Book Review: All Therapy Books

[Related: CBT In The Water Supply, Scientific Freud, Book Review: Method Of Levels, Different Worlds]


All therapy books start with a claim that their form of therapy will change everything. Previous forms of therapy have required years or even decades to produce ambiguous results. Our form of therapy can produce total transformation in five to ten sessions! Previous forms of therapy have only helped ameliorate the stress of symptoms. Our form of therapy destroys symptoms at the root!

All therapy books bring up the Dodo Bird Verdict – the observation, confirmed in study after study, that all psychotherapies are about equally good, and the only things that matters are “nonspecific factors” like how much patients like their therapist. Some people might think this suggests our form of therapy will only be about as good as other forms. This, all therapy books agree, would be a foolish and perverse interpretation of these findings. The correct interpretation is that all previous forms of therapy must be equally wrong. The only reason they ever produce good results at all is because sometimes therapists accidentally stumble into using our form of therapy, without even knowing it. Since every form of therapy is about equally likely to stumble into using our form of therapy, every other form is equally good. But now that our form of therapy has been formalized and written up, there is no longer any need to stumble blindly! Everyone can just use our form of therapy all the time, for everything! Nobody has ever done a study of our form of therapy. But when they do, it’s going to be amazing! Nobody has even invented numbers high enough to express how big the effect size of our form of therapy is going to be!

Consider the case of Bob. Bob had some standard-issue psychological problem. He had been in and out of therapy for years, tried dozens of different medications, none of them had helped at all. Then he decided to try our form of therapy. In his first session, the therapist asked him “Have you ever considered that your problems might be because of [the kind of thing our form of therapy says all problems are because of]?” Bob started laughing and crying simultaneously, eventually breaking into a convulsive fit. After three minutes, he recovered and proceeded to tell a story of how [everything in his life was exactly in accordance with our form of therapy’s predictions] and he had always reacted by [doing exactly the kind of thing our form of therapy predicts that he would]. Now that all of this was out in consciousness, he no longer felt any desire to have psychological problems. In a followup session two weeks later, the therapist confirmed that he no longer had any psychological problems, and had become the CEO of a Fortune 500 company and a renowned pentathlete.

Not every case goes this smoothly. Consider the case of Sarah. Sarah also has some standard-issue psychological problem. She had also been in and out of therapy for years, tried dozens of different medications, none of them had helped at all. Then she decided to try our form of therapy. In her first session, the therapist asked her “Have you ever considered that your problems might be because of [the kind of thing our form of therapy says all problems are because of]?” Sarah said “No, I don’t think they are.” The therapist asked “Are you sure you’re not just repressing the fact that they totally definitely are, for sure?” As soon as Sarah heard this, she gasped, and her eyes seemed to light up with an inner fire. Then she proceeded to tell a story of how [everything in her life was exactly in accordance with our form of therapy’s predictions] and she had always reacted by [doing exactly the kind of thing our form of therapy predicts that she would], only she was repressing this because she was scared of how powerful she would be if she recovered. Now that all of this was out in consciousness, she no longer felt any desire to have psychological problems. In a followup session two weeks later, the therapist confirmed that she no longer had any psychological problems, and had become the hand-picked successor to the Dalai Lama and the mother of five healthy children.

Previous forms of therapy have failed because they were ungrounded. They were ridiculous mental castles built in the clouds by armchair speculators. But our form of therapy is based on hard science! For example, it probably acts on synapses or the hippocampus or something. Here are three neuroscience papers which vaguely remind us of our form of therapy. One day, neuroscience will catch up to us and realize that the principles of our form of therapy are the principles that govern the organization of the entire brain – if not all of multicellular life.


Maybe I’m being unfair here. I’m basing this off a small sample of therapy books (five textbooks I can think of, plus scattered papers on psychodynamic and psychedelic therapies), and only a subset are quite this bad.

But my basic confusion is this: I work in a clinic with about ten therapists. Some are better than others, but all of them are competent. I send my patients to them. In a few hundred patients I’ve worked with, zero have had the sudden, extraordinary, long-lasting change that the therapy books promise. Many have benefited a little. A few would say that, over the course of years, their lives have been turned around. But sudden complete transformations? Not that much.

Of course, this fits with the therapy books’ perspective. My colleagues practice normal therapy. Sometimes it’s from a boring old school like CBT; other times it’s “eclectic” or “supportive” or any of the other words we use to describe what we’re doing when we don’t know what we’re doing. So maybe there are two sets of therapies: boring old therapies that ordinary people practice, and exciting new therapies that people write glowing books about. And maybe the first set really don’t work (or work only a little), and the second set really is that good.

The problem is, the boring old therapies that everybody uses nowadays inspired equal excitement when they first arose. This is the point that I make in CBT In The Water Supply, and that Oliver Burkeman makes more cogently in Why CBT Is Falling Out Of Favor. Look at therapy books from the 1990s, and they were all about how CBT was a new miracle therapy that would cure your anxiety forever in a few sessions. From a cognitive therapy book:

[When I first learned about cognitive-behavioral therapy, I thought] depression and anxiety seemed far too serious and severe for such a simplistic approach. But when I tried these methods with some of my more difficult patients, my perceptions changed. Patients who’d felt hopeless, worthless, and desperate began to recover. At first, it was hard to believe that the techniques were working, but I could not deny the fact that when my patients learned to put the lie to their negative thoughts, they began to improve. Sometimes they recovered right before my eyes during sessions. Patients who’d felt demoralized and hopeless for years suddenly turned the corner on their problems. I can still recall an elderly French woman who’d been bitterly depressed for more than fifty years, with three nearly-successful suicide attempts, who started shouting “Joie de vivre! Joie de vivre!” (“joy of living”) one day in my office. These experiences made such a strong impact on me that I decided my calling was in clinical work rather than brain research. After considerable soul-searching, I decided to give up my research career and become a full-time clinician. Over the years, I’ve had more than 35,000 psychotherapy sessions with depressed and anxious patients, and I’m every bit as enthusiastic about CBT as when I first began learning about it.

But look at therapy books now, and they’re all people saying “Sure, CBT barely outperforms placebo…but what about this exciting new therapy which blows CBT out of the water?”

Studies reflect this decline:

…with the average studied effect size of CBT shrinking from 2.5 to 1.0 over the course of a generation. People have come up with various explanations for this. Maybe therapist quality is falling – when CBT was the hot new thing, you had to be a really plugged-in up-to-date therapist to have heard about it and to make the effort to retrain in it, so only the best therapists would practice it, but now it’s the default therapy used by everyone who’s just clocking it in. Maybe placebo effect is falling – when people viewed it as an astounding miracle therapy, it got astounding miracle results, but now that it’s lost its luster nobody takes it seriously anymore. Maybe its ideas are spreading, so that patients come into their first session already aware of CBT insights and inoculated against them. Or maybe it’s like all science, where the first studies are done quickly by true believers, and the later studies are done carefully by the Cochrane Collaboration, and so the level of hype naturally goes down.

These explanations have different practical implications. If it’s all about therapist quality and placebo expectations, then you should go get the exciting new therapies described in therapy books, since their unusually-qualified therapists and unusually-high expectations will deliver you the miracle cure you’re looking for.

If it’s just that study quality gets better and better until we realize how crappy the exciting new therapies really are, you might as well get the boring old therapies. At least insurance probably covers them.

And they also have different philosophical implications. If it’s all about therapist quality and placebo expectations, then even if it’s hard to deliver high-quality therapy consistently at scale, it means high-quality therapy is a thing. It means that if enough factors go right at once, therapy can be the kind of powerful tool that cures someone’s life-long psychiatric issues in a few sessions with a high success rate. If this is true it would be fascinating. It would be like saying that bananas cure cancer, but only if they’re really fresh bananas. Even if there are practical issues in getting every cancer patient a banana that’s fresh enough, you still want to take a step back and think “Whoa, what’s up with this?”

I can only say that I’ve had a few patients try the exciting new therapies, and none of them have reported miracle cures. They’ve all maybe gotten a little better over long periods, same as the boring old therapies. This makes me think it’s more likely that early results from the exciting new therapies get oversold, not that some combination of therapist skill and excitement makes them go shockingly well. And the Efficient Market agrees with my low estimation, given that therapists aren’t rushing to learn these new strategies and patients aren’t rushing to use them.

But the therapy books still confuse me. They’re full of stories of incredible instant cures, with the authors assuring us that these are all real and typical of their experience. How can you get this from merely “stretching the truth”, as opposed to outright data falsification? Are therapy book authors blatantly lying? I try to have a really low prior on this sort of thing, but I’m not sure.

Therapy books are often written by the researcher who invented the therapy. I imagine if you invent a therapy yourself, then it perfectly fits your personality and communication style, you believe in it wholeheartedly, and you understand every piece of it from the ground up. You’re also probably a really exceptional and talented person who’s obsessed with psychotherapy and how to make it better. So maybe they get results nobody else can replicate?

But that still raises the philosophical implication of it being possible, for somebody, to consistently produce dramatic change through therapy. This still bothers me a lot.


Most therapy books share some assumptions, so deep as to be unspoken: current problems serve some purpose related to past traumas.

Different therapies take this in different directions. Some view problems as a passive residue of past traumas: for example you were abused as a child, that filled you with stress and rage, and now you take that out on other people and yourself. Others view them as maladaptive learning from past trauma: for example, you were abused as a child, that taught you that other people would hurt you if you opened up to them, so you never open up to anybody. I don’t know the official name for this, but let’s call it historicism: symptoms are the result of something that happened in a patient’s life history.

Some weak forms of historicism are obviously true. Many (though not all) phobias began with a clear incident where the patient was endangered by the phobic object; someone mauled by a dog as a child who then has cynophobia as an adult is hardly a medical mystery. Many (though not all) depressions are precipitated by some depressing event. And post-traumatic stress disorder has the historical perspective right there in the name; at the very least, going through trauma dysregulates something inside you. But it’s a long way from there to saying that a patient’s psychosomatic blindness is caused by persistent shame at having seen their parents having sex thirty years earlier, or something like that.

And some therapy books go beyond historicism into purposefulism: symptoms serve some quasi-logical purpose relating to the life history. I recently read a therapy book that included a case like this. Bob had a history of failing at work. He would go from job to job, making various mistakes and doing crappy work until he got fired. He went to a therapist for help. During the therapy, it came out that Bob’s abusive father had always pushed him really hard to succeed. The therapist suggested that maybe Bob failed at work to send a message to his father; ie to prove that his father’s abusive parenting had been a bad idea and would not make Bob successful. The therapist asked Bob to imagine confronting his father about this. After he worked through his anger at his father, Bob was able to succeed at work. In this story, the apparently dysfunctional symptom (failing at work) ended up having a legible purpose within Bob’s life history (it helped him send a message to his father). Only by teasing out the purpose and finding some other way to achieve it could the dysfunctional behavior be prevented.

A non-historical, non-purposeful account might argue that Bob failed at work because he was bad at work. Maybe he was bad at the specific jobs he was holding (in which case he should get more training). Maybe he was bad at social skills (in which case he should learn to communicate better). Maybe he had ADHD and kept getting distracted (in which he should get treatment for ADHD). In any case, him being bad at work isn’t related to any past traumas or serving any hidden purposes. It’s just an unfortunate fact.

I am constantly worried by the history of how many things we historically applied historical-purposeful reasoning to, totally confident at the time that our explanations made sense – which we now know are not historical-purposeful at all. Psychologists “knew” that autism was caused by distant mothers, and schizophrenia by overbearing mothers, right up until we discovered both conditions are about 80% genetic. And when they “knew” these things, they were able to come up with long lists of how exactly each individual patient fit the mold, and reported great progress by helping patients overcome their maternal attachment issues. Back when homosexuality was considered a disorder, historical-purposeful therapists would tell gay people patients they must be so angry at their mother that they had sworn off all female companionship and switched to men instead as a way of sending her a giant “F–K YOU” message; while homosexuality is mostly not genetic, few people today think this is a plausible explanation.

I sometimes see if I can come up with these kinds of historical-purposeful accounts of my patients’ symptoms. These always fit into place freakishly well – so well that either the historical-purposeful perspective is completely true, or there is some very strong bias that makes it extra-convincing despite its falsehood. But we already know there’s some very strong bias that makes it extra-convincing despite its falsehood! That bias must have been at work in all the therapists who applied historical-purposeful narratives to autistics, schizophrenics, and gays! At some point I notice the road I’m on is littered with skulls and start wondering if I should reconsider.

All therapy books propose an answer: the proof is that the patients get better. But my patients do not get better. When I tell them the historical-purposeful accounts I have devised for their symptoms, they usually shrug and say it sounds plausible and they’ve thought along those lines before, but what are they going to do? When I try all the exciting new therapies on them, they just sort of nod, say that this sounds like an interesting perspective, and then go off and keep having symptoms. It’s very rude!

I’ve told this story before: when I was a teenager, I got really into pseudohistory for a while. What snapped me out of it wasn’t the sober historians, who totally went AWOL on their job of explaining why they were right and the whackos were wrong. It was that a bunch of mutually exclusive pseudohistories all sounded equally plausible: the Pyramids couldn’t have been built by Atlanteans and Lemurians and mole-people! At that point I was able to halt, melt, catch fire, and realize there was something really wrong with my reasoning processes, which I continue to worry about and work on twenty years later.

I bring this up because I’m going to be reviewing some specific psychotherapy books. Each of them on their own can be convincing. But they should be taken in the context of All Therapy Books, which as a category are pretty worrying.

This entry was posted in Uncategorized and tagged , . Bookmark the permalink.

214 Responses to Book Review: All Therapy Books

  1. onyomi says:

    I feel like this mostly applies to all diet books as well, but rather than tracing individual cases of obesity to individual traumas they tend to spin some grander narrative about e.g. corn subsidies.

    • jhertzlinger says:

      Diet books share some common advice: They all agree that sugar is bad and green vegetables are good. Maybe that means the next fad diet will eschew green vegetables and load up on sugar.

      • siwhyatt says:

        I immediately thought of diet books too.
        Each new fad proclaiming to be the one true diet to rule them all, that only they have “the secret”.
        Again, taken in isolation any one can sound convincing, but side by side it becomes apparent that they can’t all be correct. Low fat, ketogenic, vegan, paleo – all claim to be the panacea, all completely contradictory. No doubt this phenomenon happens in all disciplines?

        • onyomi says:

          No doubt this phenomenon happens in all disciplines?

          I don’t think to this degree, but, besides diet books, I was also reminded of this “heroic narrative grant template” I read years ago and which I’ve found is indeed pretty effective.

          Basically if you want to convince people to give you money for something that may or may not work out (in other words they’re not getting e.g. a ham sandwich for their money), you want to create a coherent narrative something like:

          1. There’s a serious problem!
          2. Since this is a serious problem, many others have, of course, attempted to tackle it.
          3. But they’ve never really gotten to the heart of the matter because they’ve all overlooked factor X.
          4. I realized the importance of X because of my unique background.
          5. My [grant project, proven method] fixes the problem/addresses the research question by zeroing in on X.
          6. Here are the details, which look suspiciously similar to every other research project/miracle method (get access to these after you buy my e-book).

          For something like depression or obesity you don’t have to convince anyone the problem matters so you can kind of skip 1, but you have to devote a lot of space to it if you want money to research medieval Chinese poetry, etc.

        • pat55word says:

          In the diet case it’s probably because they all compare favorably to the baseline of the western diet which may be the worse diet in history.

          • thetitaniumdragon says:

            Ah yes, the diet that has resulted in people living longer, healthier lives and having higher IQs than any previous population. Must be terrible.

            Someone who ate nothing but convenience store junkfood for a month saw improvements.

            It seems like the only actual thing that works is eating less food in an absolute sense and exercising.

            But “eat less and exercise” is so obviously correct that you can’t really write a book about that.

            I’ve sometimes considered trying to make/print a joke book that is just called “Eat Less and Exercise – the Only Diet Book You’ll Ever Need” and the inside of the book is entirely blank pages.

        • Exa says:

          Having known people who have had dramatic, clear successes on… just about any popular diet you care to name, my current impression is some combination of “They do really work, just on some particular subset of people, a la antidepressants. People can keep trying them and have good odds of finding one that works, but any given one will do approximately nothing on the total patient population” and “Anything will work if you stick to it.”

          For a point of anecdata on the “Anything will work if you stick to it” side, so far this year I’ve lost about 40 pounds on a diet consisting mostly of caramel lattes, pastries, sandwiches, chips, and pseudo-japanese takeout. (The secret is that I just eat less food in total.)

          • In recent months, I lost fifteen to twenty pounds on a diet that wasn’t designed for weight loss at all, but to fit Bredesen’s theory of the causes of Alzheimer’s and how to prevent it. I think the reason I lost the weight is that the diet constrained what I could eat and when, with the result that I ended up eating less.

      • Don_Flamingo says:

        Not true. Carnivore books will tell you that sugar is bad and green vegetables are plant food. And then tell you to eat only meat, since that’s not plant food.

        Keto books for epileptics will tell you to never eat pasta, because you’d get epilepsy.
        They don’t differentiate between galzed donuts and pasta, because only the carbs matter and you bettter keep away from them. They’re correct, afaik.

        Other diet books however would tell you that pasta with broccoli and tomatoes and feta cheese, pesto and fish is what makes those Mediterreanaienans live so long. I’m not sure they do, but their diet sounds delicious.

        The Shangrila-diet book will tell you to always eat something else and preferably strange, so that you never grow to like something too much for your own good.
        Because there are animals who eat novel food very cautiously and…. dead animals.

        Saw an Australian vegan youtuber once, who only ate bananas. Was years ago. Sure, she could write a book (if she’s still kicking that is). Though she’d hardly claim to avoid sugar.

        Then there are many diet books about IF protocols.

        Then we’ve got diet books for people who lift like crazy and can somehow handle large amounts of protein to grow muscle like crazy.

        Then there are books about cheat days and carb refeeds and the whole slow-carb thing which I honestly have forgotten everything about.

        Point is, humans can thrive on any diet.
        That has mostly to do with the fact, that we’re not good at enforcing them 🙂

    • Scott Alexander says:

      Interesting! I hadn’t considered the comparison, maybe because therapy books seem to me more academic and official than diet books (and are also aimed at people treating problems rather than the people with problems themselves), but now that you mention it the analogy seems apt.

    • Ketil says:

      I agree, and also alternative medicine. I hear many stories from Bobs and Sarahs (i.e. patients) who have tried every treatment and remedy for years and years, until they finally found the the Ultimate Alternative Treatment, which they often didn’t really believe in, btw, and suddenly were cured for their bad back/migraines/whatever.

      One common factor to all of these – in addition to unhealthy doses of confirmation bias and selection bias, and so on – may be the Cause that is Not Your Fault. Your obesity is because of a metabolic condition, your anxiety is because of childhood abuse, your chronic fatigue is because of a virus infection – it doesn’t matter if alleged causes are undetectable by normal means as long as they exonerate the patient.

      ETA: and related, a common factor is the silver bullet. If only the real doctors can get their act together and come up with a cure for my syndrome or virus, or if the accused can confess to the transgressions, then everything will…magically fix itself, and I can eat as much ice cream as I like and become the next Nobel prize winning athlete I was meant to be.

      ETA2: Until strong evidence to the contrary, these factors are powerful priors for me to not give any credence to such theories.

      Another common factor might be an agreeable therapist. I wonder how often the evaluation of outcome depends on the patient’s own evaluation, and if there is a bias in favor of an agreeable therapist?

      Authors of diet books can perhaps be excused for not understanding placebo and cognitive biases – psychotherapists, less so.

      • jasmith79 says:

        I had an alternative medicine give me a ‘miracle cure’ for a pretty common physical ailment.

        Exactly zero of the people I know with the same issue got any benefit from it after I excitedly told them about my experience.

        I still tell people about it, sans excitement.

        • cedrus_libani says:

          I have had this experience too. I’m a biologist; the treatment in question is promoted by quacks; having to admit that they were right makes me want to take a shower. But it worked for me, when nothing else ever has.

          Even among people who take health advice from internet quacks, the general opinion is that it doesn’t work. As best as I can tell, it only works a tiny fraction of the time, but when it works…it really works. I figure that there’s an explanation in there somewhere, but I don’t have one.

          The experience has somewhat shifted my priors about what is genuine long-tail medical weirdness and what isn’t. Which is rough, because there really is a lot of nonsense out there too.

        • jasmith79 says:

          In terms of efficacy, I’ve heard the more reputable alternative medicine folks quote the “rule of thirds”: one-third will be helped a lot, one-third will be about the same, and one-third will get worse on the treatment. My anecdotal experience has been more like 5/90/5 but #whateves.

          In terms of why, my suspicion is that the ailments in question are multi-causal. The Rx will probably work well enough regardless of the actual cause, the alternative only works on people with a particular cause. But IDK, I’m no doctor/researcher.

      • Cliff says:

        One time a guy was trying to pitch oxygenated water at my gym. To verify the efficacy of the product, he asked me to first do some stretches, stretching as far as I could, and then to redo the stretches after drinking some oxygen water. I had no belief in the efficacy of “oxygenated water” and still don’t, but after drinking it I was able to stretch significantly farther, to the point that a few observers accused me of being in on the demonstration.

        • Nancy Lebovitz says:

          I assume it wasn’t just that you’d stretch farther because you were warmed up by the first session?

    • DinoNerd says:

      In my 40+ years in software engineering, I’ve encountered a number of methodologies that were advertised in the same way – just do this new thing, and all the usual problems would be solved. Sometimes, with the encouragement of consultants, managers bought into one of them in a big way, but I’ve never really noticed them working as well as was claimed by their literature.

      • martinw says:

        Joel Spolsky’s famous analogy comparing talented programmers to artisanal chef cooks, and capital-M Methodologies to fast-food restaurant chains, would seem relevant here.

        A related common pattern is: somebody develops a new programming language, and some early adopters write various successful and beautifully elegant applications in it. Then mainstream programmers try to adopt it and fail, because it turns out that the success of those first applications was not a testament to the quality of the language but to the skill of those early adopters. One of the worst things that can happen to a tech company is when an important new system needs to be written from scratch, so it is assigned to the company’s star programmer who decides to use whichever language they happen to be obsessed with that week, and then later the star programmer leaves and some ordinary mortal is asked to maintain/extend it..

      • warrel says:

        Ah.. the anti-DevOps hangover .. It’s in the air , everywhere.

    • MaureenC says:

      That was my feeling as well. I rather suspect that the doctors who come up with new diets/therapies have patients that cluster in certain traits–maybe traits that the doctors aren’t even aware of–and while the therapy/diet is very effective for people in that trait cluster, it’s less effective for random people.

      • Sebastian_H says:

        I wonder if certain therapies work better with certain *therapists*. I don’t mean just therapist skill level, I mean that certain therapy styles might match with certain therapist personality styles, but we never think to match them that way. So the therapist learns the dominant style of their school without ever thinking about it, and if much later in their career they find something that seems to work better they think that is an evolution in the science rather than better matching.

    • matthewravery says:

      My general thought on why diets tend to not work is that there’s more heterogeneity in people than a single diet can solve. There are some general rules (“don’t eat so much sugar; make sure to eat some veg”), but beyond that, you’ve got a lot of idiosyncracies. Things that work very well for some folks are actively counter-productive for others.

      Medicine is the same way, which is why it’s harder to find useful drugs these days. We’ve already got the obvious things that work in general figured out, and when we discover something new that works, it tends to only work for a subset of the population. We do statistics for global effects and find poor results.

      It seems like a similar mechanism could explain the phenomenon identified by Scott in this post. Some therapies work very well for some people and not at all for others, and it’s hard to a priori match patient with treatment.

      • Radu Floricica says:

        There’s also quite a bit of variation for each individual in time as well. I’ve done ok on various diets at various times – depends a lot on mood, motivation and social context.

      • thetitaniumdragon says:

        I think the real answer is that you just need to eat less and exercise more. It doesn’t really matter what your diet is, but if adopting some random diet encourages you to eat less (by, say, forcing you to eat less appealing foods exclusively)…

    • fibio says:

      My theory on diets is that, fundamentally, they’re all calorie counting and calorie counting works regardless of the dressing that goes on around the edges. When a diet doesn’t work it’s because people don’t or can’t follow the diet for whatever reason (not all of them under their control before anyone jumps on this). As such you get a weird cyclic effect.

      A new diet works really well for a small niche of people who were already well set up to lose weight. They rave about the power of this new diet. It gets picked up on masses, where upon it reaches all the people who can’t lose weight as easily and it does nothing at all to help. They drop out and complain about poor results and everyone goes off to search for a new diet. A few people who happen be in a better place to lose weight find another new diet and rave about it when it ‘miraculously’ works, restarting the whole cycle.

      Not sure how well this holds up but it’s my working explanation.

  2. Elo says:

    Maybe there is something in the attitude of a “knowing healer”. A helper professional that has a method that seems mysterious enough that the patient keeps trying, and the helper seems confident enough in their new, optimistic, novel, different method that they bring the required effort to cause progress on the issue.

    Maybe they all work equally well but based on what the helper is demonstrating in attitude independent of the specific method.

    • Richard Kennaway says:

      You’ve just discovered a new therapy: Knowing Healing! Start practising, do some remarkable instant cures, and write Another Therapy Book explaining that this new form of therapy will change everything and produce total transformation in five to ten sessions!

  3. romeostevens says:

    Therapy reifies a model whereby there is the person who is talking to the therapist, who ostensibly has some agency, but perhaps is frustrated by how they seemingly don’t actually, and their collection of behaviors and drives which aren’t really agents. The historical perspective tries to repair this by saying, no, really the behaviors and drives are somewhat agentic too. I think this is somewhat helpful, but not ultimately the correct point of intervention for deep rooted issues. For deep rooted issues my impression is that actually the two sides in conflict are on the same side. The conflict is theater that guards against the actual changes, because actual changes are highly threatening.

    • Scott Alexander says:

      It sounds like you’re kind of describing Games People Play, which was the Exciting New Therapy Book That Would Change Everything sixty years ago.

      • romeostevens says:

        I’ve read it, I can kind of see the comparison but I don’t think it captures the thing. The thing I’m pointing to is more closely related to selection effects whereby people who really want to change don’t wind up seeing therapists.

      • Nav says:

        Having read the majority of Eric Berne’s corpus, I’ll say that Games People Play (and really its successor, What Do You Say When You Say Hello?, which specifically discusses “script analysis”) sounds quite similar to what you’re describing. But I think to explain historicism, we have to go back to Freud. As usual.

        Freud’s first book, The Interpretation of Dreams, sets out explicitly with the sort of methodology you describe. Learn about the dream, its narrative, its manifest content, then use that to locate some “latent material” which engages with the actual cause of the symptom. Sounds bullshit, right? Let’s dig in a bit deeper, quotes from A Quick and Dirty Guide to Psychoanalysis (emphasis mine):

        Dreams provide evidence of the unconscious in that their symbols can (not ‘have to’) be traced back to everyday experiences the dreamer experienced recently, or thoughts and feelings the dreamer finds meaningful to their life narrative.

        …in the 60s-80s a big rift in psychoanalytic dream studies develops. It can be summed up as the positivists vs. the hermeneuticists. Positivists: dreams ‘have’ meaning. Each symbol has a meaning waiting to be discovered. You can map out consistencies between dreams etc. Hermeneuticists: dreams don’t take on meaning until the analysand gives them meaning via the context in which they bring up the dream, and the contexts in which the dream was created, or, until the analyst interprets the dream in a meaningful way that ‘sticks’ with the analysand. Freud was a little bit of both, which both parties seem to deny.

        I suggest that every historicist therapy finds itself straddling the same fault line as dream analysis. Some historical events in the patient’s life are inherently meaningful and directly influence the patient’s day to day life as they reflect on it. The positivist school. Other historical events are made meaningful in the process of therapy, providing raw material through which the patient forms a more complete or novel life narrative. The hermeneutic school.

        As I did in my other comment, we can insert predictive processing here: a life narrative, deeply held, is effectively equivalent to a set of extremely high level priors. The process of rebuilding a novel life narrative through a combination of positivist accounts (yes, I think about that every day!) and hermeneutic accounts (that makes sense as part of my life story, can’t believe I didn’t discover it sooner) can be seen as a readjustment of priors. I believe this is the exact sort of action that takes place under the influence of psychedelics (to tie this specifically to the REBUS paper): past experiences, viewed in a new light, form a novel narrative, which results in positive life adjustments for the taker.

        The final aspect is, why does this take time? Why doesn’t the patient’s life suddenly shift when a plausible new narrative emerges? Freud’s explanation was called “resistance”: sometimes thoughts, especially about oneself, are painful, and the patient doesn’t want to objectively consider them because it hurts them. You can see this a lot in intellectual conversations, where an interlocutor might fail to consider a certain frame that you find useful, because it goes against their image of self. “Well, other people might be bad, but I would make a good dictator.” What happens if you gently suggest they wouldn’t?

        It is through this sensitivity to suffering, caused by contradictions between our objective mind and our sense of self, that we resist psychic change. Kegan wrote a whole book on this idea (which I haven’t read) called “Immunity to Change”. And this seems like the real barrier that needs to be overcome in therapy. How do you encourage the patient to accept a more and more objective view of the world, in which the newly constructed narrative is a healthier fit? How do you, as Nietzsche defined it, help them become stronger?

        A thing could be TRUE, although it were in the highest degree injurious and dangerous; indeed, the fundamental constitution of existence [of the patient’s self] might be such that one succumbed by a full knowledge of it—so that the strength of a mind might be measured by the amount of ‘truth’ it could endure—or to speak more plainly, by the extent to which it REQUIRED truth attenuated, veiled, sweetened, damped, and falsified.

        – Beyond Good and Evil

      • Garrett says:

        My view on Games People Play is that it was a nifty enumeration of different ways in which people engage in “bad” behavior for some internalized personal return. That these are somewhat common and it provides a good language to identify the patterns and possibly stop doing them is useful. But it suffers from being limited in two ways. First, in the sense that it enumerates a set of behaviors which can be looked for, but doesn’t help much for anything which falls outside of them. The second being that it doesn’t address other causative issues like being depressed because you’re a paraplegic – attempts at doing so involve shoehorning the problem into the model rather than the model naturally addressing it.

      • kupe says:

        Woah, I’ve been using Session Games People Play as recommended reading for sessions and never knew it was based on this. Maybe I should update my psychedelic advice manual to something a little more modern 🙂

  4. Bugmaster says:

    I know absolutely nothing about therapy, but still, this post really got me curious: had there been any definitive, practical progress in therapy in the past, say, 50 years ? If so, what was it; and if not, what would progress look like ?

    • Scott Alexander says:

      It’s hard to say.

      Therapists talk about having made progress in understanding the psyche, and they currently say things that sound more reasonable to me than the things old therapists said (although modern-day therapists should beat past therapists in sounding good to modern-day me just by sharing a paradigm). Therapy today is often shorter than in the past, which at least means fewer resources are getting used.

      But in terms of whether modern day therapies outperform past therapies, I think most studies show all therapies doing equally well (the exceptions are usually studies by the proponents of various therapies which show their therapy doing better).

      • Steve Sailer says:

        I don’t have much experience with therapy, but the rise of Cognitive Behavioral Therapy in the late 20th Century strikes me, as a not-very-informed observer, as a good thing. It sounds more reasonable-sounding and thus more promising than what I recall was fashionable in the 1970s when people I knew were paying a lot of money for stuff like Primal Scream.

        Has anybody ever documented whether or not Barack Obama had CBT when he fell into depression after being beaten by Bobby Rush in a 2000 House seat primary? In his second memoir, when he’s describing how he climbed out of his long depression, he uses some CBT phrases.

        Hopefully, in his third memoir he’ll discuss how he overcame his depression, which could be helpful to the public since he certainly made quite a recovery.

      • Stringhe says:

        in terms of whether modern day therapies outperform past therapies, I think most studies show all therapies doing equally well

        I also know absolutely nothing about therapy, but still, that strikes me as an extraordinary claim

        Are you including things like SSRIs? (which seem to be 30 years old, not 50)

        • Ketil says:

          Are you including things like SSRIs?

          I’m not sure exactly where the line is drawn, but I got the strong impression we are talking about therapy that (mostly or exclusively) consists of talking with the patient.

          I.e., not pharmaceuticals, which may or may not work to various degrees, but which at least can be shown to have an effect.

          • Stringhe says:

            Oh, of course, thank you!

            So the idea is that combined with current drugs, both modern and classical therapies would have the same “performance”?

          • FLWAB says:

            So the idea is that combined with current drugs, both modern and classical therapies would have the same “performance”?

            No, not when combined with drugs. On their own just about all therapy models are equally effective. It is indeed extremely remarkable: similar to a doctor that believes in the model of imbalanced humors being as effective at treating disease as one who is a homeopath. But remarkable as it is, it is the current state of “talk therapy.” Another reason why everyone is looking for the one true therapy that will actually be better than the others, because then we might have a model that is closer to being true.

          • Bugmaster says:

            I hate to say it, but in this case, a viable alternative explanation might be that all talk therapy is just placebo. Or, if not placebo, then merely a way to allow the patient to talk out loud to a non-judgemental, sympathetic stranger, without any of the usual social constraints that limit conversations in the modern world.

          • cuke says:


            I’m a therapist and I agree with you here: “then merely a way to allow the patient to talk out loud to a non-judgemental, sympathetic stranger, without any of the usual social constraints that limit conversations in the modern world.”

            I mean, there’s a bit more to it than that, and the person you talk to isn’t a stranger for long and that’s part of it too, but I think you’ve captured a core part. It turns out it’s remarkably effective.

          • Mark Atwood says:

            without any of the usual social constraints that limit conversations in the modern world

            Don’t worry, social justice, along with the court system, and the insurance companies working on solving that problem.

            I’m starting to consider recommending prayer. Even if you are an atheist, prayer still has the advantage that who you are talking to more or less has your interests at heart, and can keep you confidences.

          • The Nybbler says:

            “Two can keep a secret, if one of them is God.”

          • Stringhe says:


            How do we know the effectiveness of modern therapies without drugs? Aren’t they included in almost all modern treatments? At least, everybody I know who had problems serious enough to go to therapy was given some

          • FLWAB says:


            I work at a Marriage & Family Therapy Clinic. We don’t prescribe drugs, because we don’t have a psychiatrist on staff. Now most of our clientele is couples working on their relationships, and there are just as many therapy models for that as there are for individual focused therapy. In fact, most of them are just the different individual focused psychotherapy models but with “marriage” taked on to it. Emotionally Focused Therapy becomes Emotionally Focused Marital Therapy, Narrative Therapy becomes Narrative Marital Therapy, etc. But even putting couples aside we see a lot of individual clients as well. Typically people with PTSD, various Anxiety disorders and disphorias, and depression. We have a Psychologist who handles personality disorders: mostly borderlines. We don’t see schizoids, OCD, or any of the “harder” cases. And we don’t do drugs: besides the fact that nobody at the clinic is allowed to prescribe them, the clinic philosophy is gently anti-medication. We don’t tell people to go off their meds, but the staff are not generally fans of people getting on them.

            I don’t think our practice is out of the ordinary in that regard. There are a lot of people who don’t want meds, and a lot of practitioners who don’t like them either. Also psychiatrists are expensive, so some people would rather stick to a psychologist or LMFT or LPC, none of whom can prescribe.

        • ADifferentAnonymous says:

          This blog had previously discussed evidence that SSRIs are worse at treating depression than older drugs. (Fewer side effects, though)

  5. Ursus Arctos says:

    A sympathetic reading of the story- Alice is a good therapist. She notices that she has a much higher rate of successfully treating patients than other therapists. She wracks her brains to think why this might be. She’s uncomfortable with the idea that it is just her innate talent, so she concludes it is due to X somewhat unusual aspect of her approach. She begins to emphasise X aspect more and more. Eventually she spins X off into a novel therapeutic approach.

    • Steve Sailer says:

      Indeed. I’ve become more sympathetic to cults as I’ve broadened my perception of what might qualify as a cult. If Alice really has a gift, why not create a school or cult of people she trains to do what she does? They likely won’t do it as well as her, but her example will boost their self-confidence, which, even if her version of therapy doesn’t actually do anything above average in the hands of random therapists, might have a positive placebo effect.

    • zzzzort says:

      This doesn’t even require that Alice is a good therapist, just that she has a run of better than average results. You could try sending in some actors who get miraculously cured in one session and see if you could induce a random therapist to write a therapy book.

  6. neworder1 says:

    [anecdotal evidence based on personal experience with many therapists + interest in body-oriented therapy: Reich, Lowen etc.]

    1. There really exist “10x therapists”. In other words, there is a HUGE gap in terms of quality of results (however you measure it) between therapists who are *outstanding* and those who are merely “competent”. Come to think of it, this is barely surprising. Personal change is very nonlinear and a therapist “finding the right key” to the client’s problems (as opposed to small incremental results based on off-the-shelf therapies and theories) can make a night and day difference. This agrees with the view of psychotherapy as largely ineffectual but potentially life-changing. And this also brings us to: —->

    2. Out of necessity, psychotherapy today is a mass profession. This creates an illusion that a therapist is someone like a car mechanic or a cardiologist, whereas in reality it’s more akin to shamanism. If instead of apprenticeship for the chosen ones with a unique gift (which is how learning shaman’s craft worked in the old days) you could learn shamanism at university, with a “shamanic certificate” after 5 years of study, the field would suffer from the same problems as psychotherapy. Personal qualities that determine a great therapist can be honed if present, but largely cannot be taught or learned.

    3. IMHO strongly intellectual types (such as Scott or yours truly) rarely make good therapists. Their overreliance on thinking make them well suited to problems which can be analyzed in a mechanistic fashion (think – psychiatric disorders with a hidden but well defined organic cause), but not so much for transformative therapeutic work, which rests much more on intuition, empathy and FEELING of the other person. This might explain your lackluster experience with practicing psychotherapy.

    4. Also IMHO, and related to 3, the specific theory and modality of therapy is largely a crutch for the therapist’s intuition and personality. This explains why various wacky New Age energetic healers can have transformative results even though their “theory” usually makes zero sense. Why an excellent therapist needs any theoretical model at all (and not just “go with the flow” using their superior intuition) is rather mysterious and a topic for a separate post (maybe a conceptual/symbolic system is needed as a bridge between the therapist’s “left and right hemispheres”).

    • All of these points seem correct to me except for the third. I may be reaching too closely for very recent experiences, but I just got back from a weekend at a CFAR reunion (as a plus one, rather than an alum, so it was my first real time at one of their events), and a lot of their whole deal feels like historicism-purposefulism formed around game theory and probability theory rather than psychology or religion.

      If this model is correct, then I think it explains a fair amount about why people have a tendency to see CFAR as being woo-y or culty. They’re in the same broad category as both shamans and therapists, but therapists are the only members of the category who have managed to maintain legitimacy, even in the face of the replication crisis.

      It also predicts that CFAR’s techniques won’t scale very well, and that already seemed to me like it would be the case. Much of the thing that appeared to make it work is that taking in a bunch of smart, educated people with a mostly-shared subculture and set of values.

    • clipmaker says:

      Interesting you mention Lowen. I read one of his books a long time ago, thought it was at best entertaining, and put it aside. Some years later someone tried to rob me on the subway and I got into a physical fight with the person, who ran away (nobody was injured and the robber didn’t get anything from me). A while later my legs started vibrating exactly like Lowen had described, and a hell of a lot of lingering frustration I had over the usual stuff cleared up. It came back over time of course. But at the time it was happening, I remembered Lowen and thought maybe he was onto something after all.

    • RubusArcticus says:

      strongly intellectual types rarely make good therapists

      I think most certified psychologists are “strongly intellectual types”, as modern psychology is an evolutional descendant of a strongly intellectual Europian community (which BTW was interested and studied shamanic culture), and thus one who joins this community as a certified psychologist is probably of the same “type”.

    • thetitaniumdragon says:

      Isn’t it more likely that the reason why some people get “miraculous” results is simply that some people’s problems are actually very simple, and a certain kind of person will remember the successes more than the failures/make excuses for the failures?

      For example, my grandmother was cured of smoking by going to a hypnotist. The hypnotist told her she didn’t want to smoke anymore. She stopped smoking and never smoked again.

      The simplest explanation is that my grandmother just needed to be told in an authoritative way that she didn’t want to smoke anymore, and after just being told that, she just stopped doing it and never started again. She quit cold turkey, and while it has a low initial success rate, quitting cold turkey has the best long-term success rate. So all she really needed was that person to tell her to stop smoking.

      This would probably only work one time in 10-20, but it would seemingly give “miraculous” results sometimes.

      It’d be easy for a hypnotist to simply say that my grandmother was more vulnerable to hypnotic suggestion than his other patients as a reason why it worked, when in reality, if you tell enough people to just stop smoking, some of them will do it permanently.

  7. sovietKaleEatYou says:

    There’s another possible explanation for why early adopters of a therapy are more successful (which you might have mentioned in your earlier post but I forgot). Namely, maybe early adopters got referred patients whose problems better match the ones the therapy is taylored to fix. For example it was probably the case that early on, there was a set of “standard CBT problems” and patients with these problems got referred to CBT specialists. These patients may have got slightly better results with CBT treatment than with conventional treatment. Then when “science proved that CBT is good for everything!” CBT became the conventional treatment and people receiving CBT treatment were no longer those for whom it was initially designed. I like this model because it is consistent with the assumption that psychiatrists (like doctors) overestimate the effect of what they do but still do better than random chance.

    • DarkTigger says:

      Shouldn’t stuff like this turn up, in the “possible cofunders” part of a well designed study?

      • sovietKaleEatYou says:

        Seems hard to control for. How would you do it?

      • notpeerreviewed says:

        In a perfect world, yes. I don’t know exactly what “standard CBT problems” might entail, but I doubt it would be something a typical study would capture.

  8. cernos says:

    Another possibility for the efficacy of The New Treatment: Current Treatment helps 80%, 20% get no effect. Then we start doing The New Treatment. The 20% of consistently untreated get a boost (turns out they needed the change), 80% still know the original thing that helped them, so maybe they hang on, or synthesis the old and new. On average stats go up. Time passes. Some new group appears, maybe people that have never heard of Original Treatment and New Treatment does nothing, perhaps the Synthesis Treatment is confusing. Stats go down. Some new thing is invented to fix the problem and the cycle starts over.

  9. Kaj Sotala says:

    All therapy books propose an answer: the proof is that the patients get better. But my patients do not get better. When I tell them the historical-purposeful accounts I have devised for their symptoms, they usually shrug and say it sounds plausible and they’ve thought along those lines before, but what are they going to do?

    How many therapy books say that just telling the patients about your guess is going to be effective, though?

    In my experience with these things, personal histories are really idiosyncratic. You may meet someone who looks like their issue is X; it matches the story they told, and maybe you had a similar thing yourself and it turned out to about X. Maybe the client even agrees that this sounds plausible. But then it turns out that the real problem was something subtly different, and both of you got it wrong because you were using logical reasoning in an attempt to figure it out, rather than really digging into the symptoms. Even if the guess was right, the fact that you just described it verbally rather than having the client dig into it on an experiental level means that you didn’t really activate the memory structure in the kind of a way that would have made a difference.

  10. denverarc says:

    Possible explanation – the point of writing a therapy book is not in fact to generate leads for or explain a new type of therapy, its a way of standing out in the marketplace for the authors personally.

    A re-write of something that already works with a shiny new label fits that plan just fine and dandy. Put a spin on it if you want to be an expert in a niche field. “CBT for Grieving Widows” by Will Pending – “Psychoanalysis for Sportsmen” by Andy Winmore and so on.

    Cookbooks are similar. Same basic things, re-ordered slightly to push the authors. Moloch again.

    That said, Uncommon Therapy by Rossi is worth a read. You’ll have 3 heart attacks at the sheer lack of ethics, but definitely worth a read.

  11. clipmaker says:

    Hmm I tried to post a comment citing a couple NYT articles and a book, it worked but I messed up the link formatting, and then when I tried to fix the formatting the comment disappeared. Did it go off to a moderation or spam queue? When I try to repost it says that’s a duplicate, but I don’t see it posted. Should I just wait?

    Added: one thing I notice is that self-help authors seem stupendously invested in whatever their technique is. I can sort of understand. They developed it to solve their own problem, so if the method doesn’t work, the problem is unfixable, and that idea is unbearable! They have not yet learned the art of pessimism ;).

  12. bobzymandias says:

    One possibility is that a drop in the placebo effect is through the therapist rather than the patient. When therapies are new and getting good results the therapists truly believe them and this increases the placebo effect felt by patients.

    As effect sizes drop therapists become less convinced and are thus less convincing to their patients, decreasing the placebo effect.

    Therapy success depending significantly on therapist convincingness would also fit in with the Dodo Bird verdict.

    Maybe this has been studied before – a very quick google didn’t find anything.

  13. fion says:

    Typo: “the only things that matters” -> “the only things that matter”

  14. joncb says:

    I personally will be glad when some acronym other than CBT becomes the new fad to talk about… that way i won’t keep having that moment where i wonder how anyone could think Cock and Ball Torture could solve even a majority of psychological problems. I don’t mean to kink shame, if you’re into that… more power to you, but i think being in a universe where getting my genitalia mashed was the guaranteed solution to all psychological problems would be my new definition of “least convenient possible world”.

    • sovietKaleEatYou says:

      I am amazed by how you can live in the modern world and not know the meaning of CBT. And then to have the gall to assume it’s some kind of therapy for men specifically. I don’t want to start a culture war and call out your sexism but you should feel ashamed and educate yourself about Clit and Bosom Teasing

  15. babarganesh says:

    re: the cbt graph, i wonder how much of the downward slope comes from a few pre-1987 studies? at a glance, if you started the analysis approximately in 1987, you would see a pretty level trend line.

  16. Jliw says:

    I’ve felt this exact same way, but from the patient side. Especially that last bit, where your patients nod, say “oh yeah that sounds plausible; I’ve had the same thought”, and then go on acting exactly the same — this has been my experience with every therapy or self-help book I’ve come across.

    Sometimes I feel a little like an alien. This is one of the areas that gives me that feeling; I don’t even understand the idea behind therapy, in a way. Like, of course I know that my behavior is maladaptive, and of course I’ve had long conversations with myself about why I might be doing it, but in the end it’s an emotional issue that Just Is, in my head, and I could no more change it by thinking about it than I could make myself not feel physical pain by thinking about it or asking myself “is this pain really helpful?”

    (i.e. maaaybe possible sort of with tons of meditation?)

    • arbitraryvalue says:

      I think I know exactly how you feel – a lot of my problems aren’t something I see only in retrospect. As I’m engaging in the problematic behavior, I know it is the wrong thing to do, I know what the right thing to do is, and I usually have a pretty good idea of why I’m doing the wrong thing rather than the right thing. But I still do the wrong thing.

      I figure that since I’m pretty good at self-awareness, the only big mental issues I have left are the ones that can’t be solved by self-awareness. So what’s left is fairly similar to physical pain, in the sense that the mental sensation of “I’m too scared to do X” or even simply “I intensely don’t want to do X” can be intense enough that my willpower is insufficient to overcome it.

      • Jliw says:

        As I’m engaging in the problematic behavior, I know it is the wrong thing to do, I know what the right thing to do is, and I usually have a pretty good idea of why I’m doing the wrong thing rather than the right thing. But I still do the wrong thing. […] the mental sensation of […] “I intensely don’t want to do X” can be intense enough that my willpower is insufficient to overcome it.

        This is exactly it, for me. I know all of the reasons to do (or not do) X, and I know why I don’t want to do it… but I still don’t want to do it, and that’s usually all she wrote for me.

    • dsd1886 says:

      Try “inner chld” or “internal family systems” therapy.

    • caryatis says:

      Yes. We should remember the possibility that *no* therapy works better than placebo. That’s what some of the research tends to suggest—that there’s no difference between talking to a trained therapist and talking to any other trusted person. FWIW, I was in therapy for 12+ years with maybe 9 different people, and none of it changed anything.

      • Statismagician says:

        Could you share some of the research you mention? My perception of therapy has long been that it’s an expensive, inconvenient replacement for family dinner conversation, but because of that I’d like to look at it more closely.

      • dsd1886 says:

        Maybe it wasn’t them, but you not wanting to change? Or do you mean your symptoms of depression/anxiety/whatever did not change?

  17. b_jonas says:

    I’m sorry but I couldn’t get your previous book review article, the one about post-traumatic stress disorder (PTSD) out of my head. And my thoughts match the cynical mood of your first section here.

    > And the Efficient Market agrees with my low estimation, given that therapists aren’t rushing to learn these new strategies and patients aren’t rushing to use them.

    The armies too. They’ve recognized by now that after they load their soldiers up with amphetamine, send them to obediently risk their lives, fight innocent conscripts on the other side, and see their comrades fall next to them, those soldiers get a psychiatric disorder. If there was a miracle therapy to erase that, then the armies would pay a lot for that sort of therapy, just so that they can send more soldiers to wars, then erase their traumatic memories with therapy and send them home to raise children as if nothing bad happened. The next batch of volunteers would be easier to convince to join too if such therapy existed. If it’s not worth for armies to pay for that, then perhaps the therapies aren’t as good as they advertise themselves.

    • alwhite says:

      Not sure the army analogy really works. Scarring and permanent damage are issues and you can be healed to an extent but be inappropriate for further combat duty. A soldier gets a broken leg and can be completely healed and returned to combat. A soldier loses a certain percentage of lung capacity and they aren’t returning to front line duty even though they can be healed. We don’t have this dichotomy of healed vs wounded, we have a spectrum of function. Wounds impair function. Therapy can improve function. This doesn’t guarantee you automatically return to base level.

      Then we also have to consider a replacement culture. Is it cheaper to make a new soldier or to repair a broken soldier? It’s probably the former. We have 2.1 million people in active and reserve status out of a country of 320 million, and all those people are volunteer. We can probably replace every soldier a couple of times before it starts getting harder and more expensive. We already do the replacement market in products, why not in people too? I don’t think the Efficient Market says anything about the effectiveness of therapy in this context.

      • b_jonas says:

        I think there can be an incentive for the army to heal the soldiers even if they won’t return to combat. There are a lot of veterans, so young people will meet them. They will consider what they saw in other veterans before volunteering to the army. If people generaly see that former soldiers can’t live normal lives because they have PTSD, they’ll be less likely to volunteer, so it will be more expensive for the army to buy replacements.

        But yes, you shouldn’t take my cynical statement too seriously.

  18. rahien.din says:

    You are preaching to the choir here : this is exactly what happens with new antiseizure drugs. Every one comes out looking gangbusters, but after some years of broad use, they all have the same response rate of 60% for the properly-selected patient.

    Some guesses :

    Maybe newness is itself an advantage in therapies.

    A patient can’t succeed in therapy unless they have buy in. When a person has significant need for therapy, it would seem to require an exceptional treatment. As such, the mundane-sounding “reframing, retraining, and persistence” seems like it won’t be sufficient. But “reframing, retraining, persistence” is the nutshell description of every form of therapy. And if a therapy method has been around for a long time, that’s actually worse – it’s the “reframing, retraining, persistence” that works for undifferentiated masses of people. The patient may think “But my pain is unique to me!” Or they may think “If this works for everyone, why doesn’t everyone have access to it and just do it for themselves?” They have no off-ramp. So they don’t buy in.

    But if you can genuinely describe a therapy method as brand-new and just-discovered, that resembles the One True Secret Therapy. Which gives them an off-ramp. They can buy in, and thus, their chance of success is improved.

    Or… maybe this is just Simpson’s Paradox.

    When a therapy method initially starts being used, you might not try it on just anybody. You might only use it for the more difficult cases, or you might only think to use it when something internally prompts you that such-and-such patient may respond. This is probably a good way to identify responders. If there are 100 patients, and 60 will respond to X Y or Z, 30 will respond to Y or Z, 8 will respond only to Z, and 2 will respond to nothing, then having failed X and Y does suggest that Z is worth a try. But as your success rate with Z increases, and you use Z more because of its increased salience, eventually you will hit that ceiling of 30:100. Your apparent “success rate” will drop from 80% to 30%. But it’s not that Z stopped working. It’s just a statistical effect.

    It’s as though someone handed you a box of nails and screws and a hammer and said “Go build.” And for a while you hated the screws because they are so much harder to hammer. Then, you discover a screwdriver, and suddenly you really like screws and the screwdriver. So you start using the screwdriver for everything, and swiftly find that it doesn’t work on the nails. There is no cause for alarm – there are just different tools for different jobs.

    Both of these ideas argue strongly that we should continue to devise new methods of therapy and write popularizing books about them. We can leverage newness to give our patients genuine off-ramps into mental health. We can increase the number of tools in our toolbox.

    Maybe the endgame is having a multitude of therapeutic frameworks, and individualizing patient care. The highest level of technique is to have no technique.

    • cuke says:

      I agree with this: “Maybe the endgame is having a multitude of therapeutic frameworks, and individualizing patient care. The highest level of technique is to have no technique.”

    • Steve Sailer says:

      Like I always cite as an example, echinacea tea helps me head off colds, but I’ve never seen it work for anybody else I’ve encouraged to try it. My guess is echinacea tea works for a tiny percentage of people, down around the level that it’s hard to get statistically significant results without GWAS-sized samples sizes.

      Maybe therapies and diets are somewhat similar: some work very well on 2% of people, enough to create a number of enthusiastic public advocates, but not enough to measure reliably.

  19. When you get an explosion in people with depression and suicidal thoughts, it’s not likely that telling people to adopt the right mindset is going to work. There’s obviously some bigger issue, and therapy is just a small wave fighting against that tide.

  20. alwhite says:

    I feel like the “sudden complete transformation” is an inappropriate expectation, and maybe that’s confusing things? I know all the books bring up these examples as success stories and maybe they shouldn’t because those aren’t what we’re supposed to be comparing too.

    I’ve had a few of those rapid successes with clients but I definitely put those successes on the client side as opposed to my side. In one particular case, I used a new exciting therapy that had just the right approach and perspective and I could see “the click” happen in the person’s face and everything was resolved. I saw that client 3 times and we were done. But again, I’m pretty sure that’s due to the client and their issue. The new therapy could potentially have cut to the chase faster than CBT might, but who knows.

    Have you read Carl Rogers? Specifically his book A Way of Being. He references all kinds of sessions with clients and all of his scenarios are things like “consider this client after 33 sessions”. I really appreciate the studies that Rogers did because you get to see transcripts of clients over long periods of time. I don’t feel like any magic is brought up in that stuff and it’s pretty straight forward what is happening. Then after dozens of sessions, you get to see changes in clients.

    I’m not a fan of historicism. The human mind loves stories to explain things regardless of whether the stories are true. Historicism is exactly that, a story for why things are hard now. For some people it works. Knowing that we fear opening up to people because our mom did such and such as a child, gives people the right kind of story to explain the thing and put it away. It’s hit or miss for who that will work for, and I know no way to predict that who it would work for. But I do think it’s useful to recognize the present comes from the past. A person who was punished as a child for making mistakes, today is highly defensive when they make mistakes. Understanding that behavior came from somewhere is helpful to avoid a “you’re just crazy” situation. It’s the ability to say “oh, that makes sense” and not feel as much shame about it. But that’s not a cure by any means. The real cure would be understanding the present moment experience. What do you feel when you make a mistake? What thoughts are happening in that moment? Can we normalize the situation and realize different behavioral choices are possible than the highly reactive one?

    • cuke says:

      I was going to mention Carl Rogers as well. Just fun to read too because his tone and presentation is so very different from the current flood of “my new model is amazing.”

      Rogers-adjacent and equally fun is Dibs in Search of Self by Virginia Axline. Definitely more showing than telling. I feel like that one ought to be required reading for all therapists whether they work with kids or not because it captures so well the quality of showing up for someone that is otherwise often lost in all the focus on technique, particularly when we know that technique doesn’t really matter.

  21. GranderDelusion says:

    If a therapeutic approach’s persuasive power is tied to its novelty, and its effective power is tied to the therapist’s faith in it (potentially up to 70%, if Wampold et al are to be believed), perhaps the Dodo Bird Verdict should be replaced with a Red Queen Verdict?

    Can we get improved outcomes by inventing more, more overhyped, more narratively pleasing therapeutic approaches? We would want to maximise their placebo effect before it tails off during the inevitable replication crisis, after all.

    Is the preponderance of All Therapy Books, seen in this light, an Efficient Market?

    • Purplehermann says:

      At some point people will wise up, you’d have to pace yourself and keep things within the realm of suspension of disbelief

  22. bagel says:

    In Michael Lewis’s book, The Undoing Project, about the lives of Kahneman and Tversky, one particular anecdote stuck out to me. In one of the wars (Yom Kippur?) where they came back to defend Israel, Tversky was sent to help an air force general. The general was ecstatic that he was there; “Tversky,” he said, “it’s good that you’re here, we have an important psychology question for you. We’re trying to train our pilots, and when they do well we compliment them and then they do worse! But when they do badly, we criticize them and they do better! Should we just criticize them?” And Amos Tversky blew up at him! “No,” he said, “they’re just regressing to the mean. You’re having no effect at all, so don’t be cruel!”

    So if we’re wondering if “miracle therapies” are effective, can we wonder a little further? How much do people get better without intervention, spontaneously or patiently over the years? Is it actually less than the population who seeks out professionals?

  23. alwhite says:

    Also this is a fairly recent literature review of the effectiveness of psychotherapy and about the Dodo Bird situation.

    • cuke says:

      Just adding in here, one of my favorite books along these lines is The Heart and Soul of Change: Delivering What Works in Therapy (Duncan and Miller).

      It’s not just that specific techniques don’t matter; it’s more that there are things that do matter, though individual therapists and clients might vary in how much they matter to each of them (empathy, trust, a shared story, client motivation, etc).

  24. simon says:

    I sometimes see if I can come up with these kinds of historical-purposeful accounts of my patients’ symptoms. These always fit into place freakishly well – so well that either the historical-purposeful perspective is completely true, or there is some very strong bias that makes it extra-convincing despite its falsehood. But we already know there’s some very strong bias that makes it extra-convincing despite its falsehood! That bias must have been at work in all the therapists who applied historical-purposeful narratives to autistics, schizophrenics, and gays! At some point I notice the road I’m on is littered with skulls and start wondering if I should reconsider.

    All therapy books propose an answer: the proof is that the patients get better. But my patients do not get better. When I tell them the historical-purposeful accounts I have devised for their symptoms, they usually shrug and say it sounds plausible and they’ve thought along those lines before, but what are they going to do? When I try all the exciting new therapies on them, they just sort of nod, say that this sounds like an interesting perspective, and then go off and keep having symptoms. It’s very rude!

    So I’m totally unqualified to talk about psychotherapy but:

    Seems to me that if there is a historical-purposeful account that fits what a patient tells you freakishly well, it’s likely that the reason is because the patient came up with that narrative themselves and organizes the facts in their head based on that. So, facts that don’t fit the narrative are omitted and those that sort of fit are told in a way that makes them fit more. Now, maybe this account is true anyway but since the patient has already been thinking it they’re in a rut and not getting anywhere with it (hence their reaction).

    Now, maybe if you had some pre-existing, bogus account for them, and dug for facts fitting the preexisting bogus account, you could find some facts that happen to fit it and maybe get the patient thinking outside of their rut.

  25. eigenmoon says:

    Here’s an optimistic model: this happens because therapy books work. In this model, somebody with issues tries first to read a self-help book, and only tries a therapist if the book didn’t work. As CBT books get better, people apply them more and more successfully and therefore the therapists get increasingly CBT-resistant people.

    • Steve Sailer says:


      A related possibility is that the public slowly gets vaguely aware of new ideas, which makes it harder for the doctor to spring an exciting insight on the patient.

      As a mirror image example, consider the notorious Milgram experiment in which a doctor in a white coat successfully ordered a lot of people to painfully (fake) electrocute a subject in the name of Science. Would the Milgram experiment replicate quite as well today? A half century later, maybe 3% of the population would say, “Oh, this is one of those Milgram experiments,” another 15% might say, “I’ve heard of experiments like this,” and another 20% would have a vague sense that this kind of thing just isn’t done anymore.

      In 2014, I was having dinner at the diner on Broadway on the Upper West Side that was used as the exterior for “Seinfeld’s” diner. At the next booth, there were two Columbia undergrads on a date. The smart Jewish boy started telling the pretty Asian girl about everything that was wrong with Malcolm Gladwell. He did a good job, about as well as I had done in 2005 when everybody else in the world seemed to be in awe of Gladwell. After about ten minutes of the kid’s dissection of Gladwell’s shortcomings, in fact, I was feeling sorry for poor Malcolm and wanted to jump in and defend the guy.

      So ideas do slowly infiltrate out to the public, although the rate of percolation has a lot to do with the prevailing winds of the Spirit of the Age.

      In the Gladwell case, the decline of his reputation can be dated rather precisely to his 2009 spat with Steven Pinker in the New York Times. Gladwell recognized that Pinker’s negative review of his book had used my criticism of Gladwell’s claim that NFL teams have no ability to predict who will be a successful NFL quarterback, and assumed he could destroy Pinker’s reputation by linking him to me. But Pinker is, so far, bulletproof, and Gladwell’s decision to publicly respond to Pinker’s critique and thus publicizing wound up damaging his own career.

      What Gladwell should have done is admit he overstated his thesis and then said that what he was trying to point out is that NFL teams are surprisingly unreliable at picking quarterbacks in the draft, which is more or less true.

      Malcolm has an unusual set of strengths and weaknesses, and I hope my criticism over the years has helped him come to understand himself better and focus more on his strengths.

    • zby says:

      Another related hyphotesis is that a particular therapy might work really good for particular parenting methods and those come and go in waves.

      By the way my sister has healed herself with diet and less stress and some other stuff from a particular medical problem that is officially uncurable. I have just written a post from the point of view of a patient – what is the rational thing to do when there is so much info available on the internet:

  26. notpeerreviewed says:

    I have seen the following claims, separately, in a variety of places, some on this blog:

    1) SSRIs are now less effective, relative to placebo, than they used to be.

    2) But wait, that’s because the placebo effect is becoming stronger, not because SSRIs are less effective.

    3) Actually, there’s no true placebo effect, for depression; what we’re measuring is just spontaneous remission.

    4) CBT has become less effective (presumably relative to placebo?) over time.

    Have we ruled out the possibility that the effectiveness of all treatments is unchanged, but cases of diagnosed depression spontaneously remit more often / quickly than they used to? It seems like that could plausibly happen if diagnosis practices or study populations have changed.

    • dokh says:

      It’s been three days and nobody’s commented on this and I find that odd.

      I mean, I haven’t either because I have nothing to add – but surely the idea that presentation of (studied) mental illness itself is changing, is either trivially debunked and somebody better informed than me would have by now, or worth regarding as really important and somebody in the field should be going “wait, what?” and channeling some relevant if not enormous fraction of their attention to thinking about the possible causes (both in changes to the world’s mental health landscape and in changes to how we study it) and their ramifications.

      Somebody please tell me this comment caught smarter eyes than mine as hard as it feels like it should have.

      • chridd says:

        As someone who is not someone in the field but just a random person reading this blog, the idea of mental illness changing over time makes sense to me, since things like one’s situation might affect things and those could be affected by changing culture, politics, economics, etc. For instance, it could be that, in the past, most cases of depression were random misfires that could be solved through therapy and drugs, but now due to economics or politics or whatever more depressed people are actually in bad situations that therapy and drugs can’t really fix. Or it could be that being online a lot and hearing about all the problems in the world has an adverse effect on mental health.

        This also means that if the idea of historical explanations is true for some significant portion of the population, then it would make sense that people keep finding new things that cause lots of problems and new therapies that work for them but not making permanent progress in general, as people keep finding the problems with whatever parenting trends were popular 10, 20, 30 years prior and as new parenting trends cause new problems that therapists aren’t prepared to deal with.

    • thetitaniumdragon says:

      We are diagnosing more people with “depression” than we used to.

      One obvious reason for a “higher rate” of “spontaneous” remission is simply that a lot of these people weren’t really depressed in the first place, at least not in the clinical sense.

      But we seem to have a higher level of mental disability today than we did historically, so I’m very skeptical of the idea of a higher remission rate.

      • Aapje says:

        We seem to have higher standards for more people, so it is plausible that mental disability is actually rarer, while more people fail to live up to society’s standards due to mental disability.

        Our higher standards may also be causing certain specific issues, like depression, to be more frequent, although this may go against the overall trend.

  27. TJ2001 says:

    A fishing lure must first catch the FISHERMAN…. 😉

    You would never buy a therapy book that claimed therapy on real people is complicated, it’s methods didn’t replicate well, and that “one size doesn’t fit all”…

    It’s interesting how many of the issues requiring treatment boil down to several specific cases (or combinations of them) such as:
    Medical intervention
    Developing long term coping strategies
    Official Permission to “Stop” or “Start” whatever is needed

    But the key seems to be developing the trust relationship with the counselor for the obvious reasons…

  28. kalimac says:

    To what extent does historicism/purposefulism, as Scott describes it/them here, equal Freudianism? Remember that Freud was the hot new, explains-everything, therapy a century ago. My impression is that Freud was just an early example of the same phenomenon and same way of reductionalizing the origin of problems, as all these subsequent theories. But I am not learned in these matters.

    • cuke says:

      I wouldn’t say equal so much as that there are always approaches in therapy that lean more or less on making meaning of the past and true, Freud was and still is a big influence in some quarters. Attachment theory and various trauma theories are more modern versions of focusing on history.

      I want to say more about this history question, but I’m going to put it somewhere else in a minute after I think more about it.

  29. sclmlw says:

    What you’ve described is consistent with the null hypothesis that these results were arrived at through random chance. The confusion could easily be explained by selection bias. Imagine a pool of 1,000 psychiatrists. Let’s say 1% of them randomly have lots of positive experiences with their patients (>2 SD). These 10 psychiatrists either look at the cases and find a common thread, or they started out with a hypothesis a priori. If one in ten of these therapists is creating their own methods de novo and then writing a book about it, and the other nine are following an established theory crafted by the first therapist, you’ll observe the following:

    1. New systems will come out every few years. The proponents of a system will swear by how effective it is in their own practice.
    2. As it becomes popular, a small subset of other therapists who adopt the system will start raving about how great it is.
    3. Most therapists won’t see these results (990/1000)
    4. Since no news isn’t news, you’ll hear about the successful therapists in the media, but all the therapists you actually know will be scratching their heads wondering why they can’t get similar results.

  30. Lambert says:

    If novelty really is part of the solution, maybe we should start doing it on purpose.

    “All your problems are caused by *rolls D8* your uncle being *flips coin* insufficiently *rolls 2D20* adventurous.”

  31. moridinamael says:

    All therapy books propose an answer: the proof is that the patients get better. But my patients do not get better. When I tell them the historical-purposeful accounts I have devised for their symptoms, they usually shrug and say it sounds plausible and they’ve thought along those lines before, but what are they going to do? When I try all the exciting new therapies on them, they just sort of nod, say that this sounds like an interesting perspective, and then go off and keep having symptoms.

    I wasn’t sure exactly if you were being tongue in cheek here, or how much, and my pedantic side wants to be clear. All Therapy Books not only tend to have a specific grand theory, but also a highly specific technique of intervention justified by the theory.

    Like, you’re not just supposed to tell the patient, “Hey, have you thought maybe this is all because of your father or something?” You’re supposed to get them do specific things like get themselves into a meditative and hallucinatory headspace and then visualize your adult self rescuing your child self from your abusive father, punching your father in the face and saying, “Never again!” and then imagining your child self transforming into a cloud of healing light. Or something.

    The point being that the overly-confident-seeming therapy books I’ve read would totally predict that patients would fail to get better if you just verbally point out a logical possibility to the patient.

  32. Squirrel of Doom says:

    I suspect therapy books have to follow this template to sell. So book publishers make sure to publish these kinds of books and not others.

    The market is speaking, but it’s just saying what sells Not what mends a broken mind.

  33. dsd1886 says:

    Hi Scott,

    Psychotherapist and “patient” here.

    Agree w a lot the general skepticism here but I gotta take a bit of umbrage.

    The example of Bob, the shitty you really think his issues w work don’t relate at all to his abusive father? I feel like you just strawmanned some of the more fantastical interpretations that psychotherapists can give to “explain” a behavior. You stated it was likely just a result of him being a shitty worker. But there likely is a pyschodynamic explanation or at least partly, for his issues at work, so why try to discredit completely any analysis of it, even if some therapists are too in love with their own wild interpretations.

    And you telling your patients where their issues stem from and them giving them medication…why would that make them better? Thats not psychotherapy..thats armchair interpretations and medicine. Are you doing psychotherapy with them as well?

  34. Dan L says:

    Therapy books are often written by the researcher who invented the therapy. I imagine if you invent a therapy yourself, then it perfectly fits your personality and communication style, you believe in it wholeheartedly, and you understand every piece of it from the ground up. You’re also probably a really exceptional and talented person who’s obsessed with psychotherapy and how to make it better. So maybe they get results nobody else can replicate?

    This rings true – an advisor of mine was friends with Francine Shapiro, and was privately of the opinion that EMDR was nonsense propped up by her formidable personal expertise. There might be quite a lot of value in studying the methods of “rockstar” therapists (Peterson comes to mind too) but that doesn’t necessarily mean their assessment of why their technique works is accurate.

  35. sourcreamus says:

    Maybe this is a combination of CBT in the Water and your comment in the PTSD post about psychological problems being symptom based.
    If one therapy is great at treating one pathology with one cause then the people who use that therapy on people with that pathology will have great results. However, people with those symptoms and a different pathology are not affected. Thus when it is first introduced it works great among the subset of patients who have that symptom caused by that problem and not at all among the patients who have that symptom caused by a different problem. Over time the solution leaks out of therapy into the mainstream and people who have that problem try the solution on their own and only the people who it doesn’t cure go into therapy.

  36. belvarine says:

    Maybe different techniques are more effective on different people (depending on all kinds of factors) and instead of trying to find The Method we accumulate a store of methods alongside vigorously tested systems to match people with the most effective technique for them?

    Why can’t we train psychs to rotate through therapy techniques until one sticks like they do with medications?

    • cuke says:

      We do know some things about this, though not enough for sure. To a certain extent clients sort themselves this way towards the therapists who match their preferred style. And then on the therapist side, I can say that the best ones I know are very flexible and adaptable from one client to the next.

  37. William75 says:

    I had a similar experience with books on insomnia. Maybe that already fits within the category of therapy books. Anyway, I was suffering from debilitating insomnia, and I remember thinking as I went home in the evening that all I wanted out of life was the ability to get a good night’s sleep.

    Among the things I tried was a book that perfectly fits Scott’s description. It was full of absurd “case studies” where someone was suffering from years of horrible insomnia, tried the book’s method, and was nearly instantly cured. I didn’t take any of them seriously. However, a weird thing happened – my insomnia got better. I wasn’t instantly “cured”, and I still have sleeping problems at times, but I got a lot better. I don’t know if it was the book, or if I just cycled out of it and stopped stressing over so much.

  38. Ozy Frantz says:

    In the case of autistics and homosexuals, there’s a plausible story in which the observations of their parents are accurate.

    Autism is 80% genetic; therefore, parents of autistic children are more likely than average to have autistic traits. That leads them to appear cold and distant.

    Homosexuals are disproportionately likely to be gender-non-conforming in childhood. Parents in the twentieth century tended to be uncomfortable around gender-non-conforming children, and often cruel to them; even if they were not, a father might have a hard time bonding with his son if the father likes hunting and the son likes fashion magazines. Parents were particularly uncomfortable around GNC boys. As one would expect, distant/uncomfortable/abusive fathers are blamed for male homosexuality, while female homosexuality was not believed to have a clear cause.

    This suggests that the historicist observations may be true even if the causality is not.

  39. clipmaker says:

    I don’t know about lasting change, but there are definitely some interventions that (sometimes) have very powerful effect even if it the effect dissipates after a while. EMDR for PTSD is one of these, another is ketamine infusion for depression, another (according to this reddit post is stellate ganglion block (some kind of injection into the neck) for PTSD.

    Some of this stuff can be observed in the lab, e.g. there is an NYT article claiming [Freudian transference can be observed with fMRI.

    Maybe the really lasting changes amount to personality changes, that happen at certain life stages, but that amount to something like a collapse of Jenga blocks that can be triggered by the right stimulus at the right time. So some of the therapy success stories come from that, but other people experience it through just normal muddling through life.

    I’ve been wanting to read Allan Schore‘s book, The Science of the Art of Psychotherapy (2012), since he has done a bunch of work on neurobiology and trauma, affect regulation, etc. But my library doesn’t have it. I’d put in an interlibrary loan request except a lot of the time I don’t get around to reading the book before it is due for return.

  40. dark orchid says:

    Scott, would you say your update part to the The Body Keeps The Score post is the same kind of thing that others are calling incredible instant cures? The bit about someone having years of therapy, then suddenly displaying signs of being cured and prompting you to wonder whether it’s all wizardry.

  41. AllAmericanBreakfast says:

    I am surprised that you’re assumption is that the placebo effect works in these cases by inoculating people against the therapeutic technique. My assumption is the reverse. People make friends with people who have similar outlooks and issues. They refer each other to the same therapists, who are drawing on a Kuhnian paradigm to devise their therapeutic approach, a paradigm established by the community they serve to explain their own problems. When a member of this population feels their therapist really gets it, and then experiences a miracle cure, they refer their friends, who come in with an expectation of the same, making the therapy work better.

    By contrast, when the technique is studied on other populations by other therapists, there’s little or no deep awareness of the technique or paradigm in the population under study. No tribal sense that the therapist is One of Us. So it doesn’t work as well.

  42. sty_silver says:

    I have undergone therapy, and it completely changed my life immediately over the course of two years. It wasn’t a behavioral therapy, though. The structure was something like “find some symptom, do some directed self-exploration to identify an underlying cause, attempt to understand the underlying cause”. Repeat every session. Almost all in dialogue form.

    I also didn’t have any diagnosed mental problem nor was I assuming that I was particularly more in need of therapy than most other people when I started. Rather, I had been convinced that most people ought to undergo therapy and they just don’t. If I hadn’t done it, I would still be alive today, but very likely be significantly less happy and less successful.

    Despite this, I’m not super optimistic about therapy as a tool. If studies say it’s mildly effective on average, II don’t doubt that. I am a success story but most definitely not a representative case.

  43. kai.teorn says:

    Corollary: Every therapist needs to found a therapy school of their own. Then we’ll finally be getting consistently amazing results!

    Seriously, compare this to art. We don’t (generally) perceive art as declining because we mostly read/listen/watch to new creations performed by their original authors. If e.g. music consisted of a handful of songwriters and thousands of singers performing someone else’s songs, we would probably also be wondering why the same song that so inspired us 20 years ago sounds so stale now.

    Another thought: expansion of the universe of patients. There’s more depression diagnosed now than 20 years ago. What if this is not genuine growth of the disease but just gradual persuading of more people that they are ill? If these new patients needed extra effort to persuade that they are ill, I would expect them to also require more effort to persuade them they got cured. Which would look to us like a gradual decline in efficiency of all treatments.

    • belvarine says:

      Corollary: Every therapist needs to found a therapy school of their own. Then we’ll finally be getting consistently amazing results!

      I am 100% on board with treating psychotherapy like kung fu.

    • sidereal says:

      Is there actually such a thing as “genuine depression”? I mean, unlike other diseases, mental illnesses are less.. objective. It’s like a hardware bug vs a software bug – depends which software you are running, what the goals are. Couldn’t it be that depression is both “genuinely increasing”, and socially mediated?

    • Kevin Carlson says:

      But of course “a handful of [composers] and thousands of [musicians] performing someone else’s [pieces]” is precisely what we have in classical music, and while classical music certainly doesn’t have the broad audience of pop music, it would be hard to argue that a key reason is that people are tired of Beethoven’s Seventh. Great pop songs also seem to have a lot of success in covers, e.g. Bob Dylan’s catalogue. The successes of great composers and great scientists are replicable; the successes of great personalities, whether in psychotherapy or popular art and music, maybe not.

  44. Selentelechia says:

    My own stab at an explanation here:

    There is no such thing as people using the same words or concepts to describe precisely the same internal [thing]. Therapy frameworks are not useful because they should scale, they’re useful because they’re a puzzle piece. [edit: this sentence is clunky and I’m not sure how to reword it]

    I disagree with this advice:

    If it’s just that study quality gets better and better until we realize how crappy the exciting new therapies really are, you might as well get the boring old therapies. At least insurance probably covers them.

    I’d say, try many until one resonates. Or until you realize that what resonates isn’t in a therapist’s office but in the woods, in psychedelics, in weightlifting or journaling or whatever. My approach to therapeutic frameworks is like the part of Gendlin Focusing where you find the verbalization that “fits” the “felt sense.” You’re just sourcing that verbalization from someone else. Roll with it while it resonates. Move on when it doesn’t. Some people will find one that resonates and continues to do so for a long time. Many won’t.

    And, on the “meta” level, solutions, accurate explanations, and precise descriptions of your internal life are moving targets. There’s a book about manufacturing processes called The Goal, in which the author describes an approach to efficiency that emphasizes identifying bottlenecks and constraints, and prioritizing them. The thing that struck me in that book was that the constraints and bottlenecks shift constantly. I think it’s similar for therapy – what resonates and works is highly specific to the individual, and for each individual, in some degree of constant flux. I’m not optimistic about finding something that scales when applied by lots and lots of therapists. Something may scale for an individual therapist, contingent on their skills, intuition, how good they are at modifying what the communicate to the patient such that they hit on what resonates.

    • cuke says:

      I agree with this. There are some aspects of therapy that “scale” — listening well, using words thoughtfully, being present in a particular kind of focused, attentive way, asking good questions, a certain kind of perceptiveness, etc. But the whole thing depends on a therapist being able to be pretty genuine and flexible in the moment, and it’s very hard to do that if you’re counting steps to the music (ie, overly focused on implementing a technique or overly attached to a story about what’s going on). And that means that therapy is going to necessarily feel pretty different from one therapist to another, even if they’re allegedly using the same approach.

      One of the other research tidbits I like is that there’s little evidence that a particular therapy modality has better outcomes if the therapist is showing a very high degree of fidelity to the technique versus only kind of halfway sorta using the technique. It’s not like fluency in a language that way or getting just the right flour measurement for the cake. It’s way more particular and fuzzy than that, and way more dependent on the quality of the interaction than on some trick that the therapist is really practiced at.

  45. Alex M says:

    Imagine you’re a therapist. You decide to secretly spy on a Trump voter. After all, “Trump voters are crazy,” so obviously this is a psychological condition in urgent need of a cure.

    After a few weeks, your patient gets very aggressive, almost violent. “Stop spying on me, you garbage,” he howls at the walls. You become very concerned about his symptoms, since “talking to nobody” is often a clear sign of mental illness, and start writing short stories on Reddit that reference his case in a very allegorical way, hoping you can heal his anguish.

    Sadly, this doesn’t seem to do the trick. The patient becomes increasingly aggressive, and starts saying things like “I’m not your lab rat, you retarded filth. Yes, you, the person secretly spying on me. I’m talking to you, you garbage.” The fact that this patient is holding detailed conversations with imaginary spies worries you and you add “possible schizophrenia” to your diagnosis. You decide to interact with him more regularly, creating dozens of Reddit accounts to reference him, but only in an roundabout way that can never be proven directly by him. After all, you don’t want to be held accountable for breaking the law and violating your patient’s civil rights when you’re only trying to HELP him.

    The patient then goes on to design some interesting techniques to sway political campaigns and uses them to support whichever candidate promises to throw people like you into death camps. They also create a religion focused on murdering intrusive therapists and begin disseminating loose blueprints for weaponized AI to their followers in order to help enact such processes. While you are impressed at his creativity and intelligence, it dismays you that such talent is wasted on somebody who is so obviously nuts.

    Ten years later, you and your family are in a death camp in the newly fundamentalist United States of America, having been convicted of spying illegally on somebody for therapeutic purposes. Before you walk up to the guillotine, you can’t help reflecting on how tragic it is that you were unable to cure your unsuspecting patient of his delusion that somebody was spying on him.

    The reason I bring this up is because a lot of therapists remind me of this parable. Many therapists are deeply damaged people, and whenever one of their patients has a viewpoint, philosophy, or way of life that they disagree with, they often choose to view this disagreement as a sign of mental illness in the patient rather than a flaw in themselves. Personally, I think that no patient should ever be treated by a therapist less intelligent than they are, because it’s too difficult for the therapist to distinguish between a situation where the patient has a legitimate mental illness or a situation where the patient is simply smarter than them and has a more accurate view of reality.

    • Garrett says:

      FWIW, I’ve never encountered a therapist like this. From what I know of their training, it is to focus on what is functional or impacting the patient’s life. It primarily focuses on ensuring that a patient can fulfill activities of daily living and then on achieving the patient’s own goals, as long as they aren’t actively harmful to themselves or others. If someone doesn’t want friends because they have a happy and fulfilling life by themselves, that’s not a problem. If someone avoids making friends because they are afraid of rejection or something and is unhappy as a result, it is something to consider addressing.

      • cuke says:

        I agree with this view and it’s consistent with nearly all the therapists I’ve seen as a client over the years, whether I found them to be helpful or not, mediocre or fabulous. I have sat as a client with one therapist who I found to be highly unethical in this way and I’ve heard my clients describe interactions with therapists that sound highly unethical in this way, but that behavior is counter to the ethical codes and training for the job.

      • sharper13 says:

        I’m closely connected to someone who has had this experience, to the point where while she had a lot of success with a therapist who “understood” her due to a similar cultural background, she’s since moved and given up on finding another one because all the ones she tries insist on judging her based on their opposed beliefs.

        To use an example people here will recognize (based on the recent open thread), she’s a committed home schooler and a “typical” therapist might not be able to disguise their adversion to “people like that” and if they do, they aren’t able to understand the situations/worries/etc… which might go along with life for someone who embraces it. Throw in 3-4 more things like that which a typical therapist doesn’t “get”, and it’s tough to find someone to really listen.

  46. J Mann says:

    Is it really the case that all talk therapies are about as effective as each other? Like if I know someone with apparently borderline personality disorder, she is as likely to benefit from Freudian therapy as DBT?

    • alwhite says:

      It’s called the Common Factors model if you want to research it. People will claim it’s not the case but those people are usually biased towards their favored approach.

    • Bugmaster says:

      If so, it sounds like maybe most people just need a sympathetic ear to talk to; someone who will listen and support them without automatically judging them. In the modern world, it takes a licensed professional to do that…

      • alwhite says:

        Active listening is actually hard to do and is exhausting. It’s not just that I listen without judgement, I listen to completely understand without injecting my own stuff. I’m supposed to reflect yourself to your self without mixing it up with myself. It’s kind of hard but it looks easy on the outside

  47. ajb says:

    One point which is worth mentioning is that general therapists have different incentives to ‘school founder’ therapists and their disciples.
    At least in the UK, the incentive structure for both private and publicly funded therapists is poor. Publicly funded therapists are under huge pressure to maximise throughput, which means a standard course of CBT is 6 weeks. Private therapists are paid as long as their patient does not recover, and have zero reputational risk due to the stigma of having a mental health problem in the first place.

    ‘School founder’ therapists presumably want recognition for successfully treating patients, and therefore have a strong incentive to at least declare patients cured, and perhaps even to actually cure them. Early adopters could well have more motivation than normal to cure their patients.

    This may seem like a very cynical view, but please note that everything I have said is literally true. Of course, weighted against this is the fact that most people who train as therapists genuinely want to help people. But they are still human beings. Incentives work in insidious ways. I’ve sometimes wondered whether it would make a difference to change the payment structure somehow – maybe hypothecate some fraction of the hourly rate, and allow the patient to award it at the end (or not).

  48. Swamplandflowers says:

    Does all of this just indicate that storytelling is the fundamental underlying mechanism for the effectiveness of therapy?

    That there are perhaps some forms of mental illness that can be remediated through teaching the subject to tell themselves new stories about themself and the world? It would fit with two of the facts that you’re pointing to here – 1) Therapists with powerful charisma can forcibly impart a new story to the patient, and 2) the ‘newness’ of a therapy can carry the energy that helps push the patient into a new activated energy state.

    The analogy that I’ve always found helpful when fighting with (admittedly undiagnosed) depression in my own life is — working through difficult mental states feels a lot like working through the ‘energy of activation’ curve in a chemical reaction. I’m stuck in a local trough that is a depressed, frustrated, or angry state. I know there is an overall lower energy state I can get into that is less anxious or frustrated, but I’ve cut a groove in my current local trough, and there is an activation energy required to get out and into that new groove. Once I’m in the new groove it takes far less energy to maintain it.

    And fundamentally those grooves are stories that my mind is telling itself. The activation energy may come from me, or it may come from a friend who invests energy into me to help push me out of the groove. Or I might get help from an outside idea or technique that excites me.

    The other thing that seems to be necessary in addition to the activation energy is for the underlying theory to include a technique that makes it self-reinforcing. You need the charmisma-lift from your therapist to get out of the original local-trough. And then once in the new, healthier trough, you need a technique that helps you make the smaller (and lower energy) adjustments required to keep you safely in the new healthier state.

    All of this also feels like we’re in the realm of talking about buddha-fields. Another mystic concept that can be applied metaphorically without relying on any mystic mode-of-action, to good effect to understand the social force of how we push and pull on each other.

    • cuke says:

      Lovely phrase here: “teaching the subject to tell themselves new stories about themself and the world?”

      I do think that’s a big part of it.

      And this is well said too: “You need the charmisma-lift from your therapist to get out of the original local-trough. And then once in the new, healthier trough, you need a technique that helps you make the smaller (and lower energy) adjustments required to keep you safely in the new healthier state.”

      One of those techniques is telling a new (usually kinder) story about oneself (and the world). In my experience, it’s also more like a whole bundle of tools rather than one technique. And it depends on the person and their situation of course.

    • Doesntliketocomment says:

      I think this is an amazing (dare I say life-changing?) comment. If we can take for granted your behavior makes sense in your current self-narrative, then the heart of making a life change is having a different narrative. But it’s hard for most people to make up a story and then go on to believe it, so they need outside help. It frequently helps if the story is novel, as it is easy to be skeptical of a story you’ve heard applied to others a dozen times before.

  49. OxytocinLove says:

    I’ve never really understood the historical-purposeful approach to therapy anyway… Like, sure, maybe all my problems are because my mother was a hamster and my father smelt of elderberries but what am I supposed to do about it now?

    • cuke says:

      I haven’t visited this research in awhile, but there is research on trauma recovery that seems to show that it’s not the severity of the trauma that determines impairment or recovery down the line but how much support/care did the person get at the time of the trauma (or not) and to what extent has the person “made sense of” the effects of that trauma in their present day life.

      So it’s not so much dwelling on the harm that was done in the past as it is noticing and naming the impact now in one’s present life and learning ways to accept and cope with that impact. Along the way, it’s also helpful to clarify goals and values for one’s life because then those become guiding lights rather than the shadow of something that came before that one had very little to no control over.

      I want to clarify here I’m speaking mainly about trauma and adversity that the client clearly identifies as a source of pain or distress. I am not talking about some pet theory the therapist has to explain whatever they think is going on with the client/patient.

    • Nancy Lebovitz says:

      Reorient to a more accurate view of the world.

    • Aapje says:


      Presumably, the idea is to try to change the coping mechanisms to something better or reduce them. For example, if a person is still very angry at a parent and takes that out on others, have them write a letter to the (dead) parent and perhaps deliver (or ‘deliver’) the letter, to provide some catharsis/vengeance, while simultaneously pointing out how unfair it is to be angry at people who aren’t to blame.

      You obviously can’t undo the past, but you may be able to change how you cope with the past in the now.

    • Doesntliketocomment says:

      Well, you can reconsider if it’s reasonable to be mad at your girlfriend for not liking the cardboard tube you got her.

  50. Andrew Hunter says:

    I have a strong suspicion that the majority of therapists are essentially rainmaking you.

    There are definitely good people in the business, I think I even went to one briefly (then immediately moved out of that city before I was “done”; drat.) But it seems the median one, possibly without knowing it, are banking on the fact that likely good things will happen to you. Most people’s mood reverts to the mean, or at the very least you’ll get a tick up in the random walk, and you’ll attribute it to them.

    There’s also substantial principal/agent problems: my therapist doesn’t remotely benefit from me getting better. (In fact, like a dating website, they actually suffer.)

    I have largely concluded that unless you have a well-founded reason to believe a given person is highly atypical and better–which reasons are hard to come by, not least because I am not an expert myself, and can barely collect data without paying $200/hour to do so–you should flatly assume any given therapist is stealing your money.

    • cuke says:

      You may be totally right in your assessment about the majority of therapists, I dunno.

      I would like to push back a little on the therapist not remotely benefiting from you getting better.

      I turn away more patients in less than a month than I have total in my practice at any given time. I don’t need to hang on to anyone. This is the situation among many of the therapists in my area.

      On the other side, if you consider therapists to be just like regular people, which we are, it’s terribly demoralizing to feel that one is utterly failing at one’s job, which is how it would feel if no one got better. That would kill me and I’d change careers really fast.

      I am sure there are cheaters and scammers in the therapy world just as there are everywhere, but the work is hard, stressful, and not that well paid. The reason I keep at it, and the reason I hear from many of my colleagues, is that it feels very meaningful to have the sense of making a difference in a way that feels like matters.

      • Andrew Hunter says:

        My mental model of this is mostly not mustache-twirling [1] therapists who are actively deceiving you. I don’t think you’re a knowing villain. These people probably think they’re helping, but they’re not measuring it, nor are they incentivized to.

        [1] Well, most are women, aren’t they? What do women twirl? (Hair, but it hasn’t the same connotations.)

    • Don_Flamingo says:

      Pretty sure the value in rainmaking lies in the respectable, calm, fatherly/motherly person hearing out your personal problems (but specifically not one, who’s part of your personal life, someone who’s apart from it, neutral).
      A distressed, dejected chimp will feel better if they can hang around a neutral, calm elder chimp.
      Everybody needs a Hare:

  51. cuke says:

    I thought this was hilarious and I’m glad you wrote it. I also feel this way reading the latest therapy technique books, and then I often wing them across the room at the wall because the obligatory attempt to try to anchor the theory/technique in science or even better, neuroscience, is so so lame.

    I do think therapy is more art than science. We could have a long conversation about how that is and what the implications of that are, but I’ll just let that sentence sit there. As an art form, there is science showing that it is effective in helping most people with a wide variety of presenting concerns.

    About looking to a person’s biography to explain causes for their present-day troubles, I want to distinguish between these things:

    1. A therapist providing unsolicited interpretations about causes in a kind of authoritative old-school Freudian way and
    2. A therapist and client in the context of a trusting relationship looking together at earlier experiences with some curiosity about whether they are relevant or not and how they might be.

    To me, there’s no need to foist causes onto a client. If you follow a client where they want to go — and if they are motivated to change — then they will tell you how important their past is to them and how important it feels to them to make sense of that past as a way to help the present. It’s not for us to say.

    There are people who come in and say, “I’m having trouble with sex in my new relationship because I keep flashing back to when I was raped…” and there are people who come in and say, “I’ve always been anxious, as long as I can remember, and it’s really getting in the way of me figuring out whether this job is a good fit for me” and there are people who come in and say “I haven’t been happy for a couple of years and I don’t know why and my girlfriend asked me to come talk to someone.” And a million other things.

    In the face of this variety, there is no one way to relate to the question of “how relevant are earlier experiences to this person’s present concern?” There’s just a lot of talking and curiosity about their experience and trying stories on and seeing what they come back with and learning what seems to light them up, and so on.

    We genuinely don’t have a f*ing clue what this person in front of us needs and we genuinely don’t know “for real” what’s “really” going on with them. So it seems we have to be incredibly careful not to tell our own stories about them, that the story-telling has to be mutual, and that we have to test with them our notions in a very tentative sort of way, so they have a lot of room to consent or not to whatever story is being told. Very often, harm has already been done to a person by the stories some other person in authority has told them about themselves, that has in some way limited in their own mind who they get to be.

    Having said that, I do see a whole lot of change. Almost none of it is instantaneous. But it’s also easily observable. When I don’t see change, it’s usually because I am not connecting well with that person or because the person is very well-defended for one reason or another — ie, removed from their feelings, shutdown, mistrustful, avoidant, very angry but unable to own their anger. And even then, I usually see change eventually; it’s just slower to come is all. Sometimes change happens very fast, but it was slow in coming — I see that maybe most of all.

  52. Anonymous` says:

    Are therapy book authors blatantly lying? I try to have a really low prior on this sort of thing, but I’m not sure.

    Damnit Jim, you’re a doctor, not some kind of niceness-to-academics maximizer. At some point you have to say “fool me a thousand times, shame on me” and stop being so trusting, to help your patients.

  53. Lucid Horizon says:

    “while homosexuality is mostly not genetic, few people today think this is a plausible explanation.”

    Excuse me, while I realize this is not the thrust of the article, I would very much like clarification on this. Surely we are not back to “homosexuality is curable actually”?

    • Protagoras says:

      Though people often confuse things being produced by environmental factors with things being easily changed, there is in fact not even much correlation between the two categories. And if it were genetic, it would be in principle eventually be “curable” by some kind of gene therapy anyway.

    • thetitaniumdragon says:

      Being an amputee is not genetic, but it isn’t curable, either.

      We don’t know what causes homosexuality. Not knowing that makes it very hard to say how alterable it is, though the fact that sexual fluidity exists suggests that some people can, in fact, be shifted. How, however, is unclear.

  54. jasmith79 says:

    To explain the ‘miracle cure’ phenomenon I propose the following conjecture:

    Dumb luck.

    Some small percentage of patients are one or two key insights away from magically putting all of their problems into perspective. Insights that pretty much any reasonably skilled therapist could provide in a handful of sessions. There are a bajillion practicing therapists. Most of them will go their entire career without having a patient like that. Some lucky few will have a handful over the course of decades. And once every generation or two, purely by chance, a therapist will have an improbable streak of such cases.

    To the plucky therapist in question, the obvious answer is that they are a genius who has invented a bold new extremely effective form of psycotherapy based on whatever random commonalities their lucky streak of patients had. From the god’s-eye view, it’s just luck.

    I don’t know that such is the case, but it makes at least as much sense as some of the alternative theories.

  55. beleester says:

    But that still raises the philosophical implication of it being possible, for somebody, to consistently produce dramatic change through therapy. This still bothers me a lot.

    I don’t mean this as snark, but hasn’t this been a pretty core part of AI risk theory for a long time? The idea that, just by having a (superhumanly competent but otherwise ordinary) conversation with the AI, it would somehow be able to work its way through whatever mental resistance you have and convince you to let it rule the world? Doesn’t that fit into “consistently producing dramatic change”?

    Now, a really good therapist isn’t superhuman (by definition), but if they’re really practiced at a particular type of therapy that works particularly well for a particular type of patient (and under much less adversarial circumstances than an AI in a box), they can probably scrape the ceiling of what it’s possible to do in an ordinary conversation.

    • kai.teorn says:

      From my experience, the more normal (in the psychiatric sense) you are, the less likely you are to want to rule the world or to have it ruled by someone at all. So it sounds like your devious AI, if it is successful, will be harming humans (making them less normal in the psychiatric sense) more than curing them. Making people less normal is generally much easier, and is already easily demonstrable, so the question of whether an AI will be able to do that is kinda moot.

    • sty_silver says:

      I would say it is a core part of the AI risk argument – if humans could resist persuasion by a superintelligent AI, I would be less worried.

      But I don’t think this is meaningful evidence. Therapists are not even trying to optimize for persuasion. I think, clearly, the AI boxing experiment gives you much better evidence since it’s a much closer analogy. (This is where two people meet in a chat room, one plays the AI, the other the gatekeeper, the second guy has to respond to what the first guy says, and he wins if he doesn’t write the sentence “I decided to let XX out of the box”). And what we find there is that the gatekeeper loses a substantial amount of time, and if asked about their beliefs on whether they could resist an AI before and after playing, they usually update downward massively.

      Even if that was different, though, an AI could be much much much more persuasive than the most gifted human. And it can make credible threats that the human can’t. In the AI boxing experiment, the gatekeeper doesn’t actually have to pretend like the other person was a real AI, that’s not part of the rules. So any threat like “I’ll torture you if I go free” or “I’ll simulate versions of you and torture them” doesn’t do anything.

      • hls2003 says:

        I know it’s off topic, but “I’ll torture your simulation” seems a lot like this. Except less convincing, because the AI is presumably still “in the box” and if “the box” means anything, it must include limited computing power, which means that near-infinite computational resources can’t be expended on such simulations anyway. “Give me what I want or I’ll vividly imagine you being on fire” just doesn’t have the same persuasive force. And if the AI says “how do you know you’re not one of them,” it seems like the answer is simple: if you could do it you wouldn’t have to talk about it.

      • beleester says:

        But I don’t think this is meaningful evidence. Therapists are not even trying to optimize for persuasion.

        Which is harder, causing someone to make one bad decision or causing a long-term shift in someone’s personality? I would think the latter. And Scott seems to agree, seeing as he believes in the former but has trouble believing the latter.

        I’m not sure where I stand on the AI-risk thing, it’s more noting an inconsistency – either you can in fact make dramatic changes with words alone, in which case it shouldn’t be surprising that a therapist is capable of doing so, or you can’t, in which case it should be surprising that an AI can.

        Yes, an AI can optimize its words better, but they’re still words. Unless we’re positing some sort of Snow Crash-style neurolinguistic hack, the AI is not doing anything qualitatively different from a debate-team champion or a therapist. It’s perfectly possible for the therapist and the AI to come up with the exact same words (even probable, if the words they need to say are something a human can easily think of), and if they do, their words will have the same effect on the listener.

        In the AI boxing experiment, the gatekeeper doesn’t actually have to pretend like the other person was a real AI, that’s not part of the rules. So any threat like “I’ll torture you if I go free” or “I’ll simulate versions of you and torture them” doesn’t do anything.

        This is the sort of thing that makes me suspect that what the AI-box winners have been optimizing for is not persuasiveness, but the ability to come up with philosophical arguments that are weird enough to attract the interest of the sort of people who play the AI-box game.

        The AI can say “I’ll torture you if you don’t let me go,” but it cannot prove such a statement, since it’s in the box and its only means of communication is a text channel. It makes no difference whether a human or an AI is the one issuing the threats – either way the listener only receives it as text on a screen. If for some reason the ability to issue computer-like threats was a vital part of their toolkit, the therapist could just say “For this session I want you to imagine that I’m a supercomputer that knows you better than you know yourself…” and they would have just as much proof of that as the AI does.

        • sty_silver says:

          I personally suspect an AI would take maybe a couple of minutes to persuade to you do literally anything it wants. I think you re vastly underestimating the space of possibilities, and I completely disagree that somehow a therapist’s abilities will be comparable to an AI’s. It’s qualitatively similar in the same way that 100 and 10000000000000 are qualitatively similar.

          But I brought up the AI boxing experiment because it’s actually empirical evidence on this. So that you don’t have to take my word. People go into that experiment thinking they have 80% chance to resist an AI or whatever and come out thinking it’s 0.1%. You are simply overconfident because you lack imagination of what the AI can do. You cannot conceive of an argument that would make the simulation thread scary.

  56. ksvanhorn says:

    “while homosexuality is mostly not genetic”

    Wait, what?

    I know it’s a sideline to the topic of the post, but I was surprised to read the above. I would be interested in hearing a bit more.

    • Douglas Knight says:

      Bailey-Dunne-Martin 2000 surveyed the Australian twin registery and Långström-Rahman-Carlström-Lichtenstein 2010 surveyed the Swedish twin registry, with similar results. About 80% of gay men with an identical twin have a straight twin, compared to virtually all of gays with a fraternal male twin. From this a broad-sense heritability of H²=40%, which is higher than I remembered. (The Swedes had 10% concordance for fraternal twins, but the same heritability? I’m confused.)

    • cedrus_libani says:

      There have been two major studies of genetic markers for homosexuality. One in the UK Biobank cohort, one in the 23andme cohort. Similar ethnic mix, but the UK cohort is significantly older; these people were coming of age while gay sex was still illegal. The 23andme cohort is mostly tech-savvy Americans with disposable income.

      In the UK cohort, there are markers that are significantly enriched in the gay population, but this finding doesn’t replicate in the 23andme cohort. The working hypothesis is this: the more social disadvantage is attached to being gay, the more the population of actively gay people is biased towards people who have a high sex drive and/or low impulse control. These “gay” markers don’t actually predict orientation; they predict whether someone will stay in the closet, given that their choice was made in 1950s-era Britain.

      Note that there’s a very wide gap between “no genetic markers” and “no biological basis”. But yeah, no genetic markers, at least in a cohort believed to have a minimal closeted population.

  57. Douglas Knight says:

    But the therapy books still confuse me. They’re full of stories of incredible instant cures, with the authors assuring us that these are all real and typical of their experience. How can you get this from merely “stretching the truth”, as opposed to outright data falsification? Are therapy book authors blatantly lying? I try to have a really low prior on this sort of thing, but I’m not sure.

    You really shouldn’t have a low prior on this sort of thing.

    But, as you said in Different Worlds, psychodynamic therapy is supposed to induce “dramatic emotional meltdowns” and you failed to do it. Maybe you’re the odd one out and everyone else is inducing them. Maybe they don’t do any good, but it’s pretty easy starting point for fooling yourself that you have a technique that works abruptly.

  58. Qiaochu Yuan says:

    The obvious hypothesis to me is that therapeutic methods become less effective over time, because people’s psychological defenses adapt to them. Thoughts?

  59. Edward Scizorhands says:

    Where’s the Amazon link to purchase All Therapy Books?

  60. acedeuceblog says:

    The prevalence of depression increased by an order of magnitude since 1970 (see figure 5: and therefore its distribution of pathophysiology probably also changed owing to new environmental causes that caused this epidemic (Anatomy of an Epidemic might be on to something) Depression is essentially umbrella for many different things that might have gone wrong in the brain and have similar symptoms but might not respond equally to different treatments. CBT might work great when you’re recovering from a bad relationship but it might not work as well when your brain has been fried by binge drinking, extremely sedentary lifestyle, poor diet, benzodiazepines destroying gaba receptors, and other more physical causes. Any form of therapy would probably lose by a mile if compared with joining a casual soccer team and fixing your diet.

  61. drethelin says:

    What if people are really quite variable, and every time a new approach is tried there is a population of patients that it fits perfectly so it seems miraculously useful and this population of course goes down to 0 as a therapy becomes common enough

  62. Reasoner says:

    Someone should use machine learning and train a model to predict what therapy will be most helpful for a particular person with a particular condition. (Maybe a particular therapist too?)

  63. armorsmith42 says:

    > Most therapy books share some assumptions, so deep as to be unspoken: current problems serve some purpose related to past traumas.

    This sounds like a dramatic assumption for books to be making. Like imagine if someone was hard-of-hearing and you assumed they’d been to too many rock concerts. Sure, that *might* be true…but couldn’t it also be true that they are missing some part of their ear from birth or that they are from Japan never learned to process the “el” phoneme reliably?

    Are there any books which assume that problems result from a missing mental process that can be learned?

    • Kaj Sotala says:

      Well, many of the books which make the trauma claim also teach learnable mental processes for fixing the effects of those traumas, so it’s not like these are mutually exclusive.

      As for why exactly they make that assumption – well, suppose that you had a special medicine which was useful specifically in the case of fixing hearing damage from rock concerts, and then you started investigating your patients to see whether they would be a good fit for this medicine, and then it empirically turned out that 95% of your patients were in fact helped by the rock concert cure. At some point you would probably start assuming that most cases of hearing damage will be because of rock concerts (or at least treatable using a medicine which was custom-tailored for rock concert damage) even if this seemed implausible to you at first.

      Putting that aside, there are also plenty of schools of therapy which involve teaching e.g. emotional regulation skills.

    • cuke says:

      About “current problems serve some purpose related to past traumas” — I would amend this a little to say that many therapeutic approaches see “present-day maladaptive coping behaviors as efforts to solve problems created in an earlier time.”

      It’s not that mental illness, writ large, serves a purpose related to past trauma. It’s that in the face of difficulty, we often turn to not-so-helpful coping responses and if we don’t consciously learn new ones, they will continue on well past their usefulness.

      I like the language of what’s in someone’s coping repertoire because a repertoire is like all the songs you know how to play and we know that one has to actually be exposed to and practice to learn a new song. So a coping repertoire, if not consciously taught or modeled, develops rather randomly and can be quite narrow. That means in the face of later stress, the person only has a few limited ways to respond, and those responses may not be in the overall interest of the person’s long-term well-being.

      Two examples:

      * A person learned to shut down when a parent yelled because they did it a lot; in their present-day marriage, they don’t hear their spouse’s concerns until they yell or make a big deal out of something because the person shifts into shutdown mode at the first sign of conflict. Shutting down as a way of solving the problem of being a helpless child raised by a chronically yelling parent needs to be replaced with functional communication skills and self-soothing in the face of normal conflict.

      * A person who had a parent who was very critical and never satisfied with their behavior or capacities; they responded by becoming highly rigid and perfectionistic in their work life in the hopes that it would make them “beyond reproach.” They react with hostility to criticism because they are working so hard all the time to be “beyond reproach.” Their rigidity and defensiveness needs to be replaced with practicing a level of performance that is “good enough” and a set of coping skills to tend to their anxiety around other people’s potential disappointment.

      Substance abuse is a kind of problem-solving effort to deal with lack of emotional regulation skills. OCD rituals are a kind of problem-solving effort to deal with high anxiety states.

      I think it’s useful to see people’s present-day behaviors as reasonable efforts to solve real problems. The link back to trauma or a past experience isn’t necessary, but sometimes it is there. Presenting concerns in therapy almost always have a complex etiology that’s a mix of genes and environment, and we don’t need to parse the causes necessarily in order to treat the presenting issue. People often show up in therapy offices when their reasonable efforts to solve real problems (ie, their existing coping repertoire) is causing more problems than it’s solving.

  64. Basil Marte says:

    I was recently contemplating self-modeling, and had this idea:
    – People have/are some kind of persistent algorithmic structure, let’s call it “self”;
    – people have/contain a model of their “self”;
    – the latter is a very compressed description of the former at the best of times, and can get way out of alignment with it;
    – people complain and go to therapy if their self-model/image [!] contains a representation of them having a problem;
    – even without assuming that the self-model is easier to change than the self, a therapy is mostly pronounced to be successful if it changes the patient’s self-model, and thus the patient reports they no longer have a problem (c.f. “no more gnomes in the mine”).

    Consequently, perhaps therapies systematically/primarily alter the patients’ self-models rather than the rest of their selves*, which alteration may or may not bring the two closer into alignment. This may not be too big a problem, since the self-model does in fact have a large causal effect on behavior.

    *: the self-model is part of the self.

  65. Douglas Knight says:

    A lot of people have proposed that people who found a school of therapy have some special gift which they are able to communicate to their students, but which is diluted by generations of teaching. But this would not apply to worksheets and books which are, at least potentially, frozen in time. (Probably people keep rewriting the worksheets, but there’s only one generation of books. But I don’t know.) Also, drugs.

    There’s a similar problem of testing out, say, acupuncture. It should be easy to do an RCT to see if a particular person can do something useful under the name “acupuncture.” But I think that most studies of acupuncture use untrained people following simple instructions. The first study is for exploratory science, for learning if there’s anything there at all, whereas the second is quality control for mass-production, determining if we’ve distilled it down to reproducible method. People seem to be confused about the difference.

  66. cedrus_libani says:

    My suspicion: there’s a subset of people who really can turn their lives around in a therapy session or two. Those who practice Exciting New Therapy are keeping a sharp eye on their patients’ outcomes; if a “miracle cure” happens, they will know about it. If this happens in Normal Therapy, the patient may simply move on with life and be lost to follow-up.

    I was one of these people. I had recurrent depression; I was cured in one session of Normal Therapy; I’m fairly certain the therapist involved would be shocked to learn of this, because it really was just a brief consultation to see if I was a candidate for SSRIs. (The specific insight: “You seem to have good coping skills, but you don’t use them when things are going poorly at work. Why is that?” Yeah, I needed to stop doing that.) Now that all of this was out in consciousness, I no longer felt any desire to have psychological problems. In a followup session two weeks later, I reported feeling better, and that’s the last he saw of me. There were no books written, because this was Normal Therapy.

    I don’t even think a Normal Therapist would categorize that experience as “cured by therapy”. Everyone knows that Normal Therapy produces gradual improvements at best. Since I was cured by a single whack with a clue-by-four, I must have never needed therapy at all.

    I’m also fairly certain that, had I been treated by a true believer in Exciting New Therapy, they’d be telling that story for decades to come. She was cured in one session! Since everyone knows that Normal Therapy produces gradual improvements at best, the difference-maker must have been whatever special sauce the Exciting New Therapy brought into play…right? It’s been proven, we have case studies.

    You find what you’re looking for, and the true believers are looking. They will find people like me, they will cure them, and their colleagues will never hear the end of it. Meanwhile, if you ask a Normal Therapist what he thinks of Exciting New Therapy, he’s not thinking about me. He’s thinking about the patient with C-PTSD and schizophrenia who will probably be in therapy for the rest of her life, and he’s not optimistic.

    • thetitaniumdragon says:

      My grandmother was cured of her desire to smoke by a single session with a hypnotherapist.

      I think some people really do just need to be told to do something by someone and then magic happens.

      Hell, I’ve seen this happen in life a few times; people realized they could do X and suddenly were free.

      • Protagoras says:

        It is my understanding that the pattern with addictions generally is that people sometimes just decide to quit and manage to stick with it (though usually they don’t), and that most treatments for addiction usually fail, but succeed about as often as the just deciding to quit approach. But the people who succeed with a given method of course tend to credit whatever method they used.

  67. tentor says:

    I’m a bit late to the party, but this post reminds of me of the single most sage advice I have ever gotten.

    At my university math class, the tutor told us that to get full credit, all proofs need to be “sufficiently explained”. A student then asked how we should know if an explanation is sufficient. The tutor replied “Well, if your proof works even when the thing you’re trying to proof is false, the explanation is *not* sufficient”

    When you have to mutually exclusive theories and both are supported by similar compelling proofs, then neither is suffciently prooven. Likewise, when a proof is used repeatedly to show a lot of things, but some of them are demonstratebly false, then the entire proof with *all* it’s conclusions must be thrown out.

    This is obviously nothing new to the rationalist community, but it is frightening how often such “proofs” can be observed in daily life.

  68. Frederic Mari says:

    EDIT : I see the question has been asked above but the replies weren’t particularly good in making the case either way.

    “while homosexuality is mostly not genetic, few people today think this is a plausible explanation”.

    Total non-sequitur to the conversation you’re having but – is this proven?

    I realise that homosexuality, like all sexualities, can’t be 100% genetic since the Ancient Greeks and modern English public school boys famously had lots of it while most societies at large don’t.

    But I would have thought sexual orientations are mostly genetic, with a dash of circumstances/free will thrown in. Am I mostly wrong?

    • thetitaniumdragon says:

      Well controlled studies suggest that if you are gay, your identical twin’s odds of being gay themselves is only somewhere around 5%-20%. That would suggest that genetics only play a small role in homosexuality – if any at all, as identical twins also are likely to have many shared environmental factors in common.

  69. Ugh. Why is the author reviewing all therapy books, when it’s clear the author hasn’t actually read all therapy books?

  70. Richard Kennaway says:

    Everyone who has had warts has a cure for warts. Painting them over with salicylic acid, freezing them, rubbing them with vinegar, rubbing them with specifically apple vinegar, rubbing them with the family Bible, rubbing them with an eel’s head and then burying the head, exposing them to the light of the full moon, having the local witch rub them with pennies, praying that they may be removed, and so on.

    Warts appear and disappear for no known reason. (Ok, actually HPV infection and the immune system overcoming it, but that doesn’t answer the when and why of it.) Suppose that none of these treatments have any effect whatever. You have some warts and try every cure you come across. At some point, they spontaneously go away. Whatever cure you were applying at the time will seem to have worked, at least for you.

    Is this a sufficient explanation of the success, such as it is, of all psychotherapeutic methods? Like pre-scientific medicine, a collection of ineffectual superstitions, with apparent successes down to fitting to the noise of spontaneous remission?

    • cuke says:

      Many of the clinical trials that test for the efficacy of psychotherapy compare it to “no treatment” and there is a lot of evidence that psychotherapy is more effective than no treatment across a wide array of disorders. We would expect it to perform the same as no treatment if recovery were due only to spontaneous remission.

      At least that’s my understanding of the current literature.

  71. alwhite says:

    In a random tangent, I followed the related link to Scientific Freud and I thought this might prove meaningful to the discussion. The post starts

    I was taught the following foundation myth of my field: in the beginning, psychiatry was a confused amalgam of Freud and Jung and Adler and anyone else who could afford an armchair to speculate in. People would say things like that neurosis was caused by wanting to have sex with your mother, or by secretly wanting a penis, or goodness only knows what else. Then someone had the bright idea that beliefs ought to be based on evidence! Study after study proved the psychoanalysts’ bizarre castles were built on air, and the Freudians were banished to the outer darkness. Their niche was filled by newer scientific psychotherapies with a robust evidence base, such as cognitive behavioral therapy and [mumble]. And thus was the empire forged.

    Now normally when I hear something this convenient, I might be tempted to make sure that there were actual studies this was based on. In this case, I dropped the ball. The Heroic Foundation Myth isn’t a claim, I must have told myself. It’s a debunking. To be skeptical of the work of fellow debunkers would be a violation of professional courtesy!

    This is almost correct but has some wrong assumptions. Albert Ellis (REBT), Fritz Perls (Gestalt), and Carl Rogers (Person Centered) all were taught Psychoanalysis and began their careers there. They all found that psychoanalysis wasn’t working on their clients and so they decided to explore different options (creating their own techniques). These guys are called the 2nd generation. Aaron Beck (CBT) is a weird middle generation guy. Marsha Linehan (DBT) and Steven Hayes (ACT) are said to be the 3rd generation. There’s also 3 waves of cognitive therapy as well, starting with Skinner, Pavlov, and Wolpe (Behaviorism), with Beck and Albert being the 2nd wave, and Hayes and Linehan being the 3rd wave. EMDR and Biofeedback might fit in this 3rd wave stuff, but I’m not sure. There are other therapies as well, such as IFS, Solution Focused, Somatic Experiencing, and Emotion Focused Therapy that are just getting ignored in this classification system. Not to mention the family systems therapies, which are a whole different ball of wax.

    There’s a lot of bias towards the cognitive therapies that I think blocks the others out. I think that bias has to do with ease of study. CBT has structured manuals that are tremendously easy to follow and replicate in a research setting. This is part of the complaint about the “evidence” around CBT. It’s not that the studies are that much better for CBT than others, it’s just CBT is so dang easy to study that there are a lot more studies that use it.

    Knowing this history seems important in trying to decipher what is happening with the Dodo Bird phenomenon.

  72. wonderer says:

    Let’s suppose that no psychotherapy worked at all. But hopeful patients come in, try out the Amazing New Therapy That Solves All Problems, and get to talk to a psychiatrist they can open up to. The psychiatrist will always support them, never reveal their secrets, and never criticize or judge them. What’s more, the patients don’t have to live with the psychiatrist every day (unlike with friends and family), and so don’t have to worry about what the psychiatrist’s opinion of them. The patient, feeling validated and supported, starts doing better. Because the actual content of the psychotherapy is total bunk, the patient gets equally better no matter which therapy is used.

    Isn’t this a sufficient explanation of the dodo bird verdict?

    • hls2003 says:

      Isn’t it rather just a restatement of the dodo bird verdict, rather than an explanation? If your example postulates that the patient really does get better, then you’ve also postulated a legitimately effective technique: “support them, never reveal their secrets, and never criticize or judge them… [make sure they] never have to worry about what the psychiatrist’s opinion of them… [make them feel] validated and supported.” That technique is employed multiple ways, thus all about equally effective, which is the dodo bird finding.

  73. Qaz says:

    (I haven’t posted before. I’m reluctant — I don’t like to post if I can’t conceivably erase the post later. Right/privilege to disappear and all that. Can this be addressed?)

    I’m a longtime lurker, fascinated by this post like others. Part of this one troubles me: it’s the “you can come up with a compelling explanation based on any theory X, so let’s be skeptical.” Sure, let’s be skeptical because skepticism is cool, but…

    I can explain a lot of my onetime problems based on this familiar “historicism,” including “this was a survival strategy you came up with in the past that doesn’t work now” (heard that a lot).

    But… my kids. One is mentally handicapped, very low IQ. I can’t come up with a past-trauma explanation for this. It predates any trauma and it never was adaptive.

    Another is bright. Had ADHD last year, some lingering. I can’t fabricate a historicism explanation based on his actual past that would lead to ADHD. He’s had some trauma (some scary bullying at school) which can explain some behaviors that developed before we found out about the bullying, but the ADHD developed later and wouldn’t help with bullying.

    We had a foster child. His psychologist absolutely explained his daily tantrums in terms of severe abuse in the original family. And I believed it, but as time went by… it wasn’t compelling. “Birth parents, overwhelmed by their own issues, couldn’t handle a child with daily tantrums” was compelling and even obvious when you say it like that.

    So I can believe people with a strong agenda could fabricate and convince themselves of an explanation… but not that a reasonable person could. At least, *I* couldn’t.

  74. hopaulius says:

    I saw an interview in which the interviewee mentioned that he practices “family systems meditation.” This intrigued me, and a few searches led me to family systems therapy. I was interested in this, because I had studied family systems theory in relation to family and institutional dynamics. So I bought a book. And it is just as you described. We all have these separate, stunted personalities inside of us, thwarting our personal progress. So we have to name them, befriend them, and ask their advice about our situation. Voila, integration!
    As I read, I noticed that the miracles happened in therapy sessions. When a patient “got it,” named the “parts,” and reported their relief after having a conversation with their “parts,” the therapist/author proclaimed them cured. But… no followup. How were they doing six months later? What are the links to neuro-science? How is this anything greater than applying personification to symptoms?
    I began to suspect that what the patient achieved was mastery of therapeutic language. The therapist taught the patient how to conceptualize their condition and describe it in the therapist’s preferred language. When they did so, obviously the patients felt validated and left the office feeling good about themselves, for a few minutes, and the therapist went home with another therapeutic notch in his/her/their gun.
    I am admittedly biased about psychiatry and psychotherapy. My older brother terrified my weak alcoholic father, who involuntarily committed him to a mental institution in about 1960. My brother experienced some of the great breakthroughs of the age, such as being fire-hosed with ice water and administered electric shocks. Not to mention pretty colorful pills by the handful. On an early visit, when I was about ten, my brother grabbed my arm and said, “Never let them take control of your mind!” It’s possible that he had a lobotomy, because whatever they did to him indeed changed his personality, robbing him of all drive and volition.
    My inexpert hunch is that people who are hurting need someone who listens to them, pays attention to them, and pretends to care about them for an hour. I think that’s the therapy, and the rest is word salad and hucksterism.

  75. Paul Torek says:

    Historicism is the hypothesis that the brain is doing what evolution designed it to do: learn from experience. As such, it bets on the base rate. It’s like hearing that someone got into a taxicab in New York City and supposing that the vehicle was probably yellow.

  76. Chaostician says:

    Are therapy book authors blatantly lying? I try to have a really low prior on this sort of thing, but I’m not sure.

    There are two different questions that can be easily confused here: Would a typical therapist who wrote a therapy book blatantly lie? Do there exist therapists who would blatantly lie in a book? Your priors should be different for these questions.

    I suspect that the market for therapy books strongly favors books with exaggerated or false success stories. In this case, the second question is more relevant. Most therapy books that someone has heard of would be by a blatant liar even though most therapists wouldn’t lie in their books.

    • Chaostician says:

      A similar confusion can happen when [someone] accuses [public figure] of [rape / discrimination / something bad that happened in private].

      People can lend credence to the claim by saying that the accuser faces [costs] by coming forward. A typical person wouldn’t make a false accusation.

      Although this argument is valid in most people’s lives, it is not valid when a public figure is involved. The question shouldn’t be: How likely is it that an arbitrary person would lie like this? The question should be: How likely is it that there exists someone who would lie like this? The more well-known the public figure, the bigger the difference in your priors for these two cases should be.

      • Aapje says:

        A typical person wouldn’t make a false accusation.

        A logical rebuttal is that people who falsely accuse may not be typical. The entire concept of ‘typical person’ is fairly questionable, anyway.

    • Chaostician says:

      If we want to steelman a lying therapy book writer:

      The only thing that really works are “nonspecific factors” like how much patients like their therapist. How do you improve the “nonspecific factors” of your patients?

      One potential strategy is to tell them that the therapy will be more effective than it actually will be. You tell them a single session will cure everything with the hope that this makes it more likely that they will make gradual progress over several years.

      I have no clue if this would work, but it is testable: Does the patient’s belief that immediate progress is likely correlate with the patient’s actual gradual progress? Is this more effective than a patient’s belief that gradual progress is likely?

  77. truthlizard says:

    Nonviolent Communication by Marshall Rosenberg is a book I would be interested in seeing you review. It’s a worthwhile sidetrack along the lines of therapy books – Rosenberg was a therapist, was unsatisfied with the constraints of therapy, and pioneered a method he could ‘give away’ to large numbers of people instead of working with individual paying clients one hour at a time. The about the author blurb:

    Dr. Rosenberg first used the NVC process in federally funded school integration projects to provide mediation and communication skills training during the 1960s. The Center for Nonviolent Communication, which he founded in 1984, now has hundreds of certified NVC trainers and supporters teaching NVC in more than 60 countries around the globe.

    A sought after presenter, peacemaker and visionary leader, Dr. Rosenberg led NVC workshops and international intensive trainings for tens of thousands of people in more than 60 countries around the world and provided training and initiated peace programs in many war-torn areas including Nigeria, Sierra Leone, and the Middle East. He worked tirelessly with educators, managers, healthcare providers, lawyers, military officers, prisoners, police and prison officials, government officials, and individual families. With guitar and puppets in hand and a spiritual energy that filled the room, Marshall showed us how to create a more peaceful and satisfying world.

    Scott – may a buy you a copy so you can see if you might want to review it?

  78. Scott, have you read The Great Psychotherapy Debate?

    It’s a rigorous dive into the evidence on the history of studying psychotherapy outcomes. I would be really interested in hearing your thoughts on it.

  79. I’ll rule out that CBT is in the water. I spent ten years trying to change my life, reading all sorts of self-help and pop-psych material. I went to a Blue Ribbon school district where our teachers occasionally sprinkled New Age psychobabble into the curriculum. Then I went to a fancy college, where the students came from similar, privileged, and discerning backgrounds and where all the teachers were versed in contemporary thought. If CBT was already in the water, then I doubt I would have gotten my first breakthrough from reading Feeling Good, five years after graduating from college.

    Compliance may also be a factor. I didn’t go to a therapist for CBT, but I have a knack for following instructions. When I get into a new method, I get over-excited about it (see: my memoir). Recently, I’ve dabbled in seeing therapists, and I’ve noticed that everyone puts CBT on their list of specialties. When you meet with said therapists, they hand you a worksheet, but don’t work the steps.

    I found the same disconnect when comparing meta-studies of psychotherapy with perceived results of talk therapy. You can cherry-pick studies that include compliant applications of CBT and wind up with absurdly high success rates for treatment overall. Meanwhile, therapy-in-the-wild is rarely rigid, and most therapists who supposedly practice CBT just “weave it in.” Therapists consider themselves artists who are building their unique brand of talk therapy. I was more impressed by the stats on the efficacy of computerized CBT. So we have to meta-review the meta-studies to make sure they’re not just aggregating noise.

    Another usual suspect is that CBT is over-prescribed now, due to its popularity.

    A collective placebo effect may be the overall takeaway, though, but it may not have the same mechanism as placebo for individuals. Early practitioners are more invested in a method’s success, and therefore will spawn a legion of success stories, only to be followed by weaker, later adopters who apply it to patients who may not even need it. So it may be true that if you tried CBT in the 80s, you were more likely to find an improvement than you would have today.