This book presents a new theory of grammatical categories - the Universal Spine Hypothesis - and reinforces generative notions of Universal Grammar while accommodating insights from linguistic typology.
AUTHOR: Olawsky, Knut J. TITLE: Aspects of Dagbani Grammar SUBTITLE: With special emphasis on phonology and morphology SERIES TITLE: LINCOM Studies in African Linguistics 41 PUBLISHER: Lincom GmbH YEAR: 2008 (1999)
David Erschler, Independent University of Moscow, Russia
The book under review is a re-edition of Olawsky's 1998 Düsseldorf doctoral thesis. The book is very far from being a reference grammar, but despite 10 years since its first publication it still remains the principal source of reference on Dagbani, a Northern Gur language of Northern Ghana. Since then, only a few works on Dagbani have appeared (Olawsky 2002; Hyman & Olawsky 2003; Olawsky 2004; Hudu 2005; Purvis 2007; Purvis 2008).
The book is addressed primarily to Gur scholars. As the presentation is often rather sketchy, a typologist in need of Gur data to include in her or his sample would probably have to look for a more comprehensive grammar.
After a short introduction there follows chapter 2, Lexicon and Syntax. This chapter lists the principal grammatical categories of the language (nouns, verbs, adjectives, adpositions, various types of pronouns). The somewhat mysterious distinction of ''true'' and ''false'' adjectives is cleared up in Olawsky (2004). A separate section of the chapter deals with verb functions (tense, aspect, and modality). Further, main features of Dagbani syntax are described: it is mostly head-initial (SVO, nouns precede adjectives, however, there exist both prepositions and postpositions). Two phenomena strike me as unusual (probably they are less surprising for experts on Gur): first, special markers (depending on the TAM characteristics of the verb) show up when the verb is clause final (p. 31, the gloss FIN is mine):
(1) a. o nyu_ri_ kom he drink.IPF water 'He drinks water.'
b. *o nyuri he drink.IPF 'He drinks.' (intended)
c. o nyu_ra_ he drink.FIN 'He drinks.'
It should be noted, however, that verbs in all examples illustrating this phenomenon are either transitive or unergative; it is unclear what happens to unaccusative ones.
The second surprising phenomenon is that interchanging the subject and the object of a sentence sometimes produces a passive reading (and does not always interchange the roles of the agent and the patient, as it might be expected)
(2) inversion producing the ''normal'' semantic effect, p. 65: a. paga maa nya adam woman DEF see Adam 'The woman saw Adam.'
b. adam nya paga maa Adam see woman DEF 'The woman saw Adam.'
(3) inversion producing the passive reading, p. 65: a. bugim di bua fire burn goat 'Fire burnt a goat.'
b. bua di bugim goat burn fire 'A goat was burnt by fire.'
The factors governing the choice of one of the readings are not discussed, but Olawsky conjectures that such 'passive inversion' is allowed only when the distribution of thematic roles can be unambiguously derived from the context.
Chapter 3, Morphology: In the first section of this chapter main grammatical categories of the language are listed again. In Section 2, the author describes the inflectional morphology. He deals first and foremost with nouns, refining the extant definition of Dagbani noun classes (Wilson 1972). Dagbani has a number of semantically non-transparent noun classes, characterized with particular singular and plural suffixes: pag-a woman-SG / pag-ba woman-PL, class 2a in Olawsky's numbering (p. 84), gab-ga rope-SG 'rope', gab-si rope-PL 'ropes', class 3a (p. 85), etc.
The subsection on the verbal morphology begins with a description of aspect markers. Then the imperative form is discussed (it is not mentioned at all in the section on verbal functions in ch. 2, and the prohibitive marker is absent from the subsection on negation in ch. 2, p. 49).
Section 3 describes the derivational morphology. Listed are suffixes used for noun derivation. For verbs, only the causative suffix is mentioned. It remains unclear from the text whether verbs form an open class and if so, how new verbs can be formed.
Section 4 lists possible types of compounding in Dagbani. This information is used later in the section on tone in chapter 4. All compounds listed here are nouns.
Section 5 discusses reduplication. Olawsky shows that reduplication is not a productive phenomenon in Dagbani.
Section 6 deals with numerals. The last section of this chapter describes an experiment on mental representation of noun classes using nonce-words.
A long chapter 4 is subdivided into 2 parts. The first part, Prosodic Structure, treats syllable types, stress, and tone (mostly in isolated nouns and noun phrases).
Olawsky shows that Dagbani usually has penultimate stress. The few exceptions with the stress on the ultimate syllable are analyzed via catalexis (i.e. positing a fictitious additional final syllable), p. 177. Later it is shown that such nouns demonstrate certain irregularities with respect to their tone pattern as well, and this effect is also explained using catalexis, p. 190.
The language possesses a two-tone system. A pre-OT autosegmental analysis of it is presented on pp. 195-229. This analysis is extended to verbs in Hyman & Olawsky (2003).
The second part, Segmental Phonology, discusses the phoneme inventory of the language. It touches on some subtle issues of phonemic status of certain sounds and describes the ATR harmony in Dagbani. The conclusion is ''that the solely phonetic status of vowel harmony must be emphasized, whereas this phenomenon seems to play a more dominant role in other Gur languages'', p. 249.
The book ends with the bibliography (pp. 272-283; books specifically on Dagbani; other linguistic literature, and books in Dagbani are listed separately), a map of Ghana (p. 284), the complete list of noun classes (pp. 285-286), a list of reduplications (p. 287), a list of contributors (p. 288), Abbreviations (pp. 289-290), and a subject index (pp. 291-293).
The book contains a lot of extremely interesting data but a number of unfortunate decisions make it less informative and harder to read than it could otherwise be.
The first of them is the tendency of the author to use the orthography without doubling it with the morphonological transcription: as Olawsky shows (see, for instance, p. 13), the orthography is rather inexact, and potentially important information is thus probably lost to non-experts on Dagbani. The decision is motivated by Dagbani's being a written language: ''literacy has been making progress recently and has to be promoted by all means'' (p. 6). I am inclined to doubt that choosing the imprecise orthography over the morphonological or phonetic representation in a book whose primary audience are linguists is a particularly efficient way of promoting literacy among Dagbani speakers.
Orthographical examples are put in angular brackets, phonological/morphological representation in slashes, and IPA phonetic transcription in square brackets. It took me some time to memorize which type of bracket stands for which. The phonetic transcription is apparently often inexact, as it tends to disregard the ATR harmony.
Another decision that seems to me unfortunate is to avoid marking tone in all sections of the book except the one specifically dealing with tone.
The reading of the book is not facilitated by very small print and the somewhat strange layout. Turning to more substantial issues, the book seems to suffer from a number of omissions and occasional unclear arguments. I will mention here only some of these points, for the purpose of illustration.
Syntax of adverbs is not treated separately anywhere, and it is only from examples that a reader may deduce that they probably follow the verb. Without a special discussion it is unclear why ''time depth markers'' (p. 34: 'earlier same day', 'one day away', etc.) are not a subclass of adverbs but a closed category on its own. In the section on focusing (3.3.7., p. 64) only examples of time and place adverbials are given. It is unclear whether others types of adverbs can be fronted as well, what is the ordering of adverbs, etc.
An argument that struck me as odd is that the ''about to'' particle _yen_ is denied the status of tense marker on the reason of its ability to co-occur with a time-depth marker (p. 33). However, examples (52 c,f) on p. 35 show that the future marker _ni_ can also co-occur with them.
On p. 19, it is stated that ''Adpositions can be either prepositions or postpositions, the latter type of which occurs directly after the noun.'' However, examples on pp. 17-18 show that 'true' adjectives also placed immediately postnominally. I wonder whether it means that ''directly after the noun'' should actually read ''directly after the noun phrase'' or that adjectives and postpositions are in complementary distribution.
p. 48: ''Whereas raising sentences admit a 'dummy' subject (<di>, cf. (23), a table of pronouns), control verbs require a 'normal' NP, i.e. the subject as the controller is a pronoun or a noun (NP). Raising verbs in this sense are not attested in Dagbani (…).'' As no examples of ''raising verbs'' are given to clarify the point the statement remains rather confusing.
The work (Vogel 1997) quoted on p. 27 is absent from the bibliography.
All of that probably does not create any difficulties for Gur scholars but make the book not an easy read for the non-initiated. However, the book presents more than enough data to whet a typologist's appetite. It remains to hope that someday a full scale reference grammar of this fascinating language will be written and published.
Hudu, Fusheini. 2005. Number marking in Dagbani. MA thesis. University of Alberta. L.M. Hyman & Knut Olawsky. 2003. ''Dagbani verb tonology''. In Chege Githiora, Heather Littlefield & Victor Manfredi (eds.), Trends in African Linguistics 4, 97-108. Trenton, N.J.: Africa World Press, Inc. Olawsky, Knut J. 2002. What is a word in Dagbani? In: Word. A cross-linguistic typology, edited by R. M. W. Dixon and Alexandra Y. Aikhenvald. Cambridge: Cambridge University Press, 205-226. Olawsky, Knut. 2004. What is a noun? What is an adjective?. Problems of classification in Dagbani. JALL 25 (2004), 127-148 Purvis, Tristan Michael. 2007. A Reanalysis of Nonemphatic Pronouns in Dagbani. In and Stephen Wechsler (eds) The Proceedings of the Texas Linguistic Society IX Conference: The Morphosyntax of Underrepresented Languages. pp. 239-264. CSLI Publications. Stanford, CA. Purvis, Tristan Michael. 2008. A linguistic and discursive analysis of register variation in Dagbani. Ph.D. thesis, Indiana University. Wilson, W.A.A. 1972. An introductory course to Dagbani. Tamale: GILLBT.
FIN marker of the clause-final position of a verb IPF imperfective aspect NP noun phrase OT optimality theory PL plural SG singular
ABOUT THE REVIEWER
ABOUT THE REVIEWER:
David Erschler holds a PhD in Mathematics from Tel Aviv University, Israel.
He is a lecturer at the Independent University of Moscow, Russia. His main
interests include Ossetic syntax, areal influences on Ossetic grammar,
Uralic languages, and syntactic typology.