Call me by multiple surnames: X00 first indicator 2
X00 fields, used for individual personal names in both bibliographic and authority records, used to have separate first indicators defined for single and multiple forenames. Today, the indicators are: 0) Forename, 1) Surname, and 3) Family name. But until 1996, they were: 0) Forename, 1) Single surname, 2) Multiple surname, and 3) Family name.
(Did you know that you can see what fields, subfields, and indicators used to be valid on the bottom of most pages of the MARC specification? Because you totally can!)
The rationale for encoding single surnames and multiple surnames differently was that certain filing rules handle names with multiple surnames differently. The different indicators went all the back to the MARC II standard from 1968 (at least according to a 1996 report), making me feel like the importance of filing rules is a holdover from a time when MARC was as much about creating catalog cards and printed library catalogs as anything else.
However, the separate indicators became a problem when the British Library and the Library of Congress started exchanging name authority records in 1994, as part of NACO. At the time, the British Library was using UKMARC, and LC was using USMARC. Both had first indicator values defined for single and multiple surnames, but the two communities used them in different ways. USMARC had a fairly confusing set of rules about when to consider something a multiple surname, depending on the type of element included in a multi-part surname and how many parts made up the whole name, while UKMARC… only cared about whether the surname had more than one part. Obviously, that led to some discrepancies. For example, UKMARC treated the name “Walter de la Mar” as having multiple surnames (first indicator 2), while USMARC treated it as a single surname (first indicator 1), because the elements that make the surname more than one part are a preposition and an article.
When the issue became apparent, the USMARC Advisory Group started looking into it, and discovered that very few systems actually used the first indicator for making a sorting distinction between the two types of surnames. So, rather than either side changing their encoding practices for multiple surnames, the distinction was eliminated altogether. USMARC got rid of it in 1996 - first indicator value 2 was rendered obsolete, and value 1 was redefined as just “Surname.” Although I can’t confirm for sure, UKMARC seems to have made the same change (see, for example, the last published version of UKMARC’s definition of field 100). So even though the British Library (and by extension, much of the UK library community) didn’t move from UKMARC to MARC 21 until 2004, differences between the two formats were being reconciled well before that - in this case, due to shared authority projects.
In a twist, the Library of Congress didn’t actually implement the change until 2000. When they did, the decision was to encode new bib and authority records correctly, but only to change old records when those records were touched for some other reason. Presumably by now, all the records with first indicator 2 have been changed - I assume that LC and OCLC have done some mass updating of that by now. But it doesn’t seem impossible that the old surname encoding is still hanging around in various local systems.
