The Cognition of Basic Musical Structures (The MIT Press) Paperback – August 20, 2004

Music cognition is an interdisciplinary field that aspires to account for the underlying mental processes that occur when people listen to music. The author presents a computational theory of music cognition that is deeply influenced by "A Generatve Theory of Tonal Music". As in GTTM, the author of this book tries to explain the cognition of common-practice music by a system that generates structural descriptions from musical "surfaces". The author's theory consists of a number of preference rule systems each containing well-formedness rules that define a class of structural descriptions and also preference rules that specify an optimal structural description for a given input. The preference rule systems are presented for six aspects of musical structure: metre, phrasing, counterpoint, harmony, key, and pitch spelling. The author then presents his theory as computer programs that take piano-roll representations of music as inputs and extract information about structure according to his models. The author then evaluates his computer models using objective tests. For example, he tested his metre program on a group of 46 excerpts from a theory workbook, comparing the output of the program with the scores of the excerpts.This book is a worthy heir to "A Generative Theory of Tonal Music" that avoids its failings via computer implementation and objective, quantitative testing. The breadth and depth of the book is impressive. The author convincingly argues that the preference rule approach can be used not only to explain aspects of musical listening, but also features of musical style perception, composition, and performance. He also makes a good effort to apply his theory to musical styles other than common practice music, such as rock music and traditional African music.The book has a few weaknesses. For example, the notion that pitch spelling is used to determine harmony and key seems to be the reverse of what happens in perception. Also, the author's melodic phrase structure model is under-developed and ad hoc. Finally, the author does not compare the performance of his resulting models with that of other systems. In spite of these weaknesses, the book is required reading for anyone who is interested in computational music analysis and cognition. The reader should already be well-versed in music theory and also have an understanding of computer programming and dynamic programming techniques in particular to get the most from this book.The author's computer programs are written in C and are freely available on the web. You can find them by typing "The Melisma Music Analyzer" into Google and selecting the first web address in the list. I notice Amazon does not show the table of contents, so I do that here for the purpose of completeness:1 Introduction 1PART I SIX PREFERENCE RULE SYSTEMS 212 Metrical Structure 233 Melodic Phrase Structure 554 Contrapuntal Structure 855 Pitch Spelling and the Tonal-Pitch-Class Representation 1156 Harmonic Structure 1377 Key Structure 167PART II EXTENSIONS AND IMPLICATIONS 2038 Revision, Ambiguity, and Expectation 2059 Meter, Harmony, and Tonality in Rock 23710 Meter and Grouping in African Music 26511 Style, Composition, and Performance 29112 Functions of the Infrastructure 325Appendix: List of Rules 357Notes 361References 381Author Index 393Subject Index 397

The purpose of this book is to explain computer models designed to simulate how the human mind perceives music. The models are heavily based on Lerdahl and Jackendoff's work with preference rules, and may be seen as a test application of that work. Yet designing a computer program that comes to the same result as the human mind is not the same as discovering the process by which the human arrives at that result. That is where the book is highly speculative. Not enough allowance is made for cultural differences in the way music is perceived, nor for the influence bodily experience of music has on its perception which must of necessity be missing from a computer model. Still, the book is an interesting read for the questions it raises, and for the possibilities it puts forth.

Good