Abstract and Keywords
This chapter provides an overview of the common syntactic features as well as the syntactic microvariation found in the Bantu languages. It particularly highlights the importance of information structure for the analysis of morphosyntax in this language family: word order, valency, voice, tense-aspect marking, subject and object marking can all be influenced and affected by the information structure expressed in the sentence. The chapter furthermore shows how Bantu languages, despite their shared basic SVO word order, noun classes and extensive verbal morphology, display a remarkable variation in the conditions determining agreement relations and word order. This has influenced syntactic theory formation in the past and should continue to do so now that more data and analyses of Bantu syntactic phenomena become available.
The name “Bantu” refers to a subfamily comprising between 500 and 680 languages, spoken by some 240 million people in sub-Saharan Africa (Nurse and Philippson, 2003: 2). The Bantu languages are conventionally classified by a letter and a number (e.g., Ndebele is S44), the letters referring to geographical zones. This reference system started in the pioneering work of Guthrie (1967–1971) and was revized lastly by Maho (2009). In referring to Bantu languages, we often find two names for the same language: an “English” name (like Swahili) and a prefixed name in the language itself (like Kiswahili). In this chapter I use the most familiar name for each language, and indicate its Guthrie number.
Before introducing the syntactic structure of Bantu languages, two caveats must be indicated. First, there is still a considerable number of languages that remain to be documented and receive a full linguistic description. Second, the research tradition in Bantu linguistics has typically focused on morphophonology, tonology, and historical linguistics, which means that syntax is not a point of focus in many descriptive grammars. Nevertheless, Bantu syntax is a growing field with, as we shall see, a number of phenomena that can influence theoretical debate.
Although one of the main points of interest is the fascinating variation in the family, a couple of generalizations can be made with respect to Bantu morphosyntax.1 First, a typical Bantu language has noun classes, indicated by numbers. Which class a noun belongs to is visible in the prefix on the noun itself and in the concord and agreement triggered on other elements in the NP and clause. To illustrate, in (1) watoto “children” is in noun class 2, as evidenced by the prefix wa-, the concord w(a)- on the modifier -ote “all,” and the agreement on the verb as the subject marker wa-.
Second, being agglutinative in nature, the verb consists of a root with optional derivational suffixes and inflectional prefixes, ordered as in the simplified template in Table 1 (Meeussen, 1967). In (1a), the verb is -fundisha “teach,” from which the passive is derived by the suffix -w-. It is inflected for tense/aspect (-na-) and noun class of the subject (wa-). Subject marking is obligatory, and for object marking there is variation in the number of prefixes and their optionality. Third, Bantu languages are so-called pro-drop languages: the subject DP can be omitted, as seen in (1b). Fourth, the canonical word order in Bantu is SVO.
Table 1 Slots in the Bantu verb
appl, pass, caus, etc.
With these basics in place, we can move on to more specific issues in Bantu syntax. This chapter does not intend to provide an overview of all aspects of the syntax of Bantu languages (for which see Bearth (2003) and Buell (forthcoming), but instead aims to highlight current areas of interest and recent developments in the field of Bantu syntactic research. Specifically, I will concentrate on two main points: 1. The (micro)variation within this family (cf. Marten et al. (2007); 2. Information structure, which I claim is crucial to understanding Bantu syntax. The first point will become clear from the crosslinguistic data used throughout the chapter. The second point will be argued for in Section 2, which considers the strong link between morphology and syntax, and Section 4, which looks at word order, while section 3 discusses valency and objecthood. These sections will be mainly descriptive. Section 5 addresses the theoretical implications of the issues raised in the previous sections, and section 6 concludes.
2. Information structure in morphosyntax
Because of their agglutinative nature, syntax in Bantu languages is always on the boundary with morphology. In this section, I show that all aspects of verbal inflection in Bantu can be affected by information structure. Specifically, after briefly introducing the concept of information structure, I consider in 2.2 the interaction of focus and tense-aspect marking, and in 2.3 I discuss the relation between topicality, givenness, and subject/object marking.
2.1. Information structure
Information structure concerns the packaging of information to facilitate the hearer’s processing of the information.2 This involves firstly the presentation of information as given or as new to the conversation: if a piece of information is already active in the hearer’s mind, it is highly accessible and can easily be referred to by the speaker, whereas new information needs to be presented and activated in the hearer’s mind. The referent that the sentence is about (generally the accessible or given information) is referred to as the “topic” of the sentence; the rest of the sentence adds information about this topic and is hence called the “comment.” For example, if we are already talking about Calvin and want to know what he did, a felicitous sentence like “Calvin conquered the tree house with Hobbes” has Calvin as the topic.
Secondly, information structure also involves the highlighting of information that the speaker wants to come back to, or wants to contrast with implicit or explicit other information. This is known as the “focus” of the sentence. Various types of focus have been proposed, the most important of which are “new information focus,” which presents a referent as new to the discourse, and “contrastive focus,” which explicitly contrasts the focused referent with a set of relevant alternatives. For example, in “CALVIN beat up Hobbes, not Suzie,” “Calvin” is contrastively focused.
2.2. TAM morphology and focus
Traditionally, finiteness and inflection have been associated with the three notions of tense, aspect and modality (TAM), that is, the grammatical(ized) expression of, respectively, time, the flow of time, and the necessity, obligatoriness, or probability of the event. By now we start to realize that this is too restrictive, as TAM marking often encodes more than just “tense-aspect-mood”: evidentiality—expressing the source of information—and information structure can also be expressed inflectionally. Crosslinguistically, not all languages have a grammaticalized expression of information structure (or TAM or evidentiality), but sometimes the morphosyntax is clearly affected by information structure. This is precisely the topic of Hyman and Watters’s (1984) article called “Auxiliary focus,” which relates focus to TAM marking, polarity, and predication (see also Güldemann, 1996, 2003). This section presents three cases of morphological markings that combine focus and TAM in Bantu.
The first case is the focus marker as reported for Kikuyu (Bergvall, 1987; Schwarz, 2007) and Kîîtharaka (Abels and Muriungi, 2008). This prefix n- orginates from a copula ni but now functions as a focus marker for both term focus (DP), as in (2a), and predicate focus (V/VP), as in (2b).
Kîîtharaka (E54, Abels and Muriungi 2008: 690, adapted)
The fact that the focus marker can be a prefix to the verb already shows how information structure is connected to morphosyntax in Kîîtharaka. Additional evidence for the role of information structure in the syntax of this language comes from the interaction of focus marking with TAM inflection. Firstly, the occurrence of the focus marker is restricted to certain tenses. Abels and Muriungi (2008: 697) remark that “on the one hand, there are tenses that FOC cannot co-occur with. These are the present perfect tense, and the future. On the other hand, the present progressive requires FOC.” Secondly, the shape of the present-tense prefix depends on the presence or absence of the focus marker, as illustrated in (3) and (4). The prefix has two allomorphs: -rî- and -kû-. When the postverbal in-situ DP is in focus, no marker is needed, and the present-tense marker -rî- should be used. When postverbal focus is impossible, the focus marker must show up, and only the allomorph -kû- can be used.
Kîîtharaka (E54, Abels and Muriungi, 2008: 697)
A similar instance of morphological marking of informational status are the so-called tone-cases found in western Bantu languages (cf. Blanchon, 1998, 1999; Kavari et al., 2012; Schadeberg, 1986). It should be made clear that, as minimal form-meaning correspondences, tones are taken to be part of morphology here. In the tone case systems, in a subset of tenses the tonal shape of a DP is determined by its syntactic function and/or its position in the clause. Compare in Otjiherero the default case (D) of otjihavero “chair” as a subject in (5a) with the complement case (C) as an object in (5b).
Otjiherero (R30; Kavari, Marten and Van der Wal 2012: 318)
When a DP is the first object after the verb, the tonal form is generally determined by the tense of the verb; for example, it is always default case in the present tense and complement case in the Remote Past Perfective. However, following a verb in the Negative Factive Habitual, either case is possible. The choice here depends on information structure: the complement case triggers a focus reading of the object (6a), which is absent for the default case (6b).
Otjiherero (R30; Kavari, Marten and Van der Wal, 2012: 325)
The Otjiherero tone cases again show us that morphology can be dependent on the interaction of TAM and information structure.
A third and final case of the interaction of TAM, morphology, and information structure is the conjoint/disjoint alternation. Some eastern and southern Bantu languages have an alternation of verbal conjugations that differ in their relation with what follows the verb. One verb form is called “conjoint” (7a), indicating a close relation between verb and following element; the other form is called “disjoint,” (7b) indicating a looser relation.
Kirundi (JD62, Ndayiragije 1999: 406)
The conjoint form cannot appear at the end of a main clause, whereas the disjoint form can but need not. When some element follows the verb, the choice between these verb forms is related to focus. The use of the conjoint verb form is associated with a focus interpretation of the postverbal element, which is absent for the disjoint form. To illustrate, in (8) the inherently focused question word kae “where” can only appear after the conjoint form.
Northern Sotho (S32, Poulos and Louwrens, 1994: 212)
What is important to note here is that this alternation is only present in a subset of tenses (e.g., in Makhuwa, only the four tenses given in (9)). Furthermore, the morphological marking of this alternation is close to, or even merged with, the TAM morphology on the verb. In Makhuwa, for example, it is difficult to separate a particular morpheme in the inflected verb that relates directly to either conjoint or disjoint (9). See Van der Wal and Hyman (forthcoming) for an overview of the conjoint/disjoint alternation in Bantu.
The picture is more complex than just sketched, and we will get back to it partly in section 3, but the key point here is that focus can influence verbal morphology and affect the TAM marking.
In a canonical sentence, the subject appears in preverbal position and determines obligatory subject agreement on the verb. In (10a), the subject alendôwo “those visitors” is in class 2, as is the subject marker a- on the verb. But many Bantu languages have one or more constructions in which the logical subject appears postverbally, and, in some of these constructions, subject agreement is not with the logical subject but with a preverbal element. In the locative inversion construction in (10b), the subject marker agrees with the preverbal locative in class 17.
Chichewa (N31, Bresnan and Kanerva, 1989: 2)
In all subject inversion constructions, the postverbal subject is non-topical (either presented as new information or in narrow focus). The “anti-agreement” on the verb has hence been connected to topicality, directly or indirectly. In a direct way, the “subject” agreement is sensitive to which DP is topical (see next paragraph). In an indirect way, it has been proposed that “subject” agreement is determined by the direction of agreement, which for Bantu would only be “upwards,” that is, with the preverbal element (Baker, 2008; Carstens, 2005; Collins, 2004). Which element is in the preverbal position is in turn determined by topicality.
Morimoto (2006) proposes on the basis of so-called subject-object reversal constructions that “subject agreement” is really “topic agreement” in some Bantu languages (see also Bokamba 1979, Whaley 1996). She argues that the preverbal theme in (11b) is not the syntactic subject of the sentence, nor is the postverbal agent the object—there is no real role reversal. The theme nevertheless triggers agreement on the verb, which is explained if agreement is sensitive to topicality: the preverbal theme is indeed the topic (just as the subject is the topic in an SVO sentence).
Kirundi (JD62, Ndayiragije, 1999: 418)
There is, however, more cross-Bantu variation in inversion constructions (see among others Bokamba, 1979; Creissels, 2011; Demuth and Mmusi, 1997; Diercks, 2011; Khumalo, 2012; Marten, 2006, 2011; Marten and van der Wal, 2015; van der Wal, 2008, 2012; Whaley, 1996; Zeller, 2012a; Zerbian, 2006b). For locative inversion, it has been established that the preverbal locative behaves like a true syntactic subject in many ways (see the overviews in Salzmann, 2011 and Diercks, 2014).
Unlike subject marking, object marking in Bantu languages is not always obligatory. The presence or absence of object marking has also been shown to correlate with information structure in some Bantu languages. This is again the case in a direct and indirect way. Direct influence of information structure is reported, for example, by Seidl and Dimitriadis (1997), Bax and Diercks (2012), Diercks and Sikuku (2015). The influence of information structure can be seen in so-called doubling, which concerns the co-occurrence of an object marker with a co-referential lexical object (which is not dislocated),3 as in (12a). This can be contrasted with (12b), where the object marker appears by itself, without doubling a lexical object (resulting in a pronominal reading).
On the basis of corpus research, Seidl and Dimitriadis (1997) show that, in Swahili, hearer-old referents are significantly more often object marked than hearer-new referents, and that this factor correlates more strongly with object marking than definiteness. Similarly, Bax and Diercks (2012) show that in Manyika, object marking triggers a non-focus interpretation of the doubled object DP. This is illustrated in (13), where (a) is felicitous when the verb, the object, or the VP is in focus (as diagnosed by a contextualising question), whereas (b) is only felicitous when the object is not included in the focus. Object marking is in these cases directly dependent on the information-structural status of the object.
Manyika (S10, Bax and Diercks, 2012: 191)
Object marking can also relate to information structure in an indirect way. If the object marker functions as a pronoun,4 similar to (12b), it is required when the coreferring DP is dislocated. Dislocation is in turn triggered by information structure, as has been argued for Zulu and other Southern Bantu languages. Van der Spuy (1993) shows that dislocated objects in Zulu require an object marker on the verb, as in (14). The dislocation of the object “bicycle” is reinforced by the clear prosodic boundary between the verb and the object. Interestingly, a DP can undergo dislocation either because it needs to be marked as given (as deduced from the context in (14)), or because it needs to “evacuate” the VP in order to focus some other referent (altruistic movement)—see further discussion in section 3.2.
Zulu (S42, Cheng and Downing, 2012: 250)
Whether direct or indirect, these cases show a clear influence of information structure, without which we cannot properly understand the restrictions on object marking.
In summary, we have seen that various aspects of Bantu inflectional morphosyntax are influenced or even determined by information structure. This holds for TAM marking as well as for subject and object marking.
As mentioned in the introduction, Bantu languages are characterized by extensive verbal morphology, not just for inflection but also for derivation. Verb-verb derivation productively happens by means of suffixation, in which the derived verb is typically associated with an increased or decreased valency, as will be shown in section 3.1. Section 3.2 discusses the status of the arguments as syntactic objects and their semantic roles, which is another source of cross-Bantu variation.
3.1. Derivational morphology
Various suffixes can be added between a verb stem and the final suffix for verbal derivation, the most productive of which are the passive (as in (1) above), causative, and applicative (see Schadeberg (2003) for an overview). The passive is valency-decreasing, removing the agent of the underived verb, whereas the causative and applicative are valency-increasing. For the causative, the added thematic role is the causer (15), but for the applicative, the role is rather unspecified. It can typically be a benefactive, instrument, or location (16) (see, among many others, Kimenyi 1980; Baker 1988; Simango 1995; Alsina & Mchombo 1990; Ngonyani 1996).
Lucazi (K13, Fleisch, 2000: 125)
Tswana (S31, Creissels 2004: 13, adapted)
Interestingly, the derived verbs can also be used without changing the valency. These non-canonical applicatives have been shown to influence the pragmatic interpretation. Marten (2003) argues, for example, that adding the applicative can “strengthen the action” expressed in the verb (17), and Creissels (2004) shows that in Tswana, a non-canonical applicative triggers a focus reading of the locative (18); see also de Kind and Bostoen (2012) on Ciluba applicatives and information structure.
Bemba (M40, Marten, 2003: 218)
Tswana (S31, Creissels, 2004: 15)
The terms “direct object” and “indirect object” actually do not seem appropriate for the internal arguments of the few inherently ditransitive predicates in Bantu and the applied and causative ditransitive verbs. Both object arguments appear without linker or preposition, and, in some languages, both DPs show object properties. There are three well-known tests to establish whether a DP functions as a (primary) object: 1. word order, 2. passivization, and 3. object marking (Hyman and Duranti, 1982). Applying these (and further) tests, Bresnan and Moshi (1990) discuss two types of Bantu languages: symmetrical ones, in which either object passes the tests; and asymmetrical ones, in which only one can simultaneously function as the primary object. Haya is a symmetrical language, as either order of objects is allowed (19), either object can be the subject of a passive (20), and either subject can be object-marked on the verb (21).
Haya (JE22, Riedel, 2009: 81)
Sambaa, on the other hand, is an asymmetrical language: the beneficiary is the primary object, as it must be closest to the verb (22), it is the only object that can promote to subject of a passive (23), and it is the first to be object-marked on the verb—only if the beneficiary is object-marked can the patient also be object-marked (24).
Sambaa (G23, Riedel 2009: 79)
Bresnan and Moshi (1990) analyse this crosslinguistic variation in Lexical Functional Grammar, which takes into account not just the syntactic (subject/object) but also the thematic roles (agent/patient/benefactive). Suitability for objecthood and passivization tends to follow the hierarchy benefactive > goal > patient > locative > instrument, and is also affected by animacy and participant hierarchies, which are together referred to as “topicality” hierarchies (Duranti, 1979). That various syntactic, semantic, and pragmatic forces are in play is supported by the observation that the situation is not as simple as just sketched: in some languages, the tests show contradictory results (Marten et al., 2007; Rugemalira, 1991; Schadeberg, 1995; Thwala, 2006), and even in otherwise highly symmetric languages, some asymmetric features may show up (Adams, 2010; Baker et al., 2012).
The question of objecthood and the functions of “valency-changing” morphology are thus very much alive and can shed light on the connections between thematic, syntactic, and discourse roles.
4. Information structure and word order
Bantu languages have been claimed to display flexible word order. That is, although the basic word order can still be given as SVO, permutations of word order are generally allowed by the syntax. Unsurprisingly, these different word orders are associated with a difference in information structure. In this section, we will look at the ban on focus in the preverbal domain, dedicated postverbal focus positions, wh question formation, and the relation with the passive.
4.1. The topic/focus symmetry
Henderson (2006b: 109) notes that
it is a well-documented fact that postverbal or VP-internal material in Bantu languages receives a new information or focus interpretation (Givon 1972, Bokamba 1976, 1979, Bresnan and Mchombo 1987, Machobane 1995, Demuth and Mmusi 1997). On the other hand, preverbal elements such as subjects tend to be interpreted as old information and function as topics.
The prototypical topic is the subject (Chafe, 1976), which appears preverbally in the canonical Bantu word order. However, objects can also move to a preverbal position when they are topical, as illustrated in (25).
Tsonga (S53: Kisseberth, 1994: 154)
Futher evidence for the tendency “preverbal = topical” comes from the fact that, in many Bantu languages, there is an absolute constraint against preverbal focal elements (Bokamba, 1976; Morimoto, 2000; Sabel and Zeller, 2006; van der Wal, 2009; Zerbian, 2006a; 2007, and many others). This is illustrated in (26a) for Northern Sotho, where the inherently focused wh subject mang “who” is ungrammatical in preverbal position. Instead, focused subjects should appear either in a postverbal (and VP-internal) position, as in the inversion construction in (26b), or in a cleft, as in (26c). The same holds for subjects modified by the focus particle fela “only” and for answers to subject questions, which are tests for focus.
Northern Sotho (S32, Zerbian, 2006a: 172, adapted)
Making the link between word order and information structure explicit, Yoneda (2011) argues that word order in Matengo is largely determined by information structure and proposes the following linear template:
(27) Matengo sentence structure
One of the questions this brings up is whether the preverbal domain should be specified positively as [topical] or negatively as [non-focal]. This is particularly relevant when preverbal elements are allowed that are not prototypical topics and that cannot be dislocated topics, either. For example, the sole argument of an intransitive verb in Matengo should appear postverbally when it is indefinite (and presented as new information), as illustrated in (28c,d). In a similar context, when both arguments of a transitive predicate are presented as new, the indefinite subject appears preverbally (28a,b).
Matengo (N13, Yoneda, 2011: 761)
This can be taken as evidence that it is possible to have a non-topic reading in the preverbal domain, which would argue for an underspecified characterization of this domain as [non-focus] rather than [topic]. Alternatively, we could say that, in the absence of a clear topic (as is the case in presentational or thetic sentences; see below), the subject is the most topic-like and will appear in the preverbal domain in this situation. The fact that this happens at all shows that there must be some syntactic constraint on word order in Matengo, according to Yoneda (2011).
A similar question is whether the postverbal domain should be characterized as [focal] or [non-topical]. An argument in favor of the latter is the underspecified interpretation of the postverbal subject: it can be the narrow (possibly exclusive) focus of the sentence, as in (29a), or it can be part of a thetic sentence, the main characteristic of which is the detopicalization of the subject (Lambrecht, 2000; Sasse, 1996), illustrated in (29b).
Zulu (S42, Buell, 2006: 23)
The generalization is that a mismatch between the typical syntactic roles and discourse roles (subject = topic, object = focus/comment) is often marked in word order. This is in some sense the other side of the generalization, known as the oft-observed “subject/object asymmetry” in focus marking: whereas objects may be focused in their canonical (postverbal) position, focused subjects are not allowed in their canonical (preverbal) position, but have to be marked in some way, for example, by inversion or by a cleft construction (see Bokamba (1976) for Dzamba; Sabel and Zeller (2006) for Nguni; Zerbian (2006a, 2007) for Northern Sotho; Van der Wal (2009) for Makhuwa). It may thus be more insightful to refer to this pattern not as a “subject/oject asymmetry,” but as the “topic/focus symmetry,” highlighting the relevance of information structure for word order (see also section 4.5).
4.2. Dedicated focus positions
It has been proposed for various Bantu languages that there is a dedicated position for the focus of the sentence, which can be sentence-final or directly adjacent to the verb. The most frequent and well-known is the so-called Immediate After the Verb (IAV) position, observed and named first by Watters (1979) for Aghem, see also Costa and Kula (2008) for Bemba; Yoneda (2011) for Matengo; Buell (2006, 2009) for Zulu; and Van der Wal (2009) for Makhuwa. In Aghem, a Grassfields Bantu language, a focused element must occur in IAV position. In (30a), the adverbial clause “in the farm” is in its typical sentence-final position. When it is the answer to a question, it is considered the focus of the sentence, and hence it occurs in IAV position (30c), just like the inherently focal question word ghɛ́ “where” (30b). The same applies to focused subjects, which cannot appear in any other than IAV position either (31).
Aghem (Watters, 1979: 147)
(Watters, 1979: 144)
Interestingly, Mbuun has been claimed to exhibit a focus position immediately before the verb (Bostoen and Mundeke, 2012), whereas Kirundi has a sentence-final focus position (Sabimana, 1986; Ndayiragije 1999).
There seems to be a relation between the conjoint/disjoint alternation mentioned in section 2.2 and the IAV position, although the two must be independent (Gibson et al. to appear): there are languages with the alternation but no IAV (Kirundi), and languages with IAV but no conjoint/disjoint alternation (Aghem); see also Buell (2011) for evidence of this independence within Zulu, a language with both the alternation and IAV. Nevertheless, we know that the conjoint verb form is only licensed in the presence of a following element, which is naturally in IAV position. Before discussing the two types of systems regarding the conjoint verb form and IAV focus, note that the relation with IAV holds for tone cases as well,5 as only the first element after the verb can receive complement case (32).
Otjiherero (R30, Kavari et al., 2012: 325)
With regard to the conjoint/disjoint alternation, it seems that there are two systems: relating directly to focus; and relating indirectly to focus (van der Wal, forthcoming). Makhuwa is a language that shows a direct relation between the use of the conjoint form and the interpretation of the IAV element as focus (van der Wal, 2011). What immediately follows the conjoint verb form is interpreted as exclusive focus: the proposition is true for this referent, excluding other referents. One argument in favor of this analysis is the behavior of DPs modified by a focus particle. Exclusive “only” must follow a conjoint verb form (33), whereas explicitly non-exclusive “even” is not allowed after the conjoint verb form (34).
Makhuwa (P31, Van der Wal, 2011: 1739)
In Zulu, on the other hand, the alternation seems to be determined by constituency, and only indirectly connected to focus. The conjoint form appears when some element follows within the relevant constituent (Buell, 2006, 2009), whereas the disjoint form is always final in this constituent. This can be seen in the phonological phrasing (penultimate lengthening indicating the right boundary), which Cheng and Downing (2009) show to correlate with vP and CP constituents. The difference between the two systems is illustrated in the contrast between (35) and (34).
Zulu (S42, Buell, 2008: 45)
This means that the information-structural status of an element correlates with word order (IAV, dislocation) in Zulu, which in turn correlates with the conjoint/disjoint alternation. Hence, the link between the alternation and information structure is indirect, unlike in Makhuwa, in which constituency alone cannot account for the distribution of the verb forms.
A rather important aside is that all languages have an alternative focus strategy, which is a (biclausal) cleft construction. This is in some languages the only available strategy to focus (or question) a subject, as in (36b). Here the verb is in a relative clause and the noun “farmers” is predicative by the absence of the augment à-, which is otherwise prefixed to the noun. See also Wasike (2007) and Cheng and Downing (2013).
Luganda (JE15, Hyman and Katamba, 2010: 92, 94)
4.3. Wh questions
Wh phrases are often said to be inherently focused, as they ask for new information. There is indeed a large overlap between the distribution of wh phrases and focus phrases, especially in languages with a dedicated focus position. This was illustrated earlier for Aghem (31), where wh words must appear in IAV position.
In other languages we find optional wh-movement, as for example in Duala (37). It may only be “optional” in a syntactic sense, as it is uncertain whether fronting induces an interpretive effect here.
Duala (A24, Sabel and Zeller, 2006: 276)
A third type of language leaves wh words in their canonical position. In Chichewa, objects are questioned in postverbal (38b) and subjects in preverbal position (38c). But Chichewa has a cleft strategy, too, which can be used for “emphasis” (38d). This is the difference between neutral and contrastive wh questions, in which the cleft question asks for an exclusive answer: “who, out of the set of alternatives, was the one smashing the pumpkins?” Note that the relative form of the verb in (38d) is only marked by tonal morphology.
Chichewa (N31, Mchombo, 2004: 45, 47)
A fourth—or perhaps mixed—type of language leaves wh words in their canonical position, unless they violate the topic/focus symmetry. Objects are questioned in situ: postverbal (39b,c) but crucially not necessarily in IAV position (39c). Subjects, on the other hand, cannot be questioned preverbally, as was illustrated in (26) earlier. Instead, they must appear in an inversion construction or a cleft.
Northern Sotho (S32, Zerbian, 2006a: 267)
These different types of languages with respect to wh questions show three points. First, there is interesting microvariation within Bantu. Second, not all wh phrases are necessarily focused; wh and focus are frequently co-occurring but independent features. And third, the strategy that is used for wh questioning can be determined or restricted by information structure.
4.4. Relative clauses
Where a language displays a biclausal cleft construction, it necessarily forms a relative clause to encode the given information. However, it is often difficult to tell the difference between a biclausal cleft construction and a monoclausal focus construction, as the latter tends to grammaticalise from the former (see Heine and Reh, 1984; Harris and Campbell, 1995, and Takizala 1972 for an illustration in the Bantu language Kihung’an), and hence there are many intermediate forms that have properties of both constructions. One formal property is the morphosyntactic marking of the relative clause—an area in which Bantu languages again show great variation. Three types of relative marking are discussed directly below; see further the extensive overview in Nsuka Nkutsi (1982), and also the theoretical comparative work in Henderson (2006b, 2007) and Zeller (2004).
A first type of relative clause is formed by a relative complementiser, followed by a clause with canonical word order and subject marking, as illustrated for Lingala, where the relative complementizer oyo introduces subject and object relative clauses.
Lingala (C30b, Meeuwis, 2010: 87)
A second type found in many Bantu languages shows a relative marker on the verb, instead of, or in addition to, the normal subject marker. The relative marker can be initial (bi- in (41)), pre-stem (–cho– in 42a), or post-cliticised to the verb (42b).
Kilega (D25, Carstens, 2005: 233)
Thirdly, there are relative strategies that make use of a structure that is not entirely verbal. One example is a relative clause that is based on the connective morpheme -a “of”, as analysed in Cheng (2006).
Another example contains a participial clause, whose nominal nature causes the pronominal subject in a nonsubject relative to appear as a possessive. This is illustrated for Makhuwa in (44).
Makhuwa (P31, Van der Wal, 2010: 210)
An interesting and much-discussed phenomenon in Bantu non-subject relative clauses is the postverbal position of the subject. This is obligatory in some languages (45) and optional in others.
Bembe (D54, Iorio 2014)
Bemba (M42, Kula & Cheng 2007: 136)
In the case of obligatory subject inversion, this is likely a syntactic requirement (unlike the information structure–driven inversion in section 2.3). Where there is optionality, the question is what determines the word order—for example, it could be the case that the preverbal subject in (46) is topical and the postverbal subject is not. This forms an interesting point for further research.
4.5. Functional passive
The word order flexibility in Bantu is in some languages also used to express what has been called a “functional passive”. A different word order can be used for passives in addition to the derivational passive strategy shown earlier in (1) or replacing (part of) this strategy if a language has lost the proto-Bantu -w- passive.6 For Mbuun, Bostoen and Mundeke (2011) show convincingly that object preposing has developed beyond the stage of simple topicalization, and non-canonical OSV order functions as a passive in this language. This is exemplified in (47), where the verb has its active form and agreement is with the subject; the fact that the initial object is weakly quantified shows that it cannot be dislocated. Similar facts are shown for Bàsàá by Hamlaoui and Makasso (2013).
Mbuun (B87, Bostoen and Mundeke, 2011: 89)
Not only object preposing, but also subject inversion can be used in Matengo as the functional equivalent of the canonical passive, resulting in OVS (48) or VS order (49) with a passive interpretation (van der Wal, forthcoming).
The passive is traditionally described as an operation demoting the subject and promoting the object. This is formulated in syntactic terms, but it should also be seen in information-structural terms: demoting the subject from its canonical topic role and encoding the object as topical (Bliss and Storoshenko, 2008; Keenan and Dryer, 2007; Siewierska, 1984). The word order strategies illustrated in this section do exactly that: they place the object in a topic position and/or detopicalize the subject. Again, it appears that information structure is an important—and perhaps the main—factor in determining the word order.
5. Theoretical implications
5.1. The direct or indirect relation between syntax and semantics/pragmatics
The effect of the Immediate After Verb position suggests that linear order is relevant to information structure (IS). A question is whether IS only operates on linear order or also on hierarchical structure. On the former standpoint, IS is only a superficial phenomenon that does not affect the syntax (cf. Berwick and Chomsky (2011)). However, considering that core syntactic processes like tense inflection and person agreement are sensitive to IS in various Bantu languages, IS must somehow be present in, or have access to, the syntax. How to model this is a hotly debated issue.
For Bantu syntax, there are generative proposals using a formal [focus] or [topic] feature, either as heading a separate projection or as a subfeature in agreement. For example, Ndayiragije (1999) proposes a focus projection focP under TP to account for the focus interpretation and the disjoint morpheme in Kirundi. Zeller (2008), on the other hand, proposes that the subject marker in Zulu starts out as part of the subject DP and has a [-foc] feature. This is motivated by the anti-agreement effect in Zulu: when the logical subject is focused, it does not have its subject marker; instead, a default ku- is inserted (50).
Zulu (S42, Zeller, 2008: 223, adapted)
Conversely, the “agreeing subject” gets an antifocus interpretation, which in turn motivates its movement out of the VP. Zeller suggests that this is behind the generalization that Bantu subject agreement is only “upwards” (cf. Baker, 2008 and section 2.3).
A problematic phenomenon to account for with IS features in the narrow syntax is “altruistic movement.” If movement results in an interpretive effect on the moved constituent (e.g., the exclusive focus in IAV position in Makhuwa), we can say that the movement was triggered by the presence of a topic or focus feature on that constituent. However, if movement is triggered by the need for another element to receive a certain interpretation, it is difficult to model this with the presence of a feature on the moved constituent. An example of such altruistic movement is found for the IAV position in Zulu and its indirect link with focus.
Instead of positing a focus feature, Buell (2009) and Cheng and Downing (2012) propose that the IAV focus interpretation in Zulu is due to the element being in a vP/CP constituent-internal position, which is the optimal locus for prosodic prominence. Anything within this constituent receives a focus interpretation.7 The analysis is motivated by the fact that a focused element cannot simply move to IAV position, but instead the nonfocal element has to be extraposed. This is illustrated in (51): the canonical word order is S V IO DO (51a), but when the DO imali “money” is focused (and hence needs to occur IAV), the “intervening” nonfocal IO uSipho is evacuated from the verb phrase (51b). The dislocated status of the IO can be seen in its separate phonological phrasing and in the obligatory object marking on the verb (see section 2.3).
Zulu (S42, Cheng and Downing, 2012: 248)
This movement can be captured in non-derivational analyses of the relation between IS and syntax. These approaches assume that syntax generates syntactically well-formed strings without taking IS into account, and the resulting representations are then mapped onto an independent IS template at the interface with semantics/pragmatics, or evaluated against phonological and/or IS constraints (e.g., Costa and Kula (2008)). Optimality Theory analyses like Zerbian (2006a) and Cheng and Downing (2012) capture the relation with prosody as well as IS, arguing for an indirect relation with focus via alignment with phonological phrase boundaries, and against a cartographic account involving focP. Cheng and Downing (2012: 266) conclude that the “‘IAV position’ is a portmanteau designation for the interaction of several factors, rather than for just one structural requirement.”
The Dynamic Syntax approach also argues for a more nuanced approach to the syntax-pragmatics interface, where “gradient and underdetermined effects are expected” (Marten 2011: 803). The information-structural interpretation is in this approach a result of particular structure building choices in the parsing of, for example, a subject marker or an auxiliary (Marten, 2007, 2011; Gibson 2012). There is thus no direct encoding of ‘topic’ or ‘focus’, but the semantic and pragmatic interpretation are encoded in the step-by-step growing representation of a string of natural language.
A related issue is the fuzzy boundary between the traditional A versus A-bar positions. A(rgument) positions are typically not associated with discourse properties, whereas A-bar positions and movement to them (wh, quantifiers, topic/focus) typically are. For example, in Henderson’s (2006b, 2011b) analysis of subject-object reversal, the preposed topical object is in an A-bar position (specCP), the A-position (specTP) not being available. Yet the functional passive and inversion constructions seriously trouble this distinction: the “passive” OSV order is motivated by the topical interpretation of the object, but it seems to involve A-movement; and the preposed constituent in subject inversion seems to be in an A position for locative inversion (10), being the grammatical subject, but in an A-bar position for patient inversion (53).
5.2. The relevance of syntactic roles (or: the universality of Case)
Especially from the discussion surrounding word order (section 4), it appears that argument licensing in Bantu languages is sensitive to discourse function, perhaps even more than to syntactic function. This relates to a proposal concerning abstract Case by Harford Perez (1985), recently revived by Diercks (2012). In generative syntax, Case is needed on every DP to license the overt appearance of that DP in a certain position and syntactic function (the Case Filter). Diercks (2012) proposes that this principle does not hold across the board, and that in Bantu languages Case does not play a role.
A straightforward argument is that Bantu nouns do not show morphological case marking. A second argument is that subject DPs can stay in a nonfinite complement clause. As nominative Case is assumed to only be assigned in finite clauses, a DP cannot be Case licensed in a nonfinite clause and should be ungrammatical in this lower subject position—as indeed it is in English. The fact that this is fine in a language like Digo (52b) suggests that Case does not play a role in argument licensing here.
Digo (E73, Diercks, 2012: 260, referring to Steve Nicolle)
A third argument provided by Diercks is that subject agreement on the verb is not determined by (nominative) Case. That is, in certain inversion constructions, the agreement is default, or with a preverbal DP, such as the patient in (53). Furthermore, the logical postverbal subject in these constructions does not seem to be Case-licensed at all, which forms another reason to suggest that Case is irrelevant here.8
Kirundi (JD62, Ndayiragije, 1999: 418)
The tests Diercks proposes do not yield the same results for the whole of Bantu, however: languages in which subject agreement is consistently with the subject, even in subject inversion like in (54), can be argued to be sensitive to nominative Case (Van der Wal forthcoming).
Makwe (G402, Devos, 2008: 382)
Even if nominative Case is inactive in several Bantu languages, Halpert (2012, 2013) and Carstens and Mletshe (2015) show that there is a structural licensing of DPs inside vP in Zulu and Xhosa, which they argue is dependent on Case. Halpert bases her analysis on the distribution of augmentless nouns such as muntu “person/anyone” (compare augmented u-muntu “person”). These augmentless nouns can only appear in vP-internal position, whether in a subordinate clause (55a) or raised to a main clause (55c).
Zulu (S42, Halpert, 2013)
These different systems underline the importance of Bantu microvariation for the study of Bantu syntax (and comparative syntax more generally). Furthermore, it may show that if Case is parameterized, it need not be present or absent for the whole language, but may be active only in a subdomain (e.g., vP, but not TP). Finally, if, compared to European languages, Case and thereby syntactic roles do not play as big a role in the grammar of (some) Bantu languages, there is space for various syntactic effects such as agreement and word order to be determined by something other than whether a DP is subject or object. This other factor could well be the status of DPs as topic or focus, and in general the presentation of information in a sentence, resulting in what has been called a discourse-configurational language (É. Kiss, 1995).
5.3. The syntax-morphology interface
The extensive verbal morphology in Bantu languages can enlighten our understanding of the relation between syntax and morphology. In fact, major insights in morphosyntactic theory were at least partly inspired by Bantu: Henderson (2011a: 17) remarks that Baker’s (1985) and Marantz’s (1984) research featuring Bantu languages “led to the conclusion that at least some derivational word formation, namely that having to do with grammatical relations, must take place under syntactic conditions and not exclusively in the lexicon.” A clear example is Baker’s (1985) Mirror principle, which states that morpheme order reflects the order of syntactic operations. This is the basis of approaches like Distributed Morphology (Halle and Marantz, 1993) and Nanosyntax (see Abels and Muriungi, 2008 and Taraldsen, 2010 for applications in Bantu languages), where syntax operates on formal syntactic features, and morphological form only comes in at the end of the derivation to spell out the syntactic features.
The agglutinative nature of Bantu morphology lends itself well to these approaches, as also shown by Julien (2002), who proposes a difference in the syntactic nature of suffixation and prefixation in Shona verbs. Prefixes are individual inflectional heads whose features are spelled out in their base positions, whereas suffixes are attached via head-movement (see Julien, 2002, chapter 4 for arguments, and Kinyalolo 2003, Buell 2005, Van der Wal, 2009 for this analysis in other languages).
This implies that the verb root only head-moves through the lower part of the derivation, picking up the applicative, causative, passive, etcetera suffixes, and stopping in a position under T. Such a “wysiwyg” analysis of verbal morphology supports the need for an extended IP projection in Bantu (hosting various heads relating to SM, OM, TAM) and attests to (variation in) verb-movement being more fine-grained than just “V-to-T.” Furthermore, the agreement facts in relatives are suggestive of agreement features on both T (subject agreement) and C (complementizer agreement) (see Carstens, 2011 and Henderson, 2011b). Relative clauses and subject inversion in relatives can also inform us on verb movement and its triggers, as Demuth and Harford (1999) argue, and about morphological adjacency requirements forcing pronunciation of a lower copy of the subject, as suggested by Henderson (2006a) and Henderson (2007).
The approach also sprouts questions about head-movement: Muriungi (2008) shows that the relative scope of derivational suffixes in Kîîtharaka (e.g., as in (58)), which is assumed to represent their respective hierarchical order, does not always match their linear order as predicted by the head-movement approach. He instead argues for phrasal movement to account for both the scope and the order of suffixes.
Kîîtharaka (E54, Muriungi, 2008: 4, adapted)
A further challenge is the influence of information structure on the inflectional morphology, as illustrated in previous sections. If there is a syntactic feature like [focus], it could form an independent head that is spelled out together with TAM/SM (as in a nanosyntactic spanning approach). But if the relation with information structure is an indirect one, for example, where the conjoint/disjoint morphology is dependent on the verb being final in the relevant constituent (vP), morphology may need to be modeled as dependent on constituent structure and/or linear order.
Just as for word order, a representational approach to morpheme order has also been proposed, having an independent morphology module. For example, Hyman (2003) presents the “CARP” template, according to which the derivational suffixes are by default ordered Caus-Appl-Recip-Pass (59). In his Optimality Theory analysis, the constraint that morphemes adhere to this template order competes with a constraint “Mirror,” which requires suffix ordering to be compositional, with the order reflecting the syntactic and semantic scope.
Makhuwa (P31, Van der Wal, 2009: 78)
If we take serious Larry Hyman’s suggestion to “believe morphology,” careful study of the morphological processes in Bantu verbal forms will keep revealing more and more of the morphology-syntax interface.
In any area of linguistics, it is essential to take an approach that conjoins language description and theoretically informed specific questioning. Especially now that we have started to cover a bit of ground in Bantu syntax, and because the Bantu languages form such a rich field of microvariation (which was only cursorily illustrated here), I think it vital to pursue that combination of description and theory in studying Bantu syntax—only then can we hope to find new interesting phenomena while also gaining a deeper understanding of the universals and variation in human languages.
Henderson (2011a: 23) rightly notes that “this descriptive and theoretical base makes formal micro-parametric work possible in ways that it simply was not 20 years ago. We now know it is not the case that once you have seen one Bantu language, you have seen them all.” I agree with Henderson that the treasure chest of Bantu syntax deserves more comparative attention (cf. Marten et al., 2007), and that this microcomparative typological work is a worthwhile strand of research for the coming years.
A particular point stressed in this chapter is that information structure is indispensable in describing and analysing Bantu languages. It is relevant not just in the prosody, but also the morphology and the syntax: word order, valency, voice, tense-aspect marking, subject and object marking can all be influenced and affected by the information structure expressed in the sentence. It is therefore important to take this into account when studying Bantu languages, and it furthermore shows that Bantu languages are a valuable field for research projects dedicated to the study of information structure and its interfaces.
To conclude, Bantu languages have substantially influenced general syntactic theory over the last decades (Henderson, 2011a), and I see every reason for it to continue that way: there is a wealth of empirical and theoretical lessons to be learned from the Bantu languages.
Abbreviations and symbols
Numbers refer to noun classes, or to persons when followed by sg or pl.
immediately after the verb
neuter or stative
Work for this chapter was supported as part of the European Research Council Advanced Grant No. 269752, “Rethinking Comparative Syntax.” I would like to thank Leston Buell, Michael Diercks, Brent Henderson, and Ian Roberts for their help and comments on previous versions.
Abels, K., Muriungi, P., 2008. The focus marker in Kîîtharaka: Syntax and semantics. Lingua 118, 687–731.Find this resource:
Adams, N.B., 2010. The Zulu ditransitive verb phrase. PhD, University of Chicago.Find this resource:
Alsina, Alex and Sam Mchombo., 1990. The syntax of applicatives in Chichewa: problems for a theta theoretic account. Natural Language & Linguistic Theory 8, 493–506.Find this resource:
Baker, M., 1985. The Mirror Principle and morphosyntactic explanation. Linguistic Inquiry 16, 373–415.Find this resource:
Baker, M., 1988. Theta theory and the syntax of applicatives in Chichewa. Natural Language & Linguistic Theory 6, 353–389.Find this resource:
Baker, M., 2008. The syntax of agreement and concord. Cambridge University Press, Cambridge.Find this resource:
Baker, M., Safir, K., Sikuku, J., 2012. Sources of (a)symmetry in Bantu double object constructions, in: Arnett, N., Bennett, R. (Eds.), 30th West Coast Conference on Formal Linguistics. Cascadilla Proceedings Project, pp. 54–64.Find this resource:
Bax, A., Diercks, M., 2012. Information structure constraints on object marking in Manyika. Southern African Linguistics and Applied Language Studies 30, 185–202.Find this resource:
Bearth, T., 2003. Syntax, in: Nurse, D., Philippson, G. (Eds.), The Bantu languages. Routledge, London, pp. 121–142.Find this resource:
Bergvall, V., 1987. Focus in Kikuyu and universal grammar. Harvard University, Cambridge, MA.Find this resource:
Berwick, R., Chomsky, N., 2011. Biolinguistics: The current state of its evolution and development, in: Sciullo, A.M.D., Boeckx, C. (Eds.), Biolinguistic investigations. Oxford University Press, Oxford, pp. 19–41.Find this resource:
Blanchon, J.A., 1998. Semantic/pragmatic conditions on the tonology of the Kongo noun phrase: a diachronic analysis, in: Hyman, L., Kisseberth, C. (Eds.), Theoretical Aspects of Bantu Tone. CSLI, Stanford, California, pp. 37–82.Find this resource:
Blanchon, J.A., 1999. “Tone cases” in Bantu Group B.40, in: Blanchon, J.A., Creissels, D. (Eds.), Issues in Bantu Tonology. Rüdiger Köppe Verlag, Cologne, pp. 37–82.Find this resource:
Bliss, H., Storoshenko, D.R., 2008. Passivization and A’-movement in Shona, Annual conference of the Canadian Linguistic Association.Find this resource:
Bokamba, E.G., 1976. On the syntax and semantics of wh-questions in Kikongo and Kiswahili. Studies in the Linguistic Sciences 6, 65–88.Find this resource:
Bokamba, E.G., 1979. Inversions as grammatical relation changing rules in Bantu languages. Studies in the Linguistic Sciences 9, 1–24.Find this resource:
Bostoen, K., Mundeke, L., 2011. Passiveness and inversion in Mbuun (Bantu B87, DRC). Studies in Language 35, 72–111.Find this resource:
Bostoen, K., Mundeke, L., 2012. Subject marking, object-verb order and focus in Mbuun (Bantu, B87). Southern African Linguistics and Applied Language Studies 30, 139–154.Find this resource:
Bresnan, J., Kanerva, J.M., 1989. Chichewa lovative inversion: a case study of factorization in grammar. Linguistic Inquiry 20, 1–50.Find this resource:
Bresnan, J., Mchombo, S., 1987. Topic, pronoun, and agreement in Chichewa. Language 63, 741–782.Find this resource:
Bresnan, J., Moshi, L., 1990. Object asymmetries in comparative Bantu syntax. Linguistic Inquiry 21, 147–185.Find this resource:
Buell, L.C., 2005. Issues in Zulu morphosyntax. University of California, Los Angeles.Find this resource:
Buell, L.C., 2006. The Zulu conjoint/disjoint verb alternation: focus or constituency?. ZAS Papers in Linguistics 43, 9–30.Find this resource:
Buell, L.C., 2008. VP-internal DPs and right-dislocation in Zulu, in: Koppen, M.v., Botma, B. (Eds.), Linguistics in the Netherlands 25. John Benjamins, Amsterdam, pp. 37–49.Find this resource:
Buell, L.C., 2009. Evaluating the immediate postverbal position as a focus position in Zulu, in: Matondo, M., Laughlin, F.M., Potsdam, E. (Eds.), Selected Proceedings of the 38th Annual Conference on African Linguistics: Linguistic Theory and African Language Documentation. Cascadilla Proceedings Project, Somerville, MA, pp. 166–172.Find this resource:
Buell, L.C., 2011. Zulu ngani “why”: postverbal and yet in CP. Lingua 121, 805–821.Find this resource:
Buell, L.C., forthcoming. The Bantu languages, in Kiss, T., Alexiadou, A. (Eds.), Handbook of syntax, part VII. De Gruyter, Berlin, 1622–1657.Find this resource:
Carstens, V., 2005. Agree and EPP in Bantu. Natural Language and Linguistic Theory 23, 219–279.Find this resource:
Carstens, V., 2011. Hyperactivity and hyperagreement in Bantu. Lingua 121, 721–741.Find this resource:
Carstens, V., Mletshe, L., 2015. Radical defectivity: implications of Xhosa expletive constructions. Linguistic Inquiry 46(2), 187–242.Find this resource:
Carter, H., Kahari, G.P., 1978. An introductory Shona reader with grammatical sketch, Part II: a short grammar of Shona, annotations to passages and bibliography. School of Oriental and African Studies, London.Find this resource:
Chafe, W., 1976. Givenness, contrastiveness, definiteness, subjects, topics and point of view, in: Li, C.N., Thompson, S. (Eds.), Subject and topic. Academic Press, New York, pp. 27–55.Find this resource:
Cheng, L. L.-S., 2006. Decomposing Bantu relatives. Proceedings of NELS 36, 197–215.Find this resource:
Cheng, L.L.-S., Downing, L.J., 2009. Where’s the topic in Zulu? The Linguistic Review 26, 207–238.Find this resource:
Cheng, L.L.-S., Downing, L.J., 2012. Against FocusP: arguments from Zulu, in: Kucerova, I., Neeleman, A. (Eds.), Information structure. Contrasts and positions. Cambridge University Press, Cambridge, pp. 247–267.Find this resource:
Cheng, L.L.-S., Downing, L.J., 2013. Clefts in Durban Zulu, in: Hartmann, K., Veenstra, T. (Eds.), Cleft structures. John Benjamins, Amsterdam, pp. 141–164.Find this resource:
Collins, C., 2004. The agreement parameter, in: Breitbarth, A., Riemsdijk, H.v. (Eds.), Triggers. Mouton de Gruyter, Berlin, pp. 115–136.Find this resource:
Costa, J., Kula, N.C., 2008. Focus at the interface: Evidence form Romance and Bantu, in: De Cat, C., Demuth, K. (Eds.), The Bantu-Romance Connection. John Benjamins, Amsterdam.Find this resource:
Creissels, D., 2004. Non-canonical applicatives and focalization in Tswana, Paper presented at Syntax of the World’s Languages, Leipzig.Find this resource:
Creissels, D., 2011. Tswana locatives and their status in the inversion construction. Africana Linguistica XVII, 33–52.Find this resource:
de Kind, J., Bostoen, K., 2012. The applicative in ciLubà grammar and discourse: A semantic goal analysis. Southern African Linguistics and Applied Language Studies 30, 101–124.Find this resource:
Demuth, K., Harford, C., 1999. Verb raising and subject inversion in comparative Bantu. Journal of African Languages and Linguistics 20, 41–61.Find this resource:
Demuth, K., Mmusi, S., 1997. Presentational focus and thematic structure in comparative Bantu. Journal of African Languages and Linguistics 18, 1–19.Find this resource:
Devos, M., 2008. A grammar of Makwe. München, LINCOM Studies in African Linguistics 71.Find this resource:
Diercks, M., 2011. The morphosyntax of Lubukusu locative inversion and the parameterization of Agree. Lingua 121, 702–720.Find this resource:
Diercks, M., 2012. Parameterizing Case: Evidence from Bantu. Syntax 15, 253–286.Find this resource:
Diercks, M., 2014. Locative inversion. Ms, Pomona College.Find this resource:
Diercks, M., Ranero, R., Paster, M., 2015. Evidence for a clitic analysis of object markers in Kuria, in Kramer, R. et al. (eds.), Selected Proceedings of the 44th Annual conference of African Linguistics . Cascadilla Proceedings Project, Somerville, MA, pp. 52–70.Find this resource:
Diercks, M., Sikuku, J., 2015. Pragmatic effects of clitic doubling: two kinds of object markers in Lubukusu. Ms, Pomona College and Moi University.Find this resource:
Diesing, M., 1992. Indefinites. MIT Press, Cambridge, MA.Find this resource:
Duranti, A., 1979. Object clitic pronouns in Bantu and the topicality hierarchy. Studies in African Linguistics 10, 31–45.Find this resource:
É. Kiss, K., 1995. Discourse configurational languages. Oxford University Press, New York.Find this resource:
Fleisch, A., 2000. Lucazi grammar. Rüdiger Köppe Verlag, Cologne.Find this resource:
Gibson, H., 2012. Auxiliary placement in Rangi: a Dynamic Syntax perspective. SOAS, London.Find this resource:
Gibson, H., Koumbarou, A., Marten, L., Van der Wal, J., to appear. Locating the Bantu conjoint/disjoint alternation in a typology of focus marking. In Van der Wal, J., Hyman, L.M. (Eds). The conjoint/disjoint alternation. Trends in Linguistics, Mouton de Gruyter.Find this resource:
Güldemann, T., 1996. Verbalmorphologie und Nebenprädikation im Bantu. Universitätsverlag Dr. N. Brockmeyer, Bochum.Find this resource:
Güldemann, T., 2003. Present progressive vis-à-vis predication focus in Bantu: A verbal category between semantics and pragmatics. Studies in Language. John Benjamins, Amsterdam, pp. 323–360.Find this resource:
Guthrie, M., 1967–1971. Comparative Bantu. Gregg International, Letchworth, UK and Brookfield, VT.Find this resource:
Hadermann, P., 2005. Eléments segmentaux et supra-segmentaux pour marquer la fonction “object” dans quelques langues bantoues, in: Bostoen, K., Maniacky, J. (Eds.), Studies in African comparative linguistics, with special focus on Bantu and Mande. Musée Royal de l’Afrique Centrale, Tervuren, pp. 397–410.Find this resource:
Halle, M., Marantz, A., 1993. Distributed Morphology and the Pieces of Inflection, in: Hale, K., Keyser, S.J. (Eds.), The View from Building 20. MIT Press, Cambridge, MA, pp. 111–176.Find this resource:
Halliday, M., 1967. Notes on transitivity and theme in English: part 2. Journal of Linguistics 3, 199–244.Find this resource:
Halpert, C., 2012. Case, agreement, EPP and information structure: a quadruple dissociation in Zulu, in: Choi, J., Hogue, E.A., Punske, J., Tat, D., Schertz, J., Trueman, A. (Eds.), WCCFL 29. Cascadilla Proceedings Project.Find this resource:
Halpert, C., 2013. Structural case and the nature of vP in Zulu, in: Sloggett, S., Keine, S. (Eds.), 42nd Meeting of the North East Linguistic Society (NELS 42).Find this resource:
Hamlaoui, F., Makasso, E.-M., 2013. Object left-dislocation, topicalization and the syntax-phonology mapping of intonation phrases in Bàsàá, Bantu5, Paris.Find this resource:
Harford Perez, C., 1985. Aspects of complementation in three Bantu languages. University of Wisconsin–Madison, Madison, WI.Find this resource:
Harris, A.C., Campbell, L., 1995. Historical syntax in cross linguistic perspective. Cambridge University Press, Cambridge.Find this resource:
Heine, B., Reh, M., 1984. Grammaticalization and reanalysis in African languages. Helmut Buske, Hamburg.Find this resource:
Henderson, B., 2006a. Multiple agreement and inversion in Bantu. Syntax 9, 275–289.Find this resource:
Henderson, B., 2006b. The syntax and typology of Bantu relative clauses. University of Illinois at Urbana–Champaign.Find this resource:
Henderson, B., 2007. The syntax of agreement in Bantu relatives, in: Hoyt, F., Seifert, N., Teodorescu, A., White, J. (Eds.), Texas Linguistics Society 9: Morphosyntax or underrepresented languages. CSLI Publications.Find this resource:
Henderson, B., 2011a. African languages and syntactic theory: impacts and directions, in: Bokamba, E., Shosted, R.K., Ayalew, B.T. (Eds.), 40th Annual Conference on African Linguistics. Cascadilla Proceedings Project.Find this resource:
Henderson, B., 2011b. Agreement, locality, and OVS in Bantu. Lingua 121, 742–753.Find this resource:
Hyman, L.M., 2003. Suffix ordering in Bantu: a morphocentric approach. In Booij, G., J. van Marle (eds.), Yearbook of Morphology 2002, 245–281. Dordrecht: Kluwer.Find this resource:
Hyman, L.M., forthcoming. Disentangling conjoint, disjoint, metatony, tone cases, augments, prosody, and focus in Bantu, in: van der Wal, J., Hyman, L.M. (Eds), The conjoint/disjoint alternation in Bantu, Trends in Linguistics. De Gruyter, Berlin.Find this resource:
Hyman, L.M., Polinksy, M., 2009. Focus in Aghem, in: Féry, C., Zimmermann, M. (Eds.). Oxford University Press, Oxford, pp. 206–233.Find this resource:
Hyman, L.M., Watters, J., 1984. Auxiliary focus. Studies in African Linguistics 15, 233–273.Find this resource:
Hyman, L.M., Duranti, A., 1982. On the object relation in Bantu, in: Hopper, P., Thompson, S. (Eds.), Syntax and Semantics: studies in transitivity. Academic Press, New York, pp. 217–239.Find this resource:
Hyman, L.M., Katamba, F.X., 2010. Tone, syntax and prosodic domains in Luganda. ZAS Papers in Linguistics 53, 69–98.Find this resource:
Hyman, L.M., Lionnet, F., 2012. Metatony in Abo (Bankon), A42, in: Marlo, M.R., Adams, N.B., Morrison, M., Purvis, T.M. (Eds.), 42nd Annual Conference on African Linguistics. Cascadilla Proceedings Project, pp. 1–14.Find this resource:
Iorio, D., 2014. Subject and object marking in Bembe, PhD at Newcastle University.Find this resource:
Julien, M., 2002. Syntactic heads and word formation. Oxford University Press, Oxford.Find this resource:
Kavari, J.U., Marten, L., van der Wal, J., 2012. Tone cases in Otjiherero: head-complement relations, linear order, and information structure. Africana Linguistica XVIII, 315–353.Find this resource:
Keenan, E.L., Dryer, M.S., 2007. Passive in the world’s languages, in: Schopen, T. (Ed.), Language typology and syntactic description. Volume I. Clause Structure. Cambridge University Press, Cambridge, pp. 325–361.Find this resource:
Khumalo, L., 2012. Passive, locative inversion in Ndebele and the unaccusative hypothesis. South African Journal of African Languages 30, 22–34.Find this resource:
Kimenyi, A., 1980. A relational grammar of Kinyarwanda. Berkeley and Los Angeles: University of California Press.Find this resource:
Kinyalolo, K.K.W., 2003. Limiting the scope of V-movement in Kilega, Handout of a talk given at Stony Brook, New York.Find this resource:
Kisseberth, C., 1994. On domains, in: Cole, J., Kisseberth, C. (Eds.), Perspectives in phonology. CSLI, Stanford, CA, pp. 133–166.Find this resource:
Krifka, M., 2007. Basic notions of information structure, in: Féry, C., Fanselow, G., Krifka, M. (Eds.), Working Papers of the SFB 632, Interdisciplinary Studies on Information Structure. Universitätsverlag Potsdam, Potsdam, pp. 13–55.Find this resource:
Krifka, M., Musan, R., 2012. Information structure: overview and linguistic issues, in: Krifka, M., Musan, R. (Eds.), The expression of information structure. Mouton de Gruyter, Berlin, pp. 1–44.Find this resource:
Kula, N.C., Cheng, L.L.-S., 2007. Phonological and syntactic phrasing in Bemba relatives. Journal of African Languages and Linguistics 28, 123–148.Find this resource:
Lambrecht, K., 1994. Information structure and sentence form. Cambridge University Press, Cambridge.Find this resource:
Lambrecht, K., 2000. When subjects behave like objects. Studies in Language 24, 611–682.Find this resource:
Maho, Jouni., 2009. “Nugl Online: The Online Version of the New Updated Guthrie List, a Referential Classification of the Bantu Languages.” glocalnet.net/mahopapers/nuglonline.pdf
Makasso, E.-M., 2012. Metatony in Basaa, in: Marlo, M.R., Adams, N.B., Morrison, M., Purvis, T.M. (Eds.), 42nd Annual Conference on African Linguistics. Cascadilla Proceedings Project, pp. 15–22.Find this resource:
Marantz, A., 1984. On the nature of grammatical relations. MIT Press, Cambridge, MA.Find this resource:
Marlo, M.R., 2014. Exceptional patterns of object marking in Bantu. Studies in African Linguistics 43, 85–123.Find this resource:
Marlo, M.R., forthcoming. On the number of object markers in Bantu languages. Journal of African Languages and Linguistics.Find this resource:
Marten, L., 2003. The dynamics of Bantu applied verbs: an analysis at the syntax-pragmatics interface, in: Lébikaza, K.K. (Ed.), Actes du 3e Congrès Mondial de Linguistique Africaine Lomé 2000. Rüdiger Köppe Verlag, Cologne, pp. 207–221.Find this resource:
Marten, L., 2006. Locative inversion in Herero: More on morphosyntactic variation in Bantu. ZAS Papers in Linguistics 43, 97–122.Find this resource:
Marten, L., 2007. Focus strategies and the incremental development of semantic representations: evidence from Bantu, in: Aboh, E.O., Hartmann, K., Zimmermann, M. (Eds.), Focus strategies in African languages. Mouton de Gruyter, Berlin, pp. 3–39.Find this resource:
Marten, L., 2011. Information structure and agreement: Subjects and subject markers in Swahili and Herero. Lingua 121, 787–804.Find this resource:
Marten, L., Kula, N.C., 2012. Object marking and morphosyntactic variation in Bantu. South African Journal of African Languages 30, 237–253.Find this resource:
Marten, L., Kula, N.C., Thwala, N., 2007. Parameters of morpho-syntactic variation in Bantu. Transactions of the Philological Society 105, 253–338.Find this resource:
Marten, L., van der Wal, J., 2015. A typology of Bantu inversion constructions. Linguistic Variation 14(2), 318–368.Find this resource:
Mchombo, S., 2004. The syntax of Chichewa. Cambridge University Press, Cambridge.Find this resource:
Meeussen, A.E., 1967. Bantu grammatical reconstructions. Annales du Musée Royal de l’ Afrique Centrale 61.Find this resource:
Meeuwis, M., 2010. A Grammatical Overview of Lingála. Lincom, München.Find this resource:
Morimoto, Y., 2000. Discourse-configurationality in Bantu morphosyntax. Stanford University, Stanford, CA.Find this resource:
Morimoto, Y., 2006. Agreement properties and word order in comparative Bantu. ZAS Papers in Linguistics 43, 161–188.Find this resource:
Muriungi, P., 2008. Phrasal movement inside Bantu verbs: deriving affix scope and order in Kîîtharaka. University of Tromsø.Find this resource:
Myers, S., 1990. Tone and the structure of words in Shona. Garland, New York.Find this resource:
Ndayiragije, J., 1999. Checking economy. Linguistic Inquiry 30, 399–444.Find this resource:
Ngonyani, D., 1996. The morphosyntax of applicatives. PhD, University of California, Los Angeles.Find this resource:
Nsuka Nkutsi, F., 1982. Les structures fondamentales du relatif dans les langues bantoues. Annales du Musée Royal de l’ Afrique Centrale 108.Find this resource:
Nurse, D., Philippson, G., 2003. The Bantu Languages. Routledge, London.Find this resource:
Poulos, G., Louwrens, L.J., 1994. A linguistic analysis of Northern Sotho. Via Afrika, Pretoria.Find this resource:
Riedel, K., 2009. The syntax of object marking in Sambaa: a comparative perspective. LOT, Utrecht.Find this resource:
Rugemalira, J.M., 1991. What is a symmetrical language? Multiple object constructions in Bantu, in: Hubbard, K. (Ed.), 17th Annual Meeting of the Berkeley Linguistics Society: Papers from the special session. Berkeley Linguistics Society, pp. 200–209.Find this resource:
Sabel, J., Zeller, J., 2006. WH question formation in Nguni, in: Mugane, J., Hutchison, J.P., Worman, D.A. (Eds.), Selected Proceedings of the 35th Annual Conference on African Linguistics. Cascadilla Proceedings Project, Somerville, MA, pp. 271–283.Find this resource:
Sabimana, F., 1986. The relational structure of the Kirundi verb, Department of Linguistics. Indiana University.Find this resource:
Salzmann, M., 2011. Towards a typology of Locative Inversion—Bantu, perhaps Chinese and English—but beyond? Language and Linguistics Compass 5/4, 169–189.Find this resource:
Sasse, H.-J., 1996. Theticity, Arbeitspapiere Institut für Sprachwissenschaft der Universität zu Köln. Cologne.Find this resource:
Schadeberg, T.C., 1986. Tone cases in Umbundu. Africana Linguistica 10, 427–445.Find this resource:
Schadeberg, T.C., 1995. Object diagnostics in Bantu, in: Emenanjo, E.N., Ndimele, O.-m. (Eds.), Issues in African languages and linguistics. National Institute for Nigerian Languages, Aba, pp. 173–180.Find this resource:
Schadeberg, T.C., 2003. Derivation, in: Nurse, D., Philippson, G. (Eds.), The Bantu languages. Routledge, London, pp. 71–89.Find this resource:
Schadeberg, T.C., Mucanheia, F.U., 2000. Ekoti. Rüdiger Köppe Verlag. Cologne.Find this resource:
Schwarz, F., 2007. Ex-situ focus in Kikuyu, in: Aboh, E.O., Hartmann, K., Zimmermann, M. (Eds.), Focus strategies in African languages: the interaction of focus and grammar in Niger-Congo and Afro-Asiatic. Mouton de Gruyter, Berlin, pp. 139–160.Find this resource:
Seidl, A., Dimitriadis, A., 1997. The discourse function of object marking in Swahili. CLS 33: The Main Session, 373–389.Find this resource:
Siewierska, A., 1984. The passive: a comparative linguistic analysis. Croom Helm, London, Sydney, Dover.Find this resource:
Simango, S.R., 1995. The syntax of Bantu double object constructions. Doctoral dissertation, University of South Carolina.Find this resource:
Takizala, A., 1972. Focus and relativization: the case of Kihung’an. Studies in African Linguistics 3, 259–284.Find this resource:
Taraldsen, T.K., 2010. The nanosyntax of Nguni noun class prefixes and concords. Lingua 120, 1522–1548.Find this resource:
Thwala, N., 2006. Parameters of variation and complement licensing in Bantu. ZAS Papers in Linguistics 43, 209–232.Find this resource:
Vallduví, E., 1992. The informational component. Garland Publishing, Inc., New York.Find this resource:
van der Spuy, A., 1993. Dislocated noun phrases in Nguni. Lingua 90, 335–355.Find this resource:
van der Wal, J., 2006. Predicative tone lowering in Makhuwa, in Van de Weijer, J., Los, B. (Eds.), Linguistics in the Netherlands. John Benjamins, Amsterdam, pp. 224–236.Find this resource:
van der Wal, J., 2008. Agreement in thetic sentences in Bantu and Romance, in: De Cat, C., Demuth, K. (Eds.), The Bantu-Romance Connection. A comparative investigation of verbal agreement, DPs and information structure. John Benjamins, Amsterdam, pp. 323–350.Find this resource:
van der Wal, J., 2009. Word order and information structure in Makhuwa-Enahara. LOT, Utrecht.Find this resource:
van der Wal, J., 2010. Makhuwa non-subject relatives as participial modifiers. Journal of African Languages and Linguistics 31, 205–231.Find this resource:
van der Wal, J., 2011. Focus excluding alternatives: conjoint/disjoint marking in Makhuwa. Lingua 212, 1734–1750.Find this resource:
van der Wal, J., 2012. Subject agreement and the EPP in Bantu Agreeing Inversion. Cambridge Occasional Papers in Linguistics 6, 201–236.Find this resource:
van der Wal, J., forthcoming. A note on the (non-existing) passive in Matengo. Linguistique & Langues Africaines 1.Find this resource:
van der Wal, J., forthcoming. What is the conjoint/disjoint alternation? Parameters of variation, in: van der Wal, J., Hyman, L.M. (Eds.), The conjoint/disjoint alternation in Bantu. De Gruyter, Berlin.Find this resource:
van der Wal, J., Hyman, L. M. (eds.), forthcoming. The conjoint/disjoint alternation in Bantu. Trends in Linguistics series. De Gruyter, Berlin.Find this resource:
Wasike, A., 2007. The left periphery, wh-in-situ and A-bar movement in Lubukusu and other Bantu languages. PhD dissertation, Cornell University.Find this resource:
Watters, J., 1979. Focus in Aghem, in: Hyman, L. (Ed.), Aghem grammatical structure. University of Southern California, Los Angeles.Find this resource:
Whaley, L., 1996. Kinyarwanda topics and object-subject reversal. Eastern States Conference on Linguistics 12, 354–363.Find this resource:
Yoneda, N., 2011. Word order in Matengo (N13): Topicality and informational roles. Lingua 121, 754–771.Find this resource:
Zeller, J., 2004. Relative clause formation in the Bantu languages of South Africa. South African Journal of African Languages 22, 75–93.Find this resource:
Zeller, J., 2008. The subject marker in Bantu as an antifocus marker. Stellenbosch papers in linguistics 38, 221–254.Find this resource:
Zeller, J., 2012a. Instrument inversion in Zulu, in: Marlo, M.R., Adams, N.B., Morrison, M., Purvis, T.M. (Eds.), Selected Proceedings of the 42nd Annual Conference on African Linguistics. Cascadilla Proceedings Project, Somerville, MA, pp. 134–148.Find this resource:
Zeller, J., 2012b. Object marking in isiZulu. Southern African Linguistics and Applied Language Studies 30, 219–235.Find this resource:
Zerbian, S., 2006a. Expression of information structure in Northern Sotho. Humboldt University, Berlin.Find this resource:
Zerbian, S., 2006b. Inversion structures in Northern Sotho. Southern African Linguistics and Applied Language Studies 24, 361–376.Find this resource:
Zerbian, S., 2007. The subject/object asymmetry in Northern Sotho, in: Schwabe, K., Winkler, S. (Eds.), On information structure, meaning and form: generalizations across languages. John Benjamins, Amsterdam, pp. 323–364.Find this resource:
(1) Nurse and Philippson (2003: 5) note that “the northwest languages form exceptions to many possible generalizations for Bantu.” The syntactic phenomena in the current chapter are predominantly from eastern and southern Bantu languages.
(3) Dislocation can generally be seen in a prosodic break between verb and object, and for some languages also in the conjoint or disjoint form of the verb.
(4) The status of the object marker as grammatical agreement, an incorporated pronoun, or some sort of clitic is a continuing debate in Bantu syntax, which dates at least to Bresnan and Mchombo 1987. For more information, see especially the overview in Riedel (2009) and references therein, and for recent additions to the debate, Diercks, Ranero, and Paster (2015). Marlo (2014, forthcoming), Marten and Kula (2012), Zeller (2012b).
(5) Predicative lowering (Schadeberg and Mucanheia 2000; van der Wal 2006) and metatony (Hadermann 2005, Hyman and Lionnet 2012, Makasso 2012) can also be mentioned as strategies indicating the relation between verb and following, see for further discussion Hyman (forthcoming).
(6) Another passive strategy found in Bantu languages is the 3rd plural form, in which a passive meaning like “the food was eaten (by the wild dog)” is literally coded as “they eat the food (by the wild dog)”. This 3pl strategy exists in a number of Bantu languages; see Kula and Marten (2010) for further references.