O.O. Oyelaran.
1. THE
PROBLEM AND ANTECEDENTS[1].
In
his 1972 work, Oke presents perhaps the first systematic attempt to subsume
under a single category, Auxiliary, all
the grammatical formatives which earlier studies (including Awobuluyi 1976)
have suggested may occur in combination
of varying lengths between the subject and the main verb of the Yoruba
sentence. Thus the “Auxiliary”, he proposes, has the following structure:
(1) AUX
à MODAL INTENSIFIER PRE-EMPTIVE.
Oke’s very useful contribution is,
however, limited in several respects. First, it excludes without any
explanation the negative marker (NEG) and the high tone syllable (HTS,
following Awobuluyi’s (1975) usage) the latter of which occurs unexceptionally
after and is assimilated to the final vowel of all non-pronominal subject,
followed by a full verb, as in (2) :
(2) àwọn
túlẹ̀ ẹ́ gba ìsinmi
they
students HTS take leave
‘The
students are going on holiday’.
Secondly, Oke excludes (1972:135,
f.n. 2) without adequate justification formatives such as dédé, gbọ́ọ̀dọ̀, on the
ground that they may be nominalized like verbs by the prefix à -.
Furthermore, the weight of counter examples to the co-occurrence restrictions
Oke proposes for the formatives call for a greater in-depth reconsideration of
the AUX in the language, given especially the tenuousness of the relationship
between the accepted usage of the terms “modal”, “intensifier”, for example and
the syntactic and semantic content of the formatives assigned to them in the
study under consideration.
Small
wonder that Awobuluyi (1975, 1978) opts to retain his own more traditional
cover-all label, “Preverbal adverb” (PVA), to subsume the following syntactic
and semantic functions:
(3) i. PVA NEG TNS ASP MODAL
MANNER
ii. TNS Past/Present Future Neutral
iii. ASP Imperfective Perfective.
In
what follows, we propose that, given the various disparate syntactic functions
involved, the cover-all-terms “preverbal” and “adverb be dropped in respect of
the formatives concerned, except, perhaps, in informal discourses. But more relevantly,
we propose, tentatively, an internal structure for the category AUX in the
Yoruba Phrase Structure which accounts grosso modo for the formative under
consideration. The present proposal must remain tentative if for no other
reason than that we provide no systematic account for the co-occurrence
restrictions among the formatives such as has been attempted by both Awobuluyi
(1967) and Oke (1972). We do, however, show that TNS is not a constituent of
the AUX and is therefore not a grammatical category of the language. We propose
that the category aspect (ASP) has a larger scope in the language than any
earlier study has accommodated. Our account of ASP finds justification in the
language although the labels for its categorial terms are inspired by Comrie
(1976).
2. DATA
Consider
now (2i) above, and the following additional examples, all from Yoruba. Formatives
to be discussed are underlined:
(3). àwọn
túlẹ̀ ẹ́ ti òní
gba ìsinmi
they students
HTS from today take leave
‘The
students go on holiday as from today’.
(4) àwọn
túlẹ̀ ń ti òní gba
ìsinmi
they
students PROG
from today take leave
‘The students ar
ar proceeding on holiday from
today’
(5) àwọn
tùlẹ̀ kò ti òní
gba ìsinmi
they
students NEG
from today take
leave
‘The students
are going on holiday (but) not from today.
(6) àwọn
tùlẹ̀ ẹ́
kúkú ti ti òní gba isinmi
they
students HTS anyway PERF from
today
take leave
‘As a matter of fact the students will be going
on holiday from today’.
(7) àwọn
tùlẹ̀ ẹ́ kúkú lè ti ti òní gba
they
students HTS anyway may PERF
from today
take
ìsinmi
leave.
‘The
students may indeed, have gone
on holiday from
today’.
(8) àwọn
tùlẹ̀ yóò
̣ ti òní gba ìsinmi
they students (
) from today take leave
‘They
students will go on
holiday from today’
(9) àwọn
tùlẹ̀ yóò
kúkú sẹ̀sẹ̀ ti òní
gba ìsinmi
they
students ( ) anyway just from
today
take leave
ni
FOCUS
‘As
a matter of fact the students will just be going on holiday from today’.
(10) àwọn
tùlẹ̀ kò kúkú níí lè tíì
ti
òní
they
students NEG anyway ( ) may
PERF
from today
gba
ìsinmi
take
leave
‘The
students may in fact not
yet be
able to go on holiday from today’.
(11)i. ògá wọn ón bá
wọn dé Ekó
master
their HTS accompany them
reach Lagos
‘Their master went with them to
Lagos’.
ii. ọ̀gá
wọn ọ́n máa ti Ilé-Ifẹ̀ bá
master their HTS ( ) from accompany
wọn
dé Ekó
them
reach Lagos
‘Their
master will go with them to Lagos
from Ile-Ife.
(12)i). bí Ayọ̀ ọ́ bá
kúkú tètè dé.
if
Ayo HTS happen indeed
early arrive.
n
ó lọ
I ( ) go
“If Ayo
happens to arrive early I shall go”.
ii Bí o bá
kúkú tètè dé n ọ́ lọ
if
you (sg) happen indeed early arrive I ( ) go
‘If you
indeed return early, I shall go.’
iii. Bí
o ọ́ báá kúkú tètè dé n
ó dúró
if
you ( ) happen indeed
early arrive I ( ) wait
‘If
you
will indeed return early, I shall wait’.
iv. Bí
o ọ̀ bá kúkú níí lè tètè dé n o
if
you NEG happen indeed ( ) able
early come I ( )
maa lọ
PROG go
‘If
you will not indeed return early, I shall be going’
(13)i. àwọn òlọ́pàá yóò sáà kọ́ tètè
they
police
( ) incidentally first early
máa
ti ọ̀dọ̀
onínǹkan da ojú
PROG/ ( ) from place proprietor make
face
ọ̀rọ̀
rú
affair
confuse
‘The police will by the way as a first step quickly confuse
the case beginning with the proprietor’.
ii. àwọn
ọlọ́pàá kọ̀ sáà níí
kó máa
they
police
NEG incidentally ( ) first
PROG
ti
ọ̀dọ̀ onínǹkàn da ojú
ọ̀rọ̀ rú
from place proprietor make face
affair confuse
‘The police will not by
the way as a first step quickly confuse the case beginning with the
proprietor’.
iii. Onínǹkan náà ọ́ mà ti máa ti
proprietor
the
( ) certainly PERF PROG from
ibẹ̀rẹ̀
di àwọn ọlọ́pàá mẹ́ru
beginning
pack they police with-baggage
‘The complainant himself shall certainly have from the start
taken the police into consideration’.
iv. Onínǹkan
kò sáà kúkú níí mọ́ọ̀mọ̀ ju
proprietor
NEG incidentally anyway (
) intentionally cast
ara
sílẹ̀ fún àwọn
ọlọ́pàá
body
to-ground for they police
‘The complainant will indeed not knowingly let the police throw
him around as they please’.
v. onínǹkan náà
yóò sáà ti kọ́kọ́
proprietor
the ( ) incidentally PERF first
ti
ẹ̀yìn agbẹjọ́rò rẹ̀ rí àwọn
ọlọ́pàá
from
back lawyer
his see they
police
‘The
complainant himself will incidentally
have
first contacted the police behind the back of his own advocate’.
3.0. THE
CATEGORY AUX DEMARCATED
3.1. Delimiting
the Auxiliary
We
would like first to delimit the category AUX in relation to other major
syntactic categories of Yoruba, then argue that AUX and the Verb Phrase are
both constituents of the PREDICATOR as far as the Yoruba language is concerned.
The
substructure ti òní (4(i) to (ii)), ti Ilé Ifè (12(ii))), ti
ọ̀dọ̀ onínǹkan (14(i)-(ii))), ti
ìbèrè are constituents of the PP
consisting of the preposition ti and
a NP. This ti introduced PP is the same as one of
Awobuluyi’s (1978) preverbal adverbials. We would like to suggest that this PP
strictly subcategorizes for the Verb Phrase (VP) in Yoruba. That is to say that
anything that follows it in the phrase is of necessity a VP. This has the
effect of excluding Awobuluyi’s 1978 bá and
fi from the class of preverbal
prepositions. Examples (11) above, and (14) below (of. Awobuluyi, 1975, section
5. 16):
(14) ó ti ibí
bá mi
fi ọkọ̀
kó o
he
from here accompany me apply
vehicle pack it
‘From here
he helped me me move it by
lorry’
are
therefore serial verb constructions with bá,
fi and kó as verbs[2]. For the purpose of the present study, then,
everything which may normally occur to the left of the preverbal ti-introduced PP is a constituent of the
category AUX. For the same purpose, the syntactic NP category and functional
subject constitutes the left hand context of the AUX. In the above examples,
therefore, all the elements which occur in the data section and brought
together here in (15) are either mere elementary symbols of the AUX or tokens
of same.
(15)
HTS NEG yóò bá
sáà kúkú níí lè ti
máa
ń
sẹ̀sẹ̀ tètè kọ́(kọ́)
Now,
because some of them commute as if they
enter into a relation of complement distribution, we shall arrange the one that
do, for the moment, in single columns as in (16):
(16) a b c d e f g
HTS yóò bá
lè ti
máa ṣẹ̀ṣẹ̀
NEG níí sáà
tíì ń tètè
kọ́
(kọ́)
Items
in each column commute and may, as in the case of a, b, c. and f, be mutually
exclusive in the sense we will explain presently.
3.2 The
Aux as Constituent
Once
again as a test of their being terms of the AUX, all of the items in (15) may
individually or in various combinations, as in (2) to (13) above, precede the
Pre-VP PP when there is one. We have avoided using the verb or the VP as the diagnostic context for the AUX
because of the ubiquitous discourse motivated deletion rule which eliminates
the verb even in the context of certain auxiliaries. Consider (17) and (18):
(17)i. A: a
ọ́ tètè dé ní ìrọ̀lẹ́.
we
shall early arrive LOC (this) evening
‘We
plan to
return early this evening’.
ii. B: ẹ tètè
you
PL early
‘Do
be early’
(18)i a ó tètè ti
oko dé ilé.
‘We
shall return
early from
the farm’.
ii. *
ẹ tètè ti oko
In
(17) and (18) the verb de with or without its complement may be
deleted. It is therefore not available as context for tètè in (17) ii) and
(18ii).
iii. ẹ
tètè
‘Be early’.
Now,
one crucial process suggests that the AUX, consisting of the class of elements
in (15), constitutes a syntactic subcomponent of the Yoruba Predicate (Pred).
That process is the deverbative nominalization. For our purpose, we consider
that the entire predicate may be nominalized in Yoruba, and that this process
applies ‘derivatively’ to the verb just in case it (the verb) constitutes by
itself the Pred in an utterance. Interestingly enough, there appears to be
three classes of nominal deverbative affixes, each distinguished by the class
of auxiliary terms they exclude, as it were. The three classes are represented
by àti -, à- and àí-. For the purpose of this exposition we put nominalization by
the reduplication of the consonant of the initial syllable of the Pred (e.g. bí – from (14)): bíbá – mi- fi-ọkọ́ - kó – o) in the same class with àì- , because
both appear to entertain similar possibilities. Consider these possibilities
with respect to (19) as exemplified in (20) for àti -, (21) for à-, and
(22) for àì-.
(19)i). Emi yóò kúkú
máa kọ́kọ́ ti oko
dé ilé
‘I shal incidentally
be getting home first from the farm’.
ii. n kò
kúkú níí gbọdọ̀ máa kọ́kọ́ ti
oko dé ilé
‘I
certainly won’t have to be getting
home first from the farm’.
(20)i. àti – dé-ilé
‘manner/fact/act
of arriving home
ii. àti
– toko-dé- ilé
‘manner/fact/act of arriving home
from the farm’.
iii. àti-kọ́kọ́-ti-oko-dé-ilé
‘Act
of arriving home from the farm first’.
iv. àti-máa-kọ́kọ́-ti-oko-dé-ilé
‘act
of the process of arriving home from the farm first’.
v. * àti – kúkú-máa-kọ́kọ́-ti-oko-dé-ilé
vi. * àti-yóò-kúkú-máa-kọ́kọ́-ti-oko-dé-ilé
(21)i). ?à-dé-ile[3]
ii. à-ti-oko-dé-ilé
‘act/process
of arriving home from the farm’.
iii. à-kọ́kọ́-ti-oko-dé-ilé[4]
‘act/process
of arriving home from the farm’.
iv. *à-máa-kọ́kọ́-ti-oko-dé-ilé
v. *à-máa-kọ́kọ́-ti-oko-dé-ilé
vi.
*à-yóò-kúkú-máa-kọ́kọ́-ti-oko-dé-ilé
(22) i) àì-dé-ilé
NEG-arrive-home
‘Failure
to arrive home’.
ii. àì-ti-oko-dé-ilé
‘Failure
to reach home from the farm’.
iii. àì-kọ́kọ́-ti-oko-dé-ilé
‘Failure
to reach home from the farm first’
iv. àì-máa-kọ́kọ́-ti-oko-dé-ilé
‘Not
arriving home from the farm first’.
v. àì-kúkú-máa-kọ́kọ́-ti-oko-dé-ilé
‘Lack
of the prospect of arriving home from the farm first’.
vi. àì-yóò-kúkú-máa-kọ́kọ́-ti-oko-dé-ilé
‘Lack
of the prospect of arriving home from the farm first’.
The asterisk (*) means ‘not
interpretable’, and, therefore, syntactically inadmissible and (?) ‘questionable’. The point is that the affixes are restricted
differently from the point of view of co-occurrence
with the elements in (19), and therefore in (15). Take NEG, for example. In àti –nominalization, NEG which is
normally kò, may only be realized as máà; à – nominalization process
as well as ai- is apparently not at
all admissible with NEG3.
Thus only (23) and (24) are possible:
(23)i. àti-máà-dé-ilé
‘manner/fact
of not reaching home’,
ii. àti-máà-ti-oko-dé-ilé
‘fact
of not arriving hme from the farm’
iii. àti-máà-kọ́kọ́-ti-oko-dé-ilé
‘fact
of not arriving home from the farm first’.
iv. àti-máà-máa-lọ́kọ́
ti-oko-dé-ilé
‘Lack
of the process of arriving home from the farm first’.
v. àti-máà-gbọ́dọ̀-maa-kọ́kọ́-ti-oko-dé-ilé
‘absence
of the obligation to be arriving home from the farm first.
vi. ?àti-máà-níí-gbọ́dọ̀-máa-kọ́kọ́-ti-oko-dé-ilé
‘the prospect of the absence of the obligation to be
arriving home from the farm first’.
vii. àti-máà-kúkú-níí-gbọ́dọ̀-máa-kọ́kọ́-ti-oko-dé-ilé
‘the incidence of the prospect of
not having to arrive home from farm first’.
viii. *àti-kò-dé-ilé
(24)i. àì-dé-ilé
ii. àì-kọ́kọ́-ti-oko-dé-ilé
iii. àì-ti-oko-dé-ilé
iv. àì-máa-kọ̀kọ́-ti-oko-dé-ilé
v. àì-gbọ́dọ̀-máa-kọ́kọ́-ti-oko-dé-ilé
vi. àì-níí-gbọ́dọ̀-máa-kọ́kọ́-ti-oko-dé-ilé
vii. àì-níí-kúkú-gbọ́dọ̀-máa-kọ́kọ́-ti-oko-dé-ilé
Argument in (20), (21), (22), (23),
(24) shows two things. First, with the exception of HTS, each of the subgroups
(columns) in (16) may be a part of a substructure which may be and is regularly
nominalized. This substructure happens to be the Predicate. The AUX is,
therefore, a subcomponent of the Pred, and we propose that the Phrase Structure
Rule which expands the Yorùbá Pred has the structure of (25) :
(25)i. Pred à AUX PP VP
ii. VP
à V NP PP VP
(See Oyelaran, 1983).
3.3. The
Structure of the AUX
The
second point made in the argument in (3.2) above is that the subcomponents of
the AUX subclassify into a finite number of syntactic subgroups. Listings by
Awobuluyi (1967, 1978) give about forty tokens of the AUX types. But for our
purpose, we do not consider a closed list to be of not more than heuristic
interest: already, we have noticed above that yóò and nii commute, and
are otherwise suppletive forms of the same formative. So are ń and máa, long accepted as suppletive forms of the imperfective (PROG)
aspect. For the rest, although HTS and NEG commute, each constitutes a unique
formative. We may now say that AUX expands as given in (26):
(26)
AUX à HTS NEG yóò kúkú le ti PROG sẹ̀sẹ̀
Now, we accept both Awobulyi’s
(1967, 1975, 1978) and Ogúnbọ̀wálé’s (1969) suggestion that ti and ń are forms for the Completed Action (PERF) and Continuous Action
(PROG) respectively. Both forms are therefore minimally terms of the aspectual
subsystem which we shall represent as ASP. Thus (26) should be rewritten as
(27):
(27) AUX
---> HTS NEG yóò kúkú le ASP sẹ̀sẹ̀.
Given that kúkú and sẹ̀sẹ̀ have
different restrictions, as by (20), (21) and (23), we suggest that they belong
to different sub-categories of the AUX. Let us call these INTENSIFIER (INT)[7]
for kúkú and SPECIFIER (SPE) for sẹ̀sẹ̀. Of particular interest in
distinguishing these two is the observation that SPE is verb-like in that it
subcategorizes for NP subject, while such subcategorization is non-relevant for
INT. For the Present, we style le and
other members of its paradigmatic class such as gbọ́dọ̀, níláti, as MODAL (MOD). Consequently, (27) may now reduce
to (28):
(28) AUX
à HTS NEG yóò INT MOD ASP SPE.
4. ASPECT
NOT TENSE
Now
in (3), recalled here, we have summarized Awobuluyi’s view of the AUX as PVA
(3i) PVA
à NEG TNS ASP MODAL MANNER
where TNS has three terms, namely,
HTS, which he considers the marker for the non-future or the past/present,
future (FUT) tense realized by yóò and
its alternants, and the Neutral tense. In what follows, we shall show that no
convincing case has hitherto been made for tense as a grammatical category in
Yorùbá.
4.1. The
Case for HTS
As
Awobuluyi (1975) himself asks, if HTS is a tense marker, why is it that the
following constraints apply to it?
(a) why
is it zero in the context of NEG, yóò, PRO,
and before the Awobuluyi-accepted verb dá,
ńkó
(b) why
does HTS co-occur with máa, a
supposed alternant of the ‘Future’ marker yóò
as in (11ii) and (29)?
(29) Ayọ̀ ọ́ máa
lọ.
Ayo
HTS FUT go
Ayo
will go.
(c) why
does HTS occur in what one would expect to
be a non-tensed context namely, in indirect command, final
constructions, and in conditional constructions which are not contrary to fact,
as in (30)?
(30)i. mo ní kí
ọmọ náà án jáde
I
say that child the HTS go-out
‘I
ordered the child to go out’,
ii. mo
fẹ́ kí
ọmọ náà án jáde
I
want that child the HTS go-out
‘I
want the child to go out
iii. bí
ọmọ náà án
bá jáde
if child the HTS happen go-out
kí Ayọ̀ ọ́ wọlé
let-that Ayo
HTS enter
‘If/when
the child goes out, let Ayo in’,
(d) We
would like to ask, in addition, how one could explain the occurrence of the HTS
in both rhetorical sayings as in (31i) and straightforward interrogative
propositions such as (31ii-iii).
(31)i. Ọtẹ̀ àgbàdo
ó ṣe lè kúrò nínú
intrigue
corn HTS do able leave
inside
ọmọ
àparò
child
bushfowl
‘How
can the bush fowl ever give up contriving to get at maize for food?’
ii. Níbo
ni ọ̀mọ ọ́n wà
lónìí
Where
FOCUS builder HTS be today
‘Where
is the builder today?’
‘Where
are builders today?’
‘Where
can one find builders today?’
‘Where were
builders today?’
. . . . . . .
iii. Njẹ
Ẹbun un wa nile nisiyii?
QUES
Ebun HTS be at-home now
Is Ebun
at home at this moment?’
It is important to note that the
context in which the HTS is zero are of two types. One is the context of
uncertainty or denial, that is, when formatives which semantically imply either
a denial or uncertainty are present. Interestingly enough, these include
contexts in which NEG ((16), (11), (12iv), (13iv), yóò ((9),
(10), (12iii), (13ii), (13iv), (13v)), the interrogative markers dà and ńkọ́ occur as in (32), and in gnomic expressions which are
semantically indeterminate, as in (33):
(32)i. àwọn dà
they how-about
‘Where
are they?’
ii. Ilé
ń kọ́?[8]
ile
how about
‘How
is home?
(33). ẹ̀wọ̀n já níbi
tí ó wù ú
chain
break where
REL it
pleas it
‘A
chain snaps where you least expect’
We wish to suggest that HTS does not
occur by the same tokes in contexts where other formatives ensure certainty or
definiteness in the VP or the Predicate. Such contexts include constructions
with the focus marker ni, and the
Pronouns (mo, o, ó ,etc.). Now, this latter context will sound strange
considering that, as Fresco (1970) shows conclusively, HTS occurs after
pronouns in certain dialects of Yorùbá such as Ọwọ̀, Ọba, and, we add,
Igbómìnà and Ọwọ́rọ̀ (Oyelaran 1978). The difference here is that in common
Yorùbá and a few other dialects, the pronouns have become cliticized to the VP,
which is why they undergo the harmonization and other intra-word assimilatory
processes in these dialects (Oyelaran 1985, Pulleyblank 1981). As deictics
(Lyons 1968, 1981), when cliticized to the VP, they do serve to definitize and
render the presence of the HTS superfluous.
As
a definitizer (DEF), then, it is not surprising that HTS is absent in these
contexts, but occurs as in (29) and (30) where Awobuluyi, who considers it a
tense marker, should not expect it. Notice incidentally that HTS in (29) asserts
Ayọ̀’s departure, as it does the Predicates in (30) from the point of view of
the speaker. Similarly, ọmọ àparọ̀’s
attitude in (31) is not subject to doubt, a fact marked by HTS. HTS cannot,
therefore, be validly considered a term of the category TNS in Yorùbá. And the
propositions questioned in (31i, ii) by means of sentential markers níbo and ǹjẹ́ are assertions.
4.2. The
Status of yóò
If,
then, HTS is not a tense marker, tense as a subsystem f the AUX in Yorùbá is
reduced to only one term, namely, Future, marked by yóò. Is this
tenable?
In
his important study on tense, Ultan (1978) groups tense systems into two,
namely the prospective system and the retrospective system.
A
tense system is prospective just in case “a present tense may ordinarily mark
an MOS (a moment of speech) future or if the latter may be unmarked”
(Ultan 1978;88). On the other hand, a tense system is retrospective “if a
present may ordinarily mark an MOS past or if the latter may be
unmarked” (Ultan 1978;60). He adds:
Thus for prospective languages, the dichotomy past-nonpast
is primary; for reterospective languages. Future-nonfuture. In the extreme
case. The former shows no distinction between present and future… Similarly,
for retrospective languages where the distinction between present and past is
lost…. Retrospective languages would tend towards greater markedness of future
time than prospective languages.
(Ultan 1978:89)
While
the study nowhere assumes that tense is of necessity present in all languages
as a grammatical category, its proposal of tense types tempts one to classify
Yorùbá as a retrospective language which marks the future tense by means of yóò
(and its variant and suppletive forms), and which leaves non-future unmarked.
This precisely is the stand taken by Ogunbowale (1970). Awobuluyi’s stand also
reduces to Ogunbowale’s if, as it has been shown above, HTS does not mark tense
at all.
Unfortunately,
any suggestion based on Ultan’s work to the effect that Yorùbá has tense system
and that it is the retrospective type appears to be unsupported by the facts of
the language on two crucial grounds which we must now examine.
4.21. Argument Based on Future Neutralization
First,
if Yorùbá was a retrospective language, then it ought to show “finer [temporal.
O.O.O] gradation in the past then in the future” (Ultan 1978:64). Alas! no
trace of gradation of the past say into recent, remote, general, or the like,
exists in the language. Secondly, Ultan’s work leads one to expect that
retrospective languages, and therefore Yorùbá, should neutralize the future
tense in the following construction types: subjunctive, negative, gapping,
subordinate (in AFTER-clauses, WHEN-clauses, probably condition clauses, and
BEFORE-clauses), back-shifting, and participles. And the study goes further to
underscore the binding force of neutralization in respect of subjunctive,
negative and indefinite constructions, saying specifically that
The relatively high incidence of neutralization of future
tenses in subjunctive and negative constructions is logically compatible with
the measure of uncertainty, indefiniteness… the unknown quality…that is of
necessity associated with the future in as much as the categories of
subjunctive (…) and negative are perhaps to a certain extent inherently
indefinite.
(Ultan
1978: 60)
Of
these construction types, gapping is yet to be identified for Yorùbá, if in
fact, it exists in the language. Again since the verb does not inflect in
Yorùbá, the subjunctive type would exist à la rigueur and then only
syntactically, namely, in kí-constructions
such as in
(34)i Mo fẹ́ kì Adé
lọ
I wish COMP Ade go
‘I like Ade to go’
ii. O yẹ kí Adé
lọ
It fit
COMP Ade go
‘It is fitting that Ade should go
in which, indeed, yóò may not occur. But the point must be
made that even such constructions could pass for subjunctive type only because
the same notion is expressed by means of subjunctive constructions in languages
in which the grammatical category is morphological.
Backshifting is not recognizable in
Yorùbá, again because there is no marking for the past tense. In any case,
contrary to what the study under review leads us to expect of retrospective
languages, the so-called future marker, yóò,
is less commonly found in the main clause of Yorùbá contrary-to-fact
conditional constructions than hypothetical or potential elements, if, then, kí… bá or ì.. bá as in (35) is taken to be an example of hypothetical
elements.
(35)i. Bí ó bá rí
bẹ́ẹ̀ yóò ti dé
if
it happen see so (
) PERF come.
‘If
it was so he would have come’
If
it is so he must have arrived’
ii. bí
ó bá
rí bẹ́ẹ̀ kì bá ti dé
if
it happen see so COMP happen PERF come
‘If
it was so he would have come’.
Ultan
also finds that retrospective languages neutralize the future-non-future
dichotomy in certain subordinate clauses (AFTER-clause, WHEN-clause, IF-clause
(probable condition), and BEFORE-clause) and employ the present subjunctive or
some atemporal marker. On the other hand. These languages prefer to use the
future in subordinate clauses expressing purpose and in clauses “that form the
subject of certain verbs of perceptions or mental activity (expectation, fear,
supposition, etc.)…” (Ultan 1978: 87). The following example would suggest that
these findings accord well with the Yorùbá use of yóò as future marker.
(36)i. AFTER-clause:
Lẹ́yìn
ìgbà tí
wọ́n dé a jáde.
behind time INT they come
we go-out
‘After
they had come we left.
ii. WHEN-clause.
Nígbà
tí wọ́n
dé a
jáde
at-time
INT they come we go-out
‘As
they arrived we left’.
iii. PROBABLE
CONDITION:
Tí
wọ́n bá dé a ó
jáde.
if
they happen come we ( ) go-out
‘If
they happen to come we shall go-out’.
iv. BEFORE-clause:
Kí
wọn tóó dé a ti jáde
INT they up:to come we PERF go-out.
‘Before
they arrived we had left’.
(37)i. PURPOSE:
A
ó yára
dé oko
ẹ dúró
we
( ) quickly
get: to farm you (P1) wait
de wa
expect
us
‘We
wish to go to the farm, do wait for us’.
ii. EXPECTATION:
Wọ́n
reti pé
a ó dé lọ́la
they
pick-ear say we ( ) come
tomorrow
‘They
expected that we would come tomorrow’.
It
is however important to note that Ultan himself makes the point, in respect of
subordinate constructions, that his data only suggest “certain tendencies worth
noting” (Ultan 1978:97). Besides, and as will be discussed further below, the
relevant construction in (36) connotes a measure of certainty such as cannot be
attributed to the structures in (37). There is a sense therefore in which one
could suggest that the use of yóò in
the latter is in fact atemporal.
Another
of the diagnostic structure for Ultan is the participle. He finds that in
retrospective languages “future is neutralized in participles resulting in a
past-nonpast contrast” (Ultan 1978:97). The problem here in respect of Yorùbá
language is deciding whether or not the “participle” exists as a category. If
one takes it literally that the “participle” is a verbal form which may
function syntactically as a determinant, then the deverbative nominal derived
through the reduplication of the initial consonant of the predicator, then
followed with a high-tone high front vowel is a participle. If this is correct,
then the sense is not clear in which “the future is neutralized resulting in a
past-nonpast contrast” (Ultan 1978:97). Consider (38), for example:
(38)i. Olú yóó jẹ
Olu
( ) eat
Olu
will eat
ii.a. jíjẹ ‘eating; edible’
b. ?
yíyóò jẹ
In Yorùbá, (38ii.a.) is unique and
shows no past-nonpast dichotomy; (38ii.b.) is questionably interpretable and,
therefore, unacceptable, presumably because, as is implied in Oyelaran’s (1982)
suggestion, reduplication even of this type makes reference only to the initial
lexical morpheme of the unit to which the process applies.
At
best, therefore, one finds that the neutralization of the ‘future’ in the
participle as non-criterion for Yorùbá.
Finally, Ultan’s data suggest to him
that in retrospective languages, the future-nonfuture tense is neutralized in
negative constructions. Since the use of replacive morphemes as in (39ii)
cannot be considered as being equivalent to neutralization, one is forced to
conclude again that the findings do not apply to the Yorùbá language. But
Yorùbá allows no grounds for doubt that yóò
is not neutralized in negative constructions because it is retained after kì, a variant of the negative
marker kò, as in (39iii).
Example (39iii) is an injunction prohibiting the act which is the referent of
its predicate.
(39)i. Ojò yóò
rọ̀
rain
( ) fall
‘It
will rain’
ii. Ojò kò níí rọ̀
rain
NEG ( ) rain’.
‘It
will not rain’
iii. Ojò
kì yóò rò
rain
NEG ( ) fall
‘It
shall not rain’.
The foregoing considerations decidedly
provide little ground for accepting yóò as a future tense marker in
Yorùbá.
4.22. Yóò as Aspectual and Modal Formative
It
should in fact surprise no-one that yóò
is commonly taken for a marker of
the future tense. In its regular usage it expresses precisely the modal notions
and “open-ended” aspectuals which Ultan finds to be semantically most closely
associated with the concept of future time as primarily expressed in the verb
and related forms in his data.
In the case of modals, the most
relevant of these notions for Yorùbá are obligation, imperative,
volition/desideration, probability, potentiality and hypothetical
consideration. The use of yóò for these notions is a commonplace. It is
not surprising, therefore, that yóò often shows the same constraints and
co-occurrence possibilities as basic Modals (e.g. lè ‘able to’: and gbọ́dọ̀ ‘must’) in the language.
For example, the so-called progressive formative ń is replaced by máa
after Modals, same as after yóò as in (40) and (41).
(40)i. Olú ń kọrin
Olu
PROG sing
‘Olu
is singing’.
ii. Olú
yóò máa kọrin
Olu ( )
PROG sing
‘Olu
will be singing’
b. *
Olù yóò ń kọrin
iiia. Olú lè máa kọrin
Olu
beable PROG sing
‘Olu
can be singing’
b. *Olù
lè ń kọrin
(41)i.a Nígbà tí wón dé
a jáde
‘As
they arrived we left’.
b.
*Nígbà tí wón yóò dé a jáde
c.
*Nígbà tí wọ́n lè dé a jáde
ii.a. Wọ́n retí pé a ó dé ní ọjọ́ keji
‘They
expected that we should arrive the following day’.
b. Wọ́n
retí pé a lè dé ní ni ọjọ́ kejì
‘They
expected that we may arrive the following day’.
Furthermore, in its volitive and
desiderative usage, students of Yorùbá have often identified yóò and fé
in the latter’s ‘so-called’ infinitive function, that is to say when fé
takes the nominalized predicate as in (42):
(42)i. a fẹ́ẹ lọ bí òjò bá ti
dá
we want(NML) go if rain happen stop
‘We
wish to set out as soon as it stops raining’.
ii. A
ó lọ
bí òjò bá ti dá
‘We
wish to
set out as soon as
it stops raining’.
By “open-ended” aspectuals, Ultan
has in mind the constructions expressing concepts which are inchoative,
durative, incompletive, gnomic, actually timeless or infinite, customary. Again
examples abound in everyday speech, and the following should suffice to
illustrate the point:
(43)i. GNOMIC:
Kòkò
tí yóò
jẹ ata
ìdí rẹ̀
yóò pọ̀n-ọ́n
pot
INTRO ( ) eat
pepper bottom its ( ) red
‘The
pot which will enjoy delicious stew will have fire-red bottom.
ii. CUSTOMARY:
Nígba
tí Olú
wà nílé
yóó jí
láàárọ̀
yóò
gbálẹ̀, yóò wẹ̀, kí ó
tóó jáde
‘When Olu was at home, he will wake up in the morning,
sweep, bathe before getting out’.
There
is a sense, therefore, in which both modal and aspectual usages exhaust the
functions of yóò in Yorùbá.
Since these usages are atemporal, there is no ground for asserting that yóò
marks the future. It may well be, of course, that yóò at some point in the evolution of the
language contrasted with other formatives, now lost to us, in order to express
tense. In the present conjecture, it no longer marks tense.
We
would like to conclude, therefore, that Tense is not a grammatical category in
Yorùbá. Instead, we hold that Aspect (ASP) is grammatical in the language, and
is a subcategory of the AUX.
4.3. Subcategories
of Aspect in Yorùbá
Distinguishing
between tense and aspect, Comrie (1976:2) has this to say:
Although both aspect and tense are concerned with time, they
are concerned with time in different ways… Tense is a deictic category, i.e.
locates situations in time, usually with reference to the present moment,
though also with reference to other situations. Aspect is not concerned with
relating the time of the situation to any other time-point, but rather with the
internal constituency of one situation; one could state the difference as one
between situation-internal time (aspect) and situation-external time (tense).
In
this acceptation, then, aspect has four terms in Yorùbá. We propose the
following to be the appropriate terms: the perfective (PERV) which is usually
unmarked, the Prospective (PROS) realized by yóò and its alternants, the
Imperfective (PROG) with ń (and its variants) as marker, and, finally,
the Perfective (PERF) realized by ti (tíi, in the negative[9]).
The data which Comrie (1976:3) uses
to define the Perfective aspect is given in (44), with its Yorùbá equivalent in
(45):
(44) ENGLISH: John was reading when I entered
FRENCH: Jean lisait quand j’entrai
SPANISH:
Juan leia cuando entré
(45) YORUBA:
Jọ̀ọ́nù ń kàwé nígbà tí mo wọlé
He comments as follows:
In each of these sentences, the first verb [read
‘predicate’, O.O.O.] presents the background to some event, while that event
itself is introduced by the second verb [predicate, O.O.O.]. The second verb
presents the totality of the situation referred to (here, my entry) without
reference to its internal temporal constituency; the whole of the situation is
presented as a single unanalyzable whole, with beginning, middle, and end
rolled into one; no attempt is made to divide this situation up into the
various individual phrases that make up the action of entry. Verbal forms with
this meaning will be said to have perfective meaning, and where the language in
question has special verbal forms to indicate this, we shall say that it has
perfective aspect (Comrie 1976:3).
We claim then that Yorùbá has the
perfective aspect, and it is unmarked as wọlé in (44). Given Comrie’s
definition, which we are adopting here, it is not surprising that this is the
‘normal’ construction for stative verbs as in (45).
(45)i. Oyin ín dùn
honey
HTS sweet
‘Honey
is sweet’.
ii. Mo
gbọ Yorùbá
I
hear Yorùbá
‘I
understand/speak Yorùbá’.
It is interesting that when verbs of
perception such as gbó takes the Imperfective marker ń, the only
permissible reading is iteration, since
reference to situation-internal time would be nonsensical. Similarly, it is not
normally interpretable, and is therefore unacceptable to use these verbs in the
Perfective aspect with the marker ti as in (46).
(46)i. *Oyin ín ti dùn
honey
HTS PERF sweet
ii *Mo ti
gbọ́ Yorùbá
I PERF hear Yorùbá
except, in the case of (46ii), with
the meaning of the assumption or accomplishment of the state referred to in the
verb.
The
Prospective aspect in the sense in which Comrie employs the term ‘Prospective’
contrasts with the Perfective. Thus he writes:
The Perfect is retrospective, in that it establishes a
relation between a state at one time end a situation at an earlier time. If
languages were completely symmetrical, one might equally well expect to find
prospective forms where a state is related to some subsequent situation for instance
someone is in a state of being about to do something.
(Comrie
1976:64)
It would be embarrassing to find
oneself in the situation where one attempts to invent categories in Yorùbá for
the sole purpose of illustrating Comrie’s theoretical postulate. Fortunately,
the debate on the status of yóò and the present writer’s point of view
about it antedates the publication of Comrie’s Aspect (1976). One observation which must be made is that the
uncertainty and the indeterminacy identified with the use of yóò in
Yorùbá accords well with the expectation of “prospective form” of Comrie’s
proposal. These same properties underlie the atemporal usages which, we have
suggested, exhaust the functions of yóò in the language.
The remaining two subcategories of
Aspect in Yorùbá namely, the Imperfective and the Perfect, are not
controversial and need not delay us here. Still Comrie’s (1976) account of both
in relation to each other and each in relation to other subcategories is
instructive.
5. CONCLUSION
In
presenting the case for Aspect in Yorùbá, we have not suggested, nor can we
suggest, had we been inclined to do so, that the HTS marks the Perfective
(PERV). One major reason that we cannot plausibly make that suggestion and why
we are unable to accept Awobuluyi’s (1978) suggestion to the same effect is
that the perfective interpretation applies even in cases where, as in negation,
HTS is zero. Furthermore, the HTS as definitizer (DEF) and the Imperfective
(PROG) are both neutralized under negation (47). This is not unexpected, since,
from the semantic point of view and as we have argued, neither continuity nor
certainty/reality is possible (i.e., is either true or assertable) under
denial, which NEG is.
(47)i. Ojò ó rọ̀
rain
HTS fall
‘It
rains/ rained’.
ii. Ojò
ń rọ̀
rain
PROG fall
‘It is raining
iii. Ojò
ń rọ̀
rain
PROG fall
‘It
is raining’.
iv. *Ojò
ó kò
rọ̀
v. *
Ojò kò ń
rọ̀[10]
The foregoing leads us to suggest
that, if one excludes the pronominal clitics, (48) must be included as part of
the Yorùbá Phrase Structure:
(48)i. AUX ---> HTS NEG INT
MOD ASP SPE
ii ASP ---> PERV PROS PERF PROG.
And
TNS (tense) is not a term within AUX or within any other auxiliary symbol in
the Yorùbá phrase structure. It is therefore not a grammatical category of the
language.
REFERENCES
Abimbola, Wande and Olasope O.
Oyelaran (1975) “Consonant Elision in Yoruba”, African Language Studies xvi: 37-60
Awobuluyi, O. (1967), “Studies in
the Syntax of the Standard Yorùbá Verb.”, Ph. D. Dissertation, Columia
University.
Awobuluyi, O. (1972), “On the
Classification of Yorùbá Verbs”, in Yorùbá
Verb Phrase, edited by Ayo Bamgbose, pp. 119-134, Ibadan: University Press.
Awobuluyi, O. (1975), “On ‘the Subject
Concord Prefix’ in Yorùbá”, Studies in
African Linguistics 6, 3:215-235.
Awobuluyi, O. (1978), Essentials of Yorùbá Grammar. Ibadan:
Oxford University Press.
Bamgbose, Ayo (ed.) (1972), The
Yorùbá Verb Phrase. Ibadan: University Press.
Bamgbose, Ayo (1983), “On Timeless
Sentences in Yorùbá”, Journal of Nigerian
Languages 1: 1-16.
Comrie. Bernard (1976), Aspect.
Cambridge: Cambridge University Press.
Fresco, Edward M. (1970), Topics in Yorùbá Dialect Phology, Studies in
African Linguistics Supplement 1.
Greenberg, Joseph A, Charles A, and
Edith A. Moravcsik (eds.) (1978a), Universals
of Human Language 1: Method and
Theory. Stanford: University Press.
Greenberg, Joseph A, (1978b), University of Human Language: Phonology. Stanford:
University Press.
Greenberg, Joseph A,(1978c), Universals of Human Language: Word Structure.
Stanford: University Press.
Greenberg, Joseph A,(1978d), Universals of Human Language: Syntax. Stanford:
University Press.
Lyons, John (1968), Introduction to
Theoretical Linguistics. Cambridge: University Press.
Lyons, John (1981), Language and
Linguistics, Cambridge: University Press.
Ogunbowale, P.O. (1970), The Essentials of the Yorùbá Language.
London: University Press.
Oke, D.O. (1972), “On the
Construction and Semantic Interpretation of Auxiliary Clusters in Yorùbá”, in The Yorùbá Verb Phrase, edited by Ayo
Bamgbose, pp. 135-162. Ibadan: University Press.
Oyelaran, O.O. (1982a), “Atunpin
Isori Oro Ise ni Ede Yorùbá: Ere abi Awada”, Paper Presented at the Dept. of
African Languages and Literatures, Obafemi Awolowo University, Ile-Ife,
Nigeria.
Oyelaran, O.O. (1982b), “On the
Scope of the Serial Verb Construction in Yorùbá”, Studies in African Linguistics 13, 2: 109-146’.
Oyelaran, O.O. (1983), “Sources and
Status of the Syllabic Nasal in Yorùbá”, An Unpublished Paper, Dept. of African
Languages and Literatures, Obafemi Awolowo University, Ile-Ife, Nigeria.
Oyelaran, O.O. (1985), “Vowel
Harmony and Vowel Change in Yorùbá”, An Unpublished Paper, Dept. of African
Languages and Literatures, Obafemi Awolowo University, Ile-Ife, Nigeria.
Oyelaran, O.O. (1992), “Tense/Aspect
in Ọwọ́rọ̀: A Yorùbá Dialect”, Research
in Yorùbá Languages and Literature 2:31-37.
Pulleyblank, D. (1980), “Clitics and Extraction in Yorùbá”,
An Unpublished Paper, M.I.T. Cambridge, Mass.
Stahlke, Herbert (1974a) “The
Development of the Three-Way Tonal Contrast in Yorùbá”, in Third Annual Conference on African Linguistics, 7-8 April, 1972,
edited by Erhard Voeltz, pp. 139-145. Bloomington: Indiana University.
Stahlke, Herbert (1974b), “Pronouns
and Islands in Yorùbá”, Studies in
African Linguistics 5, 2: 171-204.
Stahlke, Herbert (1976). “The
Pronoun Prefix in Yorùbá”, Studies in
African Linguistics Supplement 1:243-253.
Ultan, Russell (1978), “The Nature
of Future Tenses”, in Universals of Human
Language: Syntax, edited by J.H. Greenberg et al. pp85-123. Stanford:
University Press.
Wilkinson, Robert W. (1976),
“Contrast Preservation in Yorùbá”, Studies
in African Linguistics 7, 1: 65-92.
[1]
This paper was published as Oyelarn, O.O. (1992), ‘The Category Aux in the
Yoruba Phrase Structure’, Research in
Yoruba Language and Literature 3: 59-86.
[2] See Oyelaran (1983) for a different argument showing that fi in this kind of construction is a
verb and not a preposition.
(a) Mo dé
ilé gba ire
I arrive home
receive good
‘I arrive home to receive good
things’.
(b) à-dé-ilé-gba-ire
‘coming
home to receive good fortune
à- wí – ì - gbọ́
(
) say
( ) hear ‘incorrigibility’.
à-kọ́
- ì – gbà
‘an
unteachable person
à-
kúkú -ì- bí
(
) incidentally ( ) give birth to
‘an
unteachable person’.
suggest
the use of the prefix à-with an obligatory NEG-hopping, NEG being
realized as – ì – word – medially
NEG-HOPPING:
S.D.:
à- NEG
VP VP
à
S.
C.: à- VP
NEG
VP
Such
a nominalization process appears to be restricted to the Serial Verb
Construction, particularly those with two arguments (VPs).
[5] .i. The prefix àì normally renders máà unnecessary, the
co-occurrence of both àì and máà in urban Yorùbá, and the speech
of school leavers is coming into vogue. Indeed it is now not uncommon to hear
àì
– kò – dé – ilé
where
kò occurs in place of its suppletive form máà expected
pleonastically after NEG, and obligatorily after modals and in non-declaratory
constructions.
ii. níí
occurs as suppletive form for yóò after NEG.
Example
(24vii) is therefore more readily interpretable than (22vi). The latter is
heard in the speech of school leavers and of young urban speakers.
Similarly
tíì and máà occur as suppletive forms of ti PERF and ń
PROG respectively. tíì replaces ti following NEG, and máà
substitutes for ń both after NEG, yóò and before ń in
modal constructions.
[6] 5. In addition to
arguments based on co-occurrence possibilities and commutation, the deverbative
prefix à – (representing all mono-segmental syllabic deverbatives), on
one hand, and àì -, àti – and the reduplication process on the
other, differ with respect to semantic transparency. Indeed. Nominal derived by
means of à – are more readily lexicalized in that they show greater
tendencies to acquire idiosyncratic features. Consider for instance verbs such
as jọ ‘assemble (intr)’jọ̀ ‘sift’, ṣe ‘make, do, create,
came, etc’ jẹ ‘eat’.
The
following nouns and nomials may be derived from them:
(a). àjọ: ‘assembly,
society, contribution (in the sense of savings)’.
ìjọ ‘congregation (assembly of persons), the
act of assembling’.
ìjọ: ‘sieve, scraper, etc’.
èṣe: ‘injury, harm sufferd
through accidents, misfortune’.
ìjẹ: ‘food, bait, feeding trough, place which wild animals frequent
to feed and which is considered ideal for purposes of trapping them’.
(b) (i) jíjọ́, jíjọ̀, ṣíṣe, jíjẹ
(ii) àtijọ àtijò, àtiṣe, àtijẹ
(iii)
àìjọ àìjọ̀ àìṣe àìjẹ.
While items in (a) are lexical items
whose different meanings may not be a function of the original meaning of the
root verb and the prefix or may be only marginally so those in (b) are straight
forward and are interpretable from a casual examination of their constituent
morphemes. For purposes of interpretation for (b) items, there is no getting
away from the following notions:
(b)i) ‘manner/fact/act of’
‘process/fact/manner
of’
‘failure
or denial of’.
It
is significant to not also that the class which à – represents does not
nominalize structures in which aspect (ASP) is expressed. So that one is
tempted to apply Chomsky’s (1979) dichotomy “derived nominals” as opposed to
“gerundive nominals” to the Yorùbá deverbatives, identifying the à
–nominals with Chomsky type derived nominals.
But
the most important observations for us is that the distinction we are
suggesting among the three classes of deverbative prefixes is not a figment of this analyst’s own
imagination.
[7] More than any other label. INTENSIFIER is used for want of a
better term. For one thing, we are not at all sure that their function is to
“intensify”. In what sense does bá, sáà and kàn intensify, for instance?
[8] In Oyelaran (1982) we argue that dà? and ń kọ?
are in fact not verbs, contrary to Awobuluyi’s (1969, 1972, 1975, 1978)
suggestion. They function like and in fact belong to the same syntactic class
as interrogative markers bí kẹ̀
as in
(i) Eyí bi
this;
one Q
‘You
mean this one?’
(ii) Iwọ kẹ̀?
you(sg.)Q
‘You? (implying that the situation
is beyond the person questioned, and therefore incurring the questioner’s doubt
in the interlocutor’s capability)’
[9] See Bernard Comrie (1976) for an uplifting exposition on
Aspect, and an explicit distinction among these terms.
[10]In the context where this is interpretable, it may mean
“rain has never fallen” or that “rain is not in the habit of falling” in the
sense that “it never rains”. This seems to us a far cry from being a denial of
the Imperfective aspect which refers to the continuous unfolding of a single
process in a given situation.
No comments:
Post a Comment