JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. This content downloaded from 146.201.208.22 on Thu, 01 Oct 2015 23:32:09 UTC All use subject to JSTOR Terms and Conditions BOOK REVIEWS BOOK REVIEWSprovides a brief overview of the symposium and makes the telling point that one "cannot get a machine to do anything very intelligent unless it does understand the knowledge appropriate to its domain. Machine intelligence is machine understanding." A few final, general comments about this volume appear in order. First, the degree to which the book contains contributions that all approach 'cognition,' 'understanding,' and so on, in terms of information processing or computer science greatly enhances its overall impact. It is not uncommon to see a published symposium that is spread too thin and attempts a too great a diversity of orientation among its participants. Happily, this fault is not true of the present book. Second, with few minor exceptions, the papers are very clearly written and represent an excellent cross section of current attempts to model complex problem solving and similar topics by means of nontrivial simulation models and the like. Third, while this book is already the ninth of a series, it is certainly still within the category of fresh, stimulating, and highly provocative material, and one looks forward to the series continuing at this high level of excellence.