A Secular Evaluation of Rebirth

Rebirth: it’s one of those topics that defines the Secular Buddhist approach. Practitioners who accept the traditional Buddhist notions of rebirth and the kammic causation that accompanies it will be less interested in a naturalistic ‘secularization’ of the dhamma.

Discussions along the frontiers of belief tend not to be very fruitful: people find their beliefs and stick to them. That said, it can be useful simply for the purpose of openness to put forward some of the reasons for belief, the reasons for taking the approach one does. That way the reasons don’t remain hidden, appearing simply to be objects of faith or ill-consideration. To that end, I think some discussion of my reasons for rejecting rebirth are in order.

The Evidence for Rebirth

Insofar as we know anything about his dhamma, we know that the Buddha taught literal rebirth. We find this in myriad texts throughout the Canon, however for the purposes of this piece I will focus on one, from the Maha-Assapura Sutta (Majjhima Nikāya 39.19). This is a section reflected in other suttas such as MN 4.27, or Dīgha Nikāya 2.93-94.

Here the Buddha outlines what is to be expected from a monk in higher training.

He recollects his manifold past lives, i.e., one birth, two births, three births, four, five, ten, twenty, thirty, forty, fifty, one hundred, one thousand, one hundred thousand, many aeons of cosmic contraction, many aeons of cosmic expansion, many aeons of cosmic contraction and expansion, [recollecting], ‘There I had such a name, belonged to such a clan, had such an appearance. Such was my food, such my experience of pleasure and pain, such the end of my life. Passing away from that state, I re-arose there. There too I had such a name, belonged to such a clan, had such an appearance. Such was my food, such my experience of pleasure and pain, such the end of my life. Passing away from that state, I re-arose here.’ Thus he recollects his manifold past lives in their modes and details. Just as if a man were to go from his home village to another village, and then from that village to yet another village, and then from that village back to his home village. The thought would occur to him, ‘I went from my home village to that village over there. There I stood in such a way, sat in such a way, talked in such a way, and remained silent in such a way. From that village I went to that village over there, and there I stood in such a way, sat in such a way, talked in such a way, and remained silent in such a way. From that village I came back home.’ In the same way — with his mind thus concentrated, purified, and bright, unblemished, free from defects, pliant, malleable, steady, and attained to imperturbability — the monk directs and inclines it to knowledge of the recollection of past lives. He recollects his manifold past lives… in their modes and details.

Doubtless many believing practitioners take passages such as these as sufficient evidence on their own to establish the truth of rebirth. After all, the Buddha was an uncommonly bright and insightful man, one who would not construct fanciful stories for no reason. If he said it, one ought to take it as at least a reasonable hypothesis.

That said, the Buddha also told us not to take the dhamma simply on his word: we were to investigate it ourselves. To that end, a handful of contemporary Westerners have attempted to find evidence for rebirth. Most prominent among these is the psychiatrist Ian Stevenson, who claimed to discover evidence of past life memories in small children, as well as links between birthmarks and causes of death in previous lives. He gathered the stories into papers and books, which contemporary Buddhists such as Ajahn Brahm and Bhikkhu Bodhi believe are compelling further evidence for Buddhist rebirth.

Problems with the Evidence I: Internal Issues

Before turning to more contemporary interpretations, there are problems with the evidence presented, even on its own terms. The Buddha claims in MN 39 that he retains memories of aeons of prior births. If we assume a life lasted on average twenty years, a hundred thousand births takes us to a time some two million years ago. Modern humans (homo sapiens) originated some ten thousand prior births ago, on this scale, so at that time the Buddha would have been remembering prehuman ancestors. That’s to say, over ten thousand births ago, the bodhisatta (as he would have been at the time) could not have been born into the human realm, since there were no humans. And although the origins of language are foggy, that is probably well before modern languages arose. Yet there is no mention in the suttas that his appearance, food, or clan lifestyle would have diverged radically from the settled towns of 5th c. BCE India.

If we take that time period back to the aeons of cosmic contraction and expansion, the problems only ramify. Around 700,000 to a million lifetimes and we are into the pre-hominid.

At this point there is certainly no developed language, and the bodhisatta would have had no name. He could only have been one or another variety of animal, but even so, animals only go back about 600-700 million years.

Prior to that it’s not clear the bodhisatta could have been reborn on Earth, at least that would be the case if we assume that only animals have the consciousness available for kamma and rebirth.

Of course, the Buddha could have been reborn on other planes or planets, but once again there is no mention of vast divergences in body plan, language, culture, or surroundings that would indicate such a rebirth. Indeed, the evidence provided in MN 39 is consistent with a world in which humans always existed in a way much as in the Buddha’s own time. If this is evidence for rebirth, it is not very convincing. More convincing would have been some otherwise inexplicable stories about social, linguistic, and morphological change as the Buddha retreated into memories of the distant past.

Ian Stevenson’s work is similarly problematic, even on its own terms. Firstly, from within a traditional Buddhist context, the ability to see past lives is not something we should expect from young children. Instead, it’s supposed to be one of the three forms of “higher knowledge” available to a monk in advanced training. So why a traditional Buddhist should accept such stories at face value is something of a conundrum. There are also doctrinal problems surrounding whether rebirth is instantaneous or involves some extended time ‘between lives’. Stevenson’s stories support the latter conclusion, but as I understand it, some Buddhist schools reject that move.

As well, the most these stories could reasonably establish is that some people are reborn sometimes. Stevenson’s stories are not found everywhere; indeed, they are very unusual, and they do not establish more than a single rebirth for each life. Are we to assume that people who do not recall such stories are also reborn? Are we to assume that people who recall a single rebirth have been reborn countless times? If so, on what evidence?

Problems with the Evidence II: Trustworthiness of the Buddha’s Story

Contemporary studies have refuted the notion that memory is akin to a movie, retained in full fidelity if only we could get at it. Instead, memory has more in common with planning, in that it is a creative construction out of vague, encoded tracings. As psychologist Frederic Bartlett put it back in the 1930s,

… if we consider evidence rather than presupposition, remembering appears to be far more decisively an affair of construction rather than one of mere reproduction.

Remembering is not the re-excitation of innumerable fixed, lifeless and fragmentary traces. It is an imaginative reconstruction, or construction, built out of the relation of our attitude towards a whole active mass of organized past reactions or experience.

Psychologist Elizabeth Loftus (a past interviewee on The Secular Buddhist) has been at the forefront of such studies, particularly with so-called “false memories” and issues of confabulation. She and others have shown how comparatively easy it is to plant false memories by using leading questions, and further that imagining false events gave subjects more confidence in the reality of the false memories.

Buddhaghosa’s Visuddhimagga (XIII.24) suggests we are to find our first past life by tracing our memories back “in reverse order” until we arrive at our own conception, and then pushing back to before that first moment. As I understand it, this is the typical methodology applied today within traditional Buddhist practice. The problem with this method is that it assumes what contemporary psychology knows to be false: that memory is akin to a movie, which can be unproblematically ‘rewound’ in reverse order, retaining faithfulness.

In fact, long term memories don’t even appear until sometime around the first year of life. Hence any purported memory that comes from a time before one was around a year old is evidence of confabulation, not reality. Indeed, psychologist Nicholas Spanos showed that just such false infant memories could be implanted by suggestive questioning.

Spanos and his co-workers found that the vast majority of their subjects were susceptible to these memory-planting procedures. Both the hypnotic and guided participants reported infant memories. … [O]f those who reported memories of infancy, 49 percent felt that they were real memories, as opposed to 16 percent who claimed that they were merely fantasies.

Altered states of consciousness such as those one attains in deep jhāna, when accompanied by suggestions by teachers, colleagues, and texts that it is possible to regress memory through one’s first year of life into a past life, are just the kinds of situations one would expect to result in false memory confabulation.

As Loftus says,

It is highly unlikely that an adult can recall genuine episodic memories from the first year of life, in part because the hippocampus, which plays a key role in the creation of memories, has not matured enough to form and store long lasting memories that can be retrieved in adulthood.

The situation is exacerbated, of course, if we assume that such memories extend to the time of conception, before one had a functioning brain or nervous system. Further, if we are to assume that anyone trained according to the Buddha’s system had the capacity to recall uncounted numbers of past lives in their entirety, they must be able to access a store of information larger than any finite physical system. The brain only has limited storage capacity. All of these issues imply that if the claim of Buddhist rebirth is true, memories must be stored somewhere separate from the brain.

Problems with the Evidence III: Trustworthiness and Stevenson’s Story

While we can perhaps assume that Stevenson and his subjects were, like the Buddha, sincere about their claims, the question is what those claims amount to. In particular, with every given claim of an extraordinary event, there are always two options. As David Hume famously said in his Enquiry:

[N]o testimony is sufficient to establish a miracle, unless the testimony be of such a kind, that its falsehood would be more miraculous, than the fact, which it endeavours to establish.

In other words, given new evidence, we always have to weigh two likelihoods. In this case, is it more likely that the stories of rebirth collected by Stevenson are true and accurate, or is it more likely that he has been less than thorough in his analyses of the evidence? Suffice to say that there is a great deal of room for honest skepticism about both Stevenson’s evidence and the methodology he used to collect that evidence.

In particular,

There is … the obvious problem of confirmation bias. The ideal, according to Stevenson, was to seek out PLE [past-life experience] stories and then try to confirm them. Failure to confirm, however, did not count against the reincarnation hypothesis. In fact, nothing could be discovered using Stevenson’s methods that could ever disconfirm the reincarnation hypothesis. Many scientists would consider this a fatal flaw in his methodology. Another problem is that there seem to be alternative, non-paranormal, explanations for all of his data.

There is no way, outside of confining people at birth and recording all evidence they have ever been exposed to, to know for sure whether or not stories that children recounted had reached them by some other means than past-life memories. Further, in any collection of large amounts of data, there will always be apparently miraculous coincidences by chance alone. Think here of the purportedly miraculous similarities between the Lincoln and Kennedy assassinations. Snopes.com has a good rundown of the evidence in case you’re interested; the main problem here is one of the misuse of statistics. In order to know whether or not those coincidences really are miraculous, one needs to do thorough statistical analysis on the data, and this requires that the experiment be properly controlled. Stevenson’s methodology did not lend itself to such controls, and hence was worthless to establish anything much more than telling good stories.

Perhaps it’s for these sorts of reasons that the New York Times said of Stevenson on his passing,

Spurned by most academic scientists, Dr. Stevenson was to his supporters a misunderstood genius, bravely pushing the boundaries of science. To his detractors, he was earnest, dogged but ultimately misguided, led astray by gullibility, wishful thinking and a tendency to see science where others saw superstition.

Problems with the Evidence IV: Extraordinary Claims

If the Buddha’s notion of rebirth and past-life memory is true and accurate, it implies that our understanding of mind and brain is fatally flawed, and that in fact memory is not dependent upon the brain for storage or retrieval. It further implies that consciousness is not dependent upon the existence of a functioning brain or nervous system, in that a single-celled zygote must support consciousness. It is, perhaps, possible to reject such a claim from within the system, and say that the rebirth-linking consciousness or “gandhabba” (MN 38.26, MN 93.18, DN 15.21) enters the fetus when the nervous system can support some form of measurable activity. The problem with this claim is that ‘measurable neural activity’ is going to be a vague matter, not the sort of all-or-nothing claim that a ‘descent of consciousness into the womb’ (DN 15) would seem to imply.

This notion of rebirth further implies that our understanding of physical causation is flawed, in that the death of a person distant from conception in both time and space makes a real, physical difference to certain developing neural structures: at least those involved in recovering memories from wherever they reside and turning them into physical behavior. If that were to happen, we should be able to find evidence for it through violations of conservation of energy in the developing fetus. Nothing of the sort has ever been measured, of course. Alternately, sometimes traditional Buddhists explain these kinds of problems in terms of some form of “subtle matter” that carries consciousness. However unless “subtle matter” can be cashed out in terms of actual fundamental forces and particles, the explanation amounts to hand-waving.

The whole notion of a rebirth consciousness assumes either that memories are not stored in the brain’s neural network at all, or that if they are, that storage system can somehow be transmitted across time and space by nonphysical or undetectable means. This picture is only made more implausible when we recall just how much information is at issue: the Buddha claims we can recall literally aeons’ worth of lives.

This notion of rebirth also leaves unexplained how it is that the rebirth consciousness moves from life to life: how does it know where to go? If it is able to perceive things around it, such as eggs awaiting fertilization, just how does that perception take place? Does it detect photons? If so, in so doing it must itself disturb those photons and be detectable. If it processes information, it must do so by using energy of some kind, and hence once again be detectable. It perhaps needs repeating that there is no remotely compelling evidence for such ghostly apparitions.

If, on the other hand, the transmission is instantaneous, that would require temporal coincidence between death and conception that is not always guaranteed to be the case, particularly among small (e.g., proto-human) populations. And once again, the motion from life to life remains obscure: there is no cogent explanation of how the death-consciousness becomes aware of the coincident conception, nor how it moves itself over there to ‘descend into the womb’.

In sum, there is simply no plausible mechanism of action for any of this to happen, and no room in our best understanding of elementary physics that would allow for such mechanisms.

The contemporary, scientific understanding of mind, brain, memory, and mechanisms of physical change is based upon a vast amount of evidence, collected over many centuries. We have a pretty good idea of how the world works. We know that the mind is fundamentally integrated with physical processes in the brain and nervous system. We know this based on many separate but converging lines of evidence, from animal studies, to developmental studies, to brain imaging studies, to brain injury studies, to studies of healthy adults.

Given that as background, any claims of rebirth or veridical past-life memories are literally extraordinary, and following Carl Sagan’s Humean dictum we can say that “Extraordinary claims require extraordinary evidence.” The evidence provided thus far, both textually from the Canon and from modern authors like Stevenson, does not rise to that level.

The Buddha and the World

The Buddha was not a scientist any more than was Aristotle or Confucius, and we should not expect him to have had any particular insight into the pitfalls of memory confabulation. Nevertheless such confabulation is the most likely explanation for Canonical stories of past life memories. We know confabulation occurs, we know it occurs when given certain suggestions, and we know that the process involved with eliciting past-life memories, at least in the commentarial tradition, involves eliciting memories that are otherwise impossible to produce veridically: those from when one was less than a year old, indeed even those from when one was a mere zygote.

Stevenson’s collected stories of past lives similarly fails to rise to the level of credibility required to call into question our knowledge of the relation between mind and brain, and our knowledge of physical causation. As psychologist Barry Beyerstein put it in a review of a book about karma and reincarnation,

A compelling reason to doubt that a packet of personality traits and abilities could leap from a dying person, into limbo, and thenceforth to a newly conceived embryo, is the evident linkage of all psychological attributes to highly specific structures and functions in individual brains.

None of this strictly proves that rebirth does not occur, nor does it prove that some people don’t have veridical memories of past lives. All that evidence and reason can do is to illuminate the more plausible alternatives. Those alternatives may, for all that, be incorrect. If they are incorrect, however, we should at least hope for additional evidence, better controlled experimentation, showing that it is possible to recover memories that are not encoded in neural tissue, that it is possible to sustain consciousness in organisms that lack nervous systems, that it is possible for mental causation to cross time and space and alter physical substrates.

Negative experiments along these lines are legion, though they rarely find their way to the front page of your local newspaper, since nobody has any reason to promote them. One can find mention of them in places like Skeptical Inquirer magazine, or outlined at websites like the Skeptic’s Dictionary or Quackwatch.

It is for reasons such as these that any contemporary, scientifically informed Buddhist practice should reject belief in rebirth and its associated kammic causation. The Path is rich enough without them. And while we can make good use of kamma and rebirth as metaphor for our moment-to-moment lived experience of change, or of skillful and unskillful action, we simply cannot make any more of it and expect to end up with a system which is compatible with our best understanding of the way the world works.