was somewhere outside of Birmingham when the drugs began to take hold. I’d arrived earlier that day on a bus from Glasgow, seemingly jetlagged and eye-tired, wanting only oblivion. Yet my obligations as the worst saxophonist ever to bother the halls of the Paisley Youth Orchestra kept nudging me awake, urging me to practice my wholly inconsequential mumblings of forgotten bass lines and random parpings. Frankly, I needed release. It could have arrived in the form of a thousand milligrams of Swiss-pure whiteness, yet, instead of my unreachable oblivion, someone handed me the sheet music to Terry Riley’s ‘In C’. ‘Trust me’, they said, ‘you’ll like it’. Gentle Reader - I did.
It was the start of my love affair with (and I lack a better phrase for it) the mind-boggling and completely enthralling phenomenon that is ‘Generative Music’. But hold on here: introducing pretentious phrases without exposition—surely its explanation time? Well, normally, yes—but here, it’s really non-explanation time. Wait—some caveats for the interested: this short, terribly opinionated piece will not function as a guide to generative music for the unaware. Nor does it really function as an attempt to place some of the musical pieces of Brian Eno in their historical context (if you want that, you can’t do better than read the combined Weiner and Burns tour-de-force that’s been featured on Stylus this week). No, here we have, as ever, the somewhat unhinged and random ramblings of a TRUE BELIEVER. So, my dear fanatics, it is my pleasure to share, with you, my experiences with the nebulous beast that is Generative Music.
But back to ‘In C’ (and, as it helps to have the monster in front of your retinas as your read this, the score can be found here). Put simply, Riley’s genius was to place the performer’s instinct slap-bang into the treble clef of his symphony – while the piece contains 53 phrases of various lengths, the performer is free to repeat them as often or as little as they like, as long as they progress through them in order. One could repeat the first riff five times, for example, then spend twenty times on the second, knowing that one’s accompanying musicians would be making their own completely different choices (influenced, of course, by the aforementioned white granules). Pauses between phrases were also encouraged: this freedom for the performers (I’m using this deliberately in the plural: Riley always said that he felt ‘In C’ worked best with upwards of 30 people) was akin to allowing movie actors to edit their own scenes—finally, the protagonist was writing their own script. Thus the idea of six musicians in search of a composer was finally abandoned—the players were essentially free to write their own destinies between the horizontal bars of the musical narrative.
Yet they weren’t exactly free: while one could travel at will throughout this new landscape, its peaks and troughs had been pre-programmed by Riley in advance. You were free to play phrases—yet they were his phrases. You were free to advance at will through the different parts of the score—yet its order had been determined in advance by Riley. It is here that we reach the raison d’etre of generative music: something, whether it be lines of code in a laptop program or lines of semiquavers in a musician’s score (both ‘agents’), acts to constrain the possible responses that the agent can produce when they are playing the score/running the program—yet, due to the agent’s behaviour being defined and not fixed prior to the recording, one is never aware of how exactly a given reading of the said piece will pan out.
It is in this fundamental admission that one can easily define what is simultaneously these features of generative music that attract and repel people in almost equal measures: it is very unlikely that you will hear a given piece performed the same way, twice. While you may in your darker hours wish this fate on some of the more insipid clatter blocking the charts, it seems somehow tragic that every time you listen to a piece of generative music it may be the last time, ever, that the piece sounds exactly like this.
So far I have avoided any discussion of Brian the Brain (Eno, as those of you with manners might know him). For a polymath like Eno, the world is defined by its currency of ideas; and, like any currency, these can be traded or exchanged. While being aware that his explorations into the randomization of art had been preceded some time ago (for example, that old fogey Mozart once produced a score consisting of 176 measures, whose order was decided by the throwing of dice), Eno, with his background in the miasma of pop music, managed to introduce these ideas to an audience far greater than the avant-garde. While Wild Bill Burroughs had been advocating the use of ‘cut-up’ techniques (generative, again) in writing, and Jackson Pollack’s ‘action painting’ suggested that the force of gravity deserved equal billing on his paintings, Eno was listening to the tape experiments that were coming out of the American left-field in the 50s and 60s, and being radically inspired.
In particular, he became obsessed with ‘It’s Gonna Rain’ by the minimalist composer Steve Reich. The piece itself consists of two tape loops of an American preacher stating ‘It’s gonna rain’. So, right away the initial conditions of this experiment are set: we will never have anything more that the two tape loops. Yet the generative part of the performance is set by the variable speeds of the tape loops themselves: before digital sampling and editing, one always had to be careful that equipment ran at the same speed – and that it did so consistently. Here, the interest in ‘It’s Gonna Rain’ is generated by the gradual desynchronisation of the two sequences, caused by the source tapes running at slightly different speeds and thus hearing them start and stutter in and out of rhythm with one another – using the limitations of the sources in a positive light.
This approach, in a nutshell, drove much of Music For Airports. In 1/2, for example, the lengths of each of the sung notes are designed to avoid synchrony for as long as possible. To be boring: let’s say that I decide that I’m going to sing a note lasting for four beats, and you’re going to sing one lasting for nine. As there are no factors common to four or nine, it will take us the maximal 4*9=36 beats for our little pattern to being again. Yet if you are singing a note lasting eight beats, as four is a factor of eight, our little pattern repeats every eight beats – less interesting, perhaps? Certainly more structured…But you’re not limited to integers, as Eno showed in 1/2—here, there are three notes active, one 23 1/2 seconds long, one 25 7/8 seconds long, and one 29 15/16s…you get the idea. Suffice to say that it takes a damn long time for such a pattern to repeat and resynch, and we, the listeners, get to experience all of the wondrous interplay between these components as they drift between each other, washing up yet more interactions with each repetition.
But does this mean that it’s possible to just throw disparate shit together, watch it evolve, and call it generative art? Some think so, and good luck to them: but for me, the genius implicit in such endeavors lies in the choice of the initial pieces that one wants to repeat—here, Riley, Reich and Eno share an understanding. Whether they had actually done their homework on this or not is unknown (and remember, you can *never* rule out research with Brian the Brain), but work done in neuroscience has suggested that we instinctively enjoy music that is roughly positioned on an axis in between the Scylla and Charibdys of ‘too predictable’ to ‘too disorganised’.
So, it seems that generative music taps into a principle that we are all, whether we like it or not, hard-wired for: a tendency towards an appreciation of music which straddles the fence between a highly ordered and a highly disordered structure, that contains some rules yet allows the music to evolve, using these rules as a template. Eno thankfully continues to experiment with and champion such thinking in both art and music, and, as such, truly embodies his admonition that our grandchildren might look at us in wonder and say: "you mean you used to listen to exactly the same thing over and over again?”
By: Dave McGonigle
Published on: 2004-10-01