You reach out to touch someone’s arm, or perhaps you pull a trigger. What made that happen? In this extraordinary survey of the science of human behaviour, the biologist Robert Sapolsky takes the reader on an epic journey backwards through time, and through different scientific disciplines. His governing question is: what explains the fact that humans can massacre one another but also perform spectacular acts of altruistic kindness? Is one side of our nature destined to win out over the other?

The backwards time-travel is an excellent organising principle. Seconds before our action, it is neuroscience that investigates what is going on in the brain; minutes to days before is the domain of endocrinology (hormonal fluctuations). Days to months before, we focus on the brain’s ability to learn and rewire itself. Sapolsky goes back through adolescence, childhood and gestation (including genetics), and, beyond the birth of the individual, to more distant causes still – those found in culture, evolutionary psychology, game theory and comparative zoology. He makes the book consistently entertaining, with an infectious excitement at the puzzles he explains, and wry dude-ish asides. (Humans, he notes, can “delay gratification for insanely long times” compared with other animals. “No warthog restricts calories to look good in a bathing suit next summer.”) He likes to call certain facts “boggling” when he is personally amazed by them; it’s charmingly infectious.

This book is a miraculous synthesis of scholarly domains, and at the same time laudably careful in its determination to point out at every step the limits of our knowledge. Sapolsky offers a vivid account of a standard view before lining up complications or objections to it from other research, particularly in brain science. (Testosterone, for example, does not cause aggression but amplifies pre-existing tendencies for or against it. The actions of such molecules in general “depend dramatically on context”). In a phrase that has unfortunately become associated with the dishonest attempts to smuggle creationism into American schools, he is adept at “teaching the controversy”, often providing anecdotes of scientists with battling views from decades ago. Throughout, he insists on how much individual variability there is hidden beneath the statistical averages of studies, and how the explanation of nearly every human phenomenon is going to be “multifactorial”: dependent on many causes. The literature on one scientific question, he notes comfortingly, is “majorly messy”.

Along the way there are many counterintuitive ideas and stern lessons. Empathy – feeling someone’s pain – is not as likely to lead to useful action as dispassionate sympathy, or “cold-blooded kindness”. Income inequality is concretely causally bad for the health of the poorer. There is a well-established link between rightwing authoritarianism and lower IQ. Genes are not destiny, and they are not “selfish” a la Dawkins; “we haven’t evolved to be ‘selfish’ or ‘altruistic’ or anything else – we’ve evolved to be particular ways in particular settings”. (According to one astonishing survey, 46% of women would save their own dog rather than a foreign tourist if both were menaced by a runaway bus. The evolutionary explanation is that they feel more “kinship” with the dog.) In general, if our worst behaviours are “the product of our biology”, so are our best ones. That Sapolsky’s heart is evidently in the right place makes it easy to discount certain hippyish outbursts such as that the invention of agriculture “was one of the all-time human blunders”, since it led to sedentary living and social hierarchy. Sure, but it also led to wine, science and books, which I’d suggest on balance makes it rather a good thing.

More thorny is the point at which he comes to address the question of individual choice and responsibility. For nearly 600 pages, barring the odd mention of the “cognitive” aspects of human action, Sapolsky sidelines the question of what place conscious reasoning has in determining behaviour, among all the neurochemical, hormonal, developmental and evolutionary factors he has been discussing. Indeed, sometimes he writes as though it has no place at all, as when he asks what sensory input “triggered the nervous system to produce that behaviour”. He eventually nails his colours to the mast of strict determinism: every human action is inescapably caused by preceding events in the world, including events in the brain. So there can be no such thing as free will. (It follows, of course, that social systems such as that of criminal justice must be completely overhauled, as philosophers such as Ted Honderich have long suggested.) You think you can freely choose to do one thing or another? Forget it, Sapolsky says.

It’s a common view, though by no means the overwhelming philosophical consensus. Notably, he prefers to cite mainly neuroscientists and legal scholars. Sapolsky ends the chapter with a display of his pleasingly undogmatic spirit, confessing that he finds it impossible actually to live his life as though he does not have free will. It’s perhaps worth noting, too, that one study he does not mention here (by Kathleen Vohs and Jonathan Schooler in 2008) implies that the idea we have free will, whether true or not, is a crucial placebo idea for a well-functioning society: in the experiment, subjects convinced they didn’t have free will were more likely to act unethically.

But Sapolsky’s insistence on the truth of strict determinism poses wider problems for the way he frames the rest of his book. One thing he refreshingly emphasises is that reason and emotion (“cognition and affect”) always interact, and that there are advantages to “combining reason with intuition”. This is a welcome counterbalance to the recent misanthropic strain of psychology that seeks to downgrade rationality altogether, but it is not clear that, on Sapolsky’s own view, conscious reasoning can accomplish anything at all if decisions are inexorably determined by the laws of nature. Which poses a challenge to his own humanistic optimism. We are not at the mercy of our amygdala’s fearful response to human faces of a different race, he argues; we can dampen and overcome such prejudice through reflection. Yet on his own view, we cannot freely decide to do so.

For the same reasons, it is unclear how much value there is in the author’s uplifting exhortations to think more carefully about our actions, and even to imbue politics with a new kind of science-based “peaceology”. Perhaps the idea is that such encouragement will be a new part of the causal chain affecting each individual’s behaviour, so compelling his readers to act more sociably. In which case I hope this book sells several billion copies.

It remains debatable, though, whether strict determinism is compatible with Sapolsky’s final message of hope for humanity, as he tells inspiring stories about moral heroism in history – the helicopter officer who stopped the My Lai massacre, the Christmas Day football match during the first world war. Sapolsky is on the side of Steven Pinker’s argument, in The Better Angels of Our Nature, that humanity is overall getting less violent and nasty, and points to some lessons from the “social plasticity” demonstrated in troops of baboons, one of Sapolsky’s own specialities. He thus sets himself against conservative pessimism about brutish human nature. “Anyone who says that our worst behaviours are inevitable knows too little about primates, including us.”

Yet the question remains: if human beings are simply reactive robots, slaves to natural law who are causally buffeted by a zillion factors of biology and circumstance, why would we have any say in whether things get better? Either they will or they won’t, but on this magisterial account it seems that we can’t really choose to do anything about it.

Smart people = book learnin` folk

Biology, the Criminal Justice System, and (Oh, Why Not?) Free Will

Some years back a foundation sent a letter to various people, soliciting Big Ideas for a funding initiative of theirs. The letter said something along the lines of "Send us a provocative idea, something you'd never propose to another foundation because they'd label you crazy."

That sounded fun. So I sent them a proposal titled "Should the Criminal Justice System Be Abolished?" I argued that the answer was yes, that neuroscience shows the system makes no sense and they should fund an initiative to accomplish that.

"Ha-ha," they said. "Well, we asked for it. That certainly caught our attention. That's a great idea to focus on interactions between neuroscience and the law. Let's do a conference."

So I went to a conference with some neuroscientists and some legal types--law professors, judges, and criminologists. We learned one another's terminology, for example seeing how we neuroscientists and the legal people use "possible", "probable", and "certainly" differently. We discovered that most of the neuroscientists, including me, knew nothing about the workings of the legal world, and that most of the legal folks had avoided science since being traumatized by nineth-grade biology. Despite the two-culture problem, all sorts of collaborations got started there, which eventually grew into a network of people studying "neurolaw".

Fun, stimulating, interdisciplinary hybrid vigor. And frustrating to me, because I kind of meant the title of the proposal that I had written. The current criminal justice system needs to be abolished and replaced with something that, while having some broad features in common with the current system (namely keeping dangerous people far away from everyone else-just to get this one out of the way early in the chapter), would have utterly different underpinnings. Which I'm going to try to convince you of. And that's just the first part of this chapter.

You can't be less controversial than stating that the criminal justice system needs reform and that this should involve more science and less pseudoscience in the courtroom. If nothing else, consider this: according to the Innocence Project, nearly 350 people, a mind-boggling 20 of them on death row, imprisoned an average of fourteen years, have been exonerated by DNA fingerprinting.

The system is not absurd because we humans live in packs,

Despite that, I'm going to mostly ignore criminal justice reform by science.

[...]

All of these are important issues, and I think reforms are needed at the intersection of progressive politics, civil liberties, and tough standards about new science. In other words, a standard liberal agenda. Most of the time I'm a cliched card-carrying liberal; I even know the theme song to many of NPR's programs. Nonetheless, this chapter won't take anything resembling a liberal approach to reforming criminal justice.

Some humans are broken, some are evil, and some are both.

...some, many, maybe even most of the people who were prosecuting epileptics in the fifteenth century were no different from us-sincere, cautious, and ethical, concerned about the serious problems threatening their society, hoping to bequeath their children to a safer world. Just operating with an unrecognizably different mindset. The psychological distance from them to us is vast, separated by the yawning chasm that was the discovery of "It's not her, it's her disease." Having crossed that divide, the distance we now need to go is far shorter-it merely consists of taking that same insight and being willing to see its valid extension in whatever directions science takes us.

The hope is that when it comes to dealing with humans whose behaviors are among our worst and most damaging, words like "evil" and "soul" will be as irrelevant as when considering a car with faulty brakes, that they will be as rarely spoken in a courtroom as in an auto repair shop. And crucially, the analogy holds in a key way, extending to instances of dangerous people without anything obviously wrong with their frontal cortex, genes, and so on. When a car is being dysfunctional and dangerous and we take it to a mechanic, this is not a dualistic situation where (a) if the mechanic discovers some broken widget causing the problem, we have a mechanistic explanation, but (b) if the mechanic can't find anything wrong, we're dealing with an evil car; sure, the mechanic can speculate on the source of the problem-maybe it's the blueprint from which the car was built, maybe it was the building process, maybe the environment contains some unknown pollutant that somehow impairs function, maybe someday we'll have sufficiently powerful techniques in the auto shop to spot some key molecule in the engine that is out of whack-but in the meantime we'll consider this car to be evil. Car free will also equals "internal forces we do not understand yet."