Verbal Status and Serial Verb Constructions in Igala

Download Report

Transcript Verbal Status and Serial Verb Constructions in Igala

Verbal Status and Serial Verb
Constructions in Igala
Gideon Sunday Omachonu
[email protected]
Georg Forster Post-Doctoral Fellow (AvH Stiftung/Foundation)
Humboldt-Universität zu Berlin
Institut für Asien- und Afrikawissenschaften
Seminar für Afrikawissenschaften
January 3, 2012.
1
Linguistic Map of Nigeria
2
Language Classification
(Internal Structure of Niger-Congo)
Proto-Niger-Congo
Kordofanian
Proto-Mande-Atlantic-Congo
Atlantic
Mande
Proto-Ijoid- Congo
Ijoid
Proto-Dogon-Congo
Dogon
Ijo
Defaka
Proto-Volta-Congo
North Volta-Congo
South Volta-Congo =Proto-Benue-Kwa
Nyo Left Bank West Ukaan
Benue
Congo
East
Benue
Congo
Oko
Yoruboid Akokoid
Nupoid Idomoid Igboid
Kainji
Ukaan Akpes
Edoid
Cross River Platoid
Yoruba IGALA Itsekiri
Bendi Delta Cross
Bantoid
3
1.1. Introducing the Igala Language
• Ígálà language belongs to the West Benue-Congo and
more precisely one of the ‘Yoruboid’ languages in
Nigeria. It is a dominant language in Kogi state spoken
by over two million natives in nine Local Government
Areas of Kogi East Senatorial District – North Central
Nigeria.
• The language is equally spoken in some communities
outside Kogi state: Èbú in Delta state, Ólóhí & Ìfèkwù in
Edo State, Ógwúrúgwú, Ọ̀jọ́, Ìgá & Àsàbá in Enugu State,
Òdòkpè, Ńjàm, Ìnọ́mà, Àlá, Ìgbédọ̀, Ónúgwá, Òdè,
Ìgbòkènyi & Ìlá in Anambra State (See Omachonu
2011a&b).
• It is a register tone language with 3 basic tonemes
(H,M,L), isolating with agglutinative features, basically
SVO, marks aspect but no grammatical tense, major
word classes are nouns & verbs.
4
1.2.SVCs: Background Information(1)
• Although instances of SVCs (a sequence of verbs
which appears in what seems to be a single clause)
had been reported by Christaller (1875) and
Westermann (1907, 1930) in Twi and Ewe (Kwa,
Niger-Congo) languages in West Africa, the term
‘serial verb construction’ was first introduced by
Balmer & Grant (1929) and then reintroduced by
Stewart (1963).
• Even though a few alternative names for the
phenomenon appear in the literature, ‘serial verb
construction’ and ‘serial verb’ have gained more
general acceptance.
5
1.2. SVCs: Background Information(2)
• Serial verb construction (SVCs) are widely attested syntactic
phenomenon in the languages of West Africa, Creole
languages, languages of Southeast Asia, Amazonia, Oceania,
and New Guinea ( Aikhenvald 2006, Bisang 2009)
• Serial verb constructions (SVCs) are not restricted to
languages of a particular typological profile. They are
widespread, clearly recognizable robust grammatical
constructions found in nearly one-third (⅓) of the languages
of the world (c. f. Dixon 2006).
• However, none in Europe, north or central Asia, and rather
few in north America or Australia’ (Dixon 2006:338).
6
1.3.The Three Major Phases of Research in SVCs(1)
Three Major Phases of the Investigations on SVCS in
Languages (1875-2011):
• Phase 1: 1875 to early 1960s – As part of the pedagogical
issues on the grammar and initial description of some
languages (Christaller 1875, Balmer & Grant 1929,
Westermann 1907, 1930, Ward 1952, Westermann &Bryan
1952, Stewart 1963 and others).
• Phase 2: Late 1960s to the 1990s – Theorizing on the
defining features, syntactic sources of SVCs and application
of relevant theories to the syntactic representation of SVCs
(Boadi 1968, 2000, Bamgbose 1973, 1974, 1982, Awobuluyi
1967, 1973, Schachter 1974, Baker 1989, 2002, Agbedor
7
1994, Yusuf 1997,very many others).
1.3. The Three Major Phases of Research in
SVCs/Sustained Interest in SVCs(2)
• Phase 3: Mid 1990s to the present – Semantics of SVCs,
demarcation between SVCs and other verb sequence
constructions, comparative studies, typological and crosslinguistic investigations of SVCs (Schiller 1990, Lord 1993, Bearth
1999, Baker 2002, Ameka 2005, 2006 Aikhenvald & Dixon (eds.)
2006, Bisang 2009 and very many others).
• A historical account of the studies on SVCs would reveal a period
of 136 years of continued relevance and sustained interest.
• The question is why this sustained interest?
 Interest in serial verb constructions (SVCs) persists because of
the intricacies and the multi-dimensional nature of the issues
surrounding the phenomenon in languages and the crosslinguistic variations identified with such constructions.
8
1.3. The Three Major Phases of Research
in SVCs/Sustained Interest in SVCs(3)
 There are various types of serial verb constructions even in a
single language and there are cross-linguistic variations such
that the properties of SVCs in one language may not map
whole sale onto those of another language (Lord 1993,
Ameka 2005).
 The descriptions of SVCs, in many instances, had appeared
not quite adequate either because of the intricacies of the
constructions or the tendency on the part of the researcher
to address or concentrate on only one particular problem or
a few of such problems in a language or across languages.
Hence, the questions concerning the nature of verb
serialization in languages arise again and again like the
phoenix birds from the flames.
9
1.4.SVCs in Igala: Previous Studies(1)
• In Igala, the description of SVCs appears almost like an
unchartered terrain. The extant literature is very scanty. To
the best of my knowledge, apart from Omachonu (2006)
and Ejeba (2011), one hardly finds other studies in the
literature on Igala linguistics with focus on serial verb
constructions.
• Omachonu (2006): Verb Phrase Serialization in Igala:
 A peculiar VP with unique characteristics such as only one
expressed subject, shared NP object and aspect, no
intervening coordinating and subordinating conjunctions.
 Identified ten types/functions of SVCs based on only
semantic considerations.
10
1.4. SVCs in Igala: Previous Studies(2)
 Serial verb constructions in Igala derive from single sentence
in the D-structure (a kind of subordination)
 In most cases; it is verbs of action (mainly transitive) that
are serialized in the language.
 That the productiveness of verb serialization in Igala,
correlates with the fewness/scarcity of conjunctions and
prepositions in the language.
• Ejeba (2011): Posits that there are two ways in which verbs
in an SVC stand out from other verb forms; in structural
complexity and in the functional expression of a composite
notion. This marks out the serial verb as a unique exception
to the requirement that a finite clause must have one and
only one finite verb as the head of VP.
 Identified two types of SVCs in Igala based on their
structures namely; concomitant SVCs and coordinate SVCs
and gave six functions of SVCs in the language.
11
1.5. The Present Study:Goal & Structure
• Aim/Goal:
• Following up on the previous studies, the goal of the present
study is to give a more comprehensive description of SVCs in
Igala that goes beyond earlier studies of the phenomenon in
the language both in scope and depth of analysis.
• The Structure of this Presentation:
 section 2. Description of the phonology and morphosyntax
of verbs in Igala to determine their verbal status.
 Section 3. SVCs in Igala: (3.1) definition, (3.2) functions,
(3.3) types and(3.4) derivations.
 Section 4 attempts a distinction between SVCs and other
multi-verb constructions in Igala.
 Section 5 concludes the paper.
12
2. The Phonology and Morphosyntax of Verbs in Igala :
Determining the Verbal Status of Words (1)
• Most Igala verbs are monosyllabic in structure and many of the
disyllabic or polysyllabic verbs are derived through incorporation
and amalgamation.
(1a)bí ‘to give birth’ pú ‘to bend’
tà
‘to sell’
dó ‘to call’
kó ‘to write’
nọ̀
‘to grind’
Kó ‘to carry’
kpá ‘to kill’
gbẹ̀ ‘to sow’
mó ‘to drink’
nyí ‘to wear’
gọ́
‘to swear’
hì ‘to cook’
chú ‘to defecate’ jẹ
‘to eat’
(b) rúlé ‘to run’ (c) kpábíe ‘to destroy’
néjú ‘to expect’ kàjí
‘to block’
rákwú ‘to cry’
dáté̩ ‘consecrate’
Note: Tone in Igala is marked as follows: v́ for High, v̀ for Low but
Mid tone is left unmarked.
13
2. Determining the Verbal Status of Words(2)
• Apart from the phonology as seen above, determining
the verbal status of words in syntactic constructions
including SVCs, depends on three major criteria namely;
 Function : ability of the word to occur in a simplex
sentence.
 Meaning: aspects of the meaning of verb characteristic
of the usual semantic categories such as action, process,
state, attribute, experience, and so on.
 Morphology of the word. the ability of a word to take
tense and/or aspect (Potsdam, 1997).
Taking into account the following criteria: phonology,
meaning, function and morphology, the words labeled
as verbs (bold print) in (2), (3) and (4) are definitely
verbs in Igala.
14
2. Determining the Verbal Status of Words(3)
2. Áudù a là ọjẹ jẹ
Audu ASP buy food eat
‘Audu buys/is buying food to eat’
3. Àyè du ẹja wa tà
Aye carry fish come sell
‘Aye brought fish to sell’
4. Idah nána
tù
Ugwo̩lawo lè (Omachonu 2006)
Idah be: big surpass Ugwo̩lawo go
‘Idah is bigger/larger than Ugwolawo’
15
2.Determining the Verbal Status of Words(4)
• A related aspect of verbal meaning that is necessary to draw
attention to is the contrast between the meaning of the verb
in a simple sentence and its meaning in an SVC. For instance,
in (4), the verb lè ‘go’ expresses, in that context, a meaning
usually associated with adverbials than its primary meaning
as a verb. Let’s also consider the meaning of du and bié̩ne̩ in
(5) and (6).
5. Du únyi lẹ
ngà. (idiomatic)
carry house DET show ‘Show the house’
6. I
fú=u
gwó̩ bié̩ne̩ (adverb of manner, Lit. ‘irony’)
3SG MOD=3SG carve be.bad
‘He/She has carved it badly (lit., carved perfectly).
16
2.Determining the Verbal Status of Words(5)
• Irrespective of whatever secondary function a verb is made to
serve or the meaning attached to it, the most reliable indicator
of its verbness or verbal status in SVC is its morphology (the
ability to take tense and/or aspect) (c.f. Bamgbose, 1982: 7).
Observe the Akan examples below.
7. Akan (a) Kofi yɛɛ adwuma maa Amma
Kofi did work gave Amma
‘Kofi worked for Amma’
(b) Kofi gyee Amma dii
Kofi receive Amma ate
‘Kofi believed Amma’
• However, in Igala, where tense/aspect are not so overtly
coded, verb morphology is of limited use in deciding the
verbness of a word, hence we rely on the 4 criteria discussed
earlier: meaning, function, phonology & morphology.
17
3.1.SVCs in Igala: The Defining Features(1)
• The recognition of SVCs in most of the languages in which
they occur is based on a combination of formal and semantic
properties or criteria. Aikhenvald (2006:1) integrates a
considerable number of these properties into her definition
of SVC thus:
A serial verb construction (SVC) is a sequence of verbs which
act together as a single predicate, without any overt marker
of coordination, subordination, or syntactic dependency of
any sort. Serial verb constructions describe what is
conceptualized as a single event. They are monoclausal; their
intonational properties are the same as those of a
monoverbal clause, and they have just one tense, aspect,
and polarity value. SVCs may also share core and other
arguments. Each component of an SVC must be able to occur
on its own. Within an SVC, the individual verbs may have
same or different transitivity values.
18
3.1.SVCs in Igala: The Defining Features(2)
• SVCs in Igala are characterized by:
 Two or more verbs acting together as a single predicate
and marking a single event within a temporal frame.
 Same syntactic subject for all VPs but expressed only
once before VP1.
 The verbs occur in sequence without an intervening
conjunctions (subordination and/or coordination) .
 SVCs are monoclausal (allow no markers of syntactic
dependency on their components)
 They display intonational properties that equate with
those of a monoverbal clause in that there are no
intonation breaks and pause markers between the
components.
19
3.1.SVCs in Igala: The Defining Features(3)
• The following sentences illustrate frequent instances of
SVCs of the above descriptions in Igala (Example 2
represented as (8), and example 9) .
8.
Áudù a là ọjẹ jẹ
Audu ASP buy food eat
‘Audu buys/is buying food to eat’
9.
Ì
che
ómi gwẹ̀ éju
3SG scoop water wash eju
‘He/She fetched water to wash his/her face’
20
3.1.SVCs in Igala: The Defining Features(4)
• SVC cannot take separate markers of syntactic dependency.
If an SVC is embedded, its components cannot be
embedded independently. For instance, in (10) and (11), the
clause makers or indicators, ki and kà only mark
dependency once each in the SVCs (enclosed in squares
brackets).
10. Ì
dọ má [ki =i
du
únyi lẹ
3SG call 3PL COMP=3SGSCL.IND
carry house DET
ngà má.]
show 3PL
‘He called them to show them the house’
11. Ùmà
ọ̀tákáda che eu
̣ n [kà=à
nyi
Knowledge book
be thing REL=1PlSCL.IND go
á
du wa.]
ASP carry come
‘Book knowledge is what we went to bring’
21
3.1.SVCs in Igala: The Defining Features(5)
• SVC codes or describes a single event, which means
that verbs in series must code aspects of an action or
state that can be associated together.
12.(a)i. du wa
(b). i. Du ọ̀tákáda lẹ wa.
carry come ‘(to) bring’
carry book DET come
‘Bring the book’
ii.du
ngà
carry show ‘(to) show’
ii. Du únyi lẹ
ngà.
carry house DET show
‘Show the house’
iii. bò
má
cover tight ‘(to) cover’
iii. Bò ùchà lẹ má.
cover pot DET tight
‘Cover the pot’
22
3.1.SVCs in Igala: The Defining Features (6)
• SVCs in Igala share arguments in the forms of both
NP subject and object sharing. They also share
grammatical categories such as aspect and mood
except negation which applies to the entire SVC and
occurs in sentence final position( Ex. 8 as 13).
13. Áudù a là ọjẹ jẹ.
Audu ASP buy food eat
‘Audu buys/is buying food to eat’
14. Áudù a là ọjẹ jẹ n.
Audu ASP buy food eat NEG
‘Audu does not buy food to eat’
23
3.1.SVCs in Igala: The Defining Features(7)
• If these features; subject, object, aspect and negation
are repeated in the SVCs, some ungrammaticalities
and/or semantically bizzare expressions will be
generated as seen in (15. a, b and c):
15.a. *Áudù a là ọjẹ a jẹ ọjẹ
b. *Áudù a là ọjẹ Audu jẹ (Two different persons, may
be, but still ill-formed)
c.*Áudù a là n ọjẹ jẹ n
* Ill –formed structures.
24
3.1.SVCs in Igala: The Defining Features(8)
• In Igala, only a complete SVC can be questioned
and interestingly, even the response to such a
question cannot employ only one of the component
verbs but the argument (shared object) remains
constant as in (16. a & b) below.
16.a. Question: Eun Áudù là jẹ?
what Audu buy eat
‘What did Audu buy and eat?’
b. Response: Áudù là ọjẹ jẹ.
Audu buy food eat
‘Audu bought food and ate’
25
3.1.SVCs in Igala: The Defining Features(9)
• Lastly, most SVCs in Igala satisfy the requirement that
each component of an SVC must be able to occur on its
own with the same or different transitivity values. Hence,
we can derive two simple sentences from (16b) as (17) and
(18)respectively, and with the same transitivity values but
not as SVCs any longer.
17.
Áudù a là ọje.̣
Audu ASP buy food ‘Audu buys/is buying food’
18.
Áudù a
jẹ ọjẹ.
Audu ASP eat food ‘Audu eats or is eating food’
• The beauty of SVC is that these two separate actions,
when/where necessary, could be collapsed into sub-events
under an overall single event of buying the food to eat it as
in (16b).
26
3.2.Functions of SVCs in Igala (1)
i. Accompaniment(Comitative?): Accompaniment SVC has the
meaning of ‘take along with’ where the subject NP of the
initial verb takes the second NP (object) to some location.
19. Àda ko
ẹdà nẹ
wa.
Ada carry.PL shoe carry come
‘Ada brought shoes along’
This first sense of accompaniment differs slightly from another
sense of it in SVC in Igala which, strictly speaking, is
christened ‘comitative’. This expresses the notion ‘go/come
along with’ instead of ‘take along with’.
20. Ẹ̀jẹ̀ lèbọ Ónu nyi Ánkpa.
Eje go.with Onu go Ankpa
‘Eje accompanies Onu to Ankpa.’
27
3.2.Functions of SVCs in Igala (2)
ii. Benefactive: This is an instance whereby the oblique object
benefits from the action carried out by the subject of an
SVC.
21. Ákwù là ẹdà che ọyà un
ele.
Akwu buy shoe do wife 3SGPOSS gift
‘Akwu bought shoes as gift for his wife’.
iii. Cause - effect: V1 expresses the cause and the V2, the
effect.
22. Ádẹjọ fú=ú
gwo kpa. (gwo kwu)
Adejo MOD=3SG beat kill
‘Adejo beat and killed it (i.e. Adejo beat it and it died)’
28
3.2.Functions of SVCs in Igala (3)
iv. Causative: In Igala, the difference between causative
SVCs and cause-effect is not so clear-cut because of
the semantic overlap between them. However, in terms
of composition, causative SVCs appear as
asymmetrical constructions whereas cause-effect SVCs
are mostly symmetrical. Besides, causative SVCs in
Igala employs clearly causative verbs such as du
‘take/carry’ or je̩ to mean, ‘let’, ‘allow’ ‘make’ and ‘order’
respectively depending on the context. Example (23)
illustrates this.
23.
Òun je̩ wa che ò̩̩bata
3SG agree us do suffer
‘He/She caused/made us to suffer’
29
3.2.Functions of SVCs in Igala(4)
v. Comparative: In comparative SVC, two NPs are
compared to determine which of them has more or less
of certain attributes than the other.
24. Ídáh nána
tù
Ùgwo̩láwo lè. (Omachonu 2006)
Idah be. big surpass Ugwolawo go
‘Idah is bigger/larger than Ugwolawo’
vi. Direction or Motion: Such an SVC includes a verb
which specifies motion that precedes an action or
describes a direction in which the action occurs. This is
mostly ‘deitic’ with the minor verb as a typical verb of
motion or direction.
25. Mẹ lè ẹ́rẹ̀ wa.
2PL walk leg come
‘(You PL) come on foot.’
30
3.2.Functions of SVCs in Igala(5)
vii. Instrumental: This kind of SVC is characterized by an
instance where the object of the first verb is different
from that of the second which is the actual recipient of
the action of the subject NP.
26. Ákáchi du ọ̀kpọ kpa ẹ́fà.
Akachi carry boxing kill elephant
‘Akachi killed an elephant with boxing’
viii. Manner SVC: The second verb performs adverbial function
with respect to the manner in which the action of the first
verb is performed. Sometimes, manner SVCs in Igala are
used as intensifiers to express adverbs of degree by
repeating the first verb. Examples (27.a&b) and (28.a&b)
below illustrate the first and the second explanations
respectively.
31
3.2.Functions of SVCs in Igala (6)
27.a.
I
fú=u gwó̩ kpábié̩.
3SG MOD=3SG carve destroy
‘He/She has carved it badly/destroyed it due to the manner in which
it was carved’.
b. I
fú=u
gwè̩ gbó.
3SG MOD=3SG wash tear
‘He/She/it has washed and torn it due to the manner in which it was
washed’
28.a. I
fú=u
gwo a gwo kpa (kwu)
3SG MOD=3SG beat ASP beat kill
‘He/She has beaten him/she/it mercilessly’.
b. I
fú=u
kpa a kpa kwu (kpa)
3SG MOD=3SG beat ASP kill die
‘He/She has beaten him/she/it mercilessly’.
32
3.2.Functions of SVCs in Igala (7)
ix. Sequence of actions: SVC may describe two consecutive
aspects of an event, not as series of clauses or actions but
as a single overall event.
29.
Áudù a là ọjẹ jẹ.
Audu ASP buy food eat
‘Audu buys/is buying food to eat’
x. Expression of simultaneous actions: Compatible actions
that can be associated together to show that the actions
depicted by V1 and V2 occur at the same time.
30.a. Àpẹ dàchi
a
lẹ
ólu.
Ape lie.down ASP sleep sleep
‘Ape lay down sleeping’
b. Áudù dágo a
jẹ ọ̀jè.̣
Audu stand ASP eat food
33
‘Audu stood eating’
3.3.Types of SVCs in Igala (1)
• SVCs in languages have been classified based on two
major criteria; semantics and formal properties of SVCs.
The preceding section has taken care of the semantic
classification of SVCs in Igala. In classifying SVCs in Igala
based on formal features, the following parameters come
to play (We take the explanations in ascending order):
 composition,
 contiguity,
 wordhood of components and
 marking of grammatical categories
34
3.3.Types of SVCs in Igala (2)
• Marking of grammatical categories: Igala has mainly ‘single
marking’ SVCs. Even though the language does not take
inflection for tense, aspect, negation are, in most cases,
marked just once per construction.
31. Áudù a là ọjẹ jẹ n.
Audu ASP buy food eat NEG
‘Audu does not buy food to eat’
32. Ádẹjọ fú=ú
gwo kpa.
Adejo MOD=3SG beat kill
‘Adejo beat and killed it (i.e. Adejo beat it and it died)’
• Double/concordant marking of any verbal category would
connote two separate events or bring about ungrammaticality.
Whereas it is not possible to duplicate the negation, double
marking of aspect , for instance, will result in some coordinate
construction with different temporal frames/meanings as in (33).
33. Áudù a là ọjẹ a
jẹ.
Audu ASP buy food ASP eat
‘Audu buys and eats/is buying and eating food ’
35
3.3.Types of SVCs in Igala (3)
• Contiguity: SVCs in Igala exhibit both contiguous and noncontiguous relationships as well as instances of V NP V NP and
multi-verb SVCs. Examples (27) and (26) represented as (34 &
35) and example (36.a&b) illustrate these features.
34.
I
fú=u
gwó̩ kpábié̩
3SG MOD=3SG carve destroy
‘He/She has carved it badly/destroyed it due to the manner in
which it was carved’.
35.
Ákáchi du ọ̀kpọ kpa ẹ́fà.
Akachi carry boxing kill elephant
‘Akachi killed an elephant with boxing’
36
3.3.Types of SVCs in Igala (4)
36.a. Ákáchi kpa ẹ́la hì je̩
b. Àyè du ej̣ a wa tà
Akachi kill meat cook eat
Aye carry fish come sell
‘Akachi killed an animal, cooked
‘Aye brought fish to sell’
and ate it’
• Wordhood of components: Individual components of the
SVCs have their own status as individual words capable of
occurrence in simple sentences, e.g. du…wa ‘bring’,
du…ngà ‘show’ can occur in other contexts thus:
37.a. Du
eéwẹ
b. Íye
un wa ọnálẹ
carry you.POSS
mother 3SG come yesterday
‘Take/carry yours’
‘His/her mother came yesterday’
c. Ì
ngà má ọnà
3SG show 3PL road
‘He/she led/showed them the way’
37
3.3.Types of SVCs in Igala (5)
• Composition: two broad groups, symmetrical and
asymmetrical SVCs.
 Symmetrical: usually consist of two or more verbs each taken
from a grammatically and semantically open classes: the order
of occurrence of the component verbs is, in most cases, iconic
to reflect the temporal sequence of sub-events described in
the construction and the verbs in such an SVC have equal
status: Sequential SVCs (ex.29), cause-effect (ex.22), manner
(ex.27,28).
 Besides, symmetrical SVCs in Igala divide into three broad
groups based on the status and the semantic reading of the
components: concomitant (du…wa, du…nga used jointly to
express a concept), coordinate (verbs of equal status with
additive value, la…je̩, kpa…lo̩) and modifying SVCs (gwo kpa).
38
3.3.Types of SVCs in Igala (6)
 Asymmetrical SVCs must necessarily include a verb from a
grammatically and semantically closed class or a verb from
an open class which switches function to occupy a minor
verb slot: Direction/motion SVCs ,causative, benefactive,
instrumental, accompaniment/comitative, etc.
 An asymmetrical SVC is said to possess a head which
usually is the main verb. With regard to the position of
head in asymmetrical SVCs in Igala, there are two
possibilities depending on the composition.
- Head First : Left Hand Head Rule (LHHR)
 Direction/orientation: (Ex. 25) Mẹ lè ẹ́rẹ̀ wa.
 Comitative: (Ex.20): Ẹ̀jẹ̀ lèbọ Ónu nyi Ánkpa.
39
3.3.Types of SVCs in Igala (7)
 A more clearly productive set of examples for SVCs that maintain
head first (LHHR) consistently like the endocentric nominal
compounds in Igala include SVCs which contain a second verb from a
set of verbs which equally function as prepositions in Igala, e.g. kwo,
koji, tu, me̩ru and so on.
38.a. I
ra kwô aja
b. I
kwô aja
3SG run leave market
3SG leave market
‘He/She ran (away) from the market’ ‘He/She has left the market’
39.a. I che koji
ma
b. Oun koji
ma
3SG do replace 3PL
3SG replace 3PL
‘He/She did (it) on their behalf’
‘He represents them’
40.a. Nà̩
a lo tu unyi
b. U tu
unyi mè̩̩
1SG ASP go reach house 1SG reach house already’
‘I am going home’
‘I have arrived home already’
40
3.3.Types of SVCs in Igala (8)
- Head Last: Right Hand Head Rule (RHHR)
e.g. Causative: (Ex.23 ) Òun je̩ wa che o̩b
̩̀ ata
• A set of more productive examples of SVCs that maintain
head last (RHHR) in Igala apart from the use of je̩ as in 23
above is the use of du ‘take/carry´ as in 41-43.
41.
I du wawè̩.
3SG take liken
‘He/She used it as an analogy’
42.
I du che ù̩jì̩we.
3SG take do example
‘He/She used it as an example’
43.
I du kpà̩ilò̩̩.
3SG take exchange
‘He/She used it as an exchange’
41
3.4.Derivations of SVCs (1)
• The insight provided by Chomsky’s Transformational
Generative Grammar (TGG) provoked a great deal of
controversy on the origin(s) or syntactic sources of verb
serialization (Chomsky 1957, 1965, 1981,1993,1995). Whereas
some believe that SVCs (especially coordinate and modifying
SVCs) derive from two different underlying structures, others
hold that SVCs derive from single underlying structure. Thus,
two propositions emerged from the controversy.
 That SVCs are derived from multiple sentences in the base
(underlying) structure.
 That SVCs are derived from a single sentence in the
underlying structure.
42
3.4.Derivations of SVCs (2)
• Awobuluyi (1967, 1973) proposed a derivation of serial
verb constructions in Yoruba from coordinate sources.
• Bamgbose (1973, 1974): Two sources: coordinating and
embedding structures.
• Boadi 1968 and Williams 1971: Verb serialization in Twi
emanates from both coordinating and embedding sources
and in Freetown Krio, SVCs were said to subscribe to
embedding structures only.
• Schachter (1974): SVCs derive from single underlying
sentences hence he propounded single sentence rule for
serial verb constructions. He argued that serial verb
constructions behave, to a large extent, like idiomatic
expressions whose meaning(s) do not depend only on the
composition of the words in the expressions.
43
3.4.Derivations of SVCs (3)
• Yusuf (1997:20) in agreement with Schachter (1974) proposed
a rewrite PSR for serial verbs as: VP → VP (VPn), where the
superscript (n) means we may have any number of verb
phrases. In this schema, even a bare verb is coded as verb
phrase (VP) following the dictates of the projection principle
which holds that every lexical verb has the potential to project
maximally to at least the phrasal level.
S
Fig.1
NP
VP (1)
VP
VPn
Note:Thoroughout this period, the most favoured analysis on the
structure of SVCs is that that sees it as a surbordination
structure in which V2 is embedded in V1 (c.f.Johnson 2005).
44
3.4.Derivations of SVCs (4)
• In all, it is to be noted (as argued earlier) that all the
arguments or previous attempts at deriving SVCs from
underlying complex structures or sentences were influenced
by the theories of the time; TGG (with its later modifications).
• However successful some of these previous attempts might
be, the argument for the monoclausal analysis of SVCs mooted
in Foley & Oslon (1985) and consolidated by successive and
current research on SVCs seems to have put to rest, the
controversy on the derivations of SVCs (c.f. Aikkhenvald 2006,
Bisang 2009, Pavey 2010).
• In what follows therefore, we shall adopt Pavey’s (2010) model
for analyzing complex sentence structures including SVCs to
45
derive the structure(s) of SVCs in Igala.
3.4.Derivations of SVCs (5)
• Pavey (2010) gave an elaborate account of SVCs using the
argument for monoclausal analysis. According to him, simple
sentences contain one clause which contains one core, which
in turn contains one nucleus. Each nucleus therefore
corresponds to one semantic predicate as (Fig 2) below shows
(Adopted from Pavey 2010:220).
Fig 2.
SENTENCE
CLAUSE
CORE
NP NUC
NP
PRED
Olivia heard the thunder
46
3.4.Derivations of SVCs (6)
• A juncture refers to the level at which the linkage occurs and not
necessarily the type of units that are joined.
• Three levels of connections/junctures: clause, core and nucleus
 Clause: Where clauses combine; clause juncture constructions
contain two or more independent clauses which all have their own
arguments usually linked by clause linkage makers for
subordination, coordination or cosubordination.
 Core: The core juncture allows two cores to share one argument
which means that an argument occurs in the semantic
representation of both predicates but appears only once in the
syntax. For instance, in a sentence such as ‘Gideon asked David to
write the letter’, David is the shared argument; semantically, he is
both the undergoer of the verb ‘ask’ and equally the actor of the
verb ‘write’ but ‘David’ appears only once in the syntax.
47
3.4.Derivations of SVCs (7)
 Nucleus: the nuclear level represents a third level of
connection where ‘in sentences with more than one
nucleus, the nuclei combine to form a nuclear juncture
with one set of arguments expressing a single, complex
event’. In a sentence such as ‘Kerry [pushed]NUC
[open]NUC the door, Pavey explains that, ‘The arguments
Kerry and the door are syntactic arguments of the
complex nucleus pushed open’(p.222).
• Figures (3. i, ii & iii) represent the three levels of
connection in complex constructions (Culled from Pavey
2010:223).
Fig 3.i. SENTENCE
ii. CLAUSE
iii. CORE
CLAUSE CLAUSE
CORE
CORE
NUC NUC
48
3.4.Derivations of SVCs (8)
• Essentially, SVCs are complex structures in which two or more
verbs are used to express one complex event. Whereas this
description excludes sentences that are biclausal and clause
juncture constructions as in (Fig.3.i), some SVCs are core
junctures while some are nuclear junctures as in (Fig.3.ii) and
(Fig.3.iii) above respectively.
• The differences between the core level juncture and the nuclear
level juncture in SVCs, Pavey explains:
 First, at core juncture the argument (object NP) appears
between the predicates. Second, there is only one argument
(object NP) shared by the two or more predicates which is the
undergoer of both predicates but the subject NP performing the
action is not an argument to the second predicate.
 But with the nuclear juncture SVCs, though the object NP
argument is shared by both predicates, it is not allowed to come
49
in-between them.
3.4.Derivations of SVCs (9)
• The core and nuclear level junctures of this model can
be used to explain the derivations of some SVCs in Igala
thus (Fig 4 & Fig 5 represent core & nuclear levels).
44.Áudù là ọjẹ je̩.
45. Ádẹjọ fú=ú
gwo kpa.
Audu buy food eat
Adejo MOD=3SG beat kill
‘Audu bought food to eat’
‘Adejo killed it (i.e. Adejo
beat it and it died)’
Fig. 4
Fig. 5
CLAUSE
CLAUSE
CORE
CORE
CORE
NUC
NUC
Áudù là ọjẹ
je
NUC
Ádẹjọ fú=ú gwo
NUC
kpa
50
3.4.Derivations of SVCs (10)
• In (44) represented as (Fig. 4), the argument (object NP) ọjè̩̩ ̣
‘food’ which occurs between the two predicates là ‘buy’ and je̩
‘eat’ is the shared argument which is equally the undergoer of
both predicates but the slight difference between this Igala
example and Pavey (2010) explanation is that the subject NP
performing the action(s) is an argument to both the first and
the second predicate because Áudù does both the buying and
the eating. Similarly, in (45) represented as (Fig. 5), the
argument (object NP, 3SG) is not allowed to occur between
the predicates but it is the shared argument. Unlike the
separate actions of ‘buying’ and ‘eating’ in (44), even though
the subject NP performing the action is an argument to both
predicates, gwo ‘beat’ and kpa ‘kill’, the two are not to be
interpreted as separate actions because it is the beating that
leads to the killing (death).
51
3.4.Derivations of SVCs (11)
• However, if we replace kpa with another word kwu ‘die’ to
have ‘Ádẹjọ fú=ú gwo kwu’ , the subject NP performing the
action would cease to be an argument of the second
predicate because of its intransitive nature.
• Also in Igala, there are instances where in SVCs, each of
the verbs selects its own argument. In this case, the verbs
only share one subject NP argument but have different NP
objects as arguments. Example (46) illustrates this.
46. Ì
che ómi gwẹ̀ éju
3SG do water wash eju
‘He/She washed his/her face with water’
• In (46), the predicates, che and gwẹ̀ have different NP
object arguments, ómi and éju respectively as different
undergoers. The only shared argument is the NP subject
(3SG) performing the actions of carrying the water to wash
the face.
52
3.4.Derivations of SVCs (12)
• It is important also to note that the argument sharing of
symmetrical SVCs, for instance, is somewhat similar to that of
splitting verbs in Igala where such verbs get split into two and the
NP object comes in-between the fragments. Observe (47.a, b, c)
below.
47. a. kọda ‘to cut’
→ Kọ óli dá ‘cut the tree/stick’
b. kpanà ‘to break/split’
→ Kpa íji nà ‘split the firewood’
c. gwújo ‘to burn’
→ Gwu égbe jo ‘burn the grass’
• The two fragments of each of the verbs are interpreted together as
single lexical verbs in this context even though some of the
fragments are capable of generating independent meanings in
other contexts. For example, kpa could meaning ‘kill’ and nà ‘either
to branch off or flay’ (c.f. amalgamated or lexicalized compound
verbs in 1c).
53
3.4.Derivations of SVCs (13)
• The slight difference(s) noticed between Pavey’s (2010)
analysis and the analysis of data on SVCs in Igala are mere
superficial variations (language specific) consequent upon the
nature of the verbs involved in the construction.
• Pavey used data from Taba and Tukang Besi, MalayoPolynesian languages spoken in Indonesia.
• He acknowledges these kinds of cross-linguistic variations
when he posits that ‘we may sometimes find nuclear SVCs
that also allow elements to appear between them’.
• This makes the distinction between the core-level juncture
and nuclear-level juncture SVCs a bit difficult in some
languages.
• All said, the position to hold is that SVCs involve both corelevel juncture and nuclear-level juncture in most languages
including Igala as we have seen from the foregoing analysis.
54
4. Demarcation between SVCs and Other
Multi-Verb Constructions (1)
• The question of how SVCs differ from other multi-verb
constructions as well as other types of complex predicates
such as converbal complex predicates and verb
compounds, ‘remains one of the outstanding questions in
both formal and typological studies’ (Shibatani 2009:255).
• Even though uncontroversial cases of serial verbs are
found in Kwa languages and in the Benue-Congo family of
languages to which Igala belongs, there is most of the time
no obvious distinction between serial verbs and verb
sequences in which each verb constitutes a distinct
predicate. The result is that any sequence of verbs that
does not show obvious coordination or subordination is
loosely termed ‘serial verbs’.
55
4. Demarcation between SVCs and Other
Multi-Verb Constructions (2)
• Distinguishing SVCs from other multi-verb constructions
or complex predicates in general has been somewhat
difficult. As Ameka (2005:2) puts it:
Apart from various verb sequence constructions not
showing surface distinctions leading to some
constructions being mistaken for others, progress in the
analysis and typology of multi verb sequences has also
been hampered by the fact that the different types of
multi verb constructions are functionally similar. In
addition, on the basis of their function they have been
assimilated to constructions in which two or more
elements jointly constitute the predicate of a single
clause. Such constructions have been labeled complex
predicates.
56
4. Demarcation between SVCs and Other
Multi-Verb Constructions (3)
• In what follows however, I shall attempt to make a
distinction between SVCs and only two other verb
sequence constructions: consecutive constructions and
overlapping constructions/covert coordination. By this
consideration, I concentrate on only those multi-verb
constructions (including SVCs) which exhibit the following
features (c.f. Ameka 2005:2):
 With or without any marker of syntactic dependency
 Typically, at least one argument is common to all the verbs
in a sequence.
 The VPs in the sequence are seen as related.
 The individual verbs can function as independent verbs in
simple clauses.
57
4. Demarcation between SVCs and Other
Multi-Verb Constructions (4)
• Consecutive constructions: Two or more verbal clauses
which together represent related state of affairs that can be
interpreted as successive, simultaneous or alternating in
time. The different components exhibit properties of
independent clauses either just juxtaposed or linked by a
connector (See also Ameka 2005, 2006). Consecutive
construction can even be multi-clausal as in (48c)
48.a.
Mẹ gwẹ ọwọ ki =a
jẹ e̩un.
2PL wash hand COMP=1PLSCL.IND eat thing
‘You (PL) wash your hands (and) let’s eat’
b.
Mẹ hiéjúde ki =a
che àdùwa.
2PL close.eye COMP=1PLSCL.IND do prayer
‘You (PL) close your eyes (and) let’s pray’
c.
Mẹ ló, nà á wa
ki =a
jẹ e̩un.
2PL go 1SG ASP come COMP=1PLSCL.IND eat thing
‘You (PL) go, I will come/am coming (so) that we eat’
58
4. Demarcation between SVCs and Other
Multi-Verb Constructions (5)
• In (48a) and (48b), the referents of the subject of the first clauses are
included in the subjects of the second clauses. The consecutive connector
can be omitted but with a compulsory intonation break indicated with a
comma as in (48c). The individual clauses can be marked for different
aspect values as in (48c) where the VP in the second clause is marked for
the future or progressive. In (48c), the referents of the subject arguments of
the clauses are not the same. Each component of the consecutive
constructions in (48a-c) can be independently negated. Example (48c)
represented as (49.a.) and (49.b.) can illustrate the last point.
49.a.Mẹ ló n,
nà á wa ki =a
jẹ e̩un.
2PL go NEG 1SG ASP come COMP=1PLSCL.IND eat thing
‘You (PL) don’t go, I will come/am coming (so) that we eat’
b.
Mẹ ló, nà á
wa n
2PL go 1SG ASP come NEG
‘You (PL) go, I will not come’
59
4. Demarcation between SVCs and Other
Multi-Verb Constructions (6)
• Overlapping Constructions / Covert Coordination:
Overlapping constructions thought of as a kind of covert
coordination since they cannot have any overt linker. Equally, a
prosodic break is compulsory between the VP constituents (the
two clauses). Even though overlapping constructions tend to be
biclausal, there is no overt linker between the components but
the subject argument of each clause must be obligatorily
expressed. The object NP overlaps in both clauses.
50.a. Òun là ọtẹ, a
mọ.
3SG buy wine 1PL drink
‘He/She bought wine and we drank’
b.
Ma gwo ún, ì wọ ún.
3PL beat 3SG it pain 3SG
‘They beat him and it pained him’
60
4. Demarcation between SVCs and Other
Multi-Verb Constructions (7)
50.c. I
là ọjẹ̀,
I
fú=u
jẹ.
3SG buy food 3SG MOD=3SG eat
‘He/She bought food and ate it’
• Notice that in (50.c), ọ̀jẹ̀ ‘food’ is co-referenced
with a referentially identical object u ‘3SG’ as an
anaphoric pronoun. This is an important
difference between the covert VP coordination
structures and SVCs because in an overlapping
construction or covert coordination structure, if the
VPs have the same referentially identical object,
its subsequent occurrence is signaled by an
anaphoric pronoun (pronominal clitic).
61
4. Demarcation between SVCs and Other
Multi-Verb Constructions (8)
• In addition, unlike consecutive constructions, there
is no overt linker between the components of
overlapping constructions or covert coordination
structures. But like the consecutive constructions,
each verb can take different aspect, mood or
modal marking and each can likewise be negated
independently. Example (50. b) above is
rephrased as ( 51) to account for separate
marking of aspect and negation.
51.
Ma a gwo ún, ì a wọ ún n.
3PL ASP beat 3SG it ASP pain 3SG NEG
‘They are beating him (but) it doesn’t pain him’
62
4. Demarcation between SVCs and Other
Multi-Verb Constructions (9)
• The analysis of the demarcation between SVCs, consecutive
constructions and overlapping constructions in Igala could be
summarized and tabulated thus :
Table 1: Multi-Verb Constructions in Igala
Features
i. Connectives as marker of syntactic dependency
SVCs
No
Consecutive
Yes/but can be
Overlapping
No
omitted
ii. The individual verbs can function as independent verbs
in simple clauses
Yes
Yes
Yes
iii. Each verb can have different aspect and/modal marking
No
Yes
Yes
iv. Each VP can be independently negated
No
Yes
Yes
v. Same subject for all the VPs in the construction
Yes
same/different
same/different
vi. Subject argument of each verb is overtly expressed
No
Yes
Yes
vii. The number of clause
mono
bi-/multi-
bi63
4. Demarcation between SVCs and Other
Multi-Verb Constructions (10)
• From this summary, pictorial representations of the structures
of these three constructions could be generated by adopting
Pavey’s (2010) model for representing complex structures as
used earlier for the derivations of SVCs in Igala (c.f. fig.4
represented as fig 6) and Figs 7 and 8 below.
Consecutive Construction
SVC
Fig. 6.
Fig.7.
CLAUSE
CLAUSE
CLAUSE
.
CORE
CORE
CORE
CORE
NUC
NUC
NUC
NUC
Áudù là ọjẹ
je̩
Me̩ gwè̩̩ o̩wo̩ ki =a
jẹ e̩̩un
64
4. Demarcation between SVCs and Other
Multi-Verb Constructions (11)
Fig. 8.
Overlapping Construction
CLAUSE
CLAUSE
CORE
CORE
NUC
NUC
Ò̩un là̩ o̩te̩
à̩ mo̩
65
5. Conclusion (1)
• SVCs are clearly recognizable and highly productive grammatical
constructions in Igala. They are different from other closely related multiverb constructions because they are monoclausal and the VPs in SVCs are
construed as occurring in the same temporal frame (single marking of
grammatical categories per SVC). There are many types of SVCS in Igala
based on both semantic and formal classifications: symmetrical vs.
asymmetrical (involving transitive, intransitive and stative verbs), contiguous
vs. non-contiguous, etc.
• SVCs in Igala serve to convey a broad range of semantic notions which are in
tandem with the culturally recognizable activities and worldview of the
people. The versatility in function, which relates to the productiveness of
SVCs in Igala, correlates with the scarcity of conjunctions and prepositions
in the language.
• In addition, the functional motivation for SVCs in the language, one can
argue, is predicated upon discourse organization and the information
packaging efficiency of the construction(c.f. Aikhenvald 2006). For, instance,
both the symmetrical and asymmetrical SVCs are employed as invaluable
means of coherent information packing and simplification of complex
66
events in Igala thereby avoiding some unnecessarily complex constructions.
5. Conclusion(2)
• Also, serialization of synonymous and concomitant VPs could yield
some idiomatic expressions and high-flown literary styles in the
language.
• Lastly, in spite of the depth of coverage given to SVCs in this study,
there may still be need to investigate the phenomenon further in
Igala with respect to the following:
 Semantic overlaps between cause-effect and causative SVCs, manner
and event-argument SVCs as well as switch function SVCs in Igala.
 More on headedness in asymmetrical SVCs and demarcation
between SVCs and other compex predicates in Igala in relation to
other (West)Benue - Congo lanuages.
 Frequency of the occurences of SVCs in discourse to determine
whether SVCs in Igala occur more in one kind of discourse than
another.
 Grammaticalisation and Lexicalisation of SVCs in Igala.
67
References (1)
Agbedor, P.1994. ‘Verb serialization in Ewe’. Nordic journal of African studies 3(1), 115 – 135.
Aikkhenvald, A. Y. 2006. ‘Serial verb constructions in typological perspective’. In A.Y. Aikkhenvald &
R.M.W. Dixon (eds.). Serial verb constructions: a cross-linguistic typology, 1-68. New York: Oxfor
University Press.
Aikkhenvald A.Y. & Dixon R.M.W. (eds.) 2006. Serial verb constructions: a cross-linguistic typology. New
York: Oxfor University Press.
Ameka, F.K. 2005. ‘Multiverb constructions in a West African areal typological perspective’. In M.
Vulchanova & T.Afarli (eds.). Grammar and beyond: essays in honour Lars Hellan.15-42. Oslo:
Novus Press.
Anagbogu, P. N. & Omachonu, G. S. Forthcoming. ‘Headedness and demarcation in nominal
compounds: evidence from Ígálà, Ìgbò, Kọ̀ríng and Yorùbá’.
Andrews, A.D. & Manning, C.D. 1999. Complex predicates and information spreading in LFG. Stanford:
Centre for the Study of Language and Information
Awobuluyi, O. A. 1967. ‘Studies in the syntax of the standard Yoruba verb’. PhD Dissertation, Columbia
University.
---------------- 1971. Splitting verbs in Yoruba. Actes du huitieme congres international de linguistique
Africaine vol. 1: 151-164.
----------------1973 ‘Modifying serial verbs: a critique’. Studies in African linguistics (S.A.L), 4.2.87-111.
Awoyale, Y. 1987. ‘Perspectives on verb serialization’. In V. Manfredi (ed.). Niger-Congo syntax and
semantics, 1:3-35.
Baker, M. 1989. ‘Object sharing and projection in serial verb constructions’. Linguistic
Inquiry, 20:513 – 553.
Baker, M. & Stewart, O. T. 2002. ‘A serial verb construction without constructions’. Ms, Rutgers
University.
Balmer, W.T. & Grant, F.C.F. 1929. A grammar of the Fante-Akan language. London: Atlantic Press.
68
References (2)
Bamgbose, A. 1972. ‘What is a verb in Yoruba?’ In A. Bamgbose (ed.). The Yoruba verb phrase. Ibadan:
University of Ibadan Press.
---------------- 1973. The modifying serial verb construction: a reply. Studies in African
linguistics (S.A.L), 4.2.207-217.
---------------- 1974. ‘On serial verbs and verbal status’. The journal of West African languages (J.W.A.L), 9.1:1748.
----------------1982. ‘Issues in the analysis of serial verb constructions’. The journal of West African languages, xii
(2), 3-21.
Bearth, T. 1999. The contribution of African linguistics towards a general theory of focus: update and critical
review. Journal of African languages and linguistics. 20: 121-156.
Bisang, W. 2009. ‘Serial verb constructions’. Language and linguistics compass 3, 1749-1818.
Blench, R. 2004. ‘The Benue-Congo languages: a proposed internal classification’ (Ms).Cambridge, 24th June,
2004.
Boadi, L. A. 1968. ‘Some aspects of Akan deep syntax’. The journal of West African languages.5.2:83-90.
---------------- 2000. ‘Serial verb constructions’. Plenary lecture at the 22nd West African Languages Congress.
Legon: University of Ghana.
Chomsky, N. 1957. Syntactic structures. Mouton, The Hague.
---------------- 1965. Aspects of the theory of syntax. Cambridge Massachusetts: MIT Press
---------------- 1981. Lectures on government and binding. Dondnecht: Foris.
---------------- 1982. Some concepts and consequence of the theory of government and binding.
Cambridge, Mass: MIT Press
---------------- 1986. Barriers. Cambridge Mass: MIT Press
--------------- 1995. The minimalist program. Cambridge, MA: MIT Press.
Christaller, J. 1875. A grammar of the Assante and Fante language called Twi. Basel: Basel Evangelical Mission Society.
69
References (3)
Dechaine, R. 1988. ‘Towards a typology of serial constructions in Haitian’. In V. Manfredi (ed.). Niger
Congo syntax and semantics, 1:49-64.
Durie, Mark. 1997. Grammatical structures in verb serialisation. In: Alsina, Alex, Joan
Bresnan and Peter Sells eds. Complex predicates. Stanford: CSLI pp 289-354.
Ejeba, S. O. 2011. ‘Aspects of Igala syntax: a transformational generative account’. In G. S. Omachonu
(ed.). Igala language studies, 103- 151. Germany: LAMBERT Academic Publishing.
Foley, W. A. & Oslon, M.1985. ‘Clausehood and verb serialisation’. In J. Nichols & A. C. Woodbury (ed.).
Grammar inside and outside the clause, 17-60. Cambridge: CUP.
George, I. 1975. ‘Typology of verb serialization’. Journal of West African languages (J.W.A.L)
X,1.78-97.
Givón, T. 1991. ‘Some substantive issues concerning verb serialization: grammatical vs cognitive
packaging’, 137-84 Lefebvre (ed.).
Johnson, S. 2005. Revisiting the structure of serial verb constructions. MS, Department of Linguistics,
Michigan State University.
Lord, Carol. 1993. Historical change in serial verb constructions. Amsterdam: John Benjamins.
Omachonu, G. S. 2006. Verb phrase serialization in Igala. In O-M. Ndimele, C.I. Ikekeonwu & B.M. Mbah
(eds.). Language & economic reforms in Nigeria. Port- Harcourt: M & J Grand Orbit Communications
Ltd. &Emhai Press. 237-246.
------------------- 2011. (ed.). Igala language studies. Germany: LAMBERT Academic Publishing.
Pavey, E. L. 2010. The structure of language: an introduction to grammatical analysis. Cambridge:
Cambridge University Press.
Pawley, A. P.1987. ‘Encoding events in Kalam and English: different logic for reporting experience’. In R.
Tomlin (ed.). Coherence and grounding in discourse. 329-60. Amsterdam: John Benjamins.
Potsdam, E. 1997. English verbal morphology and VP ellipsis. In Proceedings of the 27th meeting of the
North East linguistic society. Amherst, Mass.: GLSA, University of Massachusetts at Amherst, 353368, 1997.
70
References (4)
Schachter, P. 1974. ‘A non-transformational account of serial verbs’. Studies in African
linguistics’. Supplement 5: 253 –270.
Schiller, E. 1990. ‘The typology of serial verb constructions’. Chicago linguistic society
26: 393- 406.
Shibatani, M. 2009. ‘On the form of complex predicates: toward demystifying serial
verbs’. In Johannes Helmbrecht, Yoko Nishina, Yong-Min Shin, Stavros Skopeteas,
Elisabeth Verhoeven, (eds.). Form and function in language research: papers in
honour of Christian Lehmann. Trends in Linguistics. Studies and Monographs.
Mouton de Gruyter.
Stewart, J. 1963. ‘Some restrictions on objects in Twi’. JWAL, 1(2), 145-149.
Yusuf, O. 1997. Transformation generative grammar: an introduction. Ijebu Ode:
Shebiotimo Publications.
Westeramnn, D. 1907. Grammatik der Ewe-sprache. Berlin: Diedrich Reimer.
------------------- 1930. A study of the Ewe language. Oxford: Oxford University
Press.
Westermann, D. &Bryan,M.A. 1952. The languages of West Africa. London: Oxford
University Press.
Williams, W. 1971. ‘Serial verb constructions in Krio’. Studies in African linguistics,
Supplement 2: 47-63.
Williams, K. & Blench, R. 2000. ‘Niger-Congo’. In B. Heine & D. Nurse (eds.). African
languages: an introducrion, 12-43. Cambridge: Cambridge University Press.
Zwicky, A. 1990. ‘What are we talking about when we talk about serial verb
constructions?’ 1-13, Joseph & Zwicky (eds:).
71