Online research in philosophy

Turing’s famous 1936 paper “On computable numbers, with an application to the Entscheidungsproblem” defines a computable real number and uses Cantor’s diagonal argument to exhibit an uncomputable real. Roughly speaking, a computable real is one that one can calculate digit by digit, that there is an algorithm for approximating as closely as one may wish. All the reals one normally encounters in analysis are computable, like π, √2 and e. But they are much scarcer than the uncomputable reals because, as Turing points out, the computable reals are countable, whilst the uncomputable reals have the power of the continuum. Furthermore, any countable set of reals has measure zero, so the computable reals have measure zero. In other words, if one picks a real at random in the unit interval with uniform probability distribution, the probability of obtaining an uncomputable real is unity. One may obtain a computable real, but that is in- finitely improbable. But how about individual examples of uncomputable reals? We will show two: H and the halting probability Ω, both contained in the unit interval. Their construction was anticipated in..

Existing work on the ultimate limits of computation has urged that the apparatus of real numbers should be eschewed as an investigative tool and replaced by discrete mathematics. The present paper argues for a radical extension of this viewpoint: not only the continuum but all infinitary constructs including the rationals and the potential infinite sequence of whole numbers need to be eliminated if a self-consistent investigative framework is to be achieved.

This volume shows Bertrand Russell in transition from a neo-Kantian and neo-Hegelian philosopher to an analytic philosopher of the highest rank. During this period, his research centered on writing The Principles of Mathematics. The volume draws together previously unpublished drafts which shed light on Russell's struggle to accept Cantor's notion of continuum as well as Russell's infinite ordinal and cardinal numbers. It also includes the first version of Russell's Paradox.

Abstract This paper challenges Christopher Ormell's claim that an explicit distinction should be drawn between a ?hard? and ?soft? sense of ?having values?. It is argued that holding values is better portrayed in terms of a continuum representing degrees of difficulty and sacrifice, for the holding of any value implies a possible tension between obligation and motivation. Making choices lacks this necessary feature and so cannot be equated with any sense of ?having values?. Ormell's claim that values but not Values are relativistic is also questioned. Finally, an important implication of this debate for moral education is drawn, concerning ways in which children may learn to hold and act upon values.

When it comes to Wittgenstein's philosophy of mathematics, even sympathetic admirers are cowed into submission by the many criticisms of influential authors in that field. They say something to the effect that Wittgenstein does not know enough about or have enough respect for mathematics, to take him as a serious philosopher of mathematics. They claim to catch Wittgenstein pooh-poohing the modern set-theoretic extensional conception of a real number. This article, however, will show that Wittgenstein's criticism is well grounded. A real number, as an 'extension', is a homeless fiction; 'homeless' in that it neither is supported by anything nor supports anything. The picture of a real number as an 'extension' is not supported by actual practice in calculus; calculus has nothing to do with 'extensions'. The extensional, set-theoretic conception of a real number does not give a foundation for real analysis, either. The so-called complete theory of real numbers, which is essentially an extensional approach, does not define (in any sense of the word) the set of real numbers so as to justify their completeness, despite the common belief to the contrary. The only correct foundation of real analysis consists in its being 'existential axiomatics'. And in real analysis, as existential axiomatics, a point on the real line need not be an 'extension'.

The paper examines Aristotle's conception of the continuum, and discusses its topological structure in contrast with modern developments by Cantor and Brouwer. The paper argues for the plausibility of Aristotle's physicalist and abstractionist philosophy of mathematics.

Our model of time is the classical continuum of real numbers, and our model of other measurable quantities that change over time is that of functions defined on real numbers with real numbers as values. This model is not derived from reality or from our experience of it, but imposed on reality; and the fit is very imperfect. In classical mathematics, the value of a function for any real number as argument is independent of its value for any other argument: the analogue is Hume's doctrine that events are loose and separate. This makes continuity in the change of any quantity a contingent law of physica, rather than a conceptual necessity. The article explores alternatives to this classical model.

This paper explores a modal analogue of Hugh Mellor''s version of McTaggart''s argument against the reality of tense. I show that if Mellor''s argument succeeds in showing that the present moment cannot be any more real than any other moment then it also shows that the actual world cannot be any more real than any other possible world.