scientific article; zbMATH DE number 1764656

From MaRDI portal
Publication:4539491





The author proposes concept expansion (like in expanding the square root function to a domain of negative numbers) as a concept that can be formalized in a ``logic of concept expansion, and essays this thesis well aware of opposite views. (Pp. 2-3): ``Here is the general structure of the argument I present in this book. While Frege claims that the idea of expansions detracts from the principles of reference and sense, and that therefore there cannot be a logic that includes this process, I claim that there can be a logic that includes non-arbitrary expansions, and that there are convincing reasons to believe that a certain type of expansion expresses human rationality. (P. 91): ``Lakatos also suggests a bolder move, which is in a certain sense the opposite of the approach that I have been presenting. He says that true research in mathematics may include attempts to refute the proofs we propose by expanding concepts. In other words, we do not expand our concepts in order to preserve the laws of mathematics, as I have been suggesting, but rather in order to refute mathematical proofs and the laws we used in formulating them. \dots{} ``This raises the question of whether his suggestion, which seems at a first glance to be exactly the opposite of mine, could actually be complementary to it. Furthermore, the author contemplates the following statement of Lakatos: ``The history of mathematics and the logic of mathematical discovery, i.e., the phylogenesis and the ontogenesis of mathematical thought, cannot be developed without the criticism and ultimate rejection of formalism (and reminds that ``Lakatos was not against all formalism). The author views forced expansion as a scientific tool on a par with induction. (P. 69): ``An invaluable incentive for non-arbitrary expansions is the creation of a hypothesis space. The importance of hypotheses for mathematical research is well known to all mathematicians. The art of creating good hypotheses is undoubtedly one of the most important talents a mathematician can have. Now expansions enable us to create a large store of well-founded hypotheses. Reviewer's comments. Ideas of ``well-founded, versus ``bold, hypotheses led in the fifties to the so-called ``Popper-Carnap controversy. In ``Knowledge of evolution and evolution of knowledge [in: E. Jantsch (ed.), The evolutionary vision, AAAS Selected Symposium 61, 129-151 (1981)], the reviewer argues that Popper's program, focusing on ``bold hypotheses, aims at modelling induction, whereas Carnap's, focusing on ``well-founded hypotheses, at describing induction. If both programs are confined to the same language -- in its holistic conception -- the two programs are shown complementary within the language (thus with an essential clarification of ``complementarity as referring to a genuine whole, namely language). The book provides interesting philosophical, mainly historical, views on the dynamics of conceptual changes which fall under the considered domain of concept expansion. Like how vague or intuitive concepts may be expanded into sharp, discussed with reference to Gödel's philosophical insights (aided by historical documents by Wang). However, concerning a general problem, namely if the proposed concept expansion could be more than an historical aid, and really help a mathematician at some new discovery -- we seem to come back to the Popper-Carnap controversy. An interesting border-line case may be seen in \textit{A. Mostowski}'s ``Recent results in set theory [in: I. Lakatos (ed.), Problems in the philosophy of mathematics. Amsterdam: North-Holland, 82-96 (1967; Zbl 0155.33603)]. Here Mostowski proposes two principles for formulations of axioms of infinity. The first, the principle of transition from potential to actual infinity, is shown to admit a variety of axioms of inaccessible cardinals. The second, called the principle of existence of singular sets, concerns a situation where all sets constructed with operations described by set-theoretical axioms which have been accepted so far, yield sets with a property \(P\). Then, if there are no obvious reasons why all sets should have property \(P\), we adjoin to the axioms an existential statement to the effect that there are sets without the property \(P\). In this form, Mostowski writes on p. 86, ``the principle is certainly far too vague to be admissible. It is an historical fact, however, that several axioms of infinity were accepted with no other justification than that they conform to this vague principle. On p. 87, Mostowski continues: ``While it is not difficult to show the independence of the axioms of infinity, proofs of their relative consistency are as good as hopeless. A straightforward application of Gödel's second incompleteness theorem shows that no such proof can be formalized within set theory. In view of what has been said above about the reconstruction of mathematics in set theory it is hard to imagine what such a non-formalizable proof could look like. Thus there does not exist any rational justification of the strong axioms of infinity. Although it seems true beyond doubt that induction, generally conceived as a paradigmatic case of generalization, is necessary for scientific as well as mathematical activity, this does not mean that we have, or can have, rational methods for describing induction. It only means that language is necessary. That is, that we do share some scientific language (in which the inductive description-interpretation processes occur).











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4539491)