[Lingtyp] Structural congruence as a dimension of language complexity/simplicity

[Alan Jones]

Hi Alan,

I don’t know how useful you will find an excursus into generative linguistics<https://en.wikipedia.org/wiki/Generative_linguistics> but generativists have something called the Principles and parameters framework which corresponds in most respects to the head-directionality principle described by Nichols and which (following generative logic) has profound implications for the simplicity aka learnability of a language.

Within the PP framework, the syntax<https://en.wikipedia.org/wiki/Syntax> of a language<https://en.wikipedia.org/wiki/Natural_language> is described in terms of general principles (or abstract rules) and specific parameters (i.e. markers or switches) that for particular languages can be either turned on or turned off. The position of heads<https://en.wikipedia.org/wiki/Head_(linguistics)> in phrases is determined by such a parameter. The notion of such a ‘switch’ makes sense to me as in learning Turkish or Japanese (or Mekeo), I found I only had to cope with the novelty of head-final word order once; after that the general ordering principle becomes quite natural and expected.

The Principles and Parameters framework, qua research program, has set out to explain the gap between linguistic knowledge and linguistic competency. I.e., given the finite, corrupt or incomplete input<https://en.wikipedia.org/wiki/Poverty_of_the_stimulus> children have access to, how do they so rapidly arrive at a (fairly) accurate and complete grammatical competence. Principles and parameters are here held to be part of universal grammar<https://en.wikipedia.org/wiki/Universal_grammar> (UG) and are thus ‘innate’ – at least as potential. Hence they do not need to be learned by exposure to any given natural language. Exposure triggers the appropriate parameters, causing the brain to adopt certain ‘settings’, e.g., it might be set to expect and produce either head-final or dependent final structures across the board. I don’t know how they account for the kinds of exceptions to the general rule of directionality that most grammars of natural languages exhibit. Perhaps someone else can help.

Alan




Alan Jones, PhD,

Adjunct Professor,

Macquarie University,

Sydney


________________________________
From: Lingtyp <lingtyp-bounces at listserv.linguistlist.org> on behalf of Alan Rumsey <Alan.Rumsey at anu.edu.au>
Sent: Sunday, 17 January 2016 2:03 AM
To: lingtyp at listserv.linguistlist.org
Subject: [Lingtyp] Structural congruence as a dimension of language complexity/simplicity


Dear All,



Francesca Merlan and I are currently working on a paper on structural congruence as a dimension of language complexity/simplicity. It is based on results from our longitudinal study of children’s acquisition of two verb constructions in the Papuan language Ku Waru, namely, serial verb constructions (SVC) and adjunct+verb constructions (AVC). We show that children learn the AVC before the SVC, and argue that this is determined in part by the greater congruence between AVC and other basic aspects of Ku Waru syntax including its strictly verb-final word order. This has got us thinking about the general issue of structural congruence as a dimension of language complexity. For example, long ago Greenberg demonstrated that there is a very strong tendency in languages with VSO word order for the adjective in NPs to follow the noun. This is presumably because there is a kind of congruence between the noun as the head of the NP and the verb as the head of the clause – an insight which led to Nichols’ later very useful typological distinction between head-marking and dependent-marking grammar. Not all languages conform to Greenburg’s generalization in this regard. But we would argue that those that do are in that respect simpler than those that don’t, because both the VSO template and the noun-adjective one can be specified in terms of a more general relationship between heads and dependents. So far in our search through the recent literature on linguistic complexity we haven’t come across any discussion of this kind of congruence as a dimension of language complexity/simplicity. Can any of you point us to any? Or to other relevant data for a comparative consideration of this issue?



Alan Rumsey,

Australian National University


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20160117/de3f16bf/attachment.htm>