David Runnion replies to me:

>>May I suggest that there are two types of rests (Ger. Pause; Fr. silence)
>>- exact and inexact? Exact rests must be strictly observed, inexacts not.
>>Much of it has to do with keeping the overall pulse of the music.  ...
>
>What's an inexact rest? Rests at the end of a piece are usually put there
>to make the thing work out mathematically, but not always.

I agree.  That's one type of inexact rest or, for that matter, an inexact
duration, sounding or not.

>will see a rest at the end of a piece with a fermata over it, to me that
>indicates that you must remain "in character" without breaking the musical
>mood for the thunderous applause.  I think those must be observed with
>their full duration.

What's the full duration of a rest with a fermata? I ask, because in the
first movement of Mahler's second, measure 4, there's a fermata over an
eighth rest in the cellos and basses, as well as a fermata over a whole
note in the first violins and violas at the same measure.

>Rests between sections (or before..ahem..those silly
>NO I WON'T SAY THE WORD signs) must be observed and counted like
>any other.

Some of them, yes.  Others, no. It depends: for example, the famous pause
between the first two movements of Mahler's 2nd - "Hier folgt eine Pause
von mindestens 5 Minuten" (here follows a pause of at least 5 minutes).
I've occasionally heard this live (sort of), although the audience wasn't
clued in, mainly because they haven't read the score, and the grand effect
of awed silence Mahler may have envisioned was obliterated by the usual
chat and moving about in the seats.  Of course, I've never heard it on a
recording, although I suppose you can program your own 5-minute pause.
I would say, both practically and from a reading of the score, this is a
prime example of an "inexact rest." The practical element of performance is
self-explanatory, and the direction (explicitly written out by Mahler) says
"von mindestens" - "at least," hardly exact.

Steve Schwartz