The Doctrine of Double Effect: Do Intentions Matter to Ethics?

Author: Gabriel Andrade
Category: Ethics
Word count: 1000

When is it morally permissible to harm someone?

One answer to this question is provided by the doctrine of double effect (or the “DDE” for short). The DDE claims that it is permissible to harm someone if it will also result in good consequences that outweigh the harm and the harm is just a “side-effect” of your action, rather than something you intend to make happen.[1]

This essay introduces the DDE.

1. Trolley Problems

To better understand the DDE, imagine this: a streetcar is going to run over five people; the brakes have failed, but you can pull a lever to divert the streetcar onto another track in which only one man will be run over.

Would you pull the lever? Most people (about 90%) say that it’s permissible to do so.[2]      

Trolley problems: should he pull the switch?
Trolley problems: should he pull the switch?

Now, imagine a different scenario: a streetcar is going to run over five people; someone is standing on a footbridge over the tracks, and you can push the person off the bridge so their body will stop the streetcar, thus saving the five, but killing the man who was pushed.

Would you push that man? Most people (about 70%) say it’s not permissible to do so.[3] 

Trolley problems: should he push the man?
Trolley problems: should he push the man?

But why would it be permissible to pull the lever but impermissible to push the man off the bridge?

Here’s one answer: in the first scenario, you foresee the death of the man, but you don’t intend it. You don’t intend it because his death isn’t part of your “plan” to save the five. You don’t need him to die: if he could (somehow) escape when the streetcar is diverted, your plan would still succeed.

In the second scenario, though, his death isn’t just a foreseen side-effect of your action. You need his death to save the other five. If he were pushed, but somehow bounced back to the bridge, then your “plan” to save the five would fail.

The DDE gives an explanation of the difference here: the DDE claims it’s permissible to harm someone if doing so will result in good consequences that outweigh the harm, and the harm is just a foreseen “side-effect” of your action, not something intended. So, in the first scenario, you foresee but do not intend the harm, whereas in the second scenario, the harm was intended, as part of your plan. And since the good consequences of saving five outweighs the harm of killing one, pulling the lever is permissible, but pushing the man is wrong.[4]

2. Applications

Let’s consider how the DDE applies to some real-life situations.

2.1. Military Ethics

Is it permissible to kill civilians in war? Suppose you drop a bomb on a terrorist training camp, killing one hundred terrorists, but, as collateral damage, killing ten innocent civilians.

According to DDE, that’s permissible. You may even foresee that those civilians will die. What matters is that you’re not intending their death: if the bomb kills the terrorists without killing the civilians, your plan would still be a success.[5]

What about killing civilians to force an enemy to surrender (as in Hiroshima)?[6] Is that permissible?

Not according to the DDE. Even if you save more lives with this action (say, by bringing a quicker end to the war), you need them to die to force your enemy to surrender. If the bomb dropped on civilians somehow fails to kill them (and so the enemy does not surrender), your plan would fail.

2.2. Medical Ethics

Suppose, for the sake of argument, that abortion is wrong because it harms the fetus.[7] Is it still permissible to remove a cancerous womb from a pregnant woman to save her life, even if you know the fetus will die as a result?

According to DDE, yes. You’re not targeting the fetus; if you could remove the womb while saving the fetus, you wouldn’t object. The fetus’ death is an unintended side-effect of your action.

Now, suppose we’re considering killing the fetus because the pregnant woman is diabetic, and her pregnancy increases her risk of dying during delivery. Here you need the fetus to die to reduce her risk of dying: the fetus’ death wouldn’t just be a side-effect. If the fetus survives inside the woman’s body, that would defeat the purpose of reducing the risk of the mother’s death during delivery.

About euthanasia, if a terminally-ill patient is in intense pain, is it ethical for a physician to administer a morphine shot, even if that action would likely lead to the patient’s death?[8] According to the DDE, yes. The physician is seeking to alleviate pain; the patient’s death is foreseen, but not intended.

In contrast, if the patient asks the physician to assist her in suicide, and the physician complies, the physician isn’t merely attempting to alleviate pain, but rather, to intentionally kill the patient. This is unethical, according to the DDE.

3. An Objection

The DDE’s verdicts about what is and isn’t permissible match many people’s intuitions. But it’s still open to objections.

A common objection to the DDE appeals to extreme cases. Suppose a terrorist knows the location of a ticking nuclear time bomb that would kill millions of people, but refuses to disclose the information. Interrogators know that the only way to make him cooperate is by torturing his innocent underage son. Torturing the child is a necessary step in this plan to prevent a nuclear disaster: you’d be intentionally harming him. So, according to the DDE, torturing the terrorist’s son is wrong.

But it’s counterintuitive to say that we must let millions of people (including many innocent children) die to avoid torturing one innocent child. So, it seems permissible to bypass the DDE to prevent disasters.

But what’s a “disaster”?[9] If 10 people will die, does that count as a disaster? If not, what’s the threshold? 100 lives? 1,000 lives? 1 million? The answer is unclear.[10]

4. Conclusion

To many, the DDE is intuitively appealing. However, some cases suggest that a strict application of the principle may sometimes be problematic.


[1]The doctrine of double effect can be traced to Medieval philosophy and discussions about the morality of war and self-defense. St Augustine believed that one is never ethically authorized to kill in self-defense, because that would be a form of egoism. However, other scholars disagreed. Thomas Aquinas argued that while self-defense may bring about the bad effect of killing an aggressor, it brings about the good effect of preventing the concerned person’s life.

For an overview of the history of the DDE, as well as alternative formulations of the DDE, see McIntyre (2018).

[2] See Greene (2014).

[3] See Edmonds (2014).

[4] For an application of the ethics of Immanel Kant (1724-1804) to trolly problems, see Deontology: Kantian Ethics by Andrew Chapman. Also see Consequentialism by Shane Gronholz.

[5] See Fotion, N., & Elfstrom (2020).

[6] In 1945, the United States dropped atomic bombs on the Japanese cities of Hiroshima and Nagasaki, killing around 200,000 civilians. Soon afterward, the Japanese surrendered and World War II came to an end. There is a historical debate about whether the bombing truly prompted the Japanese surrender. But a further ethical debate considers whether, even if the bombings did prompt the Japanese to surrender, targeting civilians was morally permissible.

[7] For an introduction to this topic, see The Ethics of Abortion by Nathan Nobis. For a discussion of the assumption that at least beginning, pre-conscious fetuses are harmed by abortion, see Ekendahl, K., & Johansson, J. (2022).

[8] For an introduction to these issues, see Euthanasia, or Mercy Killing by Nathan Nobis.

[9] See Walzer (1973).

[10] For a discussion of the potential relevance of the numbers of people or beings affected by our actions to our moral obligations, see Saving the Many or the Few: The Moral Relevance of Numbers by Theron Pummer.


Edmonds, D. (2014). Would You Kill the Fat Man? The Trolley Problem and What Your Answer Tells Us About Right and Wrong. New York: Princeton University Press.

Ekendahl, K., & Johansson, J. (2022). Does Abortion Harm the Fetus? Utilitas, 34(2), 154-166. 

Fotion, N., & Elfstrom, G. (2020). Military Ethics: Guidelines For Peace and War. New York: Routledge.

Greene, J. (2014). Moral Tribes: Emotion, Reason, and the Gap Between Us And Them. London: Penguin.

McIntyre, Alison. (2018). “Doctrine of Double Effect”, The Stanford Encyclopedia of Philosophy (Spring 2019 Edition), Edward N. Zalta (ed.).

Walzer, M. (1973). Political Action: The Problem of Dirty Hands. Philosophy & Public Affairs, 2(2), 160-180.

Related Essays

Deontology: Kantian Ethics by Andrew Chapman

Consequentialism by Shane Gronholz

The Ethics of Abortion by Nathan Nobis

Euthanasia, or Mercy Killing by Nathan Nobis

Saving the Many or the Few: The Moral Relevance of Numbers by Theron Pummer

Principlism in Biomedical Ethics: Respect for Autonomy, Non-Maleficence, Beneficence, and Justice by G. M. Trujillo, Jr.

PDF Download 

Download this essay in PDF

About the Author

Dr. Gabriel Andrade is an assistant professor at Ajman University, in the United Arab Emirates. He received his PhD in Human Sciences from the University of Zulia, Venezuela. He specializes in psychology, philosophy and the social sciences.

Follow 1000-Word Philosophy on Facebook and Twitter and subscribe to receive email notifications of new essays at