Print

Print


Bob,

What do we do retrospectively, if anything, when a second expression appears for one of those 85% or more of works?  At that point, we end up with a collocation problem, don't we?

For example, consider all of the translations that are made after an original language expression appears.  Here's a made up example:

100 1_ Surname, Forename, $e author.
245 10 Original French language expression of a novel, in its first manifestation / $c Forename Surname.

At this point, there's a single expression in the original French of this work.

Six months or a year later, an English translation is published:

100 1_ Surname, Forename, $e author.
240 10 Original French language expression of a novel, in its first manifestation. $l English
245 10 English translation title / Forename Surname.

There are now two expressions.  Do we go back to the original expression now to apply the instruction "If there is more than one expression of the work, record the expression manifested"?

What happens when a new expression in French appears, say with slight revisions by the author, or with annotations?:

100 1_ Surname, Forename, $e author.
245 10 New French language expression of a novel / $c Forename Surname ; annotated by A. Huge Fan.
250       Second corrected edition.

Since this is a new expression, I believe that Bob would record the language of expression:

240 10 Original French language expression of a novel, in its first manifestation. $l French [perhaps with other additions as well like date and or surname of annotator]

Now if we haven't gone back to the very first edition, we have a split file.  One French expression has an access point with $l French while another record is out there with no 240 at all and just the 100/245 combination.   Split file.  Does this matter?  I'm not sure.  But we need to carefully consider whether it matters and what its ramifications are.  I think the only way to avoid this potential of split filing is to always record language of express in every record that is textual in content.  But if we don't have to rely on pre-assembled AAPs, and use identifiers for every expression manifested, that may mitigate the problem.  It will require creating a lot more identifiers though!  Are we ready to create an authority record or at least some URI for every work and every expression of a work?

Adam Schiff
University of Washington Libraries


From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of Robert Maxwell
Sent: Thursday, April 27, 2017 11:48 AM
To: [log in to unmask]
Subject: Re: Queston about "Works" authority record with qualifier

Actually, not so. The core requirements for the relationship is found in 0.6.8:

When recording primary relationships between a work, expression, manifestation, and item, include as a minimum the work manifested. If there is more than one expression of the work, record the expression manifested.

So at least as far as RDA is concerned, if there is only one expression, the relationship of the resource to the work is core, i.e., the description has to indicate what the work is. I gather that something like 85% or more of all works exist in only one expression. For these, all that would be required is recording the work relationship, which, as Adam has pointed out, we record in a bibliographic record by including a 1XX (if appropriate) + 245 $a, so long as 245 $a contains the preferred title (i.e. the 1XX + 245 $a = the authorized access point)-which it does in nearly all cases where a work has only appeared in one expression. In those few cases where 245 $a doesn't contain the preferred title the preferred title for the work is recorded in 240.

It is only when there is more than one expression that the second sentence kicks in "If there is more than one expression of the work, record the expression manifested". In this case RDA expects us to identify the expression. Since we're using authorized access points to identify works and expressions, the only way to do this is to add something to the authorized access point for the work (6.27.3). I've argued that for textual works the logical thing to add is first, language, whether it's the original language or a translated language, and then add more if necessary to further distinguish the expression. In any case, to fulfill the RDA requirement we need to explicitly record the expression if there is more than one expression.

Since only a rather small minority of works exist in more than one expression, it's not at all correct to say that if we started implementing this "that *every* RDA bib record will require a 130 or 240 field". Only the small minority that exist in more than one expression would require this. I can't see why we don't think we could handle this.

Bob

Robert L. Maxwell
Ancient Languages and Special Collections Librarian
6728 Harold B. Lee Library
Brigham Young University
Provo, UT 84602
(801)422-5568

From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of Adam L. Schiff
Sent: Thursday, April 27, 2017 12:23 PM
To: [log in to unmask]<mailto:[log in to unmask]>
Subject: Re: Queston about "Works" authority record with qualifier

Yes, I've made this point many times.

Adam

Get Outlook for iOS<https://aka.ms/o0ukef>
________________________________
From: Program for Cooperative Cataloging <[log in to unmask]<mailto:[log in to unmask]>> on behalf of Kevin M Randall <[log in to unmask]<mailto:[log in to unmask]>>
Sent: Thursday, April 27, 2017 8:06:55 AM
To: [log in to unmask]<mailto:[log in to unmask]>
Subject: Re: Queston about "Works" authority record with qualifier

But if the policy were to always include the language in the expression AAP, wouldn't that mean that *every* RDA bib record will require a 130 or 240 field? Or would we do it only for things that have expression authority records? While I'm not terribly bothered by a certain (limited) amount of inconsistency in the bibliographic database due to evolving standards, this seems to be a bit much.

Kevin
(who is getting even more impatient waiting for the day when identifiers take the place of AAPs)

Kevin M. Randall
Principal Serials Cataloger
Northwestern University Libraries
Northwestern University
www.library.northwestern.edu<http://www.library.northwestern.edu>
[log in to unmask]<mailto:[log in to unmask]>
847.491.2939

Proudly wearing the sensible shoes since 1978!

From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of Stephen Early
Sent: Thursday, April 27, 2017 9:06 AM
To: [log in to unmask]<mailto:[log in to unmask]>
Subject: Re: [PCCLIST] Queston about "Works" authority record with qualifier

Bob wrote:

"... Whew. All that being said, "Lonergan, Bernard J. F. $t Works $s (Lonergan Research Institute)" is a perfectly fine AAP at the expression level, but the presence or absence of subfield coding being the only clue that it's at the expression level and not at the work level seems pretty thin to me and must be quite opaque to our users (if not most catalogers). (This comment applies to any work/expression AAP, not just ones involving the conventional collective title "Works".)

This is one reason I favor routinely including the language for textual expressions, even for original-language expressions: "Lonergan, Bernard J. F. $t Works. $l English $s (Lonergan Research Institute)". The inclusion of the language makes it completely clear that the AAP is for an expression, especially in the absence of any marker in the MARC authority format that explicitly distinguishes work-level authority records from expression-level authority records.
"

I strongly agree with Bob! I suspect that there are more than a few catalogers, including NACO contributors, who are not yet able to intuitively distinguish between expression level and work level access points. _Always_ adding language to an expression level access point might seem tedious and redundant, but I think it may be necessary until the _majority_ of catalogers finally "get it." Maybe in 10 years a task group can do a study and determine that $l (or the post-MARC equivalent) can be omitted from original language expression access points. But for now I believe: if it's expression level, always include $l .

Stephen T. Early
Cataloger
Center for Research Libraries
6050 S. Kenwood
Chicago, IL  60637
773-955-4545 x326
[log in to unmask]<mailto:[log in to unmask]>
CRL website: www.crl.edu<http://www.crl.edu>