AUTHOR: Miestamo, Matti TITLE: Standard Negation SUBTITLE: The Negation of Declarative Verbal Main Clauses in a Typological Perspective SERIES: Empirical Approaches to Language Typology 31 PUBLISHER: Mouton de Gruyter YEAR: 2005
Oliver Bond, Department of Linguistics, School of Oriental and African Studies, University of London
''Standard Negation'' by Matti Miestamo is a typology of negation in declarative verbal main clauses. Based on a stratified core sample of 240 languages, this study investigates the form and distribution of negative markers in clausal negation. The study comprises a typological classification of standard negation and a functional explanation for the structural patterns encountered in the languages under examination. Like other volumes in Mouton de Gruyter's Empirical Approaches to Language Typology Series, this book approaches language phenomena from a cross-linguistic perspective. It will appeal to a typologically/functionally-oriented audience and to those working within formal frameworks that develop theory on a wide empirical base. More specifically, this study will be of particular interest to linguists working on negation, morphosyntax and/or paradigmatic asymmetry.
The book begins in Chapter 1 with an introduction to some of the issues raised in carrying out a typological study of negation. This is followed in Chapter 2 with a discussion of the theoretical and methodological issues. Chapter 3 contains the classification and Chapter 4 comprises the quantitive analysis. In Chapter 5 some functional motivations are suggested for the observations made throughout the study, and conclusions are presented in Chapter 6.
The introduction comprises an overview of the basic concepts and conventions used throughout the work. This includes the typological goals of the investigation and establishes the functionally-oriented research perspective of the author. A number of terms relevant to the study of negation are introduced here including 'scope', 'propositional logic' and the differences between 'clausal', 'standard' and 'sentential' negation. As a preview to later discussion, examples are provided to illustrate the most common ways of marking clausal negation including the use of negative particles, affixes and auxiliaries. The main tenet of the study is that when compared to counterpart affirmatives, the formal marking of negative main clauses can be subclassified in terms of structural symmetry or asymmetry. Symmetric negative structures differ from their affirmative counterparts only by the addition of a negative marker(s), while asymmetric negative structures exhibit further structural differences in addition to the presence of the negative marker(s). Asymmetric negation is further divided into various subtypes, and a full classification is provided in Chapter 3. The introductory chapter also features a short discussion of language internal complexity, including the use of several different negative structures within a single language. Finally, an overview of previous typological research on negation is provided.
Chapter 2 establishes the theoretical and methodological concerns of the study. Following an introduction to the principles of creating a suitably sized stratified sample, Miestamo indicates that the sample devised is a genealogically stratified variety sample compiled for qualitative purposes. He identifies 413 genera among the world's languages (based on Grimes 2000) and compiles a Core Sample (CS) of 240 genera-representative languages. The 173 genera not represented in the CS were judged to have insufficient or inadequate sources for inclusion in the study. In addition to the CS, a further 57 languages were also studied (skewing the representation of certain genera) on the grounds that they were already being analysed for another project. Combined, this set of 297 languages is referred to as the Extended Sample (ES). For the purposes of compiling quantitive data, a subset of the CS, the Restricted Sample (RS), is established comprising 179 languages viewed in terms of six macroareas (following Dryer 1992): Africa, Eurasia, Southeast Asia and Oceania, Australia and New Guinea, North America, and South America. In the RS, each macroarea is proportionally represented, based on the number of genera found in that macroarea. This is undertaken to further stratify the sample on an areal basis.
After delimiting the sample, Miestamo defines the domain of inquiry, namely standard negation (SN). This is achieved by employing a mixed functional/formal (i.e. structural) definition. Miestamo's definition of SN (p.42) restricts the investigation to the negation of declarative main clauses, where negation is characterised as an operator that changes the truth value of the affirmative proposition. SN structures are further defined as general and productive means of negation within a language.
The typological classification of SN structures is presented in Chapter 3. Negative structures are classified according to whether they are symmetric (S) or asymmetric (A). Only fully grammaticalized asymmetries are taken into account and (morpho)phonological phenomena are excluded. Asymmetric negatives are further divided into instances of constructional vs. paradigmatic asymmetry. Constructional asymmetries are found when a negative structure differs from its corresponding affirmative by one or more structural changes in addition to the presence of the negative marker(s), e.g. if in the negative clause an auxiliary is required that does not occur in the affirmative counterpart. Paradigm asymmetry occurs when a one-to-one correspondence between negative and affirmative paradigms is not found. For instance, in the affirmative several temporal categories may be distinguished morphologically, while in the negative paradigm, the same distinctions may be neutralised, creating an asymmetry between a number of temporally marked paradigms in the affirmative and a single negative paradigm.
Asymmetric negation is classified into four major subtypes, all of which are exemplified and explained throughout the course of Chapter 3. They are labeled as A/Fin, A/NonReal, A/Emph and A/Cat. The motivation for this classification is discussed later in Chapter 5. In A/Fin, asymmetry is found between the verbal finiteness in affirmative and negative clauses, with negatives exhibiting reduced finiteness. In A/NonReal the negative is obligatorily marked for a category marking non-realized states of affairs, while the affirmative is not. Thirdly, in A/Emph the negative structure is marked for a category expressing emphasis elsewhere in the language, while the affirmative counterpart remains unmarked for this category. Finally in A/Cat, asymmetry is found in the grammatical categories marked in affirmatives and negatives. Further subcategorization of the types identified here are also made in Chapter 3. In this chapter Miestamo also demonstrates the grounds on which particular asymmetries are considered part of a negative structure or not. Based on the regularity of certain grammatical asymmetries between affirmative and negative structures, several implicational universals are proposed throughout the classification. These relate specifically to the subtypes identified in the sample. For instance, with A/NonReal it is suggested that ''if the affirmative is marked for a category denoting non-realized states of affairs, then the corresponding negative (if specified for TAM at all) will also be.'' (p. 96).
In Chapter 4 quantitive findings are presented for the RS. Note that 'quantitive' in this sense means percentages only, and no statistical analyses are run on the sample. It is demonstrated that symmetric negation is more common than asymmetric negation overall, and that constructional asymmetry is more frequently found than paradigmatic asymmetry. Of the asymmetries observed in the RS, A/Cat is the most common (33% of the languages examined), followed by A/Fin (25%), A/Non/Real (13%) and A/Emph (2%). A further 2% of languages exhibited asymmetry not covered by any of these categories. Miestamo also observes that only one subtype of asymmetry is usually found in a single language: 79% of languages with asymmetry exhibited only one subtype of asymmetric structure. The most common means of negating declarative main clauses are proposed to be symmetric and A/Fin structures. Miestamo also details the relationship between symmetry/asymmetry and free vs. bound negative markers, and provides numerical data relating type of SN and the position of the negator. Comments on the distribution of symmetrical vs. asymmetrical negation structures across the six macroareas are also provided in Chapter 4. For instance, it is observed that A/NonReal is only common in Australia and New Guinea.
The general principles underlying the symmetric and asymmetric patterns observed in negative structures are discussed in Chapter 5. The functional asymmetry between affirmatives and negatives is examined in light of their different semantic-pragmatic properties. Analogy is proposed as the central mechanism of Miestamo's model of explanation. It is argued that symmetric negation can be explained by analogy from form to form (from the form of affirmatives to the form of negatives). Asymmetric negation is explained as the result of analogy from function to form (functional asymmetry to formal asymmetry). It is argued that asymmetric negatives have formal structures that grammaticalize different aspects of the functional asymmetry between the affirmative and negative counterparts. A/Fin asymmetry is proposed to reflect the stative quality of negatives, and encode an asymmetry between an action described in an affirmative clause and the state described in a negative clause. Asymmetry of the type A/NonReal is proposed to reflect an asymmetry between realised and unrealised modalities. Discussion of negation in a diachronic light is also provided in this chapter.
Concluding remarks are found in Chapter 6. This is followed by four substantial appendices containing further examples and data on the sample not included in the main text.
Overall, this book represents an important and timely contribution to cross-linguistic studies of negation. It proposes links between the structural form of negatives and their functional properties, with particular reference to the formal and functional asymmetries commonly found between affirmative and negative declarative main clauses. As such, it is a useful reference tool for charting the types of recurrent asymmetries encountered across languages and provides explanations for these patterns at the semantic-pragmatic level. However, a number of issues raised by the approach of this work require further comment. In particular these concern the domain of SN as defined by Miestamo and the way in which the quantitive data is employed in the current study to render conclusions.
Negation as a functional domain is vast; therefore the number of parameters that potentially interact with this grammatical category are legion. While Miestamo is clearly sensitive to this fact when delimiting the field of study, an unavoidable consequence is the effect of excluding some of the more interesting phenomena likely to be encountered in a typological survey of this type. It is clear from Miestamo's definition of SN that he wishes to restrict the survey to declaratives, but in doing so he excludes non-indicative modalities which are of interest since they may well show different asymmetry patterning to indicative clauses. While it is clear that negative structures such as those used only for negative imperatives or questions must be excluded on this principle, in Miestamo's methodology the grounds on which non-indicative declarative categories are excluded appears is language specific, and seems to depend on which other TAM categories are found in the clause. This strikes me as an opaque and potentially inconsistent approach to delimiting the contents of the sample (p.43-44). In general, it would have been helpful to be provided with more information on language internal complexity - including the use of different negative structures within a single language. Unfortunately, this avenue of investigation is not fully explored in the book.
Some aspects of the definition of SN proposed require further clarification, particularly since some of the terminology used is contentious within the broader field of linguistics. For instance, Miestamo freely refers to verbal negation without ever defining or referring to the properties of a verb or the limits of 'verbiness'. Clarification of this type is important for languages where there may be some debate over the status of the verbal category (see Croft 1991, and also Hopper and Thompson 1984, Stassen 1997). In such instances, guidance maybe needed to help decide on what grounds a negative structure is an example of SN, and at what point a structure is no longer general enough to be included in the typology. It seems that in order for a typology of verbal negation to be constructed, a typologically valid and applicable definition of what is verbal must first be explicitly established. In a similar way, the term 'auxiliary' is not defined further than some element that takes inflection and unfamiliar terminology such as 'connegative' is also used without explanation (p. 82).
The approach to negation taken in this typology builds on earlier work in the field by expanding the sample size and structure of smaller studies. Understandably, a large-scale work of this kind requires a highly restricted domain of study. However, asymmetry is only one facet of negation in language and throughout the work the way in which the domain is delimited sometimes marginalizes the importance of other aspects of negative structures which are proclaimed to be independent of negation or secondary to a principal asymmetry. This is particularly clear when a number of different asymmetries exist between an affirmative and negative clause. For instance, in the examples in (33) from Kannada (discussed on pp. 78-79) the negative structure exhibits an asymmetry of the type A/Fin. The negative clause also lacks the person, number and gender agreement found in the affirmative counterpart. Thus, there is also asymmetry of the type A/Cat/PNG. Miestamo (p.79) proposes that ''This paradigmatic asymmetry is secondary, derived from the constructional A/Fin asymmetry''. As such, A/Fin asymmetry is the only type of asymmetry noted for these examples in the quantitive analyses. Furthermore, no discussion is made of the TAM asymmetries also evident between the Kannada affirmative and negative examples. Since 'derived asymmetry' of this kind is excluded from the numerical data in Chapter 3, the figures provided therein are a somewhat skewed representation of data in the sample.
While a functional approach is purported to be held throughout the study there are often points where evidence of this position is not clear. One such issue relates to the theme running through the study that negatives in some sense 'derive' from affirmatives. While it is tempting to endorse such a model when looking at symmetric negative structures the same cannot be said for more complex asymmetric structures involving a differences in word order, reduction of finiteness and/or asymmetric grammatical marking. It is surely for this reason that Miestamo proposes at the beginning of the study that ''rather than seeing the negative marker alone as the 'standard negation strategy' and the accompanying features as 'secondary modifications', SN is seen as a construction to which both the negative element and the relevant secondary modifications belong'' (p. 21). However, much of the time asymmetries are only accounted for if obligatory at the verbal or clausal level (p. 56-59, e.g. the use of partitive case in Finnish negatives, p. 43) and in addition to the marginalisation of 'derived asymmetries' noted above for languages like Kannada, some structural differences, such as changes in word order are considered to be facets of other functional domains (e.g. information structure or stress) and not negation (p. 67-8). However, it is exactly these sorts of asymmetries that a functional typology of negation should attempt to address.
In accounting for his observations, Miestamo identifies that mere correlations in his sample are not explanations in themselves and that each pattern must be functionally motivated by processing or discourse/pragmatic means. However, his explanation that A/Fin asymmetries are linked to the stativity of negatives is not an adequate explanation in itself. This still leaves the problem of why affirmative statives do not necessarily show the same degree of reduced finiteness as their negative counterparts e.g. 'He lives in London' vs. 'He doesn't live in London'. Although many of the observations made regarding the asymmetries encountered in the study raise a number of interesting questions regarding the functional domain of negation, the reader is left wondering why certain asymmetries are found in certain languages and not others.
Where complex structural phenomena are being discussed, the absence of minimal pairs (as is fairly common throughout the study) makes it difficult to see the structure of the argumentation. While this is a reflection of the paucity of detailed available sources rather than Miestamo's approach to the topic, it demonstrates that ascertaining typological generalizations based on sources of varying qualities is problematic. This is particularly true when examples are considered not only in terms of their structure, but also their discourse function. Although examples provided are sometimes discussed in terms of their pragmatics, links between presupposition and negation are underplayed, presumably due to the nature of the resources used. Similarly, discussion of scope is largely absent.
One of the issues identified in the study, yet not addressed in any of the explanations, is the relationship between negation and tonal phenomena, stress and intonation. While it is clear that Miestamo considers automatic phonological processes to be irrelevant in determining asymmetry, it is not transparent how suprasegmental phenomena should be treated. Certainly, differences in tone indicate negation in several languages of West Africa, including in Igbo, where certain minimal pairs of affirmative and negative structures are differentiated by tone alone (p. 119). While the negative structure exhibits a neutralised aspectual distinction (and is thus categorised as A/Cat/TAM by Miestamo), the perceptible difference between some of the minimal pairs is one of pitch. Importantly, it is not the lack of a distinction between perfective and imperfective aspect that indicates to the hearer that it is a negative structure. However, it appears that since tonal morphology does not contribute to 'structural' asymmetry in a traditional sense, this aspect of negation is not deemed to be of significance in this study.
While much of the quantitive data supplied in Chapter 3 is straightforward, the relevance of some of the figures is questionable. In general, since the sample is not statistical, only subjective comments on the significance of each numerical correlation are provided and in this respect the study is somewhat lacking. Miestamo misses a number of opportunities to make the most of the data compiled in his sample. For instance, while he points out that symmetric structures are the most common types of SN marking across his sample, he does not draw attention to the fact that most of the languages in the sample have asymmetry in their negative systems. No numerical data is provided on the observed types of internal variation. For instance, it would have been helpful to know what proportion of languages in the RS has more than one SN structure and how languages with multiple SN structures are distributed. Also from a diachronic perspective, including an analysis of data on those negative structures excluded from the sample on the grounds of being non-productive/non-general would also have been enlightening. Such a comparison may have enabled further analysis of historical aspects of this domain. This would not have involved much extra work since the data would already have been reviewed in order to be excluded from the existing sample. Additionally, one type of numerical data absent from this study are frequency counts. While correlations concerning this aspect of language use are beyond the scope of Miestamo's study and unlikely to be found in the resources available to the author, it is probable that data of this kind would reveal much about the asymmetries found in the sample, particularly asymmetries in the marking of grammatical categories within a historical perspective.
Overall, this study provides an interesting overview of asymmetry in negative declarative main clauses. As the first thorough typology to consider negation in terms of the structural and functional asymmetries between affirmative and negative clauses it is an important step towards more expansive studies of both negation and asymmetry in general.
Croft, William. 1991. Syntactic categories and grammatical relations: the cognitive organization of information. Chicago: Chicago University Press.
Dryer, Matthew S. 1992. The Greenbergian word order correlations. Language 68.81-138.
Grimes, Barbara F. (ed.) 2000. Ethnologue Vol. 1 Languages of the World, Vol. 2 Maps and Indexes. 14th ed. Dallas: Summer Institute of Linguistics.
Hopper, Paul J, and Thompson, Sandra A. 1984. The discourse basis for lexical categories in Universal Grammar. Language 60.703-752.
Stassen, Leon. 1997. Intransitive predication. Oxford: Oxford University Press.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Oliver Bond is a Postdoctoral Research Fellow in the Endangered Languages Academic Programme (ELAP) at the School of Oriental and African Studies, London. His principal research interests lie in typology, historical linguistics and language documentation, including fieldwork on Eleme (Ogonoid, Benue-Congo), an under-described language of southeast Nigeria. His current research post involves devising a fieldwork questionnaire for gathering data on negation.