|Edited by Matti Sintonen
Amsterdam-Atlanta, GA: Rodopi, 1997
ISBN 90-420-0080-5 (b)
Jaakko Hintikka was the first to present a detailed argument for the claim that Aristotle subscribed to the so-called Principle of Plenitude according to which all possibilities are eventually actualized. This paper argues that Aristotle’s “actualism,” the view that only actualities are real, determined his peculiar type of logical determinism as well as his doctrine of plenitude. The paper also argues, in part contrary to what Hintikka seems to have thought, that Aristotle’s definition of kinesis does not provide an escape from the Principle of Plenitude or from his form of logical determinism. It is also suggested that, as a result, potentialities in Aristotle’s scheme cannot have an explanatory role.
This paper argues that both Aristotle’s notion of rhetoric and dialectic, as well as his epistemological and methodological views have a greater contribution to offer to modern research in the sciences of man than recognized so far. Having presented some basic concepts of dialectic and rhetoric, the author shows with the help of a paradigmatic example the relevance of the Aristotelian model of rhetorical speech to traditional narrative historical research. She also elaborates in which ways Kuhn’s conception of interparadigmatic choice corresponds to the Aristotelian view of decision-making in rhetorical situations. In the last section, she claims that the Aristotelian concept of reputable opinions (endoxa) and the method of saving the phenomena provide fundamental epistemological solutions to problems raised by the recent criticism of positivistic philosophy of science.
Around the turn of the century, German-speaking philosophers displayed a lively interest in the nature of questions. Theorizing concentrated on the relation between questions and judgments, the distinction between propositional and why-questions, various further taxonomies, the problem of answerhood, and attempts to describe knowledge-acquisition as an interrogative process. Interest in question theory ran high because logical inquiry was non-formal and close to psychology, and because both applied and theoretical (experimental) psychology had independent reasons for focusing on the study of questions. Kant’s use of question-theoretical terms was an important resource for neo- Kantian philosophers.
This is a paper on the philosophical relations between F. P. Ramsey and Ludwig Wittgenstein.. The key question of the paper is to what extent Frank Ramsey influenced the thinking of the later Wittgenstein. It is well known that Wittgenstein strongly influenced Frank Ramsey’s philosophical thinking. But, what is less well known is that Ramsey came to change his view of philosophy and in doing so became a major force in Wittgenstein’s upheaval of the Tractarian view. From 1929 and on, one can clearly spot a remolded Ramsey in the work of Wittgenstein. Thus, it is argued that Wittgenstein probably was more influenced by Ramsey than Ramsey was influenced by Wittgenstein. The arguments are presented in three parts: (I) a few biographical glimpses are given; (2) Ramsey’s British pragmatism is outlined and discussed; and (3) it is shown where this British pragmatism can be spotted in Wittgenstein’s work after 1930.
Rudolf Carnap initiated a research program in quantitative confirmation theory, also called inductive probability theory, or inductive logic, by designing a continuum of systems for individual hypotheses. It has guided the search for optimum systems and for systems that take analogy into account.
Carnapian systems, however, assign zero probability to universal hypotheses. Jaakko Hintikka was the first to reconsider their confirmation. His way of using Carnap’s continuum for this purpose has set the stage for a whole spectrum of inductive systems of this type.
The Carnap-Hintikka program clearly has much internal dynamics; as will be indicated, its results can also be applied in several directions.
Default reasoning is that species of nonmonotonic reasoning that is ampliative. The other type of nonmonotonic reasoning is belief contravening as in reasoning from suppositions. It is commonly held that these two kinds of nonmonotonic reasoning share the same formal properties. According to the decision theoretic approach to ampliative reasoning shared by Hintikka, Hilpinen, Niiniluoto, Pietarinen and myself, formal similarity holds only when ampliative inferences are drawn with maximum boldness and when iterating the use of rules for inductive inference can yield no new information. Such “bookkeeping” with maximum boldness is hard to defend raising doubts about an important assumption made by the majority of students of nonmonotonic reasoning.
Jaakko Hintikka has recently claimed that the traditional problem of inductive generalization occupies only a modest place in the interrogative logic of scientific inquiry. It is argued in this paper that the Atomistic Postulate, i.e., the restriction of Nature’s answers to atomic propositions and their negations, can be avoided only under very strong regularity conditions. In typical curve-fitting problems the desired function is not given by Nature but discovered and supplied by the inquirer. On the other hand, Hintikka’s own system of inductive logic is not tied to the Atomistic Postulate, since it can be (and has been) applied to situations where the evidence contains a universal theory. The second part of the paper outlines a system of inductive logic which allows for observational error. In analogy to Bayesian treatments of parameter estimation, the system is based upon a probabilistic error distribution for observations.
In this paper I will introduce the notion of identifiability. The most basic properties of the notion will be stated. In my presentation I will follow, very carefully, Hintikka’s ideas. The theory of identifiability is introduced in the framework of interrogative model of inquiry. We ‘stabilize’ interrogative process by using Abraham Robinson’s diagram method. This helps us in presenting properties of the notion, and in seeing the interconnection between the notions of identifiability and definability.
Science would be impossible without the notion of natural kinds. But that notion is replete with difficulties that have long been the subject of intense philosophic inquiry. In this paper the author begins by reviewing some of these difficulties. He then proposes an analysis of his own based on two tenets. The first tenet is that belief in the kinds needed for science consists of a range of presumptions whose more complex ones entail (but are not entailed by) simpler ones. The second tenet is that natural kinds are characterizable in terms of certain questions pertaining to their members.
What are judged to be the most important questions from a survey of researchers in developmental biology are analyzed in the context of a cytological model of development and in the context of a molecular regulative model for gene expression. These models provide interpretations and a rational ordering of these questions that supersede the intuitively based opinions surveyed. The rational support behind molecular developmental genetics comes from the epistemic import of explaining cell differentiation in embryology, discovering interfield and interlevel relations, and explaining variation, physiological stability and gene conservation in evolution.
Some concepts which may be useful in the analysis of questions and questioning are defined in terms of model-theoretical semantics and multiple conclusion entailment. In particular, the following concepts are defined: soundness of a question in an interpretation of a language, partial answerhood, normal question, regular question, proper question, relative soundness of a question, relative informativeness of a question, self-rhetoricity of a question, and various kinds of presuppositions. The connections between these concepts are examined in detail. In addition, a general overview of the existing logical theories of questions and answers is presented.
This paper tries to combine ideas of Jaakko Hintikka and Jerzy Giedymin in order to achieve a general game-theoretic approach to science. The two players, Myself and Nature, taking part in scientific game are regarded as mutually dual. The same concerns their strategies. Following Giedymin, rational scientific games are defined as games without dictatorial strategies. Following Hintikka, game-theoretical semantics is used for describing the strategies of both players. In particular, strategies of Nature are described by dual logic, that is, logic preserving falsity.
In his “Four Decades of Scientific Explanation” Wesley Salmon locates the roots of all three modern notions of explanation in Aristotle. Yet his account in which the so-called erotetic or question-answer view is but one variant of the epistemic notion of explanation misses an important ecumenical message. This paper suggests that the erotetic or interrogative notion is best viewed as an umbrella notion under which all three rivals, the epistemic, modal and ontic notions of explanation, fit.
This explains why there is a unitary notion — that of a satisfactory answer to a certain form of query — behind the various intuitions of explanation, although satisfactoriness can be cashed out differently in different cases. This view not just restores the angle which somehow got lost during the four decades but also shows that Aristotle was right in thinking that explanations, different kinds of answers to why-questions, fall into distinct types. This view also builds a bridge from the science of logic to the logic of science, and hence does justice to the long interrogative tradition of inquiry.
The aim of the article is to analyze the nature of the construction process of scientific explanations. The search for scientific explanations is a type of inquiry which can be codified by means of Jaakko Hintikka’s interrogative model of inquiry. But there is more to be said; the method for constructing scientific explanations which is developed is based on the interrogative model, but also takes into account the various forms of strategic knowledge we use when constructing scientific explanations.
The object of this investigation is the nature of experimental inquiry and its logic, seen from the point of view of Hintikka’s interrogative model of inquiry. In this framework scientific inquiry is best seen as a multi-level process in which small operational questions are used to deliver answers to big research questions.
I will start with a brief description of the interrogative model and ways of classifying interrogative games in the order of their logical (quantificational) complexity. This illuminates the type of answers we can expect from Nature on different levels of an experimental inquiry. The success of any interrogative game depend largely on the strength of the principal premises and available answers. Lack of strength in initial premises may be compensated by (logically) stronger answers and vice versa. This will be expounded by reference to an experimental knowledge-seeking situation where a two-level conception of the scientific enterprise is sketched: a lower level of experiments where logically simpler answers are relied on, and a higher level of theory construction where logically more complex answers are used as an input. Induction on these two levels is different in nature, and no inductive or other rules of ampliative inference will be needed on either of them. In this connection we will also see that the old distinction between ‘the context of discovery’ and ‘the context of justification’ loses much of its plausibility.
This paper studies methodological issues in various learning models. The framework we adopt is the interrogative model of inquiry (I-model, for short), originally proposed and developed by Jaakko Hintikka. We shall first offer a survey of the structure of various contemporary learning models as well as discuss some methodological issues. Subsequently, the fundamentals of the I-model are presented and the interrogative notion of a learning process is introduced. Finally, we shall discuss some methodological problems which are prevalent in this context. For instance, it is interesting to study the problem of induction within this framework, and some of the most usual methodological restrictions of learning models could be critically assessed.