ISSN 2348-3024

AKOKOID COMPARATIVE WORDLIST

Fadoro , Jacob Oludare
Department of Linguistics and African Languages,
University of Ibadan, Ibadan

ABSTRACT
Akokoid, in this paper, refers to the nine speech forms which are spoken in Akoko North-West Local Government
Area of Ondo State in South-Western Nigeria. These speech forms are Arigidi, Erushu, Afa, Oge, Aje, Udo, Oyin,
Igashi and Uro. Since the 1970s, scholars have lumped these speech forms together as dialects of the same
language without any detailed lexicostatistic investigation. Thus, the major objective of this paper is to determine
whether the speech forms are really dialects of the same language through lexicostatistic analysis. Data were
collected from 34 informants spread across the nine geographical areas where the speech forms are spoken through
the direct interview method. In analyzing these data, Swadeshs principles of lexicostatistics were used. The
lexicostatistic figures reveal that Arigidi and Erushu are 88.5% cognate, so they are classified as dialects of Arigidi.
Afa, Oge, Aje, Udo, Oyin, Igashi and Uro are 81% cognate, so they are classified as dialects of the same language
called O won (meaning tongue). Two distinct but fairly related languages were identified within the nine speech
forms. These are Arigidi and Owon, jointly referred to as Akokoid by virtue of their Akoko root. Therefore, there is no
justification for lumping them together as dialects of a single language.

Council for Innovative Research

Peer Review Research Publishing System

Journal: Journal of Advances in Linguistics

Vol 3, No. 1
editor@cirworld.com
www.cirworld.com, member.cirworld.com

156 | P a g e

March 8,2014

ISSN 2348-3024
0.

INTRODUCTION

The Akokoid speech forms have attracted different names from scholars since the 1970s. For example, Hoffman
(1974) refers to them as Northern Akoko Cluster; Akinkugbe (1978) refers to them as Akokoid; Capo (1989) calls them
Amgbe; Akinyemi (2002) gives them the title Arigidi-Amgbe; while Fadoro (2010) retains Akokoid.
These nine speech forms are all spoken in Akoko North-West Local Government, Ondo State, Nigeria, with the local
government headquarters in Oke-Agbe. The speech forms are Arigidi, Erushu, Afa, Udo, Oge, Aje, Oyin, Igashi and Uro.
The names given to the speech forms coincide with the names of the communities in which they are spoken. Afa, Udo,
Oge and Aje are spoken in Oke-Agbe. See Fig 1, 2 and 3 below for maps.

Figure 1: MAP OF ONDO STATE

Source: Akoko North West Local Goernment Council, Ondo State, Nigeria

Figure 2: MAP OF AKOKO NORTH WEST LOCAL GOVERNMENT AREA

OF ONDO STATE SHOWING THE STUDY TOWNS
Source: Akoko North West Local Goernment Council, Ondo State, Nigeria

157 | P a g e

March 6,2014

ISSN 2348-3024

1.

RESEARCH PROBLEM

The nine speech forms explored in this work have been lumped together by scholars in the 1970s and 1980s
without a detailed lexicostatistic investigation. This resulted in the hasty conclusion that they are dialects of the same
language. However, a careful look at these speech forms reveals that they cannot be lumped together as dialects of the
same language. In fact, speakers in their respective communities do not agree that they speak the same language with
speakers in adjacent communities. This calls for a lexicostatistic investigation, which this paper has attempted.

1.1 OBJECTIVES OF THIS PAPER

The major goal of this work was determining the status of the Akokoid speech forms. It examined whether the nine
speech forms are dialects of the same language or not. It was investigated the level of intelligibility among the over
250,000 speakers who claim that they speak different languages.

1.2 METHODOLOGY
The Ibadan 400 wordlist was used to obtain data from 34 informants: Aje (5), Arigidi (5), Oyin (5), Igashi (4),
Erushu (3), Afa (3), Oge (3), Udo (3) and Uro (3). Thirty-two of them were non-mobile as well as rural. Twenty-five of
them were old, with their ages ranging from 60-80 years old. Twenty-seven of them were males. Thus, the acronym
NORM(s) (which stands for Non-mobile, Old, Rural, Males) was used with little modifications. These informants were
contacted and interviewed in their villages.

1.3 THEORETICAL PERSPECTIVE

In analyzing the data presented in this work, Pikes (1947) discovery procedure in phonological analysis was
employed in determining the phonemic status of the sound systems of Akokoid (Consonants, Vowels and Tones).
Secondly, Swadeshs (1951) principles of lexicostatistics were used in classifying the speech forms into two main
languages Arigidi and O won . Arigidi is made up of Arigidi and Erushu speech forms. O won
is made up of Afa, Aje, Udo,
Oyin, Igashi, Aje and Oge.

2.

THE CONCEPT OF VARIATION

Bright (1966:2) makes the following observation:

Within any recognizable speech community, variations are normally found on all levels of linguistic
structure Phonological, grammatical and lexical. Some of the variations are correlated with
geographical location some may depend on the identity of the person spoken to or spoken about
other variations are correlated with the identity of the speaker. These include cases of difference between
mens and womens speech linguistic variation may also be correlated with the social status of the
speakers (or) with other facts in the social and cultural context.
The excerpt above suggests that variation in language could be determined by different characteristic features.
The correlates of linguistic variation could be any of the following:
(i)

Geographical location of the speaker (where the people live)

(i)

The interlocutors (age disparity and class)

(iii)

The topic of discussion or context (what the interlocutors discuss)

(iv)

The sex of the speaker/addressee (whether they are males or females)

Brights comments above also serve as a springboard for this study. The factors spelled out by Bright manifest in
the use of language in Akoko. For instance, Dada (2006) and Oyetade (2007) note that there was a significant difference
between the different age groups in their ability in Akoko languages and between different occupations. According to
them, ability in Akoko languages is gradually dwindling; generally children are not as proficient in Akoko languages as
adults. Apart from this, the males have a slightly higher proficiency in their Akoko languages than their female
counterparts. Deliberate efforts were made to confirm these observations. This was why we modified NORM(s) in the
selection of informants. For example, five of our informants were female, seven were young and six were mobile.

2.1 FACTORS THAT CAUSE VARIATION IN LANGUAGE

Many reasons have been ascribed to variation in language. One of the prominent characteristics of language in
general is the fact that language is dynamic, that is, it is not monolithic. It changes from time to time. Brook (1973:162)
remarks that the spoken language is like a living person growing older continuously but so slowly that it is hard to point to
any one changing feature.
One of the factors responsible for variation in language is the social organization of the speech community itself.
This includes differences in age, sex, social status and the setting in which interaction takes place. Bright and Ramaanjan
(1964) capture this in the following words:

158 | P a g e

March 6,2014

ISSN 2348-3024
What is termed the social dimension of linguistic variation is correlated with the socially established identity of the
speaker/or the person addressed or mentioned.
Furthermore, regional varieties of language develop as different norms arise in the usage of groups who are
separated by some kind of geographic boundary. Geographical distance is an important factor that causes variation in
language. Fromkin and Rodman (1993) assert that regional diversity develops when people are separated from each
other geographically and socially. This is because the changes that occur in the form spoken in one area or group do not
necessarily spread to another. Dialect differences tend to increase proportionately to the degree of communicative
isolation between the groups. Communicative isolation refers to a situation such as existed between America, Australia,
and England in the eighteenth century. Other factors responsible for variation in speech are sex, time, and education.
The data presented in this paper highlights the similarities and differences attested in Akokoid.

COMPARATIVE WORDLIST
In an attempt to expose the internal relationship within the Akokoid speech forms, two hundred lexical items
consisting of one hundred nouns (such as names of body parts and natural phenomena, like sun, moon, star, fire, rain,
river, day, night etc and one hundred simple verbs (like go, come, run, eat, see, die, etc.) are presented here. Because of
their everyday significance, these words are believed to be least prone to change or borrowing, as every culture and
linguistic group has words for them. These items are presented with their Yorb and English equivalents in Table 1
below:
TABLE 1: COMPARATIVE WORDLIST

159 | P a g e

March 6,2014

ISSN 2348-3024

160 | P a g e

March 6,2014

ISSN 2348-3024

161 | P a g e

March 6,2014

ISSN 2348-3024

162 | P a g e

March 6,2014

ISSN 2348-3024

163 | P a g e

March 6,2014

ISSN 2348-3024

Note: Non-cognate items in Akokoid are put in brackets.

2.3 DISCUSSION ON THE COMPARATIVE WORDLIST/FINDINGS

When one looks at the variations exhibited above, one wonders how, in spite of these variations, there is
intelligibility among the speakers of the nine speech forms under investigation. The point is that intelligibility is a matter
of degree, ranging from total intelligibility to total unintelligibility. Our findings reveal that:
(i)

Arigidi and Erushu

(ii)

Afa, Udo, Oge and Aje

(iii)

Oyin, Igashi and Uro

One to three above are organised or arranged in a dialect continuum, in which a chain of adjacent varieties are
mutually intelligible, but pairs taken from opposite ends of the chain are not. Thus: Arigidi and Erushu are mutually
intelligible. The same thing applies to Afa, Udo, Oge, and Aje. Oyin, Igashi and Uro are also mutually intelligible.
Intelligibility is also mutual between Udo, a member of group (ii) above and Oyin, a member of group (iii) above. The
reason, according to the Oloyin of Oyin, Oba L.O. Bamisile, is that Udo and Oyin were together until 1922, when Udo
migrated to Oke Agbe to join Afa, Oge and Aje to form Oke Agbe. The story is similar for Igashi and Uro which were
together until 1955 when Uro left to join Daja, Ojo, Efifa, Iludotun, Oso, Ora and Esuku to form Ajowa community.

164 | P a g e

March 6,2014

ISSN 2348-3024
Akokoid

w n

Arigidi

Forms spoken in

Forms spoken outside

Arigidi

Erushu

Mutually intelligible
Oke Agbe

Afa Aje

Udo

Oke Agbe

Oge

Oyin

Mutually intelligible

Igashi

Uro

Mutually intelligible

Figure 4: Akokoid Classification

2.4 DISCUSSION ON THE COMPARATIVE WORDLIST

The two major objectives of this work stated in Section 1: 1 above are to:
(i)

Determine the status of the Akokoid speech forms in relation to one another.

(ii)

Investigate the level of intelligibility among the speakers.

These are the issues that shall engage our attention in the rest of the paper.

2.4.1 THE STATUS OF THE AKOKOID SPEECH FORMS

As stated in Section 1, the nine lects that constitute the Akokoid speech forms have been lumped together by
previous scholars as dialects of the same language without a detailed lexicostatistic investigation. Our major goal in this
section is to either confirm this claim or refute it. Swadesh (1951) claims that the relationship between two or more lects
can be determined on the basis of cognation percentage as outlined below:
Cognate%

Term

100-81

Language

80-35

Family

34-12

Stock

12-4

Microphylum

4-1

Mesophylum

Less than 1

Macrophylum

Our lexicostatistic count on the basis of the 200 lexical items above, shows that 152 items, which constitute 76%, are
virtual cognates. Thus by Swadeshs standard, the nine speech forms belong to the same family. We refer to this family as
Akokoid. A deeper look, produces the following breakdown:

Arigidi and Erushu average 88.5% between themselves. This figure qualifies them as dialects of the same language,
referred to as Arigidi.

165 | P a g e

March 6,2014

ISSN 2348-3024

Afa, Udo, Oge, Aje (Ese), Oyin, Igashi and Uro average 81% among themselves. This qualifies them as dialects of
the same language , which we refer to as wn . The term is particularly preferred in this work because it means
tongue in the seven speech forms, even though its previous use was restricted to Afa, Udo, Oge and Aje. The overall
title used is Akokoid. It has a general application by virtue of the fact that all the speech forms are located in Akoko.

Arigidi and O wo n average 76% between themselves; this established them as two different but related languages.
They are jointly referred to as Akokoid in this work, as shown below:

Akokoid

76%

w n

Arigidi

81%
88.5%

Arigidi

Erushu

Afa Oge Aje (Ese) Udo Oyin Igashi

Uro

Figure 5: Akokoid according to cognation percentages

2.4.2 MUTUAL INTELLIGIBILITY

Ideally, the distinction between language and dialect is based on the notion of mutual intelligibility. Dialects of the same
language should be mutually intelligible, while different languages should not be. Mutual intelligibility is a reflection of the
linguisitic similarity between the different varieties of speech. From our data above, the issue of linguistic similarity is
conspicuous. For instance, the 88.5% cognation between Arigidi and Erushu implies a considerable similary between the
two lects. The same thing applies to the Own lects, which share 81% cognation among themselves. When we compare
this with the 76% between the two groups Arigidi and Owon, it would be reasonable to draw the conclusion that the level
of intelligility within each group is very high. Let us extract some data to show this.
TABLE 2: AKOKOID/EDOID

The Edoid items are extracted from Elugbe (1989)

These are just few examples of total cognates within each group among several other ones. Apart from this, there
are many others which are examples of partial cognates. Our findings reveal that the following are intelligible:
(i)

Arigidi and Erushu

(ii)

Afa, Udo, Oge and Aje

(iii) Oyin, Igashi and Uro

I to III above are organised or arranged in dialect continuum, in which a chain of adjacent varieties are mutually
intelligible, but pairs taken from opposite ends of the chain are not. Thus Arigidi and Erushu are mutually intelligible. The
same thing applies to Afa, Udo, Oge and Aje. Oyin, Igashi and Uro are also mutually intelligible. Intelligibility is also

166 | P a g e

March 6,2014

ISSN 2348-3024
mutual between Udo, a member of group (ii) above and Oyin, a member of group (iii) above the reason, according to the
Oloyin of Oyin, Oba L.O. Bamisile, is that Udo and Oyin were together until 1922, when Udo migrated to Oke Agbe to join
Afa, Oge and Aje to form the Oke Agbe settlement. In fact, it was said that the Oloyin had agreed to join them earlier, but
as he was preparing to go with them, he was warned by his drummer through the talking drum not to follow them.
Likewise, Igashi and Uro, were together until 1955, when Uro left to join Daja, Ojo, Efifa, Iludotun, Oso, Ora and Esuku to
form the Ajo wa Community. This account was given by the traditional ruler of Uro Ajo wa, Oba Timothy Sunday
Ipinmoroti, who was one of our informants. Thus, there is mutual intelligibility between Uro and Igashi to date because
they still interact closely. However, the case is different when one compares Arigidi or Erushu with Afa or Aje (which are
members of group (ii) above) or when one compares Arigidi with Oyin (a member of group (ii) above. There is a partial
intelligibility. However, it is pertinent to note that intelligibility exists among these speech forms in varying degrees
because the overall distance is not considerable and a lot of intermarriages still go on among them. The point we are
emphasizing here is that the similarity that exists among the speech forms is not produced by accident. Obviously, they
descended from the same proto-form; therefore; they are sisters. So, we postulate a common source for them, in the
name of Akokoid. This is summarised in the tree diagram below;

Akokoid

w n
Forms spoken in

Arigidi
Forms spoken outside
Oke Agbe

Oke Agbe

Afa Aje

Udo

Arigidi

Oge

Mutually intelligible

Oyin

Igashi

Erushu

Mutually intelligible

Uro

Mutually intelligible

Having discussed the two key issues we highlighted as our objectives, what is now left for us is to discuss how
variables like age, gender, location and migration patterns have affected our result.

AGE
The ages of individual informants have a serious effect on their mastery of their speech forms. Out of the 34
informants used in this research, only 7 were young. The remaining 27 were old. The major problem we had with these
young informants was the level of mastery of their speech forms. In fact, during our sessions with them, several gaps were
created, at intervals when asked: What do you call this item in your speech form? They would respond: We dont know.
Then we would have to resort to the older informants who would quickly supply the data. Several times, the data our
young informants supplied were actually Yoruba words and, in an attempt to validate their answer, we would turn to the
elders who would tell us the answer in their speech forms. This is a confirmation that these young ones were deficient in
the mastery of their speech forms. This aligns with Oyetades (2007) observation that ability in Akoko languages among
the young ones is gradually dwindling, such that they are not as proficient in their speech forms as the adult speakers. In
fact, during our field trips, parents complained bitterly that their children could no longer speak their languages. This is a
serious indication that the process of language shift among the youth is ongoing.

SEX
Research in Western nations affirms that womens speech is considered to be more self-conscious and classconscious than mens speech, such that data collected from them are not as reliable as those from men because they
pretend to be sophisticated and artificial in their utterances. Conversely, mens speech is purer and more original
because they are rather conservative (see Chambers and Trudgill 1998 and Francis 1983). This seems to be the case
with our female informants in Akokoid. In fact, it was obvious that the men were more competent than their female

167 | P a g e

March 6,2014

ISSN 2348-3024
counterparts. We had to turn to the male informants several times because the female informants were not able to
satisfactorily supply our data. That is why only 5 informants were selected from the female, whereas 29 were males.
Oyetade (2007) has reported that the male informants have a higher proficiency in their speech forms than their female
counterparts.

LOCATIONS
Out of our 34 informants, 32 lived in their different Akoko villages, while only 2 lived in Ibadan as at the time of data
collection. Our experience confirms that these mobile informants are not as proficient as their non-mobile (stationary)
counterparts in their mastery of their speech forms. This is natural and logical. The fact that they have been away from
their root for years has detached them somehow from their mother tongues. They were now more comfortable with Yoruba
and English which their dwelling in city has better exposed them to Traditional dialectologists believe that the use of nonmobile informants guarantees that their speech is characteristic of the region in which they live, free from any external
influence.

MIGRATION
According to the traditional rulers of Uro and Oyin, all their ancestors were together in Ile-Ife at the initial stage. But
as a result of threats of war, they migrated to Benin. They later had to leave Benin, and migrated to a hilly settlement near
their present locations. In 1922, Revd. Lennan, an Anglican priest, encouraged them to leave their hilly settlement
because they were isolated. So, it was in 1922 that the first wave of migration took place. More waves of migration
followed later, as we have discussed earlier. These waves of migrations have occasioned the differences we see in our
data. A careful look at our data also shows that many lexical items are cognate with their Yoruba equivalents. For
instance, let us look at the forms for king, cow, cat, cassava, toad, tortoise, crab, bee and snail, and so on. In fact, Fadoro
(2013) forthcoming claims that Akokoid is 36% cognate with Yoruba. Also the migration to Benin has left its mark on the
speech forms. An extract from Elugbe (1989) will show this.

The Edoid items are extracted from Elugbe (1989).

A look at the above items reveals a striking similarity. It would not be out of place if one describes them as cognates.
From the foregoing, we can safely assert that the migration patterns of these people have left their marks on their speech
forms in terms of some of the similarities and variations that exist till today.

2.5 CONCLUSION
In this work, an attempt has been made to examine the relationship within the Akokoid speech forms. The data
used for this work show that the speech forms are closely related with minor differences at the phonological and lexical
levels. In most cases, the items are 100% cognate in terms of consonants, vowels and tones; in some cases, there are
minute differences. The similarity of these speech forms in the data establishes a common ancestry for the speech forms.
Apart from that, it proves that truly language is not monolithic and that variation in language is a norm rather than an
exception. Through the lexicostatistic analysis, we are able to conclude that the nine speech forms are classifiable into two
different but related languages, which are Arigidi and Owon
. This classification is also supported by mutual intelligibility.
Finally, the nine speech forms constitue a dialect continuum.

REFERENCES
[1] Akinkugbe, O.O. 1978. A comparative phonology of Yorb dialects, Itsekiri and Igala. Ph.D thesis, University of
Ibadan.

168 | P a g e

March 6,2014

ISSN 2348-3024
[2]

Akinyemi, O.V. 2002. A study of the internal relationship within Northern Akokoid. Unpublished B.A.,
University of Ibadan.

[3] Bright, W. & Rameajaan, A.K. 1964. Sociolinguistic variation and language. Proceedings of the 9
Congress of Linguistics. H. Lint (Ed.). Cambridge, Massechusetts, The Hague: Monton. 1107 12.

th

Project,

International

[4] Bright, William (ed.) 1966. Sociolinguistics. The Hague: Mouton.

[5] Brook, G.L. 1973. Varieties of English. London: Macmillan.
[6] Capo, H. B. C. 1989. Defoid, The Niger-Congo Languages. J. Bendor-Samuel (Ed.) New York and London: University
Press of America. Pp. 275 90.
[7] Chambers, J.K. & P. Trudgil 1998. Dialectology. Cambridge: Cambridge University Press.
[8] Dada, S. A. 2006. Erushu-Yorb endoglossic bilingualism and language alternation in Akoko, South-western Nigeria.
Unpublished Ph.D. thesis, University of Ibadan.
[9] Elugbe, B.O. 1989. Comparative Edoiod: phonology and lexicon. Port-Harcourt: University of port-Harcourt Press.
[10] Fadoro J.O. 2010. Phonological and Lexical Variations in Akokoid. Unpublished Ph.D thesis, University of Ibadan,
Ibadan.
[11] Fadoro 2013. (Forthcoming). The status of Akokoid in relation to the Yoruba language.
[12] Francis, W.N. 1983. Dialectology: an introdeuctin to language. Harcourt Brace Jonavanovish.
th

[13] Fromkin, V. & R. Rodman 1993. An introduction to language. (5 edition): Harcourt Brace Jovanovich.
[14] Hoffman Carl 1974. The languages of Nigeria by language family. An index of Nigerian languages, H.J. BendorSamuel and R. Stanford (Eds.) Ghana: Summer Institute of Linguistics, pp. 169 190.
[15] Hudson, J. A. 1996. Sociolinguistics. Cambridge: Cambridge University Press.
[16] Oyetade S.O. 2007. Language endangerment in Nigeria: perspectives on Akoko languages of the Southwest. Dorian.
N.C., (ed.) Small languages and small language communities. Intl. Journal of Sociology of Language. Pp. 169 184.
[17] Swadesh, M. 1972. The origin and diversification of language. London: Routledge & Kegan Paul.

169 | P a g e

March 6,2014