This Business of Interpretation

It's often said that no matter which interpretation of a physical 
theory we adopt, we get the same answers to practical questions 
(assuming we don't adopt a silly interpretation that is incompatible 
with the observed phenomena).  In view of this, we might think that 
very few working physicists would care much about interpretations 
of physics.

However, in point of fact, I've rarely encountered, either in person 
or in writing, a working theoretical physicist who was actually
disinterested in the interpretation of (say) quantum mechanics.  
Now, it's certainly true that very few physicists feel compelled to 
publish on the subject, or spend much of their time thinking about 
it, but surely the same is true for most mathematicians with respect 
to, say, the Riemann Hypothesis.  The vast majority of mathematicians 
have never published anything on the RH, and don't spend much (if 
any) time working on it, but this does not imply that they regard 
the question as either uninteresting or unimportant.  It just means
they've chosen to focus their attention on something else, usually 
something on which they can realistically hope to make some 
significant progress.

Even setting aside the factual question of whether working
physicists who "care about" interpretations of quantum mechanics
are few in number, I have to challenge the underlying implication
that interpretations of physical theories are unimportant and not
worthy of much consideration.  For example, consider the various
possible interpretations of Lorentz's electrodynamics.  It can
hardly be denied that Einstein's kinematical interpretation had
(and continues to have) a tremendous heuristic power, to the
extent that many people are tempted to regard SR as actually 
"true", because it seems to so powerfully and naturally embody 
the facts of Lorentz's theory.  

Needless to say, Einstein's is not the only possible interpretation 
that agrees with all the facts, but it seems so far superior to all 
the others that it's hard (for me at least) to avoid the impression 
that it is, in some non-trivial sense, true.  (Those unable to think 
of a non-trivial sense of "true" are invited to simply disregard that 
remark.)  In any case, the SR interpretation has been immensely 
useful and important for theoretical physics.  Similar comments 
could be made about the "curved manifold" interpretation of general 
relativity.  It isn't the only possible interpretation, and it may 
not even be unambiguously the "best", but it is undeniably powerful 
and heuristically useful.

Overall I think physics has historically advanced largely through 
attempts to rationalize our observations, i.e., to reconcile our 
raw experiences with some rational and *realistic* conceptual 
framework.  Admittedly this has sometimes involved adjustments 
to our sense of "realism", but it isn't a one-way street.  Our 
quo ante sense of realism is one of the most significant factors 
guiding us and shaping our theories.  Even in the field of quantum
mechanics, which has been the field of physics most resistant 
to complete realistic reconciliation, we nevertheless are guided 
by the "correspondence principle" and other essential links to 
our realistic roots.  QM might say of realism "Can't live with
it, can't live without it".

Needless to say, our interpretations are always provisional, but 
I think history has shown that some physical theories possess 
"realistic" interpretations that may not have been immediately 
obvious, but that can be very powerful and useful once they are 
conceived.  (I'm reminded Professor Witkowski's excited utterance 
to Professor Loria immediately after reading vol 17 of Annalen der 
Physic: "A new Copernicus is born!  Read Einstein's paper!")  This 
is why so many working physicists care so much about this business.