INTRODUCTION

1.Paradigms and paranickels

I would like to consider the contrast between the Autonomist (aka Autonomous) schools of Linguistics, on the one hand, and the Cognitive Linguistics school, on the other, in the context of Kuhn (1970). The latter work involves three central notions: (a) scientificness; (b) paradigms; and (c) revolutions. These three are not inextricably intertwined, however, and I propose that some of these can be used to some extent to describe the contrast between Autonomist and Cognitive Linguistics.

The term Autonomist Linguistics is used in the sense of Autonomous Linguistics, but without the latter's positive overtones, hopefully. Thus it is meant to include Generative Grammar, G-B Theory, Generalised Phrase-Structure Grammar, and the like -- in other words, any theory which aims to study language (at least in the first instance) without taking into account so-called "non-linguistic" aspects of mind.

Like Percival (1976), I take it that we can discuss paradigms and revolutions without taking any particular stance as to the nature of scientificness -- it does not seem to me that there is any clear notion of "scientific" such that it really matters whether Linguistics (or any other specific discipline) can rightly be characterised as such or not.

By contrast, it is widely accepted (including by Kuhn) that "revolutions" can occur in any sphere of human intellectual activity. Thus, if Cognitive Linguistics were to achieve the degree of influence that Autonomist Linguistics (chiefly: Generative Grammar) enjoyed, say, in 1970, then few would deny that this amounted to a revolution in Linguistics -- particularly if it happened relatively quickly.

The term "paradigm", on the other hand, is more controversial, as Kuhn defines it in a rather narrow way: it is said to have four components -- symbolic generalisations, models, values, and exemplars. It also results from an outstanding scientific achievement on the part of a single innovator, and it commands uniform assent among all the members of the discipline. Kuhn defines "scientific" in such a way that all and only disciplines whose activities are governed by a single such paradigm are truly scientific.

Since I am taking a sceptical stance vis a vis Kuhn's notion of what is truly scientific, I am also, by the same token, questioning whether his precise definition of "paradigm" is what distinguishes "science" from "non-science" (e.g. the Social Sciences, according to Kuhn). I take "science" to be a term that has more connotative than denotative content, and one that is too strongly associated with Mathematics.

I therefore feel the need to coin the term Paranickel, which can be seen as a watered-down version of Paradigm. A "Paranickel" has three components: models, values, and exemplars -- i.e., it differs from a paradigm by dint of not requiring symbolic formalism ( which is an optional extra, in my view).

There is no requirement that it originate in the work of a single individual (except to the extent that is explained below) . What actually happens when one Paranickel overthrows another in a Revolution, in my view, is that a latent contradiction in the established Paranickel is exposed by a particular individual (usually). However, Percival (1976) is right to point out that such individuals might actually seem to differ very minimally from their immediate predecessors and contemporaries -- with hindsight. However, at the time they were seen to make the academic community aware for the first time that one aspect of the current Paranickel was actually incompatible with another aspect of it. Thus the individuals concerned saw themselves as forced to choose between the two incompatible aspects. Historians of Science, subsequently, could see them as revolutionary if they focused on their rejection of one of the aspects, or as conservative (viz. Percival (1976)) if they focused on those aspects of the established Paranickel that were carried over into the new one.

As far as the contrast between Autonomist and Cognitive Linguistics is concerned, I see the former (in its Chomskyan variant) as containing a contradiction between the notion that Linguistics is a branch of Cognitive Psychology, and that part of the Competence-Performance Distinction (see Chapter One) which effectively protects the work of Linguists and Cognitive Psychologists from what might be called "mutual contamination". Katz (1981) resolves this contradiction by retaining the relevant aspects of the Competence-Performance Distinction, but denying that Linguistics is part of Cognitive Psychology. Cognitive Linguists, as I see it, resolve it in the converse way: by agreeing that Linguistics is a branch of Cognitive Psychology, but rejecting certain parts of the Competence-Performance Distinction.

A paranickel, unlike a paradigm, does not need to command uniform assent among all members of the relevant discipline. This has to do with the Sociology and power-structures of Academia: there exist numerous "theories", such as the Flat Earth theory and Astrology -- not to speak of various forms of Alternative Medicine -- whose adherents are excluded from membership of the mainstream discipline . In this way, "scientific" disciplines appear to be homogeneous as to Paradigm-adherence for the simple reason that alternative Paradigms have been excluded by the administrative powers that Universities, academic journals and learned societies exercise in their usual unbridled fashion. Thus this homogeneity, which is so central to Kuhn's distinction between science and non-science, is more apparent than real.

I do not consider that there is a significant distinction to be made between the notion of two distinct paranickels and the notion of paranickels that are variants of each other: the issue is purely one of degree. As stated above, each paranickel carries over aspects of the preceding one, and can therefore, if desired, be considered as a variant of its predecessor.

It should also be mentioned that discipline-boundaries are themselves less secure and sharp than Kuhn's theory might make them seem: within the field labelled "Linguistics", for example, there are subdisciplines with their own paranickels. The case of the Labov/Trudgill approach to Sociolinguistics springs to mind. It is my hope and expectation that a Cognitive approach would eventually develop a Metatheory broad enough to subsume such subparanickels under it.



2.Aims.


The present work comes into the category of what Derwing (1973) calls "foundational analysis". However, it differs from that work in at least four important respects:


. It contains a descriptive part to balance and exemplify the theoretical part;

. It claims adherence to the relatively new Cognitive school of Linguistics, and thus can be seen as part of a positive alternative to the Autonomist schools.

. It proposes positive alternatives to some of the more controversial basic assumptions of Autonomist Linguistics.

. It has nothing to say about Language Acquisition -- the focus being on Psycholinguistics, Syntax and Semantics.


I find myself, no doubt like most Cognitivists, in total agreement with the following:


...the empirical status of (Transformational Generative) grammars is sufficiently obscure and the viability of the basic assumptions sufficiently in doubt to warrant a fundamental shift in the attitude of linguists both to the nature of the work they do or ought to be doing and of the results they achieve or ought to be achieving. It is characteristic of foundations that they support the superstructure, and linguistics is no exception.... current work in linguistics along transformational-generative lines is yielding little in the way of substantive accomplishments of any empirical significance, while offering much in the way of unsupported (and, given present methods of research, unsupportable) speculative assertion, with the larger portion of current metatheoretical discussion being devoted to marginal or irrelevant "smoke screen" issues which hide deep conceptual and methodological difficulties (Derwing, op.cit.,5).


The following two quotations are also of relevance:


(a) ...those who uphold ( a system such as Classical Mechanics ) dogmatically -- believing, perhaps that it is their business to defend such a successful system against criticism as long as it is not conclusively disproved -- are adopting the very reverse of that critical attitude which in my view is the proper one for the scientist. In point of fact, no conclusive disproof of a theory can ever be produced; for it is always possible to say that the experimental results are not reliable, or that the discrepancies which are asserted to exist between the experimental results and the theory are only apparent and that they will disappear with the advance of our understanding (Popper 1959,50).


This first quotation is relevant, because there appear to be some Generativists (e.g.Chomsky 1982a,45) who take the line that Popper is concerned to discredit.


(b) ...if new data has to be introduced, we usually find that it has been barred for ideological reasons, or not even been recognized as data at all, and the new methodology must do more than develop techniques. It must demolish the beliefs and assumptions which ruled its data out of the picture. Since many of these beliefs are held as a matter of deep personal conviction, and spring from the well-established habits of a lifetime, this kind of criticism is seldom accomplished without hard feelings and polemics, until the old guard gradually dissolves into academic security and scientific limbo. (Labov 1972,99)


The relevance of this second quotation lies in the fact that Cognitive Linguistics seems (to me, at least) to entail using data collected by Psychological means in conjunction with data collected by traditional Linguistic means, in order to draw conclusions which are equally Linguistic and Psychological. In Generative Grammar, the dogma which (as one of its primary functions) serves to proscribe such direct cooperation between Psychology and Linguistics is called the Competence-Performance Distinction. In theory, Generative Competence models are intended to be embedded into Psycholinguistic Performance models, but in the thirty-odd years that this agenda has existed, I am not aware of any detailed proposal along these lines having been proposed. Nor am I aware of any Performance account of any phenomenon which demonstrably requires an Autonomist-style Grammar as an intrinsic component. Chapters One and Two take up this issue in some detail.

The present book is essentially an attempt to clarify some of the issues which distinguish the recently and currently dominant Autonomist schools of Linguistics from the emerging Cognitivist school. Thus some theoretical, or metatheoretical issues are covered in some detail in Part One, and then specific descriptive issues are addressed in Part Two in terms of the theoretical points made in Part One. I feel it is important to carry out this exercise in clarification, so that Cognitivists can build on those aspects of Autonomism which are valid and useful -- rather than having the two schools as parallel lines that never meet. The latter approach, while tactful and gentlemanly, is not likely to produce progress towards a unified paradigm or paranickel in our field.

The emerging Cognitive Linguistics school cannot at present be briefly characterised in any illuminating way. The present work is intended as a modest contribution towards the cristallisation of its central tenets. In the meantime, all I can do to help the reader understand what I mean by that term is to point him in the direction of such works as Clark and Haviland (1974), Lakoff (1982), and Langacker (1987).

I assume that a Cognitive Linguist tends to take a different line from an Autonomist as to what constitutes an "explanation" in Linguistics. For example, in his discussion of Mandarin gei, Newman (1989) writes:


The range of meanings which attach to gei is not predictable just from considerations of the basic meaning GIVE. But neither is the range of meaning arbitrary. In between these two extreme conditions lies the additional condition of being unpredictable, but motivated. The preceding discussion has shown that the range of meanings of gei can indeed be motivated in so far as they can all be traced back to some facet of the basic human experience of giving something to someone.... Unless one recognizes this experiential basis and all that is associated with it, the relationships between the various meanings of gei can not possibly be appreciated.


Since Cognitive Linguists view Language as an essentially fuzzy biological phenomenon, rather than as deriving from some quasi-Cartesian system of grammatical rules hermetically sealed in a "language faculty", the above can be accepted as explanatory and revealing. Autonomists would tend to reject such analyses as unilluminating -- but they would (I would expect) be unable to come up with a better explanation of the facts that Newman (ibid) is setting out to account for.

The present work is based on the assumption that Functionalism is compatible with Cognitive Linguistics. The language user is viewed as an organism operating in a goal-directed way in a particular environment.

Cognitive Linguistics, as such, tends to concentrate on the relationship between language and the internal (cognitive) characteristics of the human organism. Functionalism, on the other hand, concentrates on the interface between the means (language) and the goal that the organism is using language to achieve. Both perspectives are essential to a complete account of human language.

On the methodological level, I argue that a Functionalist/Cognitivist is justified in asking how plausible and probable a particular example is, and whether there are more plausible/probable alternative ways of communicating the same message.

Thus I use from time to time arguments compatible in spirit with the following:


A neo-Gricean conversational principle: If a language has two (equally simple) types of syntactic structures A and B, such that A is ambiguous between meanings X and Y while B has only meaning X, speakers of the language should reserve structure A for communicating meaning Y (since B would have been available for communicating X unambiguously and would have been chosen if X is what was intended) Dowty (1980).


Chapter One is a revised version of a paper presented at the eighth New Zealand Linguistics Society Conference at Auckland, New Zealand, 1989.

An earlier version of chapters One and Two appeared as Zohrab (1989).

An earlier version of Chapter Three appeared as Zohrab (1986).

Chapter Four originated as a talk presented to the Linguistics Association of Great Britain in 1982.

I would like to thank Nittaya Campbell for her helpful comments on an earlier draft, though she bears no responsibility for the use I made of her suggestions.

I would also like to thank Annabelle Cormack, Deirdre Wilson, Martin Davies, Max Cresswell, Robyn Carston, Laurie Bauer, and anonymous referees from the journals Humor, Cognitive Linguistics, Linguistics and Philosophy, and Linguistics for their comments on earlier versions of parts of this book.

I would like to commend Dick Hudson on his determination (albeit fruitless) to put me off the line of thinking that resulted in Chapter Five, and also my father, Balfour Douglas Zohrab, for his moral support during the writing of this book, in its various stages.

The responsibility for the final version is, however, very much my own.

Back to Contents Page