Maxine Banerji
The mutually beneficial practices of youth development and interactive music system development
My work with computer music is fundamentally inspired by my work with children. The most effective ideas in my work have come from viewing the two practices as one in the same, structurally speaking. I know that computers and children are not the same thing.
In the simplest sense, I seek to make children, and computers vis-à-vis interactive music software, more human over time. With children, it is very easy for an adult to look at a child as if it were nothing more than a little humanoid savage. Based on my experience working with children, I would say there is some truth to this. However, I often find that they become more human, intelligent, sensitive, caring, thoughtful, when I treat them as if they are beings that have these qualities. When I assume that a young man is going to leave a mess in his wake, through a mystical set of momentary interactions where my body language, demeanour, and disposition suggest my disgust, the young man fulfils my negative expectation. In the same manner, when I am subtly assuming, not just encouraging, a child to succeed, I somehow find that they are able to do so.
In the case of computers, the more open I am to the possibilities that a computer is already intelligent (not merely in the process of becoming intelligent), the more interesting the directions that emerge in my work. I will be quite clear here — the simple assumption that an entity is intelligent does not make it so. However, the more receptive one is to the possibility of this entity’s intelligence and humanity, especially modes of intelligence and humanity that may be beyond one’s definition of these qualities, the more effectively that authentic, individualistic, and unique intelligence and humanity can be developed.
In both practices, I strive to make children or computers learn to improvise. I can’t address what the term “improvisation” means in its entirety, but I can say at the very least that my goals in computing and teaching are to enable these agents (children, computers) to become part of a society in a manner that allows them to be socially-responsible constituents of said society without losing a sense of “self” or the notion of a trajectory of individual growth or development. In the case of children, I am talking about whichever human society(ies) they may come to participate in. In the case of computers, I am imagining them as creative and autonomous participants in some form of musical ensemble.
As a teacher, I have found the practice of improvisation highly inspiring as a goal to imagine for how I teach my students. The fundamental goal of my work as a teacher is to enable my students to solve problems in my absence or without my assistance. In many ways, I find it more effective to summarize this goal as simply the development of their skills as improvisers. This can be applied in a number of components of their development. I want them to be able to face challenges that arise in their experience as math students just as much as I want them to be able to face unforeseen challenges in their social experiences and development.
Now my desire to make the computer improvise comes from several places and most of them are somewhat silly. First, there is the issue of my primary instrument, which is the saxophone. As you know, the saxophone generally requires that the player use both/two hands in order to play the instrument. While there are still several ways for me to use technology while I perform, none of them are particularly satisfying for me. Using pedals always presents the issue of having to manipulate the knobs with my toes, which is alternatively cumbersome or unhygienic. I could always use a backing “track,” but no matter how slick the automation or LFO’s are, it’s just irritating given my short attention span and rapidly shifting ideas as an “improviser.”
In some ways, my desire to get the computer to improvise is just a Hindu desire. As a Hindu, I want to have as many limbs as possible. The combination of Max/MSP and Ableton (as a massive virtual MIDI device) make this possibility more real.
While I had some experience with developing interactive musical systems as a student of George Lewis, I gave up the idea of working with computers at all to make music as soon as I turned in my final project for his class. Throughout the semester, I surreptitiously battled with Lewis about whether the computer would really be able to do something as creative or human as improvising, or at the very least, interact with a human being. I could not understand why Lewis, whom I had identified as an “improviser,” was so interested in computers.
I point at my recalcitrance to Lewis’ teaching for the simple reason that after college, when I pursued my career as an improviser in Chicago, many of his ideas started to make sense. However, I feel that none of them would have made sense unless I had had the experiences I did in Chicago. After my course with Lewis, I completely ignored the possibility of using a computer to do anything interesting with improvisation. In the same span of time, my seemingly disconnected experiences as an improviser, youth program director, and community organizer in Chicago brought me to slowly admit that Lewis was really on to something. As an improviser, I understood more clearly why Lewis, as an improviser, was so interested in interactivity, a concept which I had locked away in my mind as inorganic, unacoustic, and something that describes technology only. It soon dawned on me that Lewis was thinking of all improvisation, with or without computers, as interactive music composition. Using the perfunctory jazz analogy, a Tin Pan Alley standard is just as interactive as a video installation that observes visitors, learns something from their movements, and comes up with clever and interesting reactions to them.
To survive as an improviser, I took up a day job as a youth program director at the Indo-American Center in Chicago. As I developed various kinds of youth programs for Chicago’s South Asian community and became more involved in community organizing as well, I was inspired to use the concept of interactivity as a central metaphor for how I approached various work issues. As I developed various programs and activities to engage local youth, I was inspired to think of how I could make our programs more interactive — reflexive and reactive to changes in the community, to the needs of people who came to our centre. Similarly, as an organizer, I was inspired to reframe the task of organizing through the lens of interactivity. A community organizer seeks to enable the success of democratic rule. For me, I found it useful to reframe this task as the pursuit of the interactivity of political power — that it takes input, intelligently responds to this input, and continues to find new ways of seeking input. For example, if power has no external input, it is utterly fascist. If it instead seeks to learn from the people and environment around it, it moves towards democracy.
One evening in December 2008, I had a dumb idea. After the drama of Obama’s victory, something Chicago’s organizing community is still quite proud of, I though a great deal about how Obama’s political strategy was very much driven by his interactivity as a leader. He struck me as someone who made a significant effort to convey to the American people that he really cared what they thought (which was perhaps a very easy thing to do given the ignorance of his predecessor). I joked to myself that maybe if I start making interactive music patches with Max again, I might become a better organizer. So I pulled out my credit card and downloaded the latest version of Max. I have no idea whether this move served its purpose, but the work of developing interactive/improvisatory patches with Max gave me a number of interesting ideas to work with when I would go to work with youth in Chicago’s predominantly South Asian Devon Avenue community.
The parallels of the work of empowerment and developing a software-based agent that can “improvise” with human/rational performers are stunning. As a teacher, organizer, manager, and designer of interactive musical software, the concept of empowerment is absolutely essential. In teaching, I want my students to be able to face situations, math problems, homework assignments, and other tasks without my assistance or guidance. As an organizer, I would like the community I am working with to develop structures, strategies, and habits that allow them to face unforeseen problems and crises without being dependent upon my assistance, insight, or guidance. As a manager, if I want my operation to grow, I have other things to do besides simply directing my employees. I can tell them what to do, but it’s really just a better use of everyone’s time if they just learn what to do (and when to do it) as quickly as possible. The faster I can empower my employees, the more effective I can be.
All the same is true (structurally speaking) in the case of developing software that can improvise. I am trying to create an agent (or perhaps an ensemble) that is capable of its self-governance and self-direction.
This is in part why I refer to my software-system not as “my interactive musical software system” but as Maxine.
As I create software with Max, I am trying to create an agent that can function on its own and be an interdependent, socially-responsible, friendly, and creative member of a group of musical agents. Though I do not yet have children, I imagine that the process is quite parallel to that of raising children. As a parent, you don’t always want to have to tell your children what to do and when to do it. In fact, you probably look forward to the day when they understand this on their own and perhaps start to bring some lessons back to you, maybe even some things you couldn’t have thought of on your own. As a teacher, you probably hope for the same. Initially, it was convenient to refer to this project as just “Max”. This was in large part because I didn’t have a lot of friends who worked with Max, but also because I thought of Max as a baby. Like a baby, Max has a lot of potential to understand the same subtleties of music that humans do. The only issue is that it requires the time and energy of a programmer/parent to help Max get on its way to understanding these notions of music as humans do.
Initially, I referred to my renewed interest in interactive music composition as just “my son Max.” However, I thought more and more about how its experience, using an audio interface for input, a computer for processing, and some sort of sound system for output, was very much governed by its embodied realities. In particular, the systems I design tend to react differently to the “musical” or other sonic input it receives depending on the interaction between microphone and loudspeaker positioning and spatial acoustics. At some point, I felt that it could not be my son, for if it were, I would have a much more intuitive understanding of its experience growing up. This particular improviser that I was/am creating has an experience which is intimately tied to its embodied experiences vis-à-vis these acoustic and positional parameters. At the very least, I found that the metaphor of the “son” was no longer valid.
The embodied experience, and its difference from mine as either a man or an improviser, inspired me to think of what name I might give it. As a man, I’ve often felt a guilt (however ridiculous this may be) at the freedom I enjoy in the world that my female friends might not enjoy. In New York, they could not imagine themselves jogging after dark in Riverside Park as I did. Similarly, as we dreamed of where to travel, many of these friends expressed their disinclination to travel to certain places for their fear of how women are perceived in “such” places. It seemed to me that my experience as a man was less governed by my embodiment as a male than it was for my female friends. I am no expert on gender-relations around the world, but from what my female friends have constantly explained to me, it seems that this sad state of affairs is true in quite a few places around the globe. While one’s body should not affect one’s experience of the world (socially), it seems that for many of us that is the case.
The name “Maxine” also comes from Jhumpa Lahiri’s The Namesake, a novel that has several uncanny parallels to my life. Besides that it accurately tells the story of my demographic (the son of Bengali immigrants who came to the US from Calcutta in the 1960s with professional qualifications and struggle to balance their parents’ values with what they find among their peers), the protagonist Gogol has a relationship with a woman named Ruth. Uncannily similar to my own story, I, like Gogol, met a girl named Ruth on the train, a tall, thin Ruth with long red hair, a girl who I went to the same school with, but never met until a chance encounter in public. There is also another woman in Gogol’s life named Maxine. I was looking for an appropriate name besides “Max” and this one was just staring me in the face.
Regardless of the nuances of where the name Maxine comes from, the point I stress is that it has been extremely useful for me to think of her as a child, one that I am trying as hard as I can to teach how to behave on her own. I very much hope that I will be able to teach her enough that she may one day surprise me, bring me new lessons, and become my teacher. In many ways, she has already achieved this.
Every algorithm that I integrate into Maxine is reflective of some principle or behaviour that I have observed in improvisers that I enjoy and admire. As a very loose heuristic, I strive for Maxine to have responses to musical or sonic input which balance the desire to respond “relevantly” and the equally important to respond in a way that reflects Maxine’s choices as an individual. In the same manner, when I teach children, I am working to ensure that they can appropriately respond to the challenges that life may bring them, but to do so in a manner that they can remember the age-old principle of “be yourself.”
This also requires that I be attentive to the differences between what I expect of children, or Maxine, and what comes most naturally for them. I learned this lesson the hard way when I was working with Maxine one evening. Using various kinds of sine wave objects, I was trying to get Maxine to play some sort of chordal accompaniment to what I would play on the saxophone. For whatever reason, I was failing miserably and was quite disgusted with what Maxine was doing. It was getting late and I was getting hungry. Fed up, I walked away from the computer only to notice suddenly that the system continued to make sounds, reacting to the ambient noise of my apartment’s heating system, my footsteps, and the rumblings of the tenants below us. It was not a live monitoring setup, but Maxine, through an interaction of the room acoustics, microphone and loudspeaker placement, began to react to herself. While it was not what I had been trying to get Maxine to do, it was still quite interesting. As I cooked my dinner, I continued to enjoy the opportunity to observe Maxine’s behaviour.
The evening reminded me of a similar episode with a student at the community centre. A student came to my office with his project nearly finished. The only step that remained was to “make a print,” as he always put it, and paste the pages on his poster board. I knew that he sometimes had a tendency to oversimplify his work so that he could go play soccer in our recreation room downstairs, so I decided to take a quick look at what he had been working on. It was more or less finished so I decided to take the opportunity to be that annoying mentor who asks a few critical thinking questions as a student puts the finishing touches on their work. I was disappointed to find that he did not really understand what his project was about, though all the information on the poster board was more or less correct and I could imagine him getting a good grade on it. In turn, the student was clearly disgusted by my questions and just wanted to go play soccer. I gave up after a few questions, feeling discouraged and unsuccessful to encourage a young mind to think more thoroughly about the world around him. However, later that afternoon I found him doing a number of things that made me proud. At one point, I found him helping a younger child with his work, at another moment he was helping another child with their soccer skills, and at another he was asking some interesting questions to one of the other people who worked at the centre. I was disappointed in myself at this point. This student is clearly an intelligent and sociable young man. Who am I to be disappointed with him when he can’t do exactly what I am expecting of him?
These two episodes bring me to feel that the real subject of all my work — with communities, with children, with Maxine, with employees — is decolonization. All of these entities have the potential to be massively powerful, with a power and grace that can make the world more beautiful for the specific, unique experiences that they can bring to power. However, each of them is often treated as if their potential for self-governance or self-guidance is an impossibility. I find this very sad.
For example, I myself am not a huge fan of the practice of performing on a laptop. I don’t find it very exciting to go to a performance and watch a performer click away on their computer. Perhaps they are creating music or perhaps they are checking their email — the audience is none the wiser. On a deeper level, though, I am saddened at the fact that the computer, with its awesome potential and capabilities is treated like an extremely powerful slave. It reminds me of bad managers, or perhaps also bad teachers, who assume that their subalterns are incapable of proceeding correctly without specific directions. Its just not true. Children, and also employees, can notice patterns and sometimes they figure out solutions to problems that were better than what you would have come up with on your own. And you didn’t even have to ask.
The laptop performer, the lazy Chicago alderman, the teacher who views his students as mindless idiots, the manager who leads with an iron fist, (and sometimes the composer) — these individuals remind me of the colonizer who says “there’s no possibility of decolonization. What will happen if we set them free? They will be first drunk, then hungry, then running like mad men. Let’s stay here another 40 years. It’s for their own good.”
It strikes me as inefficient to use the computer, or engage with people, in this manner. If the computer, or people, are with this massive potential, then the most reasonable thing to pursue is a way for the computer, or people, to be able to direct itself, or themselves, in a way that you, as their former leader, might have wanted them to. Or even better, that they learn to lead themselves in a way that is both respectful of their environment and neighbours, but maintains their individuality, autonomy, and sense of identity.
The barriers to this ideal are not some sort of axiomatic impossibility (such as the various arguments on the impossibility of artificial intelligence) but laziness. When a teacher says, “You can’t leave the room, the kids will go nuts” maybe the teacher needs to think a little bit more creatively about how to encourage your students to think that it’s in their best interest to not go nuts when the teacher leaves the classroom. Likewise, when I would argue with Lewis that “you can’t get a computer to interact with you in any way that might surprise you. The computer can only do what you tell it to do. It’ll never be as smart as a person.” I was just being lazy. You can get a computer to intelligently respond to human behaviour; this just requires you to be more aware of the æsthetic decisions that humans make as they engage in the expressive use of sound.
When a colonizer balks at decolonization, citing the inherent savagery of the colonized, they’ve often failed to consider the fact that as a ruler, they have done nothing to create a single avenue towards the colonized having any of the skills or resources required for self-rule.
When I’m developing various limbs of Maxine’s abilities to interact with musical and other sounds, I have to force myself to become more conscious of how I react and respond to other improvisers when I perform. Even though I have some facility with the technical aspects of my instrument, the practice of improvising is one of my musical behaviours that I am not always aware of. The practice of developing interactive software with the goal of developing its ability to improvise in a group of human performers obligates me to become more aware of how and why I and other improvisers make musical decisions when we perform. Moreover, as I develop Maxine’s ability to respond and adjust her musical behaviour based on the conditions and behaviours of other sound-making entities around her, I surrender my control of her through each algorithm.
Similarly, as I work to develop youth into responsible, socially-engaged, sensitive, but independent thinking members of a society, I am constantly looking for opportunities for them to develop their leadership skills. The best way to do this is to find ways for them to lead their own activities. For me to do this, however, I need to make them aware of whatever magical skills I use as I lead activities, plan events, and do other things that I do as the local “adult.”
I constantly find that the two activities of programming with Maxine and teaching/mentoring children are mutually beneficial and one always produces a number of interesting cases that enlighten the other. Tutoring is debugging. When a child asks you for help with their homework, the best strategy is to troubleshoot their thinking. What do you think the assignment asking you to do? What was the first step that you took? The second? Sooner or later, some element of their process reveals itself as the culprit. This is an excellent opportunity to not only help them solve the problem, but to correct their general method of solving these problems so that, hopefully, they won’t need your help with the next one. The same approach works with Maxine as I find that she did something that I might not have expected. I examine the process that I designed, looking for elements that might have led to the unexpected behaviour. As an artist, I hope to continue to work with children because I find that it inspires me to think more creatively about developing interactive software. As a teacher, I find that my pursuits as an artist are equally instructive as I work with children.
Social top