FLHIG
 

Family and Local History
Indexing Group

Selected articles from the Newsletter

This article first appeared in SIGGNL 17  pages 17 to 20 (February/March 1998)

Name Variants and Sorting
by Derek Milledge

I am told that the first rule of indexing is to index what you find, not what you think it ought to be. It is a good rule of thumb, and should be broken only after careful thought. Having listed what you have found, the next question is how to sort it.

The Standards do recommend ignoring an apostrophe when sorting, so that DARCY and D'ARCY would be the same, but computers do not understand this. Nor do they recognise that McKAY and MacKAY should be adjacent. As the Editor pointed out in SIGGnl 15, the solution is to create an additional field for a root name and use that as the first field for sorting. To keep the surnames adjacent but separate, the real surname can be the second field.

Root names

The IGI takes this approach one step further and attempts to collate different spellings together under one root name, which is a mixed blessing. It is a help, when the names are the same, to have them indexed together, with MILLEDGE, John next to MILLIDGE, John. It is not such a boon when the French name MARECHAL is lost in an endless list of MARSHALLs. It is better if genuinely different names are kept separate, especially if one name is numerous.

Use of cross references

Whatever decision is made about grouping surnames, it is always helpful to give a cross-reference "see also MILLEDGE". For forenames, the cross reference is not appropriate, and the user has to know that "William" and "Wm" will be separate lists in the IGI, and that many names have alternative forms. In SIGGNL 15 the Editor suggested sorting "Biddy" and "Bridget" together, but the user seeking "Biddy" who was not "Bridget" might then miss her.

In more complex records one person's forename may be repeated, but in a different form: it would seem perverse to index "William" and "Wm" for the same page of the same record, for one is clearly an abbreviation of the other. If the two forms had been "Harry" and "Henry", the equivalence is less certain, and the user might need either. It would seem best to combine only those forms which must be the same, defined by an equivalence list available to the user (though users seldom read the introduction).

Compound names

In some cases I see no way of satisfying conflicting requirements. In England the name DE LA MARE often became DELAMARE (and variants), and it might seem appropriate to list them together. In France the "de" was merely a title: the surname should be indexed as LA MARE, grouped with other members of the LA MARE family. [The means of dealing with prefixes to surnames varies from language to language, and from country to country; for instance, Laurens van der Post would be treated differently as a Dutchman than as an Afrikaner, a Belgian or an Englishman(2) - Ed.] I thought of indexing one person as both "DE LA MARE, Walter" and "LA MARE, Walter de", but users might then have to look him up twice. [One solution is to index a prefixed surname in full, then removing one prefix at a time, in much the same way; the same method can be applied to double- or multiple-barrelled names. Rather cumbersome, I feel, but the user is bound to find the name under one or other. It is tempting, in order to save space, to adopt inversion of all prefixed surnames, with cross-references from the prefixes, for example "Van: for names with this prefix, see the main element." - cross-references can also be used for compound names; however, the main consideration must be where the user would normally look - Ed.] I will clearly have to follow the precedents set in other Huguenot publications, but even they are not consistent in treating the "D" before an apostrophe.

Conclusion

I have come to the conclusion that name variants are a minefield through which there is no safe path, but we can still strive to assist the greater number of users to find what they seek.

References

1. The relevant standards are BS3700: 1988, 6.2.2 and ISO999: 1996; I have not seen the later as I write - Ed.

2. Relevant references in indexing texts are: R F Hunnisett. Indexing for Editors. British Records Association, 1972 (1997) at p50-51; R L Collison. Indexes and Indexing, 4th ed. London: Ernest Benn; New York: John de Graff, 1972 at p49-50 and p199; Hans H Wellisch. Indexing from A to Z, 2nd ed. New York; Dublin: H W Wilson, 1995 ay p362-5; Nancy Mulvany. Indexing Books. Chicago; London: University of Chicago Press, 1994 at p165-8; Michael Gorman. The Concise AACR2 1988 revision. Chicago: American Library Association; Ottawa: Canadian Libary Association; London: Library Assocaition: 1988 at 34D and 63A2 -Ed.

Ç TOP