Startseite / Archive / 2009 / Derek Nurse 2008. Tense and Aspect in Bantu.
Artikelaktionen

 

<1>

Written by a Bantuist who has worked in the field of Bantuistics for roughly three decades, this book presents a long awaited effort to assemble and ponder over what is known about tense, aspect, and related subjects in the Bantu languages. Bantu languages, which are spoken south of a line vaguely drawn from northwestern Cameroon to southern Somalia, typically encode a high number of tense aspect categories using a multitude of grammaticalized forms that pay much attention to semantic and functional detail. This book tries to systematize this rather complex situation and identify and describe the general characteristics of how the Bantu languages treat tense and aspect. To achieve this goal the author relied predominately on data from secondary sources and evaluated a sizable number of 210 plus languages. (To provide a quick comparative reference: the sample of Dahl’s (1985:42) study was 64 and Bybee et al. (1994:31) investigated a maximum of 94 languages.)

<2>

Chapter 1 presents the introduction, which comprises a description of the methodology used for this study. Chapter 2 explains the Bantu verb morphology. This is followed by two chapters that deal with the core topics of this book: tense (chapter 3) and aspect (chapter 4). Although the emphasis of this work lies on the grammatical categories of tense and aspect, Nurse decided to include a discussion of negation, verbal focus, and pronominal object markers. In Bantu languages these categories often form an integral part of the verbal morphology, which interacts with tense and aspect morphology and, in part, shares some of the loci of encoding on the verb forms. These issues are dealt with in chapter 5. The final two chapters deal with a more or less tentative reconstruction of Proto-Bantu tense-aspect forms (chapter 6) and the processes of change (chapter 7) that led to the current state of tense and aspect encoding in Bantu languages. The book is completed with a section in which definitions of the most central terms are given, a bibliography that includes more than just the references cited in the text, a source language reference index and a general index. Apart from the physical book two appendices are downloadable as pdf files from an open access website. They contain annotated tables that list the tense and aspect forms for 146 Bantu languages.

<3>

When looking at this overall structure, it is apparent that the subject(s) of each of these chapters might very well form the single topic of a separate book; and while the single chapters do rely on each other to some extent and are cross-referenced extensively, each can be read on its own. (That is, if one approaches the book with prior knowledge about Bantu languages’ grammar.) The following discussion of the book roughly orients itself along the broad structure introduced through the chapters. Thus at first the methodology used and the central topics of the book are presented and discussed after which a general evaluation completes this review.

<4>

The base of languages from which material has been evaluated is quite large. 210 plus languages are a lot of languages to consider in such a study. Nurse’s way to deal with this situation consists of dividing the languages into two databases: one for statistical computations and one as a source of descriptive language material. The first one numbers 100 languages chosen equally from Guthrie’s (1967-71) zones and groups. The second group contains the first database and the rest of the languages for which reasonable data was available. In preparation of this book Nurse went on to organize the highly heterogeneous data – heterogeneous in terms of quality of description, theoretical background of description, grammatical coverage – on these languages in tables (or matrices as he calls them) that are constructed with rows of tense distinctions and columns of aspect distinctions. For the tense and aspect distinctions in these tables (as well as the rest of the book) he relied to a great extent on conventional terminology such as past, future, perfective, anterior (perfect) etc. as well as terms denoting more specialized morphological categories such as persistive or inceptive. This patterning via matrices was done based on the assumption that any verb form in Bantu is likely to be marked for tense as well as a combination of aspects. To illustrate this a thinned out matrix, i.e. minus the negative forms, for the Bantu language Lega (D25) has been included.

Table 1: Lega tense-aspect matrix

Perfective

Imperfective

Progressive

Persistive

Inceptive

-ag-

-ko-

-ké-

-sǎ-/sě-

P 3

-a- ... -elɛ́

tw-a-bolót-elɛ

we pulled

tw-a-bolót-ág-á

we were pulling

P 2

-ǎ- ... -elɛ

tw-ǎ-bolót-elɛ

tw-ǎ-bolot-ag-elɛ́

tw-ǎ-bɛz-ag-elɛ to-ko-kangól-á

we were clearing

P 1

-a-

tw-a-bolót-á

we pulled

tw-ǎ-bolot-ag-a

tw-ǎ-bɛz-ag-a to-sǎ-kangol-a

we had started clearing

-(i)ko-

to-ko-bolot-ag-a

to-ko-bolót-á

we are pulling, will pull

to-ké(-le) tw-a-kangol-a

we are still pulling (sic!)

to-sǎ-kangol-a

we have already cleared, started to clear

F 1

-ǎ- ... é

tw-ǎ-bolót-ɛ́

tw-ǎ-bolót-ág-ɛ́

tw-ǎ-b-ɛ́ to-sǎ-kangol-a

we will already have cleared

F 2

-ka-

to-ka-bolot-a

we will pull

to-ka-bolot-ag-a

<5>

As should be evident from this table, Nurse embraces the structuralist view of languages as a conceptual basis for this book. As can therefore be expected, he postulates that tense and aspect represent a system based on forms whose meaning arises through contrast with other forms within the verbal paradigm. Unfortunately, further crucial information as far as methodology is concerned is not discernible in the introduction or anywhere else in this book, and the reader is left to hypothesize that the ordering and categorizing of the information in the sources somehow lead to the description in this book. But exactly how is not clear. A comparison of the individual matrices of the various languages is not taken up in the book, in contrast to what one might expect after reading the introduction. In addition, nothing is said on the statistical computations. The reader is, again, left to assume that it was a simple order-categories (or forms etc.)-and-count-in-how-many-languages-they-are-found kind of approach.

<6>

The treatment of the verbal structure is detailed and deals with different types of structures: from the imperative over single inflected verbs to multiverbal constructions. It concentrates, however, on describing the single inflected verb structure. This structure can get very intricate and allows the highly agglutinating Bantu languages to load a lot of aspectual, temporal, locational, syntactical, and other information on the verb through a series of pre- and suffixes. These affixes usually follow a set order, and each grammatical category involved in the verb tends to be encoded at a particular position (or ‘slot’) or a combination of positions. Nurse presents and discusses this structure principally in terms of a more traditional, syntactically informed sequence. This sequence is based on Meeussen (1967) and runs:

Pre-initial – initial – post-initial – formative – limitative – infix – radical – suffix – pre-final – final (final vowel) – post-final

Nurse describes these positions in some detail and talks about what kind of grammatical categories are encoded at each position. Reading this section definitely helps any non-Bantuists to get to know and understand Bantuistics’ more idiosyncratic grammatical terms. This chapter can equally serve as a quick reference for Bantuists to see, for example, what information the different positions generally carry in the Bantu languages.

<7>

The morphotactical sequence just mentioned also seems to form the dominant underlying structural concept for the descriptions in the remainder of this book. This is somewhat unfortunate, since in recent years – influenced by a more general linguistic perspective on the Bantu verb – the terminology and structure of the above sequence has been altered somewhat. See, for example, Güldemann (1999). These modifications resulted in terms that are more descriptive of the functions usually encoded at the various slots and conflated some of the above positions into a single slot. That is, the formative and limitative have been united to form a TA slot and the markers encountered at pre-final position have either been allotted to the suffixal extensions or to the final vowel position. The resulting morphotactical sequence runs as follows:

Pre-SM – SM – Neg2 – TA – OM – Root – extension – FV – Post-FV

Nurse duly acknowledges these changes, deals with them in a separate sub-chapter, and adopts the terminology on several occasions, but he does not use the terminology consistently. The mixing of terminologies belonging to the two sequences just discussed results in some confusion as the two sets of terms are not always fully compatible.

<8>

The two chapters on tense and aspect are data driven and touch upon all the important issues of the encoding of tense and aspect in Bantu languages. Nurse explains, among many things, the past and future tenses found in many Bantu languages, which are organized along degrees of remoteness from the present; it deals with the difficult descriptive issue common to many Bantu languages of determining, if a given form should rather be classified as an anterior (i.e. perfect) or a near past form; he deals with the more general aspectual categories, such as perfective, imperfective, anterior (i.e. perfect) etc., and describes how they are usually encoded in the Bantu languages; he discusses the appearance of two anterior tenses in some Bantu tense aspect systems; he deals with morphological (aspectual) categories more idiosyncratic to Bantu languages, such as the situative or persistive *-ki-; he touches upon regional tendencies of encoding tense and aspect, and he elaborates on the combination of tense and aspect in single or complex verb forms. Throughout his discussion, as well as throughout the book, Nurse interweaves comments about historical issues and, at times, he ventures into other, related topics such as questions of mood and modality, relativization, verbal derivation. The description of tense and aspect is also accompanied by several statistical observations. In general the two chapters exhibit many features of a thoroughly informed essayistic discourse fueled by 30 years of experience with Bantu languages. This approach is on the one hand his forte – one can find many interesting ideas and thoughts that might have been dropped in a more structured approach, such as his observation that perfects (or anteriors) are rarely analyzed in Bantu languages, if the author of the analysis is a native speaker of French or German – but on the other hand it is also its weakness. It does not, as a whole, present a leveled and measured account of the issues. The quality and quantity of the information presented in these chapters is certainly, as Nurse points out in the introduction, a reflex of the available data, but this does not explain Nurse’s choice of where to lay the focus of his analyses. For example, while the different instantiations of the morphological persistive/continuative categories indicated by *-ki- are dealt with over a little less than five pages and seven illustrative examples, the general discussion of the perfective is allotted little more than two pages with only one illustrative example.

<9>

Concerning the inclusion of chapter 5, I believe, together with the author, that it is helpful for the understanding of the verb to introduce a section on categories such as negation, focus, and object marking. In Bantu languages these are encoded in a similar manner to tense and aspect, and thus they interact with aspect and tense markers, and sometimes share a locus of encoding on the verb. The principal accomplishment of Nurse’s book in chapter five is that he gathers the information dispersed over many single and sometimes not easily available publications and combines them into a unified discussion. This is not only helpful for non-Bantuists who try to understand how these categories work in Bantu but also for Bantuists who can use it to orient themselves quickly as to where Bantuistics stand with regard to these issues.

<10>

Nurse’s interest in Bantu languages throughout his career always showed a very strong historical motivation and, therefore, it is not surprising that, to finish off the book, he devotes a sizable part of this work to historical reconstruction of tense-aspect morphology in Bantu and the processes of change associated with it. By far the strongest section of the book, I recommend a non-Bantuist to start reading the book with chapter 7. In this chapter Nurse discusses the principal, i.e. reconstructable, morphemes involved in tense-aspect encoding, such as *-a- for different pasts, the *-ile anterior, the itive/narrative *‑ka-, the near past vowel suffix *-ı etc., and he describes the range of meaning they show across the Bantu languages. In this manner one is introduced to the most common and recurring morphemes and the principal semantic issues involved. It is a pity, however, that the author seems reluctant to acknowledge that the idea of a reconstructable Proto-language for Bantu, which underlies this chapter, did not go unchallenged among Bantuists. Many years before the concept of ‘extreme layering’ was noted by Dixon (1997) for the Australian linguistic area, some of the strongest critiques of the possibilities of reconstructing Proto-languages came from the Bantuist camp in the publications by Möhlig (1976, 1977, 1981). It is deplorable that this critique and the accompanying publications are never mentioned in a book that puts such an emphasis on historical phenomena. (The only concession to this critique is made in a very general footnote, which simply states that the concept of a Proto language is not universally accepted.) While it is certainly legitimate to stay with the concept of a Proto-language, one starts to wonder when reading this chapter – which does not provide comparative reconstructions of its own, but relies on previous reconstructive work – how the areal and semantico-functional heterogeneity provided by the reflexes of the reconstructed elements, which Nurse neatly maps and surveys in this chapter, would lead to a homogeneous picture of linguistic developments all the way from the alleged Proto-language to the single languages. By not presenting a reconstruction of his own Nurse is able to neglect that convergence phenomena, so infinitely common in Bantu languages, certainly influence and determine, if not dominate, the processes of dispersal of the grammatical features across the Bantu languages. In this manner he avoids to address the growing problems of presenting a homogeneous picture of historic development under the auspices of the Proto-language model once more and more languages are added to the reconstruction efforts. The reader is thus left with a false sense of being presented with a more or less uncontroversial picture, and this effect is not sufficiently addressed by stating that the reconstructions are highly tentative.

<11>

In the last chapter Nurse deals with the changes from the original Niger-Congo – the phylum under which Bantu languages are categorized – grammatical system to the Bantu languages, such as the addition of tense to a system that principally marked aspect, the change from an analytic to a synthetic structure, and the change from SVO to SOV that some Bantu languages underwent. It looks specifically at how certain independent non-verbal items were introduced into the verbal structure by first becoming clitics and then suffixes. The origin of the subject markers on the verb, for example, followed such a sequence from independent pronoun to proclitic pronoun to prefix. And finally Nurse discusses the origin of the main verbal categories, such as progressive, imperfective, near past, anterior, subjunctive etc. and the grammaticalization paths they are/were involved in. The discussion of these phenomena remains sketchy and is most useful as a reference to these phenomena in the Bantu verb.

<12>

At this point in this review I deem it important to mention and elaborate a little more extensively on a central aspect of this work that I found very frustrating: the effects of Nurse’s strong reliance on the structuralist perspective. This conceptual notion is so powerful that in the discussion of the perfective, to take but one example to illustrate this point, Nurse goes through an involved argumentation on the general meaning of perfective in which he sustains a possible application for the notion of perfective to the present. He does this, ultimately, to be able to create a fundamental perfective/imperfective opposition for two languages, Lega (D25) and Bukusu (E31), whose tense aspect system he seems to view as representative for Bantu languages in general. But unfortunately, this heavy reliance on structuralism and the attempt to find a fundamental perfective/imperfective opposition causes the discussion on the perfective in Bantu to miss out on the actual semantic content(s) encountered in Bantu languages in general. The following excursion serves to elaborate this point in more detail.

<13>

His discussion of the perfective basically reiterates the more general prototypical view as a single bounded whole with the addition that it be applicable in present contexts in the case of Bantu languages. The inclusion of a present tense form in the perfective/imperfective sets which are used as illustration in the text – (1) demonstrates this for Lega – explains the motivation for Nurse to try and reconcile perfective aspect and present tense, which, as he rightly says, is viewed by many authors to be limited to the past (plus, in some cases, the future). And he never mentions the narrative functions, i.e ‘event sequencer’, often associated with and indicative of the perfective aspect (Hopper 1982:9). In this way he is able to construe a neat fundamental perfective/imperfective opposition.

Table 2

(1)

Perfective

Imperfective

tw-a-bulút-įle

we pulled (P3)

tw-a-bulút-ág-á

we were pulling

tw-ă-bulut-įle

we pulled (P2)

tw-ă-bulut-ag-įlé

(no translation given)

tw-a-bulút-á

we pulled (P1)

tw-a-bulut-ag-a

(no translation given)

tu-ku-bulút-á

we pull

tu-ku-bulut-ag-a

we are pulling

tw-ă-bulút-é

we will pull (F1)

tw-ă-bulut-ág-é

we will be pulling

tu-ka-bulut-a

we will pull (F2)

tu-ka-bulut-ag-a

(no translation given)

<14>

In (1) above the imperfective is marked by the suffix -ag- and the forms in the left hand column are, according to Nurse, either unmarked or ‘lightly marked’ for aspect, i.e. perfective, which seems to indicate that Nurse considers the forms on the left to be notionally perfective, since there is obviously no morpheme that marks perfective. In other words, while imperfective is a morphological category, perfective is, as one has to infer from his description, a mostly notional category in Bantu languages induced merely through the opposition with the imperfective forms.

<15>

This way of handling of the description of the perfective has two outcomes. Firstly, no room is left for the possibility that not all grammatical forms might carry the notion of perfective semantics only because they are systematically opposed to it. Secondly, the discussion misses out on describing the various semantic implications of the perfective category in the Bantu languages. The discussion fails to indicate just what exactly the notion of perfective entails. It also fails to clearly state, if the perfective notion in the forms above is obligatory or if the main purpose of these forms is simply to indicate the general fact that these events happened (are happening, or will happen). That is, is it used more in the sense of a general factual meaning and called perfective, or does it have more salient perfective semantics. And if it has salient perfective semantics, is the perfective in Bantu languages rather concerned with the totality of an event, to use one of Dahl’s terms, or its boundedness? The whole section on the perfective seems to be solely motivated by its structural opposition to an imperfective, but other than that it is semantically inconclusive. At this point the reader gets more information only if he looks into the definitions section of the book. There, apart from the criterion of structural opposition, Nurse states:

Perfective is here used to refer to pasts and futures. Perfective is similar or identical to what others call ponctuel (Aroga Bessong and Melcuk(sic!) 1983), factitive (sic!) (Welmers 1973), performative (Hewson and Bubenik 1997), accompli (Francophone writers), completive (Bybee et al. 1994: 57).

<16>

Thus, the semantical range of the term perfective as used here seems to be rather broad, from instances with a factual (e.g. Welmers’ (1973) factative) meaning to instances that indicate the utter completion of an event (Bybee et. al’s completive). Unfortunately, this breadth of meaning is not touched upon in the section on perfective, and it is left to the reader to try and guess what meaning of perfective is adopted whenever it occurs throughout the book. The definition above implies that Bantu languages are highly heterogeneous as far as the semantic content of perfective is concerned. The reader is left to wonder, however, how this heterogeneity is effected in the individual languages. And with respect to the grammaticalization paths in the last chapter the question arises how these can be postulated, if the extent of the functional spectra of the grammatical categories is not taken into account? A look at the language Lega in the appendix brings further inconsistencies of presentation to light. The first thing to notice is the absence of the above present tense opposition in his table (see table 1). And in addition one finds an additional present progressive form, to-ko-bolót-á ‘we are pulling, will pull’. Incidentally, this form marks aspect as a prefix, -ko-, and one notes the absence of the imperfective marker in the finite form. This makes one wonder immediately how the form relates to the present imperfective above, since both forms are translated without further comment by an English present progressive.

<17>

One might additionally remark that the perfective/imperfective opposition could be analyzed, at least for Lega, as a lexical function, since the marker -ag- seems to be treated by that language more as a derivational morpheme and thus as belonging to the realm of word formation and not, as the tense markers, to inflection. But, what I believe Nurse is trying to accommodate in his very general analysis is the fact that in other Bantu languages, e.g. Bukusu (E31c), the status as a derivational morpheme is not that clear. Still this failure to discuss the lexical/inflectional distinction in the discussion of the perfective category for some Bantu languages can be mentioned as an unwarranted omission.

<18>

Before coming to the end of this review, I would like to remark that the book contains some factual errors, which are probably unavoidable in a publication of this scale. This is taken up, however, because these errors bear on the interpretation. For example, the statement on page 22 that the hodiernal past time period for Bantu languages starts with sunrise of the same day should at least be relativized. Several, if not most, Bantu languages start the hodiernal past time period at sundown of the previous day, as indicated, for example, for Ewondo (A72a) by Angenot (1971: 71).

<19>

In Nurse’s defense it should be said that he is aware of many of the shortcomings mentioned above and he addresses them with, in my view, mixed success in the introduction. But notwithstanding the critical observations a publication of this wealth of insight is certainly a welcome addition to the study of the Bantu languages in particular, and typology in general. In the end, once the reader starts to interact with the book more thoroughly, consults some of the sources given, and looks past the sometimes difficult to follow meandering lectural style of textual composition and the rather loosely used terminology, the book turns out to be a quite wonderful trove of ideas and insights into the TA-cum-other-categories systems of Bantu languages. The book presents a good effort at taking stock, it looks at and describes where mainstream Bantuistics currently stand with relation to tense aspect and identifies gaps in the knowledge and indicates possible further avenues of research. Consequently, it is also a very good starting point to delve into the subject of Bantu TA. It assembles references to many grammatical descriptions, some of which are hard to find, and it appraises a high number of specialized articles dispersed over a multitude of journals. One should, however, always keep in mind that some of the general remarks are on occasion too general and hide the fact that many Bantu languages might turn out to behave quite different.

References

Angenot, J.-P. 1971

Aspects de la phonétique et de la morphologie de l’Ewondo. (Ph.D thesis, University of Leiden.) Brussels: Print-Express

Aroga, Bessong and I. Mel’čuk 1983

Un modele de la conjugasion bafia (à l’indicatif). Bulletin of the School of African Studies 46, 3:477-528

Bybee, Joan, Revere Perkins and William Pagliuca 1994

Evolution of Grammar. Tense, Aspect, and Modality in the Languages of the World. Chicago: University of Chicago Press

Dahl, Östen 1985

Tense and Aspect Systems. Oxford: Blackwell

Dixon, Robert M.W. 1997

The Rise and Fall of Languages. Cambridge: Cambridge University Press

Güldemann, T. 1999

The genesis of verbal negation in Bantu and its dependency on functional features and clause types. In: J-M. Hombert and L. Hyman (eds), Bantu Historical Linguistics. Theoretical and Empirical Perspectives. Stanford: Center for the Study of Language and Information, pp.545-87

Guthrie, M. 1967-71

Comparative Bantu. 4 vols. Farnborough: Gregg Press

Hewson, J. and V. Bubenik 1997

Tense and Aspect in Indo-European Languages. Theory, Typology, and Diachrony. Amsterdam and Philadelphia: John Benjamins

Hopper, Paul J. 1982

Aspect between discourse and grammar. An introductory essay for the volume. In. Paul J. Hopper (ed.), Tense-Aspect. Between Semantics & Pragmatics. Amsterdam: John Benjamins, pp.1-18

Meussen, A. E. 1967

Bantu grammatical reconstructions. Africana Linguistica 61,3:79-121.

Möhlig, Wilhelm J.G. 1981

Stratification in the History of the Bantu Languages. Sprache und Geschichte in Afrika, 3:251-316

Möhlig, Wilhelm J.G. 1977

'Zur frühen Siedlungsgeschichte der Savannen-Bantu aus lauthistorischer Sicht.' In: Wilhelm J.G. Möhlig, F. Rottland and Bernd Heine (eds.), Zur Sprachgeschichte und Ethnohistorie in Afrika. Neue Beiträge afrikanistischer Forschungen (Festschrift Köhler). Berlin: Dietrich Reimer: 166-193

Möhlig, Wilhelm J.G. 1976

Guthries Beitrag zur Bantuistik aus heutiger Sicht. Anthropos 71: 673-715

Welmers, William 1973

African Language Systems. Berkeley: University of California

Lizenz

Volltext

Kommentare

Es liegen noch keine Kommentare vor.

Möchten Sie Stellung zu diesem Artikel nehmen oder haben Sie Ergänzungen?

Kommentar einreichen.

Inhalt
Navigation