Some authors indicate their line of reasoning by the compilation of headings. This is a great help to the indexer; but if the author does not present such headings, the indexer has to make them himself, and he therefore needs the abilities of the précis-writer.
There are indexes of Books, of Transactions, Periodicals, etc., and indexes of Catalogues. Each of these classes demands a different method. A book must be thoroughly indexed; but the index of Journals and Transactions may be confined to the titles of the papers and articles. It is, however, better to index the contents of the essays as well as their titles.
Before the indexer commences his work he must consider whether his index is to be full or short. Sometimes it is not necessary to adopt the full index—frequently it is too expensive a luxury for publisher or author; but the short index can be done well if necessary.
Whatever plan is followed, the indexer must use his judgment. This ought to be the marked characteristic of the good indexer. The bad indexer is entirely without this great gift.
While trying to be complete, the indexer must reject the trivial; and this is not always easy. He must not follow in the steps of the lady who confessed that she only indexed those points which specially interested her. We have fair warning of incompleteness in The Register of Corpus Christi Guild, York, published by the Surtees Society in 1872, where we read, on page 321:
"This Index contains the names of all persons mentioned in the appendix and foot-notes, but a selection only is given of those who were admitted into the Guild or enrolled in the Obituary."
The plan here adopted is not to be commended, for it is clear that so important a name-list as this is should be thoroughly indexed. However learned and judicious an editor may be, we do not choose to submit to his judgment in the offhand decision of what is and what is not important.
There is a considerable difference in the choice of headings for a general or special index—say, for instance, in indexing electrical subjects the headings would differ greatly in the indexes of the Institution of Civil Engineers or of the Institution of Electrical Engineers. In the former, dynamos, transformers, secondary or storage batteries, alternate and continuous currents would probably be grouped under the general heading of Electricity, while in the latter we shall find Dynamos under D, Transformers under T, Batteries under B, Alternate under A, and Continuous under C.
The indexes to catalogues of libraries, etc., are among the most difficult of indexes to compile. It was not usual to attach an index of subjects to a catalogue of authors until late years, and that to the Catalogue of the Athenæum Club Library (1851) is an early specimen. The New York State Library Catalogue (1856) has an index, as have those of the Royal Medical and Chirurgical Society (1860) and the London Library (1865 and 1875). That appended to the Catalogue of the Manchester Free Library (1864) is more a short list of titles than an index.
There are special difficulties attendant on the indexing of catalogues. Books are written in many languages, and there is considerable trouble in bringing together the books on a given subject produced in many countries. The titles of books are not drawn up on the same system or with any wish to help the indexer. Titles are seldom straightforward, for they are largely concocted to attract the readers, without any honest wish to express correctly the nature of the contents of the book. They are usually either too short or too enigmatical. The titles of pamphlets, again, are often too long; and it may be taken as an axiom that the longer the title the less important the book.