Skip to main content

What’s So Bad About Harm?

New research into evaluative simulation shows how we instinctively dislike certain actions deemed to be harmful regardless of their outcomes.
(Photo: Yeko Photo Studio/Shutterstock)

(Photo: Yeko Photo Studio/Shutterstock)

Among the most well known parts of the Hippocratic Oath—the paragraph recited by medical students when they finish medical school—is the declaration to “never do harm.” It’s an axiom most people—regardless of profession—attempt to follow. In fact, our social edict against causing harm is so powerful that, in addition to forming the foundation of modern medicine, it has been used by a GOP congressman and a leading conservative economist as justification for restricting access to medical care.

What pushes us to condemn harmful acts? One factor is emotion, and the most obvious source of emotion is an act’s outcome. When there is a stabbing, for example, it’s the outcome (a wounded human) that we believe most causes us to feel bad. This feeling, or “outcome aversion,” leads us to judge the act as morally wrong. And the worse an outcome makes us feel, the worse we judge the act that produced the outcome.

But a series of recent studies suggest that an act’s outcome is not the only thing that triggers an emotional response. The research, which comes out of Brown University's Moral Psychology Research Lab, shows that certain physical actions can induce a negative emotional response that’s independent of the outcome the actions produce. This “action aversion” can occur whether the action is performed, witnessed, or imagined, and it can affect how an act is judged.

The reasoning is that we have been conditioned to associate harm with certain actions, such as pushing, hitting, or stabbing. When we think about these actions we instinctively imagine ourselves performing them, a process the researchers call “evaluative simulation.” Because we associate these actions with bad things, evaluative simulation produces a negative emotional reaction regardless of whether harm occurs.

One might expect that moral opposition to each killing method would be driven strictly by the level of suffering it was expected to cause. But that’s not what research found.

The implication is that the emotional component of our moral judgment is not strictly based on a harmful act’s outcome—it can also arise from the physical actions involved in the act. In other words, negative emotion, as well as the moral condemnation it triggers, not only arises from the presence of a stabbing victim, it also arises from exposure to a stabbing action.

In a 2012 study (PDF) led by Fiery Cushman, participants who simulated harmful actions that didn’t produce harm—firing a toy gun or smashing a fake hand with a rock, for example—produced a negative physiological response associated with stress. In a recent experiment led by Kyle Dillon, participants who imagined performing a harmful act in a conventional way—hitting a person’s foot with a hammer, for example—exhibited a greater increase in blood pressure than participants who imagined generating the same outcome in an unconventional manner, such as using a pulley system to drop a hammer on a person’s foot. The results suggest that the mere presence of certain actions can lead to more negative affect, and that can occur even if there is no bad outcome, and even if no negative affect arises when the same outcome is produced with a different action.

In a new study forthcoming in Emotion, Ryan Miller links “action aversion” to moral judgments in five experiments. In one, three groups of participants were told about 23 different ways a terminally ill man could be killed in a hypothetical, consensual “mercy killing.” One group rated how much suffering they thought each method (e.g. suffocation, gunshot, etc.) would cause—this provided a measure of outcome aversion for each method. The second group rated how much it would upset them to act out each method in a play (where no harm would occur)—this provided a measure of action aversion for each method. A third group rated how morally wrong it would be to use each method.

One might expect that moral opposition to each killing method would be driven strictly by the level of suffering it was expected to cause (i.e. outcome aversion). But that’s not what Miller found. Opposition was associated with expected suffering and aversion to performing the action, and the two measures were statistically distinct. It seems we not only condemn harmful acts because of what they produce, we also condemn them based on how the physical movements they entail make us feel. In fact, the other four experiments in the study provide evidence that action aversion is a more reliable predictor of moral judgments than outcome aversion.

The connection between physical actions and moral judgment can help explain some of our moral tendencies. Studies have found that harmful acts are seen as less moral if they involve action rather than omission and contact rather than no contact, for example. Pushing one person in front of a train in order to save five people is judged as worse than pulling a lever that produces the same outcome, which is in turn judged as worse than allowing the same outcome to occur by failing to pull a lever. The added negative emotion that can arise from stereotypically harmful physical movements provides one explanation for these findings.

Action version is also relevant to situations in which people object to behavior that seemingly causes no harm. The Brown researchers use experiments with contrived actions, but there are more natural actions—in situations involving perceived sexual immorality, for example—that lack a conventional victim but are nonetheless condemned. Miller describes evaluative simulation as a process “whereby we evaluate the moral status of another’s action by simulating the affective response that we would experience performing the action ourselves.” If that’s the case, it becomes easier to understand why a social conservative might so vehemently object to the friendly gay couple co-habitating in apartment 4F.

This research is another reminder that our moral judgments tend to be highly subjective. In addition to being swayed by our unique emotional responses to certain physical movements, our judgments can also be influenced by our own past behavior, social status, and even music. Hippocrates was wise in declaring harm morally reprehensible, but the more difficult question is what makes a doctor see something as harmful.