I’m reviewing a new history of technology book for American Scholar; it’s probably the fifteenth review I’ve written. Partly I accepted the gig for the money (though given how poorly book reviews pay, that’s a very minor reason); mainly it’s a chance to work with Adam Kirsch, the Scholar’s new book review editor. And I appreciate the opportunity book reviews afford to do some more interesting things than mere criticism and evaluation: they’re not as formal as articles, but they require you to be more thoughtful and responsible than in a blog or e-mail. A poorly-written review is a serious insult to an author: it says that their work isn’t worth thinking about.

This afternoon I spent a couple hours working with a visitor to the Institute, explaining what long-range futures work is, and how it differs from the shorter-range market research and forecasting that most companies do. Having gotten into pedagogical mode, I started thinking this evening about the question, What makes a good review? How do you write a good one? Here are some ideas.

Engage with the choices the author makes. One of the most valuable pieces of advice I got in graduate school was delivered to me in a seminar after I’d finished (what I thought at the time was) a blistering attack on a book. The professor replied that, whatever faults the book had, the more important thing to do as a reader was to understand why the author made the choices they did– to reconstruct the strategy and structure of the project, as well as its execution. What I thought about the book, in other words, was less important than what the author was up to.

The first mistake– first in the sense of first, but also simplest and most elementary– critics make it is to attack an author for writing a book different from the one they would have written. The second is to focus on what a book doesn’t do, without asking 1) how successful is it in doing what it sets out to do? and 2) how reasonable is the author’s strategy? There are plenty of books that are badly-conceived and -executed; but you can learn an awful lot more about the practice of scholarship by thinking about how a book does or doesn’t succeed, and whether its aim is important or irrelevant.

Praise the author, criticize the book. This may sound like a distinction without a difference, but it isn’t. When I write a review, whenever I have something nice to say, I aim the praise at the author; whenever I have a criticism, I aim at the book.

Why do it this way? First, it’s a small world. Second, it serves to depersonalize the criticism, to direct it away from the person and to the text, in a way that I think is useful. I think of reviews as akin to conversations with an author; structuring the discussion this way serves to put you and the author more on the same side, and to turn the discussion from “what did you do wrong?” to “where could the book be better?”

The review isn’t just about the book. I frankly love reviews that are 3,000 words long, and spend 300 words talking about the book in question. Or rather, I love them when they’re done well: when they’re not just the critic talking about what they would have done, but talking about the bigger issues that the book raises. Book reviews offer the chance to play at the speculative end of the scholarly world, to try out concepts and theories before you’ve worked them up formally, or to think about bigger questions of craft and enterprise. Conventional scholarly articles don’t allow as much space for either speculation or reflection; at the same time, I find that having to keep things short forces you to think more clearly, and to make your argument clearer and more pointed than you would if you had 10,000 words to play with.

This is also another way of taking the book seriously. Any good book is going to raise broader questions, not narrow ones; and a review that uses the book as a jumping-off point for talking about bigger issues pays that book a compliment.