Linguistic Structure and Change: An Explanation from Language Processing

Most of the studies that will be reviewed below share a conviction that is more often tacitly assumed than overtly stated. This view is made explicit by Menn and Obler ( 1982: 3), who reason that the 'explanation of linguistic phenomena must ultimately lie outside linguistics' (emphasis mine). Similar opinions have been voiced by Bybee and Moder ( 1983), Hall ( 1992), Comrie ( 1993), and others. The fundamental insight implicit in this view is that nothing is self-explanatory. We cannot understand plants unless we acknowledge their ecological functions such as providing nutrition to animals and giving rise to photosynthesis, that is, unless we establish a link between trees and the sun for instance. What are the links that should be set up in the case of language? Ohala ( 1983) sees three and only three linkages which we may conveniently abbreviate as 'the three m's'--mind, matter, and manners. The first refers to the abstract psychological dimension, the second to the concrete dimension including anatomy and the neurosciences, and the third to the social and cultural dimension. This tripartite distinction is a very useful point of departure, even though it might be rather too coarse-grained. For example, neurophysiological processes need not impose the same constraints on language as anatomical traits. It is expedient therefore to look at a maximum of approaches individually so as to be in a better position to assess the relative contribution of each. In view of a certain overlap that exists between these approaches, it will not be possible to keep all of their effects neatly apart.

The overall aim of this chapter is to evaluate the explanatory potential of several approaches by selecting representative cases from the relevant literature and gauging the depth of insight that they provide as well as their scope and limitations. This review is not primarily intended as a state-of-the-art report on the role of explanation in linguistic research. Its major function is to provide a basis from which it is possible to pinpoint the weaknesses of previous approaches and develop a novel one. The areas that will be covered are (in the order of presentation): the neurological, the phonetic, the psychological, the semiotic, the functional-communicative, the pragmatic, the socio-cultural, the historical and the system-internal approaches. The possible interactions among these approaches, whether they complement or conflict with one another, will be largely ignored. The synergetic account ( Köhler and Altmann 1986) will not be given a separate treatment.

-18-