As an old-time serials cataloger, I'd like to add my two cents worth to
Aimee's communication about title changes.  As I recall, the theoretical
justification for successive entry cataloging (that a title change
indicates a change in editorial direction) was almost an afterthought.
The pragmatic reasons for the adoption of successive entry were to reduce
cataloging workload and improve user access.  (How can that be? you cry,
for successive entry seems like more work and more confusion!)  The
problem for catalogers with latest entry was that the cataloger needed to
describe the entire history of the publication, even though the issue in
hand was only for the current title.  You know how difficult it is now
just to figure out how the current title relates to the immediately
preceeding title--imagine trying to relate it to all previous titles!
Remember too that changing the entry meant redoing all the cataloging for
every former title, not just the current and immediately preceding title.
The problem for users was that the entry in the "catalog" did not
match the citation they had.  I say "catalog" in quotes because the problem
was most evident in single-entry listings such as kardexes, union lists,
and other lists of periodicals.  The question now is whether shared
cataloging and automated systems have rendered these pragmatic
considerations anacronistic.

The thought of merging all those successive entry records back together is
more than I can bear.  But I agree with Aimee that a more liberal
application of the fluctuating title rule is imperative.  As for "not
necessarily having to change any existing rules," let's not rule out the
need for more changes to the rules and/or rule interpretations in this
area.  When we revised the definition of a title change in AACR2r, we were
pretty conservative.  Perhaps some other categories of changes which do not
require new entries should be identified.

Shared cataloging has made it more important than ever to have some
national standards we can all agree to.  On the other hand, the enormous
variation in type and level of automation within libraries has made our
needs distinctive. For example, those with manual kardexes are less likely
to embrace earliest title cataloging than those with fully automated
systems with multiple access points.

None of this is meant to discourage Aimee's pleas to do some research
on the rationale behind title changes.  I just hate to see practioners
wail against theoreticians on this topic; in real life we did it to
ourselves!

Crystal Graham
Latin American Studies Cataloger
University of California, San Diego