Writing a good book review?

In our newest "how can we help you?" thread, a reader asks:

On the topic of book reviews, I'm interested in any general advice regarding the process of writing them. Really interested in advice on any aspect -- e.g. do you read the book any differently, and if so, how? how do you choose which points to develop? what do the best book reviews do well, and what are some pitfalls or ways they could go poorly? These are just some starting points. Really any advice is helpful. This is my first time writing a book review. It's for a journal if that's relevant. Also curious how long people spend on writing them so I can get a sense of how much time to budget, though I know it'll vary a lot!

These are all excellent questions. What do you all think? For my part, I tend to think the best reviews discuss books chapter by chapter, giving the reader a fairly clear idea of what the major arguments are, and how successful the reviewer thinks they are. I tend not to like reviews that pick and choose which parts of the book to focus on (e.g. just to criticize those parts but ignoring other parts). I also think good reviews come across as 'fair minded', whatever exactly that means--though it certainly involves using the principle of charity. Here, I think Philosophy in Review's five kinds of reviewer guidelines are fairly apt:

Five Kinds of Reviewer

(adapted by Roger Shiner from Susan Swan, 'Nine ways of looking at a critic', Toronto Globe and Mail 30th November 1996. E23)

1.      The Spankers are out to administer discipline over anything from ill-conceived plot-lines to misplaced commas.

2.      The Young (and Old) Turk sees the review solely as an opportunity to demonstrate her or his own intellectual superiority and above-average intelligence.

3.      The Self-Abusers feel they could have written a better book on the subject, given half the chance, and describe it at great length.

4.      Gushers skip over discussion of the book; they just want to communicate the enjoyment of reading it.

5.      The Good Reviewer will represent the book (without lapsing into long-winded summaries) so the reader gets a sense of what the book is like whether the reviewer likes it or not. The good reviewer will also offer an interesting or revealing point of view from which the book can be perceived critically.

As for how long to spend on a review, I'm not sure: maybe a few weeks of sporadic work? You want to do a good job, obviously, but you shouldn't let writing a book review crowd out time for other things! But these are just my thoughts. What are yours?

Comments

[postduck]

I have found it useful to organize a small reading group around the book I'm reviewing so that I get a chance to discuss it with others. Then you can float potential lines of criticism within the reading group, and maybe get ideas from other people's comments.
I think it's really tempting to write a nasty, scathing review, and so it's important to check that tendency throughout the process. For example, f you find that your review skews more critical, try to ask yourself what the authors are doing *right*, and make sure to acknowledge that in a substantive way. And do a couple rounds of editing for snark at the end.

[Brad]

I am one of the editors for Metascience. In the last seven years I have commissioned, edited, and READ about 400 book reviews. A good review gives some sort of summary of the book, and provides some sort of assessment. The summary should be readable to non-specialists, and the assessment should be useful to the people who may be wanting to read or buy the book. It is fine to be critical. In fact, if the book has flaws they should be noted. But even a critical review should be respectfully written. It is nice to highlight unique features of a book, or special strengths. (for example, Chapter 7 provides an insightful analysis of the pessimistic induction). And it is also nice to really praise a book if it is a really good book.

[Bill D'Alessandro]

I'll disagree with Marcus and say that, as a reader, I don't often find chapter-by-chapter reviews to be the most interesting and enlightening. I think a useful review should instead try to answer a few big questions very clearly:

(1) Why does this book exist? (What problem is it addressing, and what sorts of things were people saying about the problem before?)

(2) What are the book's most distinctive, novel and important ideas?

(3) How well do these ideas work?

(4) Where does the debate stand, now that the book exists? What should happen next?

This is the sort of information that helps me feel I understand a book's significance and value. Given that the reviewer likely only has a couple thousand words to work with, I'd prefer they focus on doing all of the above really well -- even if it means we don't hear anything at all about Chapter 3, where the author explores a view they ultimately reject, or Chapter 8's tangential discussion of a mildly interesting side issue.

I think this approach is warranted because the typical review reader isn't a hyper-specialist on the topic of the book. (If they were, they'd just go read the book themselves and form their own opinion.) What a non-specialist wants to know is roughly "What interesting new thing is this book trying to do, and how much should I care about it?", so they can create a memorable little mental file for future reference. It seems to me that a dutiful discussion of the contents from start to finish isn't necessarily the best way to answer those questions.

[Jared]

The effect on the author matters too. A book review has a much bigger impact on an author's professional success than we might think, especially for those not already established. They're a lot easier -- and often more enjoyable -- to read than someone's actual work, and evaluators see them as a cheap and easy way to assess someone's reputation among insiders in the subfield. And that's often what hirers and especially tenure reviewers most care about; I know of at least two cases in which promotion committees scrutinized reviews of a candidate's book.

In pointing this out, of course, I'm not saying we should be dishonest and uncritical. But, for me at least, it counts as one factor to weigh in deciding how to put things.

[Review]

Perhaps there should be reviews for non-specialists and one for specialists. Chances are, if the book is relatively popular, then there’s going to be more than one review of it. Perhaps the Notre Dame website should include “In Depth Reviews” and “Brief Reviews” to correspond to different readers’ goals. I think brief reviews are good if you’re looking to teach a book while in-depth reviews are good for doing your own research on the subject.

[Michel]

I read book reviews so that I don't have to read the whole book. Consequently, I want reviews to give me an overall sense of the book and point to interesting elements that I might want to read closely. While I appreciate a rough sketch of the chapters, like Bill I prefer for the review to do more than that, and to point me in certain directions.

(Whether I've actually achieved that with my own reviews is another matter!)