Beyond the Edge of Certainty: Essays in Contemporary Science and Philosophy (review)

388 HISTORY OF PHILOSOPHY Beyond the Edge of Certainty: Essays in Contemporary Science and Philosophy. Edited with an Introduction by Robert G. Colodny. (Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1965.) This is the second volume of lectures on various current topics in the philosophy of the physical, biological, and social sciences which has been published under the auspices of the Center for Philosophy of Science at the University of Pittsburg. Volume I, Frontiers o] Science and Philosophy, also edited by Professor Colodny, appeared in 1962. Five of the seven lectures contained in this volume treat topics which fall within the philosophy of physics. The persistent student who is without a technical background in either logic or physics will find these for the most part quite readable. Considering this, it is worthy of note that, while the lectures are not of uniform quality in this respect, several problems are treated with sufficient freshness of insight to justify the interest of the philosopher of science as such. Since I cannot discuss all of the lectures, I will restrict my comments, for the most part, to those the significance of which I think I understand best and which deal significantly with some aspect of the history of philosophy. In "Newton's First Law: A Philosopher's Door into Natural Philosophy," Norwood Russell Hanson exhibits what has become a widely accepted analysis of Newton's First Law as a model to be applied to theories in science which must be regarded as, in some sense, true, but which, in principle, are not empirically demonstrable. Briant Ellis, in "The Origin and Nature of Newton's Laws of Motion," deals in a somewhat parallel way with all three of Newton's laws of motion. It is argued that these laws are primarily conceptual in origin. Evidence is marshalled that they were derived from Cartesian physics. That no amount of empirical work could establish or provide support for a principle of natural motion is held by Ellis to be due to the fact that "force" is conceived as something that exists only in order to explain effects. "Since we are at liberty to choose to regard whatever changes or states we please as natural (i.e., as not being effects) it follows that the forces which we should say are operative in nature depend upon the choices we actually make" (65). It would, I think be difficult to show that any one of the points Ellis makes, with the exception of his interesting historical note on the probable origin of Newton's laws of motion, has not previously been made. All, or nearly all, of these points were set forth, at least in outline, for example, by William Werkmeister in A Philosophy o] Science. The approach Ellis makes to his development of the concept of "force," however, is both fresh and incisive and worthy of careful study. In "A Philosopher looks at Quantum Mechanics," Hillary Putnam exhibits the Copenhagen interpretation of quantum mechanics of Bohr and Heisenberg, viewed in historical perspective, as the most adequate of several unsatisfactory theories. He finds the locus of the problem to lie in a suitable definition of "measurement." Noting that Heisenberg's definition (improperly) appeals to the macrocosm conceived in terms of classical mechanics, he seeks to clarify some of the issues that must follow the attempt to define measurement within the context of quantum mechanics. Putnam holds the relative sharpness of macroobservables for Bohr and Heisenberg to be an underived assumption of their theory. Since the quantities of which macro-observables are supposed to be averages are not (in quantum mechanics) required always to exist, this assumption cannot be derived from a plausible definition of macro-observables together with a suitable formulation of the laws of quantum mechanics. Moving from Bohr and Heisenberg's conclusion that measurement requires interaction of a system (or part of a system) measured with respect to an outside system, and that macroobservables do not exist unless measured, measurement for quantum mechanics is defined as "an interaction in which a system A (the 'measured' system), which was previously isolated , interacts with a system B (the 'measuring' system) in such a way as to cause a...