Parameter Resetting In Metrical Phonology: The Case of Setswana and English

Abstract

Hierdie artikel behandel die verwerwing van klemtoonsisteme in tweede tale. Om bespreek data van Swart Suid-Apikaanse Engels, wat verky is in 'n eksperiment waaraan 50 moedertaalsprekers van Setswana deelgeneem het. Die Principle and Parameter-teorie word as raamwerk gebruik, meer in die besonder die Metriese Fonologie, wat die terrein van die klemsisteme van natuurlike menslike tale bestudeer.

Eerstetaalverwenving is binne hierdie raamwerk die verwenving van volledige versameling parameterinstellings (‘settings’?. Normaalweg is sodanige proses volkome geslaagd. In beginsel word tweedetaalverwerwing gesien as die versameling van sulke instellings naas die' van die eerste taal. Dit geskied meestal op 'n meer gevorderde leefid. Hierdie proses is in die literatuur bekend as parameter-herinstelling (“resetting”). Die doe1 van hierdie artikel is om die voorspellings van hierdie raamwerk aan die hand van empiriese data van werklike tweedetaalverwerwing te toets. Ons toon aan hoe vroeere resultate van Pater se studie van Quebec-Engels in terme van Dresher en Kaye se parametiese leerteorie van metriese sisteme geherinterpreteer kan word as baie klein en beperkte stappies op 'n aanleerderspad. Die belang van die Setswana-geval li opgesluit in die feit dat dit 'n (non-klemtoon-) toontaat is, sodat daar vir hierdie taal as moedertaal geen parameterinstellings gemaak is nie. Tog toon ons aan dat minstens sommige van die deelnemers aan die eksperiment een, of moontlik selfs enkele, tree geneem het op hierdie leerderspad. Dit word ten slotte duidelik dat hierdie group tweedetaalsprekers van Engels se resultate soortgelyk is aan die' van die Franssprekende gebruikers van Engels.

In this paper we are concerned with the acquisition of stress systems in second languages. We discuss empirical material involving Black South African English, elicited from native spakers of Setswana. Our theoretical framework is that of Principles and Parameters theory, and within that theory the area of Metrical Phonology, i. e. the study area of the stress systems of human languages.

Within this framework, first language acquisition is the acquisition of a complete collection of parameter settings. In the normal case this process will always fully successful. In principle, second language acquisition may be seen as obtaining, usually at a more advanced age, a new collection of such settings, next to those of the first language. In the literature this process is known as parameter re-setting. One way of testing these ideas is to consider their prediction an empirical issue, and to investigate actual second language acquisition data. This is the goal of the present paper. We show how earlier results by Pater on native speakers of Quebec French using stress in their second language English, may be reinterpreted in the Dresher and Kaye parametric learning theory of metrical systems as very small and limited steps on a learning path. Next, the interest of the Setswana case resides in the fact that it is a (non-stress) tone language, so no parameter settings will have been made for this language as a first language. Even so we show, discussing the outcome of an experiment with 50 Setswana speakers for whom English is a second language, that again at least some members of this group seem to have taken one or perhaps even a few steps on the learning path, and moreover, that the second language learning picture provided by this group is similar to that encountered earlier in the French case.