ONE death is a tragedy. One million is a statistic.

You’ve probably heard this saying before. It is thought to capture an unfortunate truth about empathy: While a single crying child or injured puppy tugs at our heartstrings, large numbers of suffering people, as in epidemics, earthquakes and genocides, do not inspire a comparable reaction.

Studies have repeatedly confirmed this. It’s a troubling finding because, as recent research has demonstrated, many of us believe that if more lives are at stake, we will — and should — feel more empathy (i.e., vicariously share others’ experiences) and do more to help.

Not only does empathy seem to fail when it is needed most, but it also appears to play favorites. Recent studies have shown that our empathy is dampened or constrained when it comes to people of different races, nationalities or creeds. These results suggest that empathy is a limited resource, like a fossil fuel, which we cannot extend indefinitely or to everyone.

What, then, is the relationship between empathy and morality? Traditionally, empathy has been seen as a force for moral good, motivating virtuous deeds. Yet a growing chorus of critics, inspired by findings like those above, depict empathy as a source of moral failure. In the words of the psychologist Paul Bloom, empathy is a “parochial, narrow-minded” emotion — one that “will have to yield to reason if humanity is to survive.”

We disagree.

While we concede that the exercise of empathy is, in practice, often far too limited in scope, we dispute the idea that this shortcoming is inherent, a permanent flaw in the emotion itself. Inspired by a competing body of recent research, we believe that empathy is a choice that we make whether to extend ourselves to others. The “limits” to our empathy are merely apparent, and can change, sometimes drastically, depending on what we want to feel.

Two decades ago, the psychologist Daniel Batson and colleagues conducteda study that showed that if people expected their empathy to cost them significant money or time, they would avoid situations that they believed would trigger it. More recently, one of us, Daryl Cameron, along with the psychologist Keith Payne, conducted an experiment to see if similar motivational factors could explain why we seem more empathetic to single victims than to large numbers of them.

Participants in this study read about either one or eight child refugees from the Darfur region of Sudan. Half of the participants were led to expect that they would be asked to make a donation to the refugee or refugees, whereas the other half were not. When there was no financial cost involved in feeling empathy, people felt more empathy for the eight children than for the one child, reversing the usual bias. If insensitivity to mass suffering stemmed from an intrinsic limit to empathy, such financial factors shouldn’t have made a difference.

Likewise, in another recent study, the psychologists Karina Schumann, Jamil Zaki and Carol S. Dweck found that when people learned that empathy was a skill that could be improved — as opposed to a fixed personality trait — they engaged in more effort to experience empathy for racial groups other than their own. Empathy for people unlike us can be expanded, it seems, just by modifying our views about empathy.

Some kinds of people seem generally less likely to feel empathy for others — for instance, powerful people. An experiment conducted by one of us, Michael Inzlicht, along with the researchers Jeremy Hogeveen and Sukhvinder Obhi, found that even people temporarily assigned to high-power roles showed brain activity consistent with lower empathy.

But such experimental manipulations surely cannot change a person’s underlying empathic capacity; something else must be to blame. And other research suggests that the blame lies with a simple change in motivation: People with a higher sense of power exhibit less empathy because they have less incentive to interact with others.

Even those suffering from so-called empathy deficit disorders like psychopathy and narcissism appear to be capable of empathy when they want to feel it. Research conducted by one of us, William A. Cunningham, along with the psychologist Nathan Arbuckle, found that when dividing money between themselves and others, people with psychopathic tendencies were more charitable when they believed that the others were part of their in-group. Psychopaths and narcissists are able to feel empathy; it’s just that they don’t typically want to.

Arguments against empathy rely on an outdated view of emotion as a capricious beast that needs to yield to sober reason. Yes, there are many situations in which empathy appears to be limited in its scope, but this is not a deficiency in the emotion itself. In our view, empathy is only as limited as we choose it to be.