[Smt-talk] RN analyzers

[Michael Morse]

Forgive me for reprinting this posting in full; but my questions are directed at its every point.
If there as such a thing as traditional skepticism about the computer-based analysis and production of musical experience, it might be called the emotional complexity school. Computers can never approximate or even map the experience of music, because no machine or algorithm could ever capture the enormous but unquantifiable variety of an emotion, which in any case, not having been born in the first place, computers cannot feel. Although touchingly anthropomorphic, this school misses the trick. T'ain't that no dagnabbed IBM can ever match good ol' homo sapiens when it comes to feeling things; in fact, seeing young people listening to pile-driver rhythms on deafening IPods without moving a muscle, I'm starting to wonder about that. Rather, the core question is logical. If we can have an effective "computational model of how humans understand and process music," it must be based on analogy. There are extant models of "understanding and processing," all of them human. Neither has ever been practiced by a non-human; humanity is essential to their definition. If therefore "understanding" and "processing" is to be extended to computers or computational models--or to bats, wolves, whales, sea anemone, or lawyers--it must be through analogy to human understanding processes.
And what is the logic of that analogy? The term tends to be used with rigor, or apparent rigor, when there is an identifiable point for point correspondence between two phenomena. Roughly, that is the logic of analogic modelling; and it points up the virtual impossibility of computer analysis of musical experience. Ironically, but quite literally, and above all logically, computers lack empathy. Empathy is emotional analogy, or analogical emotion; as such, the ponderous laws of correspondence are the sole sufficient judge of its success. I'm not suggesting that computers lack the capacity to weep when they read Eric Segal's Love Story--so do I--but that the details of the processes involved, contextually rich as they are, are unlikely in the extreme to be reproducible in any sense in a non-human device. As Wittgenstein put the limits of semantic analogy very shrewdly: if a lion could talk, we would not be able to understand it.
MW MorseTrent UniversityPeterborough, Oshawa

> Date: Sat, 25 Feb 2012 20:31:21 -0500
> From: pkirlin at cs.umass.edu
> To: smt-talk at lists.societymusictheory.org
> Subject: Re: [Smt-talk] RN analyzers
> 
> One goal which this sort of analysis can contribute to is a 
> computational model of how humans understand and process music.  It is 
> plausible that having a Roman numeral analyzer will allow us to improve 
> other music-related algorithms, like those for music recommendation or 
> algorithmic composition.  Having such a program would also be incredibly 
> useful for intelligent tutoring systems for music theory, or for 
> integrating into music composition (e.g., scorewriting) software.
> 
> Following up on Michael Cuthbert's comments, having a corpus of analyzed 
> music will also enable further studies to see if patterns in functional 
> harmonic analysis are related to, for example, patterns in expressive 
> performance.
> 
> Also, from a computer science perspective, the technical challenges of 
> solving such a problem are not to be overlooked.  Studying problems like 
> this may lead us to solutions that can be generalized to solve other 
> problems.  We may obtain new ways of modeling computational problems, or 
> for performing learning or inference using those models.
> 
> --Phil
> 
> Phil Kirlin
> PhD Student
> University of Massachusetts
sictheory.org/listinfo.cgi/smt-talk-societymusictheory.org
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.societymusictheory.org/pipermail/smt-talk-societymusictheory.org/attachments/20120226/abf22045/attachment-0004.htm>