Dave Pitzer replies to me:

>My quibble was that you stated that there were two types of rests --
>internal and external.  Perhaps I misunderstood you original statement
>(actually, I hope that I did).

I hope I didn't say that.  I wanted to distinguish mainly between exact and
inexact.

>I think that we are not really disagreeing here.  I was under the
>impression that you (or perhaps someone else) had said that rests were
>"unimportant".  This is patently untrue.

Yep.  No, they're not unimportant, and, to hopelessly add to the confusion,
neither are repeats.

>>There are also notes that are held inexact durations, particularly in
>>20th-century music.  You want specific examples? How about the storm and
>>prayer scene in Gershwin's Porgy and Bess, where each character in effect
>>prays at his own speed?
>
>Isn't this more a matter of "tempo ad libitum" not rests?

Well, rests *and* note values and "tempo ad libitum." This was carried to
a constructive principle, by the way, in Britten's first church parable,
Curlew River.  Britten even invented his own notational symbol for the
score, the "curlew" (which looks like the icon of a bird in flight - sort
of a fanciful "m," since I'm confined to plain-text).  The curlew marks the
point where the all the parts come together and proceed in strict time.

Steve Schwartz