by John J. Lowe (University of Oxford)

Transitivity is typically thought of as a property of verbs, and perhaps of adpositions, but it is not a typical property of nouns or adjectives. In the influential cross-classification of syntactic categories developed by Chomsky (e.g. 1981: 48), nouns and adjectives are actually defined in opposition to adpositions and verbs by their inability to govern objects, that is by their inability to be transitive. A few authors have discussed exceptions to this generalization, but they tend to be rare and non-productive; for example in English there may be only a single transitive adjective, near, which is a historically explicable exception to an otherwise consistent synchronic rule that nouns and adjectives cannot govern ‘bare’ noun phrase complements (Maling, 1983). As a second example, in early Latin there are a few nouns and adjectives which may govern accusative case objects, but the process is not productive and is entirely eliminated by Classical Latin.

gnaruris                          vos                    volo                esse        hanc                 rem
acquainted.ACC.PL     you.ACC.PL    wish.1PL       be.INF   this.ACC          matter.ACC
‘I wish you to be acquainted with this matter.’ (Latin: Plautus Most. 100)

In the early Indo-Aryan languages, however, there is a relative wealth of transitive noun and adjective categories. In my forthcoming monograph Transitive Nouns and Adjectives: evidence from early Indo-Aryan (OUP, July 2017), I investigate the evidence from four periods of early Indo-Aryan, discussing the synchronic and diachronic explanation for this unusual phenomenon.

The majority of transitive noun/adjective categories in early Indo-Aryan fall under the traditional heading of ‘agent noun’ (including agentive adjectives, used in the same way); these are the categories whose transitivity is most clear, and most common. For example, in the sentence below the ‘agent adjective’ kāmin- ‘desirous, desiring’ governs an accusative object ‘drink’.

kāmī                                   hi       vīraḥ                            sadam    asya        pītim
desirous.NOM.SG.M   for      hero.NOM.SG.M     always    it.GEN    drink.ACC
‘For the hero (is) always desirous (of) a drink of it.’ (Sanskrit: RV 2.14.1c)

Superficially, kāmī here looks similar to a participle, i.e. to a word category which, as a non-finite verbal category, could unproblematically govern an object. However, I show that the majority of transitive nouns and adjectives attested in early Indo-Aryan cannot be analysed as non-finite verb forms, but must be acknowledged as part of a distinct constructional type in early Indo-Aryan.

Other transitive nouns fall under the traditional heading of ‘action nouns’; I show that for the most part action nouns are transitive only when used as infinitives, and hence their transitivity can be explained as the unexceptional transitivity of non-finite verb forms. There are also nouns and adjectives whose transitivity is adpositional, rather than verbal.

Crucially, I show that there is a statistical correlation between transitivity of nouns and adjectives and the syntactic context of predication: nouns and adjectives which are used as the primary predicate in a (perhaps null) copular construction are statistically more likely to be transitive than those which are used in other ways. This correlation is unique to transitive nouns and adjectives and securely distinguishes this formation from transitivity with non-finite verb categories.

The book provides a detailed introduction to transitivity (verbal and adpositional), to the categories of agent and action noun, and to early Indo-Aryan. The four periods of early Indo-Aryan selected for study are: Rigvedic Sanskrit, the earliest Indo-Aryan; Vedic Prose, a slightly later form of Sanskrit; Epic Sanskrit, a form of Sanskrit close to the standardized ‘Classical’ Sanskrit; and Pali, the early Middle Indo-Aryan language of the Buddhist scriptures. I show that while each linguistic stage is different, there are shared features of transitive nouns and adjectives which apply throughout the history of early Indo-Aryan.

The data is set in the wider historical context, from Proto-Indo-European to Modern Indo-Aryan, and a formal linguistic analysis of transitive nouns and adjectives is provided in the framework of Lexical-Functional Grammar.

References:

Chomsky, Noam (1981), Lectures on Government and Binding: The Pisa Lectures, Dordrecht: Foris.

Lowe, John J. (2017), Transitive Nouns and Adjectives: Evidence from Early Indo-Aryan, volume 25 in the series Oxford Studies in Diachronic and Historical Linguistics. Oxford: Oxford University Press. c. 400 pp. ISBN: 978-0-19-879357-1.

Maling, Joan (1983), ‘Transitive adjectives: a case of categorial reanalysis’, in Frank Heny & Barry Richard (eds.), Linguistic Categories: Auxiliaries and Related Puzzles, volume 1. Dordrecht: Reidel. 253–289.