Remerciez-le!

Remerciez @Admin pour avoir partagé cet document gratuitement, de la manière la plus simple, en partageant sur les réseaux sociaux.

Esa Itkonen What is methodology (and history) of linguistics good for, epistemo

Esa Itkonen What is methodology (and history) of linguistics good for, epistemologically speaking? In: Histoire Épistémologie Langage. Tome 13, fascicule 1, 1991. pp. 51-75. Résumé RESUME : Cet article tente de répondre à la question figurant dans son titre : Quelle est l'utilité de la méthodologie et l'historiographie de la linguistique, du point de vue épistémologique ? Pour trouver une réponse, on passe en revue les thèmes suivants : théorie grammaticale (aussi connue sous le nom de 'linguistique autonome1), linguistique causale, historiographie de la linguistique, étude du langage et de l'esprit. Finalement, on fait quelques observations sur le débat récent concernant la nature de la linguistique. Abstract ABSTRACT : This paper tries to answer the question that figures in its title : What is methodology (and history) of linguistics good for, epistemologically speaking ? An answer is sought for, in turn, in the following areas : grammatical theory (also known as 'autonomous linguistics1), causal linguistics, historiography of linguistics, and the study of language and mind. The paper concludes with some remarks on the recent discussion concerning the nature of linguistics. Citer ce document / Cite this document : Itkonen Esa. What is methodology (and history) of linguistics good for, epistemologically speaking?. In: Histoire Épistémologie Langage. Tome 13, fascicule 1, 1991. pp. 51-75. doi : 10.3406/hel.1991.2324 http://www.persee.fr/web/revues/home/prescript/article/hel_0750-8069_1991_num_13_1_2324 Histoire Épistémologie Langage 13/1 (1991) WHAT IS METHODOLOGY (AND HISTORY) OF LINGUISTICS GOOD FOR, EPISTEMOLOGICALLY SPEAKING ? Esa ITKONEN ABSTRACT : This paper tries to answer the question that figures in its title : What is methodology (and history) of linguistics good for, epistemologically speaking ? An answer is sought for, in turn, in the following areas : grammatical theory (also known as 'autonomous linguistics1), causal linguistics, historiography of linguistics, and the study of language and mind. The paper concludes with some remarks on the recent discussion concerning the nature of linguistics. RESUME : Cet article tente de répondre à la question figurant dans son titre : Quelle est l'utilité de la méthodologie et l'historiographie de la linguistique, du point de vue épistémologique ? Pour trouver une réponse, on passe en revue les thèmes suivants : théorie grammaticale (aussi connue sous le nom de 'linguistique autonome1), linguistique causale, historiographie de la linguistique, étude du langage et de l'esprit. Finalement, on fait quelques observations sur le débat récent concernant la nature de la linguistique. 1. General Remarks. Human beings are unique in that, while doing something, they generally also know what they are doing. This ceases to be the case, 52 Esa Itkonen however, when they engage in exceptionally complex activities. Science is a prime example of such an activity : it is not at all uncommon to see 'ordinary practising scientists' go about their work, while having only a vague, or downright false, idea of what it is, exactly, that they are doing. To me such a state of affairs is degrading ; and, accordingly, knowing what one is doing has an intrinsic value. This is where methodology and/or philosophy of science comes into picture. In general, philosophy of science is considered a respectable undertaking. On closer inspection, however, some differences emerge. The philosophy of natural sciences, especially astronomy and particle physics, enjoys great prestige. So does the philosophy of logic and mathematics, even if to a somewhat lesser extent. But the philosophy of linguistics is sometimes treated with mistrust. To me this is a sign of (temporary, let us hope) immaturity within the field of linguistics. It may be good to add that there are two (often related) ways of practising methodology which I reject straight away. One is to maintain that a given scientific discipline (here : linguistics) is based on some colossal misunderstanding, with the implication that the achievements of more than two thousand years just have to go down the drain. The other is to dwell upon some Utopian stage of a given scientific discipline (here : linguistics) that lies far off in the future. A sound methodology must be based on a detailed 'phenomenological' account of what the scientist is doing in fact, not what he thinks he ought to be doing. For instance, in the early phase of a typical grammatical description the grammarian observes nothing, i.e. he neither sees nor hears utterances. Rather, he invents sentences which his linguistic intuition tells him are correct, and he mentally performs some operations which either preserve correctness or change it into incorrectness. Typically, such a mental operation consists in altering either the form or the distribution of constituents, and if it produces an incorrect sentence, the grammarian becomes aware of a (possibly very low-level) rule which has been violated. That is, incorrectness never concerns one sentence alone. In all this, the grammarian has done nothing but analyze his own consciousness, or more precisely, what he himself consciously knows to be the case. Obliterating the distinction between conscious and unconscious Methodology of linguistics 53 makes here any methodological self-understanding impossible a priori. So does construing the grammarian's activity as an instance either of the observational or of the experimental method. In this paper I shall discuss some lessons that can be drawn from investigating the methodology and/or philosophy of linguistics, with the emphasis on the epistemological aspect. Understandably, I shall concentrate on questions that I have dealt with in my own work. 2. The Status of Grammatical Theory. The basic tenets of my 1978 book Grammatical Theory and Metascience1 (henceforth to be abbreviated as GTM) can be summarized as follows. A language is a system of social rules (or norms). The social nature of rules means that they exist as objects of common (or mutual) knowledge. Thus, one can become aware of rules, which means, more technically, that rules can become objects of subjective atheoretical intuition. Rules determine the correctness or incorrectness of sentences. Any rule can be described by a corresponding (atheoretical) rule-sentence, for instance, 'The precedes, as in the man, and does not follow, as in man the'2 or 'Tree means a plant, and not a number, whereas three means a number, and not a plant' or 'In a sentence like John likes him the word him cannot refer to John'. It is a peculiarity of rule-sentences that since they are known to be true, they are (known to be) unfalsifiable. Any purported counter-examples, like uttering the sentence 'Man the came in1, are simply irrelevant, because they are (known to be) incorrect. It is this, and only this, feature that distinguishes (descriptions of normative) rules from (descriptions of non-normative) regularities. Denying the unfalsifiability of rule- sentences amounts to denying that there is any 'rule vs regularity' - 1 . This is a revised and expanded version of my 1974 dissertation Linguistics and Metascience. 2. The original formulation of this rule-sentence was 'In English the definite article precedes the noun*. This created some confusion because 'definite article' and 'noun' are undeniably theoretical terms (or concepts). I pointed out from the beginning, however, that such terms can be dispensed with ; see GTM : 325-326, n. 90, and 167468. 54 Esa Itkonen opposition. In thus revealing the normative nature of the subject matter of grammatical theory, rule-sentences possess considerable metascientific interest. Because of their atheoretical character, however, they possess no scientific interest3. Such an interest appertains only to theoretical grammatical descriptions, like generative grammars of (fragments of) given languages. A rule as well as the truth of the corresponding rule-sentence is known by intuition, but since one has no prior knowledge about a theoretical description, one cannot know it by intuition. This means that a theoretical description is itself not a normative entity, even if it has been invented for the purpose of giving a systematic overview of normative data (see GTM : 8.3, esp. pp. 215-216, and 9.3, esp. 246-247). They key notions in this account are 'rule1, '(atheoretical) rule-sentence', and '(theoretical) grammar'. The question has been much debated whether linguistics is just a subbranch of psychology or whether linguistics-<?Ma-psychology should be distinguished from grammatical theory (also called 'autonomous linguistics1). In my published work I have argued at length for the latter alternative. In the present context it suffices to point out that grammatical theory, as here defined, is based on the notion of axiomaticity , or on the idea of 'describing as much as possible by means of as little as possible'. Experimental- psychological studies have established that the (untrained) human mind does not function axiomatically. (This is not surprising. Anyone who has studied axiomatics knows from bitter personal experience that he has to struggle against his natural inclination.) 3. The 'atheoretical vs. theoretical1 distinction has turned out to be surprisingly difficult to grasp. (Its general justification is given in GTM: 8.2-3.). The following should clarify the issue : "When formulating our norm-sentences [or rule-sentences] we must have some minimal trust in the intellectual capacities of our audience. In general our trust is justified. Therefore I could just as well have formulated my norm-sentence as 'the man is right, and man the is wrong', and could have let the uploads/Philosophie/ itkonen-epistemologia.pdf