Why does my book need an index?

Who reads a reference book from start to finish?  Well, some people may, but many more only have the time or inclination to dip in somewhere in the middle and extract the information that is relevant to them.  Without an index, you will lose these readers.

An index is also a sign of quality.  It tells the reader that the book is a serious work of scholarship.  By the same logic, if the index is poor (by which I mean that it does not guide the reader easily and accurately towards the information they want), it conveys the message that the book itself is not worth the reader’s time and trouble.

Which brings me on to…

Why use a professional indexer?

The short answer is because we’re good at it!

The slightly longer answer is that a professional indexer has the training and experience to identify and list the book’s themes and ideas in a clear, concise and logical fashion, and to link them together with cross-references.  A good index is not a mere list of terms (like a telephone directory) but a map that both guides the reader to where they want to go and suggests a few interesting detours along the way.

Indexers also have the skill to think like a reader.  They imagine what terms the reader is likely to look up and then direct them to the terms the author prefers.  They also know how to edit an index (boy, how we edit!) to ensure that it does not collapse under its own weight of terms or frustrate readers by referring them to scanty or irrelevant information.

These are all high-level cognitive processes, which explains why indexes (good ones) are still written by human beings and not by computers.  And the more skilful the human being, the better the index.  Stands to reason, really.

What is embedded indexing?

I have received a number of enquiries lately about embedded indexing.  Some publishers, notably Cambridge University Press, are asking authors to supply manuscripts for typesetting and final editing that have an index included.  Normally, the index is written at the very end of the publishing process from a finished 'locked' proof, but an embedded index can be produced at an earlier stage.  This is because XE (Index Entry) codes are inserted - 'embedded' - directly into the manuscript text, and these translate into the page numbers that appear in the index.  If new text is added, or text is removed, during a later edit, the embedded codes shift as well and index's page numbers update automatically.

It is possible to create an embedded index in Word, and there is plenty of guidance that is only a Google search away, but I don't recommend it.  It's a laborious process, and so not very feasible for a full-length book.  Instead, why not commission an indexer to take the strain?  We have experience of this kind of work and the software that makes it easier.

What is legal tabling?

In a serious work of legal scholarship, the index at the back of the book is complemented by the tables at the front.  Whereas an index will cite substantive references to a topic, the tables will include every mention in the book of a particular statutory provision, case or arbitral ruling, however fleeting or indirectly expressed.  So if your work requires you to research the Law of Property Act 1925 s.62 or applications of Donoghue v Stevenson, the tables are the best place to start.

Like indexing, legal tabling is a job that is best performed by a person, not a machine.  Take this example:  "In addition, art.21(1)(e) of Regulation 2018/625 states that the notice of appeal filed in accordance with art.68(1) of that regulation…”  A computer would probably table Regulation 2018/625 art.68(1) here, when, in fact, the correct citation is Regulation 2017/1001 art.68(1) - the words 'that regulation' refer back to a sentence earlier in the paragraph.  If you are going to the trouble of including tables in your work, it makes sense to ensure that their data are correct by engaging a professional tabler!

Author! Author!

The question I am asked by clients more than any other concerns the treatment of referenced authors in an index. Actually, there are two questions: should referenced authors be cited at all? And should every referenced author be cited and, if not, what selection policy applies? (OK, that’s three questions.)

The issue is contentious even among indexers, but my own view is that, in an index for an academic text, referenced authors should be cited. No academic writes in a vacuum. In fact, a key indicator of the quality of an author’s scholarship is the depth of their research and their engagement with confirmatory and opposing views. If a line of argument is based on another scholar’s work, why wouldn’t you want to direct your readers to it via this person’s name included in the index?

That said, I certainly do not believe that every referenced author should be cited. That’s the job of the bibliography. There’s little point in using the index to direct readers to another person’s work unless there is something more in the text than merely a name check.

Determining that ‘something more’ is where things get complicated. My policy is that referenced authors are cited in the index where they are quoted directly or their ideas are discussed explicitly. Direct quotes are usually easy to spot because they appear in inverted commas or in indented blocks. The quote may be as little as a single word. In a perfect world, explicit discussion of an author’s ideas would also be obvious because they would always be prefaced by such phrases as ‘Smith says…’, ‘According to Jones…’, or ‘Robinson argues…’  However, the Harvard system of referencing can obscure the distinction between an explicit and an implied discussion. Most indexers rely on the ‘outside the bracket’ rule (authors named outside the bracket are cited, those named inside the bracket are not), but this does not apply in all instances. Suffice to say, context is king. My aim here is merely to demonstrate that there is a rationale for the decision to include some author references and not others: it is by no means arbitrary.

Where multiple authors are referenced, I will normally cite the first three listed names and not the others. If ‘Smith et al.’ require a citation, only Smith will receive an entry. I also apply my author citation policy to footnotes and endnotes.  Of course, all these points are debatable and so negotiable. One last thing: sometimes we are asked to produce an ‘index of names’, citing each and every reference to an author. For me, the word ‘index’ here is a misnomer, and instead what is being described is a ‘table’ (see the above discussion ‘What is legal tabling?’). An index is a creative work, a product of the indexer’s discretion, guided by the rules and conventions laid down by the profession. I am happy to produce an ‘index of names’ if asked, but I will charge for it separately, applying my normal rate for tabling.

What am I looking for?

When indexing for a client, it’s my policy to complete a first draft of the index and send it to them for review. I then make any edits that they suggest, and create a final draft. This seemed a straightforward practice to me until, on a couple of occasions, clients have pulled me up short with a simple question: when I review an index, what am I looking for?

The best way I have found to answer this question is to ask the client to imagine a reader picking the book off a library or bookcase shelf, or selecting Amazon’s Look Inside, and turning to/clicking on the index. Does what they will see reflect the book’s contents? Will they find this information quickly and efficiently? Will they choose to buy or borrow the book?

Imagining the reader is a core skill of indexing. In one way, it’s not difficult for me to imagine the reader because I am myself a reader of the book, perhaps even its first reader without advance knowledge of its contents. But it’s also required of me to imagine readers beyond myself: the reader who is an expert on the book’s subject matter; the student who is using the book to help them with an assignment; the reader with an amateur interest in the topic; the reader who has stumbled upon the book and just might get hooked. It’s reasonable to assume that they won’t all use the index in the same way or look up the same terms, and yet it must serve them all. That’s why you may see the same pieces of information provided in the index in different places and using different wording.

There are, of course, good and bad practices in indexing from a technical perspective. The Society of Indexers has published two blog posts on the Qualities of a Good Index and the Qualities of a Bad Index by Lyndsay Marshall and, ahem, myself respectively. I recommend them to your notice.