Some Friendly Advice on Reviewing Books (or Anything Else Really)

a fun genre with a lot of pitfalls

This is the first post in (the first annual?) Lit Week at

Reviewing is fraught for the same reason podcasting is fraught: we all spend a significant amount of our daily lives forming opinions about things, just as we spend a lot of our lives in conversation, and so many assume that they’ll be good enough at these basic tasks to present them for public consumption. In both cases this assumption misunderstands why people turn to professionally-produced work when they could just get the amateur version anywhere for free: not for the opinions voiced themselves but for the quality of their expression, the unique expression of why rather than the dime-a-dozen reality of what.

I’ve reviewed a few books professionally, but not for anywhere particularly fancy. I’ve never had the title of book reviewer or anything, though that would be a thrill, and as is true in general I probably couldn’t get a review published in any big places anymore. So you are free to question my credibility on this score. If you’d like to peruse some reviews I’ve written, you can start with professionally published work here and here, academic book reviews here and here, and reviews from this space here and here. Whether you think I’m worth listening to on this topic I leave to you.

I can state two overarching rules that I believe should rule in reviewing:

  1. Everything exists to serve the opinion of the reviewer

  2. It is the expression of that opinion, not the opinion itself, that matters

You may very well say “duh.” But too many reviewers devote too much time to summary, which is not privileging the reviewer’s opinion, and too many reviewers are overly invested in recommending or not recommending, which emphasizes the summation of a reviewer’s opinion rather than its prosecution. And it is the prosecution that we are interested in. No ideas but in the expression of ideas. Probably a good philosophy above and beyond writing reviews.

The preoccupation with the reviewer’s position is not a statement of solipsism or self-obsession; in fact it’s not meant to serve the writer at all. It’s a rule meant to improve the experience for the reader, fundamentally advice about structure rather than about content. To stress the preeminence of the evaluative act rather than the summation of that act is an important antidote to several common misconceptions, the first being that the purpose of a review is to summarize and the second that the purpose of a review is to serve the reader, to guide them in their reading (or watching etc.) decisions. Wikipedia summarizes, probably better than you can. And while I’m not cynical enough to say that algorithms now do all the recommending for us, the fact is that no one is lacking for recommendations these days. They’re lacking for thoughtfulness and engagement that goes beyond good/not good. I love it when someone discovers a book thanks to something I’ve written about it, but that is a bonus. (I’m also quite sure that, while people do read books based on positive reviews, nobody ever declines to read a book they wanted to based on negative ones.) Conceiving of the role of the critic as someone who guides reading or watching decisions results in awful reviews. How did you feel about this book or movie or album or game, why did you feel that way, and how does your expression of those things deepen the experience and understanding of your reader?

Here’s some free advice, to take or leave.

Synopsis. In reviews of fiction, the rules for summarizing plot are blissfully simple: not too much. You want to summarize just enough for your audience to understand the points that you’re making in your review and no more. The summary serves the critical engagement; its purpose is to make the conversation possible. The classic mistake of inexperienced reviewers is to spend too much time laying out what happens in a story, which is boring, risks spoiling the experience for the reader, and takes up space better used for the purpose of a review, which is the thoughts, impressions, and critiques of the reviewer. If you’re writing about movies, TV, novels, etc., then the instinct to trim your synopsis down a little bit more will almost always be the right instinct.

With nonfiction it’s much more complicated. In a nonfiction review, we are certainly critiquing the book elements of a book - how well the book expresses the ideas and arguments it’s intended to express. (A book with lots of great information that’s poorly written can be a special agony.) But most readers want engagement with the book’s ideas, primarily its Big Idea, including what some of the consequences of that idea might be and how credibly it’s advanced in the text. The issue here is that sometimes a book’s idea is long and winding and can’t be rendered into a digestible form without a lot of summary. Fortunately, lots of books that are published these days are high-concept nonfiction where the title explains 90% of what you need to know. (Look out for Squirt: How Female Ejaculation Explains the World by Malcolm Gladwell this fall!) Unfortunately, most of these books are terrible because the world is complex and one-weird-trick thinking is a plague upon us. I digress.

I don’t believe, in general, that it ever makes sense for you to think of the synopsis as a discrete part of your review, found at the beginning, which has a defined endpoint after which no more summary appears. (True for fiction too.) Yes, you will frequently front an overall summary to make the immediate task comprehensible and to give your reader some scaffolding. But there’s no reason that summary can’t be sprinkled into the text as you go along, particularly if describing a plot development is the best way to illustrate a piece of praise or criticism1. The trick with book reviews is that the temptation is often to simply follow the chronology of the book, spooling out the writer’s argument step by step in more or less the author’s order. I don’t think this is always wrong to do, but I do think that it’s tricky. Scott Alexander (of Astral Codex Ten) uses something like this style at times, and I think he knows what he’s doing. But he also has the advantages of independent publishing, no space constraints, and an audience that is uniquely receptive to truly longform writing. I say that this style is tricky because, first, without discipline you will end up with a sprawling mess that your readers won’t be able to follow if they finish it at all. And it’s very easy to let your review become a book report as your “summarize then analyze” style falls into the gravity well of the book’s ideas to the point where the perspective of the reviewer gets lost - at which point we have violated our two commandments and undermined the reason for a review to exist.

I think the best you can do is to provide an an overarching summary of what you would see as the elevator pitch for the book, then articulate the details that are necessary for you to make the basic analytical steps that you want to make. You’ve got a certain degree of intellectual responsibility here: don’t introduce ideas if you don’t have the space or inclination to provide adequate justification for them; don’t express a claim without at least gesturing towards the defense of that claim. If your readers are left wondering how the book’s author could think that, when there is at least an attempt at such a rationale in the text, you’ve failed that author and your audience. It’s like anything else in conventional nonfiction writing: include that which serves a particular purpose of yours, broadly defined, in a way that demonstrates accountability to the audience and to the ideas, and otherwise trim. I am not a minimalist in any sense and many people think my stuff is too long but I believe most reviewers would do well to trim when trimming does not sacrifice meaning.

Voice. There is no point to a review that is not fundamentally voice-driven. The genre exists to access a particular book through the voice of the reviewer. This complaint might seem obscure, but you’d be surprised. In the world of GoodReads and LetterBoxd, it’s very easy to find the phenomenon I’m talking about, reviews that mechanistically summarize the plot, then provide a list of specific attributes (the plot, the characters, the style, the humor, the acting, the costumes) and the reviewer’s reaction to each - “good story, annoying characters, humor doesn’t land”). Yes, I want to know if you found the writing style frustrating, but I want to know that through your expression of the particular interplay between you as a reader and the text as a text. That is, if you write “I liked the plot,” the important word in that sentence is “I’ and not “the plot.” The plot will go on existing comfortably independent of your opinion. But without your particular you-ness there is no review. So the question is not “is the plot good,” the question is “what about the plot interacted meaningfully with your unique mindset to produce a given effect, and how can that information meaningfully inform the reader?”

I also think a lot of people adopt a style that’s artificially serious, an officious quality, when writing reviews. Adopting an ostentatiously formal or high-minded style is a very common tic of inexperienced writers in general, but it’s particularly prevalent in review writing. I suspect this stems from insecurity about judging someone else’s creative work; the formality becomes a kind of hedge against “you think you could do any better?2” and a way to distance oneself from the frequently uncomfortable position of being a judge. But we’re all judges of what we read and watch and listen to. The written review only formalizes things, and in fact typically produces more charitable and humble attitudes than those we merely think and feel. The alternative, to be clear, is not an affected jocularity or in-your-face profane style, which can be even worse. The point is to reduce affectation in general, to avoid the tendency to step outside of yourself in your writing.

Again, the only unique thing that you can bring to a review is you, so be you. I want your voice, not some strained attempt at the voice of, like, the NYT editorial board. Developing a personal style and voice is the most crucial task of growing as a writer and the one that you will never cease taking part in, no matter how acclaimed you become or how old you get. (Distinguishing between style and voice intelligently would take me an essay in and of itself, which I’ll probably write someday.) There are many books devoted to refining prose style, most of them actively harmful to that effort. So I won’t do a half-assed lesson here about how to become a better stylist. What I will say is that the review is a profoundly personal genre, one with perhaps the least interest in generalizability of any common writing task, so make your reviews personal in turn. I want to hear not just how you feel about a book but what particular aspects of yourself inspire those feelings. If you have a biographical attachment to a writer or genre, tell me about it. If you have a complicated mental relationship to the writer, say so. If you know your feelings are influenced by something extratextual, and that many people would see that as illegitimate, front it, don’t hide it. If nothing else, these things help make your writing uniquely your own, and since we can’t just decide to be good at writing, we can take comfort in writing to become more fully ourselves, which we do get to decide.

I don’t know why some writers have decided that the View From Nowhere is the proper vantage point when reviewing something, though I at least understand that it is a common convention in reviews written for scholarly journals. (This is one reason why most reviews in scholarly journals are dreadful.) I find this tendency bad enough when done by inexperienced or amateur writers, but it’s profoundly aggravating to me when it’s published in professional outlets. Perhaps it’s just the sense of self-importance that comes from writing with the voice of God, but some highly respected publications put out reviews where the critic is doing their best to convince you that they speak not from a contingent and situated subjectivity but rather from the armchair of the heavens, doling out observations from a place of pure knowledge. Personally I hate those reviews, but then I suppose I don’t begrudge professional reviewers for indulging in intangible privileges; certainly they aren’t in it for the money. I just never enjoy such reviews and my counsel is that you write from your own chair and not from anyone else’s.

One man’s opinion: no brains in a jar, please. You’re an individual human. Write reviews in that spirit.

Passage analysis. Mining a specific passage for meaning and insight is a niche thing in review writing. In print publications, obviously, there isn’t space. You won’t be able to get one past many editors at other places as well. But if you’re given a free hand then I think the practice can be fruitful for both you and your readers3. In a nonfiction book you can potentially just extract the facts and arguments that you need, although the specific construction of those elements is no less important in nonfiction than in fiction. In fiction in particular I really do need something of the actual substance in the review myself, even if it’s just a few sentences. Many people disagree, at least judging by most longform book reviews. But if you talk about incandescent prose or turgid prose or uneven prose, I want to see a little of it for myself, and then I want you to back up the adjectives you’ve just used. It’s something you can do in book reviewing that you can’t do in written movie or TV or music reviews, so why not take advantage?

Judgements. You shouldn’t necessarily feel that you need to have one summative judgment at the end of your review; sometimes a good review simply reports to you the individual impressions along the way and trusts you to derive the reviewer’s overall thoughts from the aggregate. Whether you say good or bad or in between, you should make sure that what you’re saying is

  • Personal - I’ve already explained why. I neither know nor care what books are “really” good. I know what books I think are good. Make me understand which ones you think are good and why.

  • Justified - you make reference to specific pieces of evidence to tell the reader why you feel the way you feel. Your summative judgment should feel like the aggregate or average of the specific claims you’ve already made. That’s not to say that your review can’t take a turn; I often enjoy reviews which seem at first to be negative/positive but then have a reveal that shows the critic’s overall thinking is different. But when done well this still leaves me feeling like that plot twist in Act III was adequately set up in Acts I and II. I frequently enjoy essayistic writing that’s a little mysterious or that works through evocation, folk magic. But I want to be able to understand why a review concludes what it concludes.

  • Organic - as many have said in the past, including the GOAT, negative criticism is easier and more fun to write. Brutal takedowns are also good for traffic, and there are times when they are justified, even morally necessary. But it’s also very obvious when someone is reaching deeper into their thesaurus for words that signal contempt than they really feel motivated to. Don’t let your judgments be dictated by inauthentic motives, and don’t be guided by your sense about vague rules of reviewing. Should you highlight both good and bad in a review? Only if both the good and the bad genuinely assert themselves into your opinion. Evenhanded reviews that fall between rave and rant are always welcome, and in a binary culture they can be a balm. But it would be a sin to force a balanced review if your feelings are not balanced. For me, insincerity in a critic is painfully obvious, and a failing that almost no reviews can survive.

  • Assertive - we often have conflicted or mixed or unsettled or evolving views on a book. That’s good! That’s part of the point of books and of being an adult, of wrestling with the permanent ambiguity at the heart of human life. And if you feel those complicated and impermanent feelings about a book you’re reviewing, make the opinions in your review complicated and impermanent in turn - but make the statement of those complex feelings assertive. Express your uncertainty with certainty. Own all of your feelings, here if nowhere else. So much of what we write is tentative and unsure, by necessity, in a world that routinely denies us confidence. But when writing a review, you rule. This is your place of power, your respite from the constraints of self-doubt. Tell me what you think, unapologetically, even if what you think is profoundly ambivalent, or if you are sure that your opinion is complicated by your lack of understanding. Here, you are master. I know your review is just your opinion, and you don’t have to tell me so. There is no alternative. So state that opinion with force. If you can’t feel the confidence to share your own opinion confidently, why are you writing at all?

Grades/Stars/1-10. My impression is that many people chafe at these, and I get it. They’re immensely reductive and suggest that they function on some sort of consistent scale which we can use to ordinally rank different things, which we can’t. And Metacritic and Rotten Tomatoes are indeed a plague on critical culture. That said, do you think it would be fun to give these kinds of scores in your reviews? Do you think your readers will enjoy it? If so, then go for it. If it’s low status, be low status. Who gives a shit? Have fun. If you aren’t having fun, why bother?


In my review of Weather, I mention a late-arriving character in the act of (gently) criticizing this development as somewhat artificial. This is what I mean about synopsis serving analysis; this plot detail, and any others, are useful to me as a reviewer only to the degree that they support a specific analytical move I want to make.


For the record, the “well who are you to judge” response to artistic criticism is stupid and fundamentally misunderstands the purpose of reviews. The critical impulse doesn’t stem from an assumption that the critic could do better. The critical impulse stems from the belief that the artist could have done better, that a superior work of art lay within their grasp, and they didn’t produce it. It’s entirely immaterial if the critic could do better than the artist. All that it takes to think critically is to imagine a better work of art than what we got.


I also think it’s nice to do if you figured that close textual analysis was what English professors do and were disappointed to learn that they write about the phallus or whatever the fuck instead.