An Index or a Concordance for Your Book?
I hear some publishers talk about automatic indexing software and concordance generators when I approach them to offer my book indexing services. I’ve heard them say, "Oh, Word generates a list of terms in the book for us" or even, "We don’t have indexes in our books. Do people really do that kind of work? I thought computers came up with those lists."
Only people index. And they do it well, in most cases; better than computers, in all cases. And finally, they do it without using any kind of automatic indexing software. [They do use specialized computer software however; specific tools that an indexer does use will be discussed in a future article. Check back often.]
Let us define our terms. A concordance is, according to Webster’s Collegiate Dictionary (10th ed.), "an alphabetical index of the principal words in a book or the works of an author with their immediate contexts." An index is, from the same source, "a list (as of bibliographical information or citations to a body of literature) arranged usually in alphabetical order of some specified datum (as author, subject, or keyword): as a: a list of items (as topics or names) treated in a printed work that gives for each item the page number where it may be found." The definition for index goes on to include a number of related ideas, most of which have to do with "indicating." A better definition, for this particular usage of the word index, is this: "An index is a structured sequence—resulting from a thorough and complete analysis of text--of synthesized access points to all the information contained in the text. The structured arrangement of the index enables users to locate information efficiently" (Mulvany, 1994).
Now then, we have defined our terms. Now we can determine what the needs of a particular book are. It may be that a concordance would be useful, but for most books, an index is what is needed. A concordance generator produces a list of words that occur in the book, with the exception of those on the stop list (lists of prepositions, conjunctions, common articles and the like). So, you will see that it is a listing of mainly nouns, verbs and other words not included on the stop list. It will also produce a list of pages or references where these words occur. This sounds good—it sounds like a useful thing. But, you will note that there are several things missing that a standard index should contain.
If a text contains the words roses, snapdragons, petunias, orchids and violets, there will be page references next to each in the concordance. But there will be no listing under the umbrella term, flowers. Think about that for a minute. And while the text clearly discusses what type of fertilizer to give each plant, it might never mention care and feeding. It is precisely these issues that require a human to analyze the content and organize the information. All but the simplest of books need a map. And that is what an index is.
A concordance cannot distinguish between a term that is mentioned in passing at a few locations in the book and a term that is fundamental to the theme or focus of the book. The relative importance of words and concepts can only be determined by a subjective and intelligent analysis of the context of the discussion. No computer program can yet do this.
Cross-references provide alternate pathways to the information contained in the index. There are two main types of cross-references. See and See also. A See reference indicates that there is no information contained at that entry point. It points the reader to the correct location to find information on that topic. A See also reference points readers to related information in addition to what is found at that entry. These are not the same, but the difference between them can be subtle. Regardless, they are useful and common features of a good index. Cross references can comprise between 1 and 10 percent of all the locators (page references) in an index. Mapping routes through the index is a prime responsibility of a good indexer.
Double-posting creates a duplicate entry under another, often synonymous, term. In the book that mentioned flowers, you might see that there is discussion of St. John’s wort (also known as hypericum). If it were double-posted, it might be listed under both of these in the index, with full references to all information contained in the book under each listing. Of course, double-posting is commonly eschewed in favor of See references, because of frequent space limitations in an index.
Sub-headings are the real meat-and-potatoes of a good index. A concordance does not have the intellectual and conceptual framework to support an in-depth analysis of topics in sub-heading form. In our flower book, a good index might contain a listing of all of the flowers, and specific topics under each flower noting the context of the discussion. For example: "Roses, mildew on" or "St. John’s wort, medicinal applications of" or "Corn, care and feeding of," to name but a few.
An index and a concordance share very little. They are different creatures entirely, each serving a particular need. Concordances lack the analysis that goes into an index, and are consequently less useful. They can be, and are, useful as tools for the information professional; but I wouldn’t recommend them to the general publishing field.
A Note on E-Books
The growing popularity of the many forms of portable electronic books has serious repercussions for the information retrieval industry (indexing and cataloguing professionals). Should your e-book have an index? Should you rely on the full-text search capabilities of the e-book platform itself? What are the costs involved?
These are important questions for publishers and indexers. You don’t want to spend money on an index if a search engine will do the trick. Check next month’s article for an examination of search engine technology and indexing.
Mulvany, Nancy C. (1994) Indexing Books. The University of Chicago Press. (Chicago)
Webster's Collegiate Dictionary, Tenth Edition (1997) Merriam-Webster. (Springfield, Massachusetts)