I’ve just read a fascinating article about composer, David Cope, who is known for creating music in the style of Bach, Mozart, and others with software he developed called Emmy. I first heard David Cope’s work on Radiolab, and was intrigued by his approach. The article, Triumph of the Cyborg Composer, discusses his latest application titled Emily Howell. Cope is using the computer in a more collaborative way to compose music in his own style with the help of his program. A couple of audio examples within the article illustrate the musical results.
Cope has received a lot of criticism regarding his work, including statements that his music lacks soul because it was written by a computer. But was it really written by a computer? I think a better term is generated. Cope wrote the software, so I would argue that the music generated by the software was ultimately written by the software developer. In this case Cope himself. In other instances I might argue that the music was created by the user of the software tool, rather than the developer of the software. It comes down to who is at the controls. What decisions are being made, and by whom, or perhaps what?
Since I’ve developed and am currently using software to perform and record generative music, I am curious about your opinions. You may have heard pieces on this site generated by the GMS. Perhaps you listened to the excerpts in the article. What do you think? Does music generated by computers lack soul? Does it diminish the human, communicative qualities contained in the work? Or, are we using computers simply as tools? Perhaps, as computers and software evolve we might begin to collaborate artistically with them rather than just use them slavishly. Based on Cope’s work and others, I believe that we are closer than we think to this becoming a reality.
machine based music sometimes lacks soul, yes. then again so can music composed by a human. i don’t think there’s an easy answer to this question. in my opinion it’s a matter of how artfully the machines are both constructed and utilized by the programmer/composer, and on the resulting composition.
i was particularly interested by the section of the article in which Cope describes, “asking musical questions” of Emily. or writing smaller purpose-built programs based on a stricter set of instructions. then pairing down the result, much as a sculptor might eliminate the bits of marble that don’t belong.
to me this is where the soul of computer music can come through. Cope seems to take on a sort of editorial role. in the end he is the one deciding which of Emily’s data-strings have something beautiful or moving to them, something worth recording or repeating. any other listener will have to make that decision too.
Nicely put, Luke. You have a similar view to what I was trying to state in my post. I’m looking forward to hearing what happens when the machines become artists in their own right and no longer need our input or guidance to create meaningful work.
I’m reading this with interest. As a software developer and a part-time composer, I do believe it is possible to program in some “soul”. If the extent, makeup, and pattern of what it is that we refer to as “soul” can be mapped out, that essence can be turned into an object – which in turn can be applied to the musical piece.
Let’s put it this way: Emily’s data strings relies on Cope. Cope’s editing of Emily’s output makes Cope a “Template”. I would envision that the Soul Program would be able to create soul templates out of any individual’s musical styles or preference.
In a way, we already have the tools out there. Itune’s system of rating music, and the way that the program keeps track of what you are listening to, or prefer to listen too out of your own database of songs can be a basis of such a program.
…just a thought.
@Subandi I agree. It will be fascinating to hear what happens when the machines start making aesthetic decisions of their own.
Subandi, I would disagree. I spent several years in college studying music, mostly classical, but with some 20th century composers mixed in. I don’t believe that computer generated music can have soul because it, in and of itself, lacks emotion. If there is soul in this computer generated music, then it comes from the editor of the piece, not the computer.