Paul Feyerabend's globally acclaimed work, which sparked and continues to stimulate fierce debate, examines the deficiencies of many widespread ideas about scientific progress and the nature of knowledge. Feyerabend argues that scientific advances can only be understood in a historical context. He looks at the way the philosophy of science has consistently overemphasized practice over method, and considers the possibility that anarchism could replace rationalism in the theory of knowledge. -- Amazon.com.

Michael Dummett is a leading contemporary philosopher whose work on the logic and metaphysics of language has had a lasting influence on how these subjects are conceived and discussed. This volume contains some of the most provocative and widely discussed essays published in the last fifteen years, together with a number of unpublished or inaccessible writings. Essays included are: "What is a Theory of Meaning?," "What do I Know When I Know a Language?," "What Does the Appeal to Use Do (...) for the Theory of Meaning?," "Language and Truth," "Truth and Meaning," "Language and Communication," "The Source of the Concept of Truth," "Mood, Force, and Convention," "Frege and Husserl on Reference," "Realism," "Existence," "Does Quantification Involve Identity?," "Could there be Unicorns?," "Causal Loops," "Common Sense and Physics," "Testimony and Memory," "What is Mathematics About?," "Wittgenstein on Necessity: Some Reflections," and "Realism and Anti-Realism." Serving well as a companion to Dummett's other collections, the essays in this volume are not forbiddingly technical or specialized, and have relevance to many areas of analytic philosophy. (shrink)

In this work Dummett discusses, section by section, Frege's masterpiece The Foundations of Arithmetic and Frege's treatment of real numbers in the second volume ...

Feyerabrend argues that intellectual progress relies on the creativity of the scientist, against the authority of science.

Essays discuss relativism, knowledge, creativity, progress, Aristotle, Galileo, cultural pluralism, and reason.

Putnam's internal realism entails the simultaneous rejection of metaphysical realism and (anything goes or total or cultural) relativism. Putnam argues, in some places, that relativism is self-contradictory, and in others, that it is self-refuting. This paper attempts the exegetical task of explicating these challenging arguments, and the critical task of suggesting that a full-blown epistemological relativism may be capable of surviving the Putnam attack.

Such contradictions arise “at the limits of thought” in the following sense: we have reason to set boundaries to certain conceptual processes, which, however, turn out to actually cross those boundaries. The boundaries cannot be crossed, yet they can, for they are crossed. For example, Kant regarded noumena as beyond the limit of the conceivable, yet he made judgments about them, so he did conceive of them. For another example, Russell’s theory of types cannot be expressed, yet he does express (...) it. And so on, from Aristotle’s notion of prime matter to Derrida’s différance. The boundaries that cannot be but are crossed may concern iteration, expression, cognition, or conception. In most cases, a single argument pattern is operative, according to Priest. He calls it the Inclosure Schema [=IS]. It is a contradiction-generating mechanism that works as follows: suppose we define a set Ω, on the basis of a condition φ ); suppose that Ω exists and that it has property ψ. Next, suppose we can define a function δ such that, for any subset x of Ω that has property ψ, we have both δ ∉ x and δ ∈ Ω. As Ω is a subset of itself and it has ψ by hypothesis, a contradiction follows: both δ ∈ Ω and δ ∉ Ω. The two sides of the contradiction—or perhaps the operations by which they are established—are called Closure and Transcendence. ). For example, take Burali Forti’s paradox that is greater than all members of On, and therefore not an ordinal). Here Ω = On, and δ is the function that assigns to x the least ordinal greater than all members of x is both an ordinal—δ ∈ Ω, “Closure”—and not a member of x ). The condition φ is just the property of being an ordinal. (shrink)