When it came to moral "reasoning," David Hume emphasized the quotation marks. We like to think our views on right and wrong are rational, he said, but ultimately they are grounded in emotion.
Philosophers have argued over this claim for a quarter of a millennium without resolution. Time's up! Now scientists armed with brain scanners are stepping in to settle the matter. So far it looks like Hume was onto something; though reason can shape moral judgment, emotion is often decisive, and that explains some strange quirks in our moralizing.
Harvard psychologist Joshua Greene does brain scans of people as they ponder the so-called trolley problem. Suppose a trolley is rolling down the track toward five people who will die unless you pull a lever that diverts it onto another track--where, unfortunately, lies one person who will die instead. An easy call, most people say: minimizing the loss of life--a "utilitarian" goal, as philosophers put it--is the right thing to do.
But suppose the only way to save the five people is to push someone else onto the track--a bystander whose body will bring the trolley to a halt before it hits the others. It's still a one-for-five swap, and you still initiate the action that dooms the one--but now you are more directly implicated; most people say it would be wrong to do this deal. Why? According to Greene's brain scans, the second scenario--the "up close and personal" intervention, he calls it--more thoroughly excites parts of the brain linked to emotion than does the lever-pulling scenario. Apparently the intuitive aversion to giving someone a lethal push is stronger than the aversion to a lethal lever pull.
Further studies suggest that in both cases the emotional aversion competes for control with more rational parts of the brain that take a utilitarian view, emphasizing the net savings of four lives. In the up-close-and-personal scenario the emotions are usually strong enough to win.
And when they lose, it is only after a tough wrestling match. The few people who approve of pushing an innocent man onto the tracks take longer to reach their decision. So too with people who approve of smothering a crying baby rather than catching the attention of enemy troops who would then kill the baby along with other innocents.
These cranial wrestling matches could be televised. As people ponder a moral dilemma, brain scans show changing activity levels in a part of the brain linked to abstract reasoning and cognitive control. In brains that take the utilitarian path, this part strengthens until dominant; in brains that refuse to kill one innocent to save many, it weakens until emotion has won.
Greene explains his findings in Darwinian terms. Back in the hunter-gatherer environment of human evolution, you killed people directly, not by the triple bank shot of pulling a lever that shifted a plate that rerouted a train. So an evolved aversion to the killing of an innocent might be especially sensitive to visions of direct physical assault. Imagining the triple bank shot impacts us less viscerally, causing a weaker aversion that is more easily outweighed by calculation.