Background and introduction
Vernacular parallel script is a non-Roman script for display along with the Roman script in cataloging.
Beginning 2004, OCLC made available some non-Roman language scripts besides Chinese, Japanese and Korean (CJK) which have been available for 20 years. An ad-hoc committee was formed to discuss the standard for implementation at UC Berkeley in the light of changing over to Millennium which accommodates many non-Roman scripts.
The following document was taken from the final report presented to the then Technical Services Council by the “Ad Hoc committee on Vernacular Parallel Script Standards for Cataloging Records” (committee members: Imad Abuelgasim, Adnan Malik, Noriko Nishizawa, Bob Talbott and Asya Usvitsky) in 2009. It was based on the 2009 PCC preliminary document that covers all non-Roman languages.
In April 2010, PCC revised its guidelines with little change. In May they published the guidelines.
For CJK subject heading
CONSER guidelines for CJK
Vernacula parallel script cataloging
1. Minimum standard set of MARC fields.
For the purposes of original records at an encoding level of I including vernacular data, we have determined the following fields to be mandatory if applicable:
245 Title statement
246 Alternate title information
247 Former title
250 Edition statement
260 Publication data, $a & $b portions
490 Series statement
740 02 Analyzed title added entry
Likewise, we have determined the following optional, though it is recommended that these access points be included in a record only when deemed essential by the cataloger:
1XX Main entry (controlled by authority record)
240 Uniform title (controlled by authority record)
260 $c portion (a formatting problem for right-to-left languages)
505 00 Articulated contents (useful, but not always necessary and thus optional)
600-630 Subject access (control language problematic)
700-730 Added entries (controlled by authority record)
760-787 Linking entries
8XX Series tracing (controlled by authority record)
2. When to include parallel fields
Parallel fields are included in cases where vernacular text is present on the piece in hand and the language code in 008 is set for that language. Thus, a book in English and Persian that is determined to be an English book based on the coding in the 008 (eng) will not require parallel vernacular fields (though it will still require the usual Romanized equivalents). However, the committee recognizes that there may be cases where the inclusion of vernacular in these types of records may be desirable, and leave the final decision to the cataloger’s expert opinion.
3. Vernacular entries for control points
The inclusion of parallel control fields is per §1 optional though discouraged. Should the cataloger however encounter a situation where he or she thinks a vernacular control point is required in the bibliographical record, we refer the cataloger to the report from the PCC Task Force on Non-Latin Cataloging Documentation for standards.
4. Copy cataloging standards
While original catalogers are fully responsible for all of the above in the creation of original records, it seems counter-productive to make copy catalogers and students abide by the same standards. Assuming that a person hired in the capacity to function in an environment containing a non-roman script is competent to do so, the committee recommends that the following be observed.
4a. Staff copy cataloging
We have determined that the beginning point for staff copy cataloging should be checking vernacular in the record against what is in the book in hand, and correcting typographical errors. All data in the record containing vernacular is checked, whether the parallel fields correspond to those prescribed in § 1 or not. As our catalog is an English language catalog, it is permissible for copy catalogers to admit records for non-Roman languages that do not contain vernacular parallel data. It is possible that a given staff member may be asked to provide vernacular if such responsibility is deemed appropriate, though this should be done only after consideration and is not mandatory.
4b. Student copy cataloging
Student copy cataloging involves checking vernacular in the record against what is in the book in hand, and correcting typographical errors as encountered.
5. Extant records deserving but not containing vernacular parallel
The committee recommends no retrospective cataloging be performed. There simply is not sufficient staffing for any sort of programmatic addition of this data for records inherited from GLADIS wanting vernacular.
6. Authority control
So little authority work is done in Technical Services now, that we have determined that the best approach would be for each individual cataloger to determine whether or not to include vernacular data in an authority record. As of this writing, there is no comprehensive national policy nor is vernacular data a requirement for the creation of new authority records. This may change, but for the time being, we are confident that the original catalogers who create authority records will be able to determine the most appropriate paths for their work flows.
An additional sticking point is the recent decision to discontinue UCB’s subscription to the National Authority File (NAF). It goes without saying that this will hamper public access and reference services particularly since, per §1, parallel fields for control points are optional but discouraged. Perhaps this an issue best suited for other fora, but it seems that the best and most cost effective solution would be to reinstate the subscription to the NAF.
7. Languages affected by these standards
Any non-Roman language is theoretically a candidate for the application of the above recommendations, though in practice, due to the realities of the work place not every non-Roman language can be cataloged. The languages included here represent the vast majority of non-Roman materials we receive. Also included here are languages that, once the capabilities exist in OCLC to accommodate the scripts, will include vernacular data.
Excluded languages are not listed. These languages are not done for a variety of reasons. For instance, Syriac cannot be done because Syriac script is not available on OCLC. Another example, the Cyrillic catalogers generally only include vernacular for languages that have time-saving macros in place to de-Romanize Romanized fields. Thus, Russian is cataloged in the vernacular because it is the major component of the Cyrillic work flow and also because it has a macro available to help with the native script. Kazak, on the other hand, is not a major component nor does it have a macro, thus it is not done.
Languages we can include vernacular for include (in no particular order):
Russian, Bulgarian, Belorussian, Church Slavic, Serbian/Macedonian, Ukrainian, Hebrew, Yiddish, Greek, Ladino/Judezmo, Judeo-Arabic, Arabic, Persian, Hindi, Tamil, Bengali, Urdu, Sanskrit, and the CJK languages.
Please keep in mind that this area is developing and that it is likely that the list will change as innovations appear.
Author: Charis Takaro
Approval Group: Cataloging and Metadata Council
Update Group: Cataloging & Metadata Council
Last updated date: 02/12/16
Replaced broken links
Fri, 02/12/2016 – 13:30 — Trina Pundurs
Replaced broken links to LC and PCC documentation.
Note- This procedure was
Thu, 06/30/2011 – 15:11 — Charis Takaro
Note- This procedure was excerpted from the report of the Technical Services Council Ad Hoc Committee on Vernacular Parallel Script Standards for Catalog Records. Catalogers should take a look and see if any other parts of the report should be included here. -Charis