion of /w/ in place of /t/ can be 

termed epenthetic. Hence, although several phonological processes can be subsumed 

under some major processes as discussed above, it is also possible for these processes 

to work together, overlapping in an output.  

Cleland, Gibbon, Peppe, O’Hare, and Rutherfore (2010) study the phonetic and 

phonological errors in children with high functioning autism and Asperger’s 

syndrome. Using 30 participants with high functioning autism and 39 who have 

Asperger’s syndrome and who were between the ages of 5 and 13 years, the 

researchers found that their speech was characterised mainly by developmental 

phonological processes, such as gliding, cluster reduction and final consonant 

deletion. However, non-developmental error types such as phoneme-specific nasal 

emission and initial consonant deletion were also featured in their speech. They 

conclude that the presence of these non-developmental errors pointed out speech 

disorders which could add additional communication and social barriers to affected 

children if not diagnosed and treated as early as possible. These findings are very 

relevant to studies like the current one and those concerned, however, the study still 

grouped affected children according to DSM-4 (high functioning autism and 

Asperger’s syndrome) and not DSM-5 (in accordance to their severity – mild, 

moderate or severe). Again, all these processes can be subsumed under deletion, 

substitution and epenthesis. This study is yet another from the vast expanse of 

available studies in the developed countries. 

 

 

UNIV
ERSIT

Y O
F IB

ADAN L
IB

RARY



 
 

36 

2.1.7 Appraisal of literature review  

The studies of Schoen, Paul and Chawarska (2011), Ojo (2012), Nwanze (2013), 

Nordgren (2015), Wolk, Edwards and Brennan (2016), Brennan, Wagley, Kovelman, 

Bowyer and Lajiness-O’Nell (2016), Umera-Okeke and Iroegbu (2016), Kjellmer, 

Fernell, Gillberg, and Norrelgen (2018) outline the extent of language and phonological 

studies as well as the state of intervention as it relates to autism mostly outside and within 

Nigeria.  

Also, Luce and Large (2001), Btoosh (2006), Leonard, Davis, and Deevy (2007), 

Cleland, Gibbon, Peppe, O’Hare and Rutherfore (2010), Mbah and David (2013), 

Soneye and Oladunjoye (2015), and Coady and Aslin (2017) mostly studied 

probabilistic phonotactics, or the p