A STUDY OF LEXICAL BORROWING FROM ENGLISH TO EDO

LANGUAGE

BY

300L

ODIA OSAMWONYI ANDERSON

ART1701335

DEPARTMENT OF LINGUISTIC STUDIES

FACULTY OF ART
UNIVERSITY OF BENIN
BENIN CITY.

NOVEMBER, 2022.

1
 CHAPTER ONE

GENERAL INTRODUCTION
1.0. Introduction

When languages borrow from other languages, the items are modified in such a way that they fit
into the structure of the language that has borrowed it, enabling its adoption as a loanword in that
language. This is done by some “nativization” strategies in that language. In E̩ do, the
nativization strategies include the substitution of any occurring non- E̩ do sound with an E̩ do
sound that shares features (such as the place of articulation) with the non- E̩ do sound, insertion
of an appropriate vowel in-between consonants when they occur in clusters or the deletion of one
of such consonants, addition of a vowel as a prefix to a noun when it begins with a consonant, as
well as the addition of a vowel as a suffix to any word that ends in a consonant.

This study examines the nature of borrowed items from English into E̩ do, the degrees of
nativization of the forms, the strategies that have been employed in their nativization, the roles
that the E̩ do syllable structure plays in the borrowing process, and the different structural rules in
E̩ do that the forms that are yet to be modified (and those that are yet to be fully modified) violate
in the language. The study also, by means of evaluation, show why some structural forms are
acceptable while others are not, as well as the effects that these may have on the E̩ do
phonological constraints.

1.1. E̩ do People and Language

There is no one acceptable account of the origin of the E̩ do (or Bini, Benin, as they are also
known) people (Osadolor, 2001). A number of accounts exist (most of which rely on the oral
traditions of the people). However, the popular account by Egharevba (1953) in his book, A
Short History of Benin, is notable. According to this account, “Many, many years ago, the Binis
came all the way from Egypt to found a secure shelter in this part of the world” (p. 1). E̩ do
scholars and historians have generally rejected this claim. Notably among them is the incumbent
O̩ ba of Benin, O̩ mo̩ N’O̩ ba N’E̩ do, Uku Akpo̩ lo̩ kpo̩ lo̩ , Erediauwa, CFR, O̩ ba of Benin, who, in
his landmark autobiography declared: “I personally have never accepted the account of our
illustrious historian, Jacob U. Egharevba”. (Erediauwa, 2004: 205). Among the people, there is
the general (and strongly held) belief that the E̩ do people have been in their present location
2
since the origin of the world. This is unverifiable and makes the task of understanding the origin
of the E̩ do people even more difficult. Evidences abound that the people have used several
names to refer to their geographical location at different times. Examples of such
names include “Ìdú” as found in names like “Ìdúbo̩ `” (Idu is flourishing), “Ìdúmwo̩ `úyì” (Idu is
respected), etc, and “Ìgòdòmígòdò” which reflects the name of the first of the Ògìsó kings (the
earliest dynasty of kings that ruled the kingdom). The term “E̩ do” is used to refer to the people,
their geographical location and language. The name “Bini” or Benin” came into use in the area
when the Portuguese began to establish ties with the ancient E̩ do people. According to Imasuen
(1998), the Portuguese inherited the name “Ubini” from the Itse̩ kiri people and later reduced it
to Bini. This, it seems, explains why the name of the capital city of E̩ do State is Benin City.
As Ogie (2002) noted, O̩ ba E̩ wuare (1440 – 1473) changed the name of the kingdom to E̩ do, in
memory of the slave boy who helped him during his struggle to become O̩ ba. Since then, the
people have referred to themselves as “ìvbí E̩ dó” (ie. E̩ do children or people) (Agheyisi,
1986; O̩ mo̩ zuwa, 2003).

The E̩ do language is spoken in the present day E̩ do State of Nigeria. It has an estimated 2.5
million native speakers (Yuka & Omoregbe, 2011). Imasuen (1998) reports that the language is
also spoken natively in other areas outside Benin Metropolis, including areas such as Okitipupa,
Akotogbo, Idoani and Akure in Ondo State and O̩ za N’Ogogo in Delta State.

The language is listed as Bini in the KWA sub-family of the Niger-Congo family of Greenberg
(1966). The language has also been classified as belonging to the North Central Edoid (PNCE)
sub-group of the Edoid group of Elugbe (1989), shown below.

1.2. Language Contact and Implications

Language contact refers to the use of two (or more) languages at the same place. The situation
arises when groups of people who do not share the same language need to communicate (El-dali,
2012). This has a number of possible effects. For example, one of the groups may learn the
language of the other, or a contact language may emerge (especially when none of the groups is
interested in learning the other’s language).

Borrowing is one of the several possibilities when language contact is persistent, especially when
speakers become bilingual (when a person is able to use two languages effectively in
3
communication) and/or multilingual (when a person is able to use three or more languages
effectively in communication). When words are borrowed, they are modified (or nativized)
appropriately as loanwords, such that they are made to fit into the structural pattern of the
language into which they have been loaned. Thomason (2001:10) states that up to 75% of the
total vocabulary items in English are from other languages, mostly French and Latin.

Since the colonial period, the use of English by E̩ do speakers has increased greatly. English has
become the language of education, trade, etc. The implication of this contact between the two
languages is that E̩ do speakers have to use English more frequently, leading to a situation in
which the two languages are able to influence one another. There have, therefore, been a lot of
English items borrowed into E̩ do as evident in the speeches of the native speakers.

In what Paradise & Lacharite (1997:384) refer to as “repair strategy”, borrowed items that have
forms that violate the structural rules in the borrowing language must trigger some
phonological operations that “inserts or deletes content or structures to ensure conformity to the
violated constraints”.

This study examines how borrowed items from English are modified into fully nativized forms
(i.e. loanwords) in E̩ do with the aim of showing the different phonological strategies that are
employed in the process.

1.3. Statements of the Problem

This study seeks to answer the following questions:

i. What are the strategies that are employed in the nativization of borrowed items in E̩ do?

ii. How are these strategies employed?

iii. Are English items borrowed into E̩ do with either partial or complete modification?

1.4. Purpose of Study

The purpose of the study is to:

i. account for the phonological strategies by which borrowed items are nativized in E̩ do.

ii. examine the extent to which nativization (a perceived obligatory principle) remains
4
effective in E̩ do.

5
1.5. Scope of the Study

This study is about the phonology of borrowed English words in E̩ do. It is centered on
observations from the speeches of the E̩ do native speakers, given the prolonged contact between
E̩ do and English.

This study concerns only words borrowed from English to E̩ do. It therefore excludes other forms
from other languages such as Yoruba, also found in the speeches of the respondents. Although
morphological principles are handy in our analysis, they are not the major concern of the work.

This delimitation is necessary to ensure a more comprehensive analysis and account of the
specific areas that have been carefully chosen.

1.6. Methodology

This study adopts the basic scientific methods of observation of problem, statements of
hypothesis, data collection, data analysis and data observation, testing of hypothesis and report of
findings. The data for this work include those from primary and secondary sources. The primary
data comprises tape recorded speeches of E̩ do native speakers, within the age range of 18 – 70
years, in everyday conversations. The recordings were done in several settings where the
speakers use the language freely. Such settings include market places, festivals, family
discussions, storytelling, E̩ do T.V. talk shows such as Ibota (aired on Independent Television,
Benin City, between 6.30 –7.00 pm every Monday evening), etc. The secondary data were drawn
from already existing works of E̩ do analysts and E̩ do home videos. This choice of sources is to
ensure that the work draws from a wide range of borrowing situations in the language. The
analysis of the collected data was done by observation of the borrowed words that were extracted
from the recorded conversations. The words were organized into different groups such as their
word classes and technical nature (i.e. whether a particular word has been borrowed due to
technological experiences, etc, for example). The structures of the borrowed words were checked
against the adaptation constraints in E̩ do, with the aim of observing how well, or otherwise, these
structures have obeyed the constraints. This helps in determining the extent to which the
nativization strategies of the language have affected them. This analysis is guided by the tenets
of the constraint properties of Optimality Theory, as they relate to E̩ do phonology.

6
The total recording/discussion time in the collection of data for the study is about 274 minutes
(i.e. approximately 4 hours). Within this time, it is found that a total of 902 borrowed items were
found in the speeches of the respondents who use E̩ do as their first language. The words occur in
the following word class distribution:
Nouns: 827 words
Verbs: 53 words
Adjectives: 23 words

Within the group of nouns, about 162 were found to be terms associated with new technologies
and therefore thought to have resulted from new technological experiences of the E̩ do people,
while the others resulted from non-technology driven experiences in other aspects of life; such as
education, work, security, religion, names, food, government, etc. (refer to Appendix for a list of
collected borrowed words).

In our presentation of data for discussions, the examples are presented in their transcribed forms
which represent the spoken forms in which they were used by the informants. The study uses
Nigerian English (Ugorji, 2010, 2013b) as the standard for determining the nativization of the
borrowed items, with the understanding that this is the English that the E̩ do speakers have been
in contact with and from which they have borrowed items.
1.7. Theoretical Framework
This study is based on the Optimality Theory (OT) of Prince and Smolensky (1993). The theory
shares the assumptions of the Principles and Parameters Theory (Chomsky, 1981) and accounts
for the mental operations that are involved in language processing. The theory assumes that there
are several different possible outputs or realizations of language units which compete among
themselves for the favourite spot. This implies that language users use the most preferred
possible output when they use their languages.

OT has three basic components as follows:

i. GEN.
This is the GENERATOR. Inputs are fed into this component. The GEN generates the possible
outputs that could be the end product of the entire process. These possible outputs generated by

7
GEN are known as candidates. The candidates all compete to become the optimal (or most
preferred) option.
ii. CON.
This contains the CONSTRAINTS in languages. These constraints are universal and are ranked
differently in different languages, depending on what a language permits and how strictly it
forbids certain occurences. The constraints have, among others, the following characteristics.

i. Universality: Universal Grammar provides the constraints, and they are universal in nature.
The constraints are present in all languages.

ii. Violability: Constraints can be violated, but the violation should be minimal.

iii. Ranking: The constraints are ranked in some hierarchy that depends on the different
languages. This ranking usually determines the degree of violation, and it also determines
the grammar of the language. (Prince & Smolensky, 1993: 1483; McCarthy & Prince, 1994a:
3, 2001: 6).

The notion of ranking presumes that the constraints are in hierarchical order of strict dominance. Since
the constraints can be violated, the candidates that violate the highly ranked ones perform worse than
the ones that do not violate such high-ranked constraints, even if the candidate violates more of the
lower ranked constraints. The candidate that wins does not have to satisfy all the constraints.

In a given language, a constraint may be ranked so high that it must be satisfied at all times, and may
be ranked so low that its effect is not felt when it is violated. It may also fall into any of the
intermediate ranks.
iii.EVAL.

This is the EVALUATOR component of the theory. It evaluates the operation and checks the
candidates in relation to the constraints. It, thereafter, considers the degree of violations
committed by the candidates and chooses the best candidate (that is, the optimal candidate)
which is seen as ‘the winner’. This winning candidate is the final output. This is said to be the
definition of optimality.

Any candidate that violates the highest ranked constraints is said to have committed a ‘fatal
violation’, and therefore ruled out. The optimal candidate is the one that does not violate higher

8
ranked constraints. As McCarthy & Prince (1994a) point out, “… even dominated constraints
may be visibly active under appropriate circumstances”. (p. 29). This situation is known as
‘emergence of the unmarked’, and shows that the hierarchy of the constraints is a product of their
relevance to the language.

The relevance of OT to loanword adaptation is mainly in the area of the violation, or otherwise,
of constraints in a language. This is captured by Paradis & Lacharite (1997) thus:
Often, loanwords enter the borrowing language (L1) with structures (that is, segments and sequences) that
are, from the point of view of L1, ill-formed. We maintain that such structures are repaired minimally by
universal phonological operations that are triggered by the constraints that we observed on the surface. (p.
380).
Such ill-formed structures are adapted to fit into the acceptable forms in the language by its
repair (i.e. nativization) strategies. They defined repair strategy as “a universal, non-contextual
phonological operation that is triggered by the violation of a phonological constraint”. (p. 384).
The structures serve as the inputs that are fed into the GEN. GEN generates the possible
candidates. The candidates are checked by the EVAL component, by considering how they have
fared in the competition, judging their respective degrees of violation or non-violation of existing
constraints. The best candidate is chosen as the optimal or in the Preference Grammar of Ugorji
(2010), ‘preferred’ candidate.

The result of the Evaluation process is shown on the Constraint Tableau. Prince & Smolensky
(1993) describe the Constraint Tableau as an augmented version of display. The tableau has the
constraints arranged in their ranking order on the horizontal axis (the higher ranked ones come
first) and the candidates on the vertical axis. It helps to show how the different candidates have
fared in the competition. Examine the following constraints in E̩ do in their order of hierarchy, for
example;

*BC[N] Onset and/or coder are not allowed in nouns;

*NES Non- E̩ do sounds are not allowed.

The tableau for evaluating the candidates for the term /ʧɔʧ/ ‘church’ which shows why it is
borrowed into E̩ do as /èsɔ´sì/ is shown below.

9
Tableau 1: Showing the Optimal candidate: [èsɔ´sì] ‘church’

/ ʧɔʧ / *BC[N] *NES

[ʧɔʧ] * *

☞ [èsɔ´sì]

* - violations of constraints.

! - crucial/fatal violation (shows a candidate that violates a constraint so highly that it

cannot be allowed).
☞ - (also or +) points to the optimal candidate.

Shaded cell - cells that do not contribute to the determination of the optimal candidate. The
tableau shows that the optimal candidate, [èsɔ´sì] ‘church’, is chosen because it performs better than
the other, in the sense that it does not violate any of the constraints.

10
 CHAPTER TWO

LITERATURE REVIEW
2.0. Introduction

Over the years, several linguists have become more interested in what happens when two or
more languages are used in a particular place over a long period of time. This phenomenon has
been studied under ‘Language Contact’ (Weinreich, 1953; Wardaugh, 1998; Sankoff, 2001;
Mufwene, 2007; Thomason, 2001, 2007, 2010; Matras, 2009, 2011; El-dali, 2012; Chamereau &
Leglise, 2012; etc.) or ‘Contact Linguistics’ (Winford, 2007). The major aspects of the general
literature that are relevant to this work are those of language influence (or interference),
borrowings and their nativization strategies. These will be the focus of our review. The chapter
also directly addresses the definition of relevant terms; such as loanword, codeswitching and
code mixing, among others.
2.1. Language Influence
The kinds of effects that languages have on one another in contact or second language learning
situations have been described in different terminologies at different times. Terms such as
language transfer, interference influences, contact induced change and borrowings have been
used to describe these effects, with several arguments that seek to favour the use of each term
(Bhela, 1999; Thomason, 2001; El-dali, 2012). A consideration of the different definitions and
explanations, given for the different terms, shows that the common feature is ‘influence’. In
other words, what is referred to as transfer, interference, etc is some form of influence that one
language has on another. For this reason, the term language influence is adopted in this work to
mean all kinds of such effects that contacting languages have on one another: the subject of our
discussion in this section. The relevance of this discussion is in the fact that linguistic borrowing
and loanwords (as expanded in 2.2, below) which necessitates the adoption of items are
examples of influences that languages have on one another.
As already stated above, the term language influence is here-in used in a broad sense. Hence, all
explanations given for all forms of language influences will be considered.

El-dali (2012: 11) defines transfer as “the influence second language has on a previously
acquired language”. This justifies our adoption of the term ‘influence’. This definition implies
11
that the influence is believed to be on a previously acquired language and it is motivated by a
second language. It also brings to fore, the influence of the second language(s) on the first. In
Bhela (1999)’s account of Ellis (1997), it is explained that “Ellis (1997:51) refers to interference
as ‘transfer’, which he says is ‘the influence that the learner’s L1 exerts over the acquisition of an
L2” (p. 23). This kind of flow of influence is usually governed by the first language structural
(and permissible) patterns and world view. For Winford (2003), it is not exclusively an L1 that
can influence other languages in this flow of influence, but any ‘primary language’. This point is
particularly important to our understanding of language influence, even though the last part of
that comment that associated such influence “particularly” to language shift may not always be
the cases. Since languages will always be in contact, the study of this kind of influences becomes
inconclusive.

While some of these influences may be predictable and common, others are not (Siemund,
2008). It has been stated, for example, that:

The most common result of language contact is change in some or all of the language: typically,
though not always, at least one of the languages will exert at least some influences on at least one
of the other languages. And the most common specific type of influence is the borrowing of
words (Thomason, 2001: 10).

That change is a common result of language contact is as true and interesting as the fact that
contacting languages do have influence on one another. Even more interesting in the above
statement is the characterization of lexical borrowing as one of the influences. This seems to
justify her further claim that “it is easy to find transfer in all areas of all language structure:
phonology (sound structure), morphology (word structure), syntax (sentence structure) and
lexical semantics” (p. 11). That languages always change is more strongly echoed by Mahmood,
Hussain and Mahmood (2011: 234) as follows: “Languages are bound to change, whether
internally (generate new words via existing words) or externally (by the influence of other
languages) and they inherit this quality by birth”. The externally motivated change is of interest
to this work, since that deals with the influence from other languages in its environment.
However, the genotypic quality ascribed to those changes (i.e. that they are inherited by “birth”)
seems too much to accept: changes may simply be due to social factors and exigencies such as
contact and usefulness, especially.
12
When languages influence one another, it could be the case that the first language of the speakers
influences the second or the other way round. There could also be the occurrence of code
switching and/or code mixing, as discussed below:

2.1.1. First and Second Language Influences.

Language influence is a two way phenomenon (Julca-Guerrero, 2009). The very common flow
of influence is from the first language to the second language. Kaushanskaya & Mariam (2009:
829) had pointed out that “learning a second language in adulthood can be difficult”. This is true,
and will capture the fact more accurately if our understanding of the term ‘adulthood’ includes
any person who has passed the critical stage of language learning. When the acquisition of the
first language reaches a certain stage, the structure and worldview of the language becomes
stable and established. At that point, the language has the tendency to influence the learning or
use of any other language. As Kaushanskaya & Mariam (2009) also observe, in line with Flege
(1992, 1995), “… adults’ difficulty with acquiring L2 phonology may be due to the fact that L1
phonological categories interfere with formation of L2 phonological categories, especially in
situations where L1 and L2 phonemes differ in subtle ways” (p. 829). This is fallout of the
stability in structures mentioned earlier. The mind of the learner is already conditioned to the
patterns of the L1, hence Kaushanskaya & Mariam (2009:829) continue that “as a result, L1
phonological categories interfere with L2 production, and speech in L2 carries a trace of L1
accent”. This is the case of the first language influencing the second or other languages, what we
have called “first language influence” or, as generally called, native language influence. For El-
dali (2012:11), this “involves the influence of a source language on the acquisition of a target
language; the ‘second language’ regardless of how many languages the learner already knows”.
The term –target language – in this sense refers to the language that the learner is currently
learning.

Despite the fact that first language influence has been of great interest for about a hundred years
now (El-dali, 2012), there seems to be a lot of controversy surrounding the understanding of the
nature of the influences. It is still very difficult to predict what kind of influence will take place
and when. Although there are so many disagreements on the adoption and definition of a
particular term for such contact-induced influences, it is however of common knowledge that L1
does exert significant influences on the learning and use of L2.
13
L2 on L1 influence is possible when a speaker becomes very balanced in the development of the
second language and is able to use it as well as he can use the first (or uses the L2 more
frequently than the L1); at a point when the structure of the second language is able to compete
with those of the first language in his mind. Although the answer to Lord (2008: 184)’s question
as to whether L2 actually influences L1 is yes!, El-dali (2012: 8) is of the opinion that “we are a
long, long way from finding ultimate answers to the many difficult questions we have been
asking”. This situation, which is likely due to the unpredictable nature of the attitudes of
language users towards the different languages around them, has made the understanding of L2
influence on L1 difficult.

Following the very striking claim by Lord (2008:190) that “it is possible for L1 modification to
occur, even when the learner has remained in an L1 community”, E̩ do speakers have been
observed to allow different forms of modifications to the use of their mother tongue, following
the influence of English language, a phenomenon that we have called ‘second language
influence’ (i.e. the influence of the L2 on the L1).

2.1.2 Code-Switching and Code-Mixing

Code-switching on the one hand has been defined as the use of more than one language in an
utterance “above the clause level” to appropriately convey intents (Myers-Scotton, 1992;
Cardenas-Claros & Isharyyanti, 2009; Rua, 2009; etc). It is found in bilingual and multilingual
speakers who are able to use sentences from different languages and switch between them in
communication. The switches are believed to be made possible by the occurrence of certain
words that trigger them (see Muysken, 2000).

On the other hand, code-mixing happens within sentences. It involves the use of lexical items
from different languages in the same sentence (Cardenas-Claros & Isharyyanti, 2009; Muysken,
2000; Clyne, 1991; Ayeomoni, 2006; etc). When language users code-mix, they begin a sentence
in one language and end it in another language. In this sense, the codes of the two languages are
mixed up.

As stated in Rua (2009), “any simple combination of words from different languages” will
qualify as an instance of code-mixing. While this may be true, Paula Rua did not explain how

14
this differs from other occurrences in the speeches of bilingual amd multilingual speakers, such
as borrowing (see 2.3, below), etc. When code-switching and code-mixing occurs, the forms are
usually not widely spread in the speech community, and modifications (in view of nativization)
are generally absent. This is why, it seems, Baker (1980) concluded that the words we now have
as borrowed items (or loan words) started out as forms in code-switching and code-mixing.
Following, therefore, the non-nativized forms in our data (as presented in chapter three) can be
said to belong to this group.

In view of language influence (discussed above), code-switchng and code-mixing do not seem to
fit into the claim of directionality (or should one say it is bi-directional) since language codes are
usually switched or mixed in no particular order of occurrence.

Matras (2009) has argued that language influences are noticeable, at least in the first instance, in
the sound structures of words and in prosodic elements. Although this is true, the choice of
words and their uses, it seems, are factors to consider. The enhancement of the viability of the
language that has been influenced is a major contribution of language influence to the
maintenance of languages, especially the minority languages. Against this, however, some
scholars have argued that such influences can only be negative. For example, Thomason (2001:
65) claims that they “always complicate the system, making it ‘less natural”. The complications
that Thomason (2001) noticed are not likely to remain for a long time, since the children who
will acquire the language in the future will have to acquire such influenced structures as natural
occurrences in the language. For instance, words such as àlìmóí ‘orange’ and
èkùye̩ ´ ‘spoon’ which Imasuen (1998) accounts for as borrowed items from Portuguese are
hardly recognized by E̩ do speakers as borrowed. They are acquired by E̩ do children as they
would acquire other items in the language.

2.2. Linguistic Borrowing and Loanwords

The observed items that have been moved by speakers from one language to another have been
given different names like “borrowed items”, “copying”, “loans”, “replications”, etc
(Weinreich, 1953; Matras, 2009, 2011; Johnson, 2002; Essizewa, 2007; etc). We shall make
deliberate efforts to avoid the several arguments following these terms and assume that they all
refer to the same phenomenon and, here in, use any of the terms where necessary, keeping
closely, however, to the terms borrowing and borrowed words/items to refer to the process and
15
items that are involved in that process, respectively (in line with Thomason, 2001), even though
these choices present their own implications (for example, the terms may imply the intention to
return such items – an intention that is never likely the case). Our understanding of linguistic
borrowing and loanwords is important to the purpose of this study, ie. to examine how borrowed
items from English are nativized into E̩ do.

The attempt to show an adequate understanding of the process and results of borrowing has led
many linguists (and some non-linguists) to give several definitions of the phenomenon. In what
may be considered as an elaborate definition, Imasuen (1998:43) states:

By the term loan, I mean that item borrowed or adopted from one language, similar or not, into
another, which has a reasonable degree of performance in the recipient language, and is familiar
to a wide spectrum of its speakers.

This definition tries to clarify the phenomenon by showing the readers how to identify it. The
definition posits that any item in a language that qualifies as a loan must have been borrowed:
Borrowing in this sense is a process that such a word must have gone through. A loan must
therefore not be an original item in the language within which it is defined, but must have been
“lifted” from another language. Adaptation, which is generally considered to be an
“automatic” process in borrowing, according to Imasuen (1998), may be considered not to be so.
Imasuen’s definition cited above also claims that the languages (i.e. both donor and receiving
languages) involved in the process do not need to be similar. The borrowed item must have a
“reasonable degree of performance” in the sense that they must not only be useful, but the
recipient language speakers must use them with reasonable frequency. In line with this, if such
items are not frequently used by language users, they will not be regarded as loans in the
language. Lastly, the definition claims that a large proportion of the language users must be
familiar with the item. These claims, though not exhaustive, are necessary to the identification of
loan words in languages.

After a consideration of the different terms as used by different scholars in the field, and “for fear
of contributing further to unnecessary proliferation of terminology”, Matras (2009:146) defined
the term ‘borrowing’ as “replication”. In his simple classification, the term refers to “the
replication of a linguistic structure, of any kind, in a new, extended set of contexts, understood to

16
be negotiated in a different ‘language”. This rather technical definition emphasizes the context in
which the borrowing process is triggered. Borrowing is necessary when there is the need to
express ideas that are not already existing in the recipient language, or for which they do not
already have a term. Such ideas create a contextual difference between the two languages, a gap
that is bridged by the introduction of a new term. By this definition, the words are not simply
borrowed, especially since they remain in the donor language after the process, hence they are
copied or ‘replicated’ in the recipient language. In the words of Matras (2009);

I will use the term replication to capture even more closely the fact that we are dealing not with
issues of ownership or even distinct imitation or duplication, but rather with the activity of
employing an item, in context, in order to achieve a communication goal (p.146)

In another definition, Matras (2011:204) refused (as in the previous one) a simple definition like
that of Imasuen (1998) and views “it as the removal of an invisible demarcation line that
separates subsets within the linguistic repertoire of the speakers ‘language”. What he calls
“invisible demarcation line” here may be interpreted as the communication gaps, mentioned
earlier. In this definition, however, the speaker is a bilingual (at least) with a ‘linguistic
repertoire’ that has subsets (i.e. more than one languages) that are separated by the ‘invisible
demarcation lines’ which the borrowed words are used to remove. Later, Matras (2011) gave
what seems to be an even more technical definition, even though it is assumed to be a part of an
explanatory process that sought to simplify the earlier definition. He stated thus:

Recall once again that borrowing is defined here (following Matras, 2009) as the generalization
of an item within the multilingual linguistic repertoire, and hence the partial or complete removal
of interaction set-based constraints on its occurrence (p.209).

With this latest definition, it becomes clearer that Yaron Matras’ approach to borrowing is based
on the need for a multilingual language user to remove whatever constitutes a barrier to his or
her interactional need in any of the languages. The language specific rules or constraint
hierarchies are affected in one way or the other in the course of this borrowing operation.
While Matras (2009) had opted for the view that the borrowed words are ‘replicated’ in the
receiving language, some have argued that they are ‘adopted’. Julca-Guerrero (2009:80), for
example, explained that the borrowing process “… consists of the adoption by one language of

17
linguistic elements from another language …” This view assumes that the items are moved into
the recipient language, paying attention to the adoption rules of that language. It also assumes
correctly that the items are never meant to be returned to the original language from which it was
adopted.

While explaining that borrowing is “a process that occurs any time two cultures are in contact
over a period of time” Hoffer (2002:1) defines the term simply as “the process of importing
Linguistics items from one Linguistic system into another”.

Here, borrowing is seen as involving ‘importing’ language items. It may be understood,

furthermore, from the definition that borrowing is defined from the point of view of the recipient
language, as the term ‘importing’ suggests. From Hoffer’s definition, one assumes that the
process, at any point in time is a one way process (from one language to another).

Thomason (2001) noted that borrowing may be found in the total absence of bilingualism.
Hence, language users can copy a word or name that they consider captivating and use it to refer
to the same or a different concept in their language, even though they may not be able to
construct any complete meaningful sentence in that language from which they have borrowed
such words.

In this study, the term borrowing is seen as the process by which foreign linguistic items are
made to become a part of a language, while a loanword is that linguistic item that has
completely undergone the process of borrowing. The understanding, here, is that borrowing
involves a number of operations that are aimed at nativizing an item so that it can function
appropriately as a loanword in that language.

Lexical Borrowing versus Structural Borrowing

In Thomason (2001), a distinction is made between lexical borrowing on the one hand and
structural borrowing on the other hand. She argues that, “we expect to find lexicon borrowed
without structure” (p. 69). Borrowing a lexical item without its structure implies modifying its
structure in other to make it acceptable to the recipient language’s structural combination, for
example, E̩ do will borrow English ‘school’ as ‘èsùkú’. As the contact between the two languages
stretches over a longer period, and the interaction between them intensifies, the language users of
18
the recipient language become so used to the structures of the two languages that they can use
them effectively, even in code-mixed and rapid speech forms. At this point, unmodified
borrowed items may be observed, in which the speakers are able to use the so called
unacceptable word structures of the donor language, hence, ‘èsùkú’ – school can be used after
some time as ‘èskú’. This, Thomason (2001: 69) explained, may happen unconsciously.

2.2.1. Motivations for Borrowing

When two or more languages interact in a place for a reasonable length of time, the basic factor
that is needed for language influence is set. When borrowing begins to take place, it is usually
the situation (i.e. trade, religion, war, etc) that has necessitated the contact that determines the
kind of vocabulary that is borrowed. Besides just contact, a number of other factors have been
given by different scholars as motivating factors of borrowing. These are discussed below.

2.2.1.1. Language Contact

Language contact, as stated above, is a necessary motivation for linguistic borrowing. It sets the
stage for many other sociolinguistic phenomena such as language influence, bilingualism and
multilingualism, pidgin and Creole, etc. A language is not likely to borrow words from another
language with which it has never come into contact. As Mahmood, Hussain & Mahmood (2011:
235) observed, “New words come into a language via different sources and the most important
source is borrowing words from neighboring languages”. As they have claimed, borrowing from
neighboring languages, that is, a language with which it interacts often is the “most important”.
They also emphasized the productive nature of such neighbour-to-neighbour situation to lexical
borrowing. In an earlier study, Hoffer (2002) had observed that “Many factors influence the
amount and rate of borrowing. The closer and longer the contact and neighbourliness, therefore,
the more the rate of borrowing among languages is likely to be. When this is sustained for
another longer period, as the above statement shows, there is the possibility that a word that was
borrowed from a language may be “re-borrowed” (i.e. back to its original source) in its latest
modified form.

2.2.1.2. Bilingualism and Multilingualism

Bilingualism and multilingualism are also preconditioning factors of borrowing. When a person
19
is able to speak two or more languages fluently, it is quite easy for him or her to borrow terms
from the other language(s) when using any one of them. This is useful, since there is always the
need to fill-in gaps in his use of which ever one of the languages. Such gaps may be due to the
non-existence of certain concepts in one language or due to memory failure. The ability of a
bilingual speaker to be fluent in the two languages, or be a native speaker of both, may not be a
pre-requisite for borrowing in this case. Thomason (2001) noted that “the borrowers do not have
to be native speakers. If your native language is English and you also speak Japanese fluently,
you could introduce English elements into your version of Japanese” (p. 68). She added that such
introductions are “not because you fail to master some bit of Japanese grammar, but simply
because you wanted to use a construction while speaking Japanese”, implying that there may be
no gap or memory failure, but the need to use a particular expression in a particular way.
Fluency in the use of all the languages does not apply strictly to the use of the term – bilingual
and multilingual speakers – in this case, even though the contrary may have been suggested.
Thomason (2001: 69) argues further that “you need not be at all fluent in a language in order to
borrow a few of its words; but since you cannot borrow what you do not know, control of the
source language is certainly needed before structural features can be borrowed”. Hence, a person
who is fluent in one language, but barely knows others can borrow from any of the languages in
the use of any other.
2.2.2. The Borrowing Process
The study of loanwords in particular, and borrowing in general, has led to the observations and
analysis that suggest that the process is systematic and thus predictable. Thomason (2001) had
given what she called the “most basic prediction”, stating that “vocabulary is borrowed before
structure” (p. 69), and that the features borrowed increase as the “intensity” of contact increases.
While explaining that the attitudes of the speakers usually play a lot of roles in making such
predictions difficult, she is of the opinion that the prediction, however, accounts for why “we
expect to find lexicon borrowed without structure, but not vice versa” (p. 69). Besides this social
factor of language contact, it has been posited that a linguistic factor such as how basic a set of
vocabulary is in a language is also to be considered. For example, “non basic vocabulary items
are easiest to borrow since in most languages a new noun (for instance), or even verb can be
inserted readily into existing construction. Phonological features (such as stress placement) and
syntactic features (such as word order) are the next easiest to borrow. At the opposite end of the

20
scale, “inflectional morphology is the hardest to borrow” (Thomason, 2001: 69). This

21
claim is not new. It is in line with the claim made some thirteen years earlier by Thomason &
Kaufman (1988: 74) who assert that “… we know of no exception – and would be astonished to
find any – to the rule that non basic vocabulary is always borrowed first”. These claims imply
that borrowing is systematic, a specific order which is generally referred to as the “borrowing
scale” or “borrowing hierarchy”. The borrowing scale or hierarchy is generally used to refer to
the order in which items and features in languages are borrowed as the period of contact
lengthens and the contact intensifies. Following the suggestions from existing literature (e.g.
Moravcsik, 1978; Field, 2002; etc.), Matras (2011: 208) gives the borrowing hierarchy
“pertaining to the structural autonomy and semantic transparency of morphemes” as shown
below.
Structural factors facilitating borrowing. (a) nouns > non-nouns, function words.
(b) free morphemes > bound morphemes.

(c) derivational morphology > inflectional morphology

(d) agglutinating affix > fusional affix.

This assumption suggests that nouns are borrowed first while fusional affixes are the least, likely,
to be borrowed. In her own consideration of the borrowing scale, Thomason (2001) explained
that when languages are in casual contact, only non-basic vocabulary items of nouns, verbs,
adverbs and adjectives are borrowed. With slightly more intense contact where borrowers are
mostly fluent bilinguals, function words such as conjunctions and adverbial particles like ‘then’,
as well as content words, are observed to be borrowed. When the contact becomes more intense,
there are more bilinguals and speakers’ attitudes and other social factors are there to encourage
borrowing, basic and non basic vocabulary items are borrowed. At this point, she also
observed that there is “moderate structural borrowing”. Borrowing at this point may include
closed-class items such as pronouns, numerals, and derivational affixes. When the contact
becomes so intense (to the point where there are mostly bilinguals among the native speakers of
the receptor language), there is “heavy lexical borrowing in all sections of the lexicon, heavy
structural borrowing” (p. 70 – 71). Thomason (2001) posits that structural borrowing begins at
the second stage of contact described above. She re-emphasized the unreliability of the
borrowing scale and gave a “warning” that she thought necessary, especially when her earlier
22
position that “the ‘word first’ prediction does not actually hold in all borrowing situations,
because in some cultures the borrowing of words from other languages is considered
inappropriate, though structural features may be adopted (perhaps unconsciously)” (p. 69).
Julca-Guerrero (2009) and Matras (2001) seem not to be bordered by such factors as the
speakers’ attitudes, which Thomason warns of, in their discussions and analyses of the
assumption that certain items and features are more easily borrowed than others. Matras (2011:
209) observed that the borrowing hierarchy comes with certain implications. Since some items
are higher up the hierarchy in the borrowing order, it follows that the borrowing of features that
are at the lower ranks necessarily presuppose the higher features have been borrowed. The
borrowing of inflections, for example, should imply that nouns have been borrowed.
2.2.3. Effects of Borrowing

From the above discussion, it is clear that words are borrowed for some reasons. They generally
would affect the language into which they have been introduced in one way or the other. Some of
these effects are discussed below.

2.2.3.1. Vocabulary Enrichment

The introduction of borrowed words should add to the vocabulary of a language. This leads to an
increase in the lexicon of that language. Instances abound to show this. Imasuen (1998:49) states,
for example, that “from what we have seen, we find that Portuguese has succeeded in increasing
the vocabulary of E̩ do language by at least forty-three loanwords”. This, obviously, does not give
an exhaustive account of all the loanwords borrowed into E̩ do from Portuguese, judging by the
length of contact and the fact that Imasuen (1998) admitted that he found it difficult determining
which words in the language are loanwords. In their study of Pakistani languages, Mahmood,
Hussain & Mahmood (2011) report that “words to be introduced from other cultures take a large
proportion of other words associated with them. In other words, we can say when one word is
borrowed it takes a lot of other words, eg. the word ‘mobile’ introduces a lot of other words
which have been borrowed now, such as ‘casing’, ‘keypad’, ‘sim’, ‘memory card’, ‘charger’,
‘handsfree’, etc” (p. 235). This shows, therefore, that the enrichment continues to increase. This
situation makes it even more difficult to account for the number of loanwords in any particular
language.

23
2.2.3.2. Change in Language Structure
Borrowing (especially structural borrowing) affects the permissible structural patterns of units in
a language. This is made possible by the willingness of the speakers of borrowing languages to
accept the borrowed structures when they have used it for a long time. At such points, children
are able to learn the borrowed structures along with the original structures of the language
(Matras, 2009). If this continues, such words become part of the inventory of the language and
their structures become part of the structural patterns of the language. When this happens,
Hoffer (2002) thinks that they can be said to have changed the structure of the language.
2.2.3.3. Prestige
Borrowing can also help to improve the social status of the speakers of a receptor language and
make the language more prestigious. This ‘prestige hypothesis’ will hold as long as the language
from which items have been borrowed is of prestigious status. Hoffer (2002: 18) thinks this is so,
and stated that “loanwords may serve the function of labeling the speaker as fashionable, as up-
to-date”. The ‘labeling’ effect arises from the attitudes of the speakers towards their language
and other language(s). This usually leads to the observed cases of a loanword replacing the term
or expressions, “as in the case of French-derived uncle which replaced anglo-saxon eam”
(Matras, 2009: 150). It is claimed in Matras (2009) that the prestige impart of borrowing is
related to the “conversational effects” that the use of the borrowed words has on the speaker and
hearer in certain context of use.
Due to the nature of the human mind towards social classification, the prestige effect of
borrowing may be very relevant to the drive to encourage language maintenance. It seems the
more a language borrows from the ‘prestigious’ languages, the more it becomes prestigious and
the speakers are encouraged to use it.
2.2.3.4. Broadening of Speakers’ Expressional Range

Language users constantly have to make choices of the appropriate expressions to use in
particular circumstances. With the existence of loanwords, the range of items from which the
speakers make their choices is increased. For instance, Hoffer (2002: 19)

noticed that loanwords are able to serve “as a euphemism for a native word or phrase that carries

24
negative connotations”.
Our discussion of the borrowing process suggests that it is a generally accepted phenomenon in
world languages but Hoffer (2002: 20) observes that it is “not quickly accepted into official
documents, the style of which is usually formal”. This assertion is true to the fact that such
borrowed words would have completely diffused into the donor language before it can be
allowed into official use.
There has also been the case of outright resistance to change. A situation that Thomason (1999)
described as manifested primarily in the refusal to borrow words, and latter Thomason (2007:
49) added that “resistance to lexical borrowing has been more widely noted than resistance to
structural interference”.
2.3. Loanword Adaptation
When words are borrowed, their structures are checked in line with the acceptable structural
patterns of the recipient language, and any unacceptable structure is adjusted to make it
acceptable in that language.
Paradise & Lacharite (1997) observed that the nature of loanword adaptation is not always the
focus of most works on linguistic borrowing, but they end up discussing it in one way or the
other. This has made the phenomenon a widely debated one. In the heat of the debate, a number
of defining statements have been made on the term – loanword adaptation.
According to Paperkamp & Dupoux (2001) in their discussion of the “long held” interest of
phonologists on the subject, loanword adaptation involves the “transformation of foreign words
into forms that better conform to the phonotactics of the borrowing language” (p. 1). The
term, transformation, implies changes or modifications in structure (i.e. form). Such
transformation may involve any of deletion, movement of some elements from one position to
another, etc. this is to ensure that the irregular forms are regularized into conforming structures
in the ‘borrowing language’. Julca-Guerrero (2009)’s study reveals that “borrowing involves
both cases of direct borrowing of words, and the adaptation of a word into the phonetic-
phonological and grammatical system of the other languages” (p. 80). This implies that
loanwords can be accepted into a receiving language without any form of adaptation, if there is
no need for any. When the need arises, the modifications are carried out (unconsciously) as
motivated by the constraints (or rules) of the language that such word violates.
The adaptation of irregular structures is likened to the repair of some bad forms (Paradise &

25
Lacharite, 1997). The nativization strategies usually differ from one language to another,
depending on their general structural patterns. Paradise & Lacharite (1997) define this repair
strategy as “a universal, non-contextual phonological operation that is triggered by the violation
of a phonological constraint, and which inserts or deletes content or structure to ensure
conformity to the violated constraint” (p. 384). This definition emanated from their observation
that borrowed words often enter the borrowing language with irregular structures that need to be
repaired, especially when they violate any of the highly ranked constraints (McCarthy & Prince,
1994a) in that language.
2.3.1. The Nativization Process
There is, so far, no agreement on the number of modifications that borrowed words undergo in
the nativization process. Some scholars have questioned this, and even expressed doubt about the
effects that such words have on the borrowing language (Paradise & Lacharite, 1997; Imasuen,
1998; etc). The conflict, it seems, is as a result of the peculiarities that exist in languages, since a
particular phenomenon is manifested differently across languages. In the Optimality Theory
(O.T.) of Prince and Smolensky (1993), constraints are universally present in all languages but
are arranged in language particular hierarchies. The violations of the higher ranked constraints
are, therefore, the factors that automatically trigger off the nativization process.
The nativization strategies that are commonly used in adapting English words are substitution,
epenthesis, addition (ie. prosthesis) and deletion (ie. paragoge) (Mahmood, Hussain &
Mahmood, 2011: 245). These different strategies are explained below.

26
 CHAPTER THREE
DATA PRESENTATION AND ANALYSIS
3.0. Introduction
In this chapter, the data are presented (showing their nature, the profile of the respondents from
which they were gathered, as well as their classes), the E̩ do phonological conditions are
discussed, after which the data are analyzed and evaluated with reference to the E̩ do
phonological constraints that they violate, with a view to showing the extent to which re-
syllabification (Ugorji, 2013a) plays its roles in the nativization process.
3.1. The Data
This section presents the data collected for this study. The nature of the data, the profile of the
respondents and the different classes of the data are discussed below.
3.1.1 Nature of Data
The use of borrowed items in the speech of E̩ do speakers cuts across the different ages of the
respondents that are under study. The total number of words used was 902, distributed across the
various sources as follows:
2.Interview: 552 items (61.2%)

T. V. talk shows: 57 items (6.3%)

Published E̩ do works: 126 items (14%)

Other sources (including radio/TV and billboard adverts, as well as

non-tape recorded talks from people): 167 items (18.5%)
In observing the data collected, it is seen that English items are borrowed with different degrees
of nativization. These items are, however, fewer in the speeches of the respondents from 60 and
above. While some words are borrowed without modification, some are borrowed with very
minimal modification, while others are borrowed with complete modification. Generally, there is
the imposition of tones on all items in line with the tonal system of the language (see 3.2, below),
hence the indication of tones on the data presented.
3.1.2 Classes of Data
The borrowed items collected are observed to occur in different classes. Their classes of occurrences are
presented below, defined by their degrees of nativization, with examples (refer to appendix for the full list
of data).

27
a) Items that are not nativized:
Examples of borrowed items that are not yet nativised are:

1) a) [brɔ`dá] ‘brother’
b) [àbúʤà] ‘Abuja’
c) [snâp] ‘snap’
d) [kàʃiá] ‘cashier’
e) [frɛ´ʃ] ‘fresh’
f) [klásrɛ̂ p] ‘class rep’
g) [dɔ´ktɔ̂ ] ‘doctor’
h) [áfríkâ] ‘Africa’
i) [krùséd] ‘crusade’
j) [pàtí] ‘party’

These items show the occurrence of forms that that are not in E̩ do: non-E̩ do sounds (ie. [ʤ] and
[ʃ]), consonant clusters (ie. [br], [sn], [fr], [kl], [sr], [kt], and [kr]), word final consonants (as in
1c, e, f and i) and word final consonants in nouns (as in 1a, c – g, i – j). They require
modification in these areas as part of the nativization process. There is the imposition of tones on
all items, as explained in 3.1 and 3.2, above. Even though tones are imposed, there are seen as
not yet modified because the imposition only affects the prosodic elements. Appendix (I) has
other examples of items that are not nativized.
b) Items that are partially nativized:

This class of items is observed to occur in sub-classes (refer to appendix (II) for all items in
this class):

2) i) Items involving consonant clusters:

b) [ètrélâ] ‘trailer’
c) [èfɛ´drâ] ‘federal’
d) [èskú] ‘school’
e) [èstóvù] ‘stove’
f) [ífláwâ] ‘flower’

28
The already effected modifications include the Prefixation of vowels before the word initial
consonants (as in 2a-e) and the deletion of the word final consonant (as in 2c). In 2a – b, the
cluster is such that an alveolar stop – [t] or [d] – is followed by an alveolar trill. In 5c – d, the
1)
voiceless alveolar fricative [s] is followed by the voiceless alveolar stop [t] and the
voiceless velar stop [k], respectively. In 2e, the cluster is formed by the voiceless labio-dental
fricative [f] followed by the alveolar lateral [l].
ii) Items (nouns) involving word initial syllable onsets:

3) a) [báɛ´bólù] ‘bible’
b) [búɛ´hô] ‘bore hole’
c) [sùpàgétì] ‘spagetti’
d) [rɔ`bàbágì] ‘rubber bag’

e) [màsı̂] ‘machine’

3a – b begin with the voiced bilabial stop [b], 3b – e begin with the voiceless alveolar fricative
[s], alveolar trill [r] and the bilabial nasal [m], respectively. The partial nativization that has
already taken place in 3 include the addition of surfixing vowels (as in a, c – d), insertion of
vowels between consonants that occur in clusters (as in 3c) as well as the deletion of the word
final alveolar nasal
[n] (as in 3e), even though the nasality is felt on the preceding vowel, giving risr to the
occurrence of a nasalized vowel [ĩ].
iii) items involving syllable codas:

4) a) [ɛ`zám] ‘exam’

b) [èfóm] ‘foam’

c) [èba´k̃ ] ‘bank’

d) [èrɔ´g] ‘rug’

e) [èpòlís] ‘police’

The final syllables in these forms have syllable codas: 4a – b present forms that end in the

29
bilabial nasal [m], 4c – d end with the velar stops [k] and [g], respectively, while 4e ends in the
voiceless alveolar stop [s]. it is worthy of note, here, that while the alveolar nasal [n] is deleted in
syllable coda positions (as in 7c) and 6e, above), the bilabial nasal [m] is not deleted (as in 4a –
b).

This is summarized in the table below, showing the percentage of yet to be modified forms when
borrowed.
Table 1: Frequency of non-permissible forms across different age groups.
Consonant Border English sounds
Cluster consonants [ʤ] [ʧ] [ʃ]
Ages 50- above 31.6% 54.9% 2.4% - 43.1%
Ages below 50 77.1% 16.7% 80% 64.3% 65.3%
Others 2.9% - - - -
(undetermined,
i.e. data from
published E̩ do
works)

As shown on the table, while respondents who are below 50 years are able to use borrowed
words that are not yet fully nativised in different degrees, only those with syllable onsets, coda
and forms that contain the voiceless palato-alveolar fricative [ʃ] are attested in borrowed items
used by respondents who are 50 years and above.
c) Items that are fully nativized:

5) a) [ègɔ´vúnɔ`]̃ ‘governor’

b) [èbáɛ´kì] ‘bike’[èrédíó] ‘radio’

c) [èrɛ´tɔ̂] ‘rector’

d) [èpɔ´̃dâ] ‘powder’

e) [àdátɔ`] ‘adaptor’

f) [ènɔ´̃sì] ‘nurse’

g) [èfídì] ‘field’
30
h) [èpɔ´tù] ‘pot’

31
 In 5, vowel prefixes are introduced: [e-] and [i-] as appropriate (see details in 3.3,
below), to ensure that none of the nouns has a word initial consonant. Vowel
suffixes are also observed: [-i] and [-u] as appropriate, to ensure that no word
ends with a consonant. The modification of consonant clusters is also
observed: by insertion, as in 5a, and deletion of one of the consonants that are
in clusters as in 10e – g and i. This is so, since the occurrence of consonant
clusters is not permitted in E̩ do. The items and all others in this class (see
appendix (III) for others) are fully nativized and acceptable as loanwords in
the language.
Haven presented the different classes of data, a discussion of the phonological conditions of E̩ do
will follow, giving the basis on which the expected modification of the borrowed items are
determined.
3.2. E̩ do Phonological Conditions
E̩ do basically operates an open syllable structure in which all syllables end in pronounceable
vowels. This provides the syllable condition in the language as shown below.

O R

C V

Fig. 4: E̩ do syllable condition (Adapted from Ugorji, 2013a: 184)

This condition is essential to the occurrence of any segment in the language. Since all sounds
occur within a syllable, the syllable therefore places restriction on their occurrence and triggers
the modification of borrowed item. This usually leads to an adjustment of such items in such a
way that they are made to fit into the syllable structure of the language. This process is described
as re-syllabification (Kenstowicz, 2006; Ugorji, 2013a), and determines the phonological
occurrences in the language (as shown in the following sub-sections). It also determine their
optimality, or otherwise, in the evaluation process.

32
3.2.1 E̩ do Sound Segments

There are a total of thirty-nine sounds that have been accounted for in E̩ do language (O̩ mo̩ zuwa,
2010, 2012). This number is arrived at after a consideration of “articulatory, acoustic and
perceptual” evidences (O̩ mo̩ zuwa, 2012: 71). The broad divisions of these sounds comprise
twelve vowels and twenty-seven consonants. The vowel system is made up of seven oral vowels:
[i], [e], [ɛ], [a], [u], [o], [ɔ], and five nasal
vowels: [ĩ], [ɛ̃], [ã], [ũ], [ɔ]̃ , while the consonant system is made up of twenty-two oral

consonants: [p], [b], [t], [d], [k], [g], [k͡p], [g͡b], [β], [f], [v], [s], [z], [ɹ], [r], [r̩ ], [l], [j], [x], [ɣ],

[w], [h], and five nasal consonants: [m], [n], [݃], [5], [ŋ͡ w].
Since the above thirty-nine sounds are the only significant sounds in the language, any other
sound that is contained in any borrowed items that comes into the language will trigger the
nativization process that will either replace it by the closest E̩ do sound or simply delete it, so as
to make that word fit into the E̩ do phonological system. Given this condition, a constraint is
established in the language that will reject all non- E̩ do sound segments that occur. This
constraint is, herein, stated as follows:

*NES Non- E̩ do sounds are not allowed

*NES will require the nativization of any item with foreign sound in such a way that they fit into
the E̩ do phonological structure, without which such items will not be selected as acceptable.

3.2.2 All Nouns Begin and End in Vowels.

Nouns in E̩ do obligatorily “begin and end in a vowel” (O̩ mo̩ zuwa, 1989: 321; 1997: 114; 2003:
317; etc). This means that nouns in the language can only permit vowels in its borders (word
initially and finally). This makes it necessary for borrowed word nativization process to
introduce appropriate vowels to any of the borders with consonant(s). This obligatory
requirement can be established as a constraint, stated as follows:
*BC[N] Nominal border consonants are not allowed.
Given this constraint, the nativization of borrowed items must include the modification of any
form that violates this condition. It becomes one of the conditions upon which the selection of an
item as acceptable is based.

33
3.2.3 No Consonant Clusters
Another structural rule in E̩ do is that the language does not permit the occurrence of consonant
clusters (CC). Since all words in the language are in line with this rule, loanwords in the
language must take this form. This can be stated in the form of a constraint as follows:

*CC Consonant clusters are not allowed.

Given this constraint, any borrowed item that is coming into the language with consonant
clusters will automatically trigger the nativization process that will either delete one of the
clustering consonants or insert an appropriate vowel in between them. This breaks up the clusters
in a re-syllabification process that ensures that the item is made to be in line with the
phonological condition of the language.
3.2.4 Words do not Begin with Nasal Vowels
In E̩ do phonology, words do not begin with any of the nasal vowels (Amayo, 1976; O̩ mo̩ zuwa,
1990, 2003, 2004, etc). For this reason, only oral vowels can occur in the initial position of
words in this language. This condition can be established as a constraint in the language as
shown below:
*#Ṽ Word initial nasal vowels are not allowed.
This constraint will triggers the modification of any borrowed word with a word initial nasal
vowel and initiates its nativization process that would either replace or delete the unacceptable
segments.
3.2.5 Tones
E̩ do is a tonal language in which tones are distinctive in determining the meaning of items. The
tones that are found in the language, as presented in Omozuwa (2010) are:
Level tones:
High tone [ ´]
Low tone [ ` ] Contour tones:
Falling tone [ ̂ ] Rising tone [̌]

These tones occur in different formation in the language and perform both lexical functions
(when they differentiate between words that have the same spellings) and grammatical functions
(when they are used to specify tense, aspect, conditions, etc).
This tonal condition of the language ensures that borrowed items are modified in such a way that
they are superimposed with tones.
34
The phonological conditions (or constraints) discussed above are summarized below for ease of
reference:
a)*NES Non- E̩ do sounds are not allowed
b)*BC[N] Nominal border consonants are not allowed.
c)*CC Consonant clusters are not allowed.
d)*#Ṽ Word initial nasal vowels are not allowed.
e)Tone All items are superimposed with tones.

These conditions bother on what the syllable structure of E̩ do permits, or otherwise, and must
never be violated by items that will function as loanwords in the language. It is on the basis of
the conditions that the borrowed items will be evaluated and the acceptances of the fully
nativized forms are justified.

3.3 Nativization of Borrowed Items and their Strategies

The nativization strategies employed in the items that are already nativized reveal some level of
consistencies in the choices of strategies and segments. These are presented in the following
discussion.
a) Prefixation
It is observed that items that contain word initial consonants are nativized with the addition of
prefixing vowels. This is to ensure that the syllable structure of the language is maintained,
since the language does not allow nouns to begin in consonants. The vowels observed to be
used in modifying items that require Prefixation are [e–] and [i–] with only one exception in
which [u–] is used in [úwájâ] ‘wire’. The patterns of Prefixation are discussed below.
[e–] Prefixation

The front, mid-high unrounded oral vowel [e] is used as a modifying prefix when the first
syllable of a borrowed item has a front vowel.

a) [tiʧa] [è-tísâ] ‘teacher’

b) [lɛkʧɔra] [è-lɛ´sɔ´râ] ‘lecturer’
c) [sikiuriti] [è-sìkiúrítî] ‘security’
d) [pikʧɔ] [è-písɔ̂ ] ‘picture’
35
e) [redio] [è-rédíó] ‘radio’
f) [rɛktɔ] [è-rɛ´tɔ̂ ] ‘rector’

The choice of [e], the data suggests, is motivated by the following vowel, ie, whether it is a front
or a back vowel.
Items in which the first occurring vowels are not front vowels are also observed to be modified
by [e–] prefix when their original English forms end in a front vowel as shown below:
a) [kɔstɔma] [è-kɔ´sɔ´mâ] ‘customer’
b) [lɔri] [è-lɔ´rî] ‘lorry’
c) [kɔmpiuta] [è-kɔ`̃piútâ] ‘computer’
d) [fɔti] [è-fɔ´tî] ‘forty’
e) [pɔlisi] [è-pɔ´lísî] ‘policy’

The choice here could also be said to have been motivated by the feature of the last vowel of the
item with regard to fronting or otherwise. This suggests that the choice of a prefix may be driven
by the need for uniformity in the peripheral vowels.

[i–] Prefixation

It is observed that words in which the first vowels are back, mid-high vowels are modified by the
addition of the front high vowel [i] as prefix when the original English forms do not end in
realized consonants. Consider the following:
a) [bɔl] [í-bɔ̂ ] ‘ball’
b) [mɔto] [í-mɔ´tô] ‘motor’
c) [kɔbɔ] [í-kɔ´bɔ̂ ] ‘kobo’
In this set of items, it is also observed that [i–] is used before non-coronal consonants.
Note that the introduction of a prefixing vowel increases the number of syllable(s) in a word.
This is so because the introduced segment constitutes a syllable peak.
b) Suffixation

Since E̩ do does not permit the occurrence of a syllable coda, by virtue of its open syllable
structure, borrowed items with codas may be re-syllabified by suffixation to ensure that it does
36
not violate the constraint – *BC[N] – which does not allow such occurences. Suffixing a vowel
is observed to be generally conditioned by the preceding vowels as shown in the following.

37
[-i] Suffixation

Where the preceding vowel is a front vowel, the [i] suffix is used. For example:

b) [intanɛnt] [ıt´̃ ánɛ̂ t̃ -ì] ‘internet’

b) [aksidɛnt] [ásídɛ´t̃ -ì] ‘accident’

c) [ɛdit] [ɛ`dít-ì] ‘edith’
d) [rein] [rén-ì] ‘reign’ (v)
e) [donet] [dònét-ì] ‘donate’ (v)
In cases in which there are multiple modifications, this occurrence is also consistent, as shown
below.

c) a) [polis] [èpòlís-ì] ‘police’

b) [mɛʃĩ] [èmàsín-ì] ‘machine’
c) [kɔligs] [èkɔ`lís-ì] ‘colleagues’
d) [ʤip] [èzíp-ì] ‘jeep’
e) [bil] [èbíl-ì] ‘bill’
f) [fild] [èfíd-ì] ‘field’
g) [sɔkɛt] [èsɔ´kɛ´t-ì] ‘socket’
h) [kruseid] [èkùséd-ì] ‘crusade’

c) Consonant Deletion
When structures violate the syllable condition of E̩ do to the point that they contain consonant
clusters, their modification reveals the following patterns, towards ensuring that the conditions
are met so that they will perform well in the optimality evaluation. One prominent strategy for
modifying structures with consonant clusters (i.e. where syllables contain branching onsets) is
the deletion of the second consonant in the occurrence.
d) a) [gold] [ígólù] ‘gold’
b) [prinsipl] [èpísípâ] ‘principal’
c) [flask] [èfìlásì] ‘flask’
d) [fɛstɔs] [èfɛ´sɔ´sì] ‘Festus’
e) [pastɔ] [èpásɔ̂ ] ‘pastor’
38
f) [printa] [èpıt´̃ â] ‘printer’
When this second consonant deletion is not the case, the first consonant in the
occurrences are deleted, as in the following:
e) a) [ɛsplod] [ɛ`pìlódù] ‘explode’
b) [adaptɔ] [àdátɔ̂ ] ‘adaptor’
c) [afɛkt] [àfɛ´tì] ‘affect’
d) [siks] [èsísì] ‘six’;

It can be said to have been motivated by the fact that the retained sound is central to the
understanding (or recognition) of the entire word. This justification is also believed to guide the
deletion of the second consonants as shown above.
Where items end in consonants (i.e. syllable codas), the nativization process ensures that such
structures are modified in line with the phonological condition of the language to ensure that
they do not violate the constraint that rejects such structures.
As obviously suggested in the data, the deletion of [l] (as is the case of [r] which is generally not
realized) when it occurs in syllable coda position is motivated by the need to maintain an open
syllable, hence the removal of the coda.
The realization of nasality on vowels in the environment of the alveolar nasal [n] which is
followed by the deletion of the nasal consonant is also observed in the data. Consider the
following:
a) a) [ɛlɛkʃiɔn] [ɛ`lɛ´síɔ`]̃ ‘election’

b) [aksidɛnt] [ásídɛ̂ ̃tì] ‘accident’

c) [printa] [pıt´̃ â] ‘printer’

d) [enʤɛl] [e´zɛ̂ ] ‘angel’
e) [ɛnvɛlop] [ɛ´̃vɛ´lópù] ‘shrine’

This form of modification is found at the beginning (as in 22d – e), middle (as in 22b – c) or end
(as in 22a) of given words. Recall (in section 3.2, above) that E̩ do does not permit the occurrence
of nasal vowels at word initial position, hence such forms are thought to

39
occur due to the fact that they are derived by nasalization, followed by consonant deletion (i.e. of
the alveolar nasal [n]). Consonant deletion affects the syllable to the extent of reducing the
number of sound in a word but not the number of occurring syllables. This is due to the fact that
the deleted segment (i.e. a consonant) does not feature as a syllable peak.
d) Vowel Insertion

Where structures with consonant clusters are not modified by consonant deletion, they are
usually modified by the insertion of a vowel in between the clustering consonants as a way of
changing the word to enable it fit into the E̩ do syllable structure. The choice of the vowel for
insertion is usually made between the front, unrounded high vowel [i] and the back rounded high
vowel [u]. The use of [i] is observed to be very productive in this regard, as found in the
following:

a) [wiski] [èwísíkî] ‘Whisky’

b) [grama] [ègírámɲ̂ ̃] ‘grammar’

c) [dɔktɔ] [èdɔ´kítâ] ‘doctor’

d) [greip] [ègírépì] ‘grape’
e) [batri] [èbátírè] ‘battery’

There are a few items that do not correspond to this environmental conditioning of
occurrences, however. Examples of these include:
b) a) [spageti] [èsùpàgétî] ‘spaghetti’
b) [brɛd] [èbùrɛ´dì] ‘bread’
c) [flawa] [ífúláwâ] ‘flower’

The assumption here is that this inconsistency will be taken care of in the passage of time when
these words would have been fully nativized.

40
In the following section, items in the different classes discussed earlier (3.1.3, above) are
evaluated against their violations, or otherwise, of the relevant constraints in E̩ do. Their
performance in the evaluation process justifies their nativization status, as well as shows what
remains to be modified in the form of violations.

41