1. Introduction
Reduplication is a morphophonological process that is commonly attested in many African languages. Ejagham happens to be one of them. This language is an Ekoid Bantu, classified under zone 800 following the linguistic classification of African languages (Breton & Fohtung, 1991). According to Eberhard, Simons, & Fenning (2024), the Ejagham language has an estimated population of 10.000 to 1,000,0000 speakers in both Cameroon and Nigeria. In Cameroon, the language is spoken in Manyu Division, South West Region. There are three major dialects; Western Ejagham (WE), Southern Ejagham (SE), and Eastern Ejagham (EE). Although this morphological process is found in every Ejagham dialect, special attention is given to their presence in EE dialect which has an estimated population of over 30,000 speakers.
Reduplication in this language is a productive process that may date back to Proto-Benue-Congo (PBC). Various sub-groups of Benue-Congo display at least residual forms of the process (Watters, 1990). Some strings of a word; a consonant, a vowel, a syllable, two syllables, a stem, a whole word, may be repeated to provide some particular meaning to the word in a process called “reduplication.” The reduplicated word may also have an idiosyncratic meaning (Mutaka & Tamanji, 2000). This paper addresses the problem of reduplication and accounts for the tones on the reduplicated form of adjectives in Ejagham, the aim is not just to show that reduplication is an important resource used to enrich the Ejagham vocabulary, but also to show that it is a regular phenomenon with interesting tonal processes.
It is common to find reduplication in nouns, verbs, adjectives and adverbs of the Ejagham language. The fact that this morphological process is common and occurs in different word classes, makes the discussion vast and varied to the extent that it will be seemingly unfair to handle every aspect of it. From a general look of things, Ejagham exhibits three forms of reduplication processes; Whole-word reduplication whole stem reduplication and partial stem reduplication. However, this paper focuses on the first two forms because in this language the third form is quite often attested in nouns only. Besides, studies (Watters, 1990) show the partial stem reduplication has undergone simplification process both at the segmental and tonal level in EE. Innovative measures warrant research contributions not just on the analysis of whole word and whole stem, but also on another word class.
2. Theory and Method
The general theoretical frame adopted here is Generative in the sense that surface phonetic structures are generated from underlying structures. The theory of generative grammar was introduced by Chomsky (1957) in his Syntactic Structures and cited by Allerton (2017, p. 3). This approach was applied first in syntax whereby Chomsky emphasized on the generation of surface structures of sentences from deep structures using transformational rules. He later applied this to phonology in The sound patterns of English (Chomsky & Halle, 1968).
In this paper, reduplication is analysed autosegmentally (Marantz, 1982), McCarthy and Prince (1986), Mutaka and Hyman (1990). This approach emphasizes the independence of tiers, such that reduplication is analysed as the affixation of a skeletal tier, that is, a tier not defined in the segmental tier, to a base morpheme or word. Although Goldsmith (1990) propounded the autosegmental theory to analyse tones separately from segments, the theory has been expanded to englobe other domains including reduplication. The idea of generative approach here is equally employed very broadly to refer to the generation of surface structures from deeper and more abstract structures.
As a native speaker of the language, I supplied part of the information and proceeded with verification of my own data. Reliable information was gotten from other native speakers. Some secondary sources such as Watters (1990) also helped in gathering information for analysis. A data of 115 reduplication items were collected and classified into tone groups and syllable groups in a bid to get all the possible influences on the reduplicative process in Ejagham. The forms were classified into whole word and whole stem reduplication. In the analysis proper, we shall proceed with the identification of the different morphemes, that is, the bases and the reduplicates, and posit how the reduplication process takes place at the segmental tier and at the tonal tier.
3. Syllable and Word Structure
Words in this language present different syllable types. However, the canonical shape of a syllable is CV (Nkongho, 2019). In all these, adjectives are common with the CV and CVC types.
Examples 11: V à “he/she”
CV sù “wash”
CVC fàŋ “fat/big”
CSVC kwên“carry”
The word structure will be viewed here according to the number of syllables per word and the number/types of morphemes that make up the word; monosyllabic, disyllabic and trisyllabic words.
Examples 2: a. làŋ “touch”
CVC
b. tá.kó “grass”
CV.CV
c. ὲ-là.ɣὲ “wall”
V.CV.CV
d. έ-só.ɣó.rù “orange”
V.CV.CV.CV
In (2) above, the dots indicate the syllable boundaries, the hyphens separate the prefix (which constitutes the syllable) from the stem. From a to d, we have monosyllabic, disyllabic, trisyllabic, and four syllable words respectively. Nouns, adjectives, adverbs and verbs can have any of the number of syllables per word exemplified in (2) above. But pronouns have only monosyllabic and disyllabic words.
4. Adjective Types in Ejagham
An adjective describes a noun or pronoun. In Ejagham the main structure of adjectives is Prefix+Stem. The latter type helps to mark their belonging to specific noun classes.
Examples of Adjectives include;
(3)m-fê “new”
a-ßí “two”
ἑ-zɔ.nέ “foolish”
ἑ-ɲà.ɣê “dark”
ŋ-kὲ.m “lame”
As far as the combination of morphemes to form words is concerned, Ejagham is mainly agglutinating, that is, many words consist of two or more morphemes; root and affixes (Nkongho, 2020). Some nouns and verbs in this language are derived from adjective (reduplication).
Examples: (4) blàn “lazy”
ɔ-blàn “laziness”
ἑ-blàn-έ “to be lazy”
m-blà.m-blàn“one who lazy around”
Many adjectives and adverbs overlap in Ejagham; for example.
(5)n-là mbá“first/ahead”
Looking at the above data, the first observation is that of the presence of prefixes. It is as if we were dealing with nouns since these prefixes are the same as those that mark noun classes. In fact, the prefix is a type of concord or agreement relating the adjective to a noun class. This prefix fuses with the adjective in such a way that when an adjective is uttered in isolation, it is accompanied by a prefix. Nevertheless, whether or not the prefix is there does not affect the meaning of the adjective. ßέ = á-ßέ bέ-ßí m-bí
For example: (6) à-nἑ á-ßέ “two people”
bἑ-dʒù bέ-ßέ “two yams”
n-sí m-bέ “two pots”
Ejagham adjectives use reduplication to do a lot of things. For example
To describe shape là.là “very long”
To describe size lέ.ɣέ.lέ.ɣέ “very tiny/little”
To express abundance à-ßɔ.á-ßɔ “full of children/made up of children only”
To derive adverbial expressions as in á-ßέ.a-ßέ “in pairs”
To lay emphasis tʃἑŋ.tʃἑŋ “quite foolish”
5. Tones
Generally, Ejagham exhibits four types of tones; two level (High and Low) and two contours (rising and falling). All other tones are derived through tonal processes.
Example: (7)L nù.nù “good”
H kpέ.kpέ “all”
LH băm.bȃm “dusty”
HLm-fê.m-fê “quite new”
The tone bearing unit is [+syllabic] segments which include vowels and syllabic nasals; m, n, ŋ, ɲ.
6. Data Presentation and Analysis
As mentioned earlier, there are three types of reduplication in Ejagham; Whole word, whole stem and partial stem reduplication. The first two types are common in nouns, verbs, adjectives and adverbs. The third type occurs mostly in nouns. Research has shown that the third type of reduplication occurs only in the Eastern Ejagham variety and the reduplication is becoming simplified through loss of segments alongside the tones (Watters, 1990, p. 112).
This section shows how reduplication works in Ejagham Adjectives. It is divided into sub-sections with each one devoted to the analysis of a particular reduplication in a given category.
Reduplication will be analysed here, as the addition of a phonemically bare affix to a base morpheme or word, which then triggers the mapping of the base’s phonemic melody to the affix (Marantz, 1982; McCarthy & Prince, 1986) as cited in Vernyuy (1997).
As for the nature of the reduplicative affix, we will define it in terms of what McCarthy and Prince (1986) call “authentic prosodic units”, that is; word, stem, syllable, etc. But where need be, we will specify the C-V skeleton (Marantz, 1982) composing the prosodic Unit.
The analysis in each subsection will consist in seeking answers to the following questions in (8), some of which are adapted from Mutaka and Hyman (1990).
(8) A. What is the meaning of reduplication?
B. What is the nature of the template or reduplicative?
C. Is the template prefixed, infixed or suffixed?
D. What is the melody copied?
E. What is the direction of melody mapping to the template?
F. What are the resulting phonological (segmental) processes?
G. How is the tone assigned in reduplication?
6.1. Whole Word Reduplication
Adjectives have various functions as indicated earlier. All these Adjectives reduplicate to give a sense of abundance which means “very X,” “full of X” or “like X.” where X stands for the base word. Whole word reduplication occurs in mono syllabic, disyllabic and trisyllabic words as exemplified in (9) below.
(9) a. tí “much” tí.tí “very much”
ἑ-là “long” ἑ-là.là “very long”
ἑ-mù “whole” ἑ-mù.mù “very complete”
a-yέp “water” à-yέp.à-yέp “full of water/watery”
n-tʃɔt “problem” n-tʃɔt. n-tʃɔt “full of problems/troublesome”
b. zɔ.nέ “foolish” zɔ.nέ. zɔ.nέ “foolishly”
à-wì.ɣí “hills” à-wì.ɣí.á-wí.ɣí “full of hills/hilly”
ἑ-dʒɔk “noise” ἑ-dʒɔk.ἑ-dʒɔk “full of noise/noisy”
c. káßáyá “short” káßáyá. Káßáyá “many short items”
mɔ-lòɣȏ “a bit” mɔ-lò.ɣô. mɔ-lò.ɣô “gently” (bit by bit)
ἑ-lέmέ “big/mighty ἑ-lέmέ. ἑ-lέmέ “very big/mighty”
In the above examples, the hyphen separates the prefixes from the stems while the dots show the boundary between the base and the reduplicate.
Observing the data still, from the second column in (8)a, b, c, it is not easy to say which is the base adjective and which is the reduplicate. But if we take from Marantz (1982) that reduplication is simply affixation, and we also consider that an adjective in Ejagham is either whole word (prefix + stem), or it is whole stem, then we will have grounds to say that adjective reduplication is simply prefixation and as such it has the structure in (10).
(10) word . word
[reduplicate]. [base]
(Prefix. base)
In (11) below we will give derivation for
i) tí.tí “very much”.
ii) ἑ-dʒɔk. ἑ-dʒɔk “noisy”.
iii) mɔ-lòɣô. mɔ-lòɣô “gently” (bit by bit).
To demonstrate how whole word reduplication takes place in Ejagham adjectives.
(11) We have the base words
a) i W ii) W iii) W Word
σ σ σ σ σ σ Syllable
tí Ɛ-dʒɔk mɔ-lòɣô Segments
| | | | | |
H L L L L HL Tones
“Much” “noise” “very small (progressive)”
We prefix the word reduplicative template to the base.
b) i. W W ii) W W iii) W W
σ σ σ σ σ σ
tí Ɛ-dʒɔk mɔ-lòɣô
| | | | | |
H L L L L HL
The phonemically bare prefix triggers the copying of the base melody to the left. Based on the similarity between the tone in the reduplicate and that in the base, we copy the tone alongside the segments.
c) i. W W ii) W W iii)W W
σ σ σ σ σ σ
tí tí ɛ-dʒɔk Ɛ-dʒɔk mɔ-lòɣô mɔ-lòɣô
|| | | | | || | | | |
H H L L L L L L HL L L HL
We assume that before mapping to the word template the copied material undergoes automatic syllabification after the base as follows:
d) i. W W ii) W W iii)W W
σ σ σ σ σ σ σσσ σ σ σ
tí tí ɛ-dʒɔk Ɛ-dʒɔk mɔ-lòɣô mɔ-lòɣô
| | | | | | | | | | | |
H H L L L L L L HL L L HL
We then map the syllable nodes to the word template from left to right
e) i. W W ii) W W iii)W W
σ σ σ σ σ σ σσσ σ σ σ
tí tí ɛ-dʒɔk Ɛ-dʒɔk mɔ-lòɣô mɔ-lòɣô
| | | | | | | | | | | |
H H L L L L L L HL L L HL
“Very much” “full of noise” (noisy) “very small” (progressive)
6.2. Whole-Stem Reduplication in Adjectives
Whole-stem reduplication in adjectives is noticed in monosyllabic and dissyllabic stems. In some cases, nouns reduplicate to give adjectives whereas in other cases, it marks intensification.
Example: (12) ἑ-nù “well”ἑ-nù.nù “very well”
m-fê “new”m-fê.fê “quite new”
ἑ-mù“whole” ἑ-mù.mù “full” (whole nut)
ŋ-kpἑnear ŋ-kpἑ.kpἑ quite near/close
àyέp “water” àyέp. àyέp“watery
ɲáŋέ “different” ɲáŋέ. ɲáŋέ“quite different”
The derivation of ἑ-nù.nù “quite well” and ɲáŋέ. ɲáŋέ “quite different”
The base stems are:
(13)a) i. σ ii)σ σ
| ||
nù (well) ɲá ŋέ (different)
A syllable template is prefixed
b) σ ii)σ σ
| ||
nù ɲá ŋέ
| ||
L H H
The base melody (tone and segments) is copied to the left
c) i σ σ ii)σ σ σ σ
| | |
ἑ nu nù ɲá ŋέ ɲá ŋέ
| | | | | |
L L H H H H
Mapping of the copied melody to the syllable template is done from left to right
d)σ σ ii)σ σ σσ
|| ||||
nu nù ɲá ŋέ ɲá ŋέ
| | ||||
L L H H H H
Optional prefixation takes place to give
e)σ σ ii)σ σ σσ
|| ||||
ἑ nu nù ɲá ŋέ ɲá ŋέ
| | ||||
L L H H H H
“quite well” “quite different”
To summarise whole word reduplication in Ejagham adjectives, we will answer the analysis questions initially presented in (8) thus:
Example (14)
A) Reduplication means full of X, intensifier of X, where X is the reduplicating word or base.
B) The template is a word.
C) The word template is prefixed.
D) The entire melody of the word is copied.
E) The direction of melody mapping is left to right.
F) No special phonological rule results from reduplication.
G) The tone is copied alongside the segmental melody.
7. Conclusion
The Ejagham language is very rich in reduplication and sometimes very confusing because it is attested in a significant number of word classes. However, the analyses in this paper has been carefully limited to adjectives. There are three forms of “regular” reduplicative processes in the language (where “regular” stands for those reduplications resulting from the affixation of a skeletal morpheme) These processes include: Whole word reduplication, whole stem reduplication and partial stem reduplication. The processes described in whole word and whole stem adjectives are what can be referred to as “normal” reduplication processes in Ejagham Adjectives. “Normal” because these processes are very frequently used in the language and it is easy to identify the reduplicating base. Adjective reduplication does not function derivationally. Rather, they function inflectionally, like whole word reduplication àyέp. àyέp “watery” (àyέp = water). Whole stem reduplicationŋ-kpέ.kpέ “quite near/close” (ŋ-kpὲ = near).
The reduplicative affixes or templates in Ejagham are not just any string of C-Vs, but they can be captured in terms of prosodic units; in whole word reduplication, the template is a word, in whole stem it is either a syllable or two. The templates are prefixed in adjectives. In both forms of reduplication, the tone is copied along with the segments.
NOTES
1The Ejagham examples used in this study are written phonologically rather than orthographically.