Proposal: Transcription issues associated with the Production Statement (RDA 2.7)

Tagged , , . Bookmark the permalink.

One Response to Proposal: Transcription issues associated with the Production Statement (RDA 2.7)

  1. Matthew Haugen says:

    Digest of feedback from DCRM commenters:
    We recognized that this is an area of considerable difference of opinion within the DCRM community as well, and many of the concerns brought up in JSC constituency responses are familiar.
    Several commenters wished to reiterate that unpublished resources frequently don’t bear any self-identifying information at all, or they might have vague, incomplete, illegible, false, or misleading information that was not even intended to be production information.
    Commenters also wished to clarify that the proposal does not intend to forbid transcribing information when the resource is authentically self-describing, and rather is simply not mandating transcription. Indeed, we definitely want to retain the option to transcribe statements, whether instead of or in addition to recording the data in a non-transcribed manner – but RDA needs to accommodate both options, and do so in a way that doesn’t marginalize the non-transcribed data (as is currently the case), justified especially by the nature of non-published resources.
    This is more complicated and requires more judgment than docilely transcribing everything that you see, but cataloging unpublished resources IS often (not always) harder than cataloging published resources. But there is more than one way to accomplish this.
    If we keep to RDA’s current prescriptions – noting that the data is supplied – we would have to enclose the identifying information in square brackets: for title, place, date, etc. in almost all cases. These square brackets add up to a lot of unnecessary noise, and the meaning of the brackets probably escapes a large number of our users, especially for resources for which data is almost always supplied. There is a very good reason, thus, that the archives and art cataloging communities do not give priority to transcription. Others felt that departing from the current RDA prescriptions by making an exception for unpublished resources may be misleading to patrons who may not understand the difference in treatment, especially if it’s not indicated in any other way (e.g. square brackets, manuscript GMDs, etc.), and especially because not all systems display 264_0 as “Production Statement”.

    Meanwhile, although advocates for the proposal reassert that the Production Statement should be a recorded, not a transcribed element, this also does not preclude recording in this element production information that appears on the resource. We want to discourage transcribing for the sake of transcription information that is not accurate/helpful, etc. but if it is there, and it is useful, feel free to use it. The cataloger also has the option to make a note indicating that the information appeared on the item, if considered important (“Place of production transcribed from p. 5 of manuscript/recto of drawing/label on base of sculpture/etc.”) especially since those unpublished resources which do have useful self-identifying information often do not have that information in predictable/customary locations. Or, one can make a note quoting information that appeared on the item that was NOT used in the production statement (NOTE: “Parisi” handwritten below the drawing, but we know that it was created in Vienna.

    If the discussion at CC:DA remains in favor of transcription, other options could be considered, including:
    Retain current instructions (transcription), but add an exception for cases where the information on the piece is incorrect, misleading, or vague to supply information (and bracket). Notes for specifics stay the same.
    Prescribe supplying/recording information (with brackets), but add an exception to transcribe from the piece itself if the information is self-identifying, clear, and correct.”

    We also recognize that the outcome of this proposal is intertwined with that of the 6JSC/BL rep/1 proposal. Some noted that the sharper distinction between transcribed statements and related entities advocated by that proposal leaves out the middle ground of recording production information that is often so vague that it can’t be associated with a specific place or date (“at sea” comes to mind, or “at camp near [placename illegible, but starts with Fra” or “between 1781 and 1794?”) or so obscure that it would require a great deal of additional research or authority work to identify and record in a controlled form versus simple transcription. Others found this to be more manageable. The “at sea” example might have as controlled entity something like “Atlantic Ocean,” with a Details on … note to provide more information, if needed. Ambiguous dates could still be handled in a controlled manner, perhaps in a way resembling what we do in MARC fixed fields or 046 – and here again, with a Details on … note to provide more information.

    If the 6JSC/BL rep/1 proposal is rejected, this proposal offers a way forward for emphasizing recorded data for unpublished resources while allowing transcriptions when appropriate. Alternatively, if some form of the BL proposal is adopted, we might consider leaving 2.7 as the element(s) to be used for pure transcription – but it should not be required/core or filled with extraneous [xxx not identified] fields for non-self-describing resources. If RDA establishes place and date as related entities (as in the BL proposal), together with the existing element for Producer (21.2), I think this would satisfy our needs for non-transcribed information (provided we have access to Details on… elements to elaborate on data that doesn’t fit neatly as a related entity). This approach would accommodate consistency with guidelines for Publication, Manufacture, and Distribution.

Leave a Reply