I opted to compile the index of A History of Place myself. I made this choice for various reasons, but the main one was that the index seemed to me to be an important part of the volume’s framing and presentation. Reflecting on this, it seems a little ironic, as in some ways a book’s index exemplifies the age of the pre-digital publication. Using someone’s pre-decided terms to navigate a text is antithetical to the expectations and practices of our Googleized society. Let’s face it, no one reading the e-version of A History of Place is ever going to use the index, and in some ways compiling the index manually, reviewing the manuscript and linking key words with numbers which would, in due course, correspond with dead-tree pages felt almost like a subversive act.
But like an expertly curated library catalogue, an expertly compiled index is an articulation of a work’s structure and requires a set of decisions that are more complex than they may at first seem. These must consider the expectations and needs of your readers, and at the same time reflect, as accurately as possible, the current terminologies of your field. The process of indexing made me realize it gives one a chance (forces one in fact) to reflect – albeit in a bit of a hurry – on the key categories, terminology and labels that oneself and one’s peers use to describe what they do. It thus forces one to think about what terms mean, and which are important – both to one’s own work, and to the community more broadly (some of whom might even read the book).
There is also the importance of having a reliable structure. As I outline in the book itself, and have written elsewhere in relation to crowdsourcing, some have argued that using collaborative (or crowdsourced) methods to tag library catalogues for the purposes of searching and information retrieval disconnects scholarly communities from the ‘gatekeepers of the cultural record’, which undermines the very idea of the academic source itself (Cole & Hackett, 2010: 112–23) [1]. Cole and Hackett go on to highlight the distinction between “search” and “research”; whereby the former offers a flat and acrticial way into a resource (or collection of resources) based on user-defined keywords, whereas the latter offers a curated and grounded “map” of the resource. While, in this context, Cole and Hackett were talking about library catalogues, exactly the same principle applies book indices.
I don’t wish to overthink what remains, after all, a rather unglamorous part of the writing process; however even in the digital age, the index continues to matter. Even so, there is no shame at all in busy academics (or any other writers) delegating the task of compiling an index to a student or contract worker, provided of course that person is fully and properly paid for their efforts, and not exploited. But I think it is necessary to have a conversation with that person about strategy and decision making. What follows is some examples from A History of Place which exemplify issues which authors might wish to consider when approaching their index, and/or discussing with their indexer. By discussing these examples I try to explore the decisions I made about which terms and sub-terms I decided to include, and why.
To begin with the practicalities, the wise advice provided by Routledge was:
You don’t have to wait for the numbered page proofs of your book to arrive – start to think about entries when you have completed the final draft of your typescript. The index is always the last part of the book to be put together and submission of your final copy will be subject to a tight deadline. Preparing it now may save you time later on. [emphasis added]
I would suggest that it is a good idea to think about these even before the numbered proofs turn up.
And then
On receipt of [numbered proofs], you should return to your already-prepared list of words. Use the numbered proofs to go through your book chapter by chapter and insert the page numbers against each entry on your list. (You can use the ‘Find’ function to locate words within the proof PDF.)
The gap between compiling your original list and adding page numbers will help you to evaluate your designated entries once more. Have you missed anything obvious? Are your cross references accurate and relevant? Revisit the questions under the heading ‘Choice of Entry’.
When you are satisfied that your index is complete, put it into alphabetical order.
You will come to love proper names in the early stages of this process. For example there is only one way you can represent Abraham Ortelius, or Tim Berners-Lee in your index; and no decisions involved in how to define the page limits for the references to them.
However, the process of selecting abstract terms for inclusion is more challenging. There were arguments both for and against including the word “Bias”, for example. All maps are biased of course, and in theory this could have applied to most of the examples I discuss. However, it forms an important topic of much recent literature on neogeography (for example), which address the ways in which neogeographic platforms perpetuate social bias due to their demographies (mostly white, male, Western etc). Therefore, inclusion made sense as it referenced explicit discussion of bias in secondary literature (mostly in the chapter on neogeography). It was possible to connect this to “collective bias” via the cross-referencing option of “see also”, of which Routledge advises:
- See
If the entry is purely a cross reference, the entry is followed by a single space, the word ‘see’ in italics and the cross reference. For example:
sensitivity see tolerances
Note that under the entry for ‘tolerances’ there is no cross reference back to ‘sensitivity’. Page numbers should not be stated where ‘see’ is used.
- See also
This should be used to direct the reader to additional related information.
This is a useful distinction, because it forces one to consider whether terms are synonymous versus relevant. “Bias” and “collective bias” is a good example as the original term is somewhat fluid and required some pre-hoc consideration but is clearly different from “collective bias”.
Highly specific and specialized terms presented less of a problem. Chorography, for example, features prominently in my index, but it could potentially have had any number of “see also…” references. However, given it is such a specialized term, I made a pragmatic decision (based partly on what I thought a reader using the index would need/want) to have it simply standalone, with no cross-references at all.
The most challenging terms were the big, important ones with multiple potential meanings. “GIS” is probably the most obvious example for A History of Place. Most of my arguments touch in some way on how spatial thinking in the humanities has emerged from, and been shaped by, GIS and related technologies, so the challenge was to divide the term up in to subsections which are a) useful for a potential reader, and b) reflective of disciplinary practices. My strategy was to treat branches of GIS which have been explicitly recognized and differentiated in the literature – such as Critical GIS; Qualitative GIS; Participatory GIS Historical GIS and Literary GIS – as separate index terms, linked as “see also” references. These are then tied only to specific occurrences of that term in each case. For discussions of GIS not explicitly relating to those terms, I used “and…” references which were tied to my chapter themes. This enabled me to divide the myriad references to GIS into sections which accord logically with the book’s structure – “- and archaeology” “- and and spatial analysis”, “-and text”, “-and crowdsourcing” and so on.
“Neogeography” created similar problems, but this type of term is compounded when the field moves so quickly. A recent paper by Linda See and others illustrates just how difficult this term is to pin down. I think all I can draw from this is that such index terms will need some considerable revisiting in the event of there being any future editions(!).
So, the agenda for that initial conversation with your indexer should, I would suggest, include:
- Strategies for dealing with abstract terms, and deciding which are relevant and which are not
- Highlight important, wide ranging terms, and what sub-categories you think they should have
- How to identify specific terms which may or may not need “see also” references
- Which sort of circumstances demand you to signpost between related terms using the “see” option.
- Flag terms – for your won reference if nothing else – that may not be easily “future proofed”.
[1] Cole, R., & Hackett, C. (2010). Search vs. Research: Full-text repositories, granularity and the concept of “source” in the digital environment. In C. Avery & M. Holmlund (Eds.), Better off forgetting? Essays on archives, public policy and collective memory (pp. 112–123). Toronto.
Stuart Dunn 