为什么归纳逻辑需要所罗门诺夫先验?(英文)

二十世纪五十年代,卡尔纳普发展了归纳逻辑,他把概率看作一种证据对假设对"确证度";二十世纪六十年代,所罗门诺夫用通用归纳方法进行预测。为了增强归纳逻辑的归纳预测能力以及扩展所罗门诺夫通用归纳方法的表达力,本文整合二者。本文首先将所罗门诺夫先验概率的思想引入归纳逻辑中,在这个框架下,证明一阶逻辑版本的所罗门诺夫完全性定理,然后比较二者的优略。在卡尔纳普的归纳逻辑中,不管正面证据有多少,对像"所有乌鸦都是黑的"这种全称句的支持度最终都为零,而在用所罗门诺夫先验改造的归纳逻辑中,可以证明,在任何可计算的世界中,"所有乌鸦都是黑的"可以得到确证,只要在那些世界上真的所有乌鸦都是黑的。在所罗门诺夫的模型中,要证明完全性定理需要记录所有的过去信息,在修改后的归纳逻辑中,我们可以只关注某种具体的模式而忽略其它无关信息并证明类似的收敛定理。我们甚至可以不用记录所有的相关信息而采用随机抽样的方法建立合理的信念。     

Explication of Inductive Probability

Inductive probability is the logical concept of probability in ordinary language. It is vague but it can be explicated by defining a clear and precise concept that can serve some of the same purposes. This paper presents a general method for doing such an explication and then a particular explication due to Carnap. Common criticisms of Carnap’s inductive logic are examined; it is shown that most of them are spurious and the others are not fundamental.     

An observation on Carnapʼs Continuum and stochastic independencies

We characterize those identities and independencies which hold for all probability functions on a unary language satisfying the Principle of Atom Exchangeability. We then show that if this is strengthen to the requirement that Johnson?s Sufficientness Principle holds, thus giving Carnap?s Continuum of inductive methods for languages with at least two predicates, then new and somewhat inexplicable identities and independencies emerge, the latter even in the case of Carnap?s Continuum for the language with just a single predicate.     

Uniting model theory and the universalist tradition of logic: Carnap’s early axiomatics

We shift attention from the development of model theory for demarcated languages to the development of this theory for fragments of a language. Although it is often assumed that model theory for demarcated languages is not compatible with a universalist conception of logic, no one has denied that model theory for fragments of a language can be compatible with that conception. It thus seems unwarranted to ignore the universalist tradition in the search for the origins and development of model theory. This point is illustrated by Carnap’s early semantics and model theory, which he developed within a type theoretical framework and which stand out both for their universalistic treatment and for certain idiosyncratic technicalities by which the construction is supported. One special property is that individuals are context relative in Carnap’s system. This leads to a model theory in which the model domains are more flexible than has been suggested in the literature.     

Probability logic,logical probability,and inductive support

This paper seeks to defend the following conclusions: The program advanced by Carnap and other necessarians for probability logic has little to recommend it except for one important point. Credal probability judgments ought to be adapted to changes in evidence or states of full belief in a principled manner in conformity with the inquirer’s confirmational commitments—except when the inquirer has good reason to modify his or her confirmational commitment. Probability logic ought to spell out the constraints on rationally coherent confirmational commitments. In the case where credal judgments are numerically determinate confirmational commitments correspond to Carnap’s credibility functions mathematically represented by so—called confirmation functions. Serious investigation of the conditions under which confirmational commitments should be changed ought to be a prime target for critical reflection. The necessarians were mistaken in thinking that confirmational commitments are immune to legitimate modification altogether. But their personalist or subjectivist critics went too far in suggesting that we might dispense with confirmational commitments. There is room for serious reflection on conditions under which changes in confirmational commitments may be brought under critical control. Undertaking such reflection need not become embroiled in the anti inductivism that has characterized the work of Popper, Carnap and Jeffrey and narrowed the focus of students of logical and methodological issues pertaining to inquiry.     

Evidential diversity and premise probability in young children's inductive judgment

A familiar adage in the philosophy of science is that general hypotheses are better supported by varied evidence than by uniform evidence. Several studies suggest that young children do not respect this principle, and thus suffer from a defect in their inductive methodology. We argue that the diversity principle does not have the normative status that psychologists attribute to it, and should be replaced by a simple rule of probability. We then report experiments designed to detect conformity to the latter rule in children's inductive judgment. The results suggest that young children in both the United States and Taiwan are sensitive to the constraints imposed by the rule on judgments of probability and evidential strength. We conclude with a suggested reinterpretation of the thesis that children's inductive methodology qualifies them as “little scientists.”     

Aristotle on Universal Quantification: A Study from the Point of View of Game Semantics

In this paper we provide an interpretation of Aristotle's rule for the universal quantifier in Topics Θ 157^a34–37 and 160^b1–6 in terms of Paul Lorenzen's dialogical logic. This is meant as a contribution to the rehabilitation of the role of dialectic within the Organon. After a review of earlier views of Aristotle on quantification, we argue that this rule is related to the dictum de omni in Prior Analytics A 24^b28–29. This would be an indication of the dictum’s origin in the context of dialectical games. One consequence of our approach is a novel explanation of the doctrine of the existential import of the quantifiers in dialectical terms. After a brief survey of Lorenzen's dialogical logic, we offer a set of rules for dialectical games based on previous work by Castelnérac and Marion, to which we add here the rule for the universal quantifier, as interpreted in terms of its counterpart in dialogical logic. We then give textual evidence of the use of that rule in Plato's dialogues, thus showing that Aristotle only made explicit a rule already implicit in practice, while providing a new interpretation of ‘epagogic’ arguments. Finally, we show how a proper understanding of that rule involves further rules concerning counterexamples and delaying tactics, stressing again the parallels with dialogical logic.     

Constraints and Preferences in Inductive Learning: An Experimental Study of Human and Machine Performance

The paper examines constraints and preferences employed by people in learning decision rules from preclassified examples. Results from four experiments with human subjects were analyzed and compared with artificial intelligence (AI) inductive learning programs. The results showed the people's rule inductions tended to emphasize category validity (probability of some property, given a category) more than cue validity (probability that an entity is a member of a category given that it has some property) to a greater extent than did the AI programs. Although the relative proportions of different rule types (e.g., conjunctive vs. disjunctive) changed across experiments, a single process model provided a good account of the data from each study. These observations are used to argue for describing constraints in terms of processes embodied in models rather than in terms of products or outputs. Thus AI induction programs become candidate psychological process models and results from inductive learning experiments can suggest new algorithms. More generally, the results show that human inductive generalizations tend toward greater specificity than would be expected if conceptual simplicity were the key constraint on inductions. This bias toward specificity may be due to the fact that this criterion both maximizes inferences that may be drawn from category membership and protects rule induction systems from developing over-generalizations.     

Multitude, tolerance and language-transcendence

Rudolf Carnap??s 1930s philosophy of logic, including his adherence to the principle of tolerance, is discussed. What theses did Carnap commit himself to, exactly? I argue that while Carnap did commit himself to a certain multitude thesis??there are different logics of different languages, and the choice between these languages is merely a matter of expediency??there is no evidence that he rejected a language-transcendent notion of fact, contrary to what Warren Goldfarb and Thomas Ricketts have prominently argued. (In fact, it is obscure just what Goldfarb and Ricketts claim about Carnap.) Toward the end I critically discuss Michael Friedman??s suggestion that Carnap believed in a relative a priori.     

The Early Formation of Modal Logic and its Significance: A Historical Note on Quine,Carnap, and a Bit of Church

The aim of the paper is to show that W. V. O. Quine's animadversions against modal logic did not get the same attention that is considered to be the case nowadays. The community of logicians focused solely on the technical aspects of C. I. Lewis’ systems and did not take Quine's arguments and remarks seriously—or at least seriously enough to respond. In order to assess Quine's place in the history, however, his relation to Carnap is considered since their notorious break was about the status of extensionality and modal logic (and analyticity was much more of a second issue). Since much of the works about the history of analytic philosophy is centered on the relationship of Quine and Carnap, their break about modality deserves much more attention—it also sheds some light on why should anyone wonder about Quine's early arguments against modal logic. The paper ends with some further considerations regarding the early formation of modal logic and hitherto unconsidered problematic issues.     

《逻辑哲学论》论概率和归纳

《逻辑哲学论》虽然注家甚多,此书对于概率和归纳问题的零星讨论却一直没有受到后人的重视,而本文则试图对这些讨论进行尝试性的重构。本文的写作目的是双重的。首先,本人试图揭示早期维特根斯坦的概率理论与他的整个逻辑原子论架构之间的关系,说明他是如何通过“真值根据”这个概念而将日常语言中的概率表述全部还原为真值函项理论的。在此基础上,笔者将说明为何他的这种概率理论缺乏实际的应用价值。其次,本文将讨论维氏将归纳问题和概率问题相互分离的“分离策略”,并尝试重构出他提出该策略的隐蔽动机。在此基础上．笔者将站在认知科学的立场上为这个策略本身提供一些辩护和批评。笔者认为,早期维氏虽然正确地看到了“思维经济原则”对于归纳推理的重要指导意义,并看到了该原则本身是无法被还原为逻辑真理的,却并没有意识到该原则本身可能是一切具有一定智能水准的问题解决系统所必须依赖的问题解决策略。令人遗憾的是,他对“心理学”和“逻辑”所作的简单的二分法阻止了他更为深入地对智能系统的普遍问题解决逻辑作出探究,尽管这种研究的确是可以相容于他的前述“分离策略”的。总而言之,本文认为早期维氏的概率理论的科学价值相当有限;与之相比较,他的归纳观却可能已指出了如何揭示智能系统的普遍认知机制的正确道路,尽管他在《逻辑哲学论》阶段还缺乏对于这条道路的全面自觉。     

Hempel’s logic of confirmation

This paper presents a new analysis of C.G. Hempel’s conditions of adequacy for any relation of confirmation [Hempel C. G. (1945). Aspects of scientific explanation and other essays in the philosophy of science. New York: The Free Press, pp. 3–51.], differing from the one Carnap gave in §87 of his [1962. Logical foundations of probability (2nd ed.). Chicago: University of Chicago Press.]. Hempel, it is argued, felt the need for two concepts of confirmation: one aiming at true hypotheses and another aiming at informative hypotheses. However, he also realized that these two concepts are conflicting, and he gave up the concept of confirmation aiming at informative hypotheses. I then show that one can have Hempel’s cake and eat it too. There is a logic that takes into account both of these two conflicting aspects. According to this logic, a sentence H is an acceptable hypothesis for evidence E if and only if H is both sufficiently plausible given E and sufficiently informative about E. Finally, the logic sheds new light on Carnap’s analysis.     

Revising Carnap’s Semantic Conception of Modality

I provide a tableau system and completeness proof for a revised version of Carnap??s semantics for quantified modal logic. For Carnap, a sentence is possible if it is true in some first order model. However, in a similar fashion to second order logic, no sound and complete proof theory can be provided for this semantics. This factor contributed to the ultimate disappearance of Carnapian modal logic from contemporary philosophical discussion. The proof theory I discuss comes close to Carnap??s semantic vision and provides an interesting counterpoint to mainstream approaches to modal logic. Despite its historical origins, my intention is to demonstrate that this approach to modal logic is worthy of contemporary attention and that current debate is the poorer for its absence.     

Wittgenstein and the Phenomenological Movement: Reply to Monk

Abstract

Monk’s ‘The Temptations of Phenomenology’ examines what the term ‘Phänomenologie’ meant for Wittgenstein. Contesting various other scholars, Monk claims that Wittgenstein’s relation to ‘Phänomenologie’ began and ended during 1929. Monk only partially touches on the question of Wittgenstein’s relation to the phenomenological movement during this time. Though Monk does not mention this, 1929 was also the year in which Ryle and Carnap turned their critical attention toward Heidegger. Wittgenstein also expressed his sympathy for Heidegger in 1929. Furthermore, though in 1929 Wittgenstein agrees with the early Husserl on relating logic and science to phenomenology, it is not clear that they mean the same thing by either logic or phenomenology, or that they agree on what the relation between the two should be.     

Predictive probability and analogy by similarity in inductive logic

The-continuum of inductive methods was derived from an assumption, called-condition, which says that the probability of finding an individual having propertyx _j depends only on the number of observed individuals having propertyx _j and on the total number of observed individuals. So, according to that assumption, all individuals with properties which are different fromx _j have equal weight with respect to that probability and, in particular, it does not matter whether any individual was observed having some propertysimilar tox _j (the most complete proof of this result is presented in Carnap, 1980).The problem thus remained open to find some general condition, weaker than the-condition, which would allow for thederivation of probability functions which might be sensitive to similarity. Carnap himself suggested a weakening of the-condition which might allow for similarity sensitive probability functions (Carnap, 1980, p. 45) but he did not find the family of probability functions derivable from that principle. The aim of this paper is to present the family of probability functions derivable from Carnap's suggestion and to show how it is derived.In Section 1 the general problem of analogy by similarity in inductive logic is presented, Section 2 outlines the notation and the conceptual background involved in the proof, Section 3 gives the proof, Section 4 discusses Carnap's principle and the result, Section 5 is a brief review of the solutions which have previously been proposed.     

The Logic of “Most” and “Mostly”

The paper suggests a modal predicate logic that deals with classical quantification and modalities as well as intermediate operators, like “most” and “mostly”. Following up the theory of generalized quantifiers, we will understand them as two-placed operators and call them determiners. Quantifiers as well as modal operators will be constructed from them. Besides the classical deduction, we discuss a weaker probabilistic inference “therefore, probably” defined by symmetrical probability measures in Carnap’s style. The given probabilistic inference relates intermediate quantification to singular statements: “Most S are P” does not logically entail that a particular individual S is also P, but it follows that this is probably the case, where the probability is not ascribed to the propositions but to the inference. We show how this system deals with single case expectations while predictions of statistical statements remain generally problematic.     

Where do hypotheses come from?

Why are human inferences sometimes remarkably close to the Bayesian ideal and other times systematically biased? In particular, why do humans make near-rational inferences in some natural domains where the candidate hypotheses are explicitly available, whereas tasks in similar domains requiring the self-generation of hypotheses produce systematic deviations from rational inference. We propose that these deviations arise from algorithmic processes approximating Bayes’ rule. Specifically in our account, hypotheses are generated stochastically from a sampling process, such that the sampled hypotheses form a Monte Carlo approximation of the posterior. While this approximation will converge to the true posterior in the limit of infinite samples, we take a small number of samples as we expect that the number of samples humans take is limited. We show that this model recreates several well-documented experimental findings such as anchoring and adjustment, subadditivity, superadditivity, the crowd within as well as the self-generation effect, the weak evidence, and the dud alternative effects. We confirm the model’s prediction that superadditivity and subadditivity can be induced within the same paradigm by manipulating the unpacking and typicality of hypotheses. We also partially confirm our model’s prediction about the effect of time pressure and cognitive load on these effects.