I’ve been thinking about a piece of common sense advice that I was reminded of a few months ago. You may have heard it too. It goes like this: on your first pass through the book, pick up more than is necessary, because it is quicker to go back and trim than to go back and pick up a second time.
The assumption is that reading the text and picking up entries is time-consuming. Which is true. Carefully reading the book and selecting terms does take time. Also assumed is that you may not know, on the first pass, what is indexable and what can be left. Which is also true. By picking up an abundance of entries, all bases are covered. But is over-indexing and then trimming the extraneous entries really a faster approach?
To some extent, yes. I certainly pick up entries knowing that they may not make the final cut. But I still try to choose entries within certain parameters. This advice can get you into trouble when you have no filters (or very few filters) in place.
Consider this scenario, which I have seen and done myself earlier in my career: You get a book to index and the publisher informs you that the index can be no longer than 700 lines. Which sounds like a lot. Excited to start, and worried about missing relevant discussions, you jump into the project, typing up whatever entry seems like it might mean something to someone. Finishing the full draft, you realize that the index is pushing 1,000 lines. Or 1,500 lines. Or maybe even 2,000 lines. Now what?
Trimming an index that is a little longer than it should be usually isn’t a problem and is part of the indexing process. By definition, rough drafts need to be tidied. There are entries that aren’t as important as they originally appeared, or entries that can be combined into one, or subheadings that can be removed. But an index significantly longer than the target length requires more than a trim. It requires deep cuts. Chances are, whole categories of entries need to be removed and the index structure needs to be simplified. Not only has time been spent creating all of those entries, but time will now have to be spent figuring out how to get rid of all those entries. There is also the emotional pain of having to sweep away all that you created.
It is not efficient or effective.
What I try to do instead, before I start the rough draft, is to think about what needs to be in the index. I think about the subject matter and the arguments or stories that the book is telling. I think about who might be reading the book. I think about what kind of book it is, and how readers will use the index. If there is a length limit, I think about how much space there is relative to the number of pages and to how densely the book is written, and I then try to gauge how detailed the index can be. With all of that in mind, I can more easily identify and filter out passing mentions as I read the book. This approach is not perfect—I often still come across potential entries that I am uncertain about, and I may still go back and pick up entries that I have missed—but having a plan means that my rough drafts are closer to being completed. When editing, I can focus on polishing, rather than cutting and rebuilding.
This approach works both when there is a firm length limit, and when there is no limit to how long the index can be. Just because there is unlimited space does not mean that everything is now relevant to the reader. Think ahead about what the audience needs, and filter the text through that lens.
Besides planning ahead, what else can you do to avoid over-indexing getting out of control?
- Partway through writing the rough draft, stop and assess the number of entries so far. Is the number proportional to how much space is available in the index and/or to how many pages indexed? Being a little over is okay; some trimming is normal. But if you are only halfway through the book and the rough draft is hitting the line limit, then you probably have an over-indexing problem. Stop, think again about what is important to include and what can be excluded, assess and do a preliminary cut of entries so far, and then finish the rest of the rough draft with some new filters in place.
- Try running a quick search to see how often a term or name shows up, especially if you are unsure if it is worth including. If it is only mentioned once, then it may not be important. If mentioned multiple times, then probably worth including. But be careful: number of mentions should not be the only criteria for including in the index. The quality of the discussion is also important.
- Try using some sort of label or symbol to mark entries that you are unsure about. In Cindex, I can label entries with different colors, and I use red to indicate entries that I may want to cut. These are entries that I am on the fence about—I’ll still pick up so I don’t have to go back later, but I also want to make them easy to identify and delete. Whatever you use to label, remember to remove so that it doesn’t make its way into the printed index.
- Practice. Indexing is a skill that takes time to develop. It may take several indexes to figure out how to effectively plan ahead, and how to recognize and filter out passing mentions. And that’s okay. Keep working on it with each index, and you will gradually get the hang of it.
Writing an index is hard work. We can use all the hacks and shortcuts we can get. And that’s what common sense advice is—a way to make indexing easier and hopefully faster. The problem though, is that common sense advice usually doesn’t come with instructions for how best to implement that advice. Used the wrong way, and the advice may end up creating more work.
So yes, over-index, but within limits. Assume that the index will need to be trimmed and that some entries may not make the final draft, but still think carefully about what to include on the first pass. The cleaner the rough draft, the easier and quicker editing will be. Which is the most efficient way to index.