Proposal: Works without titles

August 1, 2014

Works without titles



Print Friendly, PDF & Email

Tagged , , . Bookmark the permalink.

44 Responses to Proposal: Works without titles

  1. Robert L. Maxwell says:

    I haven’t had time to go through this whole document and may not have before next week’s deadline, but I wanted to say I wholeheartedly support the changes regarding choreographic works. The choreographer is the creator of such works and as such should be the first element in the authorized access point. I’m glad to see this change.


  2. Kathy Glennan says:

    Currently 6.2.2 has parallels with 6.14.2 and 6.19.2. Do the proposed changes for 6.2.2 need to be reflected in those subsequent instructions, or are they OK as is?

    • Dominique Bourassa says:

      I think the instructions 6.14.2 and 6.19.2 cannot be changed because the following sections are not divided between works created after 1500 and works created before 1501 like the instructions at 6.2.2. Therefore this information must remain in 6.14.2 and 6.19.2.

  3. Kathy Glennan says:

    Initial reaction to, Option A: I don’t like the new sentence under the Classical Byzantine Greek Works exception — for all users of RDA will Greek always be in a non-preferred script?

    This also raises questions with the title of the second exception below it — “… written neither in Greek nor in a preferred script…”. Thus Greek is a separate category, but it’s also a non-preferred script (based on the reference to the proposed

    The formatting of the PDF here is unfortunate (it’s a “created on a PC, converted to PDF on a Mac” problem). It’s difficult to understand if the final two paragraphs are part of the exception, or part of the main instruction. I’ll follow-up with LC to get an answer on this.

    • Kathy Glennan says:

      Dave confirmed my hunch that the final two paragraphs proposed for (either option) are *not* part of the exception.

      • Kathy Glennan says:

        Perhaps these paragraphs should be moved above the exception?

        • Allison Hausladen says:

          I think those 2 paragraphs should be moved above the exceptions. However, the 1st portion of the last paragraph could perhaps benefit from being made an exception as the use of “original language” is very confusing for non-textual resources (e.g., paintings, sculpture). See below:

          If a title in the original language is not available in modern reference sources or in resources embodying the work because there is no original language or such a title cannot be found, see

          I think it works fine in some sections of this rewrite where it’s indicated it’s the original language by which the work has become known, but in some places, it just says the original language which implies of the resource itself, but non-textual resources lack language.

    • Robert L. Maxwell says:

      In response to Kathy’s initial question, I think the new sentence under Option A is OK because it begins with “if” — if the agency prefers Roman script, then they transliterate the Greek title into Roman script; but of Greek is the preferred script of the agency, then the found Greek title will *not* be found in a script that differs from the preferred script of the agency, and so won’t apply. However, I don’t think the new sentence is necessary given the new paragraph at the end (whatever its placement).

  4. Kathy Glennan says: I’m concerned about the proposed 2nd paragraph:
    This instruction applies to works when the application of– and does not result in choosing a preferred title.

    The reference to (revised) confuses me: I think you’ve already determined the preferred title when you get to, but then that instruction allows you to transliterate it.

    The other RDA uses for “XXX in a non-preferred script” are all about names (and mostly about choosing among different forms of the name):
    RDA: Names Found in a Non-preferred Script
    RDA: Different Forms of the Same Name
    RDA: Names Found in a Non-preferred Script
    RDA: Names Found in a Non-preferred Script

    Based on a closer look at the structure of those chapters, “Recording the [element]” comes before the instruction for dealing with non-preferred scripts. Thus, I think that LC’s proposed should follow Alas, this could create quite a renumbering mess….

  5. Kathy Glennan says: While I understand the final “or” in the 1st paragraph, “or reference sources do not contain a title for the work in the original language”, there’s a possible problem here in applying this to resources that don’t have language (like a painting, sculpture, etc.). Is there some way to address this situation as well?

    • Allison Hausladen says:

      I’m realizing that is filled with the confusing wording as I noted above. I suspect that it’s not reading quite as was intended. Here are thoughts from experienced image catalogers who are part of ARLIS/NA’s CAC.

      Reply from Liz O’Keefe, Director of Collection Information Systems, The Morgan Library & Museum:

      “Maybe there should be a distinction between language material and non-language material. With the former, there is a presumption that a title will appear on the item when published–whether given by the original author or not is immaterial, it’s the title that predominates on manifestations of the item. Since titles are less likely to be given to a non-language work by their creator, and most often won’t appear on the manifestation, it’s better to rely on reference sources.

      I’m having a hard time wrapping my mind around this. If read a certain way, it sounds as if the only two options when dealing with works of art are and, because there is never a title in the “original language” for a non-language work. This rules out the use of perfectly good titles written on the art work by the artist himself, since they are not in the original language of the work. Similarly, the work may have come to be known by a particular title, but since that is not the “original language” of the title, that doesn’t apply. I don’t think it was the intention of the drafters of the proposal to rule these possibilities out, but by shaping the instructions according to “original language”, that is the effect. Since the concept of “original language” is useful for language material, it should be retained, but the scope of the instructions dealing with original language should be limited to language material, and visual works should be dealt with in a separate section, akin to choreographic works.”

      • Allison Hausladen says:

        Reply from Marie-Chantal L’Ecuyer-Coelho, Bibliothécaire, Direction du traitement documentaire des collections patrimoniales Bibliothèque et Archives nationales du Québec, ARLIS/NA CAC member:

        First, I agree with Liz: for documents which content is non-verbal, I think it makes sense to rely on reference sources. For the same reason, it would be important to ask for the exception under AACR2 22.1B to be reinstated in RDA. At the moment, if I was to apply RDA’s instructions strictly, I would have to revaluate all our artists authorities to give preference to the way they identify themselves on their works rather than the form of their name found in reference sources … Most artists solely use their surname or initials when they sign their works … I don’t understand why I should invest so much time in modifying authorities to give priority to forms of name that are less precise

        I personally find the distinction between «Titles from References Sources» ( vs «Devised Titles» ( extremely confusing when speaking about non-verbal works … The nuance between the examples of The demoiselles d’Avignon vs L’après-midi d’un faune is rather subtle … Both titles were found by cataloguers in reference sources and were attributed to the works over the course of time (almost ten years after the work was finished, in the case of the Demoiselles …). So why should they appear under two different instructions? In fact, under, it is clearly stated that « this instruction applies to works for which titles are not found in resources embodying the work or in reference source » so perhaps the Nijinnsky example should be transferred under

        As far as visual works are concerned, my understanding is that works that are sufficiently famous to appear in reference sources will probably be recorded in the language used by those sources (i.e. most of the time, in a translated form, e.g. Carré blanc sur fond blanc or White on white for the painting by Malevitch); otherwise, we would simply devise a descriptive title in the language preferred by the agency. I tend to agree with this practice, as I think: 1) it will spare cataloguers a lot of research, 2) it is, I believe, in line with CCO practices, and 3) those titles are more likely to be familiar to our users (I doubt many of them would recognize the transliterated form of the original title: Супрематическая композиция – Белое на белом (Белый квадрат). This being said, I agree with Liz: visual works should have their own subsection under (like choreographic works).

  6. Kathy Glennan says: How do compilations without a collective title fit in? Should they be explicitly excluded (a string of titles constitutes the name of the compilation, right?). But the compilation as a work itself might be perceived as not having a title.

    Note that RDA uses the term “titles proper” in this situation….

  7. Kathy Glennan says: I believe “references” should be singular in the title of the instruction.

    In the term is “in reference sources”; in, the term is “modern reference source”. I’m not concerned about the singular vs. plural; however, I wonder if “modern” should or should not be used in both cases.

  8. Kathy Glennan says:

    General comment on the instruction titles: should “preferred title” be used consistently, instead of just “title”? In the current 6.2.2, RDA regularly uses “the preferred title” and never uses just “title” in an instruction name.

  9. Kathy Glennan says:, Option A, 2nd paragraph:

    I can’t easily locate an instruction in RDA that uses a see reference to exceptions with the suggested phrasing. I think something like the following might be better:

    For anonymous works created before 1501 and written in a non-preferred script, apply the exceptions in

  10. Dominique Bourassa says:

    The proposal say it has “no impact outside of Chapter 6.” However, if this is accepted an example at will need to be changed or removed:

    Description of: Swan lake (Choreographic work)
    Resource described: Swan lake / Ann Nugent. A synopsis and history of the ballet

  11. Kathy Glennan says:

    Change 2 ( – relocating the instruction about alternative title: “Do not include an alternative title as part of the preferred title.”

    We should decide if this belongs better at (General Guidelines on Choosing the Preferred Title) or at (Recording the Preferred Title for a Work)


  12. Dominique Bourassa says:

    I am not quite done reading this proposal but here are my comments about choreographic works:

    First, like Robert, I am really glad that this proposal finally treats choreographers as creators in their full rights I am also glad that the RDA/ONIX working group plans on adding a new content type for performed movement). Second, I doubt the entire dance library community will be as enthusiastic as I am. I suspect the New York Public Library and members of the Dance Research Coalition will not support this proposal.

    Just a note: if this proposal is accepted there will be over 21,000 authority records that will need to be changed. Also, authority records that do not include the choreographer’s name but are meant to represent a group of ballets based on the same story, such as Nutcracker (Choreographic work) and Giselle (Choreographic work), will have to disappear. This alone might make some librarians unhappy. It will make subject analysis of books about such ballets a bit more complicated.

    One of my concerns is that this proposal seems to treat all choreographic works as “non-[self]-describing Resources.” Music in itself is also non-self-describing. However, we can take information from “resources embodying the work or reference sources” (RDA This is probably considered normal because music manifestations abound. But this proposal does not acknowledge the fact that manifestation of choreographic works also exist. The statement “when … resources embodying the works are not available (e.g., no manifestations of the work are known to exist)” gives the impression that we think that there is never any manifestation of dances. As a result, all examples of choreographic works appear in section or that treat works without titles.

    However, there are many choreographic works that are notated, and many more that are available in recorded formats and therefore their titles should be chosen following the instructions at Obviously, one can follow these instructions and not go farther but it would be nice to include an example of a choreographic work in section to at least show catalogers the possibility that we do have manifestations of dances. Here is one manifestation of a dance that could be used to create such an example. Note that this type of dance notation is often catalogued as score.

  13. Dominique Bourassa says:

    Comments from the dance library community

    From Beth Kerr, Theatre and Dance Librarian, Fine Arts Library, The University of Texas at Austin

    I think the change is a good one.

    From Sandi Kurtz, Seattle, WA

    Very much in favor of listing by choreographer — it’s always bugged me that we don’t.

  14. Dominique Bourassa says:

    Comments from Morris Levy, Senior Music Cataloger, Northwestern University Libary

    In general, the proposal is very sound and I agree that choreographers should be credited as creators of their works in the usual way. I do have a few concerns, though:

    1) Past practice was that titles were recorded in the language of the agency as appropriate (e.g. “Cinderella” for “Cendrillon,” “Rite of spring” for “[Le] Sacre du printemps,” “Afternoon of a faun” for “[L’]Après-midi d’un faune,” etc.), I assume for purposes of collation in catalogs. The note in the proposed for “L’après-midi d’un faune” mentions “English-language reference sources” which could be used as a continuation of this practice. Is that really what we want? Albert, Massine, and Nijinsky created titled choreographic works – shouldn’t we prefer the titles by which they knew their works over how they appear in English-language reference sources?

    2) How do we credit choreographers who based their work on a previously existing ballet, e.g.:

    Coppélia (Choreographic work : Saint-Léon)
    Coppélia (Choreographic work : Petipa, M after Saint-Léon)
    Coppélia (Choreographic work : Cecchetti and Ivanov after Petipa, M)
    Coppélia (Choreographic work : Sergeev, N after Cecchetti, Ivanov and Petipa, M)
    Coppélia (Choreographic work : De Valois and Sergeev, N after Cecchetti, Ivanov and Petipa)

    I think we should give credit to the creator of the “new” work with a reference to the previous work. Likewise, the VAPs would not include a reference to the previous work. Perhaps examples of this should be added to, e.g.:

    Coppélia (Choreographic work : Saint-Léon)
    Coppélia (Choreographic work : Petipa)

    3) I think we should also address the use of forename initials in the VAP. Typically they were added to differentiate choreographers in general, not because there was an actual conflict (e.g. Marius Petipa “Petipa, M” didn’t write ballets with the same titles as Lucien Petipa (“Petipa, L”) or Jean Petipa (“Petipa, J”) but the initial was recorded anyway). Would it be appropriate not to use initials at all even when there could be confusion, e.g.:

    Laitière suisse (Choreographic work : Petipa and Petipa)

    I may have more thoughts but these came to mind immediately.

    • Kathy Glennan says:

      Comment on Morris’ point #1 above: if the choreographic works have a title, then they aren’t covered by this instruction.

      I wonder if his question indicates that an example for a titled choreographic work should be added in the appropriate location, perhaps in (and in 6.27 somewhere?).

      If I’ve missed the point in #1, I’d appreciate some additional clarification.

      • Kathy Glennan says:

        Comment on Morris’ point #2 above — it would also be possible to relate works using chapter 25.

        • Kathy Glennan says:

          Comment on Morris’ point #3 above — I suspect that the VAP instructions are present so that the “old way” of providing access to choreographic works would still be available as a variant. I suspect that the inclusion/omission of forename initials would be more appropriate as an application decision rather than as a specific instruction in RDA.

      • Dominique Bourassa says:

        I think Morris’ comment #1 show how this proposal might be a bit confusing in that, as I said above, it seems to consider all choreographic works as not-self-describing, therefore as not having titles. If a dance is not embodied in a manifestation — if we don’t have a notated version or recording of the dance with a title written on it — when do we consider a dance to have a title or not? Massine’s and Nijinsky’s works might not have survived in notated or moving image forms, still they have titles. A reference source in English might translate these titles into English.

        I think adding an example in would help (I gave a link above to a possible example of a manifestation for a notated dance). Still, I think confusion will remain. People have been used for the last 20 years to look for instructions for choreographic works in one place (LCRI 25.5B). At first, at least, there will not be this reflex of thinking that one can look at different places for information. Someone searching the toolkit for “choreographic works” will be brought to the section; if he/she looks for “choreographic work,”, will be retrieved. One might not think to look before under instructions for works with titles. It will seem that the only thing one can do is devise a title or use a title from a reference source. To remove confusion, could we add a note to instruction (at the end of the “choreographic works” section) that says something like (even though it might seem silly), for choreographic works with known titles, see But then again, if we go to, we read:
        “For works created after 1500, choose as the preferred title the title in the original language by which the work has become known either through use in resources embodying the work or in reference sources”
        And this brings us back to the fact that if a resource is not embodied in a manifestation, we have to go back to reference sources in English that might translate the title into English.

    • Robert L. Maxwell says:

      Responding to Morris Levy’s question, I should think the preferred title of a choreographic work (which is the second part of the authorized access point) would be chosen in the same way as the preferred title for a musical work (, that is, the choreographer’s original title in the language in which it was presented.

    • Dominique Bourassa says:

      Comments from Andrea Cawelti, Ward Project Music Cataloger, Houghton Library, Harvard University

      Morris put his finger on the specific aspect of constructing choreographic titles which always bothered me, which is that of preferring the language of the cataloging agency. In the same way that we go to the trouble to ascertain the title under which a work is first published, in choreographic works why not aim for the title under which the work was first performed? Perhaps on p. 21, preferring a reference source in the language of the premiere might be preferable when not burdensome?

      To respond to Morris’ second point, in terms of crediting choreographers who based their work on previously existing choreography, I wonder if including separate VAPs for the previous work title, and for the new work title (if I’m understanding this correctly) would lose us an opportunity to express directly the relationship between the two? I also wonder if putting a VAP for the original work title with choreographer might bring in inaccurate references, but this wouldn’t be the first time.

      In change 6, on p. 28, would it be advisable to add a choreographic title example to the revision of Many choreographic titles as established include alternative titles, and it wouldn’t hurt to remind everyone that this isn’t necessary unless there is a conflict to be broken.

      Now it’s late, and I’m half blind, but one issue which will be complicated by preferring a direct creator to work relationship for choreographic works will be that of joint choreographers. Perhaps this was discussed in the 2013 meeting under the topic of “there are many possible relationships” but perhaps some guidance might be advisable in proposal 7 in the collaborative works area? Just to cause more trouble.

      • Dominique Bourassa says:

        Comments from Morris Levy, Senior Music Cataloger, Northwestern University Libary

        The comments from you [Dominique] and Bob should be adequate to clarify Kathy’s confusion about my first point. I found your comments especially helpful in describing the complications of identifying choreography as a work.

        Addressing Andrea’s comment on my second point: it’s possible that she (and Kathy) misunderstood the purpose of my examples. It’s really the second example that was significant because it no longer included a reference to the “original” choreographer. This is what I was envisioning:

        100 1 Petipa, Marius, ǂd 1818-1910. ǂt Coppélia
        430 0 Coppélia (Choreographic work : Petipa, M after Saint-Léon) ǂw nnea
        430 0 Coppélia (Choreographic work : Petipa)
        500 1 ǂi Based on (work): ǂa Saint-Léon, Arthur, ǂd 1821-1870. ǂt Coppélia ǂw r

        To address Andrea’s other point: I think joint choreographers would be treated just like joint authors of collaborative works in, e.g.:

        100 1 Petipa, Jean Antoine, ǂd 1796-1855. ǂt Laitière suisse
        400 1 Petipa, Marius, ǂd 1818-1910. ǂt Laitière suisse
        430 0 Laitière suisse (Choreographic work : Petipa, J and M) ǂw nnea
        430 0 Laitière suisse (Choreographic work : Petipa and Petipa)

    • Allison Hausladen says:

      Reply to Morris Levy’s 1st comment from Sherman Clarke, ARLIS/NA CAC member:

      “We of course have been using English-language reference sources for art work uniform titles and AAPs, whether pre-1501 or post-1500. Is there a principle in this proposal that would mean a shift in art AAPs if choreographic works shifted to the titles given by the artists? I recognize that the titles in question are often determined by scholars or the repository rather than the creator. When LCRI 25.3A was written, it grew out of the English-language preference in subject practice.”

  15. Robert L. Maxwell says:

    I am a bit concerned about the wholesale deletion of the instructions for naming manuscripts and folding them into a more general instruction about works without a title. I do think the instructions about naming manuscripts need to be looked at carefully (I commented on this when we were looking at the RDA drafts), and the current instructions are confusing and inaccurate in some places. Perhaps as an interim step this change is OK but I’d like to see a more careful look at these specific instructions, aand again, I’m concerned about just lumping manuscripts in with other kinds of works without titles. I think they have some special problems, including the repository designation type strings, which are unique to manuscripts, I think, and have problems of their own, e.g., the necessity of changing them if the repository changes. I agree with this necessity (the string is sort of an “address” of the manuscript and needs to stay current) but it’s something that’s unique about manuscript names/AAPs and so I think folding the manuscript instructions into a more general instruction may have unintended consequences. I’d prefer to keep them separate (but we DO need to have a look at them and consider clarifying them).

    Note: In the examples, the name of Browning’s work is Sonnets *from* the Portuguese, not Sonnets *for* the Portuguese (both in the example itself and in the explanatory note).


  16. Robert L. Maxwell says:

    I would like to comment that I do like the addition of the footnote to Works Created before 1501. It does explain well why there’s a different instruction for this group of works. It could be put even more succinctly: “Modern reference sources are considered a preferred source of information for works created before 1501 because the earliest form of title for works of this type is almost never the one by which the work is now commonly known.”

  17. John Myers says:

    My reading of the proposal was that it makes sense and the overall thrust of the changes are beneficial, with the caveat that my expertise in Pre-1501, Cycle, and Manuscript works is non-existent.

    I did wonder though how “Cycles and Stories with many versions” would now be addressed in the rules. My sense is that they would be rolled into “Works without Titles” (?), but in any case, it would be desirable to see those examples incorporated someplace in the revision proposal.

  18. John Myers says:

    Looking a little farther afield, the Background statement states “further development of chapter 6 was needed to enable the identification of works that are untitled and/or embodied in resources that are not self-describing.” This makes me wonder about the case of illustrations that accompany a published text, in particular different illustration sets by different artists for different editions of “classic” children’s titles, e.g. Wind in the Willows. These seem analogous to the case addressed by the section on Choreographic works in the reformulated, but is not adequately covered by which includes Illustrations in the caption but does not cover the case of illustrations in the actual rule (and in any case addresses the creation of the access point, not selection and recording of the preferred title). Is the main instruction at adequate to this situation? (In which case, I’m a little confused then why Choreographic works would need a separate section under

  19. Kathy Glennan says:

    Response from Arlene Yu and Jan Schmidt, Jerome Robbins Dance Division, New York Public Library

    We want to begin by saying it is gratifying to see the formal recognition of choreography as a work and of choreographers as creators of works. This is a goal which the Jerome Robbins Dance Division has striven to achieve for the dance community since our inception in 1944, one in which we are proud to have played a part.

    We are also in support of identifying choreographers by their first and last names, rather than merely by last names with occasional first initials, in the interest of clarity and completeness.

    In our continuing role as an advocate for the dance community, which includes both practitioners and scholars as well as enthusiasts and those just beginning to learn about the art form, we must question some of the assumptions made in 6JSC/LC/30 (http://www.rda‐‐LC‐30.pdf). First, as already stated by Dominique Bourassa on the CC:DA site (‐1969), there is an assumption in 6JSC/LC/30 that choreographic works always require devised titles: “It was also clear that the expanded instructions should apply generally to all works without titles, not just choreographic works—the same conclusion reached in 2012 when 6JSC/CCC/6 was discussed….” Indeed, the inclusion of examples of choreographic work titles only in the sections on devised titles reinforces this impression. However, “resources embodying the work or reference sources” do indeed usually exist, and we would request that
    6JSC/LC/30 be revised to reflect that fact. Choreographic works are generally titled works, certainly after their first performance.

    Second, we must point out that the origin of the current LCRI 25.5B rules
    ( #Choreographic_2._Uniform_titles), to which we contributed, lies in the recognition that collocating works by work title is reflective of dance community practices and understanding regarding the lineage
    of works. As we stated in our response to 6JSC/LC rep/4 in October of 2013,

    The idea was that titles of dance works should file together, so that users would find all of the versions of a particular work, in one list, regardless of whether the choreographer was known. This was based on the way that patrons actually search for these works. Instead of being “odd and unprincipled” as is alleged in the discussion paper, these headings were created in this way based on the principle of collocation by title and the way that patrons could reasonably be expected to search for these works. This “principle” is also found in Section 2 of RDA, “Identifying works and expressions” 6.0 Purpose and scope:

    Authorized access points representing works and expressions can be used for different purposes. They provide the means for:
    A. bringing together all descriptions of resources embodying a work when various
    manifestations have appeared under various titles
    B. identifying a work when the title by which it is known differs from the title
    proper of the resource being described
    C. differentiating between two or more works with the same title
    D. organizing hierarchical displays of descriptions for resources embodying
    different expressions of a work

    The bottom line is that a patron who wants to locate all of the various expressions of the ballet titled “The Nutcracker” should be able to locate them in one list, without having to know which choreographer’s work they are looking for….[and] in the current catalog environment, forcing patrons to search for these performances by first knowing the author would unnecessarily hinder access.

    Similarly, the removal of examples of the “distinguishing characteristic of the expression” (RDA which involve an acknowledgment of the lineage of a choreographic work:

    Nureyev, after Vaĭnonen
    A version of the choreographic work The nutcracker choreographed by Rudolf Nureyev in 1967 and derived from Vasiliĭ
    Vaĭnonen’s 1934 version

    hinders discoverability and context for patrons whose understanding of the work involves precisely those characteristics. If these examples and formulations are going to be discarded, then the information must find its way into the authority records for these titles, to aid patrons in identifying works exactly.

    We understand the desire to simplify rules for catalogers. We also understand that implementation of RDF triples and linked data may in the future allow for more sophisticated searches and/or indexes so that the order of elements in an authorized access point may become immaterial. Leaving aside the question of the tens of thousands of authorized access points which would need to be modified
    according to 6JSC/LC/30, we must ask: in our zeal to create simpler cataloging rules are we in fact reducing usability for the very people for whom we are cataloging? Why are musical, legal, and religious works granted exceptions while choreographic works are not?

    Much as we would wish otherwise, choreographic works continue to be outside the realm of most catalogers’ expertise, seekers of those works have unique needs and understandings of those works, and thus we believe choreographic work titles continue to merit a separate section in Chapter 6. As we’ve stated previously, we would be happy to contribute to the creation of such a section.

  20. Dominique Bourassa says:

    First, I must say I really am grateful that the NYPL was able to share their opinions. I think it is important to hear from the people who work with this type of material the most.

    I do not share the views of NYPL that choreographic works should be collocated under title but I do recognize that there is value in doing so for certain users. I think the view of the NYPL is one that is very ballet centric and might not be as useful for other forms of dance, for example, abstract dances. The current way of identifying dance works does not work as well either for works by choreographers who take on many creative roles such as Alwin Nikolais. Considering the fact that Nikolais created the choreography, sound, costumes, and lighting for Imago, I don’t think
    Imago (Choreographic work : Nikolais)
    is as useful as
    Nikolais, Alwin. Imago

    I think it undermines Nikolais’ creative role. I realize that, by following the instructions in this proposal, if the sound/music of a work by Nikolais was published separately, then we would need to distinguish the notated and performed music from the notated and performed movement but that is easy to do with additions to access points. It would much more logical than what we do now.

    Since RDA is meant as an international tool, I just want to point out that even the Bibliothèque nationale de France catalogs choreographic works under names of choreographers. If you do a search for the Après-Midi d’un faune in VIAF, you will find Nijinsky’s work is identified as:

    Nižinskij, Vaclav Fomič 1889-1950 L’après-midi d’un faune ‎

    Personally, I don’t see anything wrong with this. I would argue also that some researchers who do research on a particular choreographer might actually like to see a list such as this one (again from BNF in VIAF):
    Balanchine, George 1904-1983 Apollon Musagète ‎
    Balanchine, George 1904-1983 Agon ‎
    Balanchine, George 1904-1983 La chatte ‎
    Balanchine, George 1904-1983 Jack in the box ‎
    Balanchine, George 1904-1983 The prodigal song ‎
    Balanchine, George 1904-1983 The triumph of Neptune ‎

  21. Kathy Glennan says:

    Response from the Dance Heritage Coalition

    As an alliance of institutions holding major dance research collections, Dance Heritage Coalition (DHC) is keenly interested in the proposed change to cataloguing rules for treating choreographic works. Our members welcome and appreciate the recognition of choreographers as creators of works, an issue of strong importance to the dance community.

    However, DHC has some concerns about how the proposed rule change will affect the discoverability of dance records, as well as the accurate description of choreographic works that exist in many variant expressions. The current rules laid out by LCRI 25.5B ( reflect the needs of dance scholars: allowing all versions of a work such as Swan Lake to be collocated in a single place, while clearly distinguishing where choreography has been derived from an earlier version:

    Swan Lake (Choreographic work : Ashton after Ivanov and Petipa, M)

    In order to investigate such works, it is vital for researchers to be able to search for choreographic work titles without specifying a choreographer’s name, and to find multiple versions within a single list. The lineage of works with multiple expressions and the proper understanding of collaborative works by more than one choreographer are crucial areas of study for dance scholarship, and these unique aspects are well served by the existing rules (LCRI 25.5B). When two or more choreographers bear equal responsibility for a work, providing an entry for it under one choreographer’s name risks confusion and loss of appropriate credit. For example:

    Brahms/Handel (Choreographic work : Robbins and Tharp)

    While treating dance works in the same way as musical works may have advantages for consistency, there are fundamental differences in the titling conventions for dance and music. There can be multiple musical works titled, for instance, “Symphony in C Major,” but these works are not related. Nor is there a tradition of assigning numbers to the works of individual choreographers to ensure precise identification, so description of context and lineage within the name authority heading and authorized access point are crucial.

    The possibility of providing an access point for a choreographic work without assigning it to a choreographer allows for discoverability of works dealing with the broader history and lineage of a work such as Swan Lake. The filing of works by title also clearly distinguishes between alternate versions of a same-titled work by a single choreographer:

    Swan Lake (Choreographic work : Ashton after Ivanov and Petipa, M)
    Swan lake (Choreographic work : Ashton and Sergeev, N after Ivanov and Petipa, M)

    Several DHC constituents have noted that an advantage of the new rules is the use of a choreographer’s full name rather than surname only, an improvement in clarity. However, our members also voiced concerns about the vast number of records that will need to be revised to accommodate the rule change.

    DHC supports the request from the Jerome Robbins Dance Division of the New York Public Library to revise 6JSC/LC/30 in order to correct the implication that choreographic works always require derived titles. In most cases, choreographic works are assigned titles by their creators, and these titles can be found in “resources embodying the work or reference sources.” As an advocate for the communities of dance practitioners and dance scholars, DHC believes that choreographic works merit a section in Chapter 6 devoted to explicating their unique features—which unfortunately are not widely understood by cataloguers outside of dance-specific collections—rather than being treated solely under derived titles. DHC would be happy to coordinate with our members and constituents to contribute to such a section.

    DHC understands that the continuing development of RDA may create new solutions for providing the kinds of discoverability and specific information regarding variant expressions cited above. We urge that the needs of dance researchers and the dance community, as laid out in this response, be considered in the evaluation of 6JSC/LC/30 and any future adaptations to the treatment of choreographic works.

  22. Tracey L. Snyder says:

    Sorry for this rather noncommittal response, but I can see advantages either way–browsing lists of works by a choreographer, or browsing lists of different choreographic treatments of the same title. At the moment, the practical side of me is winning; we should not take lightly the incredibly burdensome work that would be involved in changing existing headings that begin with title. I fear chaos and misuse of human resources.

    • Dominique Bourassa says:

      I agree that there is value in both approaches. But I really feel listing by titles is only valuable for ballet and other similar dance works. Dance works that are not based on a story are a different case. Many (probably most) of them have unique titles that do not collocate with anything else. I also agree that if this proposal is accepted, there will be a huge amount of work need to change all these records.

      When I think of choreographers such as George Balanchine, Martha Graham, Alvin Ailey, and so many others, I think of such huge dance personalities. Their names alone conjures images of different dance style and choreographies. It feels so unfair to think that their work do not receive the same bibliographic treatment as other works, that their creative output are not valued in the same ways as writers, composers, and other creators’ outputs.

      I think we really need to proceed with caution here. If we keep the rules as they are, some cataloguing agencies will not be happy to have to chance their ways. I mention BNF above. I don’t know what other agencies do, but it would be worth looking into this. As we go ahead, we cannot impose our American views to others just because it fits our American users. Already some parts of RDA are known to be anglo-centric. RDA is supposed to be an international standard and we must ensure that it continues to evolve as such.

  23. Robert L. Maxwell says:

    I strongly agree with Dominique. I also do not share the view that choreographic works should be entered under title. Everyone appears to agree that the choreographer is the creator of a choreographic work. There is no principled reason, then, for not treating these in the same way as other works (that is, the authorized access point begins with the AAP for the creator and then the preferred title of the work) other than “that’s the way we’re used to doing it”. The argument that users might not know the name of the choreographer applies to any work, not just choreographic works. We aren’t therefore proposing that all works should be entered under title. We can certainly give *access* to the work through the title by means of variant access points formed in the AACR2 way, and it might be a fine idea for PCC to make a policy calling for this type of variant access point for choreographic works. But there isn’t any particular dance-specific reason why these works should be treated differently from all other works in the formation of their access points.

    As for the argument that we’ll have a vast number of records to revise, yes, that’s true, but it isn’t the first time this has happened. We as a profession can deal with it. For one thing, it isn’t necessary to wholesale go through and change them all at once. One way to deal with this might be that we change the access points as we need them for new or revised RDA records. “Too much revision” is often used as a reason not to change, but I think it shouldn’t be used here.

  24. Steve Kelley says:

    I agree with Dominique and Bob. There just doesn’t appear to be a principled reason for treating choreographers in a fundamentally different way than other creators of works.

    • John Myers says:

      I also agree with Dominique and Bob on the many points they raise regarding the responses from NYPL and DHC. The movement in RDA is towards standardized, principled approaches rather than extending genre-specific “case law.” While I am not unsympathetic towards the authority work involved for these two institutions, Dominique informs us that the proposal aligns with the BNF’s authority practices, so in resolving this issue SOMEONE is going to have to make changes. I would prefer that the changes imposed be in the service of principles rather than insular practice.

  25. Matthew Haugen says:

    Comment received from the DCRM(MSS) editorial group:
    The proposal’s attention to the phenomenon of works without titles is very welcome. However, the instructions in the proposal for devising a title for works embodied in manuscripts ( run counter to long-standing practice in manuscript cataloging. Manuscript cataloging rules prescribe devising a title that briefly describes the nature of the material; cf.
    AACR2, 4.1B2
    Archives, Personal Papers, and Manuscripts (APPM) 1.1B2-5

    Describing Archives: A Content Standard (DACS), 2.3.3 (

    Descriptive Cataloging of Rare Materials (Manuscripts) (DCRM (MSS)), 1B1 (

    The instructions in prescribe instead that devised titles for works embodied in manuscripts be based on the manuscript’s repository designation. This seems to reflect a misunderstanding of the purpose of using a repository designation as an access point. In almost all cases, an access point based on a repository designation refers not to the text contained in the manuscript, but to the manuscript as a physical object: its decoration, its binding, its script, etc. Such an access point is needed when describing a published manuscript facsimile, or a resource that consists of reproductions of illustrations from the manuscript, or a secondary work about a manuscript as an object. This access point, which may take the form of a repository designation or of a byname for the manuscript, complements the access point for the textual content of the work embodied in the manuscript:

    Astronomical calendar for the years 1348 and 1349 (textual content of manuscript)

    Pierpont Morgan Library. Manuscript. M.1058 (manuscript as object)

    Virgil. Aeneis (textual content of manuscript)

    Vergilius Vaticanus (manuscript as object)

    In a small number of cases, a byname for the manuscript has come to signify both the manuscript’s textual content and the manuscript as a physical object: the Domesday book and the Dead Sea scrolls would be examples of this. But generally speaking, repository designations do not have this dual significance, and should not be used as the basis of a devised title for the work embodied in a manuscript. A repository designation provides no indication of the content of the work; as Bob Maxwell has noted, it does not reliably identify a work that appears in more than one manuscript; it creates problems when ownership of the manuscript changes; and it would read very oddly in cases where the author of the work can be identified, e.g.:

    Smith, John. Repository name. Manuscript. Shelfmark.

    Moreover, the use of a repository designation would be problematic for modern manuscripts, many of which are not assigned distinctive repository numbers.

    For all these reasons, the instructions for devising titles for manuscripts should be the same as those given for devising titles for any other type of format, i.e. devise a title that briefly describes the work.


    Diary of John Ward

    Treatise on archery

    Last will and testament of Pietro Biagio

    Letter from David Garrick to Albany Wallis, 1768 April 21

    In cases where a non-distinctive title is needed as an access point, it can be qualified by the name of the repository and (if necessary) shelfmark, in the same way that an art work with a non-distinctive title is qualified by the name of the museum that owns it (cf. 7.e.

    Eyck, Jan van, 1390–1440. Saint Francis receiving the stigmata (Galleria sabauda (Turin, Italy))

    Eyck, Jan van, 1390–1440. Saint Francis receiving the stigmata (Philadelphia Museum of Art)

    Example for a manuscript:

    Ethiopian prayer scroll (Pierpont Morgan Library. Manuscript. MS M.1138)

    If it is intended for RDA to continue to provide instruction on formulating access points for manuscripts as physical objects, this should be addressed as a separate instruction.


    In the proposal, text, the new example in 7.h. should be eliminated:

    “7.h. General Guidelines on Constructing Variant Access Points Representing Works, 4th example box, add new example:

    British Library. Manuscript. Additional 43487

    Authorized access point for the work: Browning, Elizabeth Barrett, 1806-1861. Sonnets from the Portuguese.

    Variant access point using a devised title based on repository desgination of a manuscript embodying this work”

    A repository designation access point should be a variant only for an access point based on a manuscript’s byname:

    Codex Brucianus

    X Bodleian Library. Manuscript. Bruce 96

    It should not be used as a variant access point to a heading for the text, since this would result in a proliferation of variant access points for works that appear in many manuscript copies (e.g. the Bible, the Aeneid, etc.).

Leave a Reply