Journal of Logic, Language, and Information 8: 114–121, 1999. © 1999 Kluwer Academic Publishers. Printed in the Netherlands. 114 Book Review The Empirical Base of Linguistics: Grammaticality Judgments and Linguistic Methodology, Carson T. Schütze, Chicago: University of Chicago Press, 1996. Price: $28.95, £23.25 (hardback), xvi + 238 pp., ISBN: 0-226-74154-0. 1. Introduction The data on which linguists base their theories typically consist of grammaticality judgments, i.e., intuitive judgments of the well-formedness of utterances in a given language. When a linguist obtains a grammaticality judgment, he or she performs a small experiment on a native speaker; the resulting data are behavioral data in the same way as other measurements of linguistic performance (e.g., the reaction time data used in psycholinguistics). However, in contrast to experimental psychologists, linguists are generally not concerned with methodological issues, and typically none of the standard experimental controls are imposed when collecting data for linguistic theory. Carson T. Schütze’s The Empirical Base of Linguistics aims to show that such methodological negligence can seriously compromise the data obtained, and argues for a more reliable mode of data elicitation in linguistics, based on standard methods from experimental psychology. Schütze reviews the literature on linguistic judgments and identifies a set of factors that influence the judgment process, and hence have to be controlled for when collecting linguistic data. He aims to identify parallels between linguistic judgment behavior and other types of cognitive behavior, an approach that allows him to arrive at a model of the judgment process that explains linguistic intuitions as the result of the interaction of the language faculty with other cognitive faculties. Based on this model, Schütze puts forward a set of practical recommendations for eliciting more reliable linguistic data. 2. Grammaticality Judgments and Linguistic Theory Chapter 1 provides a general motivation for studying the empirical properties of linguistic data: theoretical linguists typically collect grammaticality judgment data in a naive, informal way. Psy- cholinguistic findings, on the other hand, show that grammaticality judgments are subject to a con- siderable number of biases, for which a naive approach to judgment collection fails to control. The central question is therefore one of data validity: “[i]n the absence of anything approaching a rigorous methodology, we must seriously question whether the data gathered in this way are at all meaningful or useful to the linguistic enterprise” (p. 5). The details of this problem are fleshed out in Chapter 2, where Schütze analyzes how linguists typically make use of grammaticality judgments. He points out that the difficulties with naive data collection are amplified by the fact that current linguistic research does not confine itself to cases of clear acceptability or unacceptability, but makes crucial use of subtle (and potentially controversial) judgments: “The days are over when linguistics had more than enough to worry about with uncon- troversial, commonplace judgment data, and the sophisticated and complex judgments now in use by theoreticians assume much about human abilities that remains unproven, even unscrutinized” (p. 9).