Morphological Structure in Language Processing (review)

The book under review is concerned with the nature of linguistic rules and their psycholinguistic reality, a topic that has been of considerable importance since the foundation of our discipline.¹ Recently, the discussion of these issues has been fueled both by important advances in psycholinguistics and by the availability of methodologies that allow researchers to study the role of frequency and probability in a much more principled and rigorous fashion than ever before. The existence of large corpora and lexical and syntactic databases, as well as powerful statistical software, has led to a methodological revolution that has affected even those circles that have hitherto followed the practice of building theories on the basis of asking the linguist in the office next door about the ‘grammaticality’ of a sentence or a word invented a minute ago.

The use of the rich, new arsenal of modern linguistic methodology has led to significant findings in all areas of linguistics, which forces us to rethink and reshape many of our longcherished concepts, including the ‘rule’. Probably the most prominent area of research on these questions is morphology, where books like Pinker’s (1999) Words and rules have gained a wide readership, and one that extends far beyond our field. But phonology and syntax are also more and more often subject to approaches that take gradient, probabilistic phenomena into account, without losing scientific rigor or theoretical orientation.²

The articles in the present volume investigate the structure of words and how humans process complex words. A few years ago, not many theoretical linguists would have taken notice of such a volume. Today, theoretical linguists and psychologists not only take each other’s findings into account, but also even work together. The long-term cooperation of R. Harald Baayen (the linguist) and Robert Schreuder (the psychologist) is a prime example of the successful crossfertilization of these fields, and the present book is one more outcome of this joint endeavor. The volume comprises a collection of articles that can be considered a state-of-the-art reference for questions of morphological processing and its implications for linguistic theory. Although the contributions may sometimes be difficult to read because of the statistical jargon that necessarily accompanies the presentation of experimental results, the reader is rewarded with fascinating insights into the nature of linguistic processing and its repercussions for linguistic theory.

Each article is preceded by an abstract, and the volume has a useful subject index. While the editors did a marvelous job, the copyediting and proofreading are not outstanding. There are numerous typographical errors and inconsistencies in the notation of linguistic forms in the majority of the articles. This is sometimes irritating, but I strongly encourage the reader to ignore these formal shortcomings in order to fully enjoy the contents of this book.

The book contains sixteen high-quality contributions, many written by well-known researchers, each one carefully reviewed by the editors, by other contributors, or by outside reviewers. The studies presented address a wide range of issues concerning the acquisition, representation, and processing of morphologically complex words, investigating phenomena in Dutch, French, German, English, Italian, Polish, and Serbian. Many different experimental paradigms are used, so that one also gets a good overview of the current psycholinguistic experimental arsenal. For example, one finds studies using visual and auditory lexical-decision tasks, different priming techniques (progressive demasking, forward masking, delayed and cross-modal priming), naming tasks, spelling tasks, and computer simulations.

From the linguist’s perspective, two issues are of particular relevance, and for reasons of space this review focuses on these, neglecting some issues and articles that might be more pertinent for psycholinguists and psychologists. The first of the two issues is the nature and relevance of the paradigmatic axis in the lexicon. Several experiments described in the volume address the question of how different kinds of relationships between words influence the organization...