How sleep disproves Quantum Immortality

There is a clever thought experiment, originally published in the 1980s by Hans Moravec [1], then developed by Max Tegmark [2]. It’s called Quantum Suicide, and it offers a path to immortality.

Consider Schrödinger’s famous cat-in-a-box. Now imagine that it is not a cat-in-a-box with a 50/50 chance of death, but you. You are in the box. There is a button next to the box which, if pressed by an observer, will trigger an event inside the box with a 50% chance of killing you. Let’s say that the button is pressed more than once. In fact, let’s say that an observer presses it 100 times.

The probability of you remaining alive at the end of this is 1∕2^100. To attempt to put that number into perspective, the Voyager 2 space probe is currently 18 billion kilometres away. Multiply that by 1∕2^100 and you get a distance one hundredth the width of a proton. You are very very unlikely to still be alive. And indeed you should certainly be dead, if the Copenhagen interpretation of quantum mechanics is true. There is one universe, and it is a universe in which you have the same approximate chance of being alive as you have of winning four National Lotteries in a row.

There is another interpretation of quantum mechanics: Many Worlds. In this interpretation the entire universe doubles itself at every choice. You are dead in one universe and alive in another. If, after 100 iterations of the experiment, you find yourself alive, you can be pretty damn certain that the Many Worlds interpretation is correct. (How certain? Imagine if you flipped a coin a hundred times and it came up heads every one. As certain as you would be that the coin is biased, or has two heads.)

Of course this is the point of the thought experiment. You would not be able to convince anyone else of this fact though. An observer would always find it more likely that something is wrong with the experimental apparatus and will ask for another few tries, one of which would, from their perspective, probably kill you. (Remember, for the observer each try genuinely has a 50% chance of killing you; try again ten times and you have a less than one percent chance of survival.) But what would you perceive? At each branch of the universe there is still, after all, only one you (even as your observer has been duplicated), since the other one of you will be dead. Would your subjective experience naturally continue in that universe? It wouldn’t matter if you went through 100 iterations or a million, you would come up heads every time, a rather extreme example of survivorship bias.

But then, the theory continues, shouldn’t this apply outside the box too? There will come a point when sufficient cellular damage occurs that your brain is no longer able to sustain your consciousness. Yet in one universe a unit of decay will happen, and in another universe it won’t. In one universe you will be dead, in another you will continue for one more instant, and again, and again. Your survival might appear less and less tenable, but so long as there is an infinitesimal chance of it being extended for another moment, you will continue. Subjectively, you are immortal. This is Quantum Immortality.

And yet, you are immortal in a horrible way. Gulliver, in his travels, meets the Luggnaggians. Every so often a Luggnaggian is born with a red ring on their forehead, and this is a great sadness. They are called the Struldbrugs and they are immortal, yet they still age. They get older forever.

Arthur Rackham (1867–1939)

One could say: so why don’t we see 1000 year old, 10,000 year old people? We wouldn’t; the vast majority of universes would not contain them. We are talking about the tiny sliver of space between fantastically unlikely and impossible. We could each be a Struldbrug, and if we were we would be the only one in all our universe.

The idea has — happily, should we think? — been attacked. If the Many Worlds hypothesis is wrong (and it would be difficult to think of a less parsimonious idea) then quantum immortality cannot happen. Even if it is true, dying is not a single event that can be avoided or not, it’s a path that gets ever narrower. (That is a poor objection, it merely restates what we have said about getting older and older without end.)

I want to raise another objection. What, really, is so special about death? Imagine a similar box, inside of which is a mechanism which doesn’t kill you, it just knocks you out. Think of your consciousness. In one universe you have been knocked out — your consciousness is temporarily gone, you have no subjective experience — but in another universe you haven’t been knocked out, your consciousness exists. If your subjective experience leaps to the universe where it still exists on the destruction of its underlying brain, what’s so different about the situation where the underlying brain has been rendered insensible? It should make the same leap. An observer would see you quickly fall unconscious; but you, from your perspective, would remain conscious, your subjective experience continuing each time in the universe where it exists.

This is an experiment which could conceivably be performed. It’s one for which only the subject of the experiment could possibly gain any knowledge though.

We could try taking this argument from its box and into the world too, and plausibly say that if we are subjectively immortal in the Many Worlds, Quantum Suicide sense, then we will stay awake forever.

At each moment in which you could fall unconscious, there is a chance that you remain awake. In that moment two universes are created, and to be consistent with the thesis of Quantum Immortality, you should continue in the universe in which your consciousness exists. This is true for the next moment, and the one after that. An endless chain of wakeful moments (illustrated above). The same logic that says that we are subjectively immortal says that each of us should be that version of us who has never fallen unconscious. In this way, if you woke up this morning, you disproved Quantum Immortality.

[1] Hans Moravec. Mind children: the future of robot and human intelligence. Harvard University Press, 1988.

[2] Max Tegmark. The interpretation of quantum mechanics: Many worlds or many words? Fortschritte der Physik, 46(6–8):855–862, 1998