When Meditation Makes Things Worse

Meditation is sold as something that cannot hurt you — at worst, twenty minutes spent with your eyes closed for nothing. For the great majority of people who practice it, that is true, and it is worth saying first, because nothing that follows is meant to frighten anyone. For a minority it is not true. Some people find in meditation an anxiety or a panic they did not have before, a sense of unreality about the world or about themselves, traumatic memories returning with a vividness nobody asked for, whole nights without sleep; and sometimes it does not stay on the cushion but lasts for weeks, months or longer. This site has a piece called When Meditation Goes Well, Everything Goes Well. This is the other one: the footnote the marketing of meditation almost always leaves out.

What the studies counted

For decades, research on meditation measured very little besides its benefits. One of the studies that began to change that came from Jared Lindahl, Willoughby Britton and their colleagues at Brown University; it was published in PLOS ONE in 2017 under the title The Varieties of Contemplative Experience. They interviewed sixty Western meditators from the Theravāda, Zen and Tibetan traditions who had run into difficulties in their practice, along with thirty-two teachers and clinicians, and sorted what they heard into fifty-nine kinds of experience across seven domains: thought, perception, emotion, the body, motivation, the sense of self and relationships with others. Fear, anxiety and panic; traumatic memories coming back; an oversensitivity to light and sound; insomnia; involuntary movements; the impression that nothing is quite real. The study did not measure how many people go through this — it had gone looking for people who had — but it showed something a percentage cannot: the typical duration of those difficulties was one to three years.

How many people, then. The honest answer is that it depends on what is being counted, and the term itself deserves a moment: “adverse effect” can mean an anxious afternoon that passes on its own or months of being unable to work, and studies that lump the two together, or ask about only one, arrive at very different numbers. A systematic review led by Miguel Farias and published in 2020 pooled eighty-three studies with more than six thousand participants and found adverse effects in 8.3 percent of meditators — but the figure was 3.7 percent in experimental trials and 33.2 percent in observational studies, a gap that says as much about how the question was asked as about what happened. The most common were anxiety, depression and cognitive anomalies. A year later, Simon Goldberg and colleagues surveyed a general-population sample of American adults who had ever meditated, this time symptom by symptom. Half reported at least one unpleasant effect; about one in ten said it had impaired their daily life to some degree; and 1.2 percent said the impairment had lasted a month or longer.

One more finding deserves as much attention as the rest. In 2021 Britton’s team followed ninety-six participants through three versions of an eight-week course of mindfulness-based cognitive therapy and found lasting bad effects in 6 to 14 percent of them — and concluded, too, that the rates of passing distress and of negative effects on functioning were similar to those of other psychological treatments. Psychotherapy has adverse effects as well, and nobody concludes from that that it is dangerous. The conclusion, rightly, is that it has to be done well.

Why it stays invisible

If those are the numbers, it is fair to wonder why almost nobody has heard about it. There are several reasons, and none of them requires bad faith. The first is that people who struggle tend to stop meditating without telling anyone, and so they vanish from exactly the places where it might have been noticed — the group, the course, the app. The second is that for a long time nobody asked. By the count of a 2018 review, more than 80 percent of randomized trials of mindfulness programs contained no statement about adverse effects at all, and what nobody asks about does not turn up in anyone’s statistics. The third is one Britton’s team points out: research participants, like patients in therapy, rarely volunteer a negative reaction, out of a wish to please the therapist or the researcher, so whoever waits to be told learns of far less than actually happens. And the fourth is the simplest. Meditation is also a product now, and nobody prints on the box the thing that might stop you from buying it.

A fortnight at Vesālī

The tradition did not need Brown University to find this out. One of the hardest episodes in the Pali Canon is told in the Saṃyutta Nikāya (SN 54.9) and, at greater length, in the Vinaya, the monastic code, where it becomes the origin story of the rule against a monk taking a human life. The Buddha was staying near Vesālī, in the Great Wood, and had spoken to the monks in many ways about the contemplation of the body’s unattractiveness, asubha, praising its practice. Then he announced that he would go into solitary retreat for a fortnight, and that no one was to come to him except the one who brought his food.

The monks devoted themselves to that contemplation in all its forms, and became, in the sutta’s words, “horrified, repelled, and disgusted with this body” (quotations from the suttas here are from Bhikkhu Sujato’s translations, on SuttaCentral). The Vinaya compares them to a young man or woman fond of ornaments, hair freshly washed, who finds the carcass of a snake or a dog hung around their neck. Many of them took their own lives. When the Buddha came out of retreat he asked Ānanda why the community seemed so diminished, and Ānanda told him what had happened and asked him to teach another way.

What the Buddha did next is the point here. He did not stop teaching meditation. He had every monk living in the neighborhood of Vesālī gathered in the assembly hall and taught them another kind, mindfulness of breathing, ānāpānasati:

Mendicants, when this immersion due to mindfulness of breathing is developed and cultivated it’s peaceful and sublime, a delectable and blissful meditation. And it disperses and settles unskillful qualities on the spot whenever they arise. In the last month of summer, when the dust and dirt is stirred up, a large storm out of season disperses and settles it on the spot.

The episode is not an argument against contemplating the body, which still has its place in the canon and in the forest monasteries; the piece on the ten asubhas on this site opens by warning that the practice is not for everyone. It is something more uncomfortable and more useful: a record that a sound practice, taught by the best teacher there could be, can harm the wrong person, at the wrong dose, with no one nearby to keep watch. It is worth noticing when it happened — while the teacher was away. And it is worth noticing what the answer was. The medicine was not withdrawn. It was changed.

A great rain cloud breaks over a dry, dusty plain in ancient India, its first heavy rain pressing the dust back to the earth, with a few trees and thatched huts in the distance.

When the map makes things worse

There is a way in which the tradition itself can, without meaning to, make things worse. The Visuddhimagga, the great Theravāda manual of meditation composed by Buddhaghosa in the fifth century, describes the progress of insight, vipassanā, as a sequence of knowledges, and several of those in the middle of the path are dark: the knowledge of fear, of danger, of disenchantment. The manual illustrates them with harrowing images — a mother who watches two of her three sons beheaded and gives up all hope for the third — and a good part of modern insight meditation, particularly the lines that descend from Mahasi Sayadaw, teaches that map in detail.

A map like that has its uses; it is a comfort to know that others have crossed the same difficult stretch. It also carries an obvious risk, which Bhikkhu Anālayo set out in a 2019 paper on this very subject: making fear a stage of its own easily gives the impression that distress is an indispensable ingredient of progress. And if distress is progress, the sensible thing is to keep going — in exactly the cases where the sensible thing would be to stop. Anālayo also points out that nothing in the early discourses corresponds to the knowledge of fear as a stage of meditation; the scheme took shape in later texts.

The curious thing is that the Visuddhimagga is more careful than some of its readers. It stops to ask the obvious question about the knowledge of fear (in full, the knowledge of appearance as terror):

But does the knowledge of appearance as terror [itself] fear or does it not fear? It does not fear. For it is simply the mere judgment that past formations have ceased, present ones are ceasing, and future ones will cease.

It is, the manual goes on, like a man looking at three pits of glowing coals at a city gate, who is not afraid himself, since he only judges that whoever falls into them will suffer (Visuddhimagga XXI.32, in Ñāṇamoli’s translation). Seeing that nothing conditioned offers shelter is not the same thing as a panic attack, and a map that cannot help tell the two apart has become a hazard. Ajahn Chah, describing one of his most intense experiences in meditation — lying down to rest after a long sitting, he felt his body explode, and then the whole world dissolve — remarked that “for someone with no self-possession, that kind of experience might drive them insane” (as told in Stillness Flowing, Ajahn Jayasaro’s biography of him). What separates one case from the other, by his account, is not the experience. It is the ground a person is standing on when it comes.

Who is more exposed

No one can yet predict who will struggle, but certain factors come up again and again. Intensive retreats, with many days of silence and many hours of practice a day, have been linked to adverse effects in earlier research, and in Goldberg’s study the amount of practice was itself associated with more of them. Excessive effort, which feels like virtue and produces tension, is another — Soṇa’s mistake, walking his meditation path until his feet bled. A history of adversity in childhood went with a higher risk in the same study, as did symptoms of depression; meditation takes away the distractions a person may have been sheltering behind for years, and what is behind them is not always ready to be looked at. And there is practicing alone, with no one who knows this territory and can say in time “this is normal” or “this is not.”

What helps

When practice starts to do harm, the first advice is usually less, not more: shorter sittings, or none for a while. Opening the eyes. Moving — trading sitting for walking meditation, or for some simple work with the hands. Choosing an object that anchors rather than absorbs — the sounds of the room, the feet on the floor, the weight of the body in the chair — because for some people the breath itself, the most recommended object of all and the very one taught at Vesālī, is exactly what sets the anxiety off; what calmed those monks does not calm everyone. Eating, sleeping, going back to other people’s company and the ordinary routines of the day. None of this is a defeat. It is what the Buddha did at Vesālī: change the practice when the practice was doing harm.

The hare mascot sits on its teal cushion with its eyes open, turned toward an open window full of trees and morning light.

And talking to someone who knows. If there is one point on the path where having a teacher matters, it is this one: someone experienced in this territory can tell what belongs to the practice from what does not, and adjust it in time. When symptoms persist, worsen or begin to affect work, sleep or relationships, it has become a matter for a mental health professional, and one worth seeking out without waiting for it to pass. Cheetah House, a nonprofit founded by Willoughby Britton, offers information, consultations and support groups for people who have had difficult experiences with meditation. And if thoughts of self-harm appear at any point, that is no longer a question about meditation. It is an emergency, and it is handled as one, through emergency services or a crisis line; findahelpline.com lists the free ones in each country.

When the answer is not to meditate

On a silent meditation retreat I attended at Doi Suthep, the mountain above Chiang Mai, the monk in charge gave a Dhamma talk every morning, and one of those talks was on exactly this. It was about a person who, a few days into a retreat, had broken down — sweating and shouting. The person recovered, and once they had, they were advised not to meditate. What stood out was not only the story but where it was told: on a retreat, by the monk leading it, to a hall full of people who had come precisely to meditate.

It can sound like a sad conclusion, as if a door had closed on that person. The tradition does not see it that way. In the suttas the Buddha names three grounds for making merit — dāna, sīla and bhāvanā: “Giving, ethical conduct, and meditation” (AN 8.36; also Iti 60). Thais still use the formula in that form: than, sin, phawana. And the word translated “meditation” means, strictly, cultivation, making something grow, which is why the verse that closes the Itivuttaka discourse can apply it to all three: “Giving and moral conduct, / developing a mind of love: / having developed these / three things yielding happiness, / that astute one is reborn / in a happy, pleasing world.” Ajahn Chah, in a talk to his lay supporters on an observance night, made the same point in his own way: giving and keeping the precepts are bhāvanā too, because of the contentment a gift leaves in the one who gives and in the one who receives. “Those feelings didn’t exist before. The creation of those feelings is called cultivation [bhāvanā].”

It is worth not overcorrecting. The same discourse in the Aṅguttara goes on to say where giving and ethical conduct lead when meditation is missing: to rebirth among humans or, at best, among the gods, and no further; meditation is not an ornament the path can do without. But that it cannot be done without does not mean it is for today, or for this person, or in this form, and nothing in the texts suggests that someone who stops meditating on a teacher’s advice has left the path. They are still on it, only walking a different stretch of it.

The turtle mascot in its saffron robe walks slowly along a forest path at dawn, with mist among the trees behind it.

At Vesālī the Buddha did not stop teaching meditation; he taught a gentler kind. At Doi Suthep, for one person, the advice was none at all, at least for a while. They are the same answer. The practice is there to serve the person practicing, and when it begins to harm them, it is the practice that changes.

Scroll to Top