Tom Warren wrote:

>How could a Bach fugue be "incorrect". I don't know how to write a fugue,
>but I certainly have played a few and they seem very correct to me.

Well, that's the difference between judging a fugue according to what
are known as "school fugue" rules and judging a fugue based on how well
it seems to hang together.  There are, believe it or not, "mistakes"
in a small number of Bach fugues - parallel octaves, parallel fifths,
"answering" voices not entering at the "proper" interval, and so on.  When
you're starting out to write fugues, mastering these rules actually helps
you because they limit the decisions you have to make.  Deciding takes
time and effort.  These things are generally due in a few days or even
at the end of a class hour.  On the other hand, when you write music for
real, very often you want to increase your options.  You do so at your own
risk, of course, since these "rules" are simply a compendium of effective
procedures, just as Rimsky's treatise on orchestration is a compendium of
effective sounds.  If you want new sounds, you takes your chances, and any
composers worth their salt do.  The only composer I know who wrote "school
fugues" for real was Satie.  They are delightful fugues and the constraints
are part of their artistic point.  In this case, the composer risks trying
to elevate a student exercise to a valid artistic form, while maintaining
the outward characteristics of the exercise.

Steve Schwartz