Matthieu Ricard · The habits of happiness
February 2004, Monterey, California. A Frenchman in maroon robes walks onto the TED stage and opens with a joke: the proof of globalization is that you can find Coca-Cola tins on top of Everest and a Buddhist monk in Monterey. He has just published a book on happiness, he says, and it has been attacked at home — one article ran with a line to the effect of "Don't impose on us the dirty work of happiness." His name is Matthieu Ricard. In 1972 he took a PhD in molecular genetics at the Pasteur Institute under the Nobel laureate François Jacob; the moment he had the degree he left for the Himalayas and stayed thirty years.
Then he starts taking a word apart. When we say "happiness" we usually mean pleasure, and pleasure is contingent on time, on its object, on place. "Beautiful chocolate cake: first serving is delicious, second one not so much, then we feel disgust." Sitting near a fire is wonderful until it starts to burn you. Pleasure uses itself up as you experience it, and it does not radiate outward — you can feel intense pleasure while people around you are suffering badly.
What he wants is something else. On the crest of a wave you are elated, at the bottom you hit solid rock — up, down, no depth. The high sea is different: the surface may be calm as a mirror or torn by storms, but the depth of the ocean is still there, unchanged. He calls that depth well-being, and defines it as "a deep sense of serenity and fulfillment" that pervades and underlies every emotional state. Which is why a person can be sad and not lose it.
How do you get there? By training, and his reason is plain: if anger and jealousy were the nature of mind, they would be there always, like a dye permeating the whole cloth — yet we are not always angry. Since every emotion is fleeting, there is room to change.
Then he brings in the science: the adult brain is not fixed; a violinist who has done 10,000 hours of practice shows large changes in the cortex controlling the fingers. So what about compassion, patience? Some meditators, he says, have gone into labs in Wisconsin and Berkeley with 20,000 to 40,000 hours of practice behind them — "real Olympic champions of mind training" — and he offers three pieces of evidence: a bell curve recorded under 256 electrodes on which a meditator of compassion is "totally out of the bell curve"; three and a half hours in an fMRI; and a startle experiment in which, over twenty years, nobody had ever failed to jump — while some meditators "are able not to move at all." His landing point is not the stunt: "This is not just a luxury. This is not a supplementary vitamin for the soul." We will spend fifteen years on an education and go jogging to stay good-looking, yet spend startlingly little time on the thing that determines the quality of every experience we have.
And afterwards? The big claim half held. A 2014 meta-analysis of 47 randomized trials and 3,515 participants found that eight-week mindfulness programs lower anxiety and depression by roughly 0.3 standard deviations — about what antidepressants do, and no better than any active control; on positive mood the verdict was "low evidence of no effect or insufficient evidence of any effect." Of the three exhibits from the stage: the gamma finding was partly replicated by other labs across three traditions; the horizontal axis of that bell curve — left-versus-right frontal activity — largely dissolved in a 2021 study that ran 388 people's data through 270 analyses and got 13 significant results, exactly what chance predicts; and the startle work was not published until 2012, by which time it had shrunk to a single-subject case study in which the crucial condition — the unanticipated blast — produced responses no different from 12 matched controls. As for the label the press gave him, "the happiest man in the world," he calls it "a nonsense idea."
Three minutes in, he repeats a line you would not expect a monk to relay. He has just written a book on happiness, he says, and it caused a controversy in France; someone wrote an article saying —
"Don't impose on us the dirty work of happiness."— Matthieu Ricard, relaying the French reaction, TED, 2004
The room laughs. He keeps going on his opponents' behalf, with a kind of cheerful fairness: "We don't care about being happy. We need to live with passion. We like the ups and downs of life. We like our suffering because it's so good when it ceases for a while." More laughter.
It is an unusually good opening. The easy route for a monk talking about happiness runs through compassion, the inner life, a verse. Instead he fishes the biggest doubt in the room out into the open, states it better than its owners did, and only then goes to work. For the next twenty minutes he really has one job: to show that those French intellectuals and he are not talking about the same thing at all — what they are mocking is pleasure.
Matthieu Ricard was born in 1946. His father, Jean-François Revel, was one of the best-known philosophers and political writers in France; his mother, Yahne Le Toumelin, was an abstract painter who later became a Buddhist nun. In 1972 he completed a PhD in molecular genetics at the Pasteur Institute under François Jacob, a Nobel laureate in physiology or medicine. Then he did not take a postdoc. He went to India, studied under Kangyur Rinpoche and later Dilgo Khyentse Rinpoche, and made the monasteries of the Himalayas his home. Since 1989 he has been the Dalai Lama's French interpreter; in 2000 he co-founded the humanitarian organization Karuna-Shechen with Rabjam Rinpoche, running schools and clinics in Nepal, India and Tibet — which is where the before-and-after photographs at the end of the talk come from.
This talk sits at a very specific junction, where three things happened to collide. First, his 2003 book Plaidoyer pour le bonheur had become a bestseller in France and had drawn the attack he opens with; he walks on stage in the middle of a live argument. Second, neuroplasticity had just moved from fringe hypothesis to consensus. His line that "until the last 20 years" the adult brain's connections were thought to be more or less fixed was a fresh thing to say in 2004. Third, and most important: he is himself the experimental subject. The Mind and Life Institute, driven by the Dalai Lama, had been sending experienced practitioners into Western labs, and Ricard — French, English and Tibetan, with a scientific training — became the busiest interface on that pipeline. When he speaks, the data is fresh and unpublished; he says so on stage, and the phrase he uses is "the scientific embargo." One more interval worth noting: the talk was recorded in February 2004, but TED did not post it until November 2007. By the time the world heard him say the paper had not been accepted yet, it had been in print for three years.
Once the laughter dies down he establishes a minimum shared premise: "no one wakes up in the morning thinking, 'May I suffer the whole day?'" — even the man who hangs himself, as Pascal had it, is seeking the cessation of suffering and can find no other way. The trouble is that definitions of happiness in the literature, East and West, are incredibly diverse, and he quotes Bergson's very French summary: "All the great thinkers of humanity have left happiness in the vague so that each of them could define their own terms." He treats this not as a witticism but as a problem: if happiness determines the quality of every instant of your life, we had better know what it is.
The clearing-up starts with one distinction, and it is the load-bearing wall of the whole talk: happiness is not pleasure. Pleasure is contingent on time, on its object, on place.
"Beautiful chocolate cake: first serving is delicious, second one not so much, then we feel disgust."— Matthieu Ricard, The habits of happiness, 2004
Bach is fine five times over; twenty-four hours non-stop would be exhausting. Near a fire is wonderful until you have to move back, and if you don't it starts burning. The conclusion is sharp: pleasure uses itself up in the experiencing, and it does not radiate — "you can feel intense pleasure and some others around you can be suffering a lot."
So what is the other thing? He concedes "happiness" is too vague a word, switches to "well-being," and gives what he takes to be the best definition on the Buddhist view: not a mere pleasurable sensation, but a deep sense of serenity and fulfillment, a state that pervades and underlies all emotional states. He knows how that sounds, so he asks the audience's question for them: can you have this while being sad? Why not — we are speaking of a different level. Then comes the most usable image in the talk:
"When you are at the bottom of the wave, you hit the bottom. You hit the solid rock. When you are surfing on the top, you are all elated. So you go from elation to depression -- there's no depth. Now, if you look at the high sea, there might be beautiful, calm ocean, like a mirror. There might be storms, but the depth of the ocean is still there, unchanged."— Matthieu Ricard, The habits of happiness, 2004
With the definition standing, he turns to method, and the first step is to reverse the direction of search. We look outside, assuming that if we could assemble all the conditions we would be set — and the sentence itself, he says, gives away its own doom: to have everything. If we miss something, it collapses.
The second step is the harder question: can the mind actually change, or are anger and jealousy built into it? Here he gives the closest thing to an argument in the talk, and the part least often repeated. Consciousness is like a mirror: any face can arise in it, and the mirror is not stained by those images. So it cannot be intrinsically tainted with hatred — otherwise, "like a dye that would permeate the whole cloth," it would be found all the time, "and we know we're not always angry, always jealous, always generous." Since every emotion is fleeting, there is room to work. Two routes: specific antidotes, resting on the claim that two opposite mental factors cannot occur at once — you cannot, in the same gesture, shake hands and give a blow — so rejoicing counters jealousy and loving kindness counters hatred; or a general antidote, which is to look at the emotion itself. Anger looks menacing, like a billowing monsoon cloud you think you could sit on; go there and it is only mist. "If you look at the thought of anger, it will vanish like frost under the morning sun." Do it enough and the emotion may still arise, but it "will just cross the mind, like a bird crossing the sky without leaving any track."
He adds, honestly, that this takes time, because the tendencies took time to build. And that the real meaning of meditation, for him, is familiarization — getting used to a new way of being.
The last seven or eight minutes bring in the science. The brain was thought to be more or less fixed in adulthood; recently it has been found to change a great deal — a violinist with 10,000 hours of practice shows a large change in the area controlling finger movement. Can the same be done with human qualities? That, he says, is exactly what the great meditators have been doing: some came to labs in Madison and Berkeley with 20,000 to 40,000 hours behind them, three-year retreats at twelve hours a day. "They are real Olympic champions of mind training." Then the three exhibits: first, a bell curve over 150 controls whose axis is the difference between right and left frontal lobe activity (more right, more depressed and withdrawn; more left, more altruistic and happy), on which the meditator of compassion is "totally out of the bell curve"; second, three and a half hours in an fMRI; third, Paul Ekman's startle experiment — a subject wired to physiological instruments, a bomb-like bang, and in twenty years "they never saw anyone who would not jump" — while some meditators, not trying to suppress it but simply staying completely open, "are able not to move at all."
He is the first to say this is not the point: "the whole point of that is not, sort of, to make, like, a circus thing of showing exceptional beings." The point is one sentence —
"This is not just a luxury. This is not a supplementary vitamin for the soul. ... We are ready to spend 15 years achieving education. We love to do jogging, fitness. We do all kinds of things to remain beautiful. Yet, we spend surprisingly little time taking care of what matters most -- the way our mind functions."— Matthieu Ricard, The habits of happiness, 2004
He ends without another argument, just photographs: a woman with bone TB, left alone in a tent to die, and then the same woman a year later; schools and clinics in Tibet; children's faces — "I leave you with the beauty of those looks that tells more about happiness than I could ever say." The last slide is a jumping monk. End.
What twenty-one minutes could not hold is mostly the size of those three exhibits.
Exhibit one: the gamma paper. On stage he says "a paper has been submitted to Nature, hopefully, it will be accepted." It did not appear in Nature. Nine months later, in November 2004, it came out in PNAS, authored by Lutz, Greischar, Rawlings, Ricard himself and Davidson. Open the methods and the size is this: 8 long-term practitioners against 10 controls. The practitioners averaged 49 years old (SD 15), with 15 to 40 years and 10,000 to 50,000 hours of training; the controls averaged 21 (SD 1.5) and had received one week of instruction before testing. The equipment was a 128-channel Geodesic Sensor Net — on stage he says "256 electrodes." The result is genuinely striking: the ratio of gamma to slow rhythms was already higher in the practitioners at resting baseline (p < 0.001), and once they entered nonreferential compassion the adjusted gamma increase was more than thirtyfold that of the controls. To their credit, the limitation was not dug up by critics — the authors put it in the paper themselves: the two groups "differed in many respects, including age, culture of origin, and first language," and longitudinal work would be needed to establish that the differences came from training rather than pre-existing traits. Which is to say the study can show that these two groups are different; it cannot show that meditation made them different. Between a 49-year-old Tibetan monk and a 21-year-old American undergraduate, there is rather a lot that can differ.
Exhibit two: the bell curve. Its axis is frontal asymmetry, which Davidson's lab had built into a biomarker of "affective style" through the 1990s. Ricard's own words about the position on that axis are "Here, the -0.5 is the full standard deviation of a meditator who meditated on compassion," followed by "it's something that is totally out of the bell curve" — a description that is muddled in the recording itself, since half a standard deviation and "totally out of the bell curve" do not match. What matters more is the caveat he supplied: he calls it a preliminary result and says plainly that the paper is still under embargo. What the room saw that night was not published evidence. It was a slide.
Exhibit three: the startle. The stage version is that in twenty years they had never seen anyone who would not jump, and that some meditators "are able not to move at all." That study was not formally published until 2012 — eight years after the talk — in Emotion, by Levenson, Ekman and Ricard, with four words in the title: a case study. The subject is one monk with roughly forty years of practice. The stimulus was a 115-dB, 100-ms acoustic blast, four conditions (open presence, focused, a distraction control, and unanticipated), each repeated 6 times, 24 trials in all. The findings come in three layers:
The third layer nearly inverts the sentence from the stage. Startle suppression is not a poise you carry around with you; it shows up when he knows it is coming and is deliberately holding a state. One person, 24 trials — that is the entire weight of this piece of evidence. There is also something the stage did not mention at all: gamma is the EEG band most easily contaminated by scalp muscle and eye movement, so "strong gamma measured at the scalp during intense concentration" always has a very ordinary alternative explanation — his face and eyes are moving.
Finally, his longer argument. In Plaidoyer pour le bonheur (2003; in English, 2006, as Happiness: A Guide to Developing Life's Most Important Skill) the load-bearing concept is not happiness but the Sanskrit sukha — the durable well-being that shows itself once a person is freed from mental blindness and afflictive emotions. For eighteen minutes on stage he renders it "well-being" and gives it the ocean image; in the book it is a whole epistemological claim about seeing things as they are, in which happiness is a by-product rather than a target. That shift matters: his claim was never "you can become more cheerful" but "you can see more accurately, and so suffer less easily." Nearly every retelling since has dropped the second half.
PARTLY HELD UP · as of 2026-09
The big claim — that the mind can be trained and that training leaves measurable traces — has held. But it stands on much narrower ground than the stage suggested: the reliable gain is about 0.3 standard deviations off anxiety and depression, comparable to antidepressants and no better than them, while on positive mood, the column he cared most about, the largest meta-analysis returns "low evidence of no effect or insufficient evidence of any effect." Of the three exhibits shown that night, the gamma finding was partly replicated elsewhere, the frontal-asymmetry axis has largely dissolved, and the startle work shrank on publication to a single subject whose crucial condition showed no difference at all.
January 2014: the clinical side reports. Goyal and colleagues published a meta-analysis in JAMA Internal Medicine covering 47 randomized trials and 3,515 participants. The good news: moderate evidence that mindfulness programs improve anxiety, effect size 0.38 at 8 weeks (95% CI 0.12–0.64) and 0.22 at 3–6 months; and depression, 0.30 at 8 weeks and 0.23 at 3–6 months — roughly the magnitude of antidepressants, real but undramatic. The bad news: "we found either low evidence of no effect or insufficient evidence of any effect of meditation programs on positive mood, attention, substance use, eating, sleep, and weight." And the sentence that matters most: "we found no evidence that meditation programs were better than any active treatment (drugs, exercise, other behavioral therapies)." Meditation works; it is one of the many things that work.
2017: the gamma finding is partly replicated. Braboszcz and colleagues, in PLOS ONE, compared practitioners from three separate systems — Vipassana, Himalayan Yoga and Isha Shoonya — against controls, matched by age and sex into four groups of 16. Meditators in all three traditions showed higher gamma amplitude over parieto-occipital regions in the 60–110 Hz range, and the effect was a trait effect: it appeared not only during meditation but during instructed mind-wandering, and gamma power correlated with years of practice. Crucially they addressed the contamination problem head-on, using independent component analysis to separate eye and muscle components and finding no significant between-group difference in temporal-muscle artifact activity. This is not a direct replication of those 8 people, but it moves the finding from one lab to three traditions and closes the most obvious hole. This one Ricard wins.
May 2021: the horizontal axis collapses. Kołodziej, Magnuski, Ruban and Brzezicka published in eLife: "No relationship between frontal alpha asymmetry and depressive disorders in a multiverse analysis of five studies." They went straight at the soft spot of such measures — there are too many defensible analytic paths, and a researcher wandering through them can easily arrive at significance. So they ran data from five independent studies, 388 participants, through 270 different analyses. Thirteen came out significant, exactly the number chance predicts. It is worth being precise about where this lands: it does not refute Ricard, it refutes the ruler he used that night to orient the audience. If the left-right difference in frontal activity does not reliably track depression or happiness, then "this person is totally out of the bell curve" loses the meaning it promised.
Meanwhile, something the talk never mentioned surfaced. In 2017 Lindahl, Britton and colleagues published "The varieties of contemplative experience" in PLOS ONE, interviewing 60 Western Buddhist practitioners and 32 meditation teachers and cataloguing 59 meditation-related experiences across 7 domains, many of them difficult and functionally impairing. The caveat must be stated plainly: this was a sample deliberately recruited from people who had had challenging experiences, so the proportions reported in it are not population rates. But it established something that had barely existed in public discussion — intensive meditation has adverse effects — and the "Olympic champions" frame from the 2004 stage has no place to put that.
And then the layer of misuse, which Ricard is himself fighting. The press distilled a headline out of those brain scans — "the happiest man in the world" — and stuck it to him for twenty years. He calls it "a nonsense idea" — "How can we know the state of happiness of 8 billion human beings?" Asked whether it irritates him, his answer is "I feel embarrassed." What finally stopped him fighting it was his teacher's grandson telling him: "Take it and use it. Stop fighting it." The trouble with the label is not only that it is inaccurate but that it inverts his claim: he was describing a craft anyone can practise, and the label describes a freak of nature. The second distortion is larger. After 2004 mindfulness became an industry. And among its sharpest critics is Ricard himself. In a 2013 piece he was blunt: "A sniper waiting for his victim: he can be one-pointedly concentrated, abiding unwaveringly in the present moment, calm and poised." Bare attention, however consummate, is "no more than a tool ... which can also be used to cause immense suffering." Six years later Ronald Purser reached a neighbouring conclusion from the secular side in McMindfulness — the critic and the criticized standing on the same side.
Where the field sits today is roughly this. As a method, meditation works, and works modestly, on a par with existing therapies. As a neuroscience story, the 2004 version can no longer be used as delivered — samples too small, confounds too many, one key measure no longer trusted — and the part that has best survived replication, the trait-level difference in high-frequency EEG, is precisely the part he said least about and which is hardest to interpret. As for his actual claim — that sukha is a condition-independent well-being cultivated by seeing things as they are — science has never really tested it. Testing it would take longitudinal work on the scale of decades, and eight-week trials are what we have. So the honest summary is this: the line he delivered on that stage, that happiness is a skill, may well still be true; it is just that twenty-two years on, nobody has picked up the burden of proving it.