Willard van Orman Quine - SpeedyLook encyclopedia

Biography

Quine studied the Logique with the mathematician and philosopher Alfred North Whitehead in Harvard. After its thesis in mathematical logic, it went on a journey in Europe where it met in particular Rudolf Carnap. He was professor of philosophy in Harvard of 1956 with his death.

Work

Quine contributed to the formal Logique, with the Fondation of mathematics but also with the philosophy of the language and epistemology.

Its project of naturalism or “naturalized epistemology” was that the philosophy of knowledge and sciences is itself a scientific activity, corrected by other sciences, and not a “philosophy first” founded on a metaphysics.

This naturalism is accompanied by epistemological Holisme, the thesis according to which all our knowledge is supported mutually without there being a single foundation (what it often summarized by taking again the image of the ship of Otto Neurath according to which science is a ship already at sea and which it is necessary to repair starting from materials available without rebuilding it on a dry land).

Methodological Monism

Quine was a “extensionalist” for whom all the Logique must be reduced to the extension and exclude any reference to the Intension (comprehension) to distinguish from the intention.

That led it to criticize the “Myth of the significance” and to refuse cleavage Net between analytical proposals and synthetic proposals. In the two dogmas of empiricism , it supports that the concept of analycity cannot be clearly defined: the concept of analycity is explained by that of synonymy necessary between two terms in certain contexts, but by saying that one presupposes the concept of analycity (already contained in " nécessaire"). The analysis of the concept of analycity would be thus circular.

Whereas the empiricism of Rudolf Carnap had a methodological “dualism” between logical and mathematical conventions and in addition the facts, Quine considers that one cannot distinguish in a distinct way these two sides of outside without presupposing them.

Indetermination of the translation and inscrutability of the reference

In the Word and the Thing (1960), Quine refutes the ontology of Russell and Frege in the name of the principle of indetermination of the translation.

“the philosophers and the linguists always said that the language is a social institution. They, however, immediately, forgot it and adopted concepts of significance which are not publicly accessible Quine seems to be the first to be taken with serious the public nature of the language and to explore its consequences on the significance and the communication. ”

The thesis of Quine rests on a design behaviorist of the training of the language. According to Quine, the establishment, the training and the use of the language rest strictly on publicly accessible data, i.e. on the observation of the behavior of the others.

“In psychology one may gold may not Be has behaviorist, goal in linguistics one has No choice. Each off custom learns his own language by observing other people' S verbal behavior and having his own faltering verbal behavior observed and reinforced gold corrected by others. We depends strictly one overt behavior in observable situations. Ace long ace our command off our language fits all external checkpoints, where our utterance gold our reaction to someone' S utterance edge Be appraised in the light off nap shared situation, so long all is well. Mental Our life between checkpoints is indifferent to our rating ace has master off the language. ”

Quine concludes from it that there is nothing in the linguistic significance beyond the data on the linguistic behavior publicly accessible from the speakers of a language. The significance of a linguistic expression is the joint product of all the behavioral data which help learning them and the users of the language to determine this significance.

But up to which point these data do they tend to determine the significance of the linguistic expressions? The thesis of the indetermination of the significance of Quine aims at answering this question. To work out its thesis, it compares one learning from a language with a linguist of ground which undertakes to prepare a handbook to translate in its language an indigenous language lately discovered, and it up to what point wonders the data on the linguistic behavior of the natives tend to determine the translation of the linguistic expressions of the indigenous language.

The ultimate data to which access the linguist has, like all learning from a language, are data on the linguistic behavior publicly accessible from the natives. The linguist prepares his handbook of translation of the indigenous language by conjectural extrapolation of these data. The “observational sentences” are the main door of the linguist to the indigenous language. They are sentences, as “This is a rabbit. ” or “It rains. ”, which depends rather strictly on the situations concomitant publicly accessible and on which the majority from the members from a community can agree.

The linguist initially seeks to translate such sentences. To determine that an indigenous sentence is “observational”, the linguist states itself this sentence in various situations in the presence of various natives and checks if those give their approval or their disagreement to the stating. If the natives give their approval or their disagreement to this stating only in certain publicly accessible situations, the linguist concludes that this sentence is observational and it translates it by a sentence to which it gives his approval or its disagreement in similar circumstances. Thus, the data on the linguistic behavior of the natives determine the translation of the observational sentences.

The majority of the sentences, however, are not observational, i.e. they do not achieve the unanimity in a community and that they have relationship with the publicly accessible concomitant situations little. Thus, the linguist cannot be based on the concomitant situations to translate such sentences. He is based on his translation of the observational sentences of the native. Certain segments of the nonobservational sentences, or theoretical sentences, will be in the observational sentences of the indigenous language. He will regard these segments as words and will translate them while being based on the translation of the complete observational sentences. Then, on the basis of translation of these words, he interpreted the nonobservational sentences of the indigenous language in which the words are. Thus, to translate these sentences, the linguist will have recourse to “analytical assumptions”.

Let us recapitulate. The starting point of the translation of the indigenous language, we saw it, is the translation of the observational sentences. To translate these sentences, the linguist poses assumptions resting directly on the available data, that is to say the data on the linguistic behavior of the natives. Then, in order to interpret the theoretical sentences of the indigenous language, the linguist must project analytical assumptions. These assumptions do not rest directly on the observation of the behavior of the natives in publicly observable circumstances. They draw their ultimate justification owing to the fact that they are compatible with the observational sentences.

However, according to Quine, so of the data do not refer directly to an assumption, but only to the observational consequences of this one, this assumption is under-given by the data, i.e. there can be another assumption which is incompatible with it and, nevertheless, compatible with all the conceivable data. Therefore, since the data on the linguistic behavior of the natives refer to the observational consequences analytical assumptions, that is to say with observational sentences, it follows, according to Quine, that the analytical assumptions are under-given by these data.

As the translation of the theoretical sentences of the indigenous language rests on analytical assumptions, Quine concludes from it that it is also under-given by the data on the linguistic behavior of the natives. These data are too poor to fix the translation of the theoretical sentences of this language. Various handbooks to translate these sentences can be elaborate in such a way that they all are compatible with the data on the linguistic behavior of the natives and, however, intuitively incompatible between them. (In an incalculable number of places, these handbooks will seem to diverge. Interpretations which they will respectively give of a given theoretical sentence of this language will be sentences theoretical of the basic language being the ones towards the others in any kind of relation of equivalence intuitively plausible, for coward that it is.)

As we saw, the data on the linguistic behavior of the natives determine the translation of the observational sentences. Does one have to conclude from it that the translation of the terms which are in these sentences is also determined by these data? For example, let us imagine that “Gavagai” is an observational sentence of the indigenous language which results in “Here is a rabbit”. According to Quine, there are many of other sentences of our language which all are compatible with the data on the linguistic behavior of the natives: “There is a not detached part of rabbit. ”, “There is matérialisation of the lapinitude. ”, “There is a stage of the life of a rabbit. ”, “This place is with two kilometers on the right point of the space located at two kilometers on the left of a rabbit. ”, etc Indeed, each time the native gives his approval to “Gavagai” in the presence of a rabbit, it gives simultaneously its approval to “Gavagai” in the presence of a not detached part of rabbit, matérialisation of the “lapinitude”, of a stage of the life of a rabbit, a place being with two kilometers on the right point of the space located at two kilometers on the left of a rabbit, etc

Thus, if the linguist concludes that “gavagai” is a term, then to determine if it is about a rabbit, of a not detached part of rabbit, of matérialisation of the “lapinitude” or of a stage of the life of a rabbit, it will be obliged to ask the native: “Is this gavagai the same one as the other? ”, “is this a gavagai or two? ”, etc But, to be able to put such questions, the linguist must translate the particles and grammatical constructions of its language in the indigenous language. And to translate them, it must formulate analytical assumptions. These assumptions, we saw it, go beyond the data and are thus under-given by them. Quine concludes from it that the data on the linguistic behavior of the natives are not enough to determine that “gavagai”, as a term, results by “rabbit” and not in “not detached part of rabbit”, “matérialisation of the lapinitude”, “stage of the life of a rabbit”, etc the translation of the terms of the indigenous language is under-given by the data on the linguistic behavior of the natives. Various handbooks to translate the terms of the indigenous language can be elaborate in such a way that they all are compatible with these data and however intuitively incompatible between them.

Quine wanted to know, we saw it, up to which point the data on the linguistic behavior of the speakers of a language tend to determine the significance of the expressions of this language. We are now able to see how Quine answers this question. According to him, since the translation of the theoretical sentences and terms of the indigenous language is under-given by these data and that the ultimate data which help the linguist to translate the expressions of the indigenous language are the same ones as those which help all learning how from a language to determine the significance of the linguistic expressions, that is to say data on the linguistic behavior of the speakers, it follows that the significance of the theoretical sentences and the terms of a language is under-given by these data.

The data on the linguistic behavior of the speakers are too poor to determine only one correct interpretation of the theoretical sentences and the terms of a language. In other words, there are several possible interpretations of the theoretical sentences and the terms of a language in a basic language (i.e. the metalanguage in which we speak about the language object in question, which metalanguage can be an extension of the language object or a completely different language), or, in other words, several handbooks possible of interpretation of these linguistic expressions (that it is a handbook to interpret the expressions of a language in the same language, or in another language, in which case it acts of a handbook of translation), which is all correct, i.e. all compatible with the data on the linguistic behavior of the speakers, and, however, intuitively incompatible between them. Except for the observational sentences, any linguistic expression can receive various interpretations in various handbooks of interpretation of a language which all are correct, that is to say all compatible with the data on the linguistic behavior of the speakers, and, however, intuitively incompatible between them. Such is the thesis of the indetermination of the significance of Quine.

But, isn't the thesis of Quine then simply contradictory? It estimates that its thesis is a reduction with the absurdity of our intuitive design of the significance. According to this design, the significance would be independent of the languages and given in the spirit of the speakers beyond what can be implicit in their publicly accessible behavior. Quine names this design the “myth of the museum”. According to the myth of the museum, a significance is related to an expression of the language in the same way that a painting exposed in a museum is related to its label. And two expressions of the same language or different languages are equivalent if they are related to the same significance, as two labels which are related to the same painting. Thus, in the context of the translation, it is usually allowed that an expression has a significance and that another expression is its translation if it with the same significance. This design of the significance, we see it now, leads us directly to a contradiction.

Quine concludes from the thesis of the indetermination of the significance that to speak about correct interpretation or the significance about an expression about a language, except for the observational sentences, is deprived of direction except relative to a handbook of interpretation of this language. Except for the observational sentences, we have only relative notions of the significance.

Relativity of ontology

Quine formulated an ontological criterion of engagement for all Théorie, whatever the logical formulations: a theory is committed to pose as “entities” as soon as it poses criteria of identity and distinction and as soon as the formulation in Logique first order would quantify on these terms. To be is to be the value of a dependant variable and one should not admit entity without being able to give a criterion of identity. One sees all the stake here to have the most parsimonious language and least ambiguous possible, since our ontology will depend entirely on the forms of our language. The purpose of the canonical notation that Quine seeks to establish in the word and the thing is precisely to show which are the objects whose we must necessarily pose the existence and those of which we can do, knowing that the suppression of all the useless entities is the policy of Quine. This one indeed defends the use of the Rasoir of Ockham: " not to multiply the entities without nécessité". If the significance of the sentences and the reference of the terms are unspecified, that has as a consequence that the ontology of all Théorie is relative to a “handbook of translation”, with a total system of theories which are supported mutually.

This thesis of epistemological Holisme was defended by Pierre Duhem and is called “Thesis of Duhem-Quine”: it is not possible to refute a theory by a simple isolated fact because it would be possible logically to carry out other installations in the system. Here, Quine aims at the reductionnism of the Cercle of Vienna, for which any theoretical proposal can be analyzed like bearing on only one significant experiment of the subject. The holism rejects the bijective mapping between theoretical statement and feeling. A feeling can determine several theoretical statements and reciprocally.

The scientific theories are “empirically under-given”. The experiments and the observations are informative but are not sufficient to slice between the theories. It is necessary to add pragmatic criteria of simplicity, ontological parsimony, conservation of the former theories.

Quine began its work as a “nominalist ” (“hyper-extensionalist” within the meaning of Nelson Goodman) which refused the mathematical Whole in the name of this parsimony of ontology. But he was persuaded by his naturalism which sciences of nature had need for mathematical structures. He thus formulated an argument known as of “indispensability” against Nominalism: philosophy cannot impose on sciences a criterion and if sciences need units and cannot reduce them, then they should be accepted.

Random links:

Stephen Wolfram | Canton of Saint-Germain-Laval | School of maistrance | 1806 in France