Mastermind by Maria Konnikova

I saw Maria Konnikova's Mastermind on the book lottery stacks at Science Online, and the subtitle "How to Think Like Sherlock Holmes" practically screamed "This is relevant to your interests!" Not only am I writing a book about how to think like a scientist, one of the chapters I have in mind uses mystery novels and the reading thereof as an example of scientific thinking.

I didn't score a copy of it at Science Online, but I did pick up the ebook shortly thereafter, and have been working through it during baby bedtimes for the last month or so, a process prolonged significantly by having to go and re-read a bunch of the Holmes canon for the first time in thirty-odd years (thank you, Project Gutenberg...). Which I was probably going to need to do anyway for the prospective book chapter, so it's all good. The basic idea of the book is to look at the mental tools and techniques Holmes uses in the stories, and ask how they might be applied in the real world. These discussions are further illustrated with references to cognitive science discoveries that mostly post-date the Holmes canon, and shed some light on what sort of cool things our brains are capable of, as well as what their limits are.

On the whole, this is an engagingly written book covering a lot of interesting territory. Unfortunately, it didn't work as well for me as I might've liked, which is why I'm reviewing it on a Saturday, when nobody will read this anyway...

There are two main problems that made this book miss the mark for me. One of them has to do with science writing, the other with Sherlock Holmes, though really, they're mostly about me and my reactions. Which may not make this review useful to anyone else-- another reason to bury it on a weekend-- but hey, it's my blog.

The science writing problem is, basically, that the Holmes stuff squeezes the science more than I would like. This kind of thing is basically inevitable in a short book with a pop-culture hook-- the space taken up by describing the pop-culture aspects reduces the space available for describing the science in detail. As a result, the presentation of the brain science material is kind of abbreviated, and lacks some of the detail you find in books with a more straight-up "Let me tell you about some science" presentation, like Daniel Kahneman's Thinking, Fast and Slow, or The Invisible Gorilla by Christopher Chabris and Daniel Simons (full disclosure: Chabris is a colleague of mine at Union, so I'm somewhat more predisposed to like his book...).

This problem is particularly acute given the subject matter. There are a lot of glib and superficial treatments of cognitive science out there, and a lot of stories about cool-sounding results that turn out not to be replicated in further studies. Things like the whole Jonah Lehrer mess make me extra skeptical about pop-cognitive-science, and the lack of space problem means there isn't room in this book for the additional caveats and description needed to allay that.

Please note that I am not in any way accusing Konnikova of the sort of misconduct that laid Lehrer low. As far as I can tell, from having read other books on the topic that go into more detail, the science she presents is pretty solid, and used in appropriate ways. But I need to draw on material from outside this book to make that judgement, which lessens the impact of this specific book for me. Someone with a less jaundiced view of the whole field would probably be more receptive to this approach, and really, those people are closer to the target audience than I am.

The bigger problem, though, has to do with the Holmes material itself, and became clear only because I interrupted the book to go read a big chunk of the Holmes canon, and boils down to this book treating the subject more seriously than I think the source material supports. Reading the original stories left me thinking that they're not terribly consistent or realistic. I didn't get the impression that Holmes's success was associated with any kind of realistic treatment of human mental capacities; rather, his brilliant successes are mostly due to the fact that he's a fictional character, whose author is on his side.

The best explanation of this problem, ironically, probably comes from another work of fiction, Terry Pratchett's Feet of Clay, where Commander Sam Vimes offers his thoughts on police work:

[Vimes] had a jaundiced view of Clues. He instinctively distrusted them. They got in the way.

And he distrusted the kind of person who'd take one look at another man and say in a lordly voice to his companion, "Ah, my dear sir, I can tell you nothing except that he is a left-handed stonemason who has spent some years in the merchant navy and has recently fell on hard times," and then unroll a lot of supercilious commentary about calluses and stance and the state of a man's boots, when exactly the same comments could apply to a man who was wearing his old clothes because he'd been doing a spot of home bricklaying for a new barbecue pit, and had been tattooed once when he was drunk and seventeen and in fact got seasick on a wet pavement. What arrogance! What an insult to the rich and chaotic variety of the human experience!

Konnikova spends a lot of time extolling the virtues of constant observation and carefully tended memory, which is fine as far as it goes-- I agree that the world would be a better place if more people thought more carefully about what they do all the time. Pegging that to Holmes, though, strains credibility a bit. It's all well and good to say that you should make sure to carefully store away facts that will prove useful to you later, but enough of the stories turn on Holmes knowing some improbable bit of utter trivia that it's not clear how you're supposed to decide what it is to store away in your "attic." And, as Vimes notes, it's all too possible to misinterpret careful observations-- the main reason why many of the conclusions she castigates Watson for jumping to are wrong is because Arthur Conan Doyle wanted them to be wrong. Holmes out-does Watson at every turn because the deck is stacked in his favor, not because they're realistic exemplars of human behavior.

This is, however, largely a matter of personal tastes. While I enjoyed my Holmes binge, they're not really the kind of mystery story I prefer-- my own tastes run more to the hard-boiled side of the genre-- and so I'm less inclined to cut them slack. And that, in turn, makes me less receptive to the central conceit of this book. Someone who's more a fan of the subgenre would probably feel differently about the whole business.

So, as I said, not a review that's likely to be of much general use, unless your tastes run close to mine. But it took a fair bit of thought to nail down exactly why I was dissatisfied with the book, and having put in the effort, I might as well get some use of it.

In an effort to end with something a bit more positive, I will note again that I definitely endorse the message of the book: thinking more like Holmes (or a more realistic version of what Holmes represents) is a goal worth striving for, and this book provides some useful tips on how to work toward that goal. The end result of this process would bring a lot of the benefits of thinking like a scientist that I've talked about, and the literary hook is another good approach to a worthy end.