In a scenario in which we have conscious souls (ignoring all the arguments against that for now), would the following be a statistical argument for reincarnation?

Supposing we are in something like an inflationary model of the universe where sequentially infinite universes with intelligent life can keep coming into existence and eventually there will probably always be intelligent life in some of them at any given moment.

If you find yourself conscious within a body, it's either possible that you live one life and then spend eternity dead (with no conscious experience, or some kind of afterlife, it doesn't matter for the sake of this argument) and you just happen to be experiencing your one life at this moment, or that you are eternally transferred into new bodies so any moment in time will involve experiencing life in a body.

It seems to me in this scenario it is statistically more likely that you are experiencing life in a body because reincarnation is true, otherwise it is a huge coincidence that you find yourself in this brief moment in time when you are experiencing an embodied life instead of being dead.

I also intuitively feel like there might be something wrong with this argument relating to time and moments in time but I'm not sure what?

I'm not trying to argue for reincarnation (personally I find the idea terrifying!) but I wonder if this argument works on a probability level?

I understand it's a totally different question if we don't imagine consciousness to be an immaterial soul and there are many other potential objections but purely imagining it were possible to transfer consciousness to another body, does this then make reincarnation highly statistically likely compared to just happening to be experiencing the one time in eternity when we are embodied?

  • If you want to focus on the probability argument, you could try the following exercise: reformulate your question by replacing "experiencing life" with "watching a tape" where you are in an archive room with many tapes taken by several videomakers, and asking "is it more likely that each videomaker only made one video, or that videomakers made many video each?"
    – Stef
    Commented Jan 24 at 16:23
  • 18
    "I'm currently sitting at my desk, what are the odds? Seems implausible that there is only one place in the world that is my desk, and i just happen to be seated there. It's statistically more probable that every single place in the world is actually my desk."
    – armand
    Commented Jan 25 at 2:40
  • 1
    "otherwise it is a huge coincidence..." I'm not entirely sure what you think is coinciding? Commented Jan 26 at 9:38
  • By your argument, your entire existence is you asking this question, it's all that you ever do. Wouldn't you rather do something else?
    – Scott Rowe
    Commented Jan 26 at 21:44
  • 1
    Related-to-the-point-of-duplicateness: Being alive today: the most improbable coincidence? Commented Jan 27 at 18:20

8 Answers 8


The specific statistical error you're probably intuiting is that your sample is not representative. Even if you are non-existent for a million years, and existent only for a day, there's a 100% probability that any day you ask the question "why do I exist" will be the one day that you actually do exist. (This is the basis of what's known as the anthropic principle. https://en.wikipedia.org/wiki/Anthropic_principle .) Asking a question requires existing, so the fact that you have asked presupposes your existence. Since your existence is a precondition to being able to pose the question, it's not possible to tell anything about the relative probability of your existence from the fact that you were alive to ask it.

  • You can, however, get some interesting insights and even a couple of testable hypotheses by noting that it is statistically probable that, for any statistically independent parameter, for any given question, you are within one standard deviation of the average measurement for the kind of thing that can ask the question that you are asking. Although of course not guaranteed, even if you've guessed right about the parameter's independence.
    – g s
    Commented Jan 25 at 7:09
  • Thank you. But if you look at the Botlzmann Brain problem which scientists do take seriously, if there exist vastly more conscious agents created by random fluctuations than by normal evolution then on finding yourself alive as an agent, you are more likely to have come from a random fluctuation (I think other argumetns defeat the Botlman brain problem but it's stastically valid). It isn't enough to say "I'm could only be having this experience if I'm conscious therefore it's as likely that I'm in a normal brain as a Boltzmann brain." Commented Jan 25 at 14:32
  • To follow on, if the universe has existed a set amount of time, say x millenia, I'm much more likely to be having an embodied experience at this set time if I am always having an embodied experience because I am always reincarnating, than if there is only one time at which I could have been embodied. I think the distinction between being embodied and exisitng is important here. In the same way it's valid to ask am I having an embodied experience because I'm in a boltzmann brain or because I'm in a brain that's evovled naturally (I do accept Steve Carroll's altenrative argument against BBs). Commented Jan 25 at 14:38
  • 6
    You're adding parameters that change the problem in significant and determining ways. IF you pick a time in human history, randomly, and you pick a specific soul, Cat Rat Pup, and you find that specific soul is at that time, then that is statistically meaningful in a way that starting from now, and starting from yourself isn't. Your thumb is on the scales in the original question. Commented Jan 25 at 15:20

This looks like just another form of the lottery paradox. Suppose x buys one lottery ticket and there are a million other tickets. It is rational to believe that x will not win the lottery, yet it is not rational to believe that no individual will win. Let R(P) stand for "it is rational to believe that P", and let Wx stand for "x wins the lottery", then it is the case that

forall x.R(not Wx)

but it would be faulty to conclude from the above that

R(forall x.not Wx)

Your argument amounts to a variation on this reasoning. Let H(P) stand for, "it is highly probable that P" and let Lx stand for "x is living at x's particular time and place". Then while it is arguably true that

forall x.H(not Lx)

from this you cannot conclude that

H(forall x.not Lx)

You can't conclude this, because everything has to exist at some time and place, so while for each individual, that individual may be highly improbable, this improbability does not spread out to the set of all individuals.

  • Yes, and just think about all the people who never lived! It is pretty astonishing to consider.
    – Scott Rowe
    Commented Jan 26 at 21:50

You are making the faulty assumption that if a event A is more likely under circumstance X than under circumstance Y, then circumstance X is more likely than circumstance Y if A has happened.

Suppose I pick a seven of clubs randomly from a pack of cards. The odds of me doing so are much higher if the pack is rigged and consists only of sevens of clubs, but that does not mean the pack is more likely than not to be rigged.

  • exactly. That's basically maths question, all the rest is based on wrong maths
    – user68439
    Commented Jan 25 at 13:33
  • 1
    On the other hand, if you pick two sevens of clubs randomly from the same pack of cards...
    – Stef
    Commented Jan 26 at 11:51
  • 1
    @Stef "You Only Live Twice" (if your name happens to be James Bond)
    – Scott Rowe
    Commented Jan 26 at 21:47

otherwise it is a huge coincidence that you find yourself in this brief moment in time when you are experiencing an emboided life instead of being dead.

But, no, it isn't. Experience is linked to observer bias.

This concept is closely related to the Anthropic principle.

In your thought experiment, all those eons in which the subject is dead, that subject has no experiences. Only that period during which the subject is alive does he or she have experiences. And only during that period do they make "wow, I'm alive, isn't it an amazing coincidence that it's at this particular time and place!" type observations.

  • On my breaks from work I often walk in a very large nearby cemetery. Perhaps everyone should?
    – Scott Rowe
    Commented Jan 26 at 21:40
  • 1
    @ScottRowe I know of a cemetery whose driveway has a "No Exit" road sign.
    – Kaz
    Commented Jan 27 at 1:16

Before answering, let's generalize and ask: Is any philosophical argument inevitable? It's tough to say what is meant by philosophical arguments being inevitable. If one follows the logic of Jakob von Uexküll and his notion of the Umwelt, then I believe one would be forced to conclude that reincarnation is a likely narrative to arise from the argument that human beings possess a soul. From WP:

n the semiotic theories of Jakob von Uexküll and Thomas A. Sebeok, umwelt (plural: umwelten; from the German Umwelt meaning "environment" or "surroundings") is the "biological foundations that lie at the very center of the study of both communication and signification in the human [and non-human] animal".1 The term is usually translated as "self-centered world".[2] Uexküll theorised that organisms can have different umwelten, even though they share the same environment.

While different people have different Umwelten, they all derive from reasoning about one's existence and relationship to the world. Religions seem to have similarities in reasoning. The structure of such reason seems to proceed along these lines:

  1. The difference between the living and non-living is magical substance called a soul and is either comes from the universe or from one or more gods.
  2. If one has a soul, then the soul must have a cause and have a relationship to birth and death.
  3. Either one gets a soul, and the soul disappears at death, or it survives.
  4. If it survives, it either eventually dissipates, it goes on forever in some way, or it gets reused in a new birth.

Thus, we see there are three outcomes to having a soul, its dissolution at death or sometime afterwards, its immortality after death, or its reuse which handily solves the problem of where souls come from in the first place. Thus, belief in a soul becomes part of the meta-narratives derived from one's Umwelt, in that religions play an important role of providing epistemological, ontological, and ethical theories about the world.

Given that there have been more than 10 billion human beings, and that the religious impulse seems to go back to the emergence of language and the burial of the dead among Hominid ancestors, it seems likely that it is inevitable someone somewhere would get to a narrative that settles on reincarnation.

  • I've seen estimates that something over 35 billion humans have been born, although of course many died in infancy (walk through a cemetery and see). Of those no longer living, we know zero about the vast majority of them. So yeah, stories are one way to stay sane I suppose, because even wrong stories aren't necessarily harmful.
    – Scott Rowe
    Commented Jan 26 at 21:56

Philosophy, being a non-scientific field of study, draws assumptions and not facts. Science says "It is not true until it is proven", while philosophy says "Here is an argument that is more convincing". That is no way to come to a conclusion.

First, I would argue the existence of the multiverse. It is not proven, it is hypothetically discussed about from time to time. This does not constitute a reason to debate if it does or doesn't exist. It does merit reason to research and try to prove, though. From here on, the idea is already shaky, but let's take into consideration some numbers.

In our Milky Way galaxy, there are Billions of planets (1 000 000 000). And, in our observable corner of the universe, there are Billions of galaxies who range from smaller to larger than the Milky Way, so we will assume that they are around the same size. That means that in our observable universe alone, there are quintillions of planets (1 000 000 000 000 000 000). That is quite a large number, so having discarded the idea of the multiverse, we still have a hope due to the sheer, unimaginable and amazing size of our universe. So, we can assume that for every human that ever lived, there would be a copy of them who also lived at some point in time or will live in the future, also assuming that carbon based live is the only one that can develop, narrowing down the possibilities of how life can evolve.

This also touches upon spirituality, religion. One would have to prove that these things are possible and true before continuing the argument, and that is not yet proven by any scientific method. And I am sorry, but this is where the argument falls apart.

In essence, the argument relies on speculative premises, and its conclusions are contingent upon the acceptance of these premises. Without empirical evidence or a more robust foundation for the assumptions made, the argument for eternal reincarnation remains speculative and unproven.

  • 1
    The first sentence is simply untrue - no single statement ever gets conclusively proved unless its an observational statement and we believe (or act as if we believed) in many statements that are not true, in science. We often use probabilistic reasoning to approximate the best theory given our evidence. The argument the person is presenting here might for this reason be even called "(natural-)scientific" and not "philosophical" - but it's flawed, so it doesn't constitute any significant scientific knowledge, just a curiosity.
    – user71009
    Commented Jan 25 at 18:45

It seems to me in this scenario it is statistically more likely that you are experiencing life in a body because reincarnation is true, otherwise it is a huge coincidence that you find yourself in this brief moment in time when you are experiencing an embodied life instead of being dead.

I also intuitively feel like there might be something wrong with this argument relating to time and moments in time but I'm not sure what?

There's definitely something wrong with this argument. It's a fairly common logical error that often leads to mistakes and is employed by con artists to bamboozle people.

Let's try that logic with another more straightforward example. Presumably, you reside at some specific address at some particular location on the surface of the earth. There are a vast number of other locations you could reside on this planet. If we simply look at the raw, unconditional probability that you reside at your address, it is very small. Using your argument, we can therefore conclude that it is a huge coincidence that you live there. If we follow this line of thinking, we can assert that the probability you reside at that address is so small, it can't be true. And, regardless of what address we consider, the same logic applies. Therefore, we can conclude, statistically, that you must not have a residence.

If that's not convincing, consider this. There are really incredibly large number of water molecules in the world's oceans. If scoop a teaspoon of water out, the chance that any given specific water molecule will be in that teaspoon is vanishingly small. This is true for each and every one of the molecules in that teaspoon. Using the logic of your argument, we can conclude that there's no water in the teaspoon and that is impossible to scoop a teaspoon of water out of the ocean.

The minds of all humans many other animals are 'tuned' to notice things that seem unusual or out-of-place. This is important for survival. We tend to have better survival results by assuming random coincidences are meaningful. That broken branch on the ground might indicate a predator or enemy. If it turns out that there is no danger, there's not much cost to that. These systems, which are really important in the natural world, can lead us astray if we apply them to logic and reason.

  • Yes, and factor in the very small probability that I am any good at math and the likelihood of me making a good argument using it is vanishingly small.
    – Scott Rowe
    Commented Jan 26 at 21:37

Your effort to try to think this thru statistically is not valid, as you cannot show that there is a uniformity across samples of all possible. Yes, we on this discussion board all only remember being incarnate, but that is not definitive.

Under spiritual dualist models we have to have memory wipes before entering our current bodies, in order to match our lack of memory of a prior state. So we COULD, over a full sample set, mostly experience lives as discarnate not incarnates, and just not remember most of the counterexamples of not being incarnate.

Memory wipes mess up your effort to think this thru statistically.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .