Vol. 30 Issue 3 Reviews

James Matthew Bohn: The Music of American Composer Lejaren Hiller and an Examination of His Early Works Involving Technology

Hardcover, The Edwin Mellen Press, 2004, ISBN 0-7734-6440-9, 307 pages, illustrated, foreword (by David Rosenboom), discography, sources, notes, index, US$ 119.95; The Edwin Mellen Press, P. O. Box 450, Lewiston, New York 14092-0450, USA; telephone: (+1) 716-754-2266; fax (+1) 716-754-4056; electronic mail imiller@mellenpress.com; Web www.mellenpress.com/. In Europe contact The Edwin Mellen Press, Ltd., Mellen House, Unit 17, Llambed Business Park, Lampeter, Ceredigion, Wales SA48 8LT, UK; telephone (+44) 1570-423-356; fax (+44) 1570-423-775; electronic mail iwilliams@mellenpress.com.

Reviewed by Louis Ferdinand
Meudon, France

Looking back from a technical point of view on computer music from the 1950s, one has the impression that it is ancient history; while at the same time it is surprising how contemporary the ideas from 50 years ago still appear. Descriptions of hardware and software from 50 years ago probably seem like cave drawings to most people today. While hardware and software developments were significant from the 1950s through the 1980s, for most practitioners of computer music these developments were relatively transparent. The working method and environment of a user sitting in front of a teletype, punching cards, waiting for batch jobs on a large, somewhat non-descript behemoth computer at a university did not change much for the lone "eccentric" from a music department sharing time with scientific researchers (who fortunately were sympathetic to anyone in the arts even vaguely interested in computing)

For most composers of computer music, except for the odd mini- or micro-computer user or the fortunate few to have worked in "elite" research centers, it has only been in the past 15 years or so that hardware and software have gone through a major revolution. Some might argue that this "revolution" is nothing more than a change of working habits, a change on the surface, while others would insist that this overhaul in the tools of the trade is directly related to artistic creativity in profound ways. While this discussion of the tools and technology is engaging, the fact remains that many of the prevalent ideas of the 1950s (independent of hardware and software) remain surprisingly relevant today. David Rosenboom, in his heady forward to James Bohn's The Music of American Composer Lejaren Hiller and an Examination of His Early Works Involving Technology, recaptures some of the intellectual adventurousness of the pioneer era of computer music:

Hiller's contributions to procedural thinking… he employed serial techniques and understood them as a subset… inside combinatorial mathematics… He introduced the idea of the interval row… developed a measure of linear angularity… dissonance measures… weighting schemes for parametric behaviors… geometry and topological ideas… as descriptors of forms… pseudo-random numbers, Monte Carlo-like generate-and-test loops, moving-window filters on stochastic cannons… notions of formal and perceptual hierarchy… rhythmic hierarchies… Markov transition probabilities… systems and information theory. (pp. vii-x)

Add a couple of popular terms like “chaos theory” and “genetic algorithms” to the mix, and one has a reasonable list of much of today's intellectual fodder.

If only Hiller could have done a computer search of the terms "algorithmic music" and "computer generated music" 50 years ago! The other day, Google returned about 38,000 (in 0.07 seconds) and 51,300 (in 0.11 seconds) hits, respectively, for these terms. What would be different today without his decision to abandon chemistry to pursue a career in music? What is his historical importance, and what are his major contributions to the field of computer music? Any serious study of algorithmic music must begin with Hiller’s seminal work Experimental Music, published in 1959. (Iannis Xenakis' Formalized Music from 1963 makes up the other half of seminal publications in the field from that period.)

The most striking thing about Experimental Music are the musical questions that Hiller poses. But where do we go from there? Or even better, where does algorithmic music fit into contemporary computer music society? Algorithmic composition using digital computers was on an equal footing with digital synthesis 50 years ago. Using proceedings of the International Computer Music Conference as a yardstick, one could argue that the discipline has become a backwater in a long slow fadeout. I found this rather disgruntled comment in one of the Google hits: "There are those that in agreeing with Xenakis' statement that ‘music, by its very abstract nature is the first of the arts to have attempted the conciliation of artistic creation with scientific thought’ wonder why computer music conferences are filled with engineers dedicated to cutting-edge technologies, speaking favorably about conservative and embryonic ideas of music composition, if they even mention music composition." Are there still composers bothered by the question of whether or not a computer should be involved in the composing process? As Mr. Rosenboom states in his introduction to Mr. Bohn’s book: "The term, algorithm, is already off-putting to many. The components of intuitionism, valued as a part of humanistic music making, are presumed to be absent from algorithms" (p. vi). Has algorithmic music become a dirty little secret of the computer music field: does everyone do it while no one admits it, or else do very few composers actually employ algorithms in their music? What is your algorithm?

A recent article about a very prominent American composer stated that the composer used the computer to solve certain musical problems using a variety of composition software. The reader could infer that this was not the composer’s "normal" way of working. (To be fair, the same article mentions that his most recent opera has a two-minute electronic introduction which is noise, not music. It is inferred that this electronic introduction was created by a sound designer and not by the composer himself. Just because synthesis techniques have more cachet in the computer music literature today, we should not lull ourselves into thinking that computer sounds are favored over algorithmic composition techniques by "mainstream" composers.)

While Mr. Bohn's book has the air of a published thesis, the print quality and photo reprint quality is low and it contains a number of typographical errors, it is a hugely significant book. There is a serious lack of publications on algorithmic music and on Hiller in particular. This first in-depth evaluation of Hiller's contributions to contemporary music should be forgiven its cosmetic faults. The book has introductory chapters on Hiller's life and works, and his computer music output. While Hiller studied music in a traditional academic setting with Roger Sessions and Milton Babbitt during his undergraduate days at Princeton, and while he presented himself as researcher, he clearly belongs to the American experimental tradition of Charles Ives, Henry Cowell, and John Cage. His compositions, both computer-assisted and non-computer-assisted, are grounded in the experimental tradition.

Some will be surprised to read that he composed a fairly large number of piano sonatas, string quartets, chamber pieces, incidental music, and orchestral works, well before the famous ILLIAC Suite of 1957. (Of his 75 listed works, only a fifth of his output employed computer-assisted techniques, and less than two-fifths employed electronic or computer-generated sounds.)  Mr. Bohn states that since his "work will investigate four of Hiller's compositions whose creation involved technology… a fair portion of this text is devoted to the technology itself" (p. 18). The technology used for the four compositions includes the ILLIAC I and II computers, the IBM 7090 computer, the CSX computer, and the Experimental Music Studio (EMS) founded by Hiller at the University of Illinois in Urbana-Champaign. A chapter is devoted to each of the four computers and to analog equipment in the EMS. The history, development, and operation of the computers are described in great detail. (It might be unexpected for some to learn that Hiller made use of computers for algorithmic composition, music printing, and real-time digital audio by the early 1960s.)

The chapter on the EMS covers virtually every piece of equipment that existed in that studio during the late 1950s and early 1960s. These chapters will probably be skimmed by some (they comprise one-third of the book), but will be relished by others. They are written with the same attention to detail as Paul Doornbusch's marvelous tome The Music of CSIRAC (reviewed elsewhere in this issue), which describes Australia's first computer music from the early 1950s. Anyone who enjoys reading about the mercury acoustic delay line-based memory of the CSIRAC computer and how this unusual technology influenced the music made, will find these chapters in Mr. Bohn’s book fascinating reading. Mr. Bohn understands that a detailed explanation of the tools and techniques that make up a particular technology contribute to an understanding of what was created with the technology.

Following these chapters, Mr. Bohn has a general chapter entitled "Common Stylistic Traits of Hiller's Oeuvre." Hiller was eclectic and diverse in both his influences and his style. The composer's comment that "this American experimental tradition exists… its central attribute is its emphasis on the empirical and practical rather than on the speculative and theoretical" (p. 118) illustrates his own sense of self, and contradicts some of the mythology about him. Ives, jazz, music theater, environmental soundscapes, collages, world music, the "half-baked" musical concept serialism, stylistic parody, micro-tonality, and numerical schemes all found their way into his music. We need more traces like this, of the origins of our post-modern period, to counter the simplistic notion that post-modernism was a heroic revolt against modernism.

The four detailed chapters on Hiller's early works involving technology (a chapter is devoted to each of them) are first-rate. Anyone who has been confused by Markov chains or the Monte Carlo method will no longer be, after reading the chapter on the ILLIAC Suite. While the flowcharts are sometimes difficult to read due to print quality, and the 35-page description is less detailed than Hiller's own explanation in Experimental Music, less patient readers might prefer Mr. Bohn's accurate distillation to Hiller's extremely detailed analysis. The next chapter on the 1963 composition Computer Cantata makes much use of Robert A. Baker and Hiller's 1964 article "Computer Cantata: A Study in Compositional Method" from Perspectives of New Music, but Mr. Bohn has some novel things to say about the piece. Hillers's harmonic and linear dissonance concepts are clearly explained and the discussion of form is original.

After this, we step into less charted waters with the chapter on the 1964 composition Machine Music for Piano, Percussion, and Tape.With less thorough documentation by Hiller himself to draw on, Mr. Bohn combines information from two or three documents by Hiller with his own original analysis of the work. Here, Hiller's fascination with information theory is integral to the composition and to Mr. Bohn's analysis. The final chapter of the book covers HPSCHD,composed in 1968. Mr. Bohn makes heavy use of the in-depth interview "John Cage and Lejaren Hiller: HPSCHD" by Larry Austin in the now-classic Source magazine, as well as significant use of Hiller's article "Programming the I-Ching Oracle." (This chapter contains an excellent explanation of the I-Ching for anyone who has missed out on that tidbit of ancient knowledge.) Mr. Bohn’s contributions to the record are significant from a historical perspective.

The author makes a striking point: Hiller apparently had an unusual ego for a composer. He was a collaborator. Three of the four pieces covered in the book were co-composed. The co-composers range from a laboratory chemist, to a Ph.D. student in musicology, to John Cage. In other works, he collaborated with a composition student, a performer, and the mathematician John Myhill. Mr. Bohn's discussion of the music as works of art is positive, but ambiguous. This ambiguity is refreshing. We are not treated to the word "masterpiece." Perhaps this very issue should have been discussed.

In the book Collapse: How Societies Choose to Fail or Succeed, Jared Diamond writes:  "The values to which people cling most stubbornly under inappropriate conditions are those values that were previously the source of their greatest triumphs over adversity." Replace the word values with ideas, and perhaps one could argue that Hiller clung too tightly to the idea of music as an expression of information theory and form as an expression of symmetry. These were certainly two of the most significant ideas running through his music. I am not ready to make sweeping aesthetic judgments about Hiller. It is true that his music is not performed much today. He is remembered for the ILLIAC Suite, which many consider an interesting "experiment;" he is thought of as the person who helped Cage realize HPSCHD (rather than as co-composer); and he still suffers from the reputation that plagued him all of his life: scientist first, composer second. Possibly this book will contribute to a more intelligent re-evaluation of Hiller the composer.  

Do we still need to have discussions about algorithmic composition? No one would dispute the fact that algorithmic thinking has always been an integral part of the act of composition. Is it solely the fact that computers are used to realize algorithmic processes that has caused controversy? Do architects, artists, filmmakers, and others in the arts use computers without the fears mentioned by Mr. Rosenboom? Perhaps music suffers from the fact that it was in the vanguard of computer use for artistic pursuits? Perhaps it is no longer a separate category in the discipline of computer music? Have computers been absorbed into the act of composition, as they have been absorbed into much of modern life? Is it possible that technology's transparency and ease of use, the "revolution" that has put the computer on our desk next to the musical keyboard and the pencil as just one more extension of our tools, have finally freed the term "algorithmic” to be absorbed into the larger concept of composition? If there is no longer a need to qualify composition with this label, then perhaps we are all forever in Hiller's debt. But, then again, he might be surprised to see how algorithmic music has become a less distinct discipline. He might even be disappointed at composing-by-numbers software that offers algorithmic presets. After all, Hiller was interested in using the computer to explore essential musical issues. As he pointed out on page one of Experimental Music, composing using a digital computer “immediately raises fundamental questions concerning the nature of musical communication and its relation to formal musical structures.”