Paradigm Composition

From my friend on facebook:

"Granted, I only skimmed this, but it seems like it might be right up your alley...." (link)

That is up my alley. I've been there in less significant ways. I conducted the same test type of test in the defense of my thesis on algorithmic composition. It was Bach vs. Robo Bach. I saw the tip of the iceberg of the same types uncertainty and paranoid bigotry.

The robots kick our asses on a lot of early styles. What's needed in later styles and to emulate quality "human" composers, namely breaking patterns in predefined ways, is just a variation of following a pattern, so they manage that well enough also.

But our definition of music and range of expression expanded a lot with 20th century music. Inventing new compositional paradigms (chance music, etc.) is much closer to "humanity," "creativity" etc. Robots still suck at that, and will for a good while, if not always.

What's interesting is that most humans don't give a fuck about new paradigms or expanded expression. They want a good "beat" or a nice "tune." If we stop caring about innovation, and "academics" refuse to recognize what actual innovation would mean, then all we have is "robots", "weirdos" and "composers" who lose handily to robots.

Composers will always lose to robots in "composition." If that was their only job, there would be nothing special about any of them. As it stands, I am not certain that there is anything special about current composers who strictly imitate past styles. If a "composer" is responsible not only for the generation of musical ideas, but also the creation of new compositional possibilities, then they don't lose. The trouble is that the we've got a lot of "posts" to prefix our modern by now, so anything they do in the paradigm expansion effort is going to be pretty out there.

The academics are in a tough place. They can take on the difficult challenge of expanding their definition of music to include styles which are more difficult to understand or even tolerate than CPE or pop music. Should they not do this, they are little better than the tune and beat seekers. They understand the comparatively easy and mechanical aspects of music, but they are rejecting humanity and creativity.

I don't know if or where the soul is, but it's not in the notes. Admitting this puts the emphasis in a different place. Basically, composers who did not invent a new paradigm of composition are irrelevant. All that we value is that which computers are incapable of.

Computers have largely solved the problem of automatic algorithmic composition. But if they are to solve the problem of automatic algorithmic paradigm composition (the true skill of a "composer"?), they will need a system to understand how humans break and define paradigms. This is in the works. A lot of content that is passed around online is interesting because it changes how we think about solving certain problems. So now we're down to the computer having to filter out the inane from the brilliant. Analytics are more ubiquitous than ever. Internet based products and services attempt to understand the information produced and used through them more and more by the day.

We're a ways off from the computer saying "PRODUCE NEW PARADIGMS HUMAN!" but how far, I'm not sure. "You want to be popular and make money, right?" is only slightly more subtle. I don't know if the singularity is on its way. To a large extent, I don't care. It might be horrible or benign, but if it is to come, I can't do anything anyways, so for now I say "meh."

But back to musical composition. Basically, if we don't make an effort and accept current paradigms that we may not be comfortable with(20th century, other cultures, etc.), and cultivate the creation of new ones, then we become irrelevant. But is the best way to accept old paradigms attempting to create systems to make them obsolete? I don't know. What is the best way to cultivate new paradigms of musical expression? I don't know. Should and is it possible for us to give computers insight into every aspect of our expression? I don't know.

Another question is whether what will become "human" in musical composition is limitation itself. Maybe the effort required itself becomes novel.

Or maybe the limit that we hit before computers is an ethical one. Throughout the centuries, music's dissonances have become more pronounced and intense. Is mass murder worth extracting the spectral analysis of the sounds of a knife against flesh, and seeing if over the course of many murders, it shares characteristics with the corresponding cries for help? Most musicians I know what say, "Ew. What? That's abhorrent. No." Most computers I know would say "beep boop beep 01010011100001"

So what's the lesson? Be nice to weirdos. Be accepting of things contrary to your world view, or hard to define, or ill defined, or undefined. Create new paradigms. That's our only insulation for protecting our humanity. Keep adding more "post-'s" to the post-modern. As far as I know, no computer or human has defined a human limit for where we can take that aspect of our humanity. If we keep pushing that possibly infinite limit, we have nothing to fear of analysis and automation robbing us of our humanity.