INDEXING BOOKS by Nancy C. Mulvany;
University of Chicago Press, 1994. pp xiii + 320; ISBN 0 226 55014 1 £23.95 / $29.95

Misleadingly simple a title, this one: We have here no brisk "how to do it" manual. Indeed, its author, a professional indexer who teaches indexing (at the University of California Extension in Berkeley), opens with caveats: "I do not believe that indexing can be taught ... Indexing books is a form of writing ... a mixture of art and craft, judgment and selection ... Indexing skills can be nurtured and rules can be learned, but the indexer's ability to thoroughly digest the intentions of the author and anticipate the needs of the readers, thereby producing a knowledge structure that is sensible and useful, involves the application of abilities and skills that are inherent in some individuals and not in others."

Wannabee indexers would nevertheless do well to start here, receiving good, basic guidance from "Getting Started" onwards. The book production process, relationships between the various parties involved, the skills needed by indexers, structure and arrangement of entries, typography for indexes, computer programs -- all are covered. Mulvany defines an index as "a structured sequence -resulting from a thorough and complete analysis of text - of synthesized access points to all the information contained in the text ... a network of interrelationships ultimately an interface between the author and the reader".

All kinds of LOGOS readers can profitably read this book. Publishers can learn from it how to draw up clear and professional contracts for indexing, how to brief the indexer, what the style guide should cover and of proper costs and copyright for indexes. Authors can learn how to plan ahead for the index, how to find an indexer, how to help or antagonize him or her, how to review and edit the index. Editors can learn how to assess indexes and to salvage the least worst calamities. Information technologists can study details of electronic line numbering, codes for subentry levels, formatting indention levels, embedded indexing tools, user interface design and appendices dealing with ASCII tables and generic coding for special characters. And distributors should note her comments on the effects on potential buyers of a bad index.

As for adept indexers, they can relish the detailed consideration of extreme niceties and complexities of indexing, such as intellectual versus algorithmic analysis of text; reference locator formats; listing much-married women, or names of other nationalities; transliteration; multi-author and multi-volume works; spelling out 80486 CPU for alphabetization; KWIC, KWAC and KWOC listings; tracing a name occurring only in a note reference back to its relevant text passage. There is speculation, too: should indexers scrupulously follow rules of practice not likely to be known to users? How intense should cross-referencing become? "The index should not be a vehicle for the indexer to demonstrate prowess in tracking down the genealogical roots of every individual mentioned in text," Mulvany admonishes.

But, as she emphasizes, indexing is an art, no mere set of precepts to be correctly followed: there is bound to be disagreement among concerned practitioners. My own chief dispute with Mulvany is where she dismissively rules: "There is one, and only one, reason to use a run-in format for an index: that is to save space ... it should be considered only when there definitely is not enough room for indented index." I would contrariwise declare: "Indented subheadings entail a staccato, disjointed effect and should be used only for subheadings using expected, standardized terminology, alphabetically arranged, with the keyword to the fore" (these things not usually true of biographical, historical or general humanities indexing). Oh, yes, we indexers too have our fervent disputes as to best practice. And as for Mulvany's dogmatic condemnation of strings of more than five undifferentiated reference locators and suggestion that the main topic of a book need not have its own index entry ... no, this is not the place.

Who could dispute that this is an author eminently qualified to index her own book? But instead, for reasons she cogently adduces, including exhaustion and over-familiarity with her text, she delegated this work. No doubt she followed her own precepts regarding author-indexer collaboration. She thus gained her book an excellent index, boasted at its head as "Written by Carolyn McGovern",
and expatiated in a footnote on its own first page (which is itself indexed) as "alphabetized letter-by-letter. Leading function words in subentries are not alphabetized. The page-number compression style follows that in The Chicago Manual of Style, 14th ed, section 8.69. The index was prepared with the Macrex Indexing Program. Using the formula on page sixty-four of this book, this is a 7% index. There are 2,062 entries, averaging seven entries per page."

The virtues and excitement of indexing (yes!) come through loud and clear. 'The phrase automatic indexing is an oxymoron," Mulvany insists, and demonstrates. "The index is molded and remolded through the first 'sweep' through the text ... As the indexer adds new entries, old entries are constantly being manipulated. A biological metaphor for this process would not be far off the mark. New cells grow and old cells divide; synergy is at work that results in a functioning organism in which renegade or mutant aberrations have been identified and eliminated. Index writing integrates substantive editing into the intitial creative writing process. This index goes beyond the words in a text. It provides a gateway to ideas and information ... Master the art of book indexing, and you will experience the magic of sharing knowledge."
That's indexing talk, that is.

-- Review by Hazel K. Bell in LOGOS 6/4 1995