• Korean Journal of Logic
• /
• v.7 no.2
• /
• pp.121-146
• /
• 2004

본 논문은 베이즈주의가 확률론을 이용해서 제거적 귀납을 정교하게 발전시키고 있다고 주장한다. 이를 위해 본 논문은 두 가지 작업을 진행한다. 하나는 제거적 귀납이 무엇인가 하는것이고 다른 하나는 제거적 귀납이 베이즈주의에 기여하는 바가 무엇인가 하는 것이다. 먼저 본 논문은 제거적 귀납이 참인 가설을 포함하는 가능한 가설들의 총체로부터 경쟁가설들을 연역적 또는 귀납적으로 제거하고 남는 가설을 선택하는 추론형식임을 밝히고, 이 때 베이즈주의는 제거적 귀납을 정교하게 발전시킨 모습이기 때문에 제거적 귀납으로부터 기술적으로 도움 받을 측면은 없다고 주장한다. 그 대신 본 논문은 베이즈주의가 과학방법론으로 발전되는 데에서 직면하는 여러 가지 문제점을 해결하는 방법에 대해 제거적 귀납으로부터 조언을 얻을 수 있다고 주장한다. 이와 같은 논의를 통해 본 논문은 베이즈주의와 제거적 귀납주의의 결합은 유용한 과학방법론을 만들 수 있을 것으로 전망한다.

• Korean Journal of Logic
• /
• v.17 no.3
• /
• pp.399-424
• /
• 2014

The main objective of this paper is to examine critically White's claim that there is a conflict between Perceptual Dogmatism and Bayesian Theory of Confirmation. For this purpose, this paper is structured as follows: In Section 2, I will introduce White's argument. Section 3 is dedicated to explaining some elements of Bayesian Theory of Confirmation. In particular, I will provide an explanation of confirmation measures and Bayesian Favoring. Using these two conceptual apparatuses, it will be shown that, contrary to what White has thought, there is a way of supporting Perceptual Dogmatism by means of Bayesian Theory of Confirmation - in particular, Bayesian Theory of Favoring.

• Korean Journal of Logic
• /
• v.18 no.2
• /
• pp.243-264
• /
• 2015

Roger White raised an objection, one based on Bayesianism, to the dogmatist view of perceptual justification. In his paper, "Perceptual Dogmatism and Bayesian Favoring", Ilho Park tries to show, contra Roger White, that there is no real conflict between Perceptual dogmatism and Bayesianian theory of confirmation. For this purpose, Park brings in the notions of the degree of confirmation and the favoring relation and argues that Bayesian theory, when properly understood, can yield results that are quite favorable to dogmatism. I don't think, however, that the devices that he employes actually deliver what he promises. The conflict is yet to be resolved. Probably, Bayesian theorists may be better off if they, instead of trying to resolve the conflict, consider the option of simply rejecting dogmatism.

• Korean Journal of Logic
• /
• v.6 no.2
• /
• pp.135-158
• /
• 2003

베이즈주의는 믿음의 정도라는 확률의 의미해석과 수학적 확률론에 의한 계산체계를 기초로 하여 가설과 증거간의 입증(confirmation) 관계를 명료하게 분석한다. 베이즈주의는 증거 E가 가설 H를 입증한다는 것을 PR(HIE&K)-PR(HIK)>0으로 정의한다. 그러나 이러한 분석이 과연 과학자들의 입증개념을 올바로 반영하고 있는가 하는 비판이 오래된 증거(old evidence)의 문제로부터 제기되었다 오래된 증거는 이미 알려진 정보이기 때문에 완전한 확률 값 1을 부여받는다. 이 때 오래된 증거가 가설을 입증할 수 있는가 하는 질문은 베이즈주의자와 실제 과학자 사이에 서로 다른 답변을 도출한다. 먼저 베이즈주의에 따르면 오래된 증거가 가설을 입증할 수 없다. 그것은 PR(EIK)=1일 때 PR(HIK)=PR(HIE&K)의 결과가 도출되기 때문이다. 하지만 과학사의 여러 예들로부터 제시되는 실제 과학자들의 입증개념에 따르면 오래된 증거가 가설을 입증하고 있다. 필자는 이와 같은 입증개념의 이질성 문제가 다만 어떤 증거가 입증 가능한 것인지를 구분해야 하는 질적인(qualitative) 문제일 뿐만 아니라 증거가 가설을 어느 정도 입증하는지 하는 입증도를 정확하게 측정해야 하는 양적인(quantitative) 문제라는 점을 밝힌다. 특히 필자는 양적인 문제를 해결하면 질적인 문제가 자연히 해결된다는 점을 밝히고, 반 프라센이나 가버가 제안한 전략이 모두 질적인 문제만을 다루기 때문에 부분적인 해결책에 지나지 않는다는 점을 밝힘으로써 오래된 증거의 문제의 본질은 양적인 문제에 있다는 점을 주장한다.

• Korean Journal of Logic
• /
• v.16 no.2
• /
• pp.155-187
• /
• 2013

The main objective of this paper is to vindicate the Bayesian belief updating rule, i.e. conditionalization. For this purpose, I introduce first what I call Irrelevance Principle, and show that this principle is equivalent to conditionalization. In turn, the principle is vindicated by means of Bayesian confirmation theory. That is, I suggest some theses that Bayesian confirmation theorists should accept, and prove that if Irrelevance Principle is violated, the theses cannot holds.

• Korean Journal of Logic
• /
• v.18 no.3
• /
• pp.389-436
• /
• 2015

Epistemic dogmatism claims that if it seems P to you then you have immediate justification to believe P. The view has been faced with a problem that it is incompatible with Bayesianism, especially raised by Roger White(2006). James Pryor(2013), defending epistemic dogmatism, has given a reply for the problem. In this paper, first, I show some problems on Pryor's reply. Then, I present a new kind of suggestion to deal with the problem, which avoids problems Pryor's reply has. Finally, I suggest a different diagnosis on the problem.

• Korean Journal of Logic
• /
• v.11 no.2
• /
• pp.93-125
• /
• 2008

The radical probabilitists deny that propositions represent experience. However, since the impact of experience should be propagated through our belief system and be communicated with other agents, they should find some alternative protocols which can represent the impact of experience. The useful protocol which the radical probabilistists suggest is the Bayes factors. It is because Bayes factors factor out the impact of the prior probabilities and satisfy the requirement of commutativity. My main challenge to the radical probabilitists is that there is another useful protocol, q(E|$N_p$) which also factors out the impact of the prior probabilities and satisfies the requirement of commutativity. Moreover I claim that q(E|$N_p$) has a pragmatic virtue which the Bayes factors have not.

• Korean Journal of Logic
• /
• v.8 no.2
• /
• pp.33-58
• /
• 2005

The old evidence problem raises a profound problem to Bayesian theory of confirmation that evidence known prior to a hypothesis explaining it cannot give any empirical support to the hypothesis. The old evidence problem has resisted to a lot of trials to solve it. The purpose of the paper is to solve the old evidence problem by showing that the problem originated from a serious misunderstanding about the Bayesian concept of confirmation. First, I shall make a brief analysis of the problem, and examine critically two typical Bayesian strategies to solve it. Second, I shah point out a misunderstanding commonly found among Bayesian discussions about the old evidence problem, the ignorance of the asymmetry of confirmation in the context of explanation and prediction. Lastly, 1 shall suggest two different concepts of confirmations by using the asymmetry and argue that the concept of confirmation presupposed in the old evidence problem is not a genuine Bayesian concept of confirmation.

• Korean Journal of Logic
• /
• v.20 no.3
• /
• pp.367-390
• /
• 2017

Much of literature on the paradoxes of confirmation has been focused on the problems raised by the fact that a nonblack nonraven confirms the hypothesis that every raven is black. In this paper I would like to emphasize that more interesting problems are still waiting to be explained, if we notice that a black nonraven confirms the raven hypothesis as well. For this I examine what Hempel exactly means by the paradoxes of confirmation, and show that the previous discussions on the paradoxes were at most partial solutions. Then I argue that Hempel presupposes the so-called 'converse consequence condition' regarding confirmational evidence. Finally I discuss what impact is made on the Bayesian solution to the paradoxes, if we accept a more faithful interpretation to Hempel.

• Korean Journal of Logic
• /
• v.18 no.2
• /
• pp.167-195
• /
• 2015

This paper aims to respond to Weisberg's claim that the standard Bayesian epistemology cannot model an agent's belief updating that is triggered by some undermining evidence. Our epistemological intuition seems to require that the undermining evidence decreases some particular relevant credences. According to Weisberg, however, such a belief change cannot result from the standard Bayesian belief updating rules-i.e., (Jeffrey) Conditionalization. This is because probabilistic independence between some propositions is preserved under (Jeffrey) Conditionalization on the relevant evidence. Yet I will show in this paper that this conclusion is somewhat hasty. In particular, I will show that there is another version of Conditionalization and that when one updates her credences by means of such a version, the belief updating originated in undermining evidence can be well modeled in the Bayesian framework. Some authors often call the version Higher Order Conditionalization.