Dave Wolf writes ...

>Please-- this analogy between classical music composition and photography
>doesn't work.  ...

I think that it does work as an analogy. Reasonable men may differ.

>It sounds to me more like the analogy you want is an infinite number
>of monkeys banging away at typewriters in hopes of composing King Lear.
>I hope your program is more intelligent than that!

You are in luck, it is.  The themes are not mere random permutations of
notes.  They are constructed using melodic fragments based on degrees of
the scale like 1 2 3 4, 5 6 7 8, 1 3 5 8, 2 3 4, 1 6 (a skip), 1 2 1
etc.randomly chosesn from a pool of what would work, given the underlying
harmony, which is based on standard progressions general used by all
composers, and where the harmony is going.to.  Some of the resulting themes
are pretty good, if I do say so myself.  You are quite right, just randomly
picking notes and druations would have little chance of producing a good
theme.

>While Mozart may have had a genius for instinctively finding the
>melodies that met his intentions, Beethoven had to struggle to find
>the right choices--but in both cases, and I'm sure in most others,
>there was a meta-composition each composer was striving to fill out
>with melody, harmony, rhythm, etc.  If your program begins with such a
>meta-composition--either internally or supplied by the user, then there's
>a chance it can create decent music.

Yes, it does something similar (I think).  But I am not sure that they
used the meta-composition approach you speak of.  While a writer may start
with the overall plot and characters and fill in the dialog - though some
don't - I am not conviced by my study of Beethoven that he had such a thing
in mind.  The probable problem for both Mozart and Beethoven was what to
use of the several things that came immeiately to mind from from the last
thing.  There was the struggle, resolved by asking, which of these is the
most compatible with the composition so far.  Maybe if you could define
what you mean by "meta-composition" we could discuss this more.  It sound
like an interesting idea.

>And since the desired effects, at least in the music that is most
>appealing to CM lovers--18th to early 20th Century -- are EMOTIONAL
>effects, [insert - some on the list don't agree with this - we had a big
>discussion on it ] whether or not anyone is willing to name said emotions,
>if your program can do that, then you, too, would deserve the accolade with
>which Schumann greeting the appearance of Brahms (Or maybe it was Chopin),
>"Hats off, gentlemen:  a genius!" Good luck.

I thank you for wishing me great success.  It does employ some known
effects that evoke emotion like a the resolving of a sus 4th, the deceptive
cadence, the harmony change of a IV to IV min, dramatic melodic skips, a
theme played solo repeated by the full orchestra, the theme played in the
bass register with exciting runs over it, and many others.  BTW it was
Chopin.

I should make it clear to the list that I am not trying to compete with
the great masters, as they were not competing with each other.  I am
offering a way for interested persons to play around with classical music,
learn from it, get exposed to the issues involved, turn their own themes
into larger works, try different things, and hopefully understand CM
better.  I sometimes get the impression that this is threatening to some
folks, like I am trying to add another book to the Bible.  I think I am
doing a good thing for the field of music, but one head of a composition
department, actually hung up in my face without saying good by.  I said "I
have developed a software music lab that creates classical music and I was
wondering if you and your collegues would like to see a demonstration,
listen to the music, give me your thoughts and evaluate this for its merits
as a teaching tool.  She said "I am not interested in anything like that"
and hung up.  Its a good thing that I am not a sensitive person or I would
have had my feelings hurt:-) I just thought to myself "This is the head
of a composition department of a large university who is teaching young
composers how to be creative and won't give me 15 minutes to show her new
music composition technology - wow".

Bill Pirkle