The Trolley Problem

Consider the following two scenarios:

Version 1 – A trolley is hurtling out of control along a track, bearing down on five people who have been tied to the track by a mad philosopher. You can avert the disaster by flipping a switch that diverts the trolley to another track. Unfortunately, there is one man tied to that other track. Is it morally permissible (or for that matter morally mandatory) to flip the switch?

Version 2 – A trolley is hurtling out of control along a track, bearing down on five people who have been tied to the track by a mad philosopher. You can stop the trolley by pushing a man in front of it. Is it morally permissible (or for that matter morally mandatory) to push the man in front of the trolley?

Most people would flip the switch in the first scenario, but wouldn’t push the man in the second version (according to a survey). Both scenarios have the exact same consequences, but for some (innate) reasons we prefer to kill the man by flipping a switch but not by pushing him.

Here’s slightly more convoluted versions of these hypotheticals:

Version 1 (a) – A trolley is hurtling out of control along a track, bearing down on five people who have been tied to the track by a mad philosopher. Across the track is a man sitting in front of his house, completely unaware of the trolley situation. If you say “hey, come here” to the man, he will walk towards you and in the path of that trolley, be killed and five lives will be saved as a result. You assess the situation and decide to call out. The man gets killed, and five people are saved.

Version 2 (a) – A trolley is hurtling out of control along a track, bearing down on five people who have been tied to the track by a mad philosopher. Across the track is a man walking towards you and in the path of that trolley, completely unaware of the trolley situation. If you shout “hey, watch out”, he will stop walking, be safe and five lives will be lost as a result. You assess the situation and decide not to call out. The man gets killed, and five people are saved.

Based on a survey (when people were asked to judge this behavior on a five point scale: from “morally impermissible” to “morally virtuous”), majority of the people believe that the harm caused by action (calling out) is more culpable than harm caused by inaction.

These moral biases and illusions (or contradicting moral instincts) don’t just exist in such philosophical and ethical experiments. For example, we seem to be much more comfortable with stealing music online (which involves hitting few keys on a computer keyboard) than stealing a music CD from a store. (Flipping a switch versus pushing a man.) Also, we are more forgiving when a terminally ill patient dies because of withheld medicine than when a doctor kills the patient by injecting a lethal dose of painkiller. (Action versus inaction.)

***

Talking about moral illusions, here’s one more (I promise, the last one!) moral scenario:

The Headache Problem – A billion people are experiencing fairly minor headaches, which will continue for another hour unless an innocent person is killed, in which case they will cease immediately. Is it okay to kill that innocent person?

Our moral instinct says that the answer is, quite obviously, ‘no’.

But here’s a counter-argument from an economist’s point of view: The answer is ‘yes’. We know that nobody wants to pay a dollar to avert a one-in-a-billion chance of death. (Surveys about our willingness to pay for auto safety devices have shown this.) Now, most people will happily pay a dollar to cure a headache. Which proves that most people think curing headache is more important than avoiding a one-in-a-billion chance of death.

This probably seems counter-intuitive, but think about how little we care about the risk of death when we drive. The risk of dying in a road accident in U. S. is 1 in 6,500. Every time you take your car out to drive to the grocery store, people (including you) become slightly more likely to die in a road accident. We know this. But we value our convenience more than the increased chance of death. We, consciously or otherwise, take that risk. As Steven E Landsburg argues in his book The Big Questions: Tackling the Problems of Philosophy with Ideas from Mathematics, Economics, and Physics:

[W]e all agree to kill random people all the time. We drive, install swimming pools, use drain cleaners, and drink tequila, knowing with certainty that some number of other people will die as a result. People have died so that other people can drive to the opera. Why shouldn’t they die to cure other people’s headaches?

Does that make sense, or do you think there’s a flaw in the argument?

***

The first two scenarios and The Headache Problem are taken from Steven E Landsburg’s book The Big Questions . And the other two versions of The Trolley Problem from Jonah Lehrer’s blog The Frontal Cortex.

2 responses to “The Trolley Problem”

1. I may or may not agree with the entire content, but I love reading such posts 🙂

• Vishal

Thanks for your testimonial, I appreciate it!