In no small part because of the influence of the late Thomas Kuhn, it has become increasingly common over the past two decades or so in English-speaking universities for physics departments to offer interdisciplinary courses in what might be called the informal history and philosophy of physics. Typically, such a course tends to be aimed partly at science and engineering students who may not have reflected seriously on the nature of their disciplines, and partly at non-science majors who may not have been exposed to the concepts of elementary classical physics. Such a course tends to follow a fairly standard pattern: Aristotle, the "apostolic succession" of Copernicus, Kepler, Galileo and Newton, a brisk trot through the 18th and 19th centuries with most of the emphasis on electromagnetism, then on to special and general relativity and quantum mechanics.

It is just such a course, taught by James T. Cushing at the University of Notre Dame, which has formed the basis for this book. Cushing assumes that his readers have some knowledge of elementary classical physics, but not of special relativity or quantum mechanics. He states at the outset his conviction that "it takes a lot of history of science to anchor even a little philosophy of science", and a distinguishing characteristic of the book is its emphasis on giving a historically accurate account of various episodes in the history of physics rather than the pedagogically convenient but air-brushed accounts that tend to be found in textbooks. I think this approach is well worthwhile. However, I am dubious that students who have not previously had any exposure to quantum mechanics will be able to absorb its essentials from the historically oriented material given; my experience with a similar operation is that to convey anything worthwhile to such students in the limited time available, it is essential to take a specific simple example and hammer away at it for long enough - for several lectures if needed - to get across the basic idea before trying to introduce greater generality. There is also some unevenness in the presentation, in the sense that not all the physics included pulls obvious philosophical weight.

With these reservations I think that Cushing has done an excellent job. If I again teach a course similar to his I will certainly consider using his book as a text, probably as a complement to Lawrence Sklar's less historically oriented Philosophy of Physics.

However, this is more than just a course textbook: Cushing very definitely has a thesis to advance, although this really takes centre stage only when one gets well into the section on quantum mechanics. His general thesis is that just as, for example, the ultimate dominance of a particular technology over its rivals is not determined by some transcendent teleological principle but owes much to "historical contingency" and need not necessarily correspond to the optimal outcome, very similar remarks apply to the competition of scientific theories. As a specific example he cites the victory of the so-called Copenhagen interpretation of quantum mechanics (QM) over its rivals and in particular over the "causal" theory of David Bohm; an interesting section is devoted to reconstruction of a hypothetical alternative history of the interpretation of QM in which the Bohm view eventually emerges victorious as the established wisdom.

̽»¨ÊÓƵ

While both the general thesis and the specific example cited deserve serious consideration, I suspect that many philosophically oriented physicists may find that the discussion of the Bohm "theory" and related issues misses several significant points. First, by using the word "theory" as a synonym for "formalism plus interpretation" and mentioning the question of experimental test only in passing, Cushing neglects to stress the key distinction between a description of the microworld that, at least in principle and under appropriate circumstances, makes experimental predictions different from those of standard QM, and one that is guaranteed a priori to give identical predictions under all possible conditions. This distinction is particularly crucial in the case of the Bohm "theory", since the original (1952-3) formulation was of the former type while later versions appear to belong, by design, to the latter category, and the lack of interest, justified or unjustified, shown in them by most practising physicists is probably due to this feature as much as anything else.

Second, even given that the interpretation of the quantum formalism is of interest in its own right irrespective of any experimental consequences, the bald statement that the Copenhagen view of QM is "almost universally accepted by practising physicists" seems at least 20 years out of date. A significant and arguably growing minority is deeply sceptical of the view, which is in essence unquestioned in this book, that QM can provide a complete account of the physical world; it is surprising that the existence not only of this scepticism in the abstract, but of concrete attempts to implement it such as the so-called GRWP theory, goes unacknowledged, since obviously any such programme would, if successful, render the debate obsolete. Moreover, even among those who hold that QM is the whole truth, a substantial minority adheres to the Everett-Wheeler ("many-worlds") interpretation, which does not fit Cushing's definition of a Copenhagen-type point of view. In 1998 the simple Copenhagen-Bohm dichotomy no longer rules.

̽»¨ÊÓƵ

However, the most serious point is that what has most worried the most articulate critics of the Copenhagen interpretation is not so much its lack of determinism but its lack of realism, as embodied in the measurement paradox. By focusing almost entirely on the issue of determinism and merely referring the reader to the literature for the unqualified statement that "there is no measurement problem in (the) causal (ie Bohmian) interpretation," Cushing ducks any discussion of what is to my mind a difficulty of the Bohm approach every bit as serious as anything out of Copenhagen. But that, alas, would be an essay in itself.

Anthony Leggett is professor of physics, University of Illinois at Urbana-Champaign, United States.