COVID-19 update

Stanford Libraries e-resources are available to support you during remote instruction.
Physical access to library materials and locations remain closed at this time. read more

Stanford - RDA questions and answers (upd. July 2012)

1. Relationship designators for artists in records for art/art exhibition catalogs
2. Publishers: punctuation between parallel publishers
3. Bibliography note: transcribing unnumbered pages
4. Pagination: unnumbered sequences of pages
5. Numbers (ordinal) in notes
6. Relationship designators for conference speakers
7. Relationship designators for conferences
8. Multiple copyright dates in addition to the publication date
9. Publication date: transcription of Roman numerals
10. Numerals in series numbering
11. Relationship designators for conferences
12. Multiple qualifiers order in access points
13. Relationship designator placement in name-title access points
14. Transcribing statement of responsibility for moving image works
15. Copyright date the same as publication date: fixed field coding
16. Title proper of a DVD with a predominant work and "extras"
17. ISBN: captions in parenthetical qualifying information
18. Multiple edition statements
19. Abridging long titles
20. Work manifested requirement for multiple/aggregate works
21. Work manifested requirement for compilations
22. Physical description of accompanying material
23. Preferred access points coinciding with transcribed titles
24. Transcribed titles in addition to preferred access points
25. Accompanying material as a supplementary work
26. Aggregate work and constituent work with the same title
27. Recording subunits of incomplete resource
28. Transcribed punctuation matching ISBD punctuation
29. Geographic names as subjects: use AACR2 or RDA form in 6xx?
30. Geographic names as subjects: newly established RDA form in 6xx?
31. Transcribing capital letters of varying sizes
32. Coding uncertain dates in the fixed field
33. Subfield coding for relationship designators in 856
34. Capitalization of relationship designators
35. Attribution of summaries
36. Punctuation in controlled headings with $e
37. Provider-neutral records
38. 776 second indicator
39. Production statement in lieu of publication data for an unpublished thesis
40. Relationship designators for curators of art exhibition catalogs
41. Multiple language expressions
42. Multiple language expressions: access points
43. How do we present relationship designators in $i of linking entries (76X-78X)?
44. If a series has issues in different media or carrier types, is it one series?
45. If a series changes media type, is it now two series?
46. If a series changes carrier type, do we use the existing qualifier in the series access point?
47. Which qualifier do we use if a series is produced in different carrier and media types?
48. Do we use the repeatable 260 in RDA?
49. Compilation of works by one author, collective title: work has adequate title
50. Compilation of works by one author, collective title: "essay" vs. "prose work"
51. Compilation of works by one author, collective title: "Prose works. Selections"
52. Compilation of works by one author, collective title: LCPS
53. How do we include jurisdiction terms for place names in conference access points?
54. Microform reproduction of serial: Qualifier for preferred title
55. Microform reproduction of serial: S/L fixed field
56. Microform reproduction of serial: Dates fixed fields
57. Microform reproduction of serial: Can I code fields based on the original manifestation?
58. Introductory words in titles
59. Reporting BFM to LC
60. How do we suggest a term to add to the list of relationship designators?
61. Preferred title for simultaneous publication in different languages
62. How do we code preferred title access point for a part of a work?
63. Call numbers for reproductions
64. Providing authorized acces point for related work with two or more authors
65. Related manifestation - parallel record?
66. Relationship designator "issuing body" in reproductions
67. Compilation of works which are reproductions of earlier manifestations originally published separately: use of 775 field
68. Facsimile of manuscript: relationship to item (repository designation)
69. Abbreviations in related work entries physical description (775 field $h)
70. Relationship designator "sponsoring body" at expression level?
71. Compilation with commentary as the work of the commentator?
72. Definition of "prominence" in RDA
73. Syntax of collective title access points

Authorities questions:

A1. AACR and RDA headings in NARs
A2. Breaking conflicts between 7xx and 4xx
A3. Special decisions for geographic names
A4. Adding 7xx to preliminary AACR2 NARs
A5. Earlier/later linking references in AACR2 and RDA records
A6. AACR2 "see also" references linking to RDA headings
A7. NARs for earlier names
A8. Jurisdiction qualifiers in English
A9. National Library of Medicine AACR2 NAF headings
A10. NACO Participants' Manual and RDA authority work
A11. Is date always required for authorized access points for corporate bodies?
A12. Established series authority needs "new series" numbering: RDA or AACR2 form?
A13. In adding a 7xx for the RDA form of an older series, must I take into account present-day conflicts and add a qualifier?
A14. In NAR 670s, do we transcribe capitalization as found?
A15. Place names already established in the NAF would have different forms in RDA
A16. Relationship designators for pseudonyms
A17. Post-Jan. 1, 2011 use of established AACR2 forms in RDA bibliographic records
A18. Non-jurisdictional qualifier in place name heading
A19. NARs for ongoing conferences
A20. Undifferentiated names on AACR2 NARs
A21. Canadian names (until LAC implements RDA)
A22. RDA-related 046, 37X, and 38X coding in AACR2 NARs
A23. Citing reproductions in 670
A24. Ongoing conference headings with AACR2-established place names
A25. Ellipsis in the preferred title for a work
A26. "No hybrid headings" includes access points from AACR2 bibliographic records
A27. Unnumbered main series with subseries: how many SARs in RDA?
A28. Establishing preferred titles for monographic works
A29. SARs for series entered as name-title


1. QUESTION: I am having some difficulty determining appropriate relationship designators for artists when cataloging art catalogs and art exhibition catalogs. Current practices recommend that if a book contains reproductions of a single artist’s works, unless the person who wrote the catalog is represented as the author in the chief source of information, the main entry for the catalog is under the heading [authorized access point] for the artist. In RDA., in such cases do we consider the artist the author, even if he has not providing any content for the work in hand but is only reproduced, i.e., [name or artist],$eartist,$eauthor?

ANSWER: If a corporate body is involved, start at RDA and the related LCPS section on art catalogs. If no corporate body is involved, just follow the general RDA instructions on determining if there is a creator. Is the catalog the result of collaboration? Is it a compilation? As far as the relationship designator issue, remember first that such designators are not required in RDA. If your determination is that the artist is the creator of the catalog, then the artist is given in the 1XX field. Be careful about using the term "author" as a synonym for creator. If you want to use a relationship designator in the 1XX field, "author" would not be appropriate (see the definition for this designator in appendix I) because I assume that the resource is not primarily textual. Remember also that appendix I is not a closed list; you can devise your own term as long as the type of relationship you want to express is not already covered by a term already in the appendix. (6 October 2010)

2. QUESTION: What kind of punctuation do we use for separating parallel publishers in the statement of publication? Do we separate by parallel symbol or do we separate by colon?


      Parallel symbol (=), per

ISBD 2007 edition

      , page 4-2. (There are examples on page 4.2-8.) Note that parallel statements of publication are not an RDA Core Element. (19 August 2010)


3. QUESTION: We are wondering how to transcribe unnumbered page(s) in a 504 note. Are the following examples correct?
Includes bibliographical references (unnumbered page 100-102). Formerly: (p. [100]-102).
Includes bibliographical references (page 100-unnumbered page102). Formerly: (p. 100-[102]).
Includes bibliographical references (unnumbered page 100-unnumbered page 102). Formerly: (p. [100]-[102]).


      We decided to simplify the notes because they are not transcribed information. We plan to give the numbering in notes as if the page numbers actually appeared; we think anyone looking at the resource would be able to count past the last numbered page and find the information. So, for example, our version of your first note would be: "Includes bibliographic references (pages 100-102)." However, the style of such notes in RDA is not prescribed; so, you should decide what you want to do. (14 September 2010)


4. QUESTION: We are wondering how to refer to multiple sequences of unnumbered pages: 1st group, 2nd group, etc. or 1st sequence, 2nd sequence, etc.
For example: 300 $a 4 unnumbered pages, 123 pages, 5 unnumbered pages.
504 Includes bibliographical references (2nd unnumbered sequence, page 1-5).
504 Includes bibliographical references (unnumbered page 1-unnumbered page 5, 2nd group).


      Based on policy stated above, our version of this note would be the following: "Includes bibliographic references (pages 124-129)." (14 September 2010)


5. QUESTION: Is it okay to use 1st, 2nd, 3rd, etc. in a note or should we spell out first, second, third, etc.?


      RDA sppendix B.5.11 says generally not to abbreviate. But, the "generally" gives you an out if you want to abbreviate. (14 September 2010)


6. QUESTION: What is the appropriate relationship designator to use for a person who delivered a single speech at conference when the work being described is the manifestation of the single speech in textual volume form? (Note: the person who gave the speech may or may not have written the speech.)


      If the resource content is text, then "speaker" isn't appropriate. If the person wrote the speech, "author" would be appropriate in the authorized access point. Note that appendices I-K are not closed lists. You can use a term of your own choosing if the role played is not represented in the appropriate appendix. At LC, we're asking our testers to notify us of missing terms. You might include such a comment in your record survey form and/or notify John Attig, ALA rep to the JSC. (8 September 2010)


7. QUESTION: What is the appropriate relationship designator to use for a conference where a single speech was given when the work being described is the manifestation of the single speech in textual volume form?


      "Author" would be an appropriate designator for use with authorized access point for the person. Note that appendices I-K are not closed lists. You can use a term of your own choosing if the role played is not represented in the appropriate appendix. At LC, we're asking our testers to notify us of missing terms. You might include such a comment in your record survey form and/or notify John Attig, ALA rep to the JSC. (8 September 2010)


    : "Author" would be an appropriate designator for use with authorized access point for the person. Note that appendices I-K are not closed lists. You can use a term of your own choosing if the role played is not represented in the appropriate appendix. You might notify John Attig, ALA rep to the JSC of missing terms. (4 June 2012, RJR)

8. QUESTION: How do we code Fixed Fields when there is a publication date and multiple copyright dates?

ANSWER: RDA and LCPS for 2.11 and RDA says "if the resource has multiple copyright dates that apply to various aspects (e.g., text, sound, graphics), record only the latest copyright date. LCPS says "if the copyright dates vary, give the latest copyright date." 260 $c would contain publication date and only the latest copyright date, so there are only 2 dates to code in Fixed Fields. (23-27 August 2010)
UPDATED ANSWER: Per LCPS 2.11, copyright date is recorded only if neither the date of publication nor the date of distribution is identified. Generally, the copyright date is used to infer the date of publication (in square brackets), so it is not necessary to also give the copyright date, either in 260 $c or Date2. If copyright date is given separately, it is still coded separately in Date2. (22 May 2012, CAM)

UPDATED ANSWER: As of June 12, 2012, dates of production, publication, distribution, manufacture, or copyright notice are recorded in 264 $c.

If you are recording *both* a publication date ($c in 264 with second indicator 1) and a date of copyright notice ($c in 264 with second indicator 4) for a monograph, then Type of date (008/06) is "t" (Publication date and copyright date). Both Date1 (008/7-10) and DATE2 (008/11-14) would be given.  There is an example of this in the LCPS for (LC's PSD, 19 June 2012)

9. QUESTION: We would like to confirm that following LCPS for 1.8.2, if the publication date (260 $c) is given in Roman numerals, we transcribe as found and we do not supply give date in Arabic numerals in square brackets.


      RDA 1.8.2 applies only to numerals found on the resource for elements listed at RDA1.8.1. So, the LC policy for the first alternative is to accept the form of numerals found on the source. (14 September 2010)

SUL Policy:

      Record dates in Arabic numerals.


10. QUESTION: Does 1.8.2 apply to series numbering in authorized related work access points for series (8XX fields) as well as the transcribed series statement (490)?


      No, the form of numbering in an authorized access point for a series is not covered by RDA 1.8.2 because that element is not in the list in RDA 1.8.1. The number as part of an authorized access point is in RDA at 24.6 (I know, not where you'd expect it to be; I keep losing track of it). The form of the number there is recorded according to the guidelines in 1.8. However, I would expect that libraries would still want to follow the guidance in the DCM Z1 documentation about staying with the style of the number of the issue used to create the SAR; that will need to be a PCC discussion/decision. (14 September 2010)


    The guidelines in DCM have not been updated and catalogers  are still referred to A2 and LCRI 1.6G. We are told to determine the form of numbering  from the item in hand (not necessarily the first item of the series) and then use that form in tracing all items in the series. (2 July, 2012, HGL)

11. QUESTION: What are the examples of the relationship designator terms that are appropriate for conferences?


      Two possible designators are "host institution" and "sponsoring body" for corporate body relationships to the event/publication. For a person, at times "editor of compilation" or "illustrator" may be appropriate.


      First, relationship designators are not core in RDA. So, even in situations when there are appropriate relationship designators included in appendices I-K, such designators are included or not included according to agency policy. Also, not every possible role in included in the appendices. I don't see a relationship designator that I would use in a 111 or 711. RDA says that conferences are considered to be creators; but, I don't find a term in appendix I.2.1 that fits a conference. The third paragraph of appendix I.1 says you can use your own term if you don't find an appropriate one in the appendix. Because the LC policy for the US RDA Test is to require the use of only a few relationship designators, I wouldn't try to devise a term to use for a 111 or 711. (8 September 2010)


12. QUESTION: When multiple qualifiers are needed in an authorized access point for a work, in which order are they to be added (e.g., date of work and place of origin)? Should the order reflect the order of instructions in Chapter 6? Examples under reflect the AACR2 order (e.g. "Ocean’s eleven (Motion picture : 2001)"), though the text of the instruction does not address it. Does the order even matter?


      LCPS does not explicitly prescribe order, but examples show that when date is one of the qualifiers, it comes last. (23-27 August 2010)


13. QUESTION: Can or should one use relationship designators for authors in name-title access points in bibliographic records, and if so, how does one construct such an access point? For example: 700 12 Person, Joe, $d 1927- $e author. $t Title.


      Resignator designators in authorized access points (including name-title) in bibliographic records is not required by RDA. An agency can make an overall policy decision or catalogers can decide on a case-by-case basis for certain roles or all roles. Your example is a correct formulation. (22 September 2010)


14. QUESTION: In RDA, only the first statement of responsibility is required, and LC's "Elements to be included" (at, notes "Yes (if more than one, only the first recorded is required)" and there is a further LCPS saying generally not to abridge "a statement of responsibility." There is no specific statement on So, for the testing, for institutions generally following LC's test policies, should we be transcribing only the first statement of responsibility as a default and omitting the rest? Is there an OLAC (or at least a OLAC/MLA test group) decision on how we want to handle this?

The first statement of responsibility on a video is usually a corporate name, often in the form of a logo or such before or just after the title, and often in the form of so and so "presents," so the first statement would rarely be for the director.
ABC News
producer and writer, James Benjamin
director, Al Niggemeyer
director, Dan Bessie
writer, Phyllis Harvey,
animation, B. Davis
editor, I. Dryer

They don't define what "intellectual or artistic content" means but from this example it looks like produers, writers, directors, animators, and editors. But if you go to it says: . Record the names of persons, families, or corporate bodies who have contributed to the artistic and/or technical production of a motion picture or video recording other than as performers, narrators, or presenters. Do not include the names of assistants, associates, etc., or any other persons making only a minor contribution. Preface each name or group of names with a statement of function. Example:
Credits: screenplay, Harold Pinter; music, John Dankworth; camera, Gerry Fisher; editor, Reginald Beck
Credits: script, John Taylor; calligraphy and design, Alan Haigh; commentator, Derek G. Holroyde

So from this example it looks like screenwriters and editors should be in the 508. But since the statement of responsibility example shows them there, i'm confused.
The RDA example we were provided for Black Christmas puts the producer and director in the 245 and screenwriter in 508. It has no production company at all, which seems very strange for a commercial film. Unfortunately my library does not have a copy of this because i wanted to check the record against the credits to try to understand it. (27 August 2010)
14a. ADDITIONAL QUESTION: Ok, i've been expressing my confusion about this for a couple of weeks, given the inadequate instructions at to "give preference to those identifying creators of the intellectual or artistic content," and, 7.24 "For statements identifying persons who have contributed to the artistic and/or technical production of a motion picture or video recording." As far as i can tell, the sole guidance here is in the examples, which contradict each other.
So, does anybody mind if i make an executive decision here, until OLAC or someone else comes up with something? I propose the following text for my draft instructions:

"As an interim procedure, based on RDA I.2.1 and I.2.2 relationships considered to be associated with the work, include screenwriters, filmmakers, directors, directors of photography, producers, and production companies. Feel free, however, to also include, on a case by case basis, others whose status as a creator is obvious given the nature of the work being cataloged (for example, choreographer for a ballet film, editor for a film consisting of editing found footage, animator of an animated film, etc.)"

The LCPS's are silent on the inclusion of a 508 (which is where the rest of the technical credits go). It is not a core element. It has always been included as part of full level cataloging, but was dropped for level 4,so we have not used it for many years.


      Aren't these (the relationships considered to be associated with the work) covered by RDA 1.2.1 + 7.23 + 7.24?
      Since SUL has always followed PCC/LC practice, we will generally continue doing so for the RDA test (and beyond). If no specific PCC/LC practice for the test has been recorded in RDA test documents, we will follow SUL's current cataloging practice (modified by RDA rules, of course). So yes, you can use a 508 note. (JKD, SUL MDU) (8 September 2010)


    The JSC is has deleted the conflicting example at  SUL policy for videos is now to give the first statement of responsibility.  For the subsequent statements for relationships noted above in the "interim procedure," either put them in the statement of responsibility or 508, whichever is easier. (6 June 2012, GDG)

15. QUESTION: We now have to give the copyright date even if there is a publication date (per LCPS 2.11). If the dates of publication and copyright in field 260 $c are the same, do we use “t” in the type of date fixed field (008/06) with both date 1 and date 2?

ANSWER: The MARC format isn't as clear on this point as it could be but we think in a case like this you follow the table of precedence in MARC for monographs complete in one year-- that table (attached) has "s" listed before "t", so using "s" and only Date1 would be an approach one could follow. If someone used "t" and the same date in Date1 and Date2 we wouldn't complain, but we think MARC really only expects that when the copyright date and publication date are different. (Table of precedence is found in MARC format in section 008-All materials.)
UPDATED ANSWER: Sorry to change our answer so soon, but we checked with the MARC Standards office on this question (we've gotten it a few times)-- they suggest that 't' is the appropriate value for 008/06, with the same date in Date1 and Date2 (their logic: you have two dates, even if they are the same year). We're going to include some coding for an example like this in an updated LCPS. (15 September 2010)
UPDATED ANSWER: Per LCPS 2.11, copyright date is recorded only if neither the date of publication nor the date of distribution is identified. Generally, the copyright date is used to infer the date of publication (in square brackets), so it is not necessary to also give the copyright date, either in 260 $c or Date2. If copyright date is given separately, it is still coded separately in Date2. (22 May 2012, CAM)

Note: As of  12 June 2012, dates of production, publication, distribution, manufacture, or copyright notice are recorded in 264 $c.

16. QUESTION: I had an additional question regarding predominant/subsidiary works.
By way of explanation, since you don't catalog videos at LC: It's extremely common in video cataloging to have a single film which is clearly the predominent work on a DVD. The title of this film is usually also the title on the container and on the disc label, and usually also on the DVD menu. However, also on the same DVD will be "bonus" material. This may be as little as a filmography, but more often includes bonus soundtrack materials, deleted scenes, a "making of" documentary (usually without a formal title), and interviews. Instead of or in addition to these, there is often one or more entire films added as "bonus" materials. These are usually complete works in their own right, having been previously exhibited or released separately from the "predominent" work in the resource. They may be short films or additional whole feature films. For example, in my hand i have a very typical DVD. It has a container with the title "The Italian Straw Hat (Un chapeau de paille d'italie), a film by Rene Clair." The label reads exactly the same. The DVD menu has only "The Italian Straw Hat (Un chapeau de paille d'italie)" with the menu items "play feature, chapter selection, subtitle selection, special features." If you click "Special features" you get a menu with the additional films La Tour (The Eiffel Tower) (1928) and Noce en Goguette (Fun after the wedding) (1907), plus an item instructing you go to to your DVD drive to access a text file containing the complete publication "The Leghorn Hat, by Eugene Labiche and Marc Michel, published by the Poet Lore company in 1917. Usually the extra works are accessible under the special features button on the menu, sometimes they are choices listed on the main menu along with the predominant film. But the title on the disc label and container are generally for the main film, making it clear that the intent of this DVD publication is to present that particular film.

I'm not finding in RDA the rule corresponding to AACR2 1.1.G.1 that says: If, in an item lacking a collective title, one work is the predominant part, treat the title of that work as the title proper and name the others in a contents note.. It doesn't seem to be in the RDA sections on collective titles in chapter 2. 17.3 mentions "predominant or first-named expression, but that section is on "relationships" which I take to mean added entries, and I'm trying to find the citation for the 245$a.

Discussing it with my colleagues here, it's been suggested that perhaps it's subsumed under and since the title of the main work usually is what appears on the packaging and disc label, that that would constitute a collective title. Problem with that is that, logically, you then would be using the label or packaging as the source of information and so wouldn't be able to transcribe the statement of responsibility from the credits into the 245c, since your 245a represents the "collective title" taken from the label rather than the title of the individual work contained therein. So in my simple DVD with a feature and a films and bonus text i ended up transcribing the titles and statements of responsibility from the credits in the 505 and having to split all of my notes between the two titles, rather than simply using most of the fields to catalog the predominant work (taking the title and statement of responsibility from the credits as instructed in and adding a note to describe the bonus films. This doesn't sound right to me--is there a different interpretation? And does it make a difference if the bonus materials are ones you would want to trace (as in this example) or whether it was the weakly or non-titled ancillary materials such as the "making of" and interviews that one would normally only note and not trace?

The Wizard of Oz example in looks like the title of and statement of responsibility were taken were taken from the credits with notes for the subsidiary works, but i see that it is issued on multiple discs. But... looks like says in that case you would use the lowest numbered part, which if the discs aren't numbered wouldn't apply. Subsection d does say if "there is no source of information identifying the resource as a whole (e.g., a set of locally produced videodiscs with no source of information identifying the set as a whole)" which makes it sound like a commercially issued set isn't what they have in mind. So i'm not sure if that example illustrates or the missing AACR2 1.1.G.1.


      You're correct that there isn't an RDA counterpart to AACR2 1.1.G.1. So, your choices would be to transcribe all that you have, treat one as a collective title even though it isn't (probably not an attractive choice), devise a title, or create analytic records for the parts (probably not an attractive choice on a regular basis). I note that the 2nd paragraph in RDA does refer to some situations for moving image resources.


      Also, while we were looking at, we began to wonder if the last sentence ("If the titles of individual parts ...") should really be part of the alternative preceding that sentence. We'll add that topic to our list for further consideration. (30 September 2010)


    Essentially the same at this time.  There is a CC:DA task force looking at this issue and several others in 2.1 and 2.2. but it is unclear how the issue will be resolved.  In the meantime, no longer has the sentence that LC was puzzled by in the previous answer. (6 June 2012 GDG)

17. QUESTION: When adding qualifiers to Identifier for the Manifestation in 020, should the caption agree with that used on the item, or with the extent term used (e.g., "Bd.1" vs. "v.1"). Or, since it is cataloger-supplied, does it matter? Are abbreviations acceptable, as shown in the examples?


      Since it is a cataloger-supplied element, either the found caption or the generic "v." is acceptable. (14 September 2010)


18. QUESTION: What punctuation do we use to separate multiple edition statements?


      ISBD punctuation separates multiple edition statements with a comma. (12 October 2010)


19. QUESTION: I have a question regarding the optional rule for RDA (abridging long titles).
"Optional Omission
Abridge a long title only if it can be abridged without loss of essential information. Use a mark of omission (…) to indicate such an omission. Never omit any of the first five words."
I cannot seem to find a LCPS on this. Does that mean we can apply it?


      LC leaves it up to a cataloger's judgment and we will follow LC on this so yes, you do have this option. BTW, this proposed practice is recorded in LC's document "RDA alternatives and options." (JKD, SUL MDU)


20. QUESTION: RDA Rule 17.8: LC has been interpreting the following: "If more than one work is embodied in the manifestation, only the predominant or first-named work manifested is required" to mean, "if a manifestation contains more than one work, in addition to recording the WHOLE WORK's title, etc. in the 1XX/245, ALSO record the title, etc., of the first (or predominant) included work in a 7XX with $t."
At Stanford they interpret this to mean, "if a manifestation contains more than one work, you only need to record the predominant or first-named work," and they interpret that to mean, the work named [named first?] in the title proper."
Do you see any possibility that the Stanford interpretation is correct?


      Heidi's view of the situation is just fine for Chapter 25 of RDA (Related Works), when remembering a few things about related works:
  1. one category of related works is whole/part (which is why there are examples in of what we have traditionally seen as 'contents notes')
  2. RDA allows 4 techniques for relating works:
    1. identifier (US RDA Test Coordinating Committee has rejected this technique for the test)
    2. authorized access points for the related works (what we know as "analytic" added entries)
    3. structured descriptions (which would include the kinds of formal contents notes we know)
    4. unstructured descriptions (free text notes prepared by the cataloger).

Relating works per Ch. 25 is not a 'core' activity in RDA-- all such relationships are optional in RDA. However, the LCPS for 25.1 makes related works 'LC core' for compilations, by giving contents notes or authorized access points (and allowing the cataloger to decide if it is too burdensome to do so-- this is one of the policies that we went to ABAMT to consult on before setting); we'll be adding some categories of things where it is likely too burdensome based on some of the practice records you and your fellow testers have been providing. That said, the requirement in 17.8 is different-- it is core for RDA (not just LC), and as you note is actually a "core if"-- if there is more than one work manifested, only the first or most predominant is required. These 'primary' relationships in RDA Ch. 17 can be accomplished using 3 different techniques (this is what sets it apart from Ch. 25 above):

  1. identifer (again, rejected by the US Test)
  2. authorized access point (i.e., an analytic)
  3. a composite description (this is what was called multi-level description in AACR2 and not practiced in the North American context-- we have an LCPS that says not to do this).

That leaves you with an authorized access point to satisfy the 17.8 requirement.


      So while the "Stanford Way" satisfies the LC core for 25.1 just fine, it doesn't satisfy 17.8 as we see it. We hope to do some scoping for the 17.8 to give some leeway in certain situations thanks to the practice records we've seen (and talk to the JSC about changing the core statement in 17.8 eventually). (Dave Reser, LC PSD) (2 September 2010)

ADDENDUM:The application of LCPS 17.8 is clarified below, under answer #21.

21. QUESTION: Must I give an analytical access point for the first/principal work manifested (RDA 17.8) or first/principal expression manifested (RDA 17.10) in the bibliographic record for every compilation?


      For compilations of works, give an analytical authorized access point for the predominant or first work in the compilation when it represents a substantial part of the resource. Disregard contributions such as a preface or introductory chapter. Generally, do not apply this core element to anthologies of poetry, conference proceedings, journals, collections of interviews or letters, and similar resources. (LC FAQ on the RDA Test)


    No. LCPS 25.1 states: "For compilations of works, give an analytical authorized access point for the predominant or first work in the compilation when it represents a substantial part of the resource. Disregard contributions such as a preface or introductory chapter. Generally, do not apply this core element to anthologies of poetry, hymnals, conference proceedings, journals, collections of interviews or letters, and similar resources." (4 June 2012, RJR)

22. QUESTION: I have a question regarding LCPS 3.1.4. When accompanying material is a book or booklet, i am confused as to how to record that in $e. According to, if i'm not providing details on each piece, i would give the extent, and it refers me to 3.4, and in it refers me to the list of carriers at, which gives the term "volume".
But according to For a resource consisting of a single volume, record the extent in terms of pages, leaves, or columns as appropriate to the presentation used in the resource .... The exception to this is if a volume is loose leaf or unpaginated. So it sounds like if you have a volume with paging you would always record it as, say, 23 pages. I see no option for recording it as 1 volume (23 pages.). or even just 1 volume if such detail is unnessary (as mentioned in LCPS 3.1.4 section 2).
That makes for an odd 300 field, one i'm currently working on looks like this:
1 DVD-video (76 min.; 20 min.) : $b digital, sound, black and white and color ; $c 4 3/4 in. + $e 23 pages ; 18 cm
Does that look ok? It seems to me that in this context it is confusing.
Wouldn't it be preferable to give it as +$e 1 volume (23 pages) ; 18 cm.
or even just 1 volume? Also, since this is a booklet and not really a book, is it permissible to say:
+$e 1 booklet ?
Booklet is not in the list of terms at, and i've only found one reference to a booklet in RDA, and that's in a note.


      Yes, we've discovered that RDA doesn't handle accompanying material well if you don't want to give a separate physical description for it.

                    During training sessions here, we've been telling our testers that they have three choices:

                                    repeated 300 field, note about the accompanying material, or the "+ $e ______". If using the 300 $e approach, we'd continue current practice until we/ALA can get RDA updated. So, your "+ $e 1 booklet" or "+ $e 1 booklet (23 pages ; 18 cm." is what we'd do depending upon how much info we wanted to give in $e.

                   (Also remember that has an exception for text; you're sent to 3.4.5.) (14 September 2010)


23. QUESTION: I would like clarification on a point about titles and preferred titles.
In 2.3 RDA title proper as a core element. Your spreadsheet "LC RDA CORE ELEMENTS FOR THE RDA TEST affirms this and maps it to the 245 field.
In 6.2 RDA identifies the preferred title of the work as a core element. Your spreadsheet affirms this and maps it to the 240 or 130.
From this i would infer that all records would require a 245 field and a uniform title in 1xx/240 or 130. By extension an analytical added entry would require both a 740 02 and a 730 02.
I want to be sure that i am understanding this correctly. It does not appear to me that others are coming to that conclusion and i am puzzled. Personally i would not have a problem with doing this, but i know it would be wildly unpopular, particularly if NACO members were required to make authority records for all those uniform titles.
I'm more annoyed at having to test every title for conflict rather than just routinely add the elements that would be used to differentiate (cf. LCPS probably takes less time and effort to routinely add the elements than to do all the searching for a possible conflict.
I don't find anything in the LCPSs that address this issue, though LCPS appendix 1 says that motion pictures wouldn't need a preferred title except for conflicts, additions, not original title, etc. But that's only for motion pictures and doesn't talk about other types of materials. And it contradicts RDA 6.2.


      Yes, encoding RDA in MARC raises such issues because so many fields in MARC do double-duty (e.g., the 245 $a as title proper and as uniform title if the same as the title proper). There is an LCRI saying not to give a 240 or a 130 if the uniform title is the same (except for initial article, etc.) as the title proper. At one point we had included that as an LCPS as well and then removed it, waiting to see what people would do during the Test. But, during the discussions on practice records this summer, none of the LC testers always gave a 240 or 130. We may consider reinstating that LCRI/LCPS.
      Of course, the inclusion of a 240 or a 130 doesn't necessarily mean that an authority record is required. That's an administrative decision.
      I'm not sure I understand your comment about an analytical access point needing both a 740 02 and a 730 02. The relationship expressed in an analytical access point uses the authorized access point for the related work; so, the found form of the title (740) of that related work isn't appropriate in this context -- that would be included in the related work's own description found in a separate record. Yes, there will be times when a 740 will appear in an RDA record in addition to a 7XX X2 access point: when the compilation lacks a collective title and we need the 740 fields for title access to the titles of the other works now given in $b of the 245 field. Re: LCPS We also carried forward from LCRI 25.5B the LC practice not to break conflicts for monographs unless a subject or related work added entry is needed. Again, an administrative decision related to use of cataloger's time. (14 September 2010)


    LCPS has been revised, now all titles (including monographs) should be tested for conflict.  The tagging issues are unchanged, i.e. do not routinely supply a 130 or 240 if the preferred title is identical to the 245. (6 June 2012 GDG)

24. QUESTION: If i have a collection in a single volume/piece, with or without a collective title, and i decide that i will need access points for the works contained within, i will need to determine the preferred title for each, which may be a 700 $a$t or 730, as the case may be (and usually will be in the case of moving image works). I was asking whether i also in all cases needed a 740 for the title as it appears in the manifestation, even if is identical to the 730 (on the principal that for a title main entry i would always have a 130 in addition to the 245, i would similarly pair the 730 and 740). That seems awfully redundant, so perhaps i would only need the 700 $a$t or 730 and not the 740 unless the manifestation title was a variant that needed to be traced? Since the primary means of accessing a work contained would be through the controlled access point for the work. There is no separate record for the "related work" involved, it's all on the same record because its a work contained in the resource i'm cataloging.


      You're correct that the authorized access point (as you said for your materials most likely just a preferred title with additions as appropriate) for each work inside the aggregate work would be a 7XX with second indicator 2. (Only the first or principal work is an RDA core element.) It is cataloger's judgment to give a 740 for the manifestation title proper of the works. If there is no other record (bibl. or auth.) for the work, then a 740 might be helpful for the user if it is different enough from the preferred title. (21 September 2010)


25. QUESTION: And the reason i was confused by the instruction in is that it says specifically to trace the title of an accompanying piece in the 740. That makes sense to me when the work contained in the accompanying piece is not a work that you would normally provide access to (maybe a dependent or an indistinct title), but because your piece may get separated you may want to trace it so you can look it up again. But i would think that if the accompanying material contained a work that you wanted to provide access to, that one would use a 700 $a$t or 730, since an accompanying piece is part of the resource being cataloged and so also represents a work contained in the resource i'm cataloging.


      You're referring to the LCPS for "Accompanying material" is not the same as a work in a compilation. The separate works in a compilation are equal in status although perhaps not in length/duration, importance, etc. But, when you have accompanying material, you have a main work + supplementary work situation. Most accompanying material included in a description with the main work is handled by a note or by the "+ $e ____" at end of a 300 field. If you want to give a controlled form, you can give an authorized access point in a 7XX as well -- see the second condition in this information at the beginning of the LCPS:
        There are two ways to accommodate variant titles in the MARC bibliographic record:
        1. Recording a variant title in MARC field 246;
        2. Recording a variant title for a component of an aggregate work in a form other than would be recorded as the authorized access point for that component work. This form of variant title is recorded in MARC field 740. Note that use of MARC field 740 to record a variant title does not replace the use of an authorized access point for the component. See RDA chapters 25-28 for those instructions. (21 September 2010)


26. QUESTION: Re the 130, i thought of another question regarding a collection with a collective title. The collective title represents the collection rather than the works contained in the collection (for which i'll use the 700 $a$t or 730) Does the collective title also require a 240 or 130? Aggregate works still seem rather confused in RDA, and if the aggregation is itself considered a work, then it may also need a 130. This seems really strange and potentially confusing to me, especially if the collective title happens to coincide with the title of one of the works and you'd end up needing some creative qualifiers to distinguish them.


      If there is a single creator, then there might be need for a 240 field per RDA There could be situations when a distinguishing characteristic needs to be added to a preferred title with the result given in a 130 when the collective title conflicts with the title of another work. And yes, we've had a few situations during our practice sessions this summer when the title of the compilation was the same as the title of one of the works in the compilation; we gave the LC testers the option to add the qualifier to the access point for the "smaller" work or to give a 130 with a qualifier.


27. QUESTION: Extent (RDA 3.4.1.) is core element for completed works, subunits are only recorded if readily ascertainable or considered important for identification or selection.
RDA is the specific instruction for subunits for microfiches, and links to 3.4.5. on how to record text subunits. There are no LCPSs for or
If I decide the subunit is not important for identification or selection, I can leave it out. Does it (should it) make a difference based on presence of a 787 or other 7xx related work entry? For example: If I have a microfiche which reproduces pages 9-36 from a larger work, but I can't identify the larger work at all, there will be no 787. In that case I would want to record "300 1 microfiche (pages 9-36)," wouldn't I? Or would I use a note such as "500 Reproduction of pages 9-36 of an unidentified work"? (24 September 2010)

ANSWER:      (No answer was received)


28. QUESTION: How do we transcribe found punctuation that matches ISBD punctuation?


      According to 1.7.3: "Transcribe punctuation as it appears on the source, omitting punctuation on the source that separates data to be recorded as one element from data to be recorded as a different element, or as a second or subsequent instance of an element. "Add punctuation, as necessary, for clarity."
      The LCPS that corresponds to LCRI 1.0C (LCPS 1.7.1), conspicuously does NOT include the section on "Punctuation in titles proper that is also ISBD punctuation."


29. QUESTION: For records beyond the common set, we will be assigning Library of Congress Subject Headings. Often assigning LCSH involves using names from the NAF which are formulated according to descriptive cataloging rules. This is frequently the case with geographic names which are used as geographic subdivisions.
Once we have updated an existing AACR2 authority record in the NAF, and the RDA heading (in the 7xx) differs from the AACR2 heading (in the 1xx), which one should we use in the subject heading in the RDA bibliographic record?

ANSWER:    The RDA heading. (5 October 2010)


                   Per the PCC post-test policy, use the authorized access point appearing in the 1xx of the authority record, regardless of the rule set under which it was established. (23 May 2012, CAM)

30. QUESTION: If we need to establish a name heading in the NAF because it is needed as subject heading and we establish it per RDA, can we use this as a Library of Congress Subject Heading in an RDA bibliographic record?


      Yes, because the record is an RDA record. (5 October 2010)


31. QUESTION: How are we transcribing text that is given in small caps (i.e. large capital 1st letter of word, followed by small capitals for rest)?


      I ended up using normal capitalization in my bib record for transcribing text with small caps and nobody objected to it. Big caps as caps and small caps as small letters. (RJR, SUL MDU) (5 October 2010)


32. QUESTION: Have we had any word from OCLC on how to code a date like [not before 1935] in the fixed fields? This formulation is explicitly allowed in RDA, but of course the rules don't say how to code this in DtSt and Dates.


      The consensus here is DtSt "q", Date 1 "1935", and Date 2 "uuuu". (Jay Weitz, OCLC)


      In the revised LCPS we had an example of "[not after 1980]" and we use the date type of "s" (single date) and Date1 of 1980. Can't guarantee that my judgement call on that one was correct, but here is the reasoning used:
      Since only a single date was present, i.e., not an actual date range of some type like "[between 1993 and 1999]" the date has attributes of both a single date (only one year is known) and a questionable date. In the section on date type/date1/date2 in MARC 21 there is a table "Precedence of codes (monographic items)" for when more than one code applies to an item, telling you to use the date type higher in the list. Since s comes before q, I chose the s for that example. I think the same logic could be applied to the example (presuming it is a single part monograph), although I certainly see OCLC's logic in what they suggested." (Dave Reser, LC PSD)
      We'll be checking with NDMSO. (Judy Kuhagen, LC PSD) (6 October 2010)


33. QUESTION: I am using an identifier to refer to a related manifestation (specifically, a PDF version of my book in hand). The examples under show relationship designators appearing before identifiers. However, there is no specific mapping for this component. The most appropriate subfields in 856 appear to be |z ("Public note") or |y ("Link text"). Which should be preferred? As in the examples, would a colon be appropriate to input after the designator, where normally there would be no intervening punctuation in the 856?


      A case of "where RDA meets MARC" that has some wiggle room for how to record, and probably not a 'best' way at this point-- could be a $z, $y, $3, depending on the situation-- I wouldn't worry too much at this point about order of subfields and punctuation between attributes, just do something that is understandable in your public displays. (Dave Reser, LC PSD)


      My inclination is to use |z, and include the colon after the designation, before |u. OCLC validates this order of subfields with no problem. (6 October 2010)

34. QUESTION: About capitalization in relationship designations. I see that in LCPS LC gives the formulation:
$i Reproduction of (manifestation):
In RDA they show an example of Reproduction of: ADM55/40
but appendix J lists all the relationships in lower case:
reproduction of (manifestation)
I believe all the record examples we saw earlier in our training gave them all in lower case.


      At first blush, I think the decision to capitalize would be based on where the designator appears. The one you cite is at the beginning of an access point field, so capitalization seems prudent. A relationship designator in a name access point, however, appears at the end of the name, so capitalization does not seem prudent.
      The listing in appendix J, perhaps, does not presume exactly where the designation would appear in a display, so that might be why they're lower-cased? (CAM, SUL MDU) (8 October 2010)


35. QUESTION: About summary attribution in LCPS 7.10. LCPS 7.10 has bunches of examples on how to attribute summary notes. The section i'm wondering about is this: Summaries, etc., Provided in Copy Cataloging Retain summaries, reviews, and abstracts already present in MARC field 520 in records used for copy cataloging (MARC field 042 = lccopycat, pcc, etc.). Retain any attribution already present. If none is present, add the attribution: EXAMPLE
520 ## … ?? $c Source other than Library of Congress.
I don't think we care about this in copy cataloging, but Jane and I do tons of derived records where a summary is already present. It sounds like in LC this would be attributed as above. But obviously that would be an inappropriate attribution for us. And it seems like other libraries would be annoyed at having a summary with the attribution --$c Source other than Stanford. So what should i put? --$c Summary from bibliographic record of another manifestation? --$c Summary from bibliographic record in OCLC? Or should we just accept the summary and provide no attribution as we do now?


      At this point, we accept AACR2 as well as RDA copy as long as it meets SUL requirements. When an original RDA bibliographic record is created, the same polices and practice apply regardless of whether the record was created from scratch or derived from another record. In the case of LCPS, the local instruction: "do not add $c" applies to our (MDU) "derived original" records. (And since RDA says: "Take information to be used in summarizing the content from any source," you can take a summary from another record.) (JKD, SUL MDU) (11 October 2010)


36. QUESTION: There seems to be a consensus that, in 1xx and 7xx fields, there should a comma between the parts of the field that identify a person, family, or corporate body and the subfield with the relationship designator, e.g., 100 1 Rojas Flores, Jorge, $e author.
When 1xx and 7xx fields with relationship designators are controlled on OCLC, the period at the end of the classic 1xx and 7xx fields is replaced by a comma. This is just as it should be 99% of the time. But, when a 1xx and 7xx field ends with an initial, that vanishes too. For example, 700 1 Wuhrer, L.,$e engraver.
700 1 Wuhrer, L, $e engraver.
This couldn't be right, could it?
The only way I would know how to fix this would be not to control the heading. Is there a workaround?


      Music catalogers have this same problem with name title entries for works of music. OCLC is supposed to be working on the problem. I asked an OCLC rep about it at the 2008 MLA conference and it still has not been fixed. The only fix I know is just not control the heading. (OCLC-CAT list) (11 October 2010)


    This issue has been fixed in OCLC Connexion. (23 May 2012, CAM)

37. QUESTION: How are the formal RDA testers handling provider-neutral records?


      Participants in the US RDA Test will not be creating provider-neutral records or CONSER single-record approach records. (13 October 2010)


    The PCC has received guidelines for provider-neutral RDA records from the task force, as of this writing they have not yet been officially adopted or promulgated (6 June 2012 GDG)

38. QUESTION: In following the example at LCPS, I am adding a 776 to my record describing the original which my microfiche reproduces. The example gives 776 0#. However, OCLC will not validate second indicator [blank] with the presence of |i. According to the MARC documentation, 2nd indicator 8 is used for "No display constant generated", which would seem to be the appropriate value to input in tandem with |i. Perhaps the LCPS should be amended with this in mind? Or, is this a validation change that OCLC is planning?


      Yes, sorry, our inexperience with using the linking entry fields for this purpose caught up with us-- second indicator should be 8 to be true to MARC (we did get it right in an LCPS 1.7.1 example for a 775, but I think that is something we caught later). Best to follow OCLC's validation rule and we should change our example. (Dave Reser, LC PSD) (18 October 2010)


    This issue has been corrected in the current LCPS for (July 2, 2012, HGL)?????

39. QUESTION: The original I'm describing in the 776 is an unpublished thesis. Does the requirement to record a manufacture statement apply? I notice that the "Core element" verbiage at 2.10 states "for a resource in published form". I would like to take that to mean that I am not required to record a manufacture statement, and instead my 776 0# |d will read "Place of publication not identified : publisher not identified, 2008". Is this correct?


      if you were creating an RDA record for the unpublished thesis, you would apply the instructions for Production statements (2.7), not publication statements (2.8). At 2.7, only the date of production is core, not the place of production or producer's name, nor is there any requirement to provide distribution or manufactor data in lieu of publication data. (Dave Reser, LC PSD) (18 October 2010)


40. QUESTION: I catalog art exhibition catalogs on an ongoing basis. These items represent a strong curatorial interest at my institution and it is important to provide access to people(s) who have contributed intellectually to the publication of the catalog. Very often this role is taken on in some or all degrees by curators of the exhibition. I generally cite this information in the statement of responsibility when it appears on the t.p., other preliminaries, or the colophon; I also may it cite it when it appears elsewhere in the catalog.
Currently in RDA there do not seem to be any provisions within the lists of Appendix I.2 or I.3 to express a relationship between the curator and the resources in hand (i.e. the exhibition catalog). The term “curator” does appear in appendix 1.5.2 as a relationship designator but only for use with a person … [et al.] associated with an item.
Is it permissible for me to use a term that appears in one list (in this case for relationship at the item level, as a relationship designator to express relationships between a work and a person, or a person and an expression?


      Remember that you want to give the relationship between the person and the publication. The curator of the exhibition may be the person responsible for the catalog and may be listed on the item as "curator" but the role for the publication is really more likely author (a creator role), editor (an expression role), etc.
      But your more general question: If a role is missing from any of the appendices, you may use a term you think is appropriate. At LC, we're asking our testers to notify this office of what additional terms have been used; we'll consider proposing them as additions to the appendices. You might want to do the same with John Attig, your JSC representative. (20 October 2010)


41. QUESTION: We don't understand a couple of the examples in the ppt for module 4 at We understand the multiple language expression on 79 and the conference proceedings example on 78, but we don't understand why the Greek text with Latin translation notes or the Hindu motion picture with English subtitles are not also multiple language expressions. We are also unable to extrapolate from this what to do with a DVD that has "Soundtracks in Russian, Georgian, French, with subtitles in Russian, English, German, French, Spanish, Italian, Portuguese, Dutch, Swedish, Hebrew, Arabic, Chinese, Japanese." (OCLC# 57499484) This is a bit of an extreme example, but multiple optional spoken and subtitle tracks are not uncommon and the possible combinations soon proliferate.

ANSWER: I confess that I took the two examples you question directly from the examples in RDA "Expressions Involving More Than One Language" where the basic instruction says "If a single expression of a work involves more than one language, record each of the languages." After many discussions in recent months with the LC testers, I also now question if those examples are in the correct RDA instruction.
PLUS: During a JSC teleconference today, JSC agreed to the deletion of the 2nd-4th examples in RDA I'll be submitting that request to ALA Publishing and the web developer this weekend for the next update of the RDA Toolkit (no date announced for that yet). (22 October 2010)UPDATED ANSWER: It appears that the 2nd-4th examples have been deleted from RDA (20 June 2012, MWH)

42. QUESTION: We are also wondering about the necessity of always needing to record access points for expressions by language. Since language of expression is core and 0.6.1 says "As a minimum, a resource description for a work, expression, manifestation, or item should include all the core elements that are applicable and readily ascertainable," it appears that we have to make access points for all these expressions. It seems to us that this is unlikely to be helpful for video cataloging and that in most situations 041 limiting would be more useful.

ANSWER: The core element related to identifying expressions when more than one expression of the work is manifested in the resource is explained at RDA 17.10: only the predominant or first expression is a core element. (The requirement can be met, per reference to 17.4.2, by an identifier, an authorized access point, or a composite description.) Language of expression is a separate element (6.11). The last paragraph of send you to the related instruction in 7.12 about giving what would be a note about language of content when it is a complicated situation. (22 October 2010)


43. QUESTION: How do we present relationship designators in $i of linking entries (76X-78X)? Do we capitalize the first word? Do we end $i with a colon?

ANSWER: When subfield $i for relationship designator in 76X-78X is used, it is the first subfield, the first word is capitalized, and the subfield ends with a colon. (LCPS for 1.7.1)


44. QUESTION: Trying to interpret LCPS 6.27 about different carrier/media types, it sounds like: If a series has some issues online and some in print or CD-ROM, consider it one series. Is this correct?

ANSWER: Yes. (2 November 2010)


45. QUESTION: If a series changes from print to online, consider it 2 series and add a qualifier to distinguish them -- correct?

ANSWER: Yes. (2 November 2010)


46. QUESTION: If a CD-ROM changes to online, it is considered a single series (the carrier changed but not the media type). So presumably if the CD-ROM were qualified by (CD-ROM) we would continue using that qualifier even if it switched to all online?

ANSWER: Yes, this policy is based on RDA (2 November 2010)

Variation in Media Type
If the media type varies within the serial (not a change from one type to another), do not create a new description. Make a note explaining the variation (RDA
Change in Carrier Type
If the carrier type within the same media type category changes, do not create a new description. Make a note explaining this change (RDA (8 June 2012, IG)

47. QUESTION: And if a series were produced both in CD-ROM and online, it would be considered one series because they are both computer media type. If a qualifer were needed to distinguish from the print, i'm not sure which carrier you would use in the uniform title. The first one you catalog?

ANSWER: No, this situation falls under #4 [in LCPS 6.27]. You have two manifestations and each stands alone. (2 November 2010)

UPDATED ANSWER: This situation falls under LCPS 6.27, section H, "Different carrier/media types," no. 4, which states: "If all parts of the series are each published in two or more different media types, consider that multiples series exist. Add a qualifier (e.g., the carrier type) if necessary to distinguish the series." (4 June 2012, KRT)

48. QUESTION: I am cataloging a multipart monograph, and v. 2 has a different publisher than v. 1.
Recently, LC and PCC implemented the repeatable 260. (Per LCRI 1.4: Apply the MARC 21 repeatable 260 field when there are changes in the publication, distribution, etc., information of a multipart monograph, serial, or integrating resource). However, RDA and RDA instruct catalogers to make notes regarding changes in the names of publishers, etc. For RDA should we follow the instructions and make a note regarding changes in publishers, or should we continue to use the repeatable field 260?

ANSWER: RDA doesn't specify what schema to use when giving RDA content. So, yes, overlay the RDA terminology of "note" with whatever field is appropriate for encoding that content in your schema: in MARC, the repeatable 260 for a change in publisher. LC did raise the situation you describe of using "note" in RDA when some schemas give content in fields other than notes (another example: using a 246 rather than a 5XX for a change in title proper of a monograph even though RDA says to give a note). (5 November 2010)

UPDATED ANSWER: Use the repeatable 264 field to record changes in the publication, distribution, etc., information of a multipart monograph, serial or integrating resource (20 June 2012, MWH)

49. QUESTION: I am cataloging a compilation of works by one author, and am attempting to apply faithfully. However, a number of issues give me pause.
1) Under AACR2/LCRI, my title proper would be more than "adequate", and I would not even be considering adding a collective title. I realize this notion has not been brought over into LCPS, and that that decision was most likely deliberate.

ANSWER: You're correct; that was a deliberate decision. If we want to test RDA, we shouldn't add too many "limiting" policies during the Test. (8 November 2010)


50. QUESTION: The alternative at to add a conventional collective title refers one back to .1 or .2. However, these two scenarios are not mutually exclusive, nor do they cover all cases. My compilation in hand could be said to consist of works in one form, that is, "prose works". However, "Essays" would not be an unreasonable choice (these are short-form philosophical studies). The same ambiguity exists, to my mind, with any short-form prose, especially in the humanities. What is an essay? I have selected "Prose works", since I'm more comfortable with its broad meaning than with the value judgment required to determine what "essays" are. I fear this situation will confront every cataloger applying this rule, and that the results will be inconsistent.

ANSWER: I agree that the results are likely to be inconsistent. If so, the results may point us, if LC/PCC implement RDA, in the direction of some specific policies. (8 November 2010)


51. QUESTION: Another possible collective title would be "Works. Selections." However, I cannot find a chain of logic in the RDA rules that would lead me to apply this title. AACR2 25.9 (i.e."if the person created works in one form only"), like the "adequate title proper" concept, has not been brought over. And since my works are all in one "form" (as tenuous as that notion is for this kind of literature), I seem to have no choice but to apply, selecting in this case "Prose works", then add "Selections" per the alternative at

ANSWER: I agree with your conclusion if you are going to apply the alternative. (8 November 2010)


52. QUESTION: In the spirit of the test, I will apply the title, and establish it in the NAF (per PCC). However, I am hard-pressed to see how this will help users. Perhaps LC might consider narrowing the scope of LCPS, in the similar way the application of tracing an analytical added entry for the first/predominant "Work manifested", was narrowed per the recent FAQ?

ANSWER: Despite what I said above about not limiting actions during the Test, it was the results of our practice record activities this summer that caused LC to modify its LCPS decision for RDA 17.8 given the impact on the number of LC testers who are doing all their cataloging in RDA for these three months. (8 November 2010)


53. QUESTION: A conference took place in a location with the following RDA access point: Bogotá (Colombia : Capital District). How should this appear in the $c of my conference access point? Under AACR2, per the 2nd paragraph of 24.4C1, I would not include additions to names prescribed in 24.6 (i.e. term indicating type of jurisdiction). However, I don't find a corresponding instruction in RDA/LCPS.
So, presuming the correct course of action is to include the jurisdiction term in the conference, how should it appear in the access point?
1) : $c Bogotá, Colombia (Capital District)
2) : $c Bogotá, Colombia, Capital District)
3) : $c Bogotá, Colombia : Capital District)
4) ??

ANSWER: I think the intent of the JSC was to end up with the same result as in AACR2. But, you're correct that an instruction to delete the RDA term is not present in RDA; the fourth paragraph in only says to replace the parentheses with a comma when giving the place as an addition in an authorized access point.
So, you can either apply RDA as written and pick your own choice of punctuation, noting the change from AACR2. Or you can ignore RDA gap and do what would be current practice. Either way, be sure to put a comment in your record survey. (8 November 2010)


54. QUESTION: Can I use "(Microfilm)" in my qualifier for a microfilm reproduction of a serial, to distinguish it from the original manifestation as well as from other serials with the same title?

ANSWER: We recommend that you use the reproduction publisher as a qualifier rather than the carrier. Publisher is a more common distinguishing characteristic between different resources. (18 January 2011)


55. QUESTION: My microform reproduction of a serial contains both titles of a title change under one umbrella title. In a record like this, is the S/L fixed field valid if coded "0" (successive entry)?

ANSWER: Yes. (18 January 2011)


56. QUESTION: The microfilm reels had chronological designations in their title frames (applying to the coverage of the reproduction), which I recorded in the 362. Do I still use 362 coverage information for Dates fixed fields instead of 260 $c information? Or, because under RDA the description is of the reproduction, should I use 260 $c?

ANSWER: An interesting question. However, we think the 008 dates should reflect the manifestation being cataloged (the reproduction). (18 January 2011)


57. QUESTION: In the record for my microform reproduction, can I include an 006 (to code for frequency and serial type of the original manifestation) and later/earlier title linking entries (78X) to reflect the original manifestation's title changes?

ANSWER: No, because fields other than 776 apply to what's being cataloged. This record should link to the originals, where the original frequency would be found; could also do a note about the original frequency in the record for the reproduction. Include [title change] info in the 776 field (that field is itself a complete description; just envision putting info from the record for the original in that field). Additional comment: It would be important to add 776s to the records for the originals to link all the titles to the reproduction; if only counting on recording the information in one record, may not be providing the information users need. (18 January 2011)


58. QUESTION: There is a lack of specificity on when to apply RDA The rule states very clearly "Do not transcribe words that serve as an introduction and are not intended to be part of the title." Catalogers and users of the Toolkit are not told if this rule is to be applied only to certain formats and there is no reason for them to think so. The four examples that are provided in the RDA toolkit do reflect only a few formats, but the rule as written does not state that these are the only ones to which it applies. If the rule is being applied only to certain formats, further instructions or clarifications should be provided.

ANSWER: As written, this RDA instruction does apply to all resources. During the draft stages, LC argued that it should apply only to moving image resources and to remote access serials/integrating resources. Our viewpoint was not supported by all others on the JSC.
If LC implements RDA, I don't know yet if we will have an LCPS for this instruction. (2 February 2011)


59. QUESTION: During the RDA test we were told that no BFM was to be reported to LC for RDA headings on AACR2 bibliographic records. Now that the test is over, is this still the case?

ANSWER: Yes, that is still the policy. LC won't make any decision about BFM until after the RDA implementation decision is announced by the three national libraries in June. (3 February 2011)

UPDATED ANSWER: I discussed your question with colleagues in PSD. We were not sure whether the 7xx RDA authorized access points added to AACR2 authority records during the test would automatically (and without exception) become 1xx RDA authorized access points, if RDA was adopted. That is why we did not require BFM on this category. Now, post-RDA test and pre-RDA implementation, we know that those 7xx RDA authorized access points will not automatically and without exception be made 1xx RDA authorized access points, there has been no change to this particular BFM policy (2 July 2012)

60. QUESTION: Is there a mechanism to recommend that a term be added to the list of relationship designators?

ANSWER: Send it to LCHelp4RDA with rationale for adding it, and LC will add the suggested term to their list for forwarding to the JSC. (17 February 2011)


61. QUESTION: The RDA Test "Train the Trainer" Module 4 slide set says on slide 19: "Simultaneous publication: different titles/languages - No longer a priority order of languages - use title in resource first received." I am having trouble finding the RDA rule that covers this.
RDA's instructions say to choose as the preferred title the title in the original language or the title proper of the original edition, but with a simultaneous publication in different languages, I can't determine the original language or original edition. then says: "If the work is published simultaneously in the same language under different titles, choose the title proper of the first resource received as the preferred title." It's this part of that confuses me because it's limited to works published in the same language with different titles.
I can believe that the "Train the Trainer" slide 19 reflects the intent of but where in RDA does it instruct me to do this (in the case of different language editions simultaneously published)?

ANSWER: We included the two possibilities (same language/different language) in the training materials because they were separate parts of AACR2 25.3C. You're correct that there is not a separate sentence in for different languages published simultaneously. That condition is part of the overall situation of identifying the original edition (2nd paragraph of The wording in slide 16 and slide 19 to use the title proper of the manifestation first received as the preferred title (unless there are specific instructions to do something different) is the equivalent of identifying the original edition. I agree that we need to make that point more clearly. Then if the same work is published later with a different title proper or published simultaneously with a different title in the same or a different language, the title proper of the first manifestation becomes the 130 or 240 in the bibliographic record for the other resource (with a subfield $l if language expression is involved). (25 February 2011)

UPDATED ANSWER: Per LCPS (recently revised), "If the work is published simultaneously in different languages, choose the title proper of the first resource received as the preferred title. When two or more language editions are in a compilation and the original expression cannot be determined (e.g., a document first issued with the same text in French and English), choose the title proper named first on the preferred source of information as the preferred title." (20 July 2012, CAM)

62. QUESTION: How do we code the preferred title access point for a part of a larger work?
Example: We have a new publication: The story of Turquoise, "previously published as part 2 of S.Y. Schwartz's Blue." The work, Blue, was originally published as part 1, Indigo, and part 2, Turquoise.

ANSWER: The following would be RDA as written (without $p in the 240 because Dave just reminded me that there are ISBD punctuation problems if we make it a $p):

100 1 Schwartz, S. Y.
240 10 Part 2, Turquoise
245 10 The story of Turquoise ...

However, we would bless the following application that violates RDA as written because we think that the RDA instruction wasn't updated correctly in the last review stage before publication:

100 1 Schwartz, S. Y.
240 10 Turquoise
245 10 The story of Turquoise ...

Barbara has agreed that I should do an LCPS to document this error in RDA. (25 February 2011)

UPDATED ANSWER: RDA as currently written has no such instruction, example or relevant LCPS [see for Recording the Preferred Title for a Part or Parts of a Work] (20 June 2012, MWH)

63. QUESTION: I was curious if LC during the RDA test dealt with this issue or continued with the policy outlined in the Shelflist Manual under G 140:
"Photocopy or facsimile editions. Add the date of the original edition and the work letter 'a'."

ANSWER: During the Test, we followed the policy you cite. If we change our reproduction policy (not connected to RDA; not sure when our cataloging managers will discuss -- may want to check with library community again because we received very few comments last summer), we'll look at our shelflisting policy as well. (1 March 2011)
SUL policy: MetaLawMus discussed our current practice for dates in call numbers for photocopy or facsimile editions, i.e., adding the letter "a" to the date of the original edition (per G140) and decided not to make any changes at this time. As with most other local RDA policies and practices, we will re-evaluate it after LC's and PCC's decisions are announced. (7 March 2011)


64. QUESTION: I am creating an authorized access point for a related work for which there are two creators.
The alternative for says: "Include in the authorized access point representing the work the authorized access points for all creators named in resources embodying the work or in reference sources (in the order in which they are named in those sources), formulated according to the guidelines and instructions given under 9.19.1 for persons, 10.10.1 for families, or 11.13.1 for corporate bodies, as applicable."
The LCPS tells us not to follow the alternative. Am I allowed to add an access point for the 2nd author and if yes which relationship designator would I use?

ANSWER: Stanford decided to follow the LCPS for, i.e., not to apply the alternative in the RDA rule. (JKD, 1 April 2011)

UPDATED ANSWER: Although Stanford and LC are not applying the alternative to add the name of a 2nd author in the access point for the work, it is permissible to add a separate access point for the 2nd author, with an appropriate relationship designator. This is an analogous approach to a resource embodying a single work with multiple creators. The first creator is given in the 100 (which, combined with the 245, constitutes the access point for the work), and any subsequent creator are given as separate access points (700s). (31 May 2012, CAM)

65. QUESTION: I am cataloging photocopy (reproduction) for which a parallel record exists in OCLC for the original manifestation. Do I add a 936 for the original manifestation or not?

ANSWER: I wouldn't, since the two records do not represent the same manifestation. (CAM, 4 April 2011)

UPDATED ANSWER: Per announcement from OCLC, May 18, 2012, catalogers should no longer use 936 to record parallel records (13 June 2012 GDG)

66. QUESTION: I am cataloging a photocopy (reproduction) for which I want to trace an issuing body. My question is what relationship designator would I use since the issuing body of the original manifestation did not issue the manifestation that I am cataloging.

ANSWER: I looked at Appendix I and "issuing body" is used for works and not at the manifestation level so I think that I am safe using it. (HGL, 12 April 2011)


67. QUESTION: I am cataloging a compilation of 2 works that had been issued separately, bound together by a former owner and then photocopied together by Stanford. So I am treating this as a compilation of two works which are reproductions of earlier manifestations that had been originally published separately. Under AACR2 we would make the following note:
Reprint (1st work)
Reprint (2nd work)
Reprint (3rd work)

In RDA I believe that I would only have to make an authorized access point for the first or predominant as a core element. I assume that also applies to access points for related manifestations. However, in many case other catalogers and I would prefer to provide access points for two or more works that may be contained within the manifestation. Similarly we would elect to provide formatted access points for more than 1 related manifestation.

I am looking at the LCPS for 27.1.3 and it tells me "If the manifestation is a compilation containing reproductions of works not published earlier as a compilation, give a bibliographic history note, rather than MARC 775/776 linking fields, about the individual works if you think that information would be helpful to the user."

Does that mean that I cannot make any MARC 775 linking fields? If I can, may I make more than one linking 775 field? If I can make two or more would I able to provide terminology that would indicate which number item in the compilation a particular 775 field relates to? We looked at various options on how record numeration of a reprint within a manifestation and one suggestion was the use of a repeatable $i at the beginning of the 775 field.

ANSWER: Your current compilation is a one-of-a-kind original (bound-withs are item-level relationships but, as you said, it is a work of works). That makes your photocopy also a one-of-a-kind reproduction. It also means you do have the information to be able to give 775 fields for the works used to make the reproduction. The MARC format info for subfield $i says its content may be either controlled or uncontrolled text; so, you could include an indication of which work in the reproduction is being referenced. You mentioned repeatable $i; I'd recommend repeated 775 fields, each with an appropriate $i.

The main reasons we said in LCPS that LC testers should not give, during the Test, the 775-776 fields for works/expressions in a compilation if they hadn't been previously published together were the difficulty in determining the original for each of the works/expressions and the time involved in doing so. As we said elsewhere, all policy decisions will be reconsidered if LC implements RDA. Also, remember that you don't have to follow the LCPSs.

I realize now that we would need to clarify the terminology in LCPS (and perhaps other places). When we said "manifestation" in the paragraph you're discussing, we really meant the resource being described and not that we see a compilation of works/expressions as an FRBR manifestation. A compilation is a work of works or a work of expressions.

You're correct that the core requirement for RDA 17.8/17.10 is the predominant or first-named work/expression. RDA does not yet include instructions for authorized access points for manifestations (just works and expressions are in chapter 6). That is why it is important to remember that even though you may have a one-of-a-kind compilation reproduction, the components of that compilation are still works or expressions and you would make any authorized access points for those FRBR entities. (11 May 2011)

UPDATED ANSWER:  LCPS, Update 4/10/2012: LC practice: When the decision has been made to create a separate record for the reproduction, generally follow the guidelines below for giving MARC 775 and MARC 776 fields. The guidelines do not apply when the “single-record” or “provider neutral” technique is being used.
Relationship to the original: on the record for the reproduction, use a structured description to give the attributes of the original.
1.MARC field:
a)Carrier of reproduction is the same as the carrier of the original: use MARC Field 775
b)Carrier of reproduction is not the same as the carrier of the original: use MARC Field 776
2. Use the relationship designator “Reproduction of (manifestation)”
Relationship to the reproduction for serials: on the record for the original, use a structured description to give the attributes of the reproduction; use judgment in determining how many relationships to give for separate reproductions on the record for the original. (Generally do not give this reciprocal relationship for monographs and integrating resources.)
1.MARC field:
a)Carrier of reproduction is the same as the carrier of the original: use MARC Field 775
b)Carrier of reproduction is not the same as the carrier of the original: use MARC Field 776
2.Use the relationship designator “Reproduced as”
If the manifestation is a compilation containing reproductions of works not published earlier as a compilation, give a bibliographic history note, rather than MARC 775, 776 linking fields, about the individual works if you think that information would be helpful to the user.
If the reproduction is a facsimile of a manuscript and attributes of the original for use in a MARC 775, 776 linking field are not readily ascertainable, give a bibliographic history note with the information you have. (8 JUNE 2012, IG)

68. QUESTION: I am cataloging a facsimile of a manuscript for which the old LCRI 21.30H would apply. That is, the work embodied can be described as its own entity, but an additional access point for the repository designation (which is given prominently on my t.p.) would be beneficial in my judgment. It appears most of the text of the old LCRI 25.13 has been carried over into LCPS However, there are no special instructions, analogous to LCRI 21.30H, for making an "added entry" (in RDA-speak, "related [something] access point") for the name of the physical manuscript.

Given that our local practice is to provide relationship designators whenever possible, my original inclination was to use "Facsimile of (item)", since a physical manuscript seems logically to me to fit the definition of a FRBR Item. What troubles me is that the rules in RDA for constructing an access point based on repository designation are in Chapter 6. This makes sense when the access point is needed to perform "double duty" for the work(s) and the physical manuscript. But, can I construe this access point as representing a related "item" when it's not needed to also represent the work embodied within it?

ANSWER: The answer is not in RDA/LCPS Chapter 6 Identifying Works and Expressions. It's in RDA/LCPS Chapter 28 Related Item. RDA Chapter 6 gives instructions on the preferred form of title whereas Chapters 24-28 give instructions on what in AACR2 would have been considered "related work added entries." Regarding expressing the relationship of the reproduction to the original, a 775 is used. RDA has some examples and the LCPS provides further guidance. This is an example of how it could be formulated:

775 08 $i Reproduction of (item): $a Rossiiskaia natsional'naia biblioteka. $t Manuscript. Sobranie A.F. Gil'ferdinga, no. 46.

or, using terminology from Appendix J.5:
775 08 $i Facsimile of (item): $a Rossiiskaia natsional'naia biblioteka. $t Manuscript. Sobranie A.F. Gil'ferdinga, no. 46.

Note that the manuscript numbering cannot be separately coded in the 775 (or 776) because of MARC format. However, I think this field may be of limited usefulness because it is so similar to the 710 usually given for manuscript repository, so I would recommend putting the "Facsimile of" in the 710 so it will look like this:

710 2# $i Facsimile of (item): $a Rossiiskaia natsional'naia biblioteka. $k Manuscript. $n Sobranie A.F. Gil'ferdinga, no. 46.

Using the 710 with subfield $i enables the cataloger to keep the manuscript information separately subfield coded AND get in the information about it being a facsimile.

LCRI 25.13 has this instruction "See LCRI 21.30H for situations in which an added entry for the physical manuscript is needed. See Subject Headings Manual H 1855 for situations in which a subject heading for the physical manuscript is needed." LCPS, which is the RDA equivalent to that RI, does not contain a similar reference to, and I think it's worth considering adding that next time LCPS is revised.

The JSC will be considering a discussion paper from LC for the FRBR Group 3 entities, including "object," which may provide an avenue for recording the distinctions mentioned and the disconnect in RDA that was brought forward from AACR2. (Kate James and Dave Reser, 13 May 2011)


69. QUESTION: I am working on a group of photocopies for which many items do not have cataloging copy in OCLC, or if there are records they are foreign/parallel records which are not cataloged under AACR2. Should I use abbrevations such as "p.", "v.", "ill." ... [et al.] and include a period after "cm" [when there is no series]? This might make sense when the related manifestation being described in the MARC 775 field has an AACR2 or pre-AACR2 OCLC record associated with it referenced in the MARC $w field.

Would it be preferable to write out "pages" ... [et al.] if no related record has been found in OCLC, or if only a parallel record has been found?

ANSWER: If there is a record for what is being cited in the linking field, give info as found in that record. If there isn't a record, Dave Reser and I would recommend following RDA conventions on abbreviations and ISBD punctuation conventions. If we're hoping to make the linking field via macro, it will be dependent on the info in the target record without further manipulation of data. (21 June 2011)

FOLLOW-UP QUESTION: Does this include the abbreviations and punctuations found in linking records that are not AACR2 such as those found in foreign parallel records?
FOLLOW-UP ANSWER: Yes, I think so. (21 June 2011)


70. QUESTION: I am cataloging an edition of a Jewish religious work "Mishnah. Avot" containing an interlinear translation (English and Hebrew) that has the following information on the page preceding the title page "A project of the Mesorah Heritage Foundation." This body is often traced on AACR2 bibliographic records for pieces on which this statement appears. I would like to trace it as well but I am not sure if I can use the relationship designator "sponsoring body" since that is only available for use as a Relationship Designator for persons, families, or corporate bodies associated with a Work.

The website for the foundation states that it was created to "... to remove the language barrier and make the riches of Jewish eternity available to English-speaking Jews:
Here is what we do:
Sponsor literature that celebrates the rich Jewish heritage"

It seems clear to me that sponsoring bodies can support the creation of new editions and versions of works as well as the works themselves. My preference is that the relationship designator be assigned for use for expressions as well as works.

ANSWER: I've included John Attig in this reply because you/he can work on an addition/change for appendix I. Meanwhile, you can give an access point for the foundation without the relationship designator if you think the group's responsibility is not at the work level for the resource.

Also discussed at LC during the US RDA Test was that some of the sponsoring bodies are sponsoring the event and not the resource resulting from the event. I've not been happy with the definition in appendix I including the e.g. wording "sponsoring an event;" it is mixing two levels of sponsorship. (21 June 2011)


71. QUESTION: I recently encountered a bit of dilemma when cataloging a resource that consisted of several short works by different persons, with extensive commentary by one person. The chief source gave a statement of responsibility that conveyed editorship rather than authorship ("sous la direction de..."), and, being a compilation, had a collective title. Thus, it presented itself neither as an edition of a work with commentary, nor unambiguously as a work of the commentator. This led me to consult the applicable rule in RDA,, with disappointing results. The wording in the first paragraph, "If the work consists of a previously existing work...", seems to suggest that this rule is limited in scope to resources embodying a *single* work, not compilations.

As an interesting point of comparison, the corresponding rule in AACR2, 21.13 gives a scope note (which, as it happens, restricts application to "a text, or of texts, by the same person or body") as well as an added instruction for ambiguous situations. Neither of these are present in RDA.

Ultimately, I opted to apply and treat is as a compilation, since I did not feel confident that I could "legally" apply the first paragraph of and treat it as a work of the commentator. But did I miss something here? Is the intention in RDA to carry forth the limited scope of the "commentary emphasized" instruction, or to allow it to be used more broadly? If the latter, I think clearer language would be in order.

ANSWER:A compilation is a work (i.e., a work of works); some use the terminology "aggregate work." So, the wording "work" in doesn't exclude a compilation automatically. I think the deciding words in this instruction are " consists of a previously existing work with added commentary, annotations, illustrative content, etc." Was the compilation published before without commentary?

Further info from Dave Reser: is supposed to help you decide, in a case where you have an original work and a commentary, whether to treat it as an expression of the original work or a new work in its own right. AACR2 21.13B and 21.13C pretty much required you to accept as 'fact' how the publisher represented it on the title page as to which side of the fence you landed on, you didn't really get to 21.13D (Chief source is ambiguous) if you had a clear statement on the chief source. RDA allows you to look beyond the chief source, it just uses the words "presented as" that would allow the cataloger to use their judgement of the 'facts' (including those that might be covered by AACR2 21.13D1 (preface, presentation, extent of text)) rather than limited to a statement of responsibility alone. (13 September 2011)


72. QUESTION: I am concerned that the definition or meaning of "prominently" is not found anywhere in RDA as it was in AACR2 0.8. Are there plans to include a definition of the term somewhere within RDA as a result of the rewrite?

ANSWER: The AACR2 term "prominently" applied to specific sources in the resource. In RDA the term has only its dictionary definition; I don't know of any plans to add a definition because the JSC general policy is not to define such terms. (23 September 2011)


73. QUESTION: I'm trying to come up with a systematic way to logically structure an access point containing both a language term and "Selections." RDA/LCPS appears to give no guidance as to the order in which to place these sub-elements in access points. As a point of departure, here are the conventions in AACR2 that guide my thinking:

Works. French. 2010
Selections. French. 2010


Poems. French. Selections (following AACR2 25.11)

For the most part, these conventions work fine in RDA practice, with the following exception: "Works. Selections". Should it be:

Works. Selections. French. 2010
Works. French. Selections. 2010

The first way is a logical extension of the syntax of "Selections. French. 2010", that is, language follows Selections. (Put another way, "Works. Selections" as a unit replaces "Selections") The second way, however, is a logical extension of applying the old AACR2 25.11, that is, language *precedes* Selections.

A cursory search of RDA records in WorldCat shows both syntaxes being used. Is this an inconsistency we'll have to live with? Or am I missing something?

ANSWER: You bring up a very thorny issue that we have thought a lot about-- but for now, the conundrum continues while the JSC prepares for its November meeting in Scotland to look at this very question.

At the risk of confusing you any further, please take a look at a rather lengthy discussion paper that LC prepared for the upcoming JSC meeting:

6JSC/LC rep/2 "Selections" as used in RDA Chapter 6

The original discussion paper, and the JSC constituency responses to it, are available at that link. The discussion paper attempts to frame the question in light of the FRBR attributes for "work" and "expression" in order to determine whether "Selections" as used in RDA is (or should be) at only one or both of those entities. LC's premise is that it is an attribute of the work (preferred title for a part or parts of a work). Following the RDA practice of recording work attributes in an authorized access point *before* recording the expression attributes, "Selections" would precede the language if the logic is approved. If the final answer comes back differently, as well it might, at least RDA needs to make a clearer statement than it does today.

Long way to say that you've stumbled into an interesting problem, and one for which there isn't a final answer yet. As you say, you see examples both ways, that would be understandable and acceptable at this point.

Hope this helps, and doesn't raise more questions than it answers!


Authorities Q&A

A1. QUESTION: While creating a brand new NAR to record RDA form/s of name, should we also record AACR2 form if different?

ANSWER: No. (30 August 2010)


A2. QUESTION: We were wondering whether there are any special steps to take in case the new RDA form recorded in the 7xx field in the AACR2 NAR happens to be the same as one of the existing 4xx references.

ANSWER: Following the mantra of "do what you've always done," you would need to break the conflict between the 4XX and the to-be-if-implemented heading. The practice now is to break the conflict by modifying the reference, rather than the heading. If you do not have appropriate information to add to the 4XX, leave the 4XX as is. We'll get a report, if we implement RDA, that lists the conflicts between 1XX and 4XX. (12 October 2010)


A3. QUESTION: I need to add a 7xx for the RDA heading for Germany (East) in NAR n 80-125938. The headings for both Germany (East) and Germany (West) were formulate according to special decision #2 in LCRI 23.2. Now according to RDA, one should use "the form of the name in the language preferred by the agency preparing the data, if there is one in general use." That form would most likely be East Germany or the German Democratic Republic. This would mean changing a lot of headings in authority and bibliographic records. A further complication is that both Germany (East) and Germany (West) can be used in subject headings for works about these parts of Germany before 1949 and after 1990. Has anyone given any thought of applying the old decision for Germany (as well as some of the other decisions in RI 23.2) to RDA?

ANSWER: When we were considering what content from LCRIs to move to LCPSs for the US RDA Test, we decided to ensure that the special decisions in LCRI 23.3 were recorded in the appropriate NARs rather than force people always to check the LCRI/LCPS. So, one of the general guidelines in the document for using existing authority records during the Test says to consider the information already in the NAR when determining the RDA form. (5 October 2010)


A4. QUESTION: We also came across a preliminary AACR2 NAR. Except for the addition of the 7xx, should we leave this NAR as is?

ANSWER: The instruction in DCM Z1 008/33 to upgrade a preliminary NAR still applies. (12 October 2010)


A5. QUESTION: I have established a corporate NAR using RDA. I have added a 510 $w a for the earlier form of name. The earlier form of name has been established using AACR2. I assume I add a 510 $w b for the later (RDA) form to the AACR2 NAR.

ANSWER: Yes. (12 October 2010)


A6. QUESTION: Is there any way to indicate that the 510 in the AACR2 NAR is for an RDA heading?

ANSWER: No. (12 October 2010)


A7. QUESTION: Would I add a 7XX to the AACR2 NAR record (earlier form of name) even though that heading is not used in my bib record?

ANSWER: You could give that 7XX if you want to take the time to check usage, etc., for that earlier name. We are not requiring the inclusion of a 7XX field in the earlier record. (My recommendation would be not to give such information.) (12 October 2010)


A8. QUESTION: As we add 7xx fields to geographic NARs, most forms of heading won't be any different. However, I notice that at RDA, the LCPS instructs to add the name of the jurisdiction in English. However, the corresponding instruction under (regarding constructing an access point with a jurisdiction qualifier to break a conflict) does not repeat the instruction to record the qualifier in English. Furthermore, the examples given include a number of vernacular forms (e.g. "Landkreis"). Are we to construe that the LCPS at applies to this instruction? This will be problematic in some cases, as there are often not direct English translations for the names of some of these types of jurisdictions. (14 October 2010)


A9. QUESTION: Page 92 of the NACO Participants' Manual addresses National Library of Medicine Records, specifically when the NLM heading has precedence over the LC heading.
In my case, I need to add 7XX to n87910123 which cites the NLM files in 670. Can I add a date-of-birth to my 7XX if the NLM heading does not include the date?

ANSWER: The NACO Participant's Manual has not been updated since 2005 and many things have changed since then. You should be consulting DCM Z1 for procedures; the special exception for NLM headings is no longer in DCM Z1. (21 October 2010)


A10. QUESTION: Do I ignore anything in the NACO Participants' Manual that doesn't appear in DCM Z1?

ANSWER: We should not consult the NACO Participants' Manual anymore. (JKD, SUL MDU) (21 October 2010)


A11. QUESTION: After conducting an in-house RDA training in August, we made a local decision based on the information from the training slides (Module 7, slides 12-13) to:
1) include a date in access points for corporate bodies other than conferences only if needed to break a conflict;
2) record dates in a 670 field (and not in a 046 field).
After reading your Frequently Asked Questions: #10, we realized that perhaps our no. 2 decision above is not correct and we should always be adding the date/s in the 046 $s and/or $d (alone or in addition to the 670). Could you please confirm that this is indeed what we should be doing?

ANSWER: Thanks for writing about the info in the FAQ. I was much too dogmatic in that answer. No one is required to use the 046 field. The answer should have said in an 046 or 670 field per cataloger's judgment or local policy. I'm getting ready to submit some revised versions of other documents; I'll update the FAQ at the same time. (22 October 2010)


A12. QUESTION: We are tracing a series already established under AACR2 in NAF (lccn n 93108646). The numbering restarted. We need to add [nuova serie] to the numbering. Per RDA Appendix B, we cannot abbreviate "new series", "nuova serie", and similar terms. In the AACR2 NAR, do we add a 642 using the AACR2 form, i.e.: [nuova ser.], 2 ?
Do we add a 642 using the RDA form, i.e.: [nuova serie], 2 ?
In the RDA bib we will use the RDA form in 830 $v. We want to know which form to add as a 642 in the authority record.

ANSWER: A good question. Let's give the AACR2 form in the SAR 642. (27 October 2010)


A13. QUESTION: I am adding an RDA form for a series title that was established with no qualifier. Since that time, another series was established with a reference that did conflict, and thus a qualifier was added. However, the original series heading was not revised to add a qualifier, since the conflict was sufficiently broken by qualifying the new series reference.
My question is this: in adding a 7xx for the RDA form of the older series, must I take into account the present-day conflict and add a qualifier? Or, does the presence of a qualifier on the other heading/reference mean that I don't truly have a conflict at this point, and thus can leave the 7xx unqualified (as it is in the extant 1xx)?

ANSWER: I think we'd still say to break the conflict between an authorized access point and a variant access point by adding a qualifier to the latter. If the conflict is between an authorized access point and an authorized access point, a qualifier would be needed on the later one being established in new SAR. (5 November 2010)
INTERPRETATION OF ANSWER: Inferring the answer, it seems I can leave the 730 unqualified, since my SAR is still the "earlier" one.


A14. QUESTION: In NAR 670s, do we transcribe as found (capitalization, etc.)?


UPDATED ANSWER: Generally, the capitalization, etc. in the 670 should match the transcription in the corresponding bibliographic record. (13 June, 2012, RJR)

A15. QUESTION: We have a question about adding the RDA form of heading to existing AACR2 NARs for geographic names. LCPS says: "Conflict in place names is not restricted to those already represented in the file against which the searching and cataloging is being done. Instead, if there is no existing conflict in that file, search gazetteers, etc., to determine if two or more places within the same jurisdiction have the same name or if two or more places with the same name would bear the same qualifier for the larger place."
US RDA Test Policy for the Extra Set: Use of Existing Authority and Bib Records says: "In addition to information in the resource being cataloged and any research required by RDA or LCPSs, use the information found in the existing AACR2 authority record when determining the preferred name and considering additions to that name to create the RDA form."
A great many place names established in the NAF would have different forms in RDA. The implications for file maintenance are so vast that we would like clarification before proceeding. For example, there is only one Cali (Colombia) in the NAF but more than one Cali in Colombia according to Geonet. The NAR cites an "Old catalog heading" but no gazetteers, etc. In this case, should the 751 match the form found in 151 or should it be formulated uniquely, i.e. Cali (Valle del Cauca, Colombia).
Same with Bogota. There is only one in the NAF but more than one in Colombia according to Geonet. The NAR cites an "AACR 1 form" but no gazetteers, etc. Should the 751 match the form found in 151 or should it be unique, i.e. Bogota (Capital District, Colombia)?

ANSWER: If we're going to test RDA, then we do need to make the RDA forms reflect what we know now about the entity. When the Test is over and if RDA is implemented by LC/PCC, we'll need to look at situations such as those you describe and decide if any policies need to be changed, if any situations would fall under "RDA-compatible" headings (don't know if we'd have such categories under RDA or not), etc. (5 November 2010)

UPDATED ANSWER: We are no longer adding the RDA form to AACR2 NARs although we are recoding them as RDA on a case-by-case basis (21 June, 2012, MWH)

A16. QUESTION: Given that we are endeavoring to follow LC practice in matters of encoding RDA authority records, we are hoping for a clarification of the following situation: According to the RDA Examples: Authority Records document and the MARC21 Encoding document, LC will not be using relationship designators in cases of three or more pseudonyms. However, neither document states explicitly whether LC will use them in cases of just two related identities. The Encoding document (end of p.6) states to use $w for pseudonyms and $i for other relationships. Previously, $w was not used in 5xx fields linking just two personal names, so I'm not sure if "pseudonyms" in this sentence refers only to cases of three or more pseudonyms, or all cases of pseudonyms. In any event, I have erred on the conservative side, and have not included "$i real identity $w r" in the 500 of my RDA NAR. We are curious to know what LC's practice will be for this.

ANSWER: LC's practice for the Test is to use $w nnnc for pseudonyms in cases of three or more and not to use $w when there are only two identities. (29 November 2010)


A17. QUESTION: How to establish brand new RDA authority records if that form differs from the AACR2 form on the related authority record? E.g. (example from the "testing guidelines"):

AACR2 authority record
100 1_ $a Brown, George, $c Rev.
700 14 $a Brown, George $c (Clergyman)

New RDA authority record
100 1_ $a Brown, George $c (Clergyman). $t Poems


      The 1XX forms from existing AACR2 authority records are to be used in bibliographic records. This is true even when the AACR2 form represents a "building block" in a new heading being formulated.
      RDA forms may be added to 7XX fields in the authority records, particularly when the heading form would be different. Examples:
      AACR2 authority record
      100 1_ $a Brown, George, $c Rev.
      700 14 $a Brown, George $c (Clergyman)
      New heading needed for cataloging
      100 1_ $a Brown, George, $c Rev. $t Poems
      700 14 $a Brown, George $c (Clergyman). $t Poems
      AACR2 authority record
      110 2_ $a Royal College of Physicians of London
      710 24 $a Royal College of Physicians (London, England)
      New heading needed for cataloging
      110 2_ $a Royal College of Physicians of London. $b Tobacco Advisory Group
      710 24 $a Royal College of Physicians (London, England). $b Tobacco Advisory Group
    (December 10, 2010)


      It appears best to follow AACR2 for the entire heading, if any part of it is AACR2, and to code it as such. Hybrid headings could turn out to be too challenging to get flipped into complete conformance to a single code, once an implementation decision is reached.
      If the "building block" you are using is RDA, keep the entire new heading in RDA. You may consult for help. Adding 7XX fields in authority records is considered optional.
      If you encounter a situation such as Bob Maxwell described today, where the AACR2 heading and the RDA heading are not equivalent in scope, and a portion of the required RDA heading would conflict with an established AACR2 heading, use AACR2 to catalog the item during this interim period.
    (December 16, 2010)

UPDATE: Follow the PCC Post RDA Test Guidelines at

UDATED ANSWER: Follow the PCC Post RDA Test Guidelines at

A18. QUESTION: I need to add the RDA form of heading to NAR for Lubumbashi (Congo).
The LCRI for 23.4B under Form of Larger Place says: If the heading for the larger place being added to the smaller place is established according to the provisions of 24.6, do not include in the qualifier the additions prescribed in 24.6. [See 24.6C: If the type of jurisdiction does not provide a satisfactory distinction, add an appropriate word or phrase.]
The LCPS for under Form of Larger Place says: If the heading for the larger place being added to the smaller place is created according to the provisions of RDA (type of jurisdiction), do not include the term for type of jurisdiction.
But what about Other Designation Associated with the Body? In my case, since the addition is not a type of jurisdiction, it seems the RDA heading would be: Lubumbashi (Congo, Democratic Republic). Is this correct? Not sure if the discrepancy between LCRI and LCPS was intentional or an oversight.

ANSWER: I don't remember now why we wrote the LCPS the way we did a year ago now. So, the difference may be either intentional or just an oversight. I agree with your choice for the RDA heading based on RDA/LCPS. (03 January 2011)

UPDATED ANSWER: The LCPS for under “Form of Larger Place” has been revised to include “other designation associated with the body” with an example: Seoul (Korea) not Seoul (Korea (South)). Hence, my heading should be Lubumbashi (Congo). (5 July 2012, MWH)

A19. QUESTION: According to RDA, each meeting of an ongoing conference is established separately. During the RDA test, we were adding 7XXs for each meeting of an ongoing conference to an existing NAR for a conference established according to AACR2. Now that the test is over and the addition of 7XXs for RDA forms of headings to existing AACR2 NARs is optional, we are a bit unsure of what to do when we are cataloging (according to RDA) the proceedings of an ongoing conference that has already been established as an AACR2 heading. Do we
1) use the AACR2 heading as established (and optionally add a 7XX reference)
2) or do we establish it as a separate RDA heading on its own NAR?

ANSWER: Your query (an excellent one - we've been wondering the same thing if we implement RDA) is a PCC policy decision. I've contacted the PCC Steering Committee and will respond when I have more info. (20 January 2011)
UPDATED ANSWER: Conversations with the PCC chair, the Coop Cat Team here, etc., have resulted in a PCC policy decision that will be added to the recently-posted PCC statement: "If an AACR2 name authority record exists for an ongoing conference constructed according to LCRI 24.7B (i.e. the heading for the conference does not include the number date or place of any one specific conference) use the AACR2 form as the basis for making additions to the name in bibliographic records for an individual conference." (31 January 2011)

UPDATED ANSWER: The PCC TG has recommended the creation of collective and individual NARs (see RDA PCC Proposed Guidelines and Standards at: The recommendation has not yet been implemented. (30 July 2012)

A20. QUESTION: During the RDA test period catalogers were not to create an RDA differentiated name authority for one of the individuals on an undifferentiated NAR. Is this still the case, or can we go ahead and create the new RDA authority record?

ANSWER: I think your question falls under the PCC policy to continue using AACR2 existing headings. Part of the problem is breaking out one individual from an undifferentiated name requires BFM on AACR2 records and some of the possibilities for differentiating a name in RDA are not available in AACR2. So, until we have an RDA implementation decision, we don't want to give an RDA form on an earlier AACR2 record.
UPDATED ANSWER: If an existing AACR2 authority record for a personal name is an undifferentiated name record and there is now a date of birth, a date of death, or a fuller form of name for the person related to the resource being cataloged, create an RDA differentiated name authority record for that person, remove the appropriate 670 fields from the AACR2 undifferentiated record, and report the necessary bibliographic file maintenance. For other situations, use the AACR2 undifferentiated authorized access point in the RDA bibliographic record; do not add any 7XX fields for RDA forms to the undifferentiated AACR2 authority record. (Not all RDA differentiated forms would be appropriate forms in existing AACR2 bibliographic records if RDA is not implemented.) (PCC, 11 February 2011)

UPDATED ANSWER (2 June 2012): If an existing AACR2 authority record for a personal name is an undifferentiated name record and there is now a date of birth, a date of death, a fuller form of name, or a field of activity or occupation for the person related to the resource being cataloged, create an RDA differentiated name authority record for that person, remove the appropriate 670 fields from the AACR2 undifferentiated record, and report the necessary bibliographic file maintenance. For other situations, use the AACR2 undifferentiated authorized access point in the RDA bibliographic record. Do not add any 7XX fields for RDA forms to the undifferentiated AACR2 authority record. (PCC Post-Test Guidelines, 4 June 2012, KRT)

A21. QUESTION: In cataloging of Canadian imprints we should continue using (creating as needed) AACR2 headings?

ANSWER: Yes, until Library and Archives Canada implements RDA, any records using its forms of heading should be AACR2 records. (23 February 2011)

UPDATED ANSWER: This policy has been discontinued. We no longer take into account the LAC form of heading when establishing a NAR in RDA (28 June, 2012, MWH)

A22. QUESTION: Should we remove the RDA-related 046, 37X, and 38X fields/subfields/coding from AACR2 NARs?

ANSWER: The 046, 37X, and 38X fields are not restricted to RDA; they can be included in AACR2 records (actually, in any authority record encoded in MARC). (23 February 2011)


A23. QUESTION: Although not yet addressed by PCC, I would like some thought on how we would cite a 670 for a reproduction (related manifestation) in which the date of publication of the manifestation of the resource being catalged is the date of reproduction.

ANSWER: SUL decision: Give information about the reproduction in $b of the 670, e.g.: 670 Kinder fun der Yidisher komune, 1991?: $b photocopy t.p. (???? ?????? = Line Nayman) (10 March 2011)


A24. QUESTION: I'm having a hard time reconciling two seemingly contradictory policies: I'm establishing a heading for an ongoing conference whose NAR, in RDA, would include qualifiers. Since the conference is not already established, the PCC post-test policy would have me establish it according to RDA. However, since I'm using a local place name (which is established as AACR2) as a qualifier, this would constitute a "building block", that would then trigger another post-test policy not to create hybrid headings. Thirdly, if I revert to AACR2 because of the "building block", then the NAR is in conflict with LCRI 24.7B, which instructs to omit qualifiers from the NAR for an ongoing conference. What to do?

ANSWER: You raise an interesting point, and one that does seem to intersect two contradictory policies. I am maintaining a table of PCC decisions that need to be made, and topics that need to be addressed, if RDA is adopted. I will add your point to the table. In the meantime, though, I suggest that you do not create a hybrid record in this case, and establish the conference under AACR2 24.7/LCRI 24.7B. (Paul Frank, 3 May 2011)

UPDATED ANSWER: Until RDA Day One for Authority Records (3/31/13), PCC Post-Test policies are in effect. Thus, a conference access point that would include a place name established under AACR2 would have to be established under AACR2. However, as NARs for place names are recoded to RDA (which will be happening piecemeal between now and 3/31/13), conference access points including those place names may be established under RDA. In any case, one still must check the NAR for the place name in question before establishing the conference access point. (1 June 2011, CAM)

A25. QUESTION: I am trying to create an authority record for the musical work l'aura che trema by John Croft, and am uncertain what to do about the A4.2 says "Do not capitalize the first word of a title if it is preceded by punctuation indicating that the beginning of the phrase from which the title was derived has been omitted" and has the example ...and master of none.

This rule, however, is about manifestations, not works, and is really about capitalization, not the punctuation ... I can't find the equivalent or rule specifically addressing the use of the ellipsis for an authorized access point for a work.

ANSWER:We agree that there isn't any specific instruction in RDA about punctuation in the preferred title for the work. So, we'd say follow the basic principle of representation and include the punctuation. Be sure you take the punctuation into account when you assign the nonfiling indicator. (26 August 2011)


A26. QUESTION: We would like to know if the "no hybrid headings" post-test NACO guidelines include unestablished access points (such as 245s and 130s) from AACR2 bibliographic records. If this is the case, then, for example, if we need to create a uniform title NAR for a translation of a work entered under title, if that work is represented in the bibliographic file, is it true that we can't make an RDA uniform title NAR (because it would be "hybrid")? We did not understand from the PCC post-test guidelines that any AACR2 information -- even if it comes from a bibliographic record -- makes the authority record a "hybrid" and therefore can't be established under RDA. Please clarify.

ANSWER:You've identified a gap in the PCC policy statement for this interim period (thanks for writing to us about this gap). So, what we need to do is add this category to the PCC guidelines. Yes, we'd probably -- for the sake of consistency -- state the policy as you've described it: if the bibliographic record is standing as the placeholder for a name authority record for the resource, we have to pay attention to whether the bibliographic record is an AACR2 record or an RDA record. If the bibliographic record is an AACR2 record, any authority record based on it (whether for a translation, supplement, subseries, etc.) would have to be an AACR2 record for now. (10 November 2011)


A27. QUESTION: Before RDA, if the main series was unnumbered (and therefore did not stand independently of the subseries on the publication in hand) we did not establish a SAR for the main series. Has this changed for RDA and we must establish an unnumbered main series based only on the subseries? When I have an unnumbered main series + subseries, I think the entire string is the series. I don't think it's 2 separate series. Has this changed in RDA?

ANSWER: There isn't any different understanding for subseries SARs of how to create and how many to create for RDA. (10 November 2011)


28. QUESTION: We have noticed the change in this LCPS regarding when to qualify access points representing works; specifically, it appears that for monographs, conflicts must now be broken in all cases. Aside from the fact that this constitutes a significant change from previous practice (which generally exempted monographs entered under title), we are a bit puzzled as to how this new policy relates to existing NACO policy regarding when to establish titles in the NAF.

In my case, I have a conflicting title which requires a qualifier. However, none of the other conditions which trigger the requirement to create an authority record (namely, need for a reference, need to document special research, need to make a "placeholder" for a non-analytic related work) apply. In my bib record, I'm using this title as an "analytic"; at least, that's what we used to call it. It's important to note that the new LCPS does not seem to distinguish between "analytic" related works (i.e., whole-part relationships) and other types of related works, so I'm wondering if I can even use the NACO guidelines on Uniform Titles as guidance anymore. Could it be that the intended practice going forward is to establish all preferred titles that need qualifiers, whether tagged as a 130 or 730, and whether used as an "analytic" or otherwise?

I realize it may be some time before PCC makes a revision to the Uniform Titles FAQ, given the current volume of high-priority work, so in the mean time we're interested to find out what LC's practice will be, so that we may continue to meet both LCPS and PCC requirements as best we can in our monographic cataloging.

ANSWER: Yes, differentiating all works is a change from AACR2 practice. Our returning RDA catalogers will report the impact to management by March for a permanent decision for LC.

However, this change does not signal a change in when LC RDA catalogers will make title or name/title authority records. The conditions in DCM Z1 Introduction still apply (need for a reference, etc.); so, most of the time an NAR will not be made when the authorized access point is for a work in manifestation being cataloged (either a single work or a work in a compilation). If the authorized access point is for a related work, an NAR is not made as long as there is a bibliographic record for that related work. (29 December 2011)


29. QUESTION: In the process of recoding NARs for access points that are acceptable for use under RDA, we came across a vexing situation regarding series access points beginning with the authorized access point for a person. The access point in question is:

Schoeps, Julius H. ?q (Julius Hans), ?d 1942- ?t Deutsch-ju?dische Geschichte durch drei Jahrhunderte [no2012072878]

The series "work" here is a compilation of works by a single author. In RDA, such a compilation would warrant a conventional collective title (per the alternative at; "Works. Selections. 2010" in this case). Thus, we believe this series access point is not acceptable for use under RDA (i.e., there is no option to identify the compilation by a title proper which is "adequate", as there was in AACR2/LCRI).

As a short-term solution, we decided to leave the SAR alone, use the collective title in our RDA bib record for the series as a whole (a terminal set), and code the bib record non-PCC. Going forward, is this a case in which we should routinely modify the 1xx of the SAR in order to recode it as RDA, so it may back up the access point used in the RDA bib record? Or should we continue our alternative approach for the time being?

ANSWER: (from Jessalyn Zoom) I asked PSD about this unusual situation and we had some discussions.  PSD decided to consider to apply for "commonly known title" instruction in RDA ( and do not update or recode the AACR2 heading for an RDA record.  The example of this case would be Walt Whitman's Leaves of Grass.  So we will continue to use the AACR2 heading established under LCRI "Adequate" title rule and treat the title as commonly known title under RDA for the time being, and use it in RDA bib record.  So no changes would be made to the existing AACR2 name/title NAR and use it as is in your contributed RDA record coded PCC. (26 July 2012)