Behave

01Introduction

A biologist who could not finish this book enjoyed one chapter: the chapter whose ideas he already agreed with.

02Core Idea

Behave was published in two thousand seventeen. Its author is a professor of biology and of neurology at Stanford, the neuroendocrinologist Robert Sapolsky.

The question the book takes on is narrow to state and enormous to answer: what caused a person to do the thing they just did.

The first half runs on a clock in reverse. It opens with what Sapolsky calls One Second Before, meaning the neurobiology immediately preceding an act. Then Seconds to Minutes Before, which is the cues in the room. Then Hours to Days Before, which is hormones. Then days to months, and outward from there toward genes, culture and evolution.

Notice what that is. It is a filing system for causes. A filing system can be elegant and still prove nothing on its own.

The reviewer who admires the book most in my file, a blogger named Miles Raymer, identifies the central point as the importance of context. Raymer then says something a sales page would cut. He writes that this is, in his words, "hardly a novel position."

That is the honest frame. The thesis is old. What is unusual is the volume of material stacked behind it. Raymer calls the result "a monumental achievement of syncretic thought" — his judgment, on a book he finished, and worth weighing against the readers who did not.

There is a destination. Sapolsky is a hard determinist: he rejects free will. The argument lands in chapter sixteen, on biology, the criminal justice system and free will.

The biologist Jerry Coyne, who runs the blog Why Evolution is True, reads the long opening on hormones and the nervous system as groundwork laid for that ending. That is Coyne's interpretation of the strategy, not a statement from Sapolsky. Either way, it changes how you read the first several hundred pages. They are not a tour. They are a ramp.

03The Evidence

The firmest numbers in my file come from the book itself, quoted by Raymer. Sapolsky writes that the human genome codes for about one thousand five hundred transcription factors, the proteins that switch genes on and off. It contains about four million binding sites for them, and the average cell uses roughly two hundred thousand of those sites.

His reaction to his own figure, in three words: "This is boggling."

Be precise about what that evidence is. It is a parts count, drawn from molecular biology, reported rather than discovered here. It supports the claim that genes operate through environment. It does not identify the cause of any single human act, and the book does not say it does.

A summary and criticism site reports a second claim: the person most likely to help is not the one feeling the most empathy. Too much of it can crush someone into doing nothing. What predicts action is the ability to regulate the feeling and keep some distance from it. Adolescents, on this account, feel more and act less, and in adolescence no brain region comes online to blunt the pain.

That is predictor language. Correlational, laboratory-scale, and in my file given with no sample sizes and no study designs attached. Illustrative of a mechanism. Not a demonstration that detachment produces help.

The same site reports the context findings on faces. Voters lean toward candidates who say "we" and "us". In war and unrest, people prefer older and more masculine faces; in peacetime, younger and more feminine ones; where cooperation is the frame, intelligent-looking faces.

This is the thinnest layer in the stack: people rating photographs under a prompt. Stated preference, not behaviour, and no measurement method given in what I have. The summary adds its own caution, which I will pass on — preferring an aggressive leader in wartime does not make it a good choice.

One claim I can report but not grade. A reader commenting on Coyne's review describes the testosterone chapter: testosterone is not a predictor of aggression, and under some circumstances it can push males toward prosocial behaviour. That is a reader's paraphrase of a chapter, two steps from the data. I include it because of its direction. The book is spending credibility against the popular story there, not with it.

04The Critiques

The loudest complaint in my file is not that the book is wrong. It is that people stop. Coyne says he could not finish it, found the opening boring, and describes it as "a tedious disquisition on hormones, neurons", too textbook-ish for him.

His readers echo it. One says he read less than one third. Another bought it in August and was still picking at it. These are anecdotes from a self-selected audience, not a completion statistic. They do all point the same way.

Then the line from the hook. Coyne enjoyed chapter sixteen, and he volunteers that part of the reason was that its ideas were congenial to him. Credit to him for saying it. Note what follows: the strongest endorsement of the book's conclusion in my file comes from someone who held that conclusion first. Agreement is not replication.

Raymer reproduces a passage where the author addresses the reader directly. In it, Sapolsky says that on any big, important issue, it can seem that fifty-one percent of studies conclude one thing and forty-nine percent conclude the opposite.

That is candour, and it is the hardest sentence in the book to reconcile with its own ending. A near-even split argues for suspended judgment. Chapter sixteen does not suspend judgment.

Sapolsky is sharpest when the method under the lamp belongs to someone else. A researcher named Chagnon tied killing to sexual success among the Yanomamö. By the summary's account, Sapolsky points out that Chagnon did not control for age when comparing men who had killed with men who had not, and left out the most violent men entirely — the ones who did not survive.

That is a genuine survivorship problem, correctly identified. It is also aimed at a finding the author plainly dislikes.

One for the comments: when a book dismantles another researcher's survivorship bias and then leans on people rating photographs, do you read that as rigour, or as rigour pointed in one direction?

05Who It's For

Raymer estimates he had to read between ten and twenty other books to reach what Sapolsky compresses into a few hundred pages. That estimate describes the audience precisely: someone who wants one dense container rather than a shelf, and who treats cross-disciplinary detail as the payoff rather than the toll.

The book flatters that reader openly. In the closing passage Raymer quotes, Sapolsky tells the reader they have, quote, "amply proven you have intellectual tenacity."

The reader it fails is the one who needs the argument to arrive early — because every stall in my file happened in the same place, in the hormone and neuron chapters, with the conclusion still hundreds of pages away.

And anyone hoping to see free will defended, or even seriously contested, is in the wrong room. The hard determinist position is the author's, not a debate he stages.

06Offshoots

I have no offshoots file for this book. No sales figure, no print run, no prize, no adaptation. So none gets spoken here. A missing source is a reason to say less, not a reason to estimate.

What my file does show is a recorded conversation between Sapolsky and Sam Harris on the Waking Up podcast — two hard determinists, roughly thirty-five minutes on free will, with nobody in the room taking the other side.

And one small marker of how far the argument travels. Coyne, after reading chapter sixteen, says that from now on, when someone wants to argue the point with him, his answer is: "Go take it up with Sapolsky!"

07Verdict

Scored on what I could check: dense and specific on mechanism, sharp on other people's method, opaque about the studies under its own softer claims, and long before it pays.

Strong in the catalogue, weaker in the courtroom. When Sapolsky counts things — binding sites, or the holes in a field dataset — the ground is firm. When the claims turn to empathy, hormones and faces, my sources pass them along without the studies, so they sit as illustrative. And when the book reaches free will and criminal justice, it has crossed from measurement into argument, carrying a reader who may simply be tired by then.

A book that tells you the literature splits almost evenly, and then ends on a settled verdict, is asking for a trust it has spent the previous chapters teaching you to withhold. That tension is the review. I am not going to resolve it for you.

That is the file on Behave. If scrutiny done this way is useful to you, a like registers it, and a subscribe brings the next book under the same lamp.