In this post, Ruth Martin outlines the things that shouldn’t appear in an index.
Write a bad index… us?!
Surely members of the Society of Indexers couldn’t write a bad index? Well, they could, but only in fun.
In 2022, Society members took part in a peer review, indexing the same piece of text and comparing the results. It’s a familiar training exercise, but this time there was a twist: we invited all the participants to ‘over-index’. We gave no lead on what we meant by ‘over-indexing’, but left this up to the indexers’ discretion.
The results were witty, surprising and at times bizarre, and they allowed us to conclude that a bad, that is to say, an ‘over-indexed’, index is characterised by:
As a rule of thumb, an index should be no more than 3%-5% of the length of the text.
Excessive subheadings and sub-subheadings
If the subject matter of a book is complex, then its index is likely to be complex as well, but complexity is not an end in itself. Like all good writers, indexers aim for clarity.
Long and discursive headings
It is not the job of the index to set out the author’s arguments, only to direct the reader to the points in the text where the arguments are raised. However well an index may be written, we do not intend for readers to linger and admire our work but to move on quickly (and precisely) to the sections of the text that are of interest to them.
Double-posted entries where cross-references would be better
Cross-references play various roles in an index. See also cross-references direct the reader to related content of interest. See cross-references direct the reader from a non-preferred term to a preferred term, such as in the example ‘vessels see ships’. This allows the inclusion of synonyms under different letters of the alphabet, and it also saves space: if an index entry for ‘ships’ is lengthy with subheadings, it is a waste to repeat the same information under ‘vessels’.
Excessive cross-references where double-posting would be better
If, on the other hand, the entry for ‘ships’ is a simple list of locators, then there is no space-saving advantage to not supplying the same information under ‘vessels’, sparing the reader the nuisance of having to leaf back to the S section of the index.
‘Vessels see ships’ is fine. ‘Boats see ships’ ditto. But does the index really need to include ‘barks’, ‘nautical craft’ or ‘liners’? It will depend on the text, of course, but generally we don’t strain Roget’s Thesaurus for an exhaustive list of synonyms. Such discretionary decision making is the stuff of indexing.
As the name suggests, a passing mention is a topic or entity that is mentioned by the author in passing. There is little or no substantive information provided in the text, and so it does not merit an entry in the index. Determining what is, and what is not, a passing mention is another core indexing skill.
Entries starting with insignificant words
Ever wondered why indexers use twisted phrases such as ‘Indexes, features of’ instead of ‘Features of indexes’? It’s so that the most significant word (here, ‘indexes’) is first in the phrase, making it appear in the alphabetical listing in the place where we predict most readers will look for it. An index entry filed in the wrong place might as well not be included at all.
So there you have it: the qualities of a bad index. But what of those of a good index? Is it just the opposite of all these things, or is there something more? For the answer to this, I refer you to Lyndsay Marshall’s blog post Qualities of a Good Index.
About Ruth Martin
Ruth Martin specialises in indexing books in the areas of law, politics and economics, although she has a growing list of literary criticism books on her indexing CV. She is a member of the Society of Indexers’ Executive Board, as well as its Marketing and CPD Committees. In 2021, she won the Institute of Certified Indexers’ Purple Pen award for newly accredited indexers.