In the early years of this century, a revolution in linguistic science was initiated by the Swiss linguist, Ferdinand de Saussure, who is often referred to as the ‘father of modern linguistics’. Saussure drew a distinction between diachronic and synchronic approaches to the study of language. The historical concerns of the 19th century fall under the heading of diachronic linguistics, with their emphasis on processes of language change. Despite his significant contributions to the diachronic tradition, Saussure argued for the primacy of the synchronic approach, in which the focus is on the language system as it exists at a particular point in time. The development of synchronic linguistics coincided with the growing interest in North America to record and describe adequately the numerous languages of native Americans which were under threat of extinction. As a result, the attention of many linguists turned to the study of living languages as systems in their own right, often with little regard for the formative influences which shape languages. After all, as Saussure pointed out, as far as the speakers of a language are concerned, the evolutionary history of words and structures is largely irrelevant. In any event, linguistics in the 20th century has been dominated by the synchronic approach, although historical concerns are still actively pursued.

Another important distinction in linguistic research centres on the difference between describing a particular language as opposed to describing language in general. The general approach recognizes language as a uniquely human phenomenon which distinguishes human beings from animals. A central aim is to reveal the limitations of human language and the characteristics which distinguish it from other systems of communication. In providing the general concepts and terminology with which language may be described, this approach also contributes to the investigation of individual languages. Particular languages have been studied for a variety of reasons, for example, the desire to describe norms of usage in grammars and dictionaries. Insights gained from the study of specific languages can also be of great value for confirming or refuting ideas in the study of language in general. Indeed, it has been argued that the study of a single language in depth can provide insights concerning the universal properties of language. However, the fact that this single language was almost inevitably English probably revealed more about the short-sightedness of certain linguists than about the value of this method. In fact, this extreme view has receded in recent years and now it is common for linguists to explore theoretical issues with data from a much broader spectrum of cross-linguistic data.

The prescriptivist approach to language aims not only to deduce the grammatical rules of a language, but also decrees certain norms and standards of correct usage. However, the reasons for declaring one language variant to be correct, while competing forms are incorrect, often has no basis in linguistic theory at all. Instead, decisions are often motivated by sociological concerns, such as the relative prestige of one form over another. Hence, linguists nowadays generally regard their task as one of describing how a language actually is, as opposed to promoting ideas about how it should be. In other words, modern linguistics is descriptive rather than prescriptive. It should be noted that a descriptive approach quite properly investigates what is and is not possible in a language, but only from an objective, non-judgemental point of view. Despite neglect by linguists, prescriptive issues continue to generate often heated debates among many people, as testified by the letters page of many newspapers. More seriously, prescriptive concerns are a major consideration in the process of language planning, when multilingual nations are faced with the complexities of selecting and promoting an appropriate national language. However, these kinds of issues are more properly the concern of applied linguistics, which means that theoretical linguistics generally manages to remain purely descriptive in orientation.

The quest for scientific status was initiated in large part by the American linguist, Leonard Bloomfield (1887 - 1949). It was his belief that linguistics should confine its theorizing to hypotheses which could be confirmed or refuted using objectively verifiable data. This aim was achieved by gathering a large corpus of language samples and subjecting them to a systematic analysis. This approach suffers from certain drawbacks which were first highlighted in the 1960s by Noam Chomsky, who has also been responsible for a further fundamental shift in linguistic research in the latter half of this century. Chomsky pointed out that no matter how large a language corpus is, it will never contain examples of every conceivable kind of linguistic construction. Rather than pursue this futile aim, attention has shifted towards unearthing the principles of the linguistic system which underlie the actual sentences being analysed. In this way, due consideration is given to the fact that a finite system of grammatical rules can generate an infinite number of novel sentences.

This shift of emphasis allowed large bodies of data to be supplemented by the spontaneous grammatical intuitions of native speakers about what is, and is not, grammatically permissible in a language. Linguists often rely on their own intuitions, or alternatively, make use of an informant. In either case, it is important that intuitions provide a reliable source of information, and fortunately, this is indeed the case in the majority of cases. Intuitions are not infallible, though, since there are many occasions when linguists disagree. Nevertheless, the restoration of linguistic intuitions as an acceptable form of data reveals the importance of tapping the underlying mental system which gives rise to language behaviour. As a result, the last four decades have witnessed a radical reassessment of the goals of linguistic theory. The analysis of language as an abstract object, in and of itself, has given way to the task of characterizing the knowledge of language as it exists in the minds of native speakers. MS