Ian Hacking's books are a model of clear, persuasive writing on difficult topics and this is one of his best books. In this book, he lays bare the central issues in philosophy of science (realism and rationality) in a way that makes them accessible to a non-specialist reader.



Almost unique among scholars, Hacking is equally skilled with history, concepts, and traditions, and uses these skills together well to paint a well organized and compelling picture of science; capturing both its beauty and its warts. He also does an admirable job capturing the various ways we have tried to *explain* science over the years.



This is not the usual simplistically "balanced" presentation of philosophy of science that treats all perspectives as having equal votes. Hacking makes his own moderate realist view on the issues clear at each point while he always appears to me to do justice to other viewpoints. One of the things I find most useful and special about Ian Hacking's arguments is that he seems to put great effort into successfully finding the real strengths of each viewpoint before evaluating it. He also often finds instructive points of difference between seemingly almost identical viewpoints.



Hacking discusses both realism and rationality, to place them both into perpsective in philosophy of science. We need at least a rudimentary idea of the role reasoning plays in scientific work in order to be able to speak productively about the reality of the things we are talking about. His focus is however consistently on realism: the question of the reality of theories and the question of the reality of entities and substances. This book is not yet another jeremiad on the "logic of science," or the lack thereof, it is a careful discussion of specifically the *ways* in which things are considered real in science, and the *significance* of treating things as real.



Hacking understands well the complex relationship between theory and experiment, giving us a superbly nuanced but easily understandable summary analysis of important concepts like incommensurability and the "theory boundness" of data. He then applies the results of these analyses back to the larger issues to support his own realist perspective.



The result is a view that strongly emphasizes experimentation. Entities and substances become real in science as they become a practical part of engineering and experimentation, rather than as a result of a verified theory or a long-lived theory.



Hacking is lukewarm about the reality of scientific theories compared to the reality of experimentally useful entities. He says: "Scientific realism about theories has to adopt the Peircean qualities of faith, hope, and charity. Scientific realism about entities needs no such virtues." This is one example of how Hacking uses the philosophical viewpoints of positivism and pragmatism throughout to illustrate the different ways of thinking about scientific reasoning. It would be hard to find a better, more understandable summary of the ideas of Kuhn or Feyerabend, including the specific ways their ideas have been used (or sometimes misused).



I consider this book a classic overview of the philosophy of science that accurately captures both the history of its ideas and the central conceptual issues that make it interesting and important. Hacking's lucid, clear prose and relentless attention to detailed real examples (without ever losing the larger picture) make this book both a pleasure to read and a splendid education in the topic.