|
wárí-ni=rupa
prepare-aDjvz-aDvz
lágíaa
he.told
Verb > Adverb
‘He told me how food is prepared.’
ní
I
mógopara=rupa
that.over.there=aDvz
méáwa
I.got.it
Demonstrative > Adverb
‘I got (the one) similar to that (thing) over there.’
[amá lá-wa]=rupa
[ mother say-I.did]=advz
toa
say.I.will
Clause > Adverb
‘I will tell it like I told mother.’
ába=rupa
before=aDvz
pá-lua
go-I.will
Adverb > Adverb
‘I will go like I did before.’
Apart from the example from Kewa in (9.4e), which is not attested with any property- denoting Adverb as far as I can tell from the language description, derivation from adverb to another part of speech has not been observed in the languages of the sample. Although it may of course exist, I am not aware of such a case. One reason why adverbs are unlikely to serve as a basis for derivation is the complexity of conceiving of concepts denoted by adverbs. If there are other derivational patterns at hand, it would be an unnecessarily complicated process to start from adverbs and thereby add another conceptual step.
Adverb inflection and derivation have so far been discussed as separate matters. How- ever, the issue of whether adverbs formed from adjectives should be classified as inflection or derivation has been a recurrent grounds for debate (e.g. Geuder 2000). The English -ly ending is a classical example in such discussions (Zwicky 1995). It is not surprising that this ending is often used as a textbook example for testing inflectional and derivational criteria (cf. e.g. Haspelmath 2002: 110, where this is literally the case). In Croft’s ap- proach to parts of speech, derivational morphemes are defined as structural coding, which indicates a change of function (2001: 90). Recalling the nominalizing ending -ness added to red to form redness, this indicates that the function is changed from property modifica- tion (or adjective) to property reference (or noun). If the same criterion is applied to the presumed derivation of Adverbs in English by the use of -ly, it is evident that the function is not changed in the same way. First of all, happy and its derived adverb happily are both used in the function of modification. But within modification, an important change does take place in terms of what is being modified, or what the type of modification in question is an accessory function to: reference or predication. In this sense, adjectives and adverbs are closer as categories due to their shared general function. If derivation is function-changing in its essence, the fact that adjectives and adverbs actually occur
within the same function, albeit in different domains of it, must influence how derivation from one to the other is viewed. In other words, the fact that the potential derivation from adjective to adverb appears as less derivation-like than that between other cate- gories is actually not that surprising, since it is also less of a function-changing process. A change does take place, which might be enough to accept -ly and other similar markers as derivational morphemes. However, such a change does not appear to be equal to other types of derivation, where the entire function is changed, e.g. from modification to refer- ence (as in the case of red-ness). Against this background, it is not at all surprising that one larger category may cover both sub-domains of modification in the form of general modifiers. General modifiers have also been attested as constituting a stable category for many languages in the sample, either covering the functions of adjectives and adverbs entirely (e.g. Dutch), or alongside adjectives (e.g. Kilivila), adverbs (e.g. Hdi), or both (e.g. Sango).
In conclusion, there are various issues concerning inflection and derivation in connection to adverbs. Adverbs across languages do not usually inflect, although this is attested in a number of languages. Agreement is found in Yankunytjatjara in the sample of the present study, and in a number of languages such as Archi, but also Tsakhur and Lak (Nakh-Daghestanian), Kala Lagaw Ya (Pama-Nyungan), Bhitrauti, and Gujarati (Indo- European), all discussed by Corbett (2006: 44-46). There are also cases of comparative and superlative inflection being shared by or related for adjectives and some adverbs, as illustrated by Swedish and Lithuanian. Adverbs rarely occur as the base for derivation, with Kewa as a marginal exception in the sample. One reason for this could be the conceptual complexity that adverbs imply. Finally, the debated status of the English -ly ending as derivational or inflectional serves as a case in point for illustrating the nature of the modification function. When an adverb is formed from an adjective, its function changes in terms of what is being modified, but the resulting item remains a modifier.
Do'stlaringiz bilan baham: |
