earticle

영어

We can witness the surge of interest in counterfactual reasoning for the last half century.
Not to mention logicians, metaphysicians, and philosophers of science in the analytic tradition, more and more researchers in such a varied fields as linguistics, cognitive psychology, artificial intelligence, and economics are led to this exciting and even mysterious subject area. This phenomenon is of course quite understandable in view of the ubiquity of counterfactual reasoning in any human intellectual enterprise. In the irresistible uprise of the interdisciplinary cognitive science in recent years, counterfactual reasoning must be destined to be located at the crossroad of all these scientific disciplines. Now, as I pointed out elsewhere, Baduk Studies seem to share all the essential traits of cognitive science.1) To say the least, I think it mandatory to exploit all our theories and techniques in cognitive science in order to understand Baduk, which is the most sophisticated intellectual game ever devised by humans. It is not a mere coincidence that counterfactual reasoning looms large in both playing and studying Baduk In this article, I would like to examine in somewhat cursory fashion what a cognitive scientist, who has been tormented by recurrent problems of analysing counterfactuals, might learn something from counterfactual reasoning in Baduk. At the same time, it might enhance Baduk players' self-consciousness of what exactly they are doing in their constant performance of counterfactual reasonings.
In this preliminary survey, I will concentrate on the problem of evaluation of counterfactual reasoning. For, whether it be posed as a problem of establishing the truth condition of counterfactual conditionals or as that of searching for conditions of rational support, what makes counterfactuals challenging must be found in the unexpected difficulties involved in the evaluation of them. Only when we settle the issue of the general methods, criteria, algorithms, or what not, of evaluating counterfactuals, fruitful applications and extensions of our interests to other closely related issues such as causal reasoning, dispositional properties, law statements, and scientific explanation might be adequately discussed. Nor will there be any attempt to classify counterfactual conditionals into sub-categories such as would conditionals or might conditionals.
The huge literature devoted to diverse analyses of counterfacuals is simply overwhelming.
Fortunately, in Hansson(l995) we find a nice birdeye view of the three major stages in the study of counterfactuals: (1) derivability approach, (2) world-selection approach, and (3) the belief revision approach.2) Furthermore, he was perceptive enough not only to trace their both topical and historical origin back to the so-called the Ramsey test but also to uncover the most interesting relation between these approaches, i.e., their interchangeability from a formal point of view. Equally significant for my purpose here must be the fact that he pinned down some thorny problems shared by all of them: (1) non-deductiveness, (2) circularity, and (3) shiftability. In this article I will use Hansson's analysis as a foil to fathom how and in what respects we might highlight the most salient features of the counterfactual reasoning in Baduk In section 1, I will briefly report Hansson's discussion of the basic ideas of the three major theories of counterfacuals and their relations centering around the Rarnsey test. Also a brief summary of the alleged problems of those theories of counterfactuals will be presented. In section 2, after showing how to represent counterfactual reasoning in Baduk in terms of each of these theories, we will be able to examine to what extent all those well known problems of counterfactuals could reappear in the context of Baduk Finally, in section 3, 1 will make a case for my belief that we can learn a great deal from Baduk as to how to handle counterfactuals even if there has been no self-conscious scholarly exposition of counterfactual reasoning in Baduk. I will try to hint at the possibility of securing the best Ramsey test for counterfactuals in Baduk.

한국어

반사실적 조건문 （또는 가정법적 조건문） 을 논리적으로 분석하는 문제는 현대 분석철학 자들의 관심의 표적 이 되 어 왔다． 바둑을 둘 때 매 순간마다 우리 는 여러 가지 갈래 의 수읽기를 시도하면서 고려하고 있는 착점으로 전개될 수순을 마음속에 그리는데， 그 때 우리는 무수한 반사실적 조건문들의 친 리 값이 나 지지 가능성을 평가하고 있는 것이라 볼 수 있다． Hallsson은 보기 드물게 지난 세기에 이루어진 연구를 솜씨 있게 요 약하여 우선 그 동안의 반 사실적 조건문 연구에서 크게 세 가지 접근방법을 가려내었다：(1) 도출가능성 접근법， (2) 세계－ 선택 접근법， (3) 믿음 수정 접근법． Hallsson의 장점 중 하나는 보기 드물게 그가 이 세 가지 접근법 사이의 관계를 밝히려는 노력을 통해 삼자 사이의 형식적 호환성을 증명했다는 데 있다． 또한Harlsson은 반사실적 조건문을 평가하는 데 있어서 이 세 가지 접근법이 공통적으로 마주친 난제 세 가지를 적시 한다： (1) 비-연역성， (2) 순환성， (3) 이동가능성． 우리는 이제 바둑두는 사람이 Hansson이 열거한 반사실적 조건문에 대한 세 가지 접근법을 모두 쉽게 이해할 수 있을 뿐만 아니라 실제로 바둑을 두면서 그 접근법들을 일상적으로 사용해 왔다는 점을 보인다．과학에서의 혁명적 진보나 바둑에서의 신포석 혁명에 우리가 열광한다는 것은 다시 말해서 과학과 바둑에서의 반사실적 조건문에 의거한 사유를 우리가 상상할 수 없을 만큼 중요하게 여긴다는 점을 여실히 보여준다． 우리는 과학자들의 논문 심사나 실험, 그리고 바둑두는 이들의 복기 （復 棋）등의 아름다운 제도가 보여주는 절차적 특성이 진정한 Ramsey test를 이루는 방향을 암시해준다고 믿는다.