Let's add Spirits to that list of entity classes. Is the "Ghost of Christmas Past" a spirit or a fictional character though? Since it's in A Christmas Carol, I guess it's a fictional spirit as opposed to an actual real spirit. ;)
Adam Schiff
University of Washington Libraries
-----Original Message-----
From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of Kevin M Randall
Sent: Tuesday, November 28, 2017 4:34 PM
To: [log in to unmask]
Subject: Re: LRM, RDA, and attributed authors
Stephen, thank you so much for your clear analysis of the situation and suggested solution to the problem. This would appear to be a quite satisfactory compromise between the AACR2/RDA tradition and the LRM requirements. I fully support everything you wrote.
I'm wondering, though, if all of these non-agents that don't fit into other entity classes will just be instances of "Res", or if a new class(es) should be defined in RDA. ("Fictional character", "Nonhuman being", etc. I could see an argument for just letting them be "Res", because where would you draw the lines--how many classes would need to be created? On the other hand, additional entity classes such as "Fictional character", etc. would allow for defining attributes specific to the classes.
Kevin M. Randall
Principal Serials Cataloger
Northwestern University Libraries
Northwestern University
www.library.northwestern.edu
[log in to unmask]
847.491.2939
Proudly wearing the sensible shoes since 1978!
> -----Original Message-----
> From: Program for Cooperative Cataloging
> [mailto:[log in to unmask]] On Behalf Of McDonald, Stephen
> Sent: Wednesday, November 22, 2017 6:21 PM
> To: [log in to unmask]
> Subject: [PCCLIST] LRM, RDA, and attributed authors
>
> In the thread 'King power, by Richard III', I suggested that LRM
> requires us to change the way we handle pseudonyms, spirit authors,
> etc. I would like to expand on that a bit. I apologize for posting
> this the night before a major U.S. holiday. I want to get this down
> before the topic dies and my memory of what I want to write fades.
> Our friends outside the U.S. should understand that many or most U.S.
> contributors will not be able to reply until Monday.
>
> I will start by looking at Geronimo Stilton. This is a fictitious
> character, a mouse detective, who is the attributed author of a series
> of books about his adventures. The NAF currently has 206 author-title
> AAPs for works and expressions with Geronimo Stilton listed as the
> creator. The series was created by Elizabeth Dami, and she is
> presumably the actual creator of the books, but I have not found
> confirmation of that. There is currently no authority record for Elizabeth Dami.
>
> RDA is being edited for a major update to incorporate the Library
> Reference Model. My understanding of LRM is that only agents can be
> creators, and agents must represent real human beings or groups of
> human beings who live or have lived. Specifically, the scope note for
> LRM-E6 (Agent) says:
>
> "Being an agent requires having, or having had, the potential of
> intentional relationships with instances of entities of bibliographic
> interest (works, expressions, manifestations, iems), whether that
> specific agent has ever done so or not. Human beings are directly or
> indirectly the motive force behind all such actions taken by all agents."
>
> LRM also defines creator, contributor, publisher, manufacturer, and
> distributor as relationships between a WEMI and an agent (LRM-R5
> through LRM-R11). So, in LRM, all of these persons must be agents, i.e.
> representations of real human beings. By implication, they must
> represent the actual person or persons who performed those roles.
>
> I don't know what the new RDA rules will look like. But the new rules
> will clearly require some marked changes in policy. From previous
> discussions on this list, I believe at least some members of the RDA
> Steering Committee feel that LRM does not require that every entry in
> a name authority file be an agent. That makes sense to me, too. That
> means that we can continue to have fictional and legendary characters in the NAF.
> But if we continue to use Geronimo Stilton as the AAP for the creator
> the children's mystery books, I see a big problem. If we redefine
> no2013042861 as a representation of the human author behind the books,
> we would conflate two different concepts--the creator of the books,
> and the fictional character who can be used in subject headings. It
> would get even worse if we ever see a biography of the author of the
> children's mysteries. It is confusing to use the same AAP for the
> fictional character and the author. To me, it is clear that under LRM
> we need two distinct AAPs, one for the fictional character, and one for the creator.
>
> The big argument under current rules for using Geronimo Stilton as the
> creator is discovery. People see Geronimo Stilton listed as the
> author of the books, and are likely to look for the books under that
> name. I agree that this kind of discovery is important. People
> should be able to search for a book under the author name which is
> listed on the title page. But under LRM, the creator must represent a
> real person, not a fictional mouse which can be used as a subject. So
> I believe it is important that we be able to record attributions in a
> bib record, in addition to an AAP for the real creators and
> contributors. The Geronimo Stilton books would contain an AAP for the
> author, and an AAP for the fictional character with a relationship designator identifying him as the attributed creator.
> Currently, RDA does not encompass attributed creators and contributors.
>
> Moving to Richard III, the subject of the previous thread, I believe
> that using an attributed creator relationship designator would greatly
> clarify the description of the work in question, and improve
> discovery. Again, we could have an AAP for the author, and an AAP for
> the medieval king with an attribution relationship designator. It
> would be clear to users that we understand Richard III did not really
> write the book, but it would still be discoverable under that name.
> Defining attributed creators and contributors could help in numerous
> situations, including ancient literature or literature of uncertain
> authorship. Classical works by pseudo- authors could include
> references for the attributed authors. We could describe a work both
> as it represents itself and as it really is, according to the available information.
>
> This also gives a good reason for retaining non-agents in the name
> authority file rather than the subject authorities. We can define in
> rules or policy that non-agents (but not subjects) can serve in
> certain non-agent roles such as attributed creator or contributor. We
> can define specific relationships for non-agents.
>
> As I said above, I don't know what the LRM update to RDA will look like.
> And after the update, there will be significant changes to LC-PCC policy.
> Here are some of the things I hope will result from the RDA update and
> subsequent LC-PCC policies. I am presenting these ideas to see
> whether other people agree:
>
> * Fictional characters, named animals, legendary characters, etc. can
> and should continue to be represented as AAPs, and included in the
> NAF, so that they can be recorded in non-agent roles. For the rest of
> this discussion, I will refer to everything in these categories as
> non-agent identities, as opposed to agent identities.
> * Authority records for non-agent identities should be clearly marked
> in a machine-actionable way as authorities which cannot be used in an
> agent role. Authority records for agent identities should also be
> identifiable in a machine-actionable way, either by a specific element
> value or by the lack of a non-agent element.
> * Authorized access points for all non-agent identities should
> include qualifiers which clearly mark them as non-agents. This will
> make it much easier for catalogers to identify bibliographic records
> which tag non- agent identities in the agent roles of creator or
> contributor. Using specific qualifiers will make non-agent
> identifiers easily recognizable to the eye, augmenting the
> machine-actionable element. As a side-effect, it will make clear to users that we understand that these are not real people.
> * Authority records for works and expressions should allow variant
> access points with the attributed author as the first element. For
> example, the authority record for Micekings could have a 100 with
> whatever AAP we create for the author as the first element, and a 400
> field with Geronimo Stilton as the first element. This would benefit
> both discovery and automated authority control.
> * RDA should be modified to include instructions on recording names
> of entities with non-agent roles.
> * Relationship designators for non-agent roles should be included in
> RDA. These should be separate categories from the current categories
> for creator, contributor, publisher, manufacturer, and distributor,
> and specifically open to both agent and non-agent identifiers.
> Non-agent relationship designators should encompass attributed
> creators and contributors.
> * Only agent identifiers should be valid for use in 100 fields.
> * Only agent identifiers should be valid for use with relationship
> designators for creator, contributor, publisher, manufacturer, and
> distributor.
> * Both agent and non-agent identifiers should be valid for use with
> relationship designators for non-agent roles, including attributed
> creator and attributed contributor.
> * Both agent and non-agent identifiers should be valid for use in 700
> fields which have no relationship designator.
> * NACO quality control scripts should ensure that all AAPs with non-
> agent qualifiers contain the machine-actionable element identifying
> them as non-agents, and vice versa.
> * NACO should use scripts to mark all author-title AAPs which use a
> non- agent identity as creator. Steps should be made to change all
> NAF records marked in this process. If necessary, AAPs for the true
> creator should be created, AAPs for works and expressions with
> non-agent identities in the first element should be moved to variant
> access points, and new AAPs inserted with the AAP of the human creator
> in the first element. (It is not necessary to know the actual name of
> the human creator to create an AAP.)
> * WorldCat should take advantage of the machine-actionable elements
> to flag bib records which use a non-agent identity in a 100 field or
> in a
> 700 field with a relationship designator for creator, contributor,
> publisher, manufacturer, or distributor.
>
> So, what do other people think? Am I completely off base? I am
> hoping to use this discussion to direct the eventual discussion of PCC
> policy, and possibly influence the development of the RDA update.
>
> Steve McDonald
> Cataloging and Metadata Librarian
> [log in to unmask]
|