The Forbidden Question
Posted by: Rebecca Vnuk
When someone reads the entire Encyclopaedia Britannica or Oxford English Dictionary, it is so exceptional that it often seems to be the occasion for a book about the experience. Nobody should be surprised, then, that when I reviewed Britannica for Reference Books Bulletin about ten years ago, I did not read every word. I did not read even a volume’s worth. How much less than a volume’s worth shall remain a secret (and besides, the details are lost in memory).
For me, reading only a carefully selected fraction of a reference book I am reviewing (rather than the whole thing) is the norm. I always read the introductory material and the articles or sections of interest to me, as well as the parts on which I might have a smidgen of expertise; the latter is essential because it is where I am most likely to be able to detect thoroughness and accuracy as well as their opposites. For instance, whenever I reviewed various general encyclopedias, I always read closely the articles on Houston, where I have lived most of my life, and Texas, where I have lived my entire life.
I admit that I would feel cheated if I knew that a book I had spent thousands of hours researching and writing was reviewed on the basis of an incomplete reading. Especially a negative review. I strongly suspect that almost any publisher or author would feel the same. But is a very lengthy reference work the same in that regard as, say, a scholarly monograph?
Here, then, is the forbidden question to reviewers (of reference books, but also of other types of books): How much of a book you are reviewing do you typically read?
This suggests other questions: How much should you read? What role does the length or nature of the book play? How does an online version change the equation? Who could even read updates to Wikipedia as fast as they are being produced? How do authors, publishers, and Mary Ellen Quinn, my editor at Booklist, feel about this?
September 9th, 2009 at 10:52 am
I think it is acceptable to read a work in the manner for which it was intended to be read. No editor of an encyclopedia expects a single individual to read the entire thing, so why should the reviewer?
Furthermore, I don’t think publishers would expect the book reviewer to read the whole thing in this case, so there is no disingenuousness here.
September 9th, 2009 at 11:26 am
No reviewer can be expected to read an entire encyclopedia or reference book. But long-time Reference Books Bulletin reviewers may remember that our old Editorial Board manual went into great detail about systematic sampling. A reviewer would start at a randomly chosen entry or page and then read every “nth” item after that. There is also cluster sampling (looking at all entries within a cluster, such as some part of the alphabet) and purposive sampling (for topics known to be prone to dating or bias). Whatever sampling process they used, our reviewers were expected to explain it when they turned in their reviews. Although we don’t require an explanation any more, some thoughtful method of sampling still seems to be a very good way to get a sense of a reference work without reading it all the way through.
September 14th, 2009 at 11:21 pm
As a long-time reviewer, I can say that, when reviewing a reference source, looking at a random sample of articles, the introductory matter, the credentials of the authors, editors, and contributors, and the indexes, bibliographies, resource lists, etc. gives me a good picture of the work. Knowing that it is easy for readers to use and that the content is accurate and reliable is important. Librarians want to know that they are spending their scares funds wisely.