Decategorialization
Once a linguistic expression has been desemanticized, for example from a lexical to a grammatical meaning, it tends to lose morphological and syntactic properties characterizing its earlier use but beingnolonger relevant to its new use. In this way, a number of English verbs have been desemanticized in their gerundival form (-ing) and assumed prepositional functions, for example barring, concerning, considering, etc. Consequently, they lost most of their verbal properties, such as to be inflected for tense and aspect, to take auxiliaries, etc. (see Kortmann and Ko ¨nig 1992). Decategorialization entails in particular the changes listed in (15):
(15) Salient properties of decategorialization
a. Loss of ability to be inflected.
b. Loss of ability to take on derivational morphology.
c. Loss of ability to take modifiers.
d. Loss of independence as an autonomous form, increasing dependence on some other form.
e. Loss of syntactic freedom, e.g. of the ability to be moved around in the sentence in ways that are characteristic of the non-grammaticalized source item.
f. Loss of ability to be referred to anaphorically.
g. Loss of members belonging to the same grammatical paradigm.
In accordance with this list, nouns undergoing decategorialization tend to lose morphological distinctions of number, gender, case, etc.; the ability to combine with adjectives, determiners, etc.; to be headed by adpositions; they lose the syntactic freedom of lexical nouns; and the ability to act as referential units of discourse.1 In a similar fashion, when a demonstrative develops into a clause subordinator, as has happened in many languages of the world, it loses salient categorical properties. For example, the English demonstrative that is sensitive to number, having those as its plural form, it is fairly autonomous in that it can be used both as an attribute of a noun or as a pronoun, and it belongs to a morphological paradigm which also includes this. In its grammaticalized form as a relative clause marker, however, it is decategorialized, having lost this distinction (The books that/*those I know), it has lost the distinction between use as an attributive and a pronoun, and it has lost this as a co-member of the same paradigm.
Verbs undergoing decategorialization tend to lose their ability to be inflected for tense, aspect, negation, etc., to be morphologically derived, to be modified by adverbs, to take auxiliaries, to be moved around in the sentence like lexical verbs, to conjoin with other verbs, to function as predicates, and to be referred to (e.g. by pro-verbs). Finally, they lose most members of the grammatical paradigm to which they belong by changing from open-class items to closed-class items.
In more general terms, decategorialization tends to be accompanied by a gradual loss of morphological and syntactic independence of the linguistic item undergoing grammaticalization, typically proceeding along the scale described in (16) (see also “A scenario of evolution”).
(16) Free form > clitic > affix
But decategorialization is not restricted to open-class items such as nouns and verbs, it affects in the same way closed-class items, which are also likely to lose their categorial properties; as we just observed, demonstratives in many languages show distinctions in number, gender, and/or case, or between pronominal and attributive functions; such distinctions tend to disappear when decategorialization takes place.
The generalizations just presented were hedged in the form of ‘‘tend to’’ predications, for the following reason: The changes listed are not necessarily all present in a given case of decategorialization. For example, on their way from verb to auxiliary, English items such as be going to or keep, that we discussed in “Assumptions”, have not lost their ability to combine with other markers of tense, aspect, and modality (e.g. Mary would have kept coming), and in the development from demonstrative to definite article der/die/das in German, distinctions in number, gender, and case were not lost either. There can be a number of different causes for decategorialization not taking place. One concerns language-internal factors: There may be specific structural constraints that prevent the loss of some categorical property. For example, verbal auxiliaries in many languages are the only grammatical category in the clause where distinctions of personal deixis, tense, aspect, or negation are encoded; thus, giving up these encodings might have dramatic consequences for the information structure of the clause. Another cause concerns the age of the grammaticalization process involved: Decategorialization does not happen overnight, that is, it takes some time to come in, and the younger a process is, the lower the amount of loss in categorical properties will be.
1 For an example involving English while, see Hopper and Traugott (2003: 107).
