From herbals to Hotbot: the development of journal indexing

The development of journal indexing since the early alphabetical tables to scientific works, culminating in current sophisticated cumulations, is traced. Standards for preparing indexes which have gradually evolved to cope with the increase in volume and complexity of the literature are described. Learned Publishing (2001)14, 123–130 Hazel Bell From herbals to Hotbot: the development of journal indexing 123 L E A R N E D P U B L I S H I N G V O L . 1 4 N O . 2 A P R I L 2 0 0 1 from the beginning [1665] to July 1677’. A modern critic reports of it:2 This index is in absolute alphabetical order. The entries are set out in ‘run-on’ paragraphs with much variation in the number of references given according to the specificity of the headings. ‘Aches’ has only one reference whereas the ‘Air’ references fill a column and a half. All multiple entries are arranged in chronological order by date of publication of the issues referred to; or, if several references relate to one article, in sequential order by page number. Though there are occasional internal connections and local groupings under what amount to subheadings within these blocks, no systematic subordinate rearrangement of the material, alphabetical or otherwise, appears to have been attempted at this time. Some degree of cross-referencing is evident, both of the see also type from one used heading to another, and of the see type from an unused heading to the preferred term. What strikes one, upon examining the entries grouped under ‘Air’, is the attempt by the writer to connect his material. The index has been written by a person familiar and at ease with the subject matter. The items being referred to in order of their publication, each sequence of entries under a particular heading is moulded into a brief, but detailed, history of the progress in that area of the Society’s concerns. Guidance is offered to the reader on what should be consulted; he is advised of mistakes in the original; and occasionally reference is made to material not in the running sequence. The outcome is more essay-like than ‘indexical’ in the sense that would be understood today, having a sequential interdependence which suggests that this index has been written to be read. . . . The Royal Society’s index was compiled at a time before the atomization of reference sequences had become mentally sustainable in other than an occasional or random fashion. The Royal Society was engaged in promoting those processes of observation, experimentation, analysis and inference that would produce such a sustained capacity, but empiricism had not yet overturned the older mental processes exemplified in Isaac Newton’s theological writings. The index itself must have been an instrument in that revolution as well as one of its symptoms. Over the next two centuries, we are told:3 The ‘back-of-the-book’ index, very much as we know it today, was well established by the end of the seventeenth century when the first scholarly journals were published. Cumulative and collective indexes to periodicals came a bit later – around the late eighteenth century. By the nineteenth century, the scholarship and information explosion had so grown that retrospective indexes were being initiated. William Poole, working in the library of Yale University, noticed that many of the library back-issue periodicals were never used, and concluded that indexing them would make them more useful. He developed and published such an index, 154 pages, in 1848. It was a great success, and was succeeded by the Index to periodical literature, published every five years; then by a monthly Co-operative Index to Current Numbers of Leading Periodicals , and by the Annual Literary Index, the American Library Index and the American Library Annual, all produced in cooperation with volunteer members of the American Library Association. In 1901 the indexing of general periodicals became professional with the beginning of Wilson’s Readers’ Guide to periodical literature. Other early and prominent indexing services had their origins in this period; Index Medicus was first published in 1879 by a surgeon in the Civil War. Card indexer at work From 1946 to 1974, the novelist Barbara Pym helped to edit the publications of the International African Institute in London, including four annual issues of the journal Africa, for which she compiled the indexes. Her assistant gives us a picture of the journal indexer of the period at work:4 She would sit crouched over a battered, wooden four-drawer card index of antique The index has been written by a person familiar and at ease with the subject matter 124 Hazel Bell L E A R N E D P U B L I S H I N G V O L . 1 4 N O . 2 A P R I L 2 0 0 1 design, like Miss Clothier [a character in a Pym novel], ‘moving the cards here and there with her fingers, as if she were coaxing music from some delicate instrument’. Occasionally I would catch a murmur, like the recital of some litany: ‘Abortion, administration, age-sets, agriculture, amulets, ancestors, animal husbandry . . .’. She loved the mystique of certain aspects of her job; best of all, she enjoyed the art of indexing. The great ongoing indexes were the cumulative, annual indexes for Africa. The real challenge was the Index of Tribes and Languages. Variations of tribal names gave delightful opportunities for cross-referencing, and the special characters had to find their formal place in the scheme of things – the longtailed S (= sh) or the Bushman ‘click’ (!Kung would come after Z). I think what she enjoyed most about indexing, apart from the pleasure of putting words into a certain order, was the peaceful, enclosed space an indexer inhabits. It requires a certain sort of concentration: you need to withdraw, as it were, into the world of that which is to be indexed.