The paradigm shift:From structuralism to transformationalism [Archives:2009/1226/Education]

January 19 2009

Dr. Anwar Al-thwary
[email protected]
Assistant Professor
Department of English
Faculty of Education
Sana'a University

In the issue no. 24 of the Education Supplement I talked in brief about the most significant trends and developments of Linguistics from the time of Greeks and Romans to the Modern Age. In the present article, I'll try to elaborate two linguistic approaches of the 20th century, namely the Structural Linguistics (Descriptive Linguistics) and Transformational-Generative Grammar (T-G Grammar). In holding a comparison between the two and in terms of Thomas Kuhn's concept of “Paradigm Shift”, the focus will be on the theoretical framework of the two approaches in general and how each theory views the issue of language acquisition.

In various fields of knowledge (e.g. Medicine, Physics, Education, etc.), every thing is bound to change and Linguistics is no exception. This change (in any field) is known among scientists as “Paradigm Shift”. Throughout the history of linguistics, the linguistic theories, thoughts and ideas have kept changing. Proposed by the American linguist Noam Chomsky in the 2nd half of the 20th c., T-G Grammar is a revolution against Structural Linguistics which is developed by Ferdinand de Saussure in Europe and Bloomfield in America in the 1st half of the 20th c. The Structural Linguistics itself is a reaction against what is called Historical or Comparative Linguistics which, in turn, is a reaction against the Traditional Linguistics.

The concept of “Paradigm shift” was first introduced by Thomas Kuhn in 1962 in his book “The Structure of Scientific Revolution” which represents a turning point in the people's viewing and perceiving the advances and changes in the scientific and also non-scientific aspects of human life. As a result, there has been an increasing tendency in academic world to view various developments and “shifts” in different fields of inquiry as emergence of Kuhn's type “paradigms” and “revolutions”. People have found Kuhn's terminology and ideas behind them as good approximants which could be used, not only to designate the identical characteristics of the progress made in different fields of inquiry, but also to carry out potential “wars” against the opposite “camps”. According to Kuhn the term “paradigm” is used in two different senses. On one hand, it stands for the entire constellation of beliefs, values, techniques, etc. shared by the members of a given community. On the other, it denotes one sort of element in that constellation, the concrete puzzle-solutions which, employed as models or examples, can replace explicit rules as a basis for solution of the remaining puzzles of normal science. Paradigms provide all phenomena, except anomalies, with a theory-determined place in the scientist's field of vision; they guide research by direct modeling as well as through rules. In this sense, “Paradigm” means a set of procedures, models, and hypotheses.

The first half of the twentieth century saw the emergence of a group of psychologists known as “behaviourists” (or empiricists). Among these is F. B. Skinner who is considered the father of the School of Behaviourism. He describes language acquisition process on the basis of the Association Theory in his book “Verbal Behaviour”. For Skinner and other empiricists, language acquisition is the result of stimulus-response activities. Imitation, repetition, memorization, reward, and reinforcement facilitate this process of language acquisition. They say that all knowledge is derived from experience. They think that children start out as “clean slates”. Learning a language is a process of getting linguistic habits printed on these slates. Consequently, there is no difference between learning a language and the ways one learns to do any thing else (swimming, driving, etc.). So, language acquisition/ learning is basically a mechanical process of habit formation and conditioned behaviour. This follows that as long as individuals are subjected to the same condition, they will learn in the same way and variations in learning occur because of the difference in learning experience. For behaviourists, there is not a theory of a language learning as such but merely the application of general principles of learning to language learning. At that time, the School of Behaviourism gave rise to the Structural Linguistics. The structuralists adopt almost all the above ideas of Behaviourism and apply them to language learning.

If the Behaviourist School of Psychology gives rise to Structural Linguistics, Transformational-Generative Grammar is based on the Cognitive Psychology. The structuralists believe that the mind is a blank slate, but the mind in the Cognitive Paradigm becomes of prime importance. The cognitivists contradict the empiricists at almost every point. As far as language acquisition is concerned, the cognitivists believe that a child learns or acquires a language, not because he is subjected to a similar conditioning process, but because he possesses an inborn capacity, which enables him to acquire a language as a normal process. This capacity is universal. Chomsky calls it Language Acquisition Device (LAD). This LAD consists of the hypotheses maker, linguistic universals, evaluation process and finally grammar. So, the child learns a language by exposure to it in society and by unconsciously forming certain hypotheses about language, which he goes on modifying till he comes to adult model. Thus the child goes on constructing an innate grammar, operating over generalized rules.




-Linguistic universals

-Evaluation process

For rationalists, language acquisition is species-specific (specific only to one set of species, i.e. human beings) and species-uniform (all human beings uniformly posses language except deaf and dumb people). The ability to take up and understand language is inherited genetically but the particular language that child speaks is culturally and environmentally transmitted to him. Language acquisition thus appears to be different in kind from acquisition of other skills such as swimming, driving, etc. It is really the result of nature, not of nurture and the child's imitations are not merely photographic reproductions, but artistic recreation.

One can distinguish two phases in Chomsky's revolution. The first phase begins with his dissatisfaction with American Structuralism and ends with his full recognition of the difficulties of the taxonomic theory of grammar and with his intention to replace it with his newly conceived theory of Generative Grammar based on an analogy to category systems in logic. The second phase begins with the recognition that is responsible for most of these difficulties and that new theory of Generative Grammar thus requires a new ontological interpretation for grammars. The second phase ends with Zellig Harris' transformational theory refitted with more appropriate psychological interpretation within Chomsky's conceptualist framework. T-G Grammar of Chomsky is, thus, a chemical mixture of linguistics, logic, mathematics and psychology.

Chomskyan Grammar is “theory-based” approach because it helps in understanding, in a general way, what a language is. As opposed to this, the grammar of Structural phase follows a “text-based” approach and is description oriented, both in form and function. Chomskyan Syntax provides perspectives and generates operational framework for writing specific grammar. Chomsky makes a basic distinction between “competence” and “performance”. Language is a set of rules. Competence is the knowledge of these rules and performance is the use of this knowledge. Chomskayn Grammar, therefore, is a well-defined theory of language. Structural grammarians, however, view language only as form and analyze the data of a given corpus by means of “inductive methods” and formulate a grammar through discovery procedures. Kuhn states that “what changes is not the “presence” of a paradigm but rather its “nature””. The shift, in question, takes place from purely descriptive grammar in nature to generative syntax.

This centrality of syntax in T-G Grammar with the sentence as its unit of analysis goes a long way to explain the immense creativity in Language. The discovery procedures used by structuralists make use of Immediate Constituent analysis (IC analysis), distribution criterion and other devices. Such procedures fail to capture all language ambiguities and relations. The IC analysis, as its name denotes, is not grammar; it is a kind of analysis that concerns with the surface structure of the sentences rather than their deep structure. It doesn't account for the degree of grammaticality and acceptability; nor does it stop the generation of ungrammatical utterances. It doesn't also include the idea of creativity. The grammar produced then is not predictive, projective and explicit. Discovery procedures also ignore linguistic universals and native speakers' intuition and his competence of generating infinite number of sentences out of a finite set of items. Chomskyan Syntax, on the other hand, provides a set of rules which is called Phrase Structure Rules (or Generative Rules). Such rules are able to manipulate most of, if not all, the problems mentioned above. They are precise, explicit, highly formalized, predictive and projective. They generate mainly the deep structure. Chomsky holds the view that notions of phrase structure are adequate for a small part of the language and the rest of the language can be derived by repeated application of rather simple sets of transformations (interrogative, negative, passive, etc. sentences) to the strings given by the phrase structure grammar to cover the entire language directly. This set of rules is called Transformational Rules. The notion of discovery procedures which is based on the empirical method and observation adequacy is replaced by the notion of “evaluation procedures” in T-G Grammar. They tell us which member(s) of a set of proposed generative grammars for a language is the preferable theory of the language.

It can be said that the older theories are important because it is their inadequacy, which causes new questions to be asked and new answers to be sought giving rise to a change or a shift in the paradigm. The whole Structural Paradigm underwent severe questioning. After 1957 questions were asked about the psychology of language learning, about the philosophy, about the goals and methodology, etc. The questions raised by Chomsky in his review of Skinner's Verbal behaviour strike at the very roots of Structuralism. Chomsky rejects the stimulus-response approach to the study of language because it cannot explain the originality in language. His greatest criticism of verbal behaviour is that it equates animal learning and human learning. The transformationalists discard almost every thing in structural paradigm. But the decision to reject one paradigm is always simultaneously the decision to accept another. Consequently, the linguists of T-G Grammar announced the war against the camp of structuralists. So, there is a revolution from the empirical, data-based structural description to a rational and mentalist approach to the study of language.

The paradigm shift, then, is a process that involves handling the same bundle of data as before, but placing them in a new system as before, of relations with one another by giving them a different framework. Both the Structural Paradigm and the Transformational Paradigm deal with the same data, i.e. language data. But, while structuralists look at language as an observable behaviour and a conditioned activity, Chomsky comes to consider language as a cognitive activity which is species specific and species uniform and acquired by an innate mechanism in human mind (LAD).

To sum up, there is a complete paradigm shift that takes place from the Empiricist Paradigm to the Rationalist Paradigm. There are changes in the whole context: in the basic hypotheses, the objective of analysis, the nature of language and in the procedures and methodology. “Shifts of this sort are associated with all discoveries achieved through normal science”. While Structural Grammar aims at a description of language itself as it is used by the community, i.e. considering the language itself as the locus of grammar in the domain of structures, forms and distributions, T-G Grammar aims at generating and explaining all and only grammatical sentences of the language. The locus here is the human mind (native speaker's competence) and the domain is the rules of linguistic competence.