Initial articles
On AutoCat for week 4 of January 2011, a post titled “Initial articles Do they matter anymore” generated some interest. It interested me too.
First – what are articles? I like Purdue’s definition, simple and easy to understand but of course, Wikipedia’s English articles is also great. Check out Wikipedia for more on articles in grammar for all languages.
In English, the articles are “the, a, an”. In MARC-21 Bibliographic cataloging, we use the second indicator of the 245 field to tell the system to “skip” the initial article when doing an alphabetic filing of the title. Well, that is unless the initial article is an integral part of the title or not used as an article such as “A is for Apple”. [oh, and BTW, your ILS might or might not take this into consideration, I've seen both ways in various ILS]
The posts on AutoCat start with a gentleman stating he was attempting to coordinate amongst various lists of titles he has generated. He can generate a list from his MARC Bibliographic records ‘skipping’ the initial article on the alphabetic sort but he is not having that same success with other data sources. Thus, whilst his list is alphabetical the same title will not align together on the list (his example “‘Barber of Seville” versus “The Barber of Seville”).
I wonder about this. I mean, I understand MARC Bibliographic but I wonder why we have to specific in today’s world. Surely this could be a programming issue? That is, surely we could have the programming manage the issue of the initial article. Write it so it is ignored UNLESS otherwise specified, since it would be more common to ignore rather than include.
Seems to me that journal article databases already do this – they sort and automatically ignore initial articles. Don’t they? Why can this not be ported over to other systems and programs?
Or, does it matter to the populace? Should we just stop worrying about it and allow either a mixed list or a long list of titles starting with “The…”?
What think you?
One way we’ve handled this issue with XML metadata is by creating a separate sort title element which repeats the content of the title element without the initial article.
Two different programmers requested I adjust the metadata this way, so that they could create a title index for our digital library system. I think it’s a nice solution at least using XML which is much more flexible than MARC (unless we’re talking about MARCXML).
It is a good solution Christine but it just seems like it should not be so necessary. I do wonder how article databases manage this – or title lists from databases. I mean, if you have access to EBSCO or Gale or whatever, grab one of the title lists. In EBSCO it’s in “publications”. Seems like they are alphabetic but ignoring the initial article. How are they doing this? Is it a dual-title like you are using? Is it part of the programming? Why can we not all do it?