DISCUSSION PAPER NO. 85

DATE: May 5, 1995
REVISED:

NAME: Changes to Personal Name First Indicator Values

SOURCE: NLM, LC, British Library

SUMMARY: This paper proposes the following changes in the values of the first indicator (Type of personal name entry element), of the X00 (Personal Name) fields: define new code # (blank) to indicate that the type is unknown and make value 2 (Multiple surname) obsolete.

RELATED:

KEYWORDS: X00 (Personal Name); Type of personal name entry element; Multiple surname

STATUS/COMMENTS:

5/5/95 - Forwarded to USMARC Advisory Group for discussion at the June 1995 MARBI meeting.

6/26/95 - Results of USMARC Advisory Group discussion - Discussion indicated that there is interest in a proposal for the dropping the multiple surname value for the first indicator. Only one institution represented at the meeting indicated they still use the indicator for sorting. There was more concern expressed about an indicator for no information provided. NLM stated that they could use the new generic author field instead for the conversion that required the new first indicator value and it would be very appropriate. That might suffice for their needs.


DISCUSSION PAPER NO. 85: Changes to Personal 
Name First Indicator Values

1.   INTRODUCTION

The current values for the first indicator (Type of entry element
for personal names) in fields X00 (Personal Names) have been
defined since 1968:
          0    Forename
          1    Single surname
          2    Multiple surname
          3    Family name

The indicator is primarily (but not exclusively) used for proper
filing order mandated by certain filing rules.  These rules are
used for large and small files but especially for the former.  Two
situations have arisen that call for changes to the indicator
values.  This indicator is also used on Authority and Community
Information records.

"No Information Provided" Value

The National Library of Medicine (NLM) has never distinguished the
type of entry element for personal names in their MEDLINE data
base, therefore when they distribute the MEDLINE file in the MARC
format the first indicator cannot be accurately set in records.  By
not having an "I do not know" value, the programs must guess,
misleading the receiver of the records in some percent of the field
occurrances.  NLM is uncomfortable with supplying an indicator when
the data has not really been examined.  The MEDLINE records are not
constructed using AACR2, and other non-AACR2 files may also share
this lack of differentiation of entry elements in personal names. 


Definitions of Multiple Surname Value

In working with the British Library for sharing authority records
under the cooperative NACO program, a problem occurs in the
definition of the multiple surname values in the USMARC and the
UKMARC formats.  Both formats have the same indicator with the same
values, but the definitions for single surname and multiple surname
encompass slightly different groups of headings.  The UKMARC
approach is simply to consider the number of discrete elements
preceding the inverted element, whereas USMARC tries to take into
account the type of element, with some exceptions being observed. 


USMARC:
   First Indicator Value 1 - Single Surname
     Value 1 indicates that the heading is a name with a single
     surname formatted in inverted order (single surname, forename)
     or a name without forename(s) which is known to be a single
     surname.
          100  1#$aFitzgerald, David.
          100  1#$aChiang, Kai-shek,$d1887-1975.
          100  1#$aWatson,$cRev.

     A surname beginning with a separately written prefix which is
     an article (e.g., La), a preposition (e.g., De), or a
     combination of the two (e.g., Del) is classified as a single
     surname if a single word follows the prefix (e.g., Van
     Rosenburgh). 
          100  1#$aLe Conte, John Eatton,$d1784-1860.
          600  10$aLa-Faye, Julio.
          100  1#$aO'Brien, Edna,$d1936-
          100  1#$aDe la Mare, Walter,$d1873-1956.
          100  1#$aO Neill, Sean.

   First Indicator Value 2 - Multiple Surname
     Value 2 indicates that the heading is a name with a multiple
     surname formatted in inverted order (multiple surname,
     forename) or a name without forename(s) which is known to be
     a multiple surname.  
          100  2#$aBen Omar, Saidali Bacar,$d1931-
          100  2#$aRousseau-Darnell, Lyse.
          700  2#$aVérez Peraza, Elena,$d1919-
          100  2#$aSaint-Germain, Herve de.
          100  2#$aP-Orridge, Genesis,$d1950-
          700  2#$aSalamín C., Marcel A.

     A surname beginning with a separately written prefix which is
     an article (e.g., La), a preposition (e.g., De), or a
     combination of the two (e.g., Del) is classified as a multiple
     surname if more than one word follows the prefix (e.g., Van
     Alderwerelt van Rosenburgh).
          100  2#$aDu Bois de Cendrecourt, Louis

     Names which contain a prefix that is not an article, a
     preposition, or a combination of the two are treated as
     multiple surnames.
          100  2#$aMc Intosh, Paula.

          100  2#$aEl Saffar, Ruth S.,$d1941-
          100  2#$aEl-Abiad, Ahmed H.,$d1926-
               [Anglicized Arabic surnames are treated as multiple
               surnames.]

UKMARC:
   First Indicator Value 1
     Name of person entered under a single title of nobility, or
under the last element of the name.
          100  10$aFitzgerald, David.
          100  10$aChiang, Kai-shek,$d1887-1975.
          100  1#$aWatson,$cRev.
          100  1#$aO'Brien, Edna,$d1936-

   First Indicator Value 2
     Name of person entered under a compound surname, a surname
     with a separate prefix, a compound title of nobility, or under
     any other element of the name other than the first or last
     element.
          100  2#$aBen Omar, Saidali Bacar,$d1931-
          100  2#$aRousseau-Darnell, Lyse.
          700  2#$aVérez Peraza, Elena,$d1919-
          100  2#$aSaint-Germain, Herve de.
          100  2#$aP-Orridge, Genesis,$d1950-
          700  2#$aSalamín C., Marcel A.
          100  2#$aLe Conte, John Eatton,$d1784-1860.
          600  20$aLa-Faye, Julio.
          100  2#$aMc Intosh, Paula.
          100  2#$aEl Saffar, Ruth S.,$d1941-
          100  2#$aEl-Abiad, Ahmed H.,$d1926-
          100  2#$aDu Bois de Cendrecourt, Louis
          100  2#$aDe la Mare, Walter,$d1873-1956.
          100  2#$aO Neill, Sean.

2.   DISCUSSION

NLM is requesting the addition of a new first indicator value for
the Bibliographic format 600 and 700 fields, value # (blank)
meaning "No information provided".  They suggest that is only be
used for non-AACR2 records.  In order to maintain the consistency
of the X00 fields, the new value would also be added to the 100 and
800 fields in the Bibliographic Format and to Authority and
Community Information personal name fields. 

On investigation, it appears that many systems may not carry out
any special sorting or make other use of the distinction between
the single and multiple surnames.  It may therefore be possible to
reconcile the difference and simplify the format at the same time
by eliminating the differentiation of single and multiple surnames,
keeping only values for forename, surname, and family name (value
3).

The new values for the first indicator would be:
          #    No information provided
          0    Forename
          1    Surname
          3    Family name

3.   QUESTIONS

1)  What are the uses made of the first indicator value?  Is the
type of entry element distinction important or would a "No
information provided" value be adequate for most processing (where
the attribution of surname would probably be assumed?)?  Is it
preferable that the exporter set the indicator if it cannot be
accurately set or would the importer rather guess at the value?  Is
it important, to know that is is absolutely accurate?

2)  Is the distinction between multiple and single surnames
important to any processes?  If so which ones?  Are there
alternatives for processing if the distinction were absent?



Go to:


Library of Congress
Library of Congress Help Desk (09/03/98)