Here's a piece about scholarly book reviewing. The author makes a number of good points about the importance of the practice, in defense of it against the many critics who suggest it's largely a waste of time and energy.
Her key argument, I think, is this:
Reviewing books provides a focal point for deeper thinking and an avenue through which to share those thoughts in a brief and accessible way. Rather than absorbing time better spent writing journal articles, reviewing academic books actively underpins the kind of critical thinking, deep subject knowledge and exposure to current trends that are essential to longer outputs. The practice ultimately facilitates, and is almost a natural by-product of, the kind of thorough and detailed research we naturally undertake as scholars. We should therefore view it not as an intellectual cul-de-sac but as part of a holistic scholarly process.
"Part of a holistic scholarly process"--that seems smart to me. And I agree with it. That said, however, there are good reasons to be wary of doing too many book reviews, especially if you're a younger scholar.
Book reviews are actually quite hard to do. In the space of a thousand words, or maybe two thousand if you're lucky, you're supposed to summarize a book's argument, place it in some scholarly context, and make a series of judgments about its accomplishments and weaknesses. I mean, the author of the book had to take hundreds of pages to do that--and trust me, they know their book better than you do. (Well, most of the time.) How can you manage all that in four to eight double-spaced pages?
Furthermore, if you do undertake it, the level of scrupulosity you must employ is really high. A book review is a bit of a high-wire act, and everyone can see you attempt it. So people take a lot more time getting the argument right than they normally would. And you know what? Most academic books aren't written for ease of comprehension (at least in the humanities). There's a lot of muddyness in them. Authors leave a lot of mess for you, the reader, to clean up. (We ought to be more angry at them than we normally are about doing that, by the way. But that's another post. Or maybe a whole lot of other posts.) And what if you do a great job of reviewing a book you think well of? Well, because your text is so thoroughly secondary, most people will write it off as simply directing them to the book (which it should do). And if you write a negative review of a book, what then? Well, you may well have earned a serious enemy. And while people can enjoy a nasty, slashing review, few people end up admiring the person wielding the rapier. They mostly want to stay away from them.
In general, I believe that you should review at least one book a year (and probably no more than two or three, unless you have reason to dedicate a good bit of energy to book reviewing, for reasons of letting other research lines like fallow for a bit, or something like that). It's a scholarly duty, effectively--to alert people to new work, to relate that work to work already known in the field, to participate in the field-wide conversation that occurs, in a technical sense, "occasionally"--at conferences and talks and over email and in journals--but which conversation embodies the life of the field as a whole.
What should a book review do, you ask? Here's Edmund Wilson, from his essay "The Literary Worker's Polonius," on that:
The reviewer, at the very least, should be expected to supply information. The retelling of the story of a novel, the summary of an historical or philosophical book, is the most boring part of the reviewer's business, but it is an absolutely essential part. The reader should be given a chance to judge whether or not he would be interested in the book, irrespective of what the reviewer may think of it; and it is an indispensable discipline for the reviewer, or any critic, to give the gist of the book in his own words. The reviewer, when he sets about this task, is quite likely to find that there is more in the book, or less in it, or something different in it, than he imagined when he first went through it. If the author is incoherent or woolly, the critic will be able to detect it. If the reviewer is incompetent, his incompetence will be evident to his more acute readers when they find out he cannot tell them what is in the book.
That's good advice. It influenced what I tell my students: The job of the book reviewer is to introduce new work to the intelligent reading public. By "introducing" here, I mean explicating the context into which a book intervenes, and assessing the successes and failures of that intervention. Normally this task entails the following:
1. The review must tell the reader what the piece under review is about.
2. The review must tell what the author of the piece says that the piece is about.
3. The review must tell what she or he thinks about what the author of the piece says that the piece is about.
4. The review must report on the reviewer's views on this matter, in at least two different dimensions:
4a. First, the reviewer must judge the execution of the argument--that is, are the internal steps of the argument well made? Does it offer solid evidence, make good logical moves, clear inferences, and lucidly and intelligently defend its argument against relevant challenges?
4b. Second, the reviewer must judge on the conception of the argument--that is, is it worth caring about this argument, for the field (or fields) into which the book means to intervene? This requires some "field knowledge" on the part of the reviewer; in fact, a really good book reviewer will sometimes be able to place and assess a book's intervention in multiple fields, more even than the book gives evidence of being aware of or attempting self-consciously to influence. This is often the most "visionary" and constructive dimension of a review: for in this part--which is almost certain to be very brief, a paragraph or so in a typical academic review (of 500-1000 words), the reviewer offers you a map of the field, noting perhaps the particular problems or conversations in light of which the book can best be understood and assessed. I often read reviews, not simply to learn about the book, but also to get a better understanding of a particular reviewer's report on the state of the field.
Finally, what criteria should you use to anchor these judgments? The mid-twentieth-century Polish writer Witold Gombrowicz, one of the most astounding writers of the era, gave this canny advice, which adroitly gives you permission to avoid the word "judge" (even though what he says to do instead is what I think we ought to mean by "judge"):
Literary Criticism is not the judging of one man by another (who gave you this right?) but the meeting of two personalities on absolutely equal terms.
Therefore: do not judge. Simply describe your reactions. Never write about the author or the work, only about yourself in confrontation with the work or the author. You are allowed to write about yourself.
"Simply describe your reactions"--that seems exactly right to me. It's not about proving the accuracy of your aesthetic judgments, simply showing what provoked them.
One last thing. As in all things, models are crucial, and so everyone can have favorite book reviewers, and anyone who has a scholarly vocation should have favorite book reviewers. You should develop a pantheon of people whose assessments you admire. In my personal list, it would include some more accessible "public" writers like James Wood and Daniel Mendelsohn, in a quasi-popular (but at least public) way the historians Peter Brown and Anthony Grafton, and the literary scholar Helen Vendler. More purely scholarly book reviewers I have found particularly impressive include Carolyn Walker Bynum, Arnaldo Momigliano, perhaps Christopher Ricks. All these reviewers combine, in different ways and to different degrees "expert knowledge" and outsider accessibility in good ways.
If anyone who reads this has book reviewers they especially admire, I'd love to hear from them. The excellent book review--and as a sub-set of that class, the excellent scholarly book review--should be a genre with its own Hall of Fame.