LINGUIST List 4.256

Thu 08 Apr 1993

Sum: Grammar Shifts

Editor for this issue: <>


  1. AN ELECTRIC ELK, SUM: Grammar Shifts

Message 1: SUM: Grammar Shifts

Date: Tue, 6 Apr 1993 20:40 CSTSUM: Grammar Shifts
Subject: SUM: Grammar Shifts

 I have been asked to summarize the responses to my question on grammar shifts
I posted a while ago. I asked if anyone knew of a language that had changed
from synthetic grammar structure to an analytic grammar structure or vice
versa. I was, however, most interested in any evidence of a language that had
*reversed* this trend. That is, gone from synthetic to analytic back towords
synthetic (or some variation therof). Being, for the most part, unqualified to
accurately summarize the responses I got, I am resorting to including the
majority of the replys. If this causes inconveniences, I apologize in advance.
Without further ado here they are:

 From: David Stampe <stampeuhunix.BITNET>
 Subject: grammar shifts

The Munda languages of India, one branch of the Austroasiatic family
(the other being the Mon-Khmer languages of South-East Asia), have
developed a dependent - modifier word order (Object - Verb, Adjective
- Noun, Noun - Postposition) with morphological marking of grammatical
relations (Subject and Object marked on the verb, some case marking),
and relatively free word order. Some are even polysynthetic, i.e.
whole sentences are expressed as one verb. The original grammatical
type of Austroasiatic, which is preserved in the Mon-Khmer languages,
has the opposite modifier - dependent order (Verb - Object, Noun -
Adjective, Preposition - Noun) without any inflectional morphology at
all, and therefore quite rigid word order. The two branches of
Austroasiatic thus were distinguished by a change opposite that in
Indo-European, and incidentally their now opposite structures are far
more different from each other than the opposite structures of
proto-Indo-European and modern Indo-European languages. A brief
description of the evolution of Munda grammar is given in a short
paper by Patricia Donegan and myself in John Richardson, Mitchell
Marks, & Amy Chukerman, eds., Papers from the Parasession on the
Interfaces of Phonology, Morphology, and Syntax, Chicago Linguistic
Society, 1983, pp. 337-351. The paper also describes the opposite
prosodic and phonological drifts of Munda and Mon-Khmer, and even
alludes to opposite tendencies in their verse and music structures.

 From: Steve Matthews <MATTHEWSHKUCC.BITNET>
 Subject: Grammar shifts

 I was interested in your question about direction of change.
I gave a paper at the 1988 Georgetown Round Table on Variation and Change
arguing that spoken French is undergoing a major typological shift as a
result of grammaticalization of the clitic pronouns as agreement markers.
This makes it a head-marking language in the sense of Nichols ('Language',
1986) and results in much greater freedom of word order than exists in
standard written French. In particular, it is absurd to continue to analyse
'left-dislocated' and 'right-dislocated' constructions as such, when they are
used ubiquitously as topicalisation and verb-initial (esp. interrogative)
constructions respectively.
 At that conference my paper was next to one by Bernard Bihackjian
(of Nijmegen; author of Evolution in Language) who argued forcfully for
a one-way model of typological change (evolution, in his terms) from
SOV with word order freedom to SVO without it. this view leads to a
number of paradoxes, as does the situation described in your note.
Part of the answer may involve the role of standardisation and written
forms of language, in which word order changes and variation are
sub stantially curbed. In French, incidentally, the 'spoken French' syntax
does creep into written registers.

 From: Randy LaPolla <>
 Subject: Grammar Shifts

It may not be what you are looking for, but in Sino-Tibetan the word
order is to a large extent pragmatically based, and the languages
have developed more morphology as time has gone on. Some of the
languages changed from verb final to verb medial (the Sinic lgs and
a couple of Tibeto-Burman lgs), but most often it is the morphology
or pragmatics that determines argument relations, not word order.

Those are the responses, for the most part, as received. Thank you to all who
responded, Randy LaPolla, Steve Matthews, and David Stempe.

 Dan "Toby" Williamson

Internet: Bitnet: acc_dtwvalpo
 Realnet: Valpariso, Indiana and Chevy Chase, Maryland
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue