Referential Relationships: RDA Chapters 24-28 and Appendix J

CC:DA/RBMS/2015/1/rev
July 16, 2015

Referential Relationships: RDA Chapters 24-28 and Appendix J

Earlier proposal: References to Descriptions (06/09/2015)

Submitted by Matthew Haugen, ALA/ACRL/Rare Books and Manuscripts Section

Print Friendly, PDF & Email

Tagged , . Bookmark the permalink.

14 Responses to Referential Relationships: RDA Chapters 24-28 and Appendix J

  1. Matthew Haugen says:

    Summary of significant changes from earlier version of proposal:

    updates to chapters 25 (related works), 27 (related manifestations) and 28 (related items) mirroring those initially proposed for ch. 26 (related expressions)

    New instructions at 27.2 and 28.2 for explanation of relationship to manifestation and item, respectively (mirroring existing instructions 25.2 and 26.2 for explanation of relationship to works and expressions, respectively)

    New relationship designators in J.2.3, J.4.3, J.5.3 parallel to those initially proposed for J.3.3

    updated definitions of related work, related expression, related manifestation, related item in 24.1.3 and glossary

    new definition of location within resource in glossary

  2. Steve Kelley says:

    Although I’m not a rare book cataloger, I appreciate the thinking behind this proposal and think it’s very well constructed. I’ve heard complaints from colleagues who catalog rare books that RDA does not currently address the problems and concerns that they face, and I think this does a fine job of filling that gap.

    • Amanda Sprochi says:

      I do rare book cataloging from time to time, and this is a really good proposal and way to get the reference information in a more FRBR-ized form. I like this proposal and the suggestion that we can continue to use the structured note (currently a MARC 510) or use the access point model.

  3. Kathy Glennan says:

    What is the impact on Appendix J relationship designators if we (for example) define a related expression as:
    an expression, represented by an identifier, an authorized access point, or a description, that is related to the work, expression, manifestation, or item being described

    All the existing Appendix J relationships are at the same WEMI level. This revision has much broader implications (I think) than just the designators listed in this proposal.

    • Dominique Bourassa says:

      And does this fit within the FRBR model?

    • Robert L. Maxwell says:

      I’d like to see the “broader implications” raised and resolved in favor of expanding the possible levels of relationship. I think the relationships need to be expanded so that it is possible to make appendix J relationships between different WEMI entities, not just the same. In my opinion any entity should be able to be related to any other entity; e.g., an expression should be able to relate to a work if appropriate.

      • Steve Kelley says:

        Like Bob, I would like to see the broader implication raised and resolved in favor of expanding the possible levels of relationship. I see Kathy’s point that it might result in the Appendix J stuff being killed by the JSC, but I think it’s worth bringing up the issue, even if I can’t offer a practical solution as a backup in case this fails. I was on the Task Force to Investigate the Instructions for Recording Relationships in RDA, and we kept running into this very problem of mixing relationships between WEMI levels, and found no good way to avoid it entirely.

      • Tina Shrader says:

        I agree with this, and I think that having the flexibility to make relationships across WEMI levels would facilitate mapping RDA data elements into new encoding schema like BIBFRAME. The differences in fundamental models between BIBFRAME and FRBR sometimes caused us (at NLM) considerable difficulty in figuring out how best to express RDA data in BIBFRAME properties. Opening up the relationship structure would, I think, alleviate some of those difficulties.

  4. Dominique Bourassa says:

    Is the first example in 26.1.1.3 “Recording Relationships to Related Expressions” right?
    26.1.1.3 Referenced in: Swedish imprints, 1731¬1833. Volume 57
    Can an expression have a volume number? Volumes are defined at the manifestation level.

  5. Kathy Glennan says:

    The Appendix J part of the proposal is likely to fail, unless there’s been a rethinking at the JSC level about the structure of relationship designators. Currently the relationship designators have an exclusive one-to-one reciprocal relationship. Thus, it’s a problem to have referenced in (work) with four reciprocal relationships:
    reference to (work)
    reference to (expression)
    reference to (manifestation)
    reference to (item)

    That doesn’t mean that we shouldn’t put this forward, but since it really is the entire point of the proposal, I wonder if we need a back-up plan.

    • Dominique Bourassa says:

      From John Attig

      Kathy identifies what I think is the critical issue raised by this proposal. She is correct that the structure proposed for Appendix J and the details of the reciprocal relationships are strikingly different from those currently in Appendix J.

      I believe that the changes proposed to the definitions of Related Work, Related Expression, Related Manifestation, and Related Item correct a major unintended error in RDA. The impact of this change will be dramatic, because every “Reciprocal relationship” statement in Appendix J needs to be rewritten in the form that is being proposed for the new relationship designators.

      My suspicion is that the relationships in both Chapters 24-28 and in Appendix J will need to be broken down into Work-to-Work, Expression-to-Work, Manifestation-to-Work, Item-to-Work, Work-to-Expression, Expression-to-Expression, Manifestation-to-Expression, Item-to-Expression, Work-to-Manifestation, etc. etc. etc. — and that all the reciprocal relationships will need to be rewritten in a new semantic pattern.

      This is an issue that I was planning to raise with the JSC Technical Working Group; it is something that I have been worrying about ever since I noticed the problem. I believe that this proposal is exactly what is needed to prompt work on this larger revision, because it includes a very clear example of why a new structure is needed.

      I also note that the JSC Relationship Designators Working Group is to undertake the task to “clarify the requirements of same-entity and cross-entity designators”. I believe that Gordon is referring to the same issue: whether Appendix J relationships are between instances of the same entity (work-to-work) or may also exist between instances of different entities (e.g., manifestation-to-work). This group may present relevant proposals at the JSC meeting.

      My expectation is that the JSC will undertake a discussion of the wider issues; action of this specific proposal may be deferred, but it will contribute to the wider discussion.

      Therefore, in spite of the significant issues that Kathy raises, I urge that this proposal go forward.

      John

  6. Matthew Haugen says:

    Right now, only the “primary” WEMI relationships and the work to WEMI as subject relationships are allowed across WEMI levels. There doesn’t seem to me to be a good reason for this restriction. Putting referential relationships aside for the moment, the current scope of related item (28.1.1.1) reads:

    A related item is an item, represented by an identifier or a description, that is related to the item being described (e.g., an item used as the basis for a microform reproduction).

    The reproduction relationship between facsimiles/reproductions of manuscripts or of specific copies of printed books seems to be instead an item to manifestation relationship.

    As for the one-to-many relationship structure: I can think of two possible ways of resolving it.

    Option 1. Give the referential relationships without qualifiers but still repeat them at each WEMI level in Appendix J, e.g.:

    J.2.3 Referential Work Relationships
    referenced in A work that contains a reference to the described work, expression, manifestation, or item. Reciprocal relationship: reference to
    reference to A work that is referenced in the described work, expression, manifestation, or item. Reciprocal relationship: referenced in
    not referenced in A work that does not contain a reference to the described work, expression, manifestation, or item. Reciprocal relationship: not reference to
    not reference to A work that is not referenced in the described work, expression, manifestation, or item. Reciprocal relationship: not referenced in

    And ditto for J.3.3, J.4.3, and J.5.3. Some of the relationships in CC:DA/TF/Relationship Designators in RDA Appendix K/6 appear to behave this way. E.g. “founder” and “founder of” can relate persons, families, and corporate bodies in a number of combinations but aren’t qualified differently for each possible combination when repeated in different parts of appendix K; while the designator doesn’t differ (e.g. founder of (family) vs founder of (corporate body), etc.) its domain and scope are constrained by the definition and location within the appendix.

    Option 2: Give a specific relationship for each possible WEMI-WEMI permutation, e.g.:
    J.2.3 Referential Work Relationships
    work referenced in (work) A work that contains a reference to the described work. Reciprocal relationship: work reference to (work)
    expression referenced in (work) A work that contains a reference to the described expression. Reciprocal relationship: work reference to (expression)
    manifestation referenced in (work) A work that contains a reference to the described manifestation. Reciprocal relationship: work reference to (manifestation)
    item referenced in (work) A work that contains a reference to the described item. Reciprocal relationship: work reference to (item)
    work reference to (work) A work that is referenced in the described work. Reciprocal relationships: work referenced in (work)
    expression reference to (work) A work that is referenced in the described expression. Reciprocal relationships: work referenced in (expression)
    manifestation reference to (work) A work that is referenced in the described manifestation. Reciprocal relationships: work referenced in (manifestation)
    item reference to (work) A work that is referenced in the described item. Reciprocal relationships: work referenced in (item)
    ditto for the “no reference to/not referenced in” in J.2.3 and for all in J.3.3, J.4.3 and J.5.3

    • Dominique Bourassa says:

      Comment from Robert Rendall, Principal Serials Cataloger, Columbia University

      Unattractive as it is, I think Matthew’s Option 2 or something like it should be included in the proposal in a way that will make it easy for the JSC to approve it, if they’re willing to, in order to get this through while broader implications are still unresolved.

Leave a Reply