FAQ on the 670 (Sources found) in name authority records (NARs) for NACO

1. What is the required punctuation and style in the 670 field? There is no required punctuation and style in the 670 field. There is some prescribed content (per the MARC 21 Authority Format) and some suggested punctuation (see no. 2-4 of this FAQ). Page 2 of the 670 section of the LC Descriptive Cataloging Manual (DCM) Z1 supplement to the MARC 21 Authority Format states" "conventions in regard to punctuation and style, unless a specific direction is given ... are not prescriptive ... Punctuation and style need not be consistent from record to record as long as the information is clear and accurate."

   The NACO Participants' Manual (NMP), in the section on Introduction and Basic Format, states: "As of the 1996 revision of the DCM Z1 'style and punctuation' are not prescribed." For historical purposes and because catalogers will continue to find NARs in the authority file which contain "old style" citations, the NPM and the DCM show varied examples.

   Note: DCM Z1 does ask catalogers to give the abbreviated forms of names of months when giving dates in the 670 (e.g., when recording the date a database was searched or authors' birth dates). This helps to facilitate international participation in NACO.

2. What are the prescribed elements in the 670 field? The MARC 21 Authority Format defines the 670 (sources found) field as a repeatable variable field which is comprised of non-repeatable subfields $a and $b. Subfield $a (source citation) is always required; however, subfield $b (information found) is necessary only when information is being provided in support of the formulation and/or identification of the 1XX, 4XX and sometimes the 5XX fields. DCM Z1 prescribes four elements:

   In subfield $a of the 670:

   1. The title proper of the work being cataloged.
   2. The date of publication or edition of the work.

   In subfield $b:

   1. The specific location(s) of the information found when the work cited is other than a reference source.
   2. The information found (enclosed in parentheses)

   Note: DCM Z1, reminds catalogers that: "the NAR does not serve as a biographical sketch of a person ..." and to "use judgement to determine how much data to record ..." The ideal 670 is cogent and concise yet complete.

3. Is a subfield $b always required in a 670? As noted in question no. 2 of this FAQ, the MARC 21 Authority Format requires a subfield $a; however, subfield $b is necessary only when information is being provided in support of the formulation and/or identification of the 1XX, 4XX and sometimes the 5XX fields.

   For example if the name as established in the 1XX is contained in the title of the item being cataloged (subfield $a), it is not necessary to repeat the name in a subfield $b. When creating or updating uniform title NARs it is seldom necessary to add a subfield $b to the NAR unless recording research. Nonetheless, in both these cases the inclusion/repetition of information in a subfield $b is not prohibited.

4. Is the inclusion of the main entry in the 670 totally forbidden? No. According to Z1, if the title of the item being cataloged is generic, the main entry should be included.
5. What would a "typical" 670 for an NAR look like? A typical 670 would include the following prescribed content and suggested punctuation:

   A monograph:

   670 $a La pasión de Octubre, 1996: $b t.p. (P.J. González Cuesta) back flap (Pablo González Cuesta, b. Seville, 1969)

   A database reference source:

   670 $a OCLC database, Jan. 23 2001 $b (hdg: González Cuesta, Pablo Juan; usage: P.J. González Cuesta)

6. OCLC as well as many local systems have macros to "machine assist" the creation of NARs and as a result there may be more information in the 670 citations than ever before, how much clean-up is required in this field?

   The 670 section of the DCM Z1 states: "In authority records created using an automated authority generation program, the 670 information may include the main entry name ... it is recommended that catalogers accept the additional information as generated."
   Catalogers should use judgement in deciding what other information can remain or what should be deleted.

7. When is it necessary to provide more than one 670 (Sources found field) in an NAR for a personal name heading being newly established which does not conflict with another name in the name authority file (NAF)?

   In general, when newly establishing a personal name that does not conflict with another heading in the database within which one is cataloging, it is not necessary to cite another source beyond the item-in-hand except in the following situations:

   1. When the rules for establishing personal names require consultation with a reference source (e.g., AACR2 22.1B, 22.3B2, 22.3B3)
   2. When justifying an addition to the name heading (fuller form of name, dates, title, etc.) and that information was found in a source other than the item-in-hand (i.e., during the normal course of searching in the database in which the work is being performed).
   3. When justifying a cross-reference and that information was found in a source other than the item-in-hand (i.e., during the normal course of searching in the database in which the work is being performed).
   4. When recording a variant which would not require a cross-reference (e.g., a variant in the 2nd element to the right of the comma, cf. LCRI 26.2) and that information is found in a source other than the item-in-hand (i.e., during the normal course of searching in the database in which the work is being performed).
8. When do NACO procedures require cataloger to look in other sources (beyond the item-in-hand and the database in which one is cataloging) for variants, fuller forms of the heading or dates, etc.?

   Generally, only when the heading conflicts with another in the NAF and the item-in-hand does not provide enough information to break the conflict or as noted in response to question 7 of this FAQ, when the rules call for consultation with a reference source.

9. Is it true that NACO catalogers should not use the citation "LC in OCLC" in a 670?

   PCC NACO trainers have been encouraged to de-emphasize the use of this citation. Newer training materials eliminate such examples except to alert catalogers that they may find this citation in older NARs. This is part of the simplification of the 670 field which came about as a result of the studies carried out in 1993 by the CCC's Task Group on Authorities and the subsequent CCC/CPSO 670 Task Group.

10. Should we use the designation "PCC in OCLC"" in a 670 to cite a heading found on a PCC (042=pcc) record?

    No, there is no convention for citing PCC records in the 670 and at this point it is not cost-effective to add another layer of complexity to citations in the 670 field.

11. Doesn't it "help" or give more "authority" to the heading being established if a 670 is cited showing that the heading was formulated the same as the new 1XX and has been used on bibliographic records (especially if it's an LC record)?

    No, although some catalogers seem to think so. It is AACR2r, the LCRIs, and usage which provide the authority for establishing a heading. The use of "LC in ..." is a holdover from the early days of NACO when forms were processed manually. As noted in the response to question 9 of this FAQ, since 1993 every attempt has been made to reduce the amount of time spent constructing 670 fields. To cite the occurrence of a heading that does not provide any additional information in an additional 670 (regardless of its provenance) adds to the time it takes to create an authority record and is contrary to the PCC principle of "the timely creation and maintenance of authoritative, cost-effective bibliographic and authority records."

12. If I search the LC database should I provide an "LC database" 670 citation?

    Searching the LC database is not a NACO requirement; however, NACO reviewers are aware that often catalogers will search the LC database in order to report BFM. With the implementation of LC's Online Catalog users will encounter many different types of bibliographic records that they may not have encountered when searching under the previous system. NACO catalogers are urged to read the announcement prepared by CPSO and the FAQ on reporting BFM for a more in depth explanation of records found in the current LC database.

    In response to this question, NACO cataloger may provide a 670 with a "LC database" citation but should do so only if the information provides additional information in support of the formulation of the 1XX, etc. Remember when citing headings labeled [from old catalog] to include that legend in the 670.

    Example:
    670 $a LC database, date searched $b (hdg: Poschmann, Bernhard, $d 1878-1955 [from old catalog]; usage not given)

13. Now that subfield $u has been implemented in the 670 field, it it OK to include URls at will in NARs?

    Within reason. Although subfield $u was authorized for use in NARs on February 1, 2006, NACO catalogers are expected to judiciously apply its use. Optional use of the 670$u should be for those cases when the source contains significant information related to the established heading that cannot be cited succinctly in the 670. Remember—citing a URI in 670$u does not take the place of the requirement to cite relevant data in subfields $a and $b of the 670 (i.e., enough information to support the heading/references, which will be available to future users even if the Internet site itself disappears).

14. May I also use the URI in subfield $a or $b of the 670 in NARs?

    Generally, no. Corporate names and/or title strings may have the general appearance of URIs (usually without the Internet protocol designation, e.g., http://), and may be cited as names and titles in 670$a as needed (cf. http://www.loc.gov/catdir/pcc/naco/corpfaq.html#9 for more information). In order to be “actionable,” URIs found in $u should include the protocol, e.g., “$u http://www.stephenking.com