Theory of knowledge

The theory of knowledge is the part of the Philosophie which studies nature, the origins, the contents, the means and the limits of the Connaissance, in particular of human knowledge. Most of work which concerns this discipline are devoted to the analysis of knowledge, i.e. with the determination of its requirements and sufficient. It is more precisely a question of establishing which relations maintains the knowledge with the Croyance and the Vérité, and which procedures of justification make it possible to distinguish a simple true belief (which can the being by accident) from a true knowledge.

Most of this article carries on the analytical theory of the knowledge, disciplines philosophical which essentially developed in the anglophone world.

Diversity of the theories of knowledge

The theories of knowledge are as numerous as the philosophers who are leaning on the question. One can start by distinguishing them according to the various designs from the origin from knowledge, and the nature of knowledge.

Nature of knowledge

August 1st

Origin of knowledge

The philosopher empirist (cf Locke, Hume) place the experiment sensitive at the origin of the acquisition of knowledge. For its part, the rationalist (cf Descartes, Popper) makes it rest on the exercise of the Raison.

It is indeed necessary to distinguish various theories from knowledge:

idealistic empirists
rationalist
realistic
constructivists

Classical theory of knowledge

Justified true belief

The classical theory rests on the idea that knowledge is a justified true belief and , and not only one true belief.

This additional clause makes it possible to exclude from the field of knowledge the cases in which our belief is true, but where we are not able to explain why it is true. An individual can thus believe that the ground turns around the sun (proposal p) by simple yes saying, without being able to explain it. The proposal p is true, but the individual in question does not know that the ground turns around the sun. That remains a belief.

The justification of the belief is thus the crucial factor of this traditional analysis of knowledge, and of many contemporary theories seek to precisely determine of them nature and the methods; the Théorie of the justification is one of the principal branches of the theory of knowledge.

The term of “knowledge” indicated a long time, in philosophy, of the beliefs whose truth is justified in manner some . Any belief presenting a less degree of justification constitutes on this account a “probable opinion” (or knowledge by provision). This point of view still prevails in the work of Bertrand Russell (in particular in the Problèmes of Philosophy , 1912). During decades which followed, the idea according to which the degree of justification of the beliefs must be evaluated in terms of Certitude lost in influence.

The problem of Gettier and contemporary analyzes of knowledge

In its famous article of 1963 entitled “Is Justified True Belief Knowledge? ”, Edmund Gettier affirms that there exist situations in which a belief can be at the same time true and justified, and does not constitute therefore a knowledge. More exactly, the thesis of Gettier consists in saying that the traditional analysis states the necessary conditions of knowledge, but that these conditions are not sufficient . For better appreciating the strategy of Gettier and the arguments which its contradictors will oppose to him, it is useful to leave the traditional analysis. According to this one:

S knows that p if and only if

p is true;
S believes that p ; and
the belief of S in p is justified.

The attack of Gettier rests on two premises, consistent with the traditional analysis. First premise: it is possible that a justified belief is false. (In other words, it is possible to have good reasons to believe in the truth of p and that p is false.) Second premise: if S is justified to believe that p and that p implies Q , and if S deduced Q from p and accepts Q like a result, then S is justified to believe that Q . Starting from these premises, Gettier builds two examples which express the insufficiency of the traditional definition. It will be enough to evoke the first.

Smith and Jones stand as a candidates for the same station. Smith has excellent reasons to believe that the candidature of Jones will be retained, and it knows in addition that Jones has ten coins in his pocket. That is to say p : “Jones will be engaged and it has ten parts in its pocket”. It is seen that two of the three traditional conditions are already filled: Smith believes that p , and it is justified to believe that p . Let us consider now the proposal Q : “That which will be engaged has ten parts in its pocket”. It is clear that p implies Q ; if it is supposed that Smith deduced Q from p , then (by the second premise) Smith believes that Q and this belief is justified.

Now, it is that, contrary to the prediction of Smith, it is Smith, and not Jones, who obtains the station. Although it is justified, p is thus false (case admitted by the first premise). But it is that Smith, without his knowledge , has ten coins in his pocket; Q is thus true. On the whole, Smith believes that Q , it is justified to believe that Q (by inference starting from p ), and Q , without the knowledge of Smith, is true. We are thus in a case of justified true belief which is not therefore a case of knowledge: Smith does not know not that Q is true.

Answers to Gettier

Since the publication of the article of Gettier (in the review Analysis , vol. 23,1963, pp. 121-123) a very great number of authors tried to arrive to an analysis of the knowledge which can exclude a priori from such examples. The two strategies most usually employed consist: a) to modify the clause of justification retained by Gettier, considered to be too weak; b) or to preserve the clause of traditional justification but by adding to it another, supposed to guarantee the whole of the analysis against the examples of the Gettier type. The solution suggested by Robert Nozick concerns the first strategy: the clause of traditional justification is replaced by two conditional fixing the relation between the belief of S and the truth of the contents of its belief. According to Nozick, S knows that p if and only if:

p is true
S believes that p
if p is false, S will not believe only p .
if p is true, S will believe that p .

Simon Blackburn criticized this formulation, asserting which should not admit to us with the row of knowledge of the beliefs which, although they “track the truth” (in accordance with the requirements of Nozick), are not supported by suitable reasons. Indeed, it seems possible to imagine scenarios in which the belief of S is narrowly correlated in the truth or the falseness of p , and where S is completely unable to give an account of its belief, i.e. to advance elements of justification. In other words, the true belief will be truly justified only if S knows why it is true. We will further find this idea according to which a belief is justified only if the subject has an epistemic access to the base of justification: it is the fundamental thesis of the theories internalists of the justification. The theories externalists of the justification (whose Nozick offers a first example here to us) affirm on the contrary that the base of justification of our beliefs is not necessarily accessible for us; it may be that our beliefs are not justified by other beliefs, but by reliable mechanisms connecting them to the modifications of our environment.

In another answer, Richard Kirkham explains why impossibility of arriving to an analysis of the knowledge which is perfectly with the shelter of the counterexamples of Gettier is due to the fact that only the definition of knowledge in force since Antiquity until Russell is truly satisfactory: to be a knowledge, a belief should not only be true and justified, but its base of justification must still make necessary its truth . This constraint constitutes an extremely demanding criterion (if we retain it, the majority of our “empirical knowledge” are not more), but Kirkham notices that very high standards of knowledge do not prevent from integrating the whole of our “weak” knowledge into the category of the “reasonable beliefs”.

If one chooses the second strategy (to add a fourth clause to the three traditional ones), one of the possibilities consists in requiring that the justification of the belief be “unconquered” ( undefeated ). This new theory, due in particular to Keith Lehrer and Thomas D. Paxson Jr., is not worth for any knowledge in general, but only for those which these authors call “nonbasic” ( nonbasic ). The distinction of basic” and “nonbasic” knowledge “aims at the contents of the base of justification: if S knows that p and that the base of justification of its belief does not comprise other beliefs, then its knowledge will be known as “basic”. This description corresponds in particular to noninférentielles perceptive knowledge of the type: “I perceive a pain in my left thigh”. It is seen well that this knowledge does not rest on another belief, but drift only of the contents of my experiment. According to Lehhrer and Paxson, the traditional definition of knowledge gives sufficiently an account of knowledge of this type. On the other hand, if S then knows that p is true and that another proposal Q enters the base of justification of this belief, its knowledge will be known as “nonbasic”: it is here about knowledge known as “inférentielles”, i.e. knowledge which depends logically on the truth of another beliefs, more or less many. In the case of nonbasic knowledge, a fourth clause is necessary so that the analysis is with the shelter of the counterexamples: the true and justified belief must moreover be “unconquered” ( undefeated ). In other words, if S knows that p and that the base of justification of p comprises Q , there should be no other proposal R which is true and which invalidates Q .

Random links: