The Buckingham Journal of Language and Linguistics Volume 3 pp 187-203
On Efik Prefixing Morphology
Eyo Offiong Mensah*
Abstract
Prefixation is a grammatical devise that involves the attachment of a bound morpheme to the left of a root element or stem. It functions to signal certain grammatical relationships involving categories like tense, negation, person, number, and aspect. It can also trigger off the creation of new words from existing ones. The focus of this paper is to analyse the forms and structure of Efik prefixes in relation to the different phonological and morphosyntactic operations they can signal in the language. The paper also examines the various word formation strategies involving prefixation in the language. The basic assumption, however, is that the structure of Efik prefixes vary according to agreement, and is determined by the principle of vowel harmony. The study discovers that Efik prefixes have systematic and rule-governed structures and that certain conditions, such as the phonology of the stem, the stem’s lexical category and the semantic value of prefixes stipulate their position. We wish to interpret the following abbreviations in order to facilitate our analyses: Adv(erb), Asp(ect), Aux(illiary), Conj(unction), Fut(ure tense), Mod(ality), Neg(ation), N(oun), NP (Noun Phrase), Pres(ent tense), PT(Past tense), Pro(noun), Pfx (Prefix) SC (Subject Concord), Spec(ifier), Tn (tense) and V(erb).
Introduction
Efik is the language of the Efik people who inhabit the coastal areas of the Cross River comprising Calabar South, Akpabuyo, Odukpani, Calabar Municipality and Bakassi Local Government Areas of Cross River State, Nigeria. The language has been variously classified under the Niger-Congo family of the Niger-Kordofanian Phylum (Greenberg 1963). Faraclass (1989) classifies it under the Lower-Cross sub-group of the Delta-Cross family which is an off shoot of the enlarged Cross River, which is a major constitution of the Benue-Congo sub-family. From here, it is clearly established that Efik is a Lower-Cross language which together with the Upper-Cross group of languages constitutes the Delta-Cross family, which is an immediate sub-branch of the enlarged Cross River group of languages.
According to Bussman (1996:314), the term morphology was coined by a biologist W.Von Goethe (in 1822) to designate the study of the form and structure of living organisms and was subsequently taken up in linguistics in the 19th century as a cover term for inflection and word formation. Morphology is a core area of language description that is concerned principally with the study of the internal structure of words, and word formation processes generally. An aspect of the morphology of a particular language is a set of rules with dual functions. First, these rules are responsible for word formation or the creation of new words, and they represent the speakers’ unconscious knowledge of the internal structure of the existing words in the language (O’ Grady et al: 1989). These words are not the smallest segment of meaning in a language rather, it is the morpheme. This is not to say that affixes are words. They are not words but are components of words that add new meaning to the existing words. They are accounted for within the contexts of word creation and in expressing grammatical contrasts. Katamba (1993:20) describes the morpheme as the smallest indivisible units of semantic content or grammatical function which words are made of. It is also the smallest difference in the shape of a word that correlates with the smallest difference in word or sentence meaning or in grammatical structure.
Morphemes may be free or bound. A free morpheme occurs in isolation, and may constitute a word by itself, but bound morphemes cannot convey any meaning in isolation: they need to be attached to other elements, which may be free or bound. For instance, in the English word teachers, teach is the free morpheme or the root element, while /-er/ and /-s/ are bound morphemes, which merely contribute meaning: /-er/ takes the verb teach and converts it to a noun, while /-s/ indicates plural grammatically.
Bound morphemes are generally subsumed as affixes; a useful general term for the recurrent formative morphemes of words other than roots (Robins 1989). Affixes may be classified into three categories according to the position they occupy in relation to root morphemes. Prefixes, infixes and suffixes. Prefixes always precede the root or other element, which may be bound morphemes. In other words, they occur to the left of the root morpheme. Infixes split the root and appear in between, that is they occur within the root morpheme, while suffixes are attached at the end of the root element. This implies that they occur to the right of the root morpheme. This paper is primarily concerned with prefixes, their forms, functions, and morphosyntactic properties in the Efik language. In the following analysis, we examine prefixation in the light of its role in signaling grammatical contrasts.
Prefixation in tense
The category of tense relates the time of the situation to the time of the utterance or moment of speaking about the situation (Comrie 1976, Essien 1990). A tripartite tense opposition exists in Efik, that is, the traditional present, past and future tenses. /má-/ and /ké-/ are past tense prefixes, /mé-/ and /Ø/ are present tense prefixes and /yé-/ and /dí-/ are the future tense prefixes.
Using Essien (1983) categorization of tense allomorphs in Ibibio, we represent the tense markers in Efik (a mutually intelligible language with Ibibio) as follow:
Table 1: Efik tense allomorphs, after Essien (1990:78)
|
|
Tense opposition |
S-type I |
S-type II |
(a)
|
Past
|
má- |
ké- |
|
(b) |
Present |
mé- |
ø |
|
(c) |
Future |
yé- |
dí- |
Where S- type I refers to such sentences as:
(a) A simple declarative sentence;
An affirmative sentence;
A neutral sentence;
A modally unmarked sentence
and
A sentence that asks yes-no
question.
And where S- type II refers
to sentences like:
(a) A negative sentence
A modally marked sentence
A sentence with emphasis
An interrogative sentence
that cannot take yes-no answer.
In what follows, we examine
the tense division in Efik, using Essien (1983) notion of S-type I and S-type
II model as the basis for our analysis of tense prefixes.
Tense prefixes are usually found in association with other prefixes, particularly those indicating person, number and the root element, for instance, the past tense prefix is interwoven between the subject concord (SC) and the root constituent. These prefixes function to indicate a point in time that precedes the moment of speaking, and they only occur with type1 sentence category. In indicating the past tense, series of prefixes concatenate to be associated with the root of the verb:
1. Bassey
á má- á
diá ùdiá - ‘Bassey ate food’
NP SC Pt
SC V NP
Bassey he
past he eat
food
In 1, the SC /á-/, the past tense prefix /má-/ and the SC /á- / are affixes which are linked to the verb root. The occurrence of triple subjects in 1 (the NP and two SC prefixes) is an important characterization of Efik syntax. Langacker (1972) calls this kind of phenomenon sentence trapping, which is merely required by the syntax of the language. Observe the harmonic relationship between the vowels of these prefixes with the vowel of the root verb. In 1, /má-/ occurs as a prefix and the SC vowel influences it to copy all it features. This takes place in all persons except the first person singular, which requires a syllabic nasal, which is homoganic with the following consonant. We shall make this point more explicit in our subsequent discussion of the category of person.
In contrast, the /ké-/ allomorph of the past tense morpheme occurs in type II sentence category. For instance, if we negate the simple declarative sentence in 1, the output will be 2:
2(a) Bassey
í kí- diáhá
ùdiá - ‘Bassey did not eat
food’
NP SC Pt
V Neg NP
Bassey he
past eat not
food
The /ké-/ prefix shows that the time of the situation and the time of the utterance are not simultaneous, hence indicating an imperfect correlation between the tense prefix and the time reference it describes (Essien 1983, Radford 1999). In 2, the /ma-/ past tense prefix cannot be used again because the sentence is negated. A different form of tense prefix /kí-/ is introduced to account for the phenomenon of negation. This is a case of syntactic conditioning of allomorphs. Observe that the vowel of /ké-/ prefix harmonizes with the SC prefix which becomes obligatorily /kí-/ in the negative sentence in 2. Generally, the vowel of the SC determines the harmony in all persons except the first person singular, where the verb root determines the harmony as we can see in 2(b) and (c) respectively:
2(b) Ń ké dép ké - ‘I did not buy’
SC Pt V Neg
You past buy not
(c) Ń ké kéréké ntrĕ - ‘I did not think so’
SC Pt V Neg Adv
You past hear not so
There are two allomorphs of the present tense morpheme in Efik. These are /mé-/ and /Ø/. The use of /mé-/ parallels the use of /má-/ past tense prefix, in the sense that every kind of sentence the former can be used, the latter can equally be applied. In other words, they are distributionally equivalent. /mé-/ occurs in type I sentence category. The /mé-/ present tense prefix indicates that the action or event has taken place concurrently with the moment of speaking, hence, is of present relevance:
3. (a) M mé- kèmé ń .dí nám - ‘I can do it’
SC Pres Aux SC Tn V
I can I do
(b) M- mé- yĕ - ‘I am pretty’
SC Pres adj
I Pretty
The use of /mé-/ in Efik is similar to the way sentences with have/has present tense are understood in English (Essien 1983). Observe the influence of the vowel of the root morpheme on the vowels of the present tense prefix. The use of /Ø/ present tense allomorph is quite different structurally from /mé-/. It is a null morpheme, or what Spencer (1991:20) calls a zero affixation, where the morpheme responsible for marking the tense is clearly a ghostly one. It is applicable to the type II sentence category. If we convert the examples in 3 into negative constructions, we will have 4:
4. (a) N kèmé ké ń dí nám - ‘I cannot do it’
SC Aux Neg SC Tn V
I can not I do
(b) N yĕ hé - ‘I am not pretty’
SC adj Neg
I pretty not
In 4, we have observed that the /mé-/ present tense prefix has given way to a null affixation process. In other words, there is complete absence of a tense prefix to indicate that the sentences have present relevance. Observe the different forms of the negative suffixes. In Efik, the phenomenon of NEG is syllable driven; a one-syllable root takes the suffix /-hv/ as in 4(b), while two or more syllable roots take the suffix /-ké/ as in 4(a). In Efik, adjectives exhibit similar syntactic behaviour as verbs as they can manifest concord with the subject of the sentence. They can also be inflected for tense and negation (as well as other categories) as we can see in 3(b) and 4(b). The use of /Ø/ in the imperative order does not require the SC when it occurs with the singular person, but does so if it occurs with the plural person as shown in examples 5(a) and (b) respectively:
5 (a) (Àfò) dí mí - ‘You (sg) come here’
Pro V Adv
You come here
(b) (Mbùfò) é- dí mí - ‘You (pl) come here’
Pro SCV Adv
You you come here
In 5(a) and (b), the subject
pronouns (enclosed in parenthesis) are optional elements. This implies that the
sentences could still be complete and meaningful without them, given the
agreement between the verb and the noun in respect of some grammatical
relationships..
There are two future tense
prefixes in Efik; /yé-/ and /dí-/. While the former occurs with the type I
sentence category, the latter can only be used with the type II sentence
category:
6 (a) (Àmì) ń yé-
yéné ókúk - ‘I will
have money’
Pro Sc fut V NP
I I
will need money
(b) (Àmì)
ń dì yéné ké ókúk - ‘I will not have money’
Pro SC fut V
Neg NP
I I
will need not money
In this case, the future
tense prefix is flanked by the SC prefix, and the verb root yéné ‘have’. The subject pronoun Àmì ‘I’
is an optional element.
Prefixation in negation
The phenomenon of negation has been described by Mensah (2001: 61) as when a proposition in an affirmative statement or command is reversed and the expression acquires a contrary sense with the introduction of corresponding negative affixes and phonological processes. In this way, the structure loses its positive values and is then said to be negated. In Efik, this process is basically inflectional, in which case it involves prefixation and suffixation. We focus primarily on prefixation in the following analysis. There are two types of preverbal negative prefixes in Efik: /kû-/ and /dí-/. The /kû-/ prefix which carries a consistent falling tone is attached to the root of the verb in negating an imperative construction:
7 (a) (Àfò) kû- nènì -
‘(You) do not argue’
Pro Neg
V
You not
argue
(b) (Àfò) kû- nyàm - ‘(You) do not sell’
Pro Neg
V
You not
sell
(c) (Àfò) kû- fèhé
- ‘(You) do not run’
Pro
Neg V
You not
run
Here the /kû/ prefix precedes the verb root and its characteristic falling tone is capable of influencing the inherent tone of the verb root. This is applicable if the root is either CV or CVC, where its inherent low or high tone will be changed to a rising one. This evidence shows that the syllable structure is also relevant in marking negation in Efik:
8 (a) Nó mí ùsàn - ‘Give me a plate’
V
pro NP
give
me plate
(b Kû nŏ mí
ùsàn - ‘Don’t give me a plate’
Neg V
pro NP
don’t
give me plate
9 (a) Tèm biá ìkpòñ - ‘Cook yam only’
V
NP Adv
(b) Kûtêm
biá ìkpòñ - ‘Don’t cook yam only’
Neg V NP Adv
Another instance of pre-verbal negation in Efik is the use of the prefix /dv-/, which is mostly associated with imperative constructions. The vowel of this prefixing morpheme may assimilate the features of the vowel of the preceding SC prefix.
10(a) Àfò
mùm ènyé í
dí fèhé - ‘(You)
hold him so that he may not run’
NP V Pro SC Neg V
You hold him
he not run
(b) Ku nam u du
tua - ‘Don’t do so you may not cry’
Neg V
SC Neg V
Don’t do you not cry
The vowel of the /dí-/ prefix usually copy the features of the vowel of the subject concord prefix.
Prefixation in person and number
According to Mensah (2007), the category of person is a concord category between a noun or a pronoun and the verb form. In Efik, every subject noun or pronoun has to agree in number and in person with the verb, which forms the predicate of the sentence. Prefixes are used to establish such concords:
11(a) Efioñ á-nám útóm - ‘Effiong is working’
N SC V NP
Effiong he do work
(b)Efioñ yè Afioñ é-nám útóm - ‘Effiong and Affiong are working’
N
Conj. N SC
V NP
Effiong and Affiong they do work
In 11(a) the SC prefix /á-/ is used to indicate number and person, in this case a singular person. While in (b), /é-/ is used to indicate plural person. In other words, number and person in Efik, apart from sharing the same position slots, also share the same morphemic representations (Mensah (2007). Number and person can also function as the subject of the sentence, if the referential value of the subject is already known:
12(a) Á - nám útóm - ‘He (Effiong) is working’
SC V N
S/he do work
(b) É - nám útóm - ‘They (Effiong and Affiong) are working’
SC V N
They do work
/Á-/ and /é-/ respectively serve as the subjects of the sentences in 12(a) and (b). It follows therefore that if a speaker has background knowledge about the subject of the sentence, the concord, which represents number and person, can easily substitute for the subject as the case form. This is what Allerton (1979: 139) calls “binary opposition” or binary specification of syntactic features. The case form depends on the syntactic relationship with other parts of the sentence.
The first person singular SC prefix /n-/ has three allomorphs with consistent high tones; /n-/, /ñ-/ and /m-/, and which are determined by the phonological environment:
13. (a) Ń diá ùdiá - ‘I am eating’
SC V NP
I eat food
(b) Ñ ká ùruá - ‘I am going to the market’
SC V
NP
I go
Market
(c) M brĕ ból - ‘I am playing football’
SC V NP
I play football
The alveolar nasal selects an alveolar consonant, the velar nasal takes a velar consonant and the bilabial nasal selects a bilabial consonant in the neighbouring environments 13 (a), (b) and (c) respectively. In other words, they are phonologically conditioned and occur in complementary distribution. Syntactically, they function as the subject of the sentence representing the singular person. We can represent the Efik singular person allomorphs in the diagram in Fig. 1:
morpheme
‘singular person’
allomorph
morph morph morph
/n/ /ñ/ /m/
FIG. 1 Efik singular person allomorphs.
The contrast in meaning between the second and the third persons singular and plural is brought about by the distinction in tones of the subject concord prefixes. While the second person singular and plural carry inherent low tones, the third person singular and plural bear high tones. This evidence reveals that tones can bring about grammatical contrast in Efik as we can see in 14 and 15:
14(a) À diá ùdiá - ‘You (sg) are eating’
SC V NP
You eat food
(b) Á diá ùdiá - ‘He/ she is eating’
SC V NP
He/she eat food
15(a) È diá ùdiá - ‘You (pl) are eating’
SC V NP
You eat food
(b) É diá ùdiá - ‘They are eating’
SC V NP
They eat food
Prefixation in agglutination
Agglutination is a morphological process that involves a word in a language containing separate morphemes or what Spencer (1991:224) refers to as long polymorphic words in which each morpheme corresponds to a single lexical meaning or grammatical function. In Efik, these morphemes which are all prefixes form a collocation with a root element to make up larger words or sentences. Essien (1990: 141) submits that verbs may consist of just a root and one or more affixes indicating concord of number, tense, aspect, mood and negation and in such a way that a whole sentence can consist of just a word:
16. É kpé dí mmá ń sí fiòk - ‘Had I been knowing’
SC Mod Pt Aux SC Asp V
Where the prefix /É-/ represents the SC, /kpé-/, the modal marker, and /dí-/ is the tense marker. The prefix /mmá-/ is the auxiliary element, and /n-/ is a subject concord. Aspect is indicated by the prefix /si-/ and the root of the verb is fiòk ‘know’. This state of affairs reveals a fusion where different prefixes exhibit different functions in a string in such a way that a single word can encode a meaning, which would require a fairly elaborate sentence. It also shows that all the grammatical categories in Efik can be marked by prefixes. Further examples of agglutinative constructions include:
17(a) Ú kú kpú sú nám ké - You (sg) wouldn’t have been doing it’
SC Tns MOD Asp V Neg
(b) Ń ké kémé ké - ‘I cannot do it’
SC Tns V Neg
Where the sentences are built up by strings of affixes. Apart from the verbs nám ‘do’ and kémé ‘can’ in 16 (a) and (b) respectively, all other components of the sentences are affixes which express grammatical relationships. Observe that the vowel of the SC determines the harmony in all persons except the first person singular, where the vowel of the verb root influences the vowels of the affixes.
Prefixation in derivation
So far, we have been concerned with the inflectional functions of prefixes, where their forms and structures vary in order to bring about grammatical contrasts. In what follows, we examine prefixation in derivation, that is, where it has the tendency to create new words from existing ones, hence change its formal class. In Efik, nouns can be derived from verbs through prefixation:
18. verbs derived nouns
sáñá ‘walk’ í-sàñ ‘a walk’
dé ‘sleep’ í-dáp ‘a sleep’
sèm ‘speak’ ù-sèm ‘language’
diă ‘eat ú - díá ‘food’
wèd ‘write’ ń- wèd ‘book’
kpá ‘die’ m- kpá ‘death’
kwó ‘sing’ í- kwó ‘song’
brĕ ‘play’ m- brĕ ‘masquarade’
Under this kind of derivation, nominalization is a dominant process. In which case, a verb is converted into a noun. From here, we can propose that prefixes such as /ú-/, /n-/, /m-/ and /í-/ are derivational morphemes. These forms of nouns with verb roots and the prefixes as their constituents are called deverbal nominal (Anagbogu 2003: 704). However, the process of nominalization in Efik transcends prefixation. For instance, in sáñá → í-sàñ, we observe that apart from the prefixation of the vowel in the output of the derivation, there is the delection of the final vowel and a change in the tone of the verb from high to low. Similarly, in dé → í-dáp, the process involves in addition to prefixation, the creation of syllable and vowel change (é→á). These are clearly irregular pattern of nominalization.
Agentive nominal can also be derived from verbs by prefixing the form /àndí-/ to the verb root. The use of /andi-/ is a regular pattern of nominalization, which personalizes the action of the verb:
19. verb (X) derived nouns (doer of X)
kpép ‘teach’ àndíkpép ‘teacher’
wèd ‘write’ àndíwèd ‘writer’
wòt ‘kill’ àndíwòt ‘killer’
kwó ‘sing’ àndíkwó ‘singer’
wàt ‘drive’ àndíwàt ‘driver’
The use of the prefix /àndí-/ is equivalent semantically to the way in which the /-er/ suffix is used in English, which identifies the doer of the action of the verb.
The /àndí-/ prefix can be contrasted with the prefix /á-/, which performs the same range of functions:
20 verb (X) derived noun (doer of X)
má ‘love’ ámá ‘lover’
suá ‘hate’ ásuá ‘hater’
kán ‘win’ ákán ‘winner’
The use of /á-/ prefix to form agentive nouns is potentially limited because very few verbs can be converted in this way. The verb which carries the /á-/ prefix can equally take /àndí-/ and convey the same sense, but not vice-versa. This however implies that the /àndí-/ prefix is more productive in Efik.
Prefixes as heads
From our examination of the structure of Efik words, we have found out most words in Efik are morphologically complex. Going by Williams (1981), Scalise (1984), Spencer (1991) and Owolabi (1995), one of the most influential views in contemporary morphology is that morphologically complex words have heads. Where the term head is used essentially the same way it is used in syntax, that is, the head of a construction is what determines the syntactic category of that construction (Owolabi 1995:104).
Scalise (1984: 96) argues that the head of a morphologically complex word assigns to the entire word its category by means of a mechanism referred to as percolation. He further describes the difference between the head and the other part of a complex word as follows:
… a head ‘counts’ in relation to the feature it carries, while the other elements ‘count’ in relation to the fact that they represent the presence of morphological material.
In Efik, most prefixes have the characteristics of a head, particularly the category changing prefixes. Let us examine the following synthetic compound words and their derivations:
21 verb + noun (X) derived compound noun (doer of X)
(a) míá èkòmò á- míá ékòmò - ‘drummer’
(beat + drum)
(c) kpép nwèd é-kpép nwèd - ‘teacher’
(teach + book)
Going by the basic methodology of lexical phonology, the VPs, míá èkòmò, ‘beat drum’ and kpép nwèd ‘teach book’ in 21 are respectively reintroduced into the lexicon for nominalization through the prefixation of á and é respectively . The prefix functions as the agent or doer of the action described by the VP; hence it is the head of the derived compound noun. It is modified by the verb, and the post verbal noun functions as the complement of the verb. The verb + noun string is dependent on the prefix, which determines the syntactic property of the entire structure. In other words, the prefix determines the lexical category of the whole construction.
Observe the tonal implication of the nominalizing prefix. Its inherent high tone influences the low tone of the initial syllable of the nominal root to become high. This is a form of tonal assimilation. According to Chomsky and Halle (1968), this kind of affixes are said to have a strong boundary, since the can have phonological effects on the base to which they are attached. They are called non-neutral affixes in contrast with neutral affixes, which do not trigger such effects.
We can represent 21(a) structurally as 22:
22 N¹¹
N¹ V¹¹
Pfx V¹
á V N¹¹
miá N
ékòmò
Here, we claim that the prefixes in 20(a), (b) and (c) are heads by virtue of the fact that they assign the category label N to the entire complex nouns of which they are part in 21(a), (b), (c). This implies that these prefixes must also be dominated by the category labelled N. The derivation in 20 and structural representation in 21 clearly show that Williams’ (1981: 248) Right-hand Head Rule (RHR), which forbids prefixed heads and allows only the right hand member of a complex word the morphological privilege of functioning as the head of the complex word in question cannot apply universally. This is because we have shown that in Efik, prefixes are not just heads of complex words, they are also left headed.
By regarding prefixes as heads, the complex nouns of which these prefixes are part demonstrate a striking structural similarity to phrases which are left-headed in Efik. However, a problem on whether or not all morphologically complex words in Efik are heads arises. This is because some complex words in the language are reduplicated, in which the syntactic categories of the root and the syntactic category of the complex nouns are exactly the same. This is the class maintaining forms. In the case of the class–changing forms, the syntactic category of the root differs from that of the complex noun. In other words, where we have for instance the structure noun + noun, it is difficult to determine the head of the complex word. This is because the root is repeated either to the left or to the right.
According to Owolabi (1995:106), the device which enables a complex word to inherit the syntactic properties (or feature) of its head is referred to as percolation. Scalise (1984:96) support this definition thus: “if a constituent a is the head of a constituentb, a and b are associated with an identical set of features (syntactic or diacritic)”. Following this definition and the fact that prefixes as heads must be dominated by the category label N, the percolation of the category label N subsumes the head to the entire complex noun in each of the categories.
The motivation for making the prefixes in 16 heads is nominalization, which merely converts verbs into nominal expressions. Further evidence that prefixes can serve as heads of constructions is where they function as subjects of the sentence. This automatically qualifies them as heads of the NPS. Let us examine the following examples:
23(a) É-nyéné úwák íbàn - ‘One with many wives’
SC V Spec NP
One have many wives
(b) Á- dá úkàrà - ‘The head of government’
SC V NP
One stand government
Using the X-bar framework, we can represent 22 (a) and (b) structurally as 23 (a) and (b):
24a) N''
(b) N''
N'
N'
N V''
N V''
É- V' Á- V'
V NP V N
nyéné Spec N dá úkàrà
úwák íbàn
We however wish to admit that not all prefixes in the language can function as heads. Where the root has more than one prefix, especially in agglutinative structures, it is difficult to determine the status of head. Let us consider the following examples:
25 a) É mé sí dí mí -‘Have you been coming here?’
SC Tn Asp V Adv
You pres come here
b) Í kpi sí
i nám ké -‘You (pl) would
not have been doing it’
SC
Mod Asp SC V Neg
You would
do not
The constituents of the
sentences in 24 are mostly affixes, mainly expressing grammatical relationships
such as person, tense, modality, aspect and negation. Only the SC prefix can
function as the head of the NP. Other prefix types cannot function as heads of
syntactic constructions.
Conclusion.
This paper has been concerned with prefixation as one of the most productive word formation processes in Efik. We have taken into account the structure of Efik prefixes, which constitute a syllable of the forms V, CV, CVC, or VCCV and whose phonetic shapes are determined by the SC or root element it co-occurs depending on the person involved. The paper examines the role of the prefix in signaling certain grammatical contrasts involving such categories like tense, negation and number, among others. The role of prefixes in word formation and the functions of prefixes as heads of some morphosyntactic constructions have been examined. A general assumption has been made; that is, apart from the principle of vowel harmony, which is a dominant process, other grammatical processes like nasalization, nominalization, syllabification, and tonal contrast, are important features of the morphology of Efik prefixation, and that stems and prefixes do not occur randomly, but systematically in a rule-governed order.
References
Allerton, David. (1979). Essentials of Grammatical Theory: A Consensus View of Syntax and Morphology. London: Routledge & Kegan Paul.
Anagbogu, Philip (2003) Nominalizing Prefixes of Koring. In Ndimele, Ozo-Mekuri (ed) Four Decades in the Study of Languages and Linguistics in Nigeria: A Festschrift for Kay Williamson. Aba: NINLAN Press. 701-706.
Bussman, Hadumod (1996) Routledge Dictionary of Languages and Lingusitics. London: Routledge
Chomsky Noam and Halle, Morris.(1968). The Sound Pattern of English. New York: Harper and Row.
Comrie, Bernard (1976) An Introduction to the Study of Verbal Aspect and Related Problems. Cambridge: Cambridge University Press.
Essien, Okon E. (1983) The Tense System of Ibibio. In Kaye, I.E (ed.) Current Approaches to African Linguistics. 2 (20) 329-344.
Essien, Okon E. (1990) A Grammar of the Ibibio Language. Ibadan: University Press Ltd.
Faraclass, Nicholas (1989) The Cross River Languages In J. Bendor-Samuel (ed.) The Niger-Congo Languages. University Press of America, 377-394.
Greenberg, Joseph H. (1963) Languages of Africa. The Hague: Mouton.
Katamba, Francis (1993) Morphology. New York: Palgrave.
Langacker, R. (1972). Fundamentals of Linguistic Analysis. New York: Harcourt, Brace, Jovanovich.
Mensah, Eyo O. (2001) Negation in Efik. Kiabara: Port-Harcourt Journal of Humanities 7 (2): 61-67
Mensah, Eyo O. (2002) A Morphological Analysis of Efik Verbs. Ndunode: Calabar Journal of Humanities 1 (3) 94 -105
Mensah, E. O. (2007) Number and Person in Efik. Lingua et Linguistica 1(2), 45- 55.
O’Grady, William M. Dobrovolsky and M. Aronoff (1989). Contemporary Linguistics: An Introduction. New York, St. Martins Press
Owolabi, Kola (1995). More on Yoruba Prefixing Morphology. In Owolabi, K. (ed.) Language in Nigeria: Essay in Honour of Ayo Bamgboe. Ibadan: Group Publishers, 93 – 112.
Radford, Andrew (1999). Syntactic Theory and the Structure of English: A Minimalist Approach. Cambridge: Cambridge University Press
Robins, Robert H. (1989). General Linguistics. London: Longman
Scalise, S. (1984). Generative Morphology. Dordrecht: Foris
Smith, George (1991). Computers and Human Language. Oxford: Oxford University Press.
Spencer, Andrew (1991). Morphological Theories. Cambridge: Blackwell
William, E. (198) On the notion of ‘lexically related’ and
‘head of a word’. L1 12, 245 –
274.