It's claimed in An Introduction to Parapsychology, 5th ed.:

The survival hypothesis concerns the notion of postmortem survival, that is, that a disembodied consciousness or some such discarnate element of human personality might survive bodily death at least for a time. The survival hypothesis has obvious religious connotations, but these are of no concern in scientific parapsychology.

The authors claim that:

Even if the survival hypothesis is vague, at least specific forms of it could be open to empirical evaluation.

Is it possible to use empirical evidence to build a case that reincarnation really happens?

(Note 1: This is separate from the question "can belief in reincarnation be justified in general?" See: Ought we only form beliefs based on sufficient empirical evidence?)

(Note 2: This question was partially borrowed from a deleted Skeptics.SE question by njfife. My answer will be copied from there)

  • Of course it is open to empirical confirmation, it is just that the confirmation has not been forthcoming so far. If we could be made to remember experiences of past lives that would make it far more plausible. If, on the other hand, reincarnation is postulated to have no detectable consequences then the question becomes why bother with it at all. – Conifold Nov 5 '17 at 0:07
  • @Conifold See the answer I wrote. – Avery Nov 5 '17 at 0:13

No, reincarnation is not open to empirical validation:

The problem of how life after death can be scientifically studied is not a simple one. Parapsychologists believe progress can be made in a scientific fashion, but they have received doubt both from the skeptic community and from sympathetic, neutral philosophers of science. It is this latter criticism I will focus on. The remainder of this answer is an extremely brief and hopefully accurate discussion of Michael Sudduth's new book A Philosophical Critique of Empirical Arguments for Postmortem Survival (Palgrave Macmillan, 2016).

The example of Ian Stevenson's work on reincarnation is pretty straightforward: he proposes to simply find case studies of alleged reincarnations who possess factual information about their past lives that they couldn't have come across by normal means. This would seem to be evidence of some kind of paranormal activity, and indeed, much of the criticism of Stevenson has focused on trying to disprove his dozens of examples.

However, what Stevenson provides is data, not a hypothesis. The "survival hypothesis," the statement "The soul survives death", does not cause us to expect any this-worldly phenomena in itself. We could be living in an orthodox Christian world where the future states of the soul do not affect us in any way. That doesn't make the claim unusual; according to the Durhem-Quine thesis, there are very few scientific claims that can be falsified by themselves. To restate Wikipedia's example, Galileo seeing rings around Jupiter could be explained by Jupiter being surrounded by a dozen small moons and thousands of tiny objects in a ring, or it could be explained by a scratch on the lens of Galileo's telescope, as one of the other intellectuals of the day suggested. So, a truer statement of Galileo's hypothesis is actually "Jupiter has rings and I can see them perfectly well through my telescope". The second half of this is called an auxiliary hypothesis. It's not essential to the main hypothesis but we need it to get from the data we currently have to the main hypothesis.

Similarly, the survival hypothesis in a more testable form must make more specific claims that cause us to expect data, like "The soul survives death and goes to Heaven, and people can experience this themselves through near death experiences," or perhaps "The soul survives death, reincarnates sometimes on Earth, and retains at least some of its memories, and we can confirm this and rule out cold reading through carefully controlled interviews." The latter is closer to Stevenson's actual view.

But we are not done with auxiliary hypotheses yet, because at some point they lose explanatory power. For example, if Stevenson were to narrow down his claim to "I myself spoke to a reincarnated person once, in a unique event that cannot be duplicated," that's not particularly helpful for our lives. In the more general form he gives in his books, Stevenson's claims can be explained in all sorts of other ways:

  1. Children like making up stories and sometimes hit on several real facts about other people through totally random chance.
  2. Children can access the memories of relatives of a deceased person through some unknown mechanism. (Living psi)
  3. Memories drift around in the ether at death and sometimes fall into other people's heads totally independent of the question of survival of the soul. (René Guénon's "psychic residues")
  4. Stevenson was systematically deceived by his interpreters, or for some other reason didn't see what he thought he saw. (Cartesian demon)
  5. Regardless of what country you're in, for some reason -- pursuit of fame, confirmation of community beliefs, etc. -- people have a strong inclination to fake reincarnation-like incidents.

None of these alternative explanations actually falsify the claim "The soul survives death, reincarnates sometimes on Earth, and retains at least some of its memories, and we can confirm this and rule out cold reading through carefully controlled interviews." They just tell us we haven't truly tested it yet.

But alas, that tells us that we need additional auxiliary hypotheses which reduce the explanatory power of the survival hypothesis so much that there's no particular reason to accept it as an explanation of Stevenson's data over any of the other hypotheses. In particular the living psi hypothesis would seem to cover any anomalous data that could possibly be claimed to support the survival hypothesis. So according to Michael Sudduth's book, the argument that "the soul survives death" necessarily has no explanatory power, regardless of the more specific data it is meant to analyze, or the specific test meant to prove or disprove it. There is no way to state it in a way that becomes scientifically valuable.

Sudduth concludes that the survival hypothesis is not a scientific hypothesis, meaning that it can't be evaluated empirically. This is the problem with much of parapsychology: experiments may suggest to a reader that something weird is going on, but it's hard to turn "something weird" into a scientific hypothesis, especially when it's an anomaly that is not systematically reproducible and varies its nature widely between experiments. Such deficiencies reduce the explanatory power of any hypothesis significantly.

  • In my (lay) opinion, this answer would be much improved by inclusion of a clear, one-sentence TL;DR (perhaps with a little boldface to draw attention), ideally starting with “Yes, reincarnation is open to empirical validation ... “ or “No, reincarnation is not open to empirical validation ...” . – Dan Bron Nov 4 '17 at 7:44
  • How common is it for a person to ask a question only to answer it himself on the same day? – jeffreysbrother Nov 4 '17 at 14:49
  • @jeffreysbrother It is explicitly permitted. You can do this to "share your knowledge Q and A style". In that case, you're not expecting others to answer (and there's no need to wait for them to do so), since your plan was to answer yourself anyway. – barrycarter Nov 4 '17 at 17:24
  • +1 for the addition of the first sentence. Well done. – Dan Bron Nov 5 '17 at 16:56

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

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