The Library of Congress >> Especially
for Librarians and Archivists >> StandardsThe Library of Congress >> Especially
for Librarians and Archivists >> Standards
HOME >> MARC Development >> Discussion Paper List
DATE: May 22, 2025
REVISED:
NAME: Adding an Indicator for a Preferred Heading in 4XX Fields in the MARC 21 Authority Format
SOURCE: Library of Congress, Network Development and MARC Standards Office
SUMMARY: This paper proposes adding an indicator to See From tracing fields (400, 410, 411, 430, and 451) in the MARC21 Authority Format to indicate the preferred heading in an alternate graphic representation when multiple headings use the same alternate script.
KEYWORDS: Field 400, 2nd indicator (AD); See From Tracing-Personal Name (AD); Field 410, 2nd indicator (AD); See From Tracing-Corporate Name (AD); Field 411, 2nd indicator (AD); See From Tracing-Meeting Name (AD); Field 430, 1st indicator (AD); See From Tracing-Uniform Title (AD); Field 451, 2nd indicator (AD); See From Tracing-Geographic Name (AD); Preferred heading (AD); Preferred variant heading (AD)
STATUS/COMMENTS:
05/22/25 – Made available to the MARC community for discussion.
06/26/25 – Results of MARC Advisory Committee discussion: MAC was supportive of the paper, with suggestions for several changes. In particular, the lack of clarity introduced by the indicator value label "Preferred heading" which invites confusion with the authorized access point label for the 1XX entry. The authors will incorporate "variant" in the label as recommended by MAC and there will be further opportunity to incorporate the approval of the bcp47 coding. Additional comments referenced the PCC Task Group on Evaluation Guidelines for Non-Latin Script References in Name Authority Records final report. The task group's input will be incorporated into a subsequent paper. The paper will return as a proposal.
In the MARC Authority Format, the following indicators are defined for fields 400, 410, 411, 430 and 451:
400 – See From Tracing-Personal Name (R)
First Indicator: Type of personal name element
0 - Forename
1 - Surname
3 - Family nameSecond Indicator - Undefined
# - Undefined410 – See From Tracing-Corporate Name (R)
First Indicator -Type of corporate name entry element
0 - Inverted name
1 - Jurisdiction name
2 - Name in direct orderSecond Indicator - Undefined
# - Undefined411 – See From Tracing-Meeting Name (R)
First Indicator - Type of meeting name entry element
0 - Inverted name
1 - Jurisdiction name
2 - Name in direct orderSecond Indicator - Undefined
# - Undefined430 – See From Tracing-Uniform Title (R)
First Indicator - Undefined
# - UndefinedSecond Indicator - Nonfiling characters
0-9 - Number of nonfiling characters451 – See From Tracing-Geographic Name (R)
First Indicator - Undefined
# - UndefinedSecond Indicator - Undefined
# - Undefined
Currently, MARC authority records may contain "see from" tracings in multiple scripts in the 4XX fields. Since authority records commonly use Model B for multiscript descriptions, the ability to include language and script designations in field 880 (Alternate Graphic Representation) subfield $6 is lost.
The creation of data provenance subfields (usually $7) introduced the ability to add language, script, and transliteration information for a single MARC field. Proposal 2025-05 explores this in greater detail.
It is common in Knowledge Organization Systems to have a preferred lexical label for a specific concept or entity. This equates cleanly with the MARC Authority 1XX field, which is understood to be the preferred form of a name, title, topic, etc., represented by a MARC Authority record. But without a means to record language or script information in a MARC Authority record until recently, datasets (or files) such as the LC/NACO file or the LCSH file have been treated as monolingual vocabularies, even when records contained (variant) labels in different scripts and languages. This is logical given the preeminence of the non-repeatable 1XX field – i.e., there can only be a single preferred label in a MARC Authority record – when combined with the lack of language/script support.
Although it would be exceedingly difficult, if not impossible, to take the MARC Authority Format and use it for multilingual vocabularies, it is increasingly advantageous to not just know the language and/or script of a label in a MARC Authority record, but to also know if a specific variant label in a MARC Authority record is preferred over other variants/labels in the same language/script. For example, knowing which variant label per language/script is the preferred label makes it possible to elevate it with respect to indexing criteria and search weighting. And knowing which variant label per language/script is preferred enables displays to show the best representation given the material. For example, a book in Hebrew about Mark Twain can show the preferred Hebrew variant for Mark Twain out of the seven Hebrew labels in the related MARC Authority record.
It would also provide a way to enrich MARC records with information to use in downstream applications that are natively multilingual, bringing MARC a little closer to common non-MARC data formats. LC's MADS/RDF and W3C SKOS both permit multiple preferred labels so long as no two preferred labels share the same language code (a BCP 47 language code, to be specific) and these vocabularies are used extensively in the broader library community. VIAF has long demonstrated that an agent, or name, can have multiple preferred languages both in SKOS and in FOAF, another RDF vocabulary that supports multilingual labels. The UN's AGROVOC Linked Data representation is another well-known example of a multilingual vocabulary leveraging SKOS's preferred-label-per-langtag mechanism. LC's ID.LOC.GOV service has presented datasets, in multiple languages since 2011, notably the ISO language lists, which maintain preferred labels in English and French.
Recently, at ID.LOC.GOV, we have experimented with displaying non-Latin variants of an authorized heading more prominently. For example, see Mark Twain's authority record at https://id.loc.gov/authorities/names/n79021164.html
Lacking any other meaningful data point, the selection of the alternate script heading to display is based on the length of each variant heading. For now, the longest heading "wins" and displays. This may or may not be correct based on usage and preferences of the agent – sometimes it is, but sometimes it isn't.
What is lacking is a way to determine which "see from" heading is actually the preferred heading in a specific script. This paper offers a solution to this problem.
It should be noted, however, that, while this paper proposes a technical solution to the format for the recordation of a preferred variant label, it is an implementation matter for individual communities regarding the specifics of its use.
Advantages of the proposed changes: Display of authorized access points in discovery system could match the language of the resource in a way that does not require the creation or maintenance of 880 fields in a bibliographic record.
Disadvantages of the proposed changses: Retrospective cleanup of the existing LCNAF data file to utilize the new indicators could be challenging. One cataloger's choice of a preferred heading may not be shared by others.
In the MARC Authority Format, define a new second indicator in fields 400, 410, 411, 451 to indicate the preferred heading for a language or script when multiple versions exist in the authority record.
In fields:
400 – See From Tracing-Personal Name (R)
410 – See From Tracing-Corporate Name (R)
411 – See From Tracing-Meeting Name (R)
451 – See From Tracing-Geographic Name (R)define:
Second Indicator - Preferred heading
# - No information provided
1 – Preferred heading
In the MARC Authority Format, define a new first indicator in field 430 to indicate the preferred heading for a language or script when multiple versions exist in the authority record.
In field:
430 – See From Tracing-Uniform Title (R)
define:
First Indicator - Preferred heading
# - No information provided
1 – Preferred heading
In the above fields, define the proposed "Preferred heading" indicator value as follows:
1 - Preferred heading
Preferred variant heading in an alternate language and/or script to indicate relative importance compared to other variants in the same language and/or script to guide use in discovery, display, etc. Only a single 4XX field should be marked as "preferred" per language and script combination. When identifying a preferred variant, language and script must be recorded as a BCP 47 code in a $7 pertaining to the entire field.
Multiple Hebrew and Yiddish variant headings.
010 ## $a n 79021164 $z sh 89001267 $z no 98029431
100 1# $a Twain, Mark, $d 1835-1910
400 1# $a מרק ,טבןַ, $d 1835-1910 $7 (bcp47) he-hebr
400 11 $a מארק ,טוויין, $d 1835-1910 $7 (bcp47) yi-hebr
400 1# $a מרק ,טוויין, $d 1835-1910 $7 (bcp47) yi-hebr
400 1# $a מארק ,טווין, $d 1835-1910 $7 (bcp47) yi-hebr
400 1# $a מרק ,טווין, $d 1835-1910 $7 (bcp47) yi-hebr
400 1# $a מארק ,טווען, $d 1835-1910 $7 (bcp47) yi-hebr
400 11 $a מרק ,טוין, $d 1835-1910 $7 (bcp47) he-hebr
Variant headings in Traditional and Simplified Chinese. In this example, the variant heading with the most occurrences in the Library of Congress OPAC is selected as the preferred variant heading.
010 ## $a n 81092441
100 1# $a Shen, Congwen, $d 1902-1988
400 11 $a 沈从文, $d 1902-1988 $7 (bcp47) zh-hans
400 1# $a 沈岳煥, $d 1902-1988 $7 (bcp47) zh-hani
400 1# $a 沈崇文, $d 1902-1988 $7 (bcp47) zh-hani
400 1# $a 沈從文, $d 1902-1988 $7 (bcp47) zh-hant
Variant headings in multiple scripts, with a designation of preferred script for each:
010 ## $a n 79046204 $z n 89149387
130 #0 $a Qurʼan
430 #0 $w nnea $a Koran
430 10 $a קוראן $7 (bcp47) zxx-hebr
430 10 $a قرآن $7 (bcp47) zxx-arab
430 #0 $a Каран $7 (bcp47) zxx-cyrl
430 10 $a Коран $7 (bcp47) zxx-cyrl
In all of these examples, usage of the preferred heading indicator would enable a cataloger to indicate which heading should appear in an 880 field.
Use of a preferred heading will enable the BIBFRAME-to-MARC conversion to more accurately construct 880 fields for access points.
Since BIBFRAME is an RDF standard, language, script, and transliteration scheme are tagged at the equivalent of the field or subfield, rather than encompassing the entire record. Enabling more options in MARC to support a more granular level of tagging in MARC could help make conversion from MARC to BIBFRAME and from BIBFRAME to MARC smoother, with less data loss.
6.1. Do you agree with the proposed addition of this indicator?
6.2. Are there any unintended consequences to implementing a preferred heading indicator in these fields?
6.3. Are there any other issues that should be addressed?
HOME >> MARC Development >> Discussion Paper List
| The Library of Congress >> Especially
for Librarians and Archivists >> Standards (11/26/2025) |
Legal | External Link Disclaimer | Contact Us |