Say one was given a trolley problem such that one outcome was certain to kill 10 people, and the other had a 1% chance of killing 100 people depending on the outcome of some quantum event, similar to Schrodinger's cat box. Under the Copenhagen interpretation of quantum mechanics the expected mortality of the first is 10 people and the expected mortality of the second is one person, therefore it could be moral to choose the second. Under the many worlds interpretation of quantum mechanics the first is sure to kill 10 people in both universes after the decoherence, the second is sure to kill 100 people in one universe and zero in the other universe after the decoherence, therefore it could be moral to choose the first.

Is this a problem that has been examined and/or solved by the philosophy community? I read this question, but the answers seemed to focus on the fact that there is always one world that something bad happened, not that we get to choose between different bad things happening.

  • 3
    Schrodinger's cat box. Now that's funny.
    – user4894
    May 25 at 15:08
  • 2
    Since you're obviously trying to minimize the mean number of deaths, the question is really whether probability alters moral decisions, as it surely must viz. risk management (where for some problems we might decide mean-minimization is the wrong criterion, but that's another issue). @kutschkem's answer points out you may have miscalculated the details, but that's separate from the philosophical aspect.
    – J.G.
    May 25 at 19:53

7 Answers 7


Your question relies too much on there being only two worlds - one where the 1% event happened, and one where it didn't. I think the more reasonable many-worlds interpretation is that out of the infinite existing worlds (that share a past up to that point), in 1% of them the event happened, and in the 99% it didn't. Which means that the expected value of killed people really is the average amount of killed people, which resolves the conundrum and means there should be no difference between the moral judgement in a Copenhagen universe vs. a many-worlds universe.

From wikipedia:


The many-worlds interpretation implies that there are most likely an uncountably infinite number of universes.

  • Some people may disagree: en.wikipedia.org/wiki/Quantum_suicide_and_immortality May 25 at 13:15
  • @MoreAnonymous I am honestly not sure what that has to do with the answer.
    – kutschkem
    May 25 at 13:39
  • Some many-worlders apparently believe that you who dies. Dies in their world the 99%. While you live on in the 1%. So is it truly immoral to kill someone then? And if so from whose perspective? May 25 at 13:42
  • @MoreAnonymous From the perspective of everyone else in the 99% worlds. But the questions asks a trolley problem and basically counts each person killed in each world separately.
    – kutschkem
    May 25 at 13:52
  • 1
    What is 1% of infinity? What is the average of infinitely many cases? You might start needing serious calculus here ... May 26 at 16:24

In my view, the thing you refer to as schrodinger's cat box is exactly that- not something you want to play in, because of the particular nature of the lumps buried in it by the cat.

It is good to remember that schrodinger's half-dead cat-in-a-box parable was not intended to furnish an actual description of a macroscopic physical system. Note that all interpretations of QM are required to reduce to the classical newtonian picture as you zoom out of the QM scale length and return to the ordinary world of macroscopic objects like boxes, cats, and the contents of cat boxes.

This means that there will never be any connection between one's moral philosophy and any particular formulation of quantum mechanics- because moral philosophy is inapplicable to the study of atoms and electrons and photons, and QM is inapplicable to the study of what human beings think about the abstract concepts they invent while digging in the sand box of philosophical pursuits.

Sand boxes are fine to dig around and play in. Cat boxes are a completely different thing.


I don't understand how you concluded that under a decoherence/many worlds model, the trolley with a 1% chance of killing is "sure to kill 100 people in one universe and zero in the other universe after the decoherence". There's only a 1% chance of killing 100 people in both universes.

Moreover, "probability" as we conventionally understand it doesn't exist in the same way under a many worlds hypothesis. A many worlds hypothesis would, I think, take the view that a switch with a 1% chance of killing 100 people is actually describing an outcome scenario of 100 world "splits," with one of those worlds killing 100 people and the other 99 killing zero.

So the adequate comparison, if we can make one, is that across 100 worlds, the switch that kills 10 people will have killed 1000 people, whereas the second switch would only kill 100 people, making the first switch 10X as deadly and the clear immoral act.

Under the Copenhagen interpretation, one comes to a similar conclusion, albeit probabilistically. Switch #1 delivers an average of 10 deaths per roll of the die, switch #2 delivers an average of 1 death, making switch #1 10X as lethal.


If one were to believe in the MWI of QM, one could conclude ethically that no choice matters. There is no morality: in another layer of the universe the alternative outcome happened.

Don't do this. There is not enough energy in the universe to allow MWI to be true. The proper interpretation of the wave equations is that these are potentials. But not even GOD knows which one will become the Truth.

  • 1
    Well thank God for that!
    – Scott Rowe
    May 27 at 10:00

Does ones interpretation of quantum mechanics alter one's moral philosophy?

You have at play the intersection of physical theory and ethical theory, each distinctive and large projects. Your question asks after how presumptions about physicalism affect one's theory of right and wrong action. This is clearly a metaphysical issue insofar as it touches upon a host issues related to how theories themselves relate. From the philosophy of science, let us introduce the idea of theory-ladenness. From WP:

Semantic theory-ladenness refers to the impact of theoretical assumptions on the meaning of observational terms while perceptual theory-ladenness refers to their impact on the perceptual experience itself. Theory-ladenness is also relevant for measurement outcomes: the data thus acquired may be said to be theory-laden since it is meaningless by itself unless interpreted as the outcome of the measurement processes involved.

Insofar as science is a form of empiricism, and a moral theory has empirical presumptions, then the answer to your question is absolutely, ones interpretation of quantum mechanics can alter one's philosophy. That is because ethics in a very real sense is about formulating normative statements about action in the physical world. That even applies to a misunderstanding of physical theory. Let's take a Schrödinger's cat as a moral dilemma.

A simplistic and (IMNSHO) wrong interpretation is to believe in the claim that the cat is alive and dead at the same time in the vein of scientific realism. Suspending the logical challenge of accepting such a situation (a scientific instrumentalist need not grapple with such an ontological contradiction), one now has a very real moral dilemma. Is it okay to run an experiment putting a cat in a box to see what happens? We now have two possible theoretical presuppositions: the realist case and the instrumentalist case. Let's see the outcome.

Realist case: By conducting the experiment, you are killing exactly 1 cat. So, if anyone is pondering the ethics of such an experiment, and one rejects the act of killing a cat under these circumstances immoral, then clearly the experiment is immoral. Thus, some disciple of Peter Singer may outline quite the sophisticated and persuasive argument based on an interpretation of quantum mechanics to arrive at the conclusion such an experiment is unethical.

Instrumentalist case: An instrumentalist would say that the logical contradiction to having an unobserved cat alive and dead simultaneously has nothing to do with physical reality at all. In fact, such an experiment with a detector cannot be realized for this or that reason, and putting the cat in the box, covering him, and executing the protocol results in the limits of our theorization in predicting outcomes. Hence, the cat being alive and dead is a conceptual reality, not a physical one. Thus, since no cats can possibly die in such an experiment, the cat will remain unharmed, and no unethical action has occurred.

So, who is right? Well, to the extent you agree with one ethical argument or the other is going to be a function of both your views on physical and ethical theory combined, with the former determining the outcome of the latter!


So it's worth pointing out a few things. Decoherence does not solve the measurement problem.

the claim that decoherence solves the measurement problem of quantum mechanics, and discusses the exacerbation of the problem through the inclusion of environmental interactions. It is thus important to consider not decoherence by itself, but the interplay between decoherence and the various approaches to the foundations of quantum mechanics that provide possible solutions to the measurement problem and related puzzles.

Secondly, lets say you actually do this as an experiment where you put the observer himself into a superposition and then the observer makes a choice on the basis of a measurement outcome.

Then this is a deja-vu of another thought experiment: Wigner's friend which a leading quantum information theorist concludes the source of the paradox is that the human evolves in a manner which is not unitary (Only unitary evolution or evolution via the Born rule is allowed) not human choice based upon a measurement outcome!

Personally to truly accept the premise of the question means and understand whatever quantum mechanics is trying to tell us, solving Wigner's friend would be a good first step.

  • The Frauchiger-Renner paradox is an extension of the Wigner's friend which has some very interesting implications: "This indicates that quantum theory cannot be extrapolated to complex systems, at least not in a straightforward manner."
    – JimmyJames
    May 25 at 20:34

The ultimate question posed by the trolly problem is not how many are killed, but by making a choice do you become responsible for the outcome? Morals are also interpreted through the lens of the community/culture, so if you are concerned with morality, you'd have to temper your answer with all possible cultures across the (uncountable) possible worlds.

  • 1
    Yes, but I think the 'hook' in the problem is that you become responsible as soon as you can see it. There is no "not deciding" there. From the standpoint of morality, I don't think it matters how the choice point arrived. You have the same choices whether an evil person set it up, or a boulder started the trolley rolling. Responsibility is for a court to determine. Maybe the boulder scenario should have been prevented by someone? You just have the Yoda Choice: do, or don't do. Similarly, morals of various people are not significant, only the outcome of your singular action. As I see it...
    – Scott Rowe
    May 27 at 9:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.