NonLatin Script References in Name Authority Records University

Non-Latin Script References in Name Authority Records University of Washington Libraries Adam Schiff, Principal Cataloger June 1, 2009 1

Overview • Non-Latin script data may now be entered in NACO records • Scripts are currently limited to: Arabic, Chinese, Cyrillic, Greek, Hebrew, Japanese, Korean • Non-Latin script data are currently limited to the 4 XX “see from tracing” fields and 667, 670, 675 note fields • Pre-population of data by OCLC began on July 13, 2008 and ended in late 2008 • Addition of non-Latin data is optional 2

Pre-population • OCLC pre-populated name authority records with non. Latin script references based on non-Latin headings found in bibliographic records • Made use of data-mining techniques used for the World. Cat Identities project • Only for personal names and corporate bodies tagged 110 • The references added reflect a variety of (inconsistent) practices in formulating non-Latin name headings • Easier to remove incorrectly formulated references later on once LC/PCC policies on how they should be formulated are developed 3

4

5

Chiang, Kai-shek, 1887 -1975 Same characters, one in direct order with dates added, one in inverted form, and one in direct order without dates 6

Pre-population process • 400 and 410 references were added to NARs by an automated process • 670 s were not added • Reference evaluation code (008/29) in fixed field (Ref status in OCLC) was changed to “b”, meaning the references are not all evaluated 7

Pre-population process • Two 667 fields were added: 667 Machine-derived non-Latin script reference project. 667 Non-Latin script reference not evaluated. or 667 Non-Latin script references not evaluated. 8

NAR after pre-population 9

NAR after prepopulation 10

Some non-Latin references have dates added and some do not, reflecting various practices in formulating headings 11

Add Non-Latin Script References • You may add non-Latin script references to any name authority record: NARs tagged 100, 111, 130, 151 12

Including Non-Latin Data in NARs • Change the Ref status value from “a” to “b” • Add 667 “Non-Latin script reference not evaluated” or “Non-Latin script references not evaluated” • Justify any non-Latin references you add with 670 s • Follow LC guidelines on formulating 670 citations • Only romanized forms of names in non-Latin scripts are required in NARs; addition of the non. Latin script form is optional 13

DCM Z 1 670 “Data must be given in romanized form. Normally it is understood that the cataloger has provided the romanization; therefore, when transcribing romanization found in the source, add after it the bracketed phrase [in rom. ]. If a cataloger chooses to provide nonroman script references, the 670 $b should contain both the nonroman script transcription(s) found in the source, and the systematically romanized form(s). ” 14

670 Guidelines • Found on PSD’s Non-Latin Script Data web page 1. Romanize the title citation in 670 $a. 2. In 670 $b, use the equals sign to equate the found non. Latin script form found on the item to the systematically romanized form (or to another supplied non-Latin form, e. g. , Hangul characters supplied for a Korean entity represented only by Chinese characters in the source). 3. Romanized forms found on an item should be recorded as found on the source, followed by an indication that this was a found romanization (e. g. , add "[in rom. ]" after the found romanization). 15

670 Guidelines - Examples 670 $a Xin Gui xi ji shi. Xu bian, 2005 - : $b t. p. (中国 人民政治�商会�广西壮族自治区委�会文史和 学�委�会 = Zhongguo ren min zheng zhi xie shang hui yi Guangxi Zhuangzu Zizhiqu wei yuan hui wen shi he xue xi wei yuan hui) 670 $a Nirdaf, c 2007: $b t. p. (' = רוני דונביץ Roni Dunevits') t. p. verso (Roni Dunevich [in rom. ]) 670 $a Hangukhwa non, 1989: $b t. p. (金 鍾太 = 김 종태 = Kim Chong-t’ae) 16

Formulating Non-Latin References • Formulate non-Latin references in the same way you do when including them as parallel headings in bibliographic records (e. g. , if you add dates to the non-Latin heading in a bib record, include them in the non-Latin reference in the authority record) • No attempt will be made at this time to identify any given non-Latin reference as an authorized form for a particular language and script; there can be multiple references in the same script 17

Formulating Non-Latin References LCRI 26. 1 General Rule “In general, construct a reference in the same form in which it would be constructed if chosen as the heading. Exception: Until practices related to the form and style of nonroman script references have been established, it is not necessary to construct nonroman references in the same form as the heading. . ” 18

LCRI 26. 2 Names of Persons Referring from Variant Forms “Generally, make only one reference from each variant in each script, normally constructing it in the form in which it would be constructed if chosen as the heading (cf. specific instructions in LCRI 26. 1, including the exception for nonroman script references). In addition, if applicable, make one reference under each surname element, other than particles and prefixes, not already referred from. Normally, do not otherwise make references that are ‘variants of the variant. ’” 19

Editing Non-Latin References • Until guidelines are developed formulating non-Latin references, there is no “right” or “wrong” form for them • Obvious errors (e. g. , typos) may be corrected and references that have clearly been added to the wrong authority record can be deleted • Don’t change an existing reference to make it match the form preferred here; instead add an additional reference in that form 20

Editing Non-Latin References • Until reference evaluation guidelines are published (perhaps later in 2009), all authority records with non-Latin references should have Ref status value of “b” 21

Documentation • PSD website “Non-Latin Script Data in Name Authority Records”: http: //www. loc. gov/catdir/cpso/nonlatingeneral. html • Includes: – FAQ – Revisions to LCRIs, DCM Z 1 670 (“yellow pages”), and LC Guidelines Supplement to MARC 21 Authority Format 008/29 (“blue pages”) – LC guidelines on formulating 670 citations 22

The Near Future • Catalogers will see a variety of practices that have been used to formulate non-Latin headings • Right now we are in an interim period to gain experience working with non-Latin references and to make recommendations on best practices • In 2009, LC and the PCC will develop guidelines on formulating and evaluating non-Latin references • After those guidelines are issued, catalogers will be able to modify or delete non-Latin references to make them conform to the guidelines 23

The Long Term Future • Additional fields besides 4 XX, 667, 670, 675 may be used in the future based on user feedback; however, 1 XX and 5 XX must always be in romanized form • Additional scripts will be added • In a later phase it may be possible to designate a preferred non-Latin form for a given language and script; this would probably be done using the 7 XX linking fields rather than the 4 XX reference fields 24