Double-effect theory recognises that a course of action might have a variety of ethical effects, some ‘good’ and some ‘bad’.

It can be seen as a way of balancing consequentialist and deontological approaches to ethics.

According to the theory, an action with both good and bad effects may be ethical as long as:

  • Only the good consequences are intended (we don’t want the bad effects to occur, they’re just inescapable, even if they can be foreseen).
  • The good done by the action outweighs the harm it inflicts.
  • The bad effect is not the means by which the good effect occurs (we can’t do evil to bring about good – the good and bad consequences have to occur simultaneously).
  • The act we are performing is not unethical for some other reason (for example, an attack on human dignity).
  • We seek to minimise, if possible, the unintended and inadvertent harm that we cause.

Double-effect is best explained through the classic thought experiment: the Trolley problem.

Imagine a runaway train carriage is hurtling down the tracks toward five railroad workers. The workers are wearing earmuffs and unable to hear the carriage approaching. You have no way of warning them. However, you do have access to a lever which will divert the train onto a side-track on which only one person is working. Should you pull the lever and kill the one man to save five lives?

Take a moment to think about what you would do and your reasons for doing it. Now, consider this alternative.

The train is still hurtling toward the five workers but this time there’s no lever. Instead, you’re a lightweight person standing on a bridge above the railroad. Next to you is a very large man who would be heavy enough to stop the train. You could push the man onto the tracks and stop the train, but it would kill the heavy man. Should you push him off the bridge?

Again, think about what you would do and why you would do it.

Did you say ‘yes’ in the first scenario and ‘no’ in the second? That’s the most common response, but why? After all, in each case you’re killing one person to save five. According to many consequentialists that would be the right thing to do. By refusing to push the man off the bridge, are we being inconsistent?

Double-effect theory provides a way of consistently explaining the difference in our responses.

In the first case, our intention is to save five lives. An unintended, foreseeable consequence of pulling the lever is the death of one worker. But because the stakes are sufficiently high, our intended act (pulling a lever to redirect a train) isn’t intrinsically wrong. The good consequences outweigh the bad. The negative outcomes are side-effects of our good action, and so, we are permitted to pull the lever.

In the second case, the death of the heavy man is not a side-effect. Rather, it is the means (pushing the man off the bridge to stop the train) by which we achieve our goal (saving the five men). The negative outcomes are not unavoidable side-effects that occur at the same time as the good deed. It is causally prior to and directly linked to the good outcome.

This fact has ethical significance because it changes the structure of the action.

Instead of ‘saving lives whilst unavoidably causing someone to die’, it is a case of ‘killing one person deliberately in order to save five’. In the lever scenario, we don’t need the one worker to die in order to save the five. In the latter, we need the heavy man to die. Which means when we push him, we are intentionally killing him.

Double-effect is used in a range of different contexts. In medical ethics it can be used to explain why it would be ethical for a pro-life pregnant woman to take life-saving medicine even if it would likely kill her unborn child (unintended side-effect). It also explains the actions of doctors who increase the dose of opiates to end pain – even though they know that the dosage will end the patient’s life.

In military ethics it explains how an air strike which causes some unavoidable ‘collateral damage’ (the death or injury of non-combatants) might still be permissible – assuming it meets the criteria described above and involves the proportionate and discriminate use of force.

Join the conversation

Can an action with both good and bad effects be ethical?