Look, Hofstadter is an amateur. He is obviously a music lover, but he's not a music expert by any stretch. This was plainly obvious when he came to my university and gave his little talk.
He assumed that because his talk was at the music school's concert hall, his audience must be comprised of music students and faculty, without actually ever checking if that were the case. Those of us from the music school in the audience were all very aware that the majority of the audience were students from mathematics, physics, computer science, etc. Most music students don't know who Hofstadter is, and don't care.
So when he jumps to his conclusion that he "fooled" 1/3 of the audience with his recording, it's way way off base. After the talk, we all (music faculty & grad students) went out for a drink with Hofstadter and pointed out the errors in his assumptions, but he refused to listen. His methodology is totally bent, and his conclusions are what he wants to think, not the actual data.
I --and all my fellow musicians-- were pretty much all totally nonplussed by his demonstrations, and underwhelmed indeed by the supposed "quality" of the computer compositions. Don't get me wrong, it's impressive that a computer can do what EMI can do. But it ain't Bach, and it sure ain't Chopin.
Also, do not underestimate the human intereference in the process. A human is involved in "tweaking" EMI and "filtering" its output. EMI probably produces 1000 pieces per minute, 999 of which are total crap, and one of which is vaguely interesting. Run it for a few minutes and you'll have a few vaguely interesting pieces. Pick the best one, and say "the computer composed this, isn't it great?". What about the other N-thousand pieces that were shite?
Edit: (It was to long) I think the program is fairly close to the human creative process. While it probably produces a lot of junk so do we. I would consider it a success when presented with a few good ideas out of the countless possibilities. Saying it lacks good taste is different than saying it's not creative.
Thanks for the up-vote. My point is that it's easy to fool an amateur. But classical music, which is what Hofstadter thinks he understands/knows, is often beyond the realm of the amateur. It's not that an amateur can't love classical music, but to understand why a piece of music is great when another similar one isn't is often a rather complicated question.
I think my point would be illustrated by this experiment, similar to his original experiment. Take a piece by Chopin, and a piece by a contemporaneous unknown composer. No doubt to the amateur, they would sound like they could have been written by the same person. Add in the piece by EMI and an amateur would not be able to tell which one was which. The reality is that in order to really understand great music, you need to know a lot more than Hofstadter does. I am confident that I could play him 10 pieces written in Poland in 1835 and he would think they were all by Chopin, even if all of them were utter crap.
Human musicians and composers can riff off a single rhythm or melody and produce something wonderful the very first time. Engineers can visualize a solution, implement it, and have it basically work, with only surface elements changing. Artists can have a vision of what they want to achieve in their minds and cast it out, bit by bit, ever remaining the work they intended it. Sure, each incorporate many patterns already learned and practiced and processed. But computers are a long, long way from even doing that. This is a very different type of creativity.
The other way to look at it would be to see the beauty in simplicity, and gather hope from it. Why is it so important that the human brain is especially "deep"? Does he imply that all "less deep" things have no dignity?
Personally I have for a long time now been fascinated by the complexity that results from simple computations - all of nature around us is an example of it, and who would claim that nature is not beautiful? As an example, I wonder what the world looks like for the spiders in my cellar. Presumably, the cellar is a whole microcosm that looks incredibly strange and alien, if we were to look at it from the perspective of a spider living there. Lot's of complex computations go on there, without an explicit brain being involved. We humans are completely oblivious to it. I hope some biologists at least have looked into such things, though. But there are so many "small worlds" that I doubt biologists can cover them all.
As for music, I don't think it is a god-given thing. Rather, I think it was a by-product of evolution. Because of the brains ability to draw analogies, sequences of tones invoke associations in the brain. It is only an example of the brilliance of evolution that it managed to use this by product for something useful.
Reading that article made me expect a future in which a musician will not be known for a particular composition, but for the algorithm he devised for creating music of a particular flavor. Could be fun, no?
Reading that article made me expect a future in which a musician will not be known for a particular composition, but for the algorithm he devised for creating music of a particular flavor.
This only goes to show that eventually, all fields of knowledge will be nothing except mathematics!
I think he's being a bit too histrionic about EMI here. It produces good music, but it has to be highly "inspired" by a human composer first. Without the works of a genius artist to consume and process, EMI would not be able to produce anything lovely itself. It can only elaborate on a composer's style, not have one of its own. For now, computers still don't have real creativity.
The fact remains that without Chopin, the EMI can't do anything. It can't create, it can only emulate. Rather like a lot of other, lesser composers, for that matter. ;-)
I liked the article, but I think Hofstadter is overrating the philosophical importance of his own subjective experience and emotions here. Just because you had a profound emotional experience listening to human-composed music doesn't mean it's the end of the world or human creativity is ruined if you can get the same feeling from somewhere else. He could probably get the same feelings from the right drugs.
"Spiritual experiences" aren't that rare or objectively profound and Hofstadter would do well not to generalize too much from his own.
I don't understand why he is so morose. I could understand if it had been revealed that Chopin was actually a computer. Being upset about this seems akin to being upset that the world's fastest sprinter can be equaled or bettered by a motor car. Surely what makes the music so great is that it is a product of human ability.
That's what makes the musician great, but it doesn't make the music any better. Usain Bolt is very impressive but I'm not going to rely on him to carry me to the supermarket.
I agree but personally I view art in the sense that it is an intermediary between the artist and the audience, a symbolic representation of the true value of it: the semantics of the piece and how the artist chose to express themselves and not the physical manifestation itself. In a vacuum I don't think artworks are special. They are after all just noise or shapes and colors.
(Apologies if this sounds very pretentious but I couldn't figure out how to say it any other way :) )
Edit: This view was cemented for me when I read the book 'The Power of Art' by Simon Schama. After I read the book I realised that what made most of the great paintings great was the context in which they were created. The fact that some of them are aesthetically attractive or required consummate skill to create is besides the point. Their value comes from the messages in the paintings. For example there was a time painting divine biblical characters with dirty feet was seen as humanizing them, blasphemy. Something which I would never have noticed about the painting if it hadn't been pointed out. And even then I would not have deduced its significance which was that it supported a deeply subversive movement of religious reform.
That being said, traveling forward 400 years or so. The music of Gordon Sumner, AKA Sting is self professed 'doodling' yet it can probably evoke a stronger emotional response than arguably more significant works.
So I retract my previous position and now I don't know where I am :/
I haven't listened to any of them yet, but I am surprised by how few there are. There's only one Chopin-like piece, for instance. This causes me to worry: how many bad Chopin-like pieces did this thing spit out before coming up with a passable one? If there's still a human intelligence sitting there and sorting the good compositions from the bad ones, the program is significantly less impressive.
Amen Mr Hofstadter. And it forces us to think: the universe is so orderly that even deep feeling can be reproduced mathematically! How can we then say that the universe was not designed? There is so much order, so many patterns! When we see this in Apple products, we marvel at Apple's mastery of design. But when we see even greater level of attention to detail in nature, we say it just happened...
As a lover of good music, such as classical and soundtrack music, and traditional Christian hymns, it is troubling. but I say this: No matter how great this program gets, it will NEVER, ever be able to communicate the intricacy of feeling that the human soul produces. It will always be a mechanical analysis of what has already been created - NOT true creation.
It was new, it was unmistakably "Chopin-like" in spirit, and it did not feel emotionally empty. I was truly shaken. How could emotional music be coming out of a program that had never heard a note, never lived a moment of life, never had any emotions whatsoever?
Most emotions don't come directly from music, they come from a listener's interpretation of it. It's not romantic at all. There's nothing magical about music, it's just an abstract way for people to communicate. Composers might pour their emotion into a piece but if no one understands then people won't think it's emotional. And obviously, as he discovered with EMI, simply knowing the psychological tricks a computer program can generate music that does evoke emotion.
My question would be: how deterministic is this algorithm? On a large corpus the program outputs something in the "style" of the composer. (This happens to human composers too, sometimes: late in life they start turning out self-parodies.) But given a single piece, can the EMI algorithms (without jiggering) produce the same piece, or a piece indistinguishable from it? Unless it could do that, I don't see in what way it replaces or replicates the composer.
From a pragmatic point of view, it would be really neat if I could input the songs in my Instrumental Hard Rock Radio on Pandora.com (http://pandora.com/stations/30edcc89cca783c32f86a5cddf5ee0c7...) into the EMI program mentioned, then vote on the resulting songs.
What is art/creativity after all, other than the EMI's basic modus operendi:
1) Chop-up
2) Re-assemble
For example, a unicorn, takes a horn from say, a picture of a longhorn, and sticks that onto a picture of a horse in the mind's eye, and voila, there is a unicorn. Getting bits of information from disparate places and re-assembling them in ways which are harmonious in some sense underlies all art, be it music, or painting, or whatever.
Truth is, in my own opinion anyway, there is no "soul". There is only a complex neural net which can do the above algorithm pretty darn well, or at least, can do it well in the case of artists. Get over it, I say. The earth isn't the center of the universe either. Too bad.
He assumed that because his talk was at the music school's concert hall, his audience must be comprised of music students and faculty, without actually ever checking if that were the case. Those of us from the music school in the audience were all very aware that the majority of the audience were students from mathematics, physics, computer science, etc. Most music students don't know who Hofstadter is, and don't care.
So when he jumps to his conclusion that he "fooled" 1/3 of the audience with his recording, it's way way off base. After the talk, we all (music faculty & grad students) went out for a drink with Hofstadter and pointed out the errors in his assumptions, but he refused to listen. His methodology is totally bent, and his conclusions are what he wants to think, not the actual data.
I --and all my fellow musicians-- were pretty much all totally nonplussed by his demonstrations, and underwhelmed indeed by the supposed "quality" of the computer compositions. Don't get me wrong, it's impressive that a computer can do what EMI can do. But it ain't Bach, and it sure ain't Chopin.
Also, do not underestimate the human intereference in the process. A human is involved in "tweaking" EMI and "filtering" its output. EMI probably produces 1000 pieces per minute, 999 of which are total crap, and one of which is vaguely interesting. Run it for a few minutes and you'll have a few vaguely interesting pieces. Pick the best one, and say "the computer composed this, isn't it great?". What about the other N-thousand pieces that were shite?