1This paper discusses two particles in Bambara – dè and dɛ́. As will be argued below, these two particles have parallel functions, namely they are used to distinguish between two different types of scope of focus – constituent and operator focus.
2Before we proceed a terminological issue needs to be briefly mentioned. Since both particles are used for focus, I will refer to them as focalization particles. Calling them ‘focus particles’ would be equally accurate, however this term is widely used with a slightly different meaning. It is mainly used for English words such as also, too, only, etc. that are claimed to have special inherent association with focus (König 1991), but are not focus markers per se. In English these particles have scope over the focused constituent, which is highlighted by the sentence stress.
- 1 Henceforth in English examples and translations I will capitalize the word that takes sentence stre (...)
(1)
|
Only FRED1 regrets that he lost. (König 1991: 4)
|
3Thus in (1) only has scope over the subject Fred, which is in focus and is marked by the sentence stress.
4Unlike these examples, the particles dè and dɛ́ in my view are genuine focus devices, just like English sentence stress itself.
5The function of the two particles has been discussed and described by Masiuk (1986; 1994) and in (Dumestre 2003; 2011) and (Bailleul 2007).
6The connection between dè and focus is well established and accepted by all major sources. Bailleul (2007) describes dè as a “particule de mise en relief”, while Dumestre (1987) and Masiuk (1986; 1987) use the term “particule de focalisation”; Dumestre (2011) recognizes focalization as the main value of the particle.
7In contrast, the particle dɛ́, although frequently mentioned, has received much less attention and, to my knowledge, its connection to focus has not been proposed before. Bailleul (2007) describes dɛ́ as a “particule exclamative.” Dumestre (2011) defines its value as “intensive” and “exclamative.” Compare examples (2) and (3).
(2)
|
Intensive
|
|
à
|
ká
|
júgu
|
dɛ́!
|
|
3SG
|
QUAL
|
nasty
|
OF
|
|
‘He is very nasty!’ (Dumestre 2003: 321)
|
(3)
|
Exclamative
|
|
í
|
tɛ́na
|
táa
|
dɛ́!
|
|
2SG
|
FUT.NEG
|
go.away
|
OF
|
|
‘Don’t go away!’ (Bailleul 2007: 96)
|
- 2 (Masiuk 1994: 4): “…étant donné qu’elles [= “particules monovolantes”, incl. dɛ́] n’ont pas de va (...)
8Masiuk (1994) leaves dɛ́ (among other particles) without any discussion. In her remarks concerning those particles she says that they lack “proper semiotic value” and rely more on discourse mode and the individual language habits of speakers.2
9As argued in this paper, dɛ́’s basic function is the marking of contrastive truth-value focus and its other uses (like exclamative and intensive) can be derived from this basic function.
10Before we proceed with Bambara material, a brief introduction to the framework in which the further discussion is developed is necessary.
11My understanding of focus is in line with Dik (1989: 277):
… information that is relatively the most important or salient in the given communication setting, and considered by S [=speaker, KP] to be the most essential for A [=addressee, KP] to integrate into his pragmatic information.
12Focus typologies are usually built around several parameters that include the scope of focus (cf. Dik 1989; Lambrecht 1994; Kiss 1998). The scope of focus characterizes the entity the focus ranges over. In this paper the following classification of scope categories will be used (cf. Dik 1989; Güldemann 2009).
Table 1. Typology of the scope of focus
Term focus
|
|
Subject-focus
|
|
Non-subject focus
|
Predicate-centered focus
|
|
Lexical-verb focus
|
|
Truth-value focus
|
|
TAM focus
|
13Following Dik (1989) and Güldemann (2009), the main division here is drawn between term and predicate-centered focus. Term focus embraces the cases where the scope of focus ranges over a ‘term’ that is a non-predicative (e.g. nominal, adverbial) constituent. The term-focus domain is further divided into subject and non-subject. Predicate-centered focus on the other hand serves as a cover term for the focus types that are characterized by a focus scope over semantic components typically hosted by the predicate, such as the lexical meaning of the verb, truth value and TAM.
Figure 1. Constituent and operator focus
14Watters (2010), following Dik (1989), suggests two further meta-categories – constituent focus and operator focus. In constituent focus “the scope of focus ranges over the lexical constituents” while in operator focus it “ranges over any sentential operator” (Watters 2010: 355). Among “sentential operators” are truth-value or polarity, and tense, aspect and mood (TAM). Thus, the two categories cut across the distinction between term and predicate-centered focus as shown schematically in Figure 1.
15As we will see below, the distinction between constituent and operator focus is relevant for Bambara and shows up in the distribution of the focalization particles.
16Bambara lacks any distinction within the term-focus domain. The particle dè equally follows any focalized constituent, be it a subject, a direct object, a postverbal dative, an oblique or even a verb. Compare examples (4–6). Curly brackets before the translation are used to indicate the context.
(4)
|
Subject
|
|
Ámadu
|
dè
|
yé
|
sàgá`
|
fàga.
|
|
PN
|
CF
|
PFV.TR
|
sheep:ART
|
kill
|
|
{Who slaughtered the sheep?} ‘AMADOU slaughtered the sheep.’ (Prokhorov, f. n.)
|
(5)
|
Direct object
|
|
à
|
yé
|
sàgá`
|
dè
|
fàga.
|
|
3SG
|
PFV.TR
|
sheep:ART
|
CF
|
kill
|
|
{What did Amadou slaughter?} ‘He slaughtered the SHEEP.’ (Prokhorov, f. n.)
|
(6)
|
Verb
|
|
à
|
bɛ́nà
|
à
|
fàga
|
dè.
|
|
3SG
|
FUT
|
3SG
|
kill
|
CF
|
|
{What is he going to do with the sheep?} ‘He is going to SLAUGHTER it.’ (Prokhorov, f. n.)
|
17In (4) dè is used after the subject Amadu, thus marking focus on that constituent. In (5) it follows and marks focus on the direct object sàgá` ‘the sheep’. Finally, in (6), an answer to a question about the lexical semantics of the verb, dè follows the verb fàga ‘beat, kill, slaughter’, which is the focus of the sentence.
18As can be seen from these examples, morphosyntactically dè can be described as a “floating” particle with scope over the constituent immediately to its left. Since dè is used with all major constituent types including the verb, it seems plausible to describe the particle dè’s function as “constituent focus.”
19In this section, I argue that dɛ́ functions as a marker of contrastive truth-value focus, and that its other uses that have been noted in the literature (including intensive and exclamative ones) do not contradict this analysis.
20Notions of “intensivity” and “exclamation” fail to explain an important feature of the particle dɛ́ that the sources do not mention: the particle is sensitive to the truth value of the clause. Consider examples (7) and (8).
(7)
|
àyí,
|
à
|
má
|
nà
|
dɛ́.
|
|
no
|
3SG
|
PFV.NEG
|
come
|
OF
|
|
OK {Did Amadou come?} B: ‘No, he didn’t come.’
|
|
*{Amadou didn’t come.} ‘No, he didn’t come.’ (Prokhorov, f. n.)
|
(8)
|
à
|
nà-nà
|
|
dɛ́.
|
|
3SG
|
come-PFV.ITR
|
OF
|
|
|
*{Did Amadou come?} ‘(Yes), He did come!’
|
|
OK {Amadou didn’t come.}. ‘(No) He did come.’ (Prokhorov, f. n.)
|
- 3 Examples like (7) a (8) should not be taken the ultimate evidence for a grammatical constrain on dɛ (...)
21In question-answer pairs the particle dɛ́ is only allowed in answers that have a truth value which is the opposite of that of the question/stimulus (in curly brackets). Example (7) features the negative perfective auxiliary má. Because of this dɛ́ is allowed only in an answer to the positive-polarity question Did Amadou come?, but not as a reaction to a negative utterance Amadou didn’t come. In the same way an affirmative clause with dɛ́ (8) is not allowed as an answer to a preceding positive-polarity question, but can only be used as a contradiction to a negative utterance.3
22Examples like (7) and (8) constitute my main piece of evidence in favor of a definition of dɛ́ as a marker of focus on the truth-value operator, but there are some further facts that are in accordance with the explanation just proposed.
23It is crucial to mention that dɛ́ doesn’t occur in clauses with constituent focus (9) marked by the particle dè or in true (non-rhetorical) questions, either in WH- (10) or polar ones (11).
(9)
|
Subject focus
|
|
Ámadu
|
dè
|
yé
|
sàgá`
|
fàga (*dɛ́).
|
|
|
PN
|
|
CF
|
PFV.TR
|
sheep:ART
|
kill
|
|
{Seydou killed the sheep} ‘(No,) AMADOU killed the sheep.’ (Prokhorov, f. n.)
|
|
(10)
|
WH-questions
|
|
Ámadu
|
yé
|
mùn
|
kɛ́
|
(*dɛ́)?
|
|
PN
|
PFV.TR
|
what
|
do
|
OF
|
|
‘What did he do?’ (Prokhorov, f. n.)
|
- 4 The reverse ordering *wà dɛ́ is also ruled out.
(11)
|
Polar questions
|
|
à
|
má
|
sàgá`
|
nìn
|
fàga
|
(*dɛ́)
|
wà4?
|
|
3SG
|
PFV.NEG
|
sheep:ART
|
DEM
|
kill
|
OF
|
Q
|
|
‘Didn’t he slaughter a sheep?’ (Prokhorov, f. n.)
|
24What sentences (9–11) have in common is that in each of them there is either a specialized focus marker dè or an element which is inherently connected with (a certain type of) focus, viz. a WH-word, like mùn ‘what’, and the polar-question particle wà. The fact that dɛ́ is not used in these cases is in accordance with the hypothesis about dɛ́’s connection with truth-value focus. Since dè marks constituent focus, its incompatibility with dɛ́ is expected. The same is true of WH-words like mùn ‘what’ in (10), since WH-words in questions are in focus by default. In contrast, the incompatibility with the polar-question particle wà can be explained by wà’s inherent association with focus on the truth value, which would make the use of dɛ́ redundant. Alternatively, it can be argued that dɛ́ doesn’t occur in (non-rhetorical) polar questions because of its contrastive nature. This argument will be discussed below in Section 4.2.
25Concerning dɛ́’s relation to questions, it is also important to mention that there is one type of question, namely rhetorical questions, in which the use of dɛ́ is allowed. In my view this fact can also be explained from the basic assumption about dɛ́ as a contrastive truth-value focus particle. See Section 4.3 for details.
26Based on the data presented above I propose that dɛ́ functions as a marker of contrastive truth-value focus. We saw that dɛ́ is sensitive to focus and to the truth value of the clause, but its relation to contrast needs further demonstration.
27In my treatment of contrast I follow Zimmermann (2007: 154), according to whom:
Contrastive marking on a focus constituent α expresses the speaker’s assumption that the hearer will not consider the content of α or the speech act containing α is likely to be(come) common ground. (italics in the original)
28Zimmermann states this as the Contrastive Focus Hypothesis, which is primarily intended to capture the facts about contrastive term focus. Applying this definition to truth-value focus, this would mean that the speaker focuses on the truth value of the utterance (e.g. positive) because s/he assumes that the hearer holds the opposite value (e.g. negative) to be or likely to become part of the common ground. This explanation indeed fits well with the observation that dɛ́ cannot be used in question-answer or stimulus-reaction pairs to confirm the speaker’s assumption, but only to mark statements that contradict it.
29As has been noted above, the contrastive nature of dɛ́ can be evoked to explain the particle’s non-occurrence in true (non-rhetorical) polar questions. Such questions typically inquire whether or not the proposition stated is true (according to the hearer’s knowledge), and thus can be claimed to have an inherent truth-value focus. This truth-value focus, however, is not contrastive, since it doesn’t express the speaker’s assumption about whether a certain proposition is true or not true in contrast to the hearer’s assumption, but rather the speaker’s unawareness of the actual situation.
30If one accepts that the basic function of dɛ́ is to mark truth-value focus, both its “exclamative” and “intensive” readings can be explained from the point of view of this general assumption.
31In Sadock & Zwicky’s (1985: 162–163) approach, which I adopt here, an exclamative like a declarative statement “represents the proposition as being true,” but also emphasizes the speaker’s “strong emotional reaction to what he takes to be a fact.”
32This definition of exclamation is compatible with my understanding of dɛ́’s basic function as truth-value focus. Pragmatically a strong emotional reaction is appropriate when the speaker assumes that the content of the statement is not known to the hearer or at least the hearer doesn’t take this information to be relevant in the current speech situation. In other words the content of the statement is not part of the common background. This makes exclamation and contrastive truth-value focus very similar (cf. Table 2).
Table 2. Exclamation and contrastive truth-value focus
Pragmatic function of statement X
|
Contrastive truth-value focus
|
Exclamation
|
Emphasis on truth value of X
|
+
|
+
|
Information in X is not part of the common ground of speaker and hearer
|
+
|
+
|
Emphasis on speaker’s emotional reaction to what he takes to be a fact
|
–
|
+
|
33As shown in Table 2, the two categories differ only in the emotional aspect which exclamatives contribute to the speech act. Thus, in the case of dɛ́ it is plausible to explain the exclamative reading by pragmatic factors.
34The same can be shown for the intensive reading of dɛ́. As can be seen from the examples provided by Dumestre (2011) and Bailleul (2007), the intensive reading of dɛ́ is found where the particle occurs with quality predicates.
(12)
|
à
|
ká
|
júgu
|
dɛ́!
|
|
3SG
|
QUAL
|
nasty
|
OF
|
|
‘He is very nasty!’ (Dumestre 2003: 321)
|
(13)
|
à
|
ká
|
jàn
|
dɛ́!
|
|
3SG
|
QUAL
|
big
|
OF
|
|
‘He is very big’ / ‘What (a) big (man) he is!’ (Bailleul 2007: 96)
|
35It is reasonable to suppose that in these examples, the property concept denoted by the quality predicate gets intensified as a result of pragmatic reinterpretation of the focus on the truth value, as represented schematically in Figure 2.
Figure 2. ‘Truth-value focus → intensive’ reinterpretation
36The emphasis on the truth value of the proposition with a quality predicate expressed in (a) by the adverbial indeed is reinterpreted as intensification of the property concept denoted by the adjectival predicate very nasty in (b).
37Such a semantic development is well known from studies of grammaticalization of adjectival intensifiers. Thus, Lorenz (2002: 146–147) concerning the origin of the English intensifier very says: “it is derived from Latin verus through old French verai and Middle English verray, all with a modal meaning of ‘tru(ly)’ ‘truthful(ly)’.”
38The intensifier reading of dɛ́ has not been grammaticalized in Bambara since it is not used exclusively with adjectival predicates but has several other functions. It is hence safer to suppose that this is another pragmatic reading of dɛ́ as a truth-value focus marker.
39Some examples of the use of dɛ́ provided by Dumestre (2003; 2011) can be understood as rhetorical questions. Unfortunately he doesn’t give any context, but at least in the examples (14)–(16) the most natural interpretation seems to be rhetorical.
(14)
|
í
|
y'
|
à
|
dɔ́n
|
mùso
|
ní
|
jànfa
|
má
|
bán
|
dɛ́?
|
|
2SG
|
PFV.TR
|
3S
|
know
|
woman
|
with
|
betrayal
|
PFV.NEG
|
be.over
|
OF
|
|
‘You know (well), women don’t stop betraying, do they?’ (Dumestre 2011: 232)
|
(15)
|
wágati
|
má
|
sé
|
dɛ́?
|
|
time
|
PFV.NEG
|
arrive
|
OF
|
|
‘Wouldn’t it be the time (now)?’ (Dumestre 2003: 321)
|
(16)
|
kása
|
t'
|
à
|
lá
|
dɛ́?
|
|
Smell
|
be.NEG
|
3SG
|
LOC
|
OF
|
|
‘Doesn’t it have a strong smell?’ (Dumestre 2011: 232)
|
40Following Quirk et al. (1985) and Koushik (2005) I understand rhetorical questions as being “conducive,” that is, as showing that the speaker is predisposed to receive a particular answer to his/her question. In the case of rhetorical polar questions, this means the speaker’s predisposition to one of the two possible truth values of the sentence. In this sense rhetorical questions can be seen as carrying a strong assertion, which makes them similar in a way to sentences with contrastive truth-value focus. In both cases the truth value of the assertion is the most important information that the speaker wants to become part of the common background. Examples (14)–(16) conform to this definition.
41The rhetorical nature of (14) is clear since it starts í y'à dɔ́n ‘you know well’, which unequivocally shows the conduciveness of the following question. The question itself is a reference to a common (sexist) belief about the character of women. It is important to note that (14) refers to a “common truth” and the answer is assumed to be known by the hearer too, as all sexists share the same belief about the character of women. In my understanding however it is not this common truth itself that the speaker wants to utter, but rather the relevance of this common truth to the current speech situation. What the speaker really wants to say in (14) is that this woman or these women (not mentioned in the question itself) will cheat as all women do. That is why using a conducive, assertion-carrying question is appropriate here: the reestablishing of a “common truth” as being true in the common background activates its relevance for the current speech situation.
42Example (15) doesn’t refer to some “common truth” but rather to a single event. According to the speaker, the current reference time (now) is exactly the moment when this event should happen, but it has not happened yet. The speaker’s main intention is now to convince the hearer that the time for this event has come or, in other words, that the proposition this would be the time is true. To do this the speaker uses a conducive negative question with final dɛ́, which presupposes a positive-polarity answer. Similarly in (16), a negative question with dɛ́ at the end is used to elicit a positive-polarity answer. The speaker believes that the smell is really strong and wants this to become part of the common background.
43Thus, in my understanding the use of dɛ́ in rhetorical questions doesn’t contradict the assumption about its relation to truth-value focus, but only shows another possible pragmatic reading of the particle in addition to the exclamative or intensive uses discussed above.
44If one accepts dɛ́ as a marker of contrastive truth-value focus, its syntactic features can be understood as being parallel to those of the constituent-focus particle dè. Syntactically both particles can be described as elements with scope. Like other elements with scope over other constituents in Bambara, the particle dè immediately follows the element in its scope. Take as an example the relativizer mín (17):
(17)
|
Relative clause: relativized direct object
|
|
[í
|
bɛ́
|
cɛ́`
|
mín`
|
dɔ́n]
|
né
|
yé
|
ò
|
yé.
|
|
2SG
|
IPFV
|
man:ART
|
REL:ART
|
know
|
1SG
|
PFV
|
3SG
|
see
|
|
‘I saw the man that you know.’ (Prokhorov, f. n.)
|
45Interestingly, both the focus particle dè and the relativizer mín are insensitive to the constituent structure of the clause. Thus, both occur between a postposition and its complement in constructions with a postpositional phrase, as in (18) and (19).
(18)
|
PP focus
|
|
|
|
|
|
|
|
|
Ámadu
|
yé
|
wári`
|
dí
|
à
|
[dɔ́gɔ-muso dè
|
mà]
|
|
|
PN
|
PFV.TR
|
money:ART
|
give
|
3SG
|
younger-woman
|
CF
|
DAT
|
|
{Who did Amadou give the money to?} ‘Amadou gave the money to his younger SISTER.’ (Prokhorov, f. n.)
|
(19)
|
PP relativization
|
|
Kéyítà
|
tùn
|
yé
|
bàtakí`
|
cí
|
[móri`
|
|
PN
|
PST
|
PFV.TR
|
letter:ART
|
send
|
marabout:ART
|
mín`
|
mà],
|
ò
|
sà‑ra.
|
REL
|
DAT
|
ANA
|
die-PFV.ITR
|
‘The marabout who Keita sent a letter to is dead.’ (Vydrin 2008: 96)
|
46In contrast, the clause-final slot, where the particle dɛ́ is found, is occupied by particles that have scope over the truth value of the clause, as for example the polar-question particle wà (21).
(20)
|
yíri`
|
bɛ́
|
bìn sísan
|
wà?
|
|
tree:ART
|
IPFV
|
fall now
|
Q
|
|
‘Is the tree going to fall now?’ (Prokhorov, f. n.)
|
47In this paper I have argued that the particle dè can be described as a constituent focus marker, while dɛ́ is a contrastive truth-value focus marker. The exclamative, intensive reading of dɛ́ and its use in rhetorical questions found in the literature can be derived from dɛ́’s basic truth-value-focus function. The syntax of the two particles is parallel in that both occur in a position that is typical for scope elements of their type. Like elements with constituent scope (e.g. the relativizer mín), dè follows the constituent, while dɛ́ occurs in the clause-final position typically occupied by elements with scope over a clausal truth-value operator (like the polar-question particle wà).
48In this view Bambara’s focus system constitutes an example of a focus alignment with a basic distinction between constituent and operator focus (recall Figure 1), thus lending further support to the relevance of these categories in the typology of the scope of focus.