Historical Change & Repair of Aktionsart: the development of compound verbs/absolutives in Indo-Aryan, esp. Hindi & Nepali

 

This paper investigates the historical development of compound verb [CV] constructions in modern Indo-Aryan [I-A] as motivated by the collapse of the Indo-European [I-E] system of preverbal marking of Aktionsart properties. We proceed via comparative analysis of relevant forms drawn from various languages and stages of I-A.

I-A CV constructions, in which a main verb in a stem form is followed by an inflected form of a ‘vector’ verb, are connected with, and ultimately derived from, Sanskrit absolutives [ab.] in –ya. Modern I-A languages show increased use of absolutive/conjunctive participles, and additionally use the same participles in CV constructions, which results in some sentences which are thus indeterminate between CV complex predication and conjunctive clause-chaining [see (1)]. In earlier I-A we also find ‘absolutive’ participial constructions which are most felicitously read as complex predicates [see (2)]. 'Absolutive’ complex predication exists in even Old I-A [see (3)], with a more limited range of Aktionsart semantics (restricted to durative and transitional).

Part of this development might be explained as syncretism with other South Asian languages (esp. Dravidian) which use ‘serial verb’ constructions. However, in both I-A and English/Norse, the original I-E Aktionsart system of verbal prefixes [see (4)], was obscured and eventually lost due to phonological changes. In Norse and English, this system is ‘repaired’ by the extension of use of the pre-existing verbal-particles (e.g., eat up, wind down). In I-A, it is ‘repaired’ by the extension of the pre-existing system of ‘absolutive participles’ as CVs: the increased phonological ‘decay’ is complemented by the increased use (and increased range of Aktionsart meanings) of ‘absolutive’ verbal complexes. Hindi and Nepali have morphologically ‘renewed’ a (partial) distinction between clause-chaining and Aktionsart-marking.

In the Modern period, I-A languages in general show remarkable regularities in the set of ‘vector’ verbs, yet the extent and usage of vectors is not identical: Nepali, for instance, does not permit the use of COME or TAKE as vectors, which are basic and frequent vectors in other South Asian languages [see (5)]--displaying interesting differences in CV-development.

Thus we observe some cases of linguistic change which work to preserve certain semantic features (e.g. Aktionsart): when the original system (e.g. verbal prefixes) deteriorates, a pre-existing system (e.g. I-A ‘absolutives’, Eng./Norse postverbal particles), which expresses such features in a more limited manner, is expanded. In I-A, this results in some ambiguity between clause-chaining and complex predication—which itself is partially 'repaired' in Hindi and Nepali by additional suffixes marking the ‘conjunctive’ against CV ‘absolutives’.

 

(1) Hindi ambiguous ‘absolutives’

(a) vah aa (kar) gayaa

he come (CP) go-past-3p-sg.-masc (conjunctive reading)

‘Having come, he went’.

(b) vah aa gayaa

he come go-past-3p-sg.-masc (CV-reading)

‘He came’ (CV with jaa- indicating transition/change of state Aktionsart)

 

(2) daruni aahari-tvaa aggim ka-tvaa dassati [Buddhist Pali, Sri Lanka]

sticks bring -CP fire-acc make-CP give-fut-3sg

'Bringing sticks, he’ll make a fire (for him)'. [Trenckner 1879:77, cited in Hook 1993:97]

 

(3) ime ta indra te vayam puru stuta [Vedic Skt]

pron-dual dem.pro.3sg. indra-voc pro-2sg-gen pro-1pl full-praise-PP.voc.sg.

ye tvaarabhya caraamasi .... [Vedic Skt.]

rel.pro pro-2sg.acc-grasp.CP go-pres-1p-pl

(a) (?*) 'We are yours, O much-praised Indra, who wander about having taken hold of you...'

(b) 'We are yours, O much-praised Indra, who hold to you (constantly)...'

[ Rgveda I.57.4; cf. Butt & Lahiri 2002:22; Tikkanen 1987:175; Delbrück 1888:406]

 

(4) Examples of preverbal Aktionsart particles

(a) Sanskrit ava- 'down, off'

k rt 'cut' ----- ava-k rt 'cut off'

jñaa 'know' ----- ava-jñaa 'despise'

gam 'go' ----- ava-gam 'understand'

(b) Old English

brecan 'break' ----- a-brecan 'destroy'

slean 'strike' ----- of-slean 'kill'

winnan 'strive, compete’ ----- ge-winnan 'win, succeed'

 

(5) Modern Indo-Aryan vectors [selected] (**=exists as lexical, but not as ‘vector’ verb)

Meaning of lexical verbHindiBengaliMarathiNepaliMeaning of ‘vector’‘go’jaa-jaa-dzaa-jaa-transition/completion‘give’de-de-de-di-benefaction, completion/transition‘come’aa-ash-ye-(**aau-)self-directed action, continuation‘take’le-ne-ghe-(**li-)reflexion/self-direction action

Selected References:

Aryal, Bala Ram (2001). nepālī bhā sh ā 1. Kathmandu : ShiSam.

Butt, Miriam & Aditi Lahiri (2002). 'Historical Stability vs. Historical Change'. Ms., Uni. of Konstanz. [online @ http://ling.uni-konstanz.de/pages/home/butt/stability.pdf ]

Chatterjee, Suniti Kumar (1926). The Origin and Development of the Bengali Language. Calcutta Uni. [reprinted, New Delhi: Rupa, 1985].

Hendriksen, Hans (1944). Syntax of the Infinite Verb Forms of Pali. Copenhagen: Einar Munksgaard.

Hook, Peter Edwin (1974). The Compound Verb in Hindi. Ann Arbor: Uni. of Michigan.

------ (1993). 'Aspectogenesis and the compound verb in Indo-Aryan'. in Verma 1993:97-114.

Kachru, Yamuna (1993). 'Verb serialization in syntax, typology & historical change'. in Verma 1993:115-134.

Masica, Colin P. (1991). The Indo-Aryan Languages. Cambridge UP.

Nespital, Helmut (1997). hindī kriyā-kosh. Allahabad : Lokbharati.

Pandharipande, Rajeshwari (1993). ‘Serial verb construction in Marathi’. In Verma 1993:177-195.

Samuels, M.L. (1972). Linguistic evolution, with special reference to English. Cambridge UP.

Srivastava, Dayanand (1962). Nepali language, its history and development. Calcutta Uni.

Tikkanen, Bertil (1987). The Sanskrit Gerund: a synchronic, diachronic and typological analysis. Helsinki: Finnish Oriental Society.

Verma, Manindra K., ed. (1993). Complex Predicates in South Asian Languages. New Delhi: Manohar.