|Vol. 29 Issue 3 Reviews||Reviews > Publications >|
|Heinrich K. Taube: Notes from the Metalevel: Introduction to Algorithmic Music Composition|
Softcover, 2004, Studies in New Music Research, Volume 6, ISBN 90-265-1975-3, 338 pages, illustrated, with accompanying CD-ROM, US$ 50 (hardcover, 2004, ISBN 90-265-1957-5, US$ 100); available from Taylor & Francis Group/Routledge, 270 Madison Avenue, New York, New York 10016-0602, USA; telephone (+1) 800-634-7064, international (+1) 859-525-2230; fax (+1) 800-248-4724, international (+1) 859-647-5027; electronic mail email@example.com; Web www.routledge-ny.com/.
Reviewed by Dave Phillips
Many books are currently available that describe the use of the computer in various music-making processes. However, the majority of these books deal with technical details of sound synthesis or MIDI operations. Only a few authors have attempted a detailed approach to the specific topic of computer-based music composition, and the works of authors such as David Cope, Phil Winsor, and Iannis Xenakis have already attained the status of classic texts. The list of significant texts is now expanded by the publication of Heinrich K. Taube's Notes from the Metalevel, a guide to algorithmic music composition through the Lisp programming language and the author’s own Common Music, a Lisp-based language designed for what he calls the metalevel of music composition.
Notes from the Metalevel is divided into 24 chapters. Chapters 1 through 9 introduce the elements of Lisp programming most relevant to computer-assisted music composition, while the remaining chapters gather and extend the lessons of the previous chapters, applying those lessons to various interesting musical procedures such as microtonality, Markov processes, and loop-phase structures. The book's tutorial approach takes the reader from the most basic levels to more advanced activity, with clear and complete explanations along the way. The book is very well written and assumes no previous knowledge of Lisp or computer music composition, although prior knowledge in those domains will certainly be helpful.
The book can be considered as an in-depth presentation of the Common Music programming language. The example code is intended for evaluation in Common Music, which is essentially an implementation of Lisp with predefined elements specially designed for the purpose of computer music composition. The tutorial design is intended for hands-on use, and the book includes a CD with the complete Common Music system available for installation on Windows, Macintosh, and Linux platforms. After installing the software the interested reader can immediately interact with the book's example code.
Common Music is a sophisticated system that has been in use for many years by composers around the world. It is easy to learn, inviting exploration and experimentation, and the user will quickly find possibilities beyond the many tutorial examples throughout the book. Common Music is also very powerful, capable of creating complex output, in formats that include MIDI files, score files for the Common Lisp Music and Csound audio synthesis environments, and encapsulated Postscript notation files (via Common Music Notation).
MIDI files can also be used as input files, giving Common Music the ability to function as a roundabout "MIDI to notation" machine; however, the more typical use of Common Music involves the creation of processes and functions for the computer-assisted creation of small- and large-scale musical forms.
The majority of the tutorials prepare their output as MIDI files. Some readers might object to the decision to use MIDI as the base output format for the tutorial examples, but MIDI's ubiquity and ease of use afford a greater guarantee that the book (and Common Music itself) will be of immediate value to the broadest range of interested readers. Most modern soundcards include a MIDI synthesizer, so in most instances the new user should be able to simply install the system and run the example code. If your soundcard doesn't have a built-in synthesizer the book explains how to configure your Common Music installation for use with a software synthesizer such as the TiMidity MIDI-to-WAV renderer. And as the preceding paragraph indicates, it is a relatively easy process to edit the examples for output in a sound synthesis score format or standard music notation.
Like Lisp, Common Music is a graphics-independent language. The author advises us that Common Music is best run in a Lisp-aware text editor such as emacs or Xemacs. The user can then compose, edit, and evaluate Common Music code without leaving the editor. With the proper setup (for both Common Music and the editor) it is possible to immediately play the results of the evaluated code, or even to stream the results for real-time playback (thanks to Common Music's support for the GRAME MidiShare environment).
The book's chapters vary between presentations of fundamentals (normal chapters) and the application and extension of those fundamentals in a specific musical context (etude chapters). The etude chapters relieve the sense that one is reading through a rather specialized Lisp tutorial. They serve the dual purpose of clarifying code application precepts while providing musically useful examples, and the reader is encouraged to experiment with and alter the examples to explore their extensible possibilities.
Many early examples utilize well-known procedures applied to 12-tone sets, but the book assumes no particular musical bias. I consider this lack of bias to be one of Common Music's greatest strengths: it can be used for wholly tonal and deterministic composition just as well as for composition based on algorithms associated with chance, random, and chaotic procedures. To echo Charles Dodge's assessment, Common Music is indeed a truly general language for music composition with computers. This generality is a blessing to composers working in various musical genres and styles, and Notes from the Metalevel is blessedly short on theory and long on application.
Incidentally, although 12-tone processes are often presented as source material, there is no discussion of any particular theory of musical serialism. Common harmonic and contrapuntal processes are often demonstrated, but there is no explication of any particular compositional practice or system. The text assumes a basic literate musicianship, but if the reader is unprepared for the book's musical terminology there is plenty of remedial material freely available on the Internet. Similarly the mathematics employed and referenced throughout the book requires only a basic understanding of concepts familiar to most high-school graduates. Calculus is not required, nor is a degree in digital signal processing necessary.
From Chapter 10 onward the reader is presented with increasingly complex concepts and implementations. Real music is used for the basis of some chapters: chapter 14 presents phase-loop composition by using concepts derived from Steve Reich's Piano Phase, and chapter 22 presents an algorithmic model of György Ligeti's Désordres. Other directly musical material is drawn from work by composers Tobias Kunze, Erik Flister, and the author himself. As a further amenity, the author has included MP3 audio realizations of the more complete examples and etudes on the accompanying CD. Readers who are not yet inclined to install Common Music can still hear the results of using the system (and are likely to be persuaded by those recordings).
The later material is also more complex in its presentation, and here the author's talents as an educator truly shine. Musicians often feel poorly prepared to study exotic concepts such as Markov tables, computer music improvisation, and composition procedures derived from the frequency spectra of audio sources, yet Mr. Taube presents this material in his usual clear, uncluttered prose, always keeping the reader focused on main points while taking care not to summarily dismiss important ancillary material. This readability is another great strength of Notes from the Metalevel, giving it utility and appeal to the widest possible audience, and the author has succeeded in the formidable task of presenting and explaining complex subjects in such a way that the reader is continually drawn onwards.
As is the case with most books about computer music software, the system described has already developed beyond the versions available on the accompanying CD. Thankfully Common Music is easy to update, and the more recent versions offer graphic tools and new language extensions (available from www.sourceforge.net/projects/commonmusic).
I have only a few minor complaints about this book. Despite extensive proofreading a number of grammatical errors remain in the text, though none are particularly offensive or potentially bewildering. A more serious complaint has to do with the book's price. US$ 100 for the hardcover edition is rather forbidding for potential buyers without funding resources. The softcover edition is priced at US$ 50, which is still expensive for many interested readers and students. Nevertheless, I must emphasize that this book is worth every penny of its purchase cost, and even the paperback edition is a sturdy volume. It should be noted that Mr. Taube maintains a Web site for the book, housing primarily code examples and audio examples of the tutorials (pinhead.music.uiuc.edu/~hkt/nm/).
It is customary to recommend a valuable book for a permanent place on your bookshelf. I shall not do so for Notes from the Metalevel. Instead, I recommend that this book remain on your desktop, near your computer, opened to any chapter, ready to be studied and applied. This book is above all a hands-on, practical guide for users, and it is in use that it attains its full purpose.
(Note: The reviewer read the original text of Notes from the Metalevel before publication and assisted with light editorial advice. However, this review has been written after reading the published material).