(Previous entries in this series: Pts. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, and 12.)
It’s appropriate that this entry is the thirteenth in the series. Academics generally hate compiling an index, even though it’s one of the most important tools that they (at least historians) use when reading.
There are software programs available that can create a book index for you. I understand that they are good at indexing terms, but not concepts, but they might be useful for you. I know some people also prefer to hire someone to compile the index for them.
I prefer to create an index the old-fashioned way (sort of). My system this time went like this:
- Between the copy-edit and the page-proof stages, I compiled a skeleton index of names and concepts from the copy edits. (The page numbers were still not set at that stage, so it was pointless to include them.)
- When the page proofs arrived, I read through them, highlighting terms in different colors. I used separate colors for Jackson-related events, Jackson-related concepts, non-Jackson proper names, and geographic names. (That turned out to be overkill-see below for why.)
- I then searched the .pdf page proof file for proper names as I came to them in the hard copy of the page proofs. I took this approach just in case I missed something in my skeleton index. (I found a handful of missed terms that way.)
- Once I completed the proper names, I moved to concepts. Searching in .pdf files for concepts will not find every mention, but by this stage, I was familiar with the flow of names, events, etc. That allowed me to do a basic search in the .pdf file for a concept (e.g., kinship), then work from the places in the hard copy where that word (and its variations) appeared.
A word of caution:
When searching .pdf files, make sure to use every variation of a term. For example, searching simply “secession” would miss “secede” and “seceding” (and “Secession” if you were using the “match case” search function”). The same goes for proper names. For example, I used both “Andrew J. Donelson” and “Andrew Donelson,” so searching for only one variation wouldn’t have worked. You also have to be careful if there are brackets within a word (e.g., “[J]ohn Donelson”).
The index that I completed for Andrew Jackson, Southerner isn’t as detailed as the one for Old Hickory’s Nephew. To keep from having lots of extra blank pages, which looks unsightly and adds to the book’s cost, the publisher gave me the following parameters: 34 manuscript pages, 25 lines per page, 41 characters per line. That comes out to ten published index pages, which isn’t a lot for a history book. My original index manuscript was 52 pages, so I had to eliminate a lot of detail to make it under the required length. In hindsight, I spent a lot of time on detailed entries that won’t appear in the book’s index. That’s unfortunate, because conceptual indexes are more useful, in my opinion.
Since I’m not an indexing expert, you might find find it useful to search for threads on the topic at the CHE forums. Seniorscholar‘s system, for example, is very much like my own. I would also be interested to hear readers’ experiences with what works and doesn’t work for them when they index.
Pt. 14 is here.
3 thoughts on “The Evolution of a Book, Pt. 13: Indexing the Book”
Oddly enough, I’m doing my index right now too. It actually came in short the first time through because I haven’t been nearly as thorough as you are being. I was basically obsessing over place names and refrigerator brands. When I go back and break up giant entries like “ice” and “refrigeration” and add the stuff I missed the first time around I think it will be alright.
I farmed out the index on my first two books but never again. This isn’t that hard if you have a week or so to spare. Besides, since I don’t have nearly as healthy an attitude about typos as you do, this gives me another excuse to read the whole darn thing a couple of times more. [I got 37 corrections and I still have 3 left!!!]
I didn’t mention the time factor in my post, but I spent at least 40 hours on reading the proofs and doing the index. I still think it’s worth it, but it sure is a time suck.
I’m on my fourth (and last) time through now. I haven’t bothered to count hours, but I do know that any mistakes that will remain will not be obvious ones and that’s probably the best I can hope for.