When I was in graduate school, a hot topic in musicology was the
authorship of a set of quartets known as Haydn's opus 3. There were good
reasons to suppose that Haydn did not actually compose those works. To
make a long story short, many scholars argued that the real composer was
someone named Romanus Hoffstetter. A friend of mine used this controversy
to develop techniques of computer-aided analysis of music. Analyzing the
op. 3 quartets along with a set undisputably by Haydn and another
undisputably by Hoffstetter, he found a high statistical probability that
Hoffstetter was the composer. He successfully defended his dissertation in
1977.
There is no question that a computer can uniquely identify an author.
There is also no question that whatever computers could do thirty years
ago, today's computers can do faster and with less cumbersome programming
tools. The only question I have is whether they can do it quickly enough
and on a large enough scale to give librarians a really practical tool. I
suppose it is only a matter of time if it is not possible already. But
then that brings up another question: at whatever time it becomes
technologically feasible to use a computer to identify an author, will the
same sort of people who gave us our crummy OPACs be doing the programming,
or will we get a really useful product?
^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*
David Guion
Music Cataloger
University of North Carolina, Greensboro
Jackson Library
320 College Ave.
Greensboro, NC 27412
(336) 334-5781
dmguion_at_uncg.edu
The early bird may get the worm, but the second mouse gets the cheese.
Received on Thu Sep 06 2007 - 08:43:59 EDT