FLHIG
 

Family and Local History
Indexing Group

Selected articles from the Newsletter

This article first appeared in SIGGNL 17  pages 20 to 22 (February / March 1998)

Indexing of Periodicals - Introduction
by Tony Rydings

To the uninitiated it may seem that there is no obvious difference between indexing a book, which probably contains a series of chapters on various topics, and a volume or volumes from a periodical publication, which contains a number of articles of varying length and again on diverse subjects. Superficially this is true, but the way in which the index is compiled should produce a result that reflects the different nature of the two kinds of writing.

One might look further, and consider that not all periodicals are similar, and that the indexes to them should also vary in style and content. For the purpose of the current thesis we shall limit our study to two main categories of periodicals. The first, and probably the kind with which the majority of our readers may be concerned, are the publications of family history societies; and the second group covers genealogical materials from a more scholarly point of view. Many family historians will regularly read both categories, but will probably agree that they do not expect the indexes to cover the first kind in such detail as those which index the second kind. It is a principle of good indexing that the depth of indexing should vary with the class of reader, and this should be borne in mind when the two kinds of family history periodicals are discussed.

Planning

Among the basic decisions to be made when planning an index is the approximate length of the index relative to the text to which it relates. This is often determined by the publisher or editor, and the indexer should beware of a tendency to underestimate the space required. for books it is estimated that the percentage of total pages required for the index varies between 2 and 15, depending on the type of material, and the number of entries per page will vary between 3 and 10. The factors which will influence the index length include the size of type, the layout of entries, whether the entries are arranged in one or more columns, and whether sub-headings are run on or not.

Type size

It is common for the text of a book to be set in 12-point type, and for the index to be in 10-point. This is an acceptable variation, but occasionally where space is short the index may be squashed up to the extent that it is hard to read, which should be avoided.

Layout of entries

Indexes are frequently arranged in two columns to the page, as this results in space saving without cramming, so long as the average length of entry is less than half a page width.

Style of sub-headings

The choice here is between indented, thus:

The latter evidently takes less space, but the former may be preferable as regards clarity.

Index length

There are two consequences of the length of an index: the time taken to compile it, and the space needed for printing it. I have myself avoided the worst handicaps of either, having always worked in my own time without limits imposed by a publisher, so cannot expound on the usual problems and how they should be overcome. The question of time should be agreed before an assignment is accepted, and space can probably be best estimated by actually indexing a sample proportion of the whole, and calculating pro rata.

Problems of ensuring uniformity

It is desirable that indexing a series of periodical volumes should all be carried out by one person, or at least edited by an individual, since an important object of the exercise should be uniformity. The difficulty here is that the articles are contributed by different persons, each probably with varying views and opinions, and expressing them in different words. The aim of the indexer should be to combine entries for related materials, even though they originate from various contributors. This is not too difficult for entries for persons or places, but when it comes to indexing composite terms it is another matter. Often even the same author may use different phrases to mean the same thing, and even more frequently different writers will do this. For example, one will write of agricultural labourers and another of farm workers (or vice versa), while the indexer should make a single entry for both, with the required cross reference. We shall frequently refer to this need to standardize in indexing.

When indexing a single issue or volume of a periodical, it is not too difficult to relate entries which refer to the same or allied concepts; but as the amount of material to be covered grows, so the task becomes more difficult. This is a matter of practice and experience. To have a limited subject range is obviously an advantage, but it is surprising how wide a scope the contributors to family history journals will cover, so that the indexer is occasionally obliged to use a general heading, whereas the preference may be for more specific headings. Thus an article on bricklayers should be entered under that heading rather than under building industry. In the next instalment we shall look more closely into different kinds of entry and how to choose them.

Article continues in SIGGNL 18 p.9

Ç TOP