Chapter 5

The Constitution of FOCUS:

Wolof

1. Introduction
Wolof is a member of the Niger-Congo family of languages. The sub-
grouping goes: Niger-Congo -> Atlantic -> Northern Atlantic ->
Senegambian. There is a Gambian Wolof which is distinct from Senegalese
Wolof. The latter is the language dsescribed here (International Encyclopedia
of Languages 1:131-132):1

2,000,000 to 3,000,000 first-language speakers reported in 1989, in western and

central Senegal, on the left bank of the Senegal River to Cape Vert. Also spoken
in Mauritania and in France. Also called Yallof, Walaf, Volof, or Waro-Waro;
speakers call themselves Wolof. It is the main language of Senegal, and the
second language of at least 500,000 people, predominantly urban. Different from
Wolof of Gambia, although related.

2. Some preliminaries
The simple sentence in Wolof is in some instances SVO and SV:

1 The Wolof data come from Sadibou Sow, a native of Njarab, Senegal, and from Alioune
Deme, a native of Kaolack, Senegal. These data were collected during a two-semester class
in fieldmethods and supplemented by correspondence with Sadibou Sow. I much appreciate
their patience with our questions. All mistakes of presentation and interpretation are my
responsibility. Occasionally, the initials “AD” or “SS” will appear in the margin to identify
when an example is assigned to one or the other speaker. Where comments appear in “double
quotes,” they come from a speaker of Wolof. If they are not attributed, SS is the source.
See also Becher (2002), Dunigan (1994), Kihm (1999), N’Dyiaye Corréard (1989), and
Robert (1986).
2 SYNTAX & SEMANTICS

The verb-like predicate element displays the following inflection for person
and number of the Subject in the ‘past’ tense. The following is for the EVENT
‘to kiss’:

(3) Sg Pl

Noun-like predicates inflect in a similar way:2

(4) Sg Pl

Adjective-like predicates have still another inflection:

2 Kihm (1999) in his discussion of focus in Wolof concentrates primarily on the role of this
paradigm (Kihm 1999.248), particularly , and argues that it is a “verbal element”
(Kihm 1999.247) claiming that “la compacts an entire VP” (Kihm 1999.248). He goes on to
segment the form into l that is “identified with the noun class prefix L- of generic reference”
and the “a copula” (Kihm 1999.249). Here and below, no such segmentation is proposed, and
whole forms such as those in (4) are taken to be stem plus suffix and to constitute (because
inflected for person and number of the Subject) a “verb”, because that is what “verbs” do.
But this is not really germaine to understanding Wolof. It is the function of the forms that is
at issue.
 FOCUS in Wolof 3

When an overtly named Subject PARTICIPANT occurs with a predicate (verb,

noun or an adjective), it commonly precedes, and other PARTICIPANTS follow:

2.1 Nouns
PARTICIPANTS in a PROPOSITION fall into several classes depending upon
the shape of the ‘definite’ article that occurs with them. Indefiniteness is
marked by the absence of the suffix:

The following classes have been identified:

4 SYNTAX & SEMANTICS

(8) Class Affix Class Affix

déx -bi

(g) nít -ki

Some PARTICIPANTS may accept more than one shape of the article:

(9) (a) fas-bi (b) dyígen-dyi

fas-wi dyígen-bi
‘the horse’ ‘the woman’

Others may not:

(10) gel-bi
*gel-si

Lengthening the vowel of the ‘definite’ article signals a proximal

demonstrative sense which contrasts with a distal demonstrative:

(11) (a) dyigén-dyii fón-n xále-bi

dyigén-bii fón-n xále-bi
‘This woman kissed the boy’
‘This woman kissed the boy’
 FOCUS in Wolof 5

(b) xále-bii fón-n dyigén-dyi

‘This boy kissed the woman’

(c) k r-gii
‘this house’

(d) fas-bii
fas-wii
‘this horse’

(e) ndóx-mii
‘this water’

and with the distal demonstrative:

Here a difference emerges between the shapes. further

away than it seems to “minimize the person ... You don’t know
the person” (AD). “I’d think of it in terms of distance” (SS). Lengthening
occurs elsewhere:

Plurality can be signaled by choice of ‘definite’ article:

6 SYNTAX & SEMANTICS

The plural demonstrative:

is the plural of both:

2.2 Pronouns
In addition to the inflections of (3), (4), and (5), which indicate ‘subject’,
Wolof seems to have a set of object pronouns which are post-posed to the
Verb. The pronominal object shapes are as follows:

Following Prepositions, e.g. ak ‘with’, the set of Pronouns in (18) occurs:

 FOCUS in Wolof 7

3. ROLES and VOICE

Wolof exploits three syntactic positions for the expression of ROLES: _V,
V_1, and V_ _2 :

The immediately preverbal position signals an Agentive-like ROLE, and the

EVENT will normally encode person and number information for the
PARTICIPANT filling it. When the AGENT is an overt noun, the inflection and
the noun co-occur. The first and second persons are not coded as independent
pronouns, but do appear as verbal inflections. The two postverbal positions do
not appear as agreement affixes. Sentences will either contain nouns filling
the two postverbal ROLES or an affix appropriate to the person and number of
the PARTICIPANT filling the ROLE. When expressed pronominally, the
8 SYNTAX & SEMANTICS

PARTICIPANTS appear post-verbally as suffixes following the inflection for the

AGENT.

Wolof appears to grade the postverbal content semantically in the

following way. The position immediately following the verb is the most
involved, defined, and precise. The second position following is less involved,
defined, and precise. Neither of these positions is associated with an overt
mark. While the V_1 and V_2 positions are frequently associated with a given
ROLE, the position by itself is not sufficient to identify the ROLE, which is
then signalled by a combination of position, EVENT semantics, and verbal
morphology. Cp. xále-bi ‘the boy’, yááy-am ‘his mother’, and ‘my
young sibling’ in (19) - (21). The third position in V_3 (and the least involved,
defined, and precise PARTICIPANT) is marked by one of several prepositions:
‘for’, tyi ‘to’, ak ‘with, and’, si ‘at’.

3.1 -ee-
The claims of the preceding section are most easily seen in the
examination of two verbal suffixes (-ee- and -al-), which manipulate the
semantics of ROLE and VOICE (i.e. position in the PROPOSITION). We will
begin with -ee-.
As noted above, the V_ 1 position appears to be appropriate to Recipients
(in [19]), a second sort of Recipient in (20), and Beneficiaries (in [21]). Both
(20) and (21) are possible because of -ee- and -al-, resp. Without those
derivational affixes, the expressions differ:
 FOCUS in Wolof 9

‘The man killed the pig for the sake of/to honor his mother’

The comparison of (24) with (21) and of (23) with (20) suggests that in each,
the ‘mother’ and ‘Awa’ are more intensely involved when they appear in the
V_1 position. In (21), the mother is present at the site of the EVENT, whereas
in (24), she can be absent, and the sense of a memorial is possible. Although
the English glosses of (20) and (23) are identical, there is a difference in that
baay-ee implies a focus on the PARTICIPANT in V_1 .
The suffix -ee- produces a second effect in these:3

the “focus is more on the tool” in (25b). Both bááyi ‘leave’ and dóór ‘hit’ are
grammatically transitive verbs. The -ee- affix, when added to them, does not
seem to produce the same effect in terms of ROLE. The result in (20) seems to
be more Recipient-like, and in (25b), more Instrument-like. Note also that
(25c) indicates that the order of PATIENT and INSTRUMENT does not change as
the order of PATIENT-RECIPIENT changed to RECIPIENT-PATIENT in (20). This
suggests that the content of a tyi phrase and what precedes are more tightly
bound than is the content of a phrase introduced by ak.4
Still a third result emerges when the affix -ee- is added to ditransitive
EVENTS:

3 Mbam-bék is a morphophonemic variant of mbam-bi ak , and ak is one of the Wolof

prepositions, as noted above.

4 One speaker (AD) accepts (25c). Another, (SS) asserts “Here mbam-bi should come before
bant-bi”.
10 SYNTAX & SEMANTICS

The semantic effect of -ee- is expressed in the English ‘away’. There is a

sense of disposition in (26), which is absent from the alternate expressions
without -ee-. In (26a), the child was given up for adoption, and in (26b), the
impression is that the money was not given to someone just to hold, but its
ownership was relinquished.
Figure 1 depicts the three effects of -ee-. In that figure, (a) represents the
effect of -ee- in (20). The relation of RECIPIENT is expressed in the V_1
position. The alternate expression is one with tyi in the V_3 position. The
effect described by (b) is that of (25), in which an INSTRUMENT is expressed
in V_ 2, and that is an alternate to one in which the INSTRUMENT is expressed
in V_3 with ak. In (c), the PATIENT is drawn to the V_1 position as in (26), and
First position Second position Third position
following following following
EVENT

tyi (a)

ak (b)

(c)

Figure 1: The effects of -ee-.

and the Recipient has no possible expression:5:

5 One speaker (AD) will accept

(i)
 FOCUS in Wolof 11

This RECIPIENT has no nominal nor pronominal expression as the RECIPIENT

in (28) allows both:

When -ee- expresses a relation of a PARTICIPANT for which the alternative is

an expression in V_3 , i.e., (a) and (b) in Figure 1, the effect is not so severe,
and the PARTICIPANT can — in contrast with (c) in Figure 1 — be expressed
as a noun or a pronoun:7

7 The order of pronominal suffixes, especially in (30b), confirms that the effect of -ee- is to
place a PARTICIPANT in V_2, even though the order of this ‘new’ PARTICIPANT may be
variable (Cf. foontote 6).
12 SYNTAX & SEMANTICS

These patterns suggest several conclusions. First, -ee- is clearly not the
mark of a ROLE. Its function seems to be more one of VOICE . It identifies a
PARTICIPANT in a less involved ROLE

(i) The Recipient in (23)

(ii) The Instrument in (25a)
(iii) The Patient in (26)

and it makes each more heavily enveloped by the EVENT. Second, the V_1 ,
V_2, and V_3 positions are graded semantically in such a way that V_2 has
more definition than V_3. Cf. the speakers’ reactions cited above and contrasts
such as:

In (31a) with the preposition si, the location is less a precise ‘somewhere’. In
answer to the question fán nékk ‘Where are you?’, (31) is a “vague
response”. In (31b), the location is more focused, and the speaker is precisely
in Senegal as opposed, say, France. The greater precision of V_1 over V_2 can
be further seen in the inability of the language to exploit the indefinite
grammar in V_1 , while it is possible in V_2 (cf. [7]):
 FOCUS in Wolof 13

The V_1 requires the preciseness of benn ‘one’ to achieve a kind of

indefiteness.
Certain relations remain beyond the reach of -ee-. Let us begin with (33):

From (33), it would seem that the si of location cannot interact with -ee-, but
the sentences of (34) indicate otherwise:

The contrast in (34) shows three things: (1) -ee- is possible with times and
places, (2) the time or place has to be FOCUS (Cf. section 5), and (3) it must be
deictically indicated (i.e., ‘here’ as opposed to ‘in this house’). The second
and third constraints on the interaction of place with -ee- support the ‘precise’
character of -ee-, and speakers will note the impression of “starting point”
with sentences such as (34a). This pattern extends to other uses:
14 SYNTAX & SEMANTICS

The sense of ‘why’ also has expression with -ee-, but again only when the
cause is FOCUS.
The behavior of -ee- in (33) - (35) now suggests a further organization of
material expressed by preposition, an organization in which tyi is more closely
allied with the preceding content, ak less so, and si, least. Some content, e.g. a
comitative relation, is beyond the pale of -ee-:

This suggests that a relation like the Comitative is not a ROLE since it has only
the peripheral expression marked with ak.

3.2 -al-
The suffix -al-, like -ee-, seems to interact with word order, and to
elaborate on the semantics of V__2. Sentences (24) and (21) are repeated here
as (37a) and (37b):
 FOCUS in Wolof 15

In (37a), the AGENT’s mother is a remote Beneficiary, i.e. ‘for the honor of’. It
expresses a commemoration, whereas in (37b) with -al-, the mother is present
and is more directly involved and affected: “His mom came to visit him and
he killed the pig to celebrate.” The alternative order of (37c) maintains a
greater involvement than in (37a), but the focus is on the pig. 8 In (37c), yááy-
am ‘his mother’ is involved only because she cannot kill the pig, which was
killed then by someone else. Note that in

both the BENEFICIARY and the PATIENT can have pronominal expression as
the INSTRUMENT with -ee- could in (29). The BENEFICIARY and the
INSTRUMENT differ in that the INSTRUMENT can also exploit the V_1 position
(cf. [37b] and [37c]), a possibility denied to the INSTRUMENT (cf.[25b] and
[25c]).9

8 Note here that both speakers accept the alternate orders of (37b) and (37c), whereas there
was some disagreement with respect to parallel expressions with -ee-. Cf. (25b) and (25c)
and footnote 6.

9
16 SYNTAX & SEMANTICS

3.3 Summary of ROLES

Overall, the PARTICIPANT relations have a graded relationship to the
EVENT:

Of the relations which follow the EVENT in the grammatical expression, the
RECIPIENT seems to be the ROLE most closely associated with the EVENT,
followed by the PATIENT. The V_2 expressions of the INSTRUMENT and the
BENEFICIARY approximate the semantic relation of the PATIENT , with the
BENEFICIARY being perhaps slightly more bound to the EVENT.10 The
RECIPIENT (e.g. [23]), BENEFICIARY (e.g. [24]), and INSTRUMENT (e.g. [25a])
lie in the next tier. They are all related by prepositions to the EVENT, but they
may also occur with a tighter bond (with the appropriate affix, -ee- or -al-).
Finally, there are those relations which give no evidence of having the status
of ROLE; they are only expressed by preposition, e.g. expressions of Time,
Location, and the Comitative.11
There are further observations that can be made about the organization of
Figure 2. Grammatically, there is an opposition between expression of
PARTICIPANTS with a preposition and without. There is an opposition between
the expression of a pronominal PARTICIPANT by verbal suffix and by a
pronominal shape following a preposition. This dichotomous opposition
suggests a binary opposition of this sort:

10 Recall the order options available to -al- which were not possible for -ee-.

11 Cf. also section 5.1.1, which suggests a PERIPHERAL relation for TIME and PLACE .
 FOCUS in Wolof 17

The fact that a PATIENT may be expressed along with the BENEFICIARY
created by -al- as in (37), while no PARTICIPANT could accompany the
PATIENT created by -ee- from ditransitive verbs in (27), suggests that there
may be additional organization to the NULCEAR ROLES. The V_1 PATIENT
created with -ee- and ditransitive EVENTS is more tightly entwined with the
EVENT than is the V_1 BENEFICIARY created by -al-. The former does not
admit other PARTICIPANTS to be in the NUCLEUS (nor perhaps in the
PROPOSITION at all), while the latter allows another PARTICIPANT in the
NUCLEUS but does not require them; for example, (39) has only two identified
PARTICIPANTS , not three as in the English gloss:

This suggests a scale of ROLES in the V_1 position.

4. Varieties in aspect and time

To this point, most of our examples have been cited in the past perfective
form of (3). But other aspects and times exist in Wolof. Having described the
outline of a Wolof simple sentence, we can now turn to some of the variety in
the encoding of Aspect, which exploits this syntax.

4.1 Imperfective
There is an imperfective that is said to be “present or past”:
18 SYNTAX & SEMANTICS

When the inflection appears not on di-, but on another constituent because of
the requirement to express a FOCUS , the shape don appears:

Compare the forms of (40) to (46) and (56) below.

The following forms illustrate a contrast between di and don: 12

with di being present (and past? Cf. [40].) imperfective, and the don being just
past time imperfective.13

4.2 Future time

Note that the forms in (40) lack a suffix encoding the AGENT wáá-dyi ‘the
man’. An inflection for AGENT can be added, and the effect is “the real
future”:

12 Although the two seem parallel in (42), they differ in that don has been to this point
always an uninflected form. Cf. the co-occurrence of the two forms in (43c).

13 The use in (40) and (43) suggests the existence of questions parallel to (41a):

(i) kan moo nyu di door

[who us hit]
‘Who is hitting us?’
 FOCUS in Wolof 19

In this regard, re-examine these expressions based on

In (44), the clause ‘I be calling him/her’ is heard as past

time because of the past context ‘When you arrived ...’ In
another context, the same form could be heard as ‘I am calling him/her’.

4.3 Past Time

A past time sense is commonly expressed with -oon (with a varying first
consonant):
20 SYNTAX & SEMANTICS

5. FOCUS in Wolof
As noted in (1) and (2), word order in Wolof appears to be SVO. The
behavior of questions and responses to them reveals another pattern in which
FOCUS is grammatically encoded in sentential initial position.

5.1 The expression of FOCUS in PARTICIPANTS

Consider the question of (46) and its answer:

The response to this question is neither sentence in (47), but (48) or (49)14:

14
 FOCUS in Wolof 21

While the sentences of (47) are valid Wolof utterances, they do not properly
respond to the question of (46). Only those in (48) and (49) do. The question
of a post-verbal PARTICIPANT takes this shape:

and its answer is not (51), but either expression in (52):

 22 SYNTAX & SEMANTICS

The sentences of (47) not only do not answer the who-question of (46), they
do not answer the whom-question of (50). Only an expression that places the
content sentence initially and marks with with the proper inflection will
correctly answer (50).
Questions about content coded grammatically with Prepositions show a
similar pattern:

The assertion of (55) answers (53), not the SVO pattern of (54).
Two observations are primary here. Initial position is employed both for
the questioning content of a PROPOSITION and for the answering material, but
position alone is not the signal of this content. When the content has the
function of ‘subject’, the element mo ~ moo follows, and when the content is
‘object’ of Verb or Preposition, the content is followed by l Note that the
question forms also follow this pattern as well: kan mo ‘Who’ in (46),
Whom?’ in (50), and ‘To whom?’ in (53). Subject Nouns and
Pronouns share a form mo(o) when they function as FOCUS; other Nouns and
Pronouns use :15

15The paradigm of pronominal forms in this function is very similar to those in footnote 14:
 FOCUS in Wolof 23

‘S/hé kissed the boy’

(b)

The form which appears with FOCUS , is the third person marker of Noun
predicates. Cf. (4) above: g He’s a man’.

5.2 FOCUS in time and place

Both these markers (a mo following a Subject Noun answer and a
following a non-Subject Noun answer) indicate person and number of a
Subject PARTICIPANT, but other lexical material can be placed in initial
position without accompanying information of person and number:
24 SYNTAX & SEMANTICS

The alternative position for temporal content is the final position:

The difference is again one of FOCUS, with initial position being its
grammatical mark. In (57c), it is precisely last night that the speaker did not
sleep, implying that the speaker slept on other nights; but in (58), the speaker
may have been sleepless on other nights as well.
There are two ways of saying ‘I am in Houston’:

If there is a contrast between ‘I am in Houston, but you are in Dakar’, it is the

latter expression with the place in initial position that is more acceptable:

The association of initial position with the focussing of place seem is

confirmed by the question of (61) and its possible answer:
 FOCUS in Wolof 25

Only (63) sensibly responds to (60).16

5.3 Degrees of FOCUS

If the grammar of FOCUS is related to the use of Subject inflection, then
FOCUS has various grammars. In this section we further elicidate the grammar
of FOCUS and its semantics.

5.3.1 FOCUS on EVENTS

If agreement is the mark of FOCUS (in the context of sentence initial
position), we might expect that the agreement marks on EVENTS /Verbs to also
signal FOCUS, even though EVENTS does not appear in initial position. What
we find, however, is that sentences such as (1) and (2), and others on the S-
initial model with Subject inflection on the Verb do not behave as if they
contain FOCUS , at least not in the way we found it in sections 5.1 and 5.2. Let
us begin by looking at how such SV and SVO utterances behave in the context
of questions.
Given a question such as (64), we might expect that the SVO usserance of
(65) would be an appropriate answer:

16
26 SYNTAX & SEMANTICS

But it is not. Given the question ‘What did Sadibou do yesterday?’, the
expression of (65) is an insufficient answer. In response to (64), only (66) will
do. SV sentences such as (67)

(67)

may constitute answers to such rhetorical questions as

[should have-3rd.sg-3rd.sg cook yesterday]

‘Did someone cook yesterday?’

The inability of SV and SVO grammars to respond to questions suggests that

FOCUS of the sort identified by that diagnostic is absent from them.
When the context requires that the EVENT in a PROPOSITION be
highlighted beyond its normal condition, constructions such as the following
are used:

The expression is an alternative expression to b ‘S/he

killed it’. And it is the former that is appropriate to the contrast of (69),
‘release, not kill’ as opposed to ‘release’ alone. The grammar FOCUS in
EVENTS turns upon a root de, which has this paradigm for the persons and
numbers of Wolof:
 FOCUS in Wolof 27

Speakers comment on these forms as follows “‘We do not know what he [the
man in (69)] would do if he didn’t kill it ... sell it? ... uncertainty ... a lot of
possibilities ... Expecting the pig to be killed ... After you give information
first, then ( ” An alternative to (69) with exists:

and speakers are “More concerned with giving information than responding to
expectations ...” The extraordinary combining of FOCUS with EVENT is
frequently heard as offering explanations so that

is used “when you’re trying to explain a cause ... you’re supposed to be here
and you say ” as a reason for not being there. And

is more than ‘S/he is sick’; it asserts a frequent occurrence or a characteristic

of the person. To express just the observation that ‘It’s raining’ or ‘S/he’s
sick’, the forms of (74) are used:

17 A contraction exists.
28 SYNTAX & SEMANTICS

5.3.2 Degrees of FOCUS : Absence of S inflection

The previous section raises the question of “degrees” of FOCUS. Run-of-
the-mill SV and SVO grammars seem not to signal FOCUS in that such
sentences are not adequate answers to wh-questions. While not containing a
degree of the concentrated FOCUS needed to respond to a pointed question,
they are suitable responses to more general inquires such as:

(75)

‘What’s going on?’

In answer to a query such as (75), that does not single out a particular
component of content, the SVO grammar of (76) is suitable:18

(76)
[what happen Sadibou.3rd.sg-NGI cook chicken]
‘What’s happening is Sadibou’s cooking a chicken’

The grammar of SV and SVO with the inflections of (3), (4), and (5)
contrast with certain alternatives which amplifies the presence of “degrees” of
FOCUS. Compare these:

[hit-1st.sg man-the man-the hit woman-the]

‘I hit the man; the man hit the woman’

[hit-1st.sg man-the man-the hit-3rd.sg woman-the]

‘I hit the man; the man hit the woman’

18 Compare (76) with sentence (44) above. is a morphophonemic

variant of
 FOCUS in Wolof 29

The sentences of (77) contrast grammatically in that the (a)-version lacks the
inflection present in the (b)-version. The semantic difference is that in the (a)-
sentence, the man hitting the woman is somehow a consequence of ‘I hit the
man’. In the (b)-version, the two actions are disconnected: “It could be the
other way around. You don’t know what came first.” But in the first
expression, it is an “action following an action.” Similarly, in (78), the
showering, dressing, and eating unfolded in that order, but in (78b), it is
“some actions you did ... not in a linear progression. I did these acts, I don’t
care how they happened.” Because the missing inflection implies a necessary
sequence, the expression of (79a)

[when 1st.sg-get.up eat.breakfast-1st.sg. dress shower]

[when 1st.sg-get.up eat.breakfast-1st.sg. dress-1st.sg.

shower-1st.sg]

strikes the Wolof ear as distinctly odd. One just does not get up, eat breakfast,
dress, and then shower in that order. Sentence (79b) is better since it is more
of a recitation of what I do when I arise in the morning, among which are ...
“It just lists things that the person did without specifying the order.”
Confirming this are sentences such as (80a):

[Awa steal book-the]

‘Awa stole the book’
30 SYNTAX & SEMANTICS

[Awa steal-3rd.sg. book-the]

‘Awa stole the book’

Sow remarks that (80a) is “like you’re telling a story”, and Deme says that
“There is always something before or after.”
The sentences of (81) continue the semantic contrast of inflection and its
absence:

[know-1st.sg woman-the see boy-the]

‘I know the woman who saw the boy’
*‘I know that the woman saw the boy’

[know-1st.sg woman-the see-3rd.sg boy-the]

‘I know the fact that the woman saw the boy’
*‘I know the woman who saw the boy’

The inflectionless EVENT in (81a) must necessarily follow and be linked to

what precedes. What precedes is dyigén-di ‘the woman’, and the result is the
equivalent of a relative clause. In (81b), the inflection lends greater
separation: ‘I know (something) ... the woman hit the boy’. And the English
equivalent is ‘I know that the woman hit the boy’. Comparing the grammar of
dyigen-dyi gis xále-bi in (81) with Awa saty teere-bi in (80a), we may
conclude that Wolof does not have a syntax dedicated solely to expressing the
sense of ‘relative clause’; it simply allows the close semantic connection
associated with the lack of Subject inflection. Thus in (82)

[see-1st.sg. dog-the bite child-the]

‘I saw the dog that bit the child’
‘I saw the dog bite the child’

I can have seen the dog and it bit the child, or I can have seen the dog biting
the child. I certainly saw the dog, but what else I personally saw is vague.
How that vagueness is resolved in real life gives one or the other English
 FOCUS in Wolof 31

glosses.19 The EVENT xam ‘know’ in (81a) is not one that supports a gloss like
the second one in (82), so there is the single ‘relative clause’ gloss.20

5.3.3 Degrees of FOCUS : Prefixed S inflection

When the propositional content of a clause is mated with a preceding
clause that contains a PARTICIPANT that would be the same PARTICIPANT as
the Subject of the following clause, Subject inflection can be absent from the
second clause, e.g. matt xále-bi ‘(it) bite the child’ in (82), That is, the O of
gís-na is the S of matt xále-bi. Null S inflection requires that the S be present
nearby, e.g. in the preceding clause or elsewhere in the close context. That
proximity is necessary but not sufficient for null inflection because prefixed S
inflection can also appear when the condition of an immediately present S is
satisfied. We will see below that the three grammatical patterns contrast: null
S inflection, prefixed S inflection, and suffixed S inflection.
When the condition necessary for null S inflection is not satisfied, then
overt S inflection must occur, but the inflection is not always the suffixed kind
from (3). It can be prefixed, and any pronominal O will be prefixed as well:21

19 The potential vagueness is also present in (i):

20 Such inflectionless utterances as (80a) are then undetermined out of context, e.g. the
expression of (i)

(i)

21 Although both the S-inflection and the O-inflection seem to be prefixed in (83), if they
occur together, the O-inflection continues in its position as a suffix:

(i)
[hit-1st.sg-2nd.sg. 2nd.sg.-hit-3rd.sg.]
‘I hit you and you hit him/her’

(ii)

[hit-1st.sg-2nd.sg. hit-2nd.sg.-3rd.sg.]

Sentence (i) “insist[s] on the succession” and is used when “Something is already
32 SYNTAX & SEMANTICS

S-Inflection O-Inflection

Keeping in mind the necessary requirement that the PARTICPANT which is

the S of the second clause must be immediately present elsewhere if null S
inflection is to occur, compare these four sentences:

[want-1st.sg. go]
‘I want to go’

done.” Placing both the S- and the O-inflection before the verb is unacceptable.
The semantic conjoining of (i) requires the prefixed Subject inflection, but in (iv)

[hit-1st.sg-2nd.sg. but hit-2nd.sg.-3rd.sg.]

‘I hit you, but you kissed him/her [anyway]’

the two propositions are opposed and the suffixed Subject inflection is appropriate, and the
prefix is not:

(v)
 FOCUS in Wolof 33

[want-1st.sg. 2nd.sg.-go]
‘I want you to go’

but not

(c)
[want-1st.sg. 1st.sg.-go]

(d)
[want-1st.sg. go-1st.sg.]

Sentences “(c) and (d) are completely wrong.” In (84c), S of is

present in the preceding clause, but because the connection between the two is
of the tighter sort, the S prefixed inflection of (84c) will not pass, nor will the
suffixed S inflection of (84d). Where the S of dem ‘go’ is not present in the
preceding clause, the necessary condition for null inflection is missing, and
the prefixed inflection occurs, e.g. (84b).
In PROPOSITIONS bound in a looser semantic fashion, S-inflection is
necessarily present, and it is prefixed. The “tightness” is found when there is a
Subject present and when there is a sense of ‘not-one-without-the-other’
between the PROPOSITIONS . Sentences (77a), (78a), (79a), (80a), (81a), (82),
(83), and (84a) are examples of this as are the expressions of concentrated
FOCUS in (48), (49), (66), (70), etc. These are all examples of the most tightly
bound connection between PROPOSITIONS , and complementarily, examples of
the least presence of FOCUS. Sentence (84b) illustrates the lesser degree of
“tightness” prompted by an independent Subject, but the looseness can
nevertheless be present when there are PARTICIPANTS in common:
34 SYNTAX & SEMANTICS

Sadibou is present in ‘I gave Sadibou the book’ and again in ‘He gave it to
Deme’, yet unlike earlier examples where such a common PARTICIPANT
elicited null S inflection, here prefixed S inflection is present. And that is
because the connection between the two PROPOSITIONS is a looser ‘in order
to’, which is equivalent in its ‘looseness’ to a non-shared PARTICIPANT across
PROPOSITIONS (as in [84b]). The ‘in order to’ relation facilitates the looser
relation between the clauses.
The form ndax occurs in sentences with the meaning ‘because’, ‘so’, or ‘in
order to’:

[Aux-NEG 1sg.sg. go.out because rain-the]

‘I can’t go out because of the rain’

[hit-3rd.sg so 3rd.sg.-study]
‘Hit him/her so s/he’ll study!’

When ndax mediates the relation between two PROPOSITIONS , as in (86b), the
prefixal version of S inflection is present. The other two inflectional choices
do not work:

[hit-3rd.sg so study-3rd.sg.]

[hit-3rd.sg so study]

The sense of consequence imparted by ndax is also present in its absence, but
with prefixed S inflection:

(86) (e)
[hit-3rd.sg.-3rd.sg. 3rd.sg.-study]
‘S/he hit him/her so that s/he’ll study’

[hit-3rd.sg.-3rd.sg. 3rd.sg.-give-2nd.sg. book-the]

‘S/he hit him/her so that s/he would give you the book’
 FOCUS in Wolof 35

(b)
[my-mother give-3rd.sg.-1st.sg money

1st.sg.-buy ]
‘My mother gave me money for him/her to buy ...’

The temporal joining of propositional content also employs prefixed S

inflection:22

[be-you.sg come time I-call-3rd.sg.]

‘When you arrived, I called him/her’

But when ‘when’ is intended as a consequence, and not a time, then the
absence of S inflection is invoked to signal the meaning:

[do-1st.sg want/love-PST woman-the kiss-PST boy-the]

‘I loved the woman when/after she kissed the boy’

When the prefixed forms occur independently of a preceding utterance,

they impart an uncertain assertion, a conjecture that is perceived as a question:

[1st.sg-call-3rd.sg.]
‘Shall I call him?’

[1st.sg-tell.story-AL-2nd.pl.]
‘Can I tell you a story?’

[3rd.sg.-speak]
‘Should s/he speak?’

22 The sense of ‘when’ is also conveyed by the form bi, which is followed by Verbs with
prefixed S inflection:
36 SYNTAX & SEMANTICS

[3rd.sg.-give-2nd.sg. book-the]
‘Should s/he give you the book?’

[2nd.sg.-go or NEG .2nd.sg. go]

‘Should you go or not?’

There are uncertainties that are felt as wonder. In (91),

(91)
[2nd.sg.-beat-3rd.sg.]
‘You dare beat him!?’

the sense is both a question and a challenge to the fact: incredulity, but not
assertion. These contrast with the given, unasserted free standing utterances
with null S inflection, and the free statements with suffixed S inflection.
The Verb báyi ‘release/let go’ occurs both with following EVENTS that are
null S inflected and prefixed inflected:

[release/let.go-IMPERATIVE Philip 3rd.sg.go]

‘Let Philip go!’

(b) bayi-l Philip dem

[release/let.go-IMPERATIVE Philip go]
‘Let Philip go!’

Sentence (92a) “implies that you are preventing ...” Philip from going and a
greater oomph is required to get him released, while (92b) “Just means I’m
suggesting somebody else ... suppose I want you to go somewhere and you
think I’m not the best person ... suggest somebody else”. This seems to be
close to ‘Let P. be the one to go!’.

6. Conclusion
Wolof employs a grammar that scales the sense of contingency from lesser
to greater, paralleled by a sense of lesser assertion to greater. The grammar
which signals the scale exploits Subject inflection in various ways. Cf. Figure
4. In (a), one’s purview is distributed across content that exceeds a single
 FOCUS in Wolof 37

PROPOSITION. FOCUS is maximally diluted in (a), but as one moves from (a)

Lesser contingency and assertion

(a) No inflection on the Verb

(b)Prefixed inflection on the Verb
(c) Suffixed inflection on the Verb -
(d) Inflection elsewhere

Greater contingency and assertion

Figure 4: The scale of FOCUS in Wolof.

to (d), one’s field of view becomes progressively narrowed, and FOCUS is

more concentrated and more defined. As the FOCUS is compressed, the degree
of assertion becomes increasingly intense. One moves from referencing
propositional content in (a), to uncertain conjecture in (b), to statement in (c),
and finally to pointed assertion in (d). As one moves from (a) to (d), the
semantics of FOCUS is more and more formed and centered until in (d), it is so
compacted that it selects a minimal component of the PROPOSITION. FOCUS in
Wolof is then not one thing. It is a principle that comes in four degrees. 23
We noted in the discussion of Kinyarwanda especially, and elsewhere, that
the semantics of FOCUS, across languages, avoids association with the
equivalent of the AGENT ROLE. This disassociation is one of the empirically
abiding properties of FOCUS. Its semantics may gravitate to EVENT (as in Bella
Coola, Yogad, and West Greenlandic Eskimo), the less motile ROLE of
PATIENT (or the equivalent, as in Kinyarwanda), but it never settles on the
AGENT. Wolof then poses an interesting potential conflict. Looking at
sentences such as (1) above, we found that sentence initial word order was one

23 This interpretation generally conforms with a received analysis of Wolof that finds five
“sous-modes” in the “indicatif affirmatif” forms of the verb (N’Diaye Corréard 1989.179).
Our (a) is there labelled “le narratif”; our (b) is “l’énonciatif”; our (c) appears to be the “le
présentatif”; and our (d) is “l’emphatique du verbe”. Our FOCUS on the S and other
PARTICIPANTS is apportioned between “l’emphatique du sujet” and the “l’emphatique du
complément”. It is not completely clear that the form in (c) is correctly equated with N’Diaye
Corréard’s “présentatif” since that sub-mode is illustrated with forms we have introduced in
(5) above.
Of the presentative, N’Qiaye Corréard (1989.180) writes approvingly, “Quant à la forme
Pr[ésentatif], elle [Robert 1984] lui attribue une fonction toute différente: celle de situer
l’agent du procès dans l’espace du sujet énonciateur.” While the terms labelling these sub-
modes are suggestive of our interpretation, e.g. “narratif” for (a) and “présentatif”, there
seems to be no integration of the variation in inflectional choices of the sort proposed here.
38 SYNTAX & SEMANTICS

of grammatical marks of the Wolof AGENT, and the “normal” order seemed to
be SVO. The conflict may seem to arise when we add the observation that
sentence initial position is also the mark of the most intense degree of FOCUS.
These two observations do not, however, result in the association of FOCUS
with AGENT ... for several reasons. The SVO order we observed in (1) exists
in those expressions in which FOCUS has less than its maximum degree. There
is no FOCUS concentrated on any one component of those sentences, and
FOCUS seems to be distributed equally throughout. It is a semantics of the
clause as a whole. FOCUS is not sentence initial when the AGENT is sentence
initial in these sentences. Second, when FOCUS is sentence initial and
associated with some PARTICIPANT, the expression actually consists of two
grammatical clauses. 24 The first clause, which signals the FOCUS, consists of
the FOCUSED content in combination with the appropriate inflection, e.g.
, ‘It’s the woman’.25 This may be followed by a second clause,
with null inflection, e.g. fon xále-bu-góór-bi ‘kissed the young man’

(93) dyigén-dyi móó fon xále-bu-góór-bi

[woman-the kiss young-MODIFY-man-the]
‘It’s the wóman who kissed the young man’
‘The wóman kissed the young man’

It is true that FOCUS is sentence initial in (93), and it is also true that the
FOCUSED person who performs the act of kissing is also sentence initial, but
FOCUS and AGENT (S) do not merge because they are in separate clauses. The
FOCUSED is, in its clause, not AGENT because there is no AGENT
relation in . The FOCUSED content ‘woman’ is linguistically
coreferential with and is, in the real world, the same person as the
unmentioned AGENT in fon xále-bu-góór-bi. But there is no initial AGENT in
(93).
Unlike some of the languages we have examined to this point, Wolof does
not preferentially associate FOCUS with any one particular component of a
PROPOSITION. Wolof is like Haida in not selecting some one propositional
component as the favorite son of FOCUS . But Haida did not otherwise employ
position (initial or otherwise) as a consistent mark of AGENT (S) as Wolof
does. Haida was more sensitive to the balance between the semantics of the

24 N’Diaye Corréard (1989.180) and others see these as “des énoncés complexes composés
d’une proposition nominale suivie d’une proposition verbale ...”

25 See the paradigm in footnote 14.

 FOCUS in Wolof 39

PARTICIPANTS and the functions to which they were put. When the
combinations were not “normal” for Haida, initial position was invoked as a
mark of that unusual circumstance.26 Haida and Wolof are also similar in
ranking the presence of FOCUS so that it comes in degrees,yet Wolof differs
from Haida in merging the semantics of FOCUS with the semantics of
ASSERTION so that the answers to wh-questions (our first diagnostic of FOCUS )
is just the extreme of a scale of ASSERTION. The association of FOCUS with
ASSERTION in Wolof further enriches our general concept of FOCUS by adding
to the (dis)associations between FOCUS and other semantics observed in
Haida, Kinyarwanda, etc.

26 Recall that Haida employs ¢uu following sentence initial content to signal the maximum
presence of FOCUS. Otherwise, Haida places content in sentence initial position without ¢uu,
following a principle that recognizes the greater “surprise” in the packaging of certain
content with certain functions, for example, combining a less ‘potent’ PARTICIPANT acting
with the function of AGENT to affect a more ‘potent’ PARTICIPANT acting as OBJECT . In Haida,
one expects the reverse, and when the unexpected happens, the language signals it with a
FOCUS lesser than that conveyed by ¢uu, i.e., simply initial position for the less ‘potent’
PARTICIPANT.