Indexing Tips

Well, I just finished reading through the page proofs for my Crater book.  Now all that stands between me and a finished book is the index.  I have the formatting guidelines in front of me, but I would love to know what is the best way to proceed from those of you who have done this before.  I was going to take the traditional route and use index cards.  One index card for each term and place them in alphabetical order.  Seems simple enough.  The only thing I would have to worry about is a surprise visit from Felix or Jeb.

This is likely the last time I will ask my readers for assistance with this project so thanks a bunch.  I assure you that if I thought there was anything glamorous about publishing a book before this week it is a distant memory.  Why exactly did I put myself through this?

26 comments… add one
  • Mark Stoneman Apr 16, 2012 @ 15:55

    I’ve been happy with Twin Oaks Indexing for the stuff I edit for my institute, though I understand the impulse to do one’s own index. After all, it’s another chance to emphasize certain structural elements of the text. Still, I think it would drive me crazy.

  • pbinkley Apr 9, 2012 @ 20:37

    Word and other word processors have indexing functions, intended to be usd by embedding index term markers in the text; but you can use them independently of the text by creating a dummy file with page breaks to create the pages, and enter the index markers on the appropriate page. Compiling (and recompiling) the index into its final form is then easier than the sort-and-deduplicate method.

  • Vicki Betts Apr 4, 2012 @ 5:15

    When I’ve done indexing I’ve noticed that it becomes the final and ultimate proof-reading, because I’m focusing on individual names and words rather than flow. Mostly I find variations on the spellings of names of people and places, and incomplete names. I’m glad you are doing topical indexing as well. I’ve noticed that many recent history books are proper name only, and at most maybe ten topics, leaving it up to the researcher using the book later to basically come up with an additional index penciled in on the end papers. Or at least that’s what I’ve had to do.

    Vicki Betts

  • Joshua Horn Apr 4, 2012 @ 3:45

    I have used Adobe Indesign for book layout, and it has some very nice indexing tools. You can have it automatically add index entries based on a search term, such as James Longstreet. If you find a phrase you wish to index, you can add entries for every time that phrase is used. I am not sure if this would work for you with your workflow with your publisher, but it has worked well for me.

    • Kevin Levin Apr 4, 2012 @ 3:47

      I will check it out. Thanks.

    • Michael Douglas Apr 4, 2012 @ 5:18

      I’d second the use of InDesign’s indexing tools as well as those of QuarkXpress. The problem is that Kevin’s got to have the software, and it’s not inexpensive.

  • Matt Gallman Apr 3, 2012 @ 18:56


    First off, I would really urge you not to hire an indexer. (That is, since you’ve said you aren’t going to hire an indexer, I would say “well done.”)

    Second, it is worth thinking about how readers actually approach indexes. I pretty much read books for a living, and I generally scan the index and footnotes before I read a word of the text. The latter gives me a sense of what you looked at and who you read, the latter gives me a sense of what you talked about. I’d say to index for those readers, and also – for your book – index for the person who is going to pick up the book and check to see if X is in it.

    Third, what do you include? Well, some of that is obvious (and, needless to say, the Press has a good guide and the Chicago Manual of Style is determinative). Essentially you index nearly all of the proper nouns: names, places, organizations, holidays, regiments, pieces of legislation. They all get in there.

    Third, part B, and this is more tricky. You should be careful about including some key groups and concepts: memory, women, African Americans, patriotism, union, Lost Cause etc. The point here is to mark those topics that you actually engage. Note that some readers will draw conclusions based upon what is not in the index. [“Women only appear on three pages …”] Note, also, that it is useful to cite the same thing in multiple places: “blacks: see African Americans; see U. S. Colored Troops.”] That is, the reader should find the same stuff even if they use four different labels.

    Fourth, how do you mechanically do this? Here is my strategy: I read the page proofs with a colored pen. I circle every word I wish to index. I write up in the upper right-hand corner every “concept” I wish to index. I do this on a hard copy, Usually in a coffee shop or under a tree. Then, I open up a Word file and just start typing the words that appear, page by page in a column, followed by a comma and the page number. If the “concept” spans several pages, I note that, If the same name appears on several consecutive pages, I might note that [ie Smith, John E., 19-22.] When all the dust settles I have a long word file in a column, with single hard returns after each term.

    Then I just alphabetize the whole thing and combine it into an index

    That is what I do

    • Kevin Levin Apr 4, 2012 @ 1:46

      This is incredibly helpful, Matt. Thanks so much for taking the time to share your approach to indexing.

  • Woodrowfan Apr 3, 2012 @ 16:50

    I paid a professional freelance indexer. The going rate was about $4.00 per indexable page. It cost me about $500-$600.

    • Kevin Levin Apr 3, 2012 @ 17:18

      Unfortunately, I can’t afford to pay a professional indexer.

  • Eric Wittenberg Apr 3, 2012 @ 16:39

    My method works best, Kevin. I will give you an ironclad guarantee.

    Pay someone else to do it. You will be much happier.

    Works every time for me.

  • Jonathan Mahaffey Apr 3, 2012 @ 15:13

    It took me about 30 seconds of trying to figure out why Jeb Stuart would be in a book on the Crater (and why that would mess up the indexing) before it dawned on me that Felix and Jeb must be your pets! 🙂

  • Jimmy Dick Apr 3, 2012 @ 14:11

    That brings up the professional index folks. How much do they cost?

    • Andy Hall Apr 3, 2012 @ 14:28

      I hear Levin got a six-figure advance on this one, so money should be no object.

    • Mike Musick Apr 3, 2012 @ 14:39

      I would urge you to index regiments and other units the correct way – something that’s been largely lost over the years. Examples are in the “Official Records, Armies” General Index volume, viz: Georgia Troops (Confederate) – Infantry – Regiments: 48th. Or, United States Colored Troops – Cavalry – Regiments: 5th. Definitely not Forty-eighth Georgia Infantry, under “F.” All sorts of other unfortunate methods have been used, especially in the latter part of the last century. Thanks for allowing me to air a pet peeve!

      • Kevin Levin Apr 3, 2012 @ 14:53

        Thanks for the suggestion, Mike. The only problem is that regiments have already been formatted as Forty-eight Georgia in the text as opposed to 48th Georgia. I followed the publisher’s guidelines on this.

        • Mike Musick Apr 3, 2012 @ 15:03

          I understand completely, Kevin. Que sera, sera. I’ll refrain from berating your publishers and trying to point out to them the error their ways.

          • Kevin Levin Apr 3, 2012 @ 15:06

            I will remember it for the next project.

            • Terry Johnston Apr 3, 2012 @ 17:19


              But you might still use that format, perhaps. Such as:

              Georgia Troops (Confederate)


              Forty-eighth Regiment, p. X

              • Kevin Levin Apr 3, 2012 @ 17:23

                Hi Terry,

                I thought the same thing and will definitely look into it tomorrow morning. Thanks.

                • Terry Johnston Apr 3, 2012 @ 17:25

                  You might check one of Garold Cole’s Civil War Eyewitnesses volumes as inspiration. If you don’t have access to either, I’d gladly scan you a few sample pages of his index. It’s really good. And easily doable, I think.

                  • Kevin Levin Apr 3, 2012 @ 17:27

                    Thanks. I will take you up on that offer.

          • Vicki Betts Apr 4, 2012 @ 10:46

            Kevin–this is entirely off-topic so you may not want to approve it in this thread. Seeing Mike Musick posting here, I’d like to ask him if, during his years at the National Archives, he has ever run across any Civil War documents or set of documents that just called out for someone to tackle them as the basis for a really intriguing project, but he never found the time to do it himself.

            Vicki Betts

            • Mike Musick Apr 10, 2012 @ 17:25

              Hi, Vicki! Sorry to take so long to get back to you. Yes, there are all sorts of potential projects in the NARA records. I would just say that what fascinates one person might well not fascinate someone else. If Kevin will pass along your e-mail address, we can discuss it. Thanks for thinking of me.

  • Ken Noe Apr 3, 2012 @ 11:57

    It’s an unpleasant chore any way you do it, but here’s the quickest way I’ve found. Read through the mss and type names, subjects, and page numbers into a Word document. Then use the sort command to create a very messy alphabetical index. From that, create the actual index by combining similar entries. For example:

    Floyd, John 34
    Floyd, John, 112
    Floyd, John 213

    can easily become: Floyd, John, 34, 112, 213.

    • Kevin Levin Apr 3, 2012 @ 14:53

      Thanks, Ken.

Leave a Reply

Your email address will not be published. Required fields are marked *