Standard negation can be defined as the basic way (or ways) a language has for negating declarative verbal main clauses. Negative constructions that fall outside standard negation include the negation of existential, copular or non-verbal clauses, the negation of subordinate clauses, and the negation of non-declarative clauses like imperatives (see Chapter 71). These negatives are not taken into account here, but it is of course possible that languages use their standard negation constructions for the negation of these clause types too.
This map shows how symmetric and asymmetric standard negation are distributed among the languages of the world. In symmetric negation the structure of the negative is identical to the structure of the affirmative, except for the presence of the negative marker(s). In asymmetric negation the structure of the negative differs from the structure of the affirmative in various other ways too, i.e. there is asymmetry between affirmation and negation. Affirmative and negative structures can be symmetric or asymmetric in two ways: there can be (a)symmetry either between the affirmative and negative constructions, or between the paradigms that the affirmative and negative constructions form. Symmetric negative constructions do not differ from the corresponding affirmative constructions in any other way than by the presence of the negative marker(s), whereas asymmetric negative constructions show structural differences in comparison to the corresponding affirmative constructions. In symmetric paradigms, all (verbal) categories or forms have corresponding affirmative and negative forms, whereas in asymmetric paradigms such one-to-one correspondences do not obtain.
Three types of languages are distinguished on the map: Type Sym — languages where standard negation is always symmetric; Type Asy — languages where standard negation is always asymmetric; and Type SymAsy — languages where both symmetric and asymmetric standard negation is found.
|Go to map|
|Symmetric standard negation only: Type Sym||114|
|Asymmetric standard negation only: Type Asy||53|
|Symmetric and asymmetric standard negation: Type SymAsy||130|
The structural differences, i.e. the asymmetry, can show up in different grammatical domains, and according to the domain and nature of the asymmetry, subtypes of asymmetric negation can be established. In the present map the subtypes of asymmetric negation play no role; these subtypes will be discussed in Chapter 114. I will now give examples of the Types Sym, Asy and SymAsy. Examples of asymmetric negatives can also be seen in Chapter 114.
In Type Sym negation is always symmetric, i.e. no asymmetry is found — neither constructional nor paradigmatic. In German (see 1a-b) the negative construction is symmetric since only the negative marker nicht is added to the corresponding affirmative and no other structural differences are found between the affirmative and the negative.
(1) German (personal knowledge)
‘I do not sing.’
singen ‘to sing’,
ich singe nicht
ich sang nicht
ich habe gesungen
ich habe nicht gesungen
ich hatte gesungen
ich hatte nicht gesungen
The small sample of the German verbal paradigm given in (1c) illustrates the fact that all affirmative verbal forms have a corresponding negative form, i.e. all forms can be negated and the negative paradigm is thus symmetric with the affirmative one. No asymmetry, constructional or paradigmatic, is found, and German is thus classified as a language of Type Sym.
‘She’s going seawards.’
‘She’s not going seawards.’
(3) Shipibo-Konibo (Pilar Valenzuela, p.c.)
‘Rono went to Caco.’
‘Rono did not go to Caco.'
In Taba the negative particle te is simply added to the positive clause and there are no further changes in the structure. In Shipibo-Konibo the negative suffix -yama appears on the verb with no further changes. These symmetric negative constructions are used in all standard negation contexts, and all affirmatives can be negated with these constructions in these languages. No constructional or paradigmatic asymmetry is thus found in Taba or Shipibo-Konibo, and they belong to Type Sym.
In Type Asy standard negation is always asymmetric. In Finnish (see (4)) negation is marked by a construction where the negative verb e- carries the marking of person-number, and the lexical verb loses its finiteness, appearing in the connegative form in the present (see (4a-b)) and in the past participle in the past (see (4c-d)). There are thus structural differences between the negative and the affirmative in addition to the presence of the negative marker; the negative construction is therefore analysed as asymmetric.
(4) Finnish (personal knowledge)
‘I am coming.’
‘I am not coming.’
‘I did not come.’
The paradigm is symmetric in Finnish since all affirmatives can be negated with the asymmetric negative verb construction. But as all negatives use the asymmetric construction, negation is always asymmetric, and Finnish is a language of Type Asy.
The Burmese standard negation construction involves a combined prefix and suffix ma- -bû (see (5)). This negative suffix -bû replaces the tense-aspect-mood markers used in the affirmative.
‘((S)he) goes, went.’
‘((S)he) will go.’
‘((S)he) has gone.’
‘((s)he) does/did/will not go, has not gone’
As the negative markers are not simply added to the corresponding affirmative, the construction is asymmetric. In Burmese the paradigm is also asymmetric: the affirmative paradigm makes a distinction between actual, potential (future) and perfect, but the distinction is lost in the negative since the negative construction replaces the suffixes marking these categories. No symmetric constructions being found in Burmese, it is a language of Type Asy.
For a language to qualify as belonging to Type Asy, it is sufficient that every negative construction in the language is asymmetric. The paradigm can of course be symmetric or more or less asymmetric, but as the negative construction is always asymmetric, there are no instances of symmetric standard negation.
In languages of Type SymAsy both symmetric and asymmetric standard negation are found. Standard negation is sometimes symmetric, i.e. at least some negative constructions are symmetric, but there is also some asymmetry to be found, either in some negative constructions or somewhere in the paradigm. This is an intersection of the two types of standard negation within the same language. In Lezgian (Nakh-Daghestanian; eastern Caucasus; see (6)), the negative suffix -č appears in standard negation.
‘The villagers take advice from him.’
‘The villagers do not take advice from him.’
‘was not going’
The negative construction is symmetric in nonpast imperfectives (6a, b). In past imperfective verb forms (6c, d), past tense is marked differently in the affirmative and the negative, and the negative construction is thus asymmetric. Lezgian belongs to Type SymAsy.
In Maung (Yiwaidjan; Northern Territory, Australia) the negative clause is obligatorily marked for the irrealis category (see (7d)), whereas the affirmative can make a distinction between realis and irrealis (see (7a-c)).
‘I put [it].’
‘I was putting [it].’
‘If I put [it].’
‘I did not put [it].’
The negative construction simply adds the negative marker marig to the non-negative irrealis, and the construction is thus symmetric. There is paradigmatic asymmetry since the realis-irrealis distinction is lost in the negative. There are symmetric negatives in Maung, but some asymmetry is also found, and the language thus belongs to Type SymAsy. Note that the punctual-continuous distinction made in the affirmative (7a, b) is also lost in the negative, but this distinction is lost in the irrealis in general, and the loss is thus not directly due to negation.
As seen above, a language can have different standard negation constructions used in different contexts (e.g. with different tense-aspect categories) or in free variation. Furthermore, the negative paradigm can be symmetric or asymmetric, and this is partly independent of the type of negative construction(s) the language has. Thus, different combinations of symmetric and asymmetric negative constructions and paradigms are found in type SymAsy.
It is worth noting that not all asymmetry found in declarative verbal main clauses is taken into account in this classification. First, asymmetry that is not grammaticalized but is only a tendency does not figure on the maps; e.g. in Russian there is a preference for imperfective aspect under negation, but both perfective and imperfective aspect can appear under negation (see for example Schmid 1980) — Russian is classified as having only symmetric negation. Secondly, asymmetries that are outside the limits of the standard negation construction are disregarded. Since standard negation expresses the negation of verbal clauses, the negative construction is delimited in such a way that categories belonging to the verb or more globally to the clause are relevant, but categories belonging to individual non-verb constituents (like case marking on noun phrases) usually are not. For example, the changes in case in Finnish (accusative changing into partitive under negation) and the changes in the article in French (indefinite article changing into partitive de under negation) are not taken into account. They belong primarily to the marking of noun phrases and are not considered part of the negative construction here. Similarly, negative polarity effects on indefinites are not taken into account. Thirdly, changes that are (morpho)phonological and do not involve any substantial differences in the meaning-bearing elements are not relevant for the classification, either.
The map says nothing about the proportion of the different types in individual languages. In some cases the proportion can be rather unequal; standard negation in a language can be predominantly of a given type, but there can be a small domain where the language shows a different type. Nevertheless, the symbol given to the language represents both types equally strongly. Haitian Creole, for example, has a symmetric negative construction and most verbal categories can appear under negation (D'Ans 1968). But the particle ava marking future cannot occur in negated clauses. This asymmetry concerns only a minority of the negative clauses in the language, but still Haitian Creole is treated as belonging to Type SymAsy.
Symmetric negation (Types Sym and SymAsy) is frequent in all parts of the world. There are two areas where only symmetric negation (Type Sym) is found: much of Continental Europe and a large part of Southeast Asia. Languages with only symmetric negation (Type Sym) are the least common in Africa, but on the whole symmetric negation is not uncommon in Africa, since a large number of African languages have both symmetric and asymmetic negation (Type SymAsy). Asymmetric negation (Types SymAsy and Asy) is also frequent in all parts of the world. It is absent in the two areas just mentioned and most common in Africa. Languages with only asymmetric negation (Type Asy) are not dominant in any area. Note that some connections between Maps 113A and 114A are discussed in Chapter 114.
Why have I made this division into symmetric and asymmetric negation? The main reason is that it allows us to find different functional motivations for the different types of negative structures. Symmetric negatives copy the formal linguistic structure of the affirmative, and are thus language-internally analogous to the affirmative structure; they are motivated by the pressure for cohesion in the system. As to asymmetric negation, there is functional (semantic and pragmatic, i.e. in a strict sense language-external) asymmetry between affirmation and negation: they differ as to their stativity/dynamicity, reality, and discourse context, for example (see Givón 1978, 2001: 369ff.; Miestamo 2003). Asymmetric negative structures copy (or grammaticalize) aspects of this functional asymmetry, and are thus language-externally analogous to the just-mentioned functional asymmetry. The different subtypes of asymmetric negation discussed in Chapter 114 have grammaticalized different aspects of the functional asymmetry. For more detailed discussion, see Miestamo (2000, 2003: 169-212).