Buddhism and Science

"…if scientific analysis were conclusively to demonstrate certain claims in Buddhism to be false, then we must accept the findings of science and abandon those claims." — the Dalai Lama, 2005¹

"In the largest and most rigorously controlled study to date, we failed to replicate prior findings and found no evidence that MBSR produced neuroplastic changes compared to either control group." — Kral, Davidson and colleagues, 2022²

Summary

Buddhism is often presented as the religion that science has vindicated:  a "science of mind" whose meditation practices are now confirmed in the laboratory.  This article states that claim in its advocates' own words, traces where it came from, reads the research record paper by paper, and asks what that record can and cannot show.  It then applies the same test to Christian practices, and ends with the one place where Buddhist texts make measurable claims about the physical world:  Mount Meru.

1. The claim has a strong form and a weak form. The weak form is that Buddhist-derived meditation is good for people. The strong form is that meditation is an instrument, like a telescope, that confirms Buddhist doctrine. B. Alan Wallace states it plainly: memories are carried in a "substrate consciousness" from "lifetime to lifetime," a conclusion "based on the experiences of highly trained contemplatives."³ The Dalai Lama commits Buddhism to abandon any claim that science "conclusively" disproves,¹ but adds that science must show a thing "to be non-existent (which is not the same as not finding it)."¹
2. The "science" framing is modern, and it took shape in a contest with Christian missions. Henry Steel Olcott's Buddhist Catechism of 1881, modelled on missionary catechisms, told Sinhalese children that "True science entirely supports" karma and rebirth, and that the science in question was "Evolution."⁴ Later editions said Baron von Reichenbach had "proved" the Buddha's aura and that the Buddha's illusions were "familiar to all students of mesmerism and hypnotism."⁵ Each generation has attached the claim to the science of its day: evolution, the aura, radioactivity and relativity, quantum physics, and now neuroscience.⁶ The English word "mindfulness" for the Pali sati dates from 1881.⁷
3. The clinical record shows small real effects that shrink against fair comparisons. The 2014 review for the US Agency for Healthcare Research and Quality pooled 47 trials with active controls and 3,320 people. Mindfulness programmes reduced anxiety by an effect size of 0.38 at eight weeks and 0.22 at three to six months, depression by 0.30 and 0.23, and pain by 0.33. It "found no evidence that meditation programs were better than any active treatment (drugs, exercise, other behavioral therapies)."⁸ A 2021 review of 136 trials in healthy adults found moderate effects against doing nothing and "no statistically significant evidence" of superiority to specific active controls.⁹ The most favourable review found mindfulness equal to evidence-based treatments (d = −0.004) and slightly better than other active controls (d = 0.23).¹⁰
4. The best clinical result is for preventing depressive relapse, and it equals antidepressants rather than beating them. Across nine trials and 1,258 patients, 38% relapsed within 60 weeks after mindfulness-based cognitive therapy against 49% without it.¹¹ In the largest head-to-head trial, 44% relapsed over two years after mindfulness therapy and 47% on maintenance antidepressants, a difference that was not significant.¹²
5. The largest school trial found nothing. MYRIAD randomised 85 British secondary schools and analysed 8,376 pupils. After a year, the differences on depression risk, behaviour and well-being were 0.005, 0.02 and 0.02 standard deviations.¹³ An exploratory analysis found pupils already at risk did slightly worse.¹⁴
6. The famous brain findings came from small studies and did not replicate. The 2005 report that meditation thickens the cortex compared 20 meditators with 15 non-meditators at one point in time.¹⁵ The 2011 report that eight weeks of mindfulness training grows grey matter compared 16 people with 17, without randomisation.¹⁶ In 2022 Richard Davidson's own laboratory scanned 218 people randomised to mindfulness, an active health programme or a waiting list, and concluded that "MBSR does not differentially affect brain structure."² When the same laboratory measured the brain waves of Tibetan meditators said to remain in meditation after death, it found "no recognizable EEG waveforms" in any of 13 cases.¹⁷
7. Meditation has side effects, and the tradition knew it. Interviews with 60 Western Buddhist practitioners, selected for having had difficulties, catalogued fear, panic, changes in the sense of self and impairment that sometimes lasted years.¹⁸ A review of 83 studies found adverse events in 8.3% of meditators.¹⁹ Trials rarely asked: fewer than a quarter of meditation trials actively assess adverse effects.²⁰
8. None of this tests rebirth, karma or non-self. That a breathing practice helps some people a little is evidence about the practice. It says nothing about whether consciousness passes between lives. The best-known popular defender, Robert Wright, concedes the point in his book's first pages: he is "not talking about the 'supernatural' or more exotically metaphysical parts of Buddhism—reincarnation, for example."²¹ Buddhist critics make the opposite complaint: that secular mindfulness has been "stripped of the teachings on ethics that accompanied it."²² The founder of the best-known programme wrote that he "bent over backward" to keep it from "being seen as Buddhist."²³
9. The same standard applies to Christian practice, and it cuts the same way. Women in the Nurses' Health Study who attended services more than weekly had a 33% lower death rate over 16 years than those who never attended.²⁴ That finding is observational, and the mindfulness record is mostly randomised, so on design the Buddhist-derived practice has the better evidence. Neither record can show that its religion is true. Christianity rests its case on a historical claim, the resurrection, that no health study can test, and the controlled trials of distant prayer were null.
10. Where Buddhist texts make physical claims, they fail as physics, and modern teachers say so. A discourse in the Pali canon has the Buddha say that Mount Sineru rises 84,000 leagues above the ocean.²⁵ The standard scholastic cosmology puts the sun and moon at half Meru's height.²⁶ On the text's own unit, Meru would rise one and a half times as high as the Moon's orbit, and the sun would sit about 520 times too close.²⁷ The Dalai Lama's verdict is that "Buddhism must abandon many aspects of the Abhidharma cosmology."¹

Verdict:  the weak claim is true.  Mindfulness programmes are a modest, fairly safe psychological treatment, roughly as good as the alternatives, and their benefits for healthy people and schoolchildren have been oversold.  The strong claim is not supported.  No trial has tested rebirth, karma or a mind independent of the brain, and the few Buddhist physical claims that could be tested, from Meru to post-mortem meditation, failed or went unconfirmed.  The "science" in "Buddhism and science" has changed its meaning every generation since 1881;  what has stayed constant is a practice with real but ordinary effects and a doctrine that the practice cannot reach.

Part 1: The claim, in its advocates' words

The Dalai Lama's rule

The most quoted statement of the claim is the Dalai Lama's, from The Universe in a Single Atom (2005).  Its context matters.  He recalls an American woman who warned him that "science is the "killer" of religion," and replies that he has nothing to fear:  "My confidence in venturing into science lies in my basic belief that as in science so in Buddhism, understanding the nature of reality is pursued by means of critical investigation: if scientific analysis were conclusively to demonstrate certain claims in Buddhism to be false, then we must accept the findings of science and abandon those claims."¹

Two qualifications follow, and both are his.  First, he grants that in physics Buddhism lags:  "there are many areas of traditional Buddhist thought where our explanations and theories are rudimentary when compared with those of modern science."¹  Second, he limits what counts as disproof.  Science must show a thing "to exist or to be non-existent (which is not the same as not finding it)."¹  He applies this at once to rebirth:  "just because science has not proven beyond the shadow of a doubt that beings take rebirth doesn't mean reincarnation isn't possible."¹

The second qualification decides how much the rule risks.  Science rarely proves a negative about an unobserved process.  Under this rule rebirth can be abandoned only if someone shows that it does not happen, and absence of evidence does not count.  A Christian apologist who said "the resurrection must be abandoned only if science shows it did not happen" would be told, rightly, that this is a very safe bet.  The rule is sincere, and the Dalai Lama does apply it to cosmology (Part 8).  But for the doctrines that matter most to Buddhists it puts almost nothing at stake.

Mind and Life, and "contemplative science"

The Dalai Lama's dialogues with scientists became an institution in 1987.  The Mind & Life Institute describes its origin as "a meeting of three visionaries":  the Dalai Lama, the scientist and philosopher Francisco Varela, and the lawyer and entrepreneur Adam Engle.²⁸  It claims credit for what followed:  "an entire field of contemplative science and research emerged—catalyzing scientific laboratories and applied research centers."²⁸  That claim is largely fair.  Richard Davidson, whose Madison laboratory produced many of the studies below, brought his team to Dharamsala to study meditators at the Dalai Lama's request.¹

The Institute is also an interested party.  It exists to show that "science and contemplative wisdom together could advance human flourishing,"²⁸ and many trials in this field are run by people who teach or practise what they test.  The authors of the main relapse-prevention review say so themselves:  "the constituent trials were largely conducted by proponents of MBCT."¹¹  Christian researchers who study prayer face the same charge, as this site's audit of the prayer trials records, and it is a fair one in both cases.

Meditation as an instrument

The strongest form of the claim comes from B. Alan Wallace, who spent fourteen years as a Buddhist monk ordained by the Dalai Lama³ and taught the three-month retreat in one of the studies below.  In Contemplative Science (2007) he argues that trained attention is a scientific instrument:  "Such focused attention is to the scientific investigation of mental phenomena what the telescope is to the scientific investigation of celestial phenomena."³  Buddhism, "like science, can be said to be primarily an empirical, not a philosophical or metaphysical approach to understanding reality."³

He then reports what the instrument has found:  "Memories are stored, in a manner of speaking, in the continuum of the substrate consciousness, which carries on from lifetime to lifetime. This conclusion is based on the experiences of highly trained contemplatives who have refined their attention in ways unknown to modern science."³  This is a claim that meditation confirms doctrine.  It is tested in Part 6.

Matthieu Ricard, who did "several years of research at the Pasteur Institute" before becoming a disciple of Tibetan teachers,²⁹ carries the doctrine into cosmology.  In The Quantum and the Lotus (2001) he says that "The idea of an absolute beginning of time and space is therefore flawed according to Buddhist thinking," and that "Consciousness has no start, either."²⁹

The naturalistic version

Robert Wright's Why Buddhism Is True (2017) is the most widely read recent defence, and it makes a much smaller claim than its title.  Wright argues that natural selection built minds that "mislead us, even enslave us," and that Buddhism diagnosed this delusion correctly.²¹  But he limits "true" at the outset:  "I'm not talking about the "supernatural" or more exotically metaphysical parts of Buddhism—reincarnation, for example—but rather about the naturalistic parts: ideas that fall squarely within modern psychology and philosophy."²¹  Near the end he writes:  "I don't believe in reincarnation or related notions of karma, and I don't bow before the statue of the Buddha upon entering the meditation hall, much less pray to him or to any Buddhist deities."²¹

Wright also corrects a Western picture of Buddhism that his own book might feed:  "most Asian Buddhists do believe in gods, though not an omnipotent creator God, and don't meditate."²¹  The religion practised by most Buddhists is not the one in the laboratory.

The clinical package

The practice that most trials test is Mindfulness-Based Stress Reduction, begun by Jon Kabat-Zinn at the University of Massachusetts Medical School in September 1979,²³ and its offshoot, Mindfulness-Based Cognitive Therapy.  Kabat-Zinn has been candid about what he built.  MBSR "was developed as one of a possibly infinite number of skillful means for bringing the dharma into mainstream settings."²³  He wanted his book "to articulate the dharma that underlies the curriculum, but without ever using the word 'Dharma' or invoking Buddhist thought or authority."²³  And:  "from the beginning of MBSR, I bent over backward to structure it and find ways to speak about it that avoided as much as possible the risk of it being seen as Buddhist, 'New Age,' 'Eastern Mysticism' or just plain 'flakey.'"²³

So the claim reaches the clinic in a form designed to carry Buddhist teaching without Buddhist vocabulary.  That matters twice over.  It explains why the trials say almost nothing about Buddhist doctrine:  the doctrine was taken out on purpose.  And it is the root of complaints from two directions:  that the dharma was lost in the translation (Part 6), and that it was carried into hospitals and schools under another name.

Part 2: Where the claim came from

A datable beginning

Before the nineteenth century no Buddhist argued that Buddhism agreed with European science, because the two had barely met, and when they did meet they disagreed.  Donald Lopez, a professor of Buddhist and Tibetan studies at Michigan and no critic of Buddhism, has written the history.  The first encounters, over Mount Meru, were "encounters not of compatibility but of conflict."⁶  At the Pānadura debate in Ceylon in 1873, between a Buddhist monk and a Wesleyan clergyman, the clergyman held up a globe against Mount Meru.  For the monk, Guṇānanda, "Science was a heliocentric cosmology ascribed to Newton and rejected as false."⁶

The turn came within a decade, and an American made it.  Henry Steel Olcott, a Civil War veteran who had co-founded the Theosophical Society, read an account of the Pānadura debate, sailed for India in 1879, and took the lay vows in Ceylon the next year.⁶  He then wrote A Buddhist Catechism (1881) for Sinhalese schools, "modeled on works used by the Christian missionaries."⁶  Its first edition already made the claim:

Q. Is this Buddhistical doctrine supported or denied by the teachings of modern science?  A. True science entirely supports this doctrine of cause and effect. Science teaches that man is the result of a law of development, from an imperfect and lower, to a higher and perfect, condition.  Q. What is this doctrine of science called?  A. Evolution.⁴

The doctrine was karma and rebirth, and the science was a progressive picture of evolution that biologists no longer hold.  By 1897 the Catechism had a whole part titled "Buddhism and Science."  It opens:  "Has Buddhism any right to be considered a scientific religion, or may it be classified as a "revealed" one?  Most emphatically it is not a revealed religion."⁵  It then asks which "great scientist has proved the existence of this aura" around the Buddha's body, and answers "The Baron von Reichenbach," adding that "not only all human beings, but animals, trees, plants and even stones have it."⁵  The Buddha's power to make others see illusions was "Very well known; it is familiar to all students of mesmerism and hypnotism."⁵

The Sinhalese reformer Anagārika Dharmapāla, for a time Olcott's closest associate, took the message to the World's Parliament of Religions in Chicago in 1893 and kept updating it.  In New York in 1925 he said the Buddha taught "a cosmogony which is in harmony with geology, astronomy, radioactivity and relativity."⁶  His 1924 essay "Buddhism, Science and Christianity" opened with "Every new discovery in the domain of Science helps us to appreciate the sublime teachings of the Buddha Gautama," and closed, after pages of attack on Christianity, "We condemn Christianity as a system utterly unsuited to the gentle spirit of the Aryan race."⁶  The Aryan language reflects the race science of the day, which Lopez treats at length.  No modern advocate holds it.  It is cited here only to show that the "science" in "Buddhism and science" has meant whatever was current.

The pattern

Lopez sums up a century of such claims:  "The claim that an itinerant teacher of Iron Age India understood the theory of relativity, quantum physics, or the big bang theory (each of which has been asserted) would seem to be preposterous. Yet such claims have been made for more than a century, substituting whatever is regarded as the most advanced scientific knowledge of the day as a component of the Buddha's enlightenment."⁶  He notes that Hindu reformers made parallel claims in the same decades, and that "the encounter with European colonialism played some role in the process."⁶  The sibling article Hindu Cosmology and Vedic Science Claims tests those.

The best-known endorsement is not genuine.  "If there is any religion that would cope with modern scientific needs, it would be Buddhism" circulates under Albert Einstein's name.  Lopez opens his book with it and then writes:  "But it seems that Einstein never said this."⁶

"Mindfulness" is a translation choice

The word at the centre of today's research has its own short history.  Lopez traces the English rendering of the Pali sati (Sanskrit smṛti, "memory") as "mindfulness" in a Buddhist sense to 1881, in Thomas Rhys Davids's Buddhist Suttas, after a Wesleyan missionary in 1845 had rendered it "meditation" and another "conscience."⁷  In the classical texts mindfulness is one factor among many, and its aim is not calm.  The mindfulness of the body ends in the charnel-ground contemplations, and "is intended to result in the understanding that the body is a collection of impure elements that arise and cease in rapid succession, utterly lacking any kind of permanent self."⁷

The modern research record therefore tests something that Buddhist tradition would not recognise as its goal.  Lopez puts it as a question:  "But is stress reduction a traditional goal of Buddhist meditation?"  His answer:  "A glimpse at any number of forms of Buddhist meditation, however, suggests that stress reduction is often not the aim."⁷  He also marks the change in the kind of claim.  The meditation research is "qualitatively different from the assertion that the Buddha understood the theory of relativity. The claim here is that Buddhist meditation works."⁷  That claim can be tested, and the next three parts test it.

How much is modern

Three things are old:  the practices of meditation, the Abhidharma analysis of mental states, and a strand of the early discourses that tells hearers to judge teachings by their fruits (Part 6).  Everything else in the "science" framing is modern.  The claim that Buddhism agrees with science dates from about 1880.  Two of its earliest and most influential voices were an American Theosophist and a Sinhalese reformer, both writing against Christian missions.  "Mindfulness" as a technical term in English dates from 1881, the clinical package from 1979, and the Mind and Life programme from 1987.  The historian David McMahan calls the movement "Buddhist modernism":  "forms of Buddhism that have emerged out of an engagement with the dominant cultural and intellectual forces of modernity."³⁰  He traces its science discourse to "two crises of legitimacy," the Victorian crisis of faith in the West and colonial rule with its missions in Asia, and judges that "perhaps no major tradition has attempted to ally itself with scientific discourse more boldly than Buddhism."³⁰

Part 3: What the clinical trials show

Why the comparison group decides the answer

A mindfulness course gives people eight weeks of attention, a hopeful teacher, a group, and a daily habit.  Any of those might help on its own.  A trial that compares mindfulness with a waiting list measures all of them together.  A trial that compares it with another credible programme, such as exercise or a health class, isolates what mindfulness adds.  Richard Davidson, the field's leading laboratory scientist, makes the same point in his popular book:  "Better: an "active control," where one group engages in an activity unlike meditation, one that they believe will benefit them, such as exercise."³¹  Most of the record below sorts itself by that one question.

The 2007 baseline

In 2007 the US Agency for Healthcare Research and Quality reviewed the whole field.  It found "813 predominantly poor-quality studies" and concluded:  "Firm conclusions on the effects of meditation practices in healthcare cannot be drawn based on the available evidence."³²  The trials since then are much better, and the reviews below rest on them.

The large reviews

Goyal and colleagues, 2014.  The same agency funded a second review, published in JAMA Internal Medicine.  The Johns Hopkins team screened 18,753 citations and kept only randomised trials with an active control:  47 trials with 3,320 participants.⁸  Against nonspecific active controls, mindfulness programmes had "moderate evidence to improve anxiety" with an effect size of 0.38 at eight weeks and 0.22 at three to six months, and depression 0.30 and 0.23.  The effect on pain was 0.33.⁸  An effect of 0.3 means the average treated person ends up better off than about 62% of the comparison group.³³  The authors judged these "small to moderate reductions."⁸  Against specific treatments the result was different:  "We found no evidence that meditation programs were better than any active treatment (drugs, exercise, other behavioral therapies)."⁸  Those comparison trials were small, averaging 37 people per group, and "none appeared adequately powered to assess noninferiority or equivalence."⁸  So the review could not say that mindfulness was as good as the alternatives either.

Goldberg and colleagues, 2018.  A review of 142 samples and 12,005 patients with psychiatric disorders, with Davidson among the authors, gives the most favourable picture.¹⁰  Effect sizes fell steadily as the comparison got stronger:  0.55 against no treatment, 0.35 against nonspecific active controls, 0.23 against specific active controls, and −0.004 against evidence-based treatments.¹⁰  Its summary:  "Mindfulness conditions did not differ from evidence-based treatments (d = −0.004)."¹⁰  The authors note that their small advantage over specific controls "diverge[s] from those of Goyal et al. (2014)."¹⁰

Galante and colleagues, 2021.  For healthy adults in workplaces, universities and communities, a Cambridge team pooled 136 trials with 11,605 participants.  Against doing nothing, mindfulness reduced anxiety, depression and distress by moderate amounts (standardised differences about 0.45 to 0.56) and improved well-being by 0.33.⁹  Against specific active controls, "there is no statistically significant evidence of MBPs' superiority."⁹  The authors' plain summary:  "mindfulness is neither better nor worse than other feel–good practices such as physical exercise, and RCTs in this field tend to be of poor quality, so we cannot be sure that our combined results represent the true effects."⁹

Goldberg, Sun and Davidson, 2022.  An umbrella review of 44 meta-analyses covering 336 trials and 30,483 people reached the same pattern from the top:  "Effects were typically smaller and less often statistically significant when compared to active controls."³⁴  Against evidence-based treatments, every estimate except smoking ran from d = −0.18 to 0.02.³⁴

Depression relapse: the strongest result

Mindfulness-based cognitive therapy was built for one job:  keeping people with recurrent depression from relapsing once they have recovered.  Here the evidence is strongest.  An individual-patient meta-analysis of nine trials, 1,258 patients, found that 38% relapsed within 60 weeks after the therapy against 49% without it, a hazard ratio of 0.69.¹¹  Against other active treatments the hazard ratio was 0.79, and against antidepressants 0.77.¹¹  The protective effect faded with time.¹¹

The largest single trial, PREVENT, put the question directly.  It randomised 424 British patients on maintenance antidepressants to stay on them or to take the therapy and taper off.  Over two years, 44% relapsed in the mindfulness group and 47% on antidepressants, a hazard ratio of 0.89 with a confidence interval from 0.67 to 1.18.  The authors concluded:  "We found no evidence that MBCT-TS is superior to maintenance antidepressant treatment for the prevention of depressive relapse."¹²  That is still a useful result.  A patient who wants to stop medication has an alternative that did about as well.

Pain

A trial of 342 adults with chronic low back pain compared mindfulness, cognitive behavioural therapy and usual care.  At 26 weeks, 61% of the mindfulness group, 58% of the therapy group and 44% of the usual-care group had a meaningful improvement in function.  For pain, the figures were 44%, 45% and 27%.³⁵  Mindfulness beat usual care and tied with the established therapy.

Schools

The MYRIAD trial was the field's largest test of universal mindfulness in schools.  It randomised 85 British secondary schools to ten lessons of mindfulness training or to their usual social and emotional teaching, and analysed 8,376 pupils at one year.  The differences on risk of depression, behaviour and well-being were 0.005, 0.02 and 0.02 standard deviations.  The authors' conclusion:  "Findings do not support the superiority of SBMT over TAU in promoting mental health in adolescence."¹³  Their confidence intervals were narrow enough for them to write that "we can rule out the possibility of important effects."¹³  An exploratory analysis found that pupils already at risk of mental-health problems had slightly worse depression and well-being scores after the training, differences the authors called "small and not clinically relevant," and they added:  "Universal SBMT is not recommended in this format in early adolescence."¹⁴  The teachers trained to deliver it had less burnout at first, but the effect had faded by one year.³⁶

Publication bias and hype

Two cautions apply across the record.  First, the published trials look too good.  A 2016 audit of 124 published mindfulness trials found that 108 (87%) reported a positive result, 1.6 times the number expected even if mindfulness worked as well as individual psychotherapy for depression.  Thirteen of 21 registered trials were still unpublished 30 months after they ended.³⁷  Second, the field's own researchers have warned against its publicity.  Fifteen of them, led by Nicholas Van Dam, wrote in 2018 that "Misinformation and poor methodology associated with past studies of mindfulness may lead public consumers to be harmed, misled, and disappointed."²⁰  Goleman and Davidson concede the same in their popular book:  "That left us surprised by how little we can say with certainty, given the great excitement (and, okay, hype) about meditation as a way to boost health."³¹

What replicates, what shrinks, what failed

The fair summary is that mindfulness training is a legitimate psychological treatment of ordinary strength.  It is not a placebo, and it is not a breakthrough.

Part 4: Brains, genes and the body

The claims that reached the public fastest were about the brain:  meditation thickens the cortex, grows grey matter, shrinks the amygdala, and produces extraordinary brain waves in monks.  These claims are the most often cited and the most fragile.

Thicker cortex, more grey matter

In 2005 Sara Lazar's group at Harvard compared brain scans of 20 Western insight meditators with 15 people who had never meditated, matched for age and sex.  Two small regions, in the prefrontal cortex and the right anterior insula, were thicker in the meditators.¹⁵  The design was cross-sectional:  one scan, two groups.  People who choose to meditate for nine years may differ in many ways from people who do not.

In 2011 Britta Hölzel and Lazar scanned people before and after an eight-week MBSR course and reported more grey-matter density in the hippocampus and other regions.  The title said mindfulness practice "leads to" the change.  The study had 16 people in the course and 17 on a waiting list, and it was not randomised.¹⁶  Its authors wrote:  "Finally, the current study employed a rather small sample size and replication is necessary."¹⁶

By 2014 there were 21 structural studies of about 300 meditators.  A meta-analysis by Kieran Fox and colleagues found differences in eight regions, but warned:  "Publication bias and methodological limitations are strong concerns, however."  Very large effects "were found exclusively in studies with relatively small sample sizes (n < 40)."³⁸

The replication

In 2022 Tammi Kral, Richard Davidson and colleagues at Wisconsin ran the test the earlier studies had called for.  They pooled two randomised trials:  218 meditation-naive adults scanned before and after eight weeks of MBSR, an active Health Enhancement Program built to match it in everything but mindfulness, or a waiting list.  They looked across the whole brain and in the specific regions from the earlier papers.  They found no difference:  "In the largest and most rigorously controlled study to date, we failed to replicate prior findings and found no evidence that MBSR produced neuroplastic changes compared to either control group, either at the whole-brain level or in regions of interest drawn from prior MBSR studies."²  Their verdict on the earlier work:  "Given the low sample sizes of prior work and the larger samples and lack of replication in the current study, there is a possibility that prior results suffered from inflated effect sizes and low positive predictive value."²  And on their own:  "These null effects, in conjunction with our large sample size and rigorous matched comparison condition, allow us to conclude that MBSR does not differentially affect brain structure."²

This was Davidson's own laboratory, the most prominent in contemplative neuroscience, publishing against a claim its allies had made famous.  It deserves credit.  The limit of the result is also real:  it tested an eight-week course, not decades of monastic practice.  The cross-sectional differences in long-term meditators remain unexplained, and they may reflect practice, self-selection or both.

Gamma waves in monks

In 2004 Antoine Lutz, Matthieu Ricard, Davidson and others recorded EEG from eight long-term Tibetan practitioners and ten student volunteers during a compassion meditation.  The practitioners had trained for "10,000 to 50,000 h over time periods ranging from 15 to 40 years."  Their high-frequency gamma activity was much larger, ">30-fold greater" by the authors' adjusted measure.³⁹  The finding is real for those eight people on that day.  What caused it is uncertain, and the authors list why:  "Our practitioners and control subjects differed in many respects, including age, culture of origin, and first language, and they likely differed in many more respects, including diet and sleep."³⁹  The practitioners averaged 49 years of age and the students 21.  Ricard, who carries the doctrine into cosmology above, is a co-author.

Immune function and telomerase

A 2003 trial randomised 25 employees to MBSR and 16 to a waiting list, then gave all of them a flu vaccine.  The meditators showed more left-sided frontal brain activity and a larger rise in antibodies, with p-values just under 0.05.  The authors noted:  "First, there was a relatively small number of subjects who participated and this limited our statistical power."⁴⁰

Telomerase, the enzyme that maintains the ends of chromosomes, became a popular marker after a 2011 study of a three-month retreat taught by B. Alan Wallace.  Retreatants had higher telomerase than a waiting-list group.  But the enzyme was measured only once, after the retreat, so no change was observed;  only 17 retreatants and 25 controls entered the analysis;  and the result, F(1,36) = 3.01, was called significant under one-tailed tests.⁴¹  Two-tailed, it would be about p = 0.09.⁴²  The claim that meditation slows cellular ageing rests on evidence of this kind.

A claim from the tradition, tested

One Buddhist claim about the body is old, specific and testable.  The Dalai Lama writes that his teacher Ling Rinpoche "remained in the clear light of death for thirteen days; although he was clinically dead and had stopped breathing, he stayed in the meditation posture and his body showed no sign of decomposition."  He asked scientists to study the state, called tukdam:  "When Richard Davidson's team came to Dharamsala, they were keen to do some experiments on this phenomenon, but when they were there — I am not sure whether I should say fortunately or unfortunately— no meditators died."¹

Davidson's team kept at it.  Between 2013 and 2018 they were notified of 13 cases in India and recorded EEG from 13 practitioners in apparent tukdam, against 14 living volunteers.¹⁷  The result:  "No recognizable EEG waveforms were discernable in any of these tukdam cases, thus we failed to find support for the hypothesis of residual brain activity following the cessation of cardiorespiratory function in tukdam cases recorded beyond 26 h postmortem."¹⁷  The authors are careful about what this does not show.  They did not try to test whether awareness persisted, "as asserted in the Tibetan Buddhist tradition," and they did not measure the rate of decomposition, which they call difficult to operationalise.¹⁷  So the claim that bodies in tukdam resist decay remains untested rather than refuted.  What was tested, a living brain sustaining the state, found nothing.

Part 5: Harms

Mindfulness is sold as gentle.  For most people it is.  But the traditions themselves warn about meditation's dangers, and modern research has only recently begun to count them.

Jared Lindahl, Willoughby Britton and colleagues at Brown University interviewed 60 Western Buddhist practitioners, 20 each from Theravāda, Zen and Tibetan lineages, most of them meditation teachers, plus 32 teachers and clinicians.¹⁸  The sample was chosen for people who had difficulties, so its percentages are not rates for meditators in general.  Within it, "The majority of the sample (73%) indicated moderate to severe impairment in at least one domain, with 17% reporting suicidality, and 17% requiring inpatient hospitalization."  "The median duration of symptoms was 1–3 years, ranging from a few days to more than 10 years."¹⁸  The paper notes that the traditions have names for these states:  "Zen sickness" in Zen, and the "corruptions of insight" in Theravāda.¹⁸

A second study by the same group interviewed 96 people after an eight-week mindfulness-based cognitive therapy course.  Fifty-eight percent reported at least one unpleasant meditation-related effect, 37% one that hurt their functioning, and 6% to 14% effects that lasted more than a month or more than a day, depending on the cut-off.  The authors conclude that these rates are "similar to other psychological treatments."⁴³  Open-ended questions of the kind most trials use found only about a third of the people with side effects.⁴³

A 2020 systematic review by Miguel Farias and colleagues pooled 83 studies with 6,703 meditators.  Adverse events were reported in 8.3% of participants overall, 3.7% in experimental studies and 33.2% in observational ones;  the most common were anxiety and depression.  The authors concluded that such events are "not uncommon, and may occur in individuals with no previous history of mental health problems."¹⁹

That is the honest frame.  Meditation is a psychological intervention with side effects comparable to psychotherapy's.  The trials mostly did not look:  Van Dam and colleagues report that "less than 25% of meditation trials actively assess AEs."²⁰  Goyal's 47 trials included nine that reported on harms at all.⁸

Part 6: What the record can and cannot show

The practice is not the doctrine

Suppose every trial above had come out twice as strong.  It would show that a set of attention exercises reduces anxiety and depressive relapse.  It would not show that a stream of consciousness passes from one body to the next, that deeds ripen across lives, that there is no self, or that nirvana is real.  Those are claims about what happens after death and about the nature of persons.  None of the trials measured them, and most could not.

The advocates mostly concede this.  Wright excludes rebirth and karma from what he defends.²¹  Kabat-Zinn removed the doctrine from the clinical package on purpose.²³  The Dalai Lama keeps rebirth while granting that science has not shown it.¹  David McMahan describes how modernist teachers handle the problem:  "many modernist Buddhist teachers present the wheel of rebirth, the traditional doctrine of the various realms into which beings are reborn, neither as obsolete nor as literal cosmology but as a psychological reality, with each realm representing a state of mind."³⁰  A doctrine reinterpreted as a state of mind can no longer be tested as a claim about the world, which is the point of reinterpreting it.

The philosophy of non-self, and whether rebirth without a self is coherent, belongs to the sibling article Non-Self, Rebirth, and Nirvana.  The best empirical case for rebirth, children who seem to remember past lives, is audited in Reincarnation and Past-Life Memories.  Neither case rests on meditation research.

Is meditation an instrument?

Wallace's stronger claim deserves a direct answer, because it is the one place where meditation is said to confirm doctrine.  A telescope earned trust in two ways.  Different observers pointed different telescopes at Jupiter and saw the same four moons.  And the telescope's readings could be checked by other means:  the moons' orbits predicted eclipses that anyone could time.

Contemplative reports have not yet met either test.  The reports Wallace cites come from contemplatives trained within the doctrine the reports confirm.  And trained contemplatives do not agree about what deep meditation shows about the self, even within Buddhism:  a "substrate consciousness" that carries memories between lives sits awkwardly beside the classical teaching that no continuing self exists at all.  The sibling article Non-Self, Rebirth, and Nirvana sets out those disputes.  Where a contemplative claim has been checked against a third-person measurement, the tukdam study, the measurement found nothing.¹⁷  This does not prove the contemplatives wrong.  It means their reports are testimony, to be weighed as testimony, and not yet observations from a calibrated instrument.

The same rule binds Christians.  A Christian mystic's report of union with God is also testimony, shaped by the mystic's training.  Christians who cite religious experience as evidence have to meet the same standard of independent agreement and outside checks, and the site's article on miracles and religious experience treats that question.

What the Buddha's "test it yourself" covers

Modern Buddhists often quote the Kālāma Sutta as a charter for empirical inquiry.  Olcott's Catechism cited it as the discourse in which "the Buddha tells us not to believe in an alleged revelation without testing it by one's reason and experience."⁵  The passage reads:  "Please, Kālāmas, don't go by oral transmission, don't go by lineage, don't go by testament, don't go by canonical authority, don't rely on logic, don't rely on inference."⁴⁴  But the next sentence names what the Kālāmas are to test:  "'These things are unskillful, blameworthy, criticized by sensible people, and when you undertake them, they lead to harm and suffering', then you should give them up."⁴⁴  The test is about conduct:  does greed lead to harm?  It is not a method for checking cosmology or rebirth.  The discourse even closes with four "consolations" that are explicitly agnostic about the next life:  "'If it turns out there is another world, and good and bad deeds have a result'" and "'If it turns out there is no other world, and good and bad deeds don't have a result, then in this very life I'll keep myself free of enmity and ill will, untroubled and happy.'"⁴⁴  Read in full, the text supports the weak claim, that the practice can be judged by its fruits, and says nothing for the strong one.

Buddhist objections to the secular package

The sharpest critics of clinical mindfulness are Buddhists.  In 2013 Ronald Purser, a professor of management at San Francisco State University and an ordained teacher in the Korean Buddhist Taego Order, wrote with David Loy:  "Uncoupling mindfulness from its ethical and religious Buddhist context is understandable as an expedient move to make such training a viable product on the open market."  They warned that "decontextualizing mindfulness from its original liberative and transformative purpose, as well as its foundation in social ethics, amounts to a Faustian bargain," and noted that "Buddhists differentiate between Right Mindfulness (samma sati) and Wrong Mindfulness (miccha sati)."⁴⁵  In a 2019 essay adapted from his book McMindfulness, Purser wrote that the practice, "Although derived from Buddhism," has been "stripped of the teachings on ethics that accompanied it, as well as the liberating aim of dissolving attachment to a false sense of self."²²

Lopez frames the same worry in Buddhist terms.  Buddhist scriptures predict a gradual disappearance of the dharma, and he asks:  "If the practice of the four foundations of mindfulness is reduced to stress reduction and the visualization of Vajrayoginī is understood as a technique for raising one's body temperature, where do we stand in the process of the disappearance of the dharma?"⁶  McMahan notes a deeper shift.  When Daniel Goleman said of meditation research that "it's beginning to look like the Buddha just might have had it right," the remark implied that the scientist, not the sutra, now has the final word, and "honest scientific investigation may find that in some respects the Buddha, as it were, got it wrong."³⁰

These objections cut against the popular claim from inside Buddhism.  If the trials test a practice stripped of Buddhist ethics and aims, their success cannot vindicate Buddhism.  If the trials test Buddhism, then Buddhism has accepted a judge that can rule against it.

Part 7: The same standard, applied to Christian practice

A Christian reader may enjoy watching "Buddhism is science" shrink.  The same test has to be run on Christian claims, and it gives the same kind of answer.

Church attendance and health

The best-known Christian-friendly finding comes from Tyler VanderWeele's group at Harvard.  In the Nurses' Health Study, 74,534 women were followed from 1996 to 2012, during which 13,537 died.  Compared with women who never attended religious services, those who attended more than once a week had a 33% lower death rate:  "Hazard Ratio [HR]=0.67, 95% confidence interval [CI]: 0.62–0.71."²⁴  A companion paper found that women who attended weekly or more had about one-sixth the suicide rate of those who never attended, a hazard ratio of 0.16, though from only 36 suicides in all.⁴⁶  Earlier meta-analyses point the same way.  Michael McCullough's 2000 review of 42 samples found an odds ratio of survival of 1.29 for religious involvement.⁴⁷  A 2009 review by Yoichi Chida and colleagues found a hazard ratio of 0.82 in healthy populations, but 0.98, no effect, in people who were already ill.⁴⁸

These are observational studies.  Nobody can randomise people to church, and the authors say so:  "Clearly a randomized trial of religious service attendance is neither ethical nor feasible. Our study is an observational study. Although we adjusted for major confounders for the association between religious service attendance and mortality, the results may still be subject to unmeasured confounders and residual confounding."²⁴  The study did more than most to meet the objection.  It calculated that a hidden confounder would need to raise both attendance and survival by 2.35-fold to explain away the result.²⁴  VanderWeele judges that "the evidence that some of the association is causal is quite strong."⁴⁹  Critics such as Richard Sloan at Columbia have argued that much of the older literature was weak or misrepresented:  in a 2002 audit, only 17% of 266 articles on religion indexed in one year were even relevant to claims of health benefit.⁵⁰  A 2021 study by the Finnish psychologist Markus Jokela tested one confounder directly, family background, by comparing siblings in several cohorts.  The link between attendance and lower psychological distress "was attenuated by almost half in the sibling analysis," though the link with well-being held.⁵¹

On design, then, the Buddhist-derived practice has the better evidence.  Mindfulness has been tested in hundreds of randomised trials;  church attendance cannot be.  A fair reader should hold the attendance findings more loosely than the relapse-prevention findings for mindfulness, not less.

Christian practices in randomised trials

Where Christian practices have been randomised, the results look like the mindfulness results:  ordinary or null.

• Distant intercessory prayer. The large blinded trials of strangers praying for patients found no effect. This site's audit of them is Does Prayer Heal?. That article also argues that such trials test a mechanical model of prayer that Scripture does not teach. A Buddhist could say the same about MBSR, and the article's argument has to be granted to both or to neither.
• Religiously integrated therapy. Harold Koenig and colleagues randomised 132 depressed, chronically ill patients to conventional cognitive behavioural therapy or a religiously integrated version of it. At 12 weeks there was "no significant difference in outcome between the two groups (B = 0.33; SE, 1.80; p = 0.86)." The religious version did slightly better among the more religious patients, a pilot-sized interaction at p = 0.048.⁵²
• Centering Prayer. A pre-registered online trial randomised 702 Christians to 28 days of Centering Prayer with Christian framing, the same practice without it, or a passive control. "We found no group differences between the conditions on our primary outcome measure of well-being at one-week post-intervention."⁵³ A Christian contemplative practice, tested the way mindfulness is tested, did no better than a passive control of daily reflection, which the authors concede may have been "stronger than expected."⁵³

What the comparison shows

The rule from Part 6 applies without change.  That church attendance goes with longer life is evidence about attendance:  community, purpose, fewer cigarettes, a reason not to despair.  VanderWeele notes that many of the mechanisms "are seemingly not distinctively religious and could be operative in other contexts as well."⁴⁹  It is not evidence that Jesus rose from the dead, any more than lower relapse after MBCT is evidence of rebirth.  A Christian apologist who cites the Nurses' Health Study as proof of Christianity is making the same move as an advocate who cites cortical thickness as proof of the dharma, on weaker data.

The two religions stake different things on evidence.  Paul tied Christianity to one historical event:  "If Christ has not been raised, your faith is vain; you are still in your sins."⁵⁴  That claim is tested by historical evidence, which this site weighs in The Historical Case for the Resurrection, not by health outcomes.  Buddhism, as the Kālāma discourse shows, invites a pragmatic test of conduct and leaves the next life open.  Each religion should be judged on the claim it actually makes.  On the pragmatic test, both have practices that help people a little.  On the metaphysical claims, neither the meditation trials nor the attendance studies decide anything.

Part 8: Mount Meru and the realms

What the texts say

Buddhist texts describe a world with a measurable shape.  In a discourse of the Pali canon, the Buddha teaches impermanence by describing the end of the world, and he begins with its central mountain:  "Sineru, the king of mountains, is 84,000 leagues long and 84,000 leagues wide. It sinks 84,000 leagues below the ocean and rises 84,000 leagues above it."²⁵  Seven suns then appear one after another until the oceans dry and the mountain burns.  The point of the discourse is moral:  "Mendicants, conditions are impermanent."²⁵

The fourth- or fifth-century scholar Vasubandhu systematised this picture in the Abhidharmakośa, the cosmology that "The various Buddhist traditions of Asia generally subscribe to," in Lopez's words.⁶  Its commentary gives the numbers.  Meru stands in water "to a depth of eighty thousand yojanas" and "rises out of the water for the same number of yojanas," and the mountains "are as wide as they extend out of the water."²⁶  The sun and moon rest "Upon the wind" and "revolve around Meru as if transported by a whirlpool";  they "move at a level with the summit of Yugandhara," the first ring of mountains, at half Meru's height;  "The disk of the moon is of fifty yojanas; the disk of the sun is of fifty-one yojanas."²⁶  The text defines its unit too:  twenty-four finger-widths make a cubit, four cubits a bow, five hundred bows a krośa, and "eight krośas make what is called one yojana."²⁶  Lopez summarises:  "the human realm that Buddhist texts describe is a flat earth, or perhaps more accurately a flat ocean, its waters contained by a ring of iron mountains."  In this scheme the sky is blue over our continent "because the southern face of Mount Meru is made of lapis."⁶

The numbers, priced

Because the text defines the yojana, its claims can be checked.  At a cubit of about 45 centimetres, a yojana is about 7.2 kilometres.  On that unit:

The longer nineteenth-century Sri Lankan yojana of about 16 miles makes Meru more than five times as high as the Moon.²⁷  One number comes out close:  the Moon's height on the short unit is three-quarters of its real distance.  And the near-equal disks of sun and moon at the same height match what the eye sees, since the two look the same size in the sky.  The scheme is a careful description of the sky as it appears from the ground, wrapped in numbers far too large for the mountain and far too small for the sun.

How Buddhists have handled it

Buddhists met this problem as soon as European astronomy reached Asia, and they answered it in the two ways Christians answered questions about Genesis.

The first answer is accommodation.  The Japanese scholar Tominaga Nakamoto wrote in the 1730s:  "The teachings about Mount Sumeru were all handed down by brahmans. Though Śākyamuni used them to expound the Way, they are to be regarded as cosmic theory."⁶  The Buddha used the cosmology of his day as a backdrop for teaching liberation and did not vouch for it.  The Seven Suns discourse fits that reading, since its point is impermanence, not geography.

The second answer is defence.  The monk Entsū Fumon (1755–1834) wrote five volumes defending the Meru cosmology against European astronomy,⁶ and at Pānadura in 1873 the Buddhist side rejected Newton.⁶

The Dalai Lama takes the first.  "By the age of twenty, when I began my systematic study of the texts that discuss Abhidharma cosmology, I knew that the world was round."¹  He writes that the Abhidharma's figures "are flatly contradicted by the empirical evidence of modern astronomy," cites a Buddhist dictum "that to uphold a tenet that contradicts reason is to undermine one's credibility; to contradict empirical evidence is a still greater fallacy," and concludes:  "My own view is that Buddhism must abandon many aspects of the Abhidharma cosmology."¹  He adds that Vasubandhu "was presenting systematically the variety of cosmological speculations that were then current in India," and that the cosmos, "the container," is "secondary to the account of the nature and origins of sentient beings, who are "the contained.""¹

This is the rule of Part 1 actually applied, and it should be credited.  Two things limit it.  First, the measurements are not only in Vasubandhu.  The Seven Suns discourse puts them in the Buddha's own mouth, so assigning the cosmology to later scholastics does not fully settle the matter;  it requires the accommodation reading of the Buddha as well.  Second, the rule is applied where the evidence is overwhelming and the doctrine is, by the Dalai Lama's own account, secondary.  McMahan observes that "There has, however, been no large-scale jettisoning of Buddhist doctrine," and that the Dalai Lama "still maintains many traditional beliefs, such as karma and rebirth."³⁰

The Christian parallel

Christians face the same question about the sky of Genesis 1, and this site's article Ancient Cosmology and Genesis concludes that the author probably pictured a solid sky while the text never asserts one.  John Calvin's answer, that Moses described "the visible form of the world" and that anyone who "would learn astronomy" should "go elsewhere,"⁵⁵ is the same move as Tominaga's.  It is legitimate for both or for neither.  Two differences are worth stating without pressing them too hard.  Genesis gives no measurements, while the Buddhist texts give exact dimensions in a defined unit, which makes them easier to test and harder to read as poetry.  And the Christian claim that matters most, the resurrection, is a historical claim about one event, not a cosmology.

A beginningless universe

One cosmological doctrine survives in modern Buddhist teaching:  the universe has no beginning.  Ricard says that "The idea of an absolute beginning of time and space is therefore flawed according to Buddhist thinking."²⁹  Standard cosmology traces the observable universe back about 13.8 billion years to a hot, dense state.  Whether that state was an absolute beginning is not settled by observation, and cyclic and eternal models remain under discussion.  Christians who cite the Big Bang as proof of creation should not overstate it either.  The site's article on the Big Bang separates what the evidence for it does and does not establish.  On the question Ricard raises, present data exclude neither a beginning nor a beginningless cycle, so the Buddhist doctrine is not refuted here, and the Christian doctrine of creation is not proved.

Objections and responses

1.  "Mindfulness is not Buddhism, so its trials say nothing about Buddhism either way."  Partly true, and the article agrees that the trials cannot test Buddhist doctrine.  But the popular claim that "science confirms Buddhism" rests largely on these trials, and Kabat-Zinn designed MBSR as a way of "bringing the dharma into mainstream settings."²³  If the trials do not bear on Buddhism, the popular claim loses its main evidence.  If they do, they show a modest practice, not a confirmed metaphysics.

2.  "Equal to antidepressants and CBT is a success, not a failure."  Agreed.  A treatment with no drug side effects that prevents depressive relapse about as well as maintenance medication is useful, and the article says so.  The objection is to the marketing, not the method:  equal to existing treatments is not evidence of a uniquely deep science of mind.

3.  "Active controls are unfair, because they borrow ingredients from meditation."  Some do.  But the active controls in the strongest trials were built to match everything except mindfulness:  time, attention, group support, expectation.  If mindfulness adds little beyond those, the honest conclusion is that much of its benefit comes from them.  That is also why Christian practices should not be credited with health effects that a secular club would produce.

4.  "Kral 2022 studied beginners.  Long-term meditators really do differ."  Possibly.  The cross-sectional differences in long-term meditators may be real, but people who meditate for decades differ from others in many ways, and Lutz's monks differed from his controls in age, culture, language, diet and sleep, as the authors said.³⁹  Until someone follows people into long practice with a matched comparison, the cause of those differences is unknown.  The claim that eight weeks of MBSR reshapes the brain, the one that reached the public, did not replicate.

5.  "The Dalai Lama keeps rebirth because science has not disproved it.  That is ordinary agnosticism, not special pleading."  It is consistent, and Christians make the same move about miracles.  The point is only that a rule which waits for proof of non-existence puts very little at risk, so it should not be presented as Buddhism submitting to science.  Where the Dalai Lama did apply the rule, to Meru, he gave up a teaching the canon puts in the Buddha's mouth, and that is to his credit.

6.  "Meru is symbolic, not physical."  Tominaga said so in the eighteenth century, and Christians read parts of Genesis the same way.  But the Abhidharmakośa gives dimensions in a defined unit, sets the sun's size and height, and, in Lopez's summary of it, explains the colour of the sky by the mountain's lapis face.^{26 6}  For most of Buddhist history the scheme was read as geography, and Entsū Fumon defended it as such against Western astronomy.⁶  The symbolic reading is a legitimate later accommodation, not the original sense.

7.  "Christian health studies are observational too, so the article is harder on Buddhism."  The article says the reverse:  the mindfulness evidence is mostly randomised, the attendance evidence observational, and the mindfulness record is the stronger.  Both records are held to the same rule, that a practice's benefits do not prove its religion's doctrines.

8.  "The adverse-effects studies are biased samples of troubled meditators."  Lindahl's sample was selected for difficulties, and the article says so.  Britton's MBCT sample and Farias's review were not selected that way, and they found unpleasant effects in a minority and lasting harm in a few percent, similar to psychotherapy.⁴³  Meditation is not dangerous for most people.  It is an intervention with side effects, like other interventions.

9.  "This is a Christian site attacking another religion."  The test applied here is the one the Dalai Lama proposed, and every critic quoted on Buddhism's side is a Buddhist or a scholar of Buddhism:  Lopez, McMahan, Purser and Loy.  The same test is applied to Christian practices in Part 7, and they do no better.

Issues and unanswered questions

1. Locators from author manuscripts. Several papers (Goyal 2014, Kuyken 2016, Goldberg 2018 and 2022, Van Dam 2018, Lazar 2005, Hölzel 2011, Cherkin 2016, Li 2016, VanderWeele 2016) were read in their author manuscripts, so their page numbers are manuscript pages, not journal pages. The wording was checked against those copies.
2. The PREVENT correction. The Lancet corrected the online PREVENT paper in September 2016. The copy read here is the April 2015 version, and the content of the correction was not found. The headline figures (44% against 47%, HR 0.89) should be checked against the corrected version.
3. Two reviews disagree on specific active controls. Goyal 2014 found no advantage over specific active treatments; Goldberg 2018 and 2022 found a small one (d about 0.13 to 0.26) for depression, substance use and smoking. The reviews define "specific active control" differently, and Goldberg's group includes Davidson, so the two are not independent. Which is right is open.
4. Long-term meditators. No randomised or prospective matched study has followed people into thousands of hours of practice. The cross-sectional differences in Lutz 2004 and Lazar 2005 are unexplained.
5. Telomeres. The 2014 meta-analysis of four trials (190 participants) found a pooled effect on telomerase of d = 0.46,⁵⁶ and a 2020 meta-analysis addressed telomere length. Neither full text was obtained, so this article does not lean on them either way.
6. Tukdam and decomposition. The 2021 study tested brain activity, not decay. A study measuring decomposition in tukdam bodies against matched controls in the same climate has not been published.
7. Christian contemplative trials. Only one well-powered randomised trial of a Christian contemplative practice (Rohde 2024) was found. A 2018 trial of an internet contemplative prayer programme by Joshua Knabb and colleagues was identified but not read.
8. Meru's unit. The conversion uses a cubit of about 45 cm. Estimates of the yojana in the literature vary, and the longer Sri Lankan unit makes the mismatch larger. No plausible unit makes the numbers fit.

Verdict

The weak claim, that Buddhist-derived meditation helps people, is true in a modest form.  Mindfulness programmes reduce anxiety, depression and pain by small to moderate amounts against no treatment, about as much as other active treatments, and they prevent depressive relapse about as well as maintenance antidepressants.  Their benefits for healthy adults are no greater than other "feel–good practices," and in schools they did nothing measurable.  They have side effects at rates like psychotherapy's.

The strong claim, that science confirms Buddhism, is not supported.  The brain-change findings that made the claim famous did not survive a large randomised replication run by the field's leading laboratory.  The one tested Buddhist physiological claim, persisting brain activity in tukdam, found nothing.  The cosmology of Mount Meru, which the canon puts in the Buddha's mouth, is wrong by factors of hundreds to thousands, and the Dalai Lama says Buddhism must abandon it.  The doctrines that matter most, rebirth, karma and non-self, have not been tested by any of this research and mostly cannot be.

The framing itself has a history.  It began around 1880 with an American Theosophist's catechism that equated karma with evolution and the Buddha's halo with Reichenbach's aura, and it has borrowed whatever science was current ever since.  Christian practices, tested the same way, show the same pattern:  real but ordinary benefits that say nothing about whether the religion is true.  The question of truth is settled elsewhere, by history for Christianity and by argument and testimony for Buddhism, and the sibling articles take it up there.

Sources