What kinds of ethical implications should we expect from the Many-Worlds Interpretation of Quantum Mechanics (MWI)? I’ll argue that we shouldn’t expect decision-making to change. The implications are more about how we should think or feel about events in our lives, and the virtues of taking a cosmic perspective.

According the many-worlds interpretation (MWI) of quantum mechanics, the universe is constantly splitting into a staggeringly large number of decoherent branches containing galaxies, civilizations, and people exactly like you and me^[1].

You might think such a metaphysically radical theory should have pretty radical implications for how one should live. The quantum physicist John Bell once wrote “if such a theory were taken seriously, it would hardly be possible to take anything else seriously”. 

But proponents of MWI - including both Eliezer and David Wallace - have concluded the opposite. Eliezer quotes Egan's Law: It all adds up to normality. There are no major ethical implications at all^[2].

I’ll argue that it's important to distinguish between two kinds of ethical implications we might expect MWI to have, which I’ll call ‘decision-theoretic’ and ‘virtue-theoretic’ (I’ll explain what I mean by these names). 

Eliezer and Wallace are right that MWI doesn’t have decision-theoretic implications. But they overlook the fact that MWI plausibly has implications for virtue theory.

Decision theory in the multiverse

The main reason for thinking MWI has no significant implications for ethics comes from decision theory. 

In standard decision theory you try to calculate the expected value for a particular course of action by multiplying the utility of each possible consequence by the probability of that consequence occurring. 

The main difference in MWI is that each possible consequence corresponds to a world that actually ends up existing, rather than being just hypothetical.

But this makes no difference to the expected value that the calculation spits out. As Eliezer puts it:

“Your decision theory should (almost always) be the same, whether you suppose that there is a 90% probability of something happening, or if it will happen in 9 out of 10 worlds. Now, because people have trouble handling probabilities, it may be helpful to visualize something happening in 9 out of 10 worlds. But this just helps you use normal decision theory.”

So it seems that we end up making the same decisions in MWI as we would otherwise, which in turn seems to imply that MWI has no significant implications for ethics.

Problems for decision theory

Now, one can question this argument. Those who argue that decision theory does work differently in MWI, usually start from a version of MWI in which it makes sense to count the number of worlds.

If you allow ‘branch-counting’, as this approach has been called, then decision theory seems to break down. To take a simple example, suppose you do a quantum experiment with two possible outcomes - your measurement apparatus makes a beep or it doesn't. 

On the branch-counting approach, where there was previously one world, there are now two: each containing the same people, animals and other valuable objects - except for the beep.

The question is: do you now also have lots more value? Consequentialist approaches to ethics (ones which say for example that two happy people are better than one) would seem to imply that you do, and that therefore doing the experiment is extremely desirable - or possibly extremely bad, if you think total disvalue outweighed total value before the split.

And of course from a decision-theoretic perspective it seems this evaluation of consequences should inform the utilities we assign in order to calculate expected value.

It’s true that this simple example ignores the fact that in MWI branching is happening all the time. 

But that fact just makes the decision-theoretic situation worse. If we're allowed to count branches, the number of worlds, and therefore the amount of value and disvalue, is rapidly increasing all the time.

Of course this seems to be a reductio ad absurdum^[3], but which premise do we let go of? Do we reject the branch-counting approach to MWI or reject the consequentialist approach to calculating value?

Indefinite numbers of worlds

Fortunately for consequentialists, David Wallace has developed a detailed version of MWI that does not involve branch counting. The number of worlds that result from quantum processes, on this view, is in fact undefined. As he puts it:

“Decoherence causes the Universe to develop an emergent branching structure. The existence of this branching is a robust (albeit emergent) feature of reality; so is the mod-squared amplitude for any macroscopically described history. But there is no non-arbitrary decomposition of macroscopically-described histories into ‘finest-grained’ histories, and no non-arbitrary way of counting those histories.”

Importantly though, on this approach it is still possible to quantify the combined weight (mod-squared amplitude) of all branches that share a certain macroscopic property, e.g. by saying:

“Tomorrow, the branches in which it is sunny will have combined weight 0.7”

This allows Wallace to build up a detailed model of how decision theory works in MWI - and how it produces the same results as classical decision theory, as Eliezer suggests.

Wallace shows that by choosing a specific set of sensible axioms, you can formally prove the Born rule in quantum mechanics, which states that mod-squared amplitudes can be treated as probabilities.

And one of Wallace’s axioms, which he calls ‘branching indifference’, essentially says that it doesn’t matter how many branches there are, since macroscopic differences are all that we care about for decisions.

So Wallace’s proof confirms that in order for decision-theory to give sensible results in MWI, you need to stop thinking about numbers of branches; and he thinks that’s OK because numbers of branches are in fact undefined.

All of which is to say that there is indeed a way MWI ‘adds up to normality’. You can still get decision theory to give you the same results as before - it just takes a bit of housekeeping to iron out branch-counting wrinkles.

Virtue theory and virtue ethics

So Wallace and Eliezer are plausibly right that MWI doesn’t have ethical implications in the decision-theoretic sense outlined above. 

The mistake is to conclude that MWI has no ethical implications at all. The focus on decision theory leads us to overlook other kinds of ethical implications MWI could have.

A sizable portion of (both contemporary and historical) ethical theory is not about decisions at all, but rather: what kind of person to be, what kinds of character traits are desirable and how one should think and feel about situations.

It’s common to think of ‘virtue ethics’ - understood as the approach to ethics (deriving ultimately from Aristotle) in which such things are treated as fundamental - as one of the three main approaches to ethical theory, the others being the deontological (Kantian) and consequentialist (deriving from Bentham and Mill’s Utilitarianism) approaches.

But you don’t need to be in the ‘virtue ethics’ camp to think virtue is worth understanding. Consequentialist and deontological approaches to virtue exist as well. 

For instance a consequentialist might propose that what makes a trait virtuous is its tending to lead to good consequences. And the well-known difficulties involved in actually quantifying the utilities of all possible consequences are among the reasons for consequentialists to show interest in virtue^[4].

In short, it’s possible for those from a variety of perspectives to agree that the focus on decision-making leaves out a great deal of ethics.

So even if MWI doesn’t have decision-theoretic implications, it could still have virtue-theoretic ones. For example, it could have implications for how we should think or feel about our location within the multiverse.

Wisdom in the multiverse 

To make such implications seem not only technically possible but also plausible, I'll now  sketch out some specific virtue-theoretic implications of MWI (I'll aim to go into more detail on these in future posts).

Firstly, consider that there's a certain kind of anxiety and regret associated with having to choose between two mutually exclusive good options. It seems plausible that MWI could help us feel better about such choices, if it's true there's a world in which you actually experience the other good option.

Secondly, consider the simple fact that if you find yourself in an extremely unlucky personal situation - a car crash, say, or getting cancer -  then MWI implies that there are other worlds, with high quantum weights, in which you are not so unlucky. 

Again this is plausibly a consoling thought, similar in kind to the consoling thoughts recommended by philosophical traditions like stoicism.

Likewise, if you're in what you estimate as the bad end of the spectrum of physically possible 'timelines' of global history, it seems consoling to know that those other timelines are real.

From a virtue-theoretic perspective, we could say that it’s good to develop the disposition to take the wider cosmic perspective these thoughts assume, and so enhance one’s equanimity (a standard goal of classical virtue theory^[5] which is arguably linked to both well-being and effective action).

Our cosmic situation

Zooming out further, there is also the ability to take the widest possible cosmic perspective, and consider one’s place in the quantum multiverse as whole. 

Other fundamental scientific theories have been taken to have implications of this sort.

For instance, the second law of thermodynamics, and the theory of evolution have both been taken to imply that the universe as a whole is essential hostile to human interests.

In 'A free man's worship'. Bertrand Russell proposed that the second law of thermodynamics is a kind of foundation for one's overall philosophical perspective:

all the labours of the ages, all the devotion, all the inspiration, all the noonday brightness of human genius, are destined to extinction in the vast death of the solar system, and that the whole temple of Man's achievement must inevitably be buried beneath the debris of a universe in ruins - all these things, if not quite beyond dispute, are yet so nearly certain, that no philosophy which rejects them can hope to stand.

In a similar vein, Eliezer's discussion comparing evolution to a blind, idiot God ends with the suggestion that our basic stance towards the universe should be confrontational^[6]:

Well, more power to us humans. I like having a Creator I can outwit. Beats being a pet. 

Similarly, MWI plausibly has implications for our assessment of the goodness or badness of the cosmos as a whole: whether we should feel at home in nature, or set against a hostile universe.

Again taking a virtue-theoretic perspective, we might say that the virtue of wisdom requires an accurate appreciation - both intellectually and emotionally - of our place in the multiverse.

Summary

As I said at the start, MWI is a metaphysically radical theory that we might reasonably expect to have ethical consequences. 

We’ve seen that proponents of MWI like Eliezer and Wallace, have arrived at the somewhat surprising conclusion that it actually doesn’t.

I’ve argued they're only half-right. They're right that MWI doesn’t have significant decision-theoretic implications. But it plausibly does have significant implications for virtue theory.

1. ^{^}

   I'm not going to argue for this view as that was done very well by Eliezer in his Quantum Physics sequence. And in fact since that sequence was written MWI has become increasingly mainstream, so you can also read for example a major edited volume (2012) and David Wallace's 'The Emergent Multiverse' (2014) for painstaking academic support of the view.

2. ^{^}

   Wallace makes a similar claim in his book: “But do [the many worlds in MWI] matter to ordinary, banal thought, action and language? Friendship is still friendship. Boredom is still boredom. Sex is still sex.” (p273)

3. ^{^}

   Though like many a reductio ad absurdum the conclusion has been taken seriously, eg in this post.

4. ^{^}

   A reminder here that Eliezer has a post called 'The Twelve Virtues of Rationality'.

5. ^{^}

   See for example discussions of Ataraxia in Stoicism and Epicureanism.

6. ^{^}

   Joe Carlsmith's sequence Otherness and control in the age of AGI is a good exploration of these and related ideas under the heading of  'deep atheism'.