The fourteenth-century English philosopher and theologian Richard Kilvington (1302/5–61) presents a useful correction to popular views of medieval philosophy in two ways. On the one hand, he reminds us that to think of medieval philosophy in terms of Aquinas, Duns Scotus and Ockham, or to think of medieval logic in terms of Aristotelian syllogistic, is to overlook vast areas of intellectual endeavour. Kilvington, like many before and after him, was deeply concerned with problems that would now be assigned to philosophy of language; philosophical logic and philosophy of science. He discussed topics in epistemic logic, semantic paradoxes, problems of reference, particularly those connected with the interplay between quantifiers and modal or temporal operators, and problems arising from the use of infinite series in the analysis of motion and change. On the other hand, this very account of his work raises the important issue of conceptual domain. I have spoken as if Kilvington's work can be neatly classified in terms of contemporary interests; and the temptation to read medieval philosophy in modern terms is only strengthened when one recognizes Kilvington as the first member of the group of Oxford calculatores, men such as William Heytesbury and Richard Swineshead, whose discussions of mathematics and physics have caused them to be hailed as forerunners of modern science.

John Earman's new book, World Enough and Space-Time (WEST), is a brisk account of the controversy between space-time absolutists and relationists. The book is intended, one is told, to be “appropriate for use in an upper-level undergraduate or beginning graduate course in the philosophy of science” (p. xi), but Earman's no-holds-barred approach to the mathematics of space-time theories will have bludgeoned most philosophical readers, undergraduate or beyond, into submission long before it is revealed that Pirani and Williams “have studied the integrability conditions for Born-rigid motions in curved space-times and have shown that space-times of Petrov types II, III, and N do not admit of nonrotating Born-rigid motions” (p. 101). I say this sadly, because Earman's book is a discerning review of an important literature, and most of its main arguments can be grasped even if some technical details remain out of reach. The more you reach for those details, the more compelling the book will become.