Composition? There’s an App for That… (Part 1)

As the academic year comes to a close, I realize it has been a quite while since I have written a post. Initially, I thought I would just share some fun music-generating web-links that I ran across:

But then I got to thinking about the music-making involved and started asking myself questions such as “What does it to take to make an application that can generate more-or-less pleasing music regardless of musical ability on the part of the ‘composer’?” That’s just about when I began reading Jason Freeman’s NYTimes opinion piece “Compose Your Own, Part 2” and its prequel “Compose Your Own.” This led to a number of other questions that I will address in separate posts in the upcoming week.

The ‘Building Blocks’ of Music?

As I pondered what makes composition apps work in a musical sense, my attention was drawn to the issue of sequence or—perhaps more appropriately—the lack thereof. The composer/designer of the app relinquishes the decision-making power over the order of musical events and must therefore accommodate the potential musical outcomes. Each app is designed in such a way that more-or-less pleasing music will result regardless of the actions of the user.

Such an approach to composition is by no means novel. In many ways, these apps conceptually resemble Earle Brown’s Available Forms I, in that the lines between audience, performer and composer are blurred by giving more responsibility for the sequence of musical events in the final musical product to someone other than the initial composer (Überkomponist?). In this sense, the initial composer has produced musical ‘building blocks’ that may be put together in any ‘legal’ (i.e., permitted by the rules of the initial composer) way such that the result will always be effective.

Freeman’s app is perhaps the most clear realization of this conceptual approach. He denatured composition into a pseudo-visual/auditory task of piecing together blocks of musical gestures in a web-based platform called Piano Etudes. Unlike Brown’s piece, however, the ‘composer’ need not read music given that all of the musical fragments are represented visually by pseudo-registral/durational notation.

Beyond Musical Sequence

The simplistic impression suggested by the ‘building block’ analogy is perhaps misleading. Much Western music written before the twentieth century has some significant sequential component that cannot be arbitrarily dismissed. An obvious example might include the so-called ‘sonata form,’ with the resolution of secondary material in a principal key area upon its return. In any other sequence, it would simply not be a sonata (especially by definition, although the musical impact would also be lessened).

Even the popular music of earlier times had specific sequential determinants. A performer could not simply piece together the various phrases from a Baroque dance suite and hope that the outcome would make sense. Rather, the harmonic and cadential schemata require accurate sequencing. This differs greatly from modern popular dance music. If there happens to be a differentiation between ‘verse’ and ‘chorus,’ the order of presentation likely makes little difference. Even more explicitly sequence-free is the product of the live DJ that combines various repetitive patterns in overlapping layers with other musical gestures that need not suggest any particular musical event.

The composition of musical fragments that can truly go in any order is an entirely different matter.

Non-Sequential Music and the Audience

Music that can come in any sequence must essentially be more-or-less effective regardless of when the audience begins listening to the materials. In this sense, the listening should be able to begin at any point. An audience member could theoretically walk in during the middle of a performance and still appreciate the music because any component can sound like a ‘beginning’ just as much as an ‘ending.’ That is, no component of the music will rely on any previous component and can thereby serve as an entry or exit point to the audience’s attention.

This seems to have great implications for keeping an audience in their seats according to the traditional model. Part of the value of sitting through an entire classical symphony, for example, is hearing the logical conclusion of that which comes before. Missing some earlier portion of the music may preclude appreciation of a later portion. Therefore, rapt attention is incredibly useful, if not necessary. If there is no sequence to be perceived, is such rapt attention the most useful audience model? And yet, much can be appreciated in each new performance of such a work.

As many new music ensembles seek to maintain or increase audiences, one notable trend is the shifting of attention away from the music. Such ensembles play in bars or clubs where talking and mingling is not only accepted, but also encouraged. Concerts are paired with dinners or other artistic endeavors that deserve their own attention. Music is no longer the focus at such events, so much as is the sensory experience.

In a way, this also reflects popular consumption. Students listening to their iPods between classes are by no means paying rapt attention to the music. Music is everywhere and paired with every sort of experience from movies to museums, elevators to telephones, and the like. Even classical music on NPR is often transformed into ‘background’ music that can be entered into or exited from with ease. Without understanding the logic and sequence of the music, it too becomes non-sequential.

Could it be that listening habits are changing to better appreciate the music? Does the music change to meet the listening habits? It’s hard to know for sure, but musicians and music advocates alike must take these issues into consideration.

Sequence vs. Non-Sequence

I will not suggest that either sequence-dependent or non-sequential music is somehow superior [I definitely enjoyed BallDroppings!]. Rather, I merely want to examine the related issues. However, what I will end with is an interesting note from the results of Freeman’s project. He allowed users to submit ‘compositions’ to be judged (by himself). The winning ‘compositions’ would then be prepared for performance and recorded (you can hear the results in the NYTimes article).

Two interesting features emerged in the winning ‘compositions’: 1) minimalistic repetition; 2) goal orientation. On the one hand, this could simply reflect Freeman’s tastes as a composer and audience. However, the comments of the ‘composers’ were also rather revealing in these respect:

“I’m interested in patterns in nature, and I was thinking about them when I composed the etude. A pattern is a sequence that repeats in time, space, or both. Because our world is finite, patterns must have boundaries. What does the edge of a pattern sound like? What about the boundary between related but different patterns?”

“I began with what I felt like were the more ‘pop’ elements of the score and created a loose musical narrative around those ideas.”

“I organized the material to create shifting states of rhythmic and harmonic tension within an overall hypnotic, pensive space. My intention was to allow this ebb and flow of tension to gradually unwind into a closing series of calming, pacific breaths.”

“I actually took a course from John Cage in the ’60’s at UC Davis, and am familiar with such ‘alternate’ forms of composition. … The choices were made with an aesthetic in mind — definitely not aleatoric!”

It is intriguing to me to see these ‘composers’ describing the opposing forces of order and chaos, sequence and non-sequence as they composed. They were interested in the moods created by repetition and yet were driven to seek some goal. There seems to be a satisfaction in both the appreciation of a single item in detail just as there is in recognizing global relations over the scale of a composition. Semper idem, sed non eodem modo (always the same, but not in the same way), invoking the memory of Schenker.

Leave a Comment