Therapy works? So . . . ?


A woman I used to know defined an intellectual as someone whose first question about anything is, “What’s the truth?” She contrasted this with the first question she thought most people ask: “What’s going to make me feel better?”

That’s probably not a good definition of an ‘intellectual,’ but it points toward a significant conflict we all face in deciding what to believe: the more reassuring ideas, the ones that make us feel good about ourselves and those we love or depend on, the ideas that make us optimistic about our prospects, may not be true.

Social science research on the conflict between truth and good mood is extensive. Depressed people tend to be more accurate in their estimations of how much control they have over their environments. Happy people tend to process challenging information in a highly biased manner to preserve their good cheer—mild sadness promotes accuracy. Most of us believe we’re better, smarter, more high-ranking than we are.

Still accuracy matters—over the long run, we’re more likely to do well if our beliefs generally allow us to coordinate our actions with each other’s and with reality. Truthfulness is a universal virtue for a reason.

In mental health care, the conflict between truth and good feeling takes many forms. Some are obvious, like the temptation to affirm a client’s self-justifications, rationalizations, and other wishful thinking. The most important conflict between truth and good feeling, though, is the central role that outcome studies play in “evidence based” practice.

Indeed, the very idea that outcome studies can “scientifically validate” a form of therapy fatally conflates good feeling and truth, so that citing “scientific validation” through clinical trials is more misleading than enlightening.

An outcome study, no matter how well designed, answers only one sort of question: does the therapy “work”? It never tells us whether the ideas behind the therapy or the beliefs induced by the therapy are true—or therefore, whether the practices and behaviors based on those ideas and beliefs are sound.

An idea does not have to be true to make us feel better. Practices do not have to be based on true ideas to structure lives agreeably to those who live them. Thus, knowing that a therapy “works” tells us nothing about whether the ideas and practices it conveys deserve respect.

Why does this matter? Therapy is not an autonomous country. Both therapists and patients exist within larger society, and they need to know and do things that have little direct bearing on the suffering that brings them into care. But a therapy that “works” can lead to beliefs and practices that set patients and therapists at odds with their social contexts and their more comprehensive intellectual needs.

Historically—and even now, outside of health care—science aims to describe and explain nature. Outcome studies don’t. They simply say, “If we do x, it has y affect.”

In English, we distinguish between “knowing that” and “knowing how.” In Greek, this distinction was captured in the difference between “episteme” and “techne.” Episteme, or “knowing that,” is the business of science and other attempts to describe and explain nature. Techne, or “knowing how,” has to do with craftsmanship and action.

Outcome studies should be seen as serving techne. In essence, outcome studies are a step in product development—a form of engineering research, not science. At best, they are part of quality assurance. But like all engineering, and no science, they have no claims to our assent unless we happen to want to buy the product.

When we pretend that outcome studies “scientifically validate” therapy, we confuse a product that can be used to specific ends with knowledge of how the world works. That’s a pretty serious confusion.

Truth commands assent. Products don’t. The fact that a therapy makes its practitioners and clients happy imposes no obligation on anyone else to cooperate with those happy sellers or consumers. It doesn’t even tell us that being happy in that way deserves anyone’s respect.

When we are crafting and choosing therapies, we need to take into account a great many things other than how it affects the suffering that brings us into care. That’s why the correct response to any positive outcome study is, “What significance does that have?”—or, more colloquially, “So what?”


Mad in America hosts blogs by a diverse group of writers. These posts are designed to serve as a public forum for a discussion—broadly speaking—of psychiatry and its treatments. The opinions expressed are the writers’ own.


  1. Dr. Fancher thank you for this compelling article. I have read it twice and will read it again a couple more times. Your comment about the value of those who are feeling depressed or sad was insightful. As a biblical counselor (much different from a Christian counselor) I often point out to my clients that every emotion has a unique purpose and should have a time and place to be expressed. It is a biblical principle to be authentic in every way. Setting aside the obvious issues of potential abuses, there is value in all that life offers; including, but not limited to, conflict. There are many products available (as you suggested) to the consumer to marginalize, minimize, and escape conflict; to be obtain happiness. In the process, these products often lead to missed opportunities of personal growth and community interdependence. These “happy” products may serve the individual (presently), but they tend to do little for the community or the individual’s future generations. Perhaps the Creator has given us the ability to feel sad – to mourn – in order that our lives can be reordered; realigned for optimum productivity and effectiveness. I’m amazed that among the goodness of the Creator’s work we find that the default position of the universe is decay (by His design). Unless energy is enacted or invested everything will eventually fall apart. Maybe the feelings of sadness or being depressed is initially a means of awakening the individual to an opportunity to recreate the world. Do the “happy” products undermine this opportunity?

  2. I agree that outcome studies to not validate the veracity of of the process/mechanisms of a therapy’s affect on someone’s mood/well-being. As you say, outcome studies do show that participation in a therapy can lead to particular outcomes. I want to be clear about what I mean by outcomes-I define an outcome as an intended benefit a participant experiences.

    Without a strong understanding of suffering/well-being, why people feel worse or better, and the mechanisms of changing them, outcome studies are a place to start to assess at least what people people can expect to get when they often seek from therapy, feeling/living better or getting some relief from psychiatric symptoms.

    There is so much therapy happening, all kinds of therapy, and even ones that are validated to show effect for particular outcomes (reduced depression, reduced panic, etc.), mechanisms of change are still elusive (and the proposed mechanism for that therapy is not proven by outcomes occurring). Because we are unlikely to scientifically get at the mechanisms of suffering wellbeing without some kind of strong theory (not have really held up on its own), I think outcome studies are a perfectly reasonable place to start to assess all of the therapy that is happening out there now. I am not yet asking anyone to have evidence of WHY something works, but I think it is essential to have evidence THAT something works and works well and often enough beyond any risks that may be involved. Many popular psychotherapies (analysis, dynamic, etc.) do not even have this information, while at least the “scientifically validated” studies do.

    Doing outcomes research at least lets people make informed decisions about what they can expect from participating in a particular kind of therapy, and can identify ineffective or potentially harmful therapies and help limit people from being exposed to them. Hopefully, I have some faith in that in working more inductively, trying to build and test theory starting from what works (as opposed to theorizing mental health and designing treatments to fit that theory), we can start to get some more insight in the mechanisms of suffering and the mechanisms of its alleviation.

    To be honest, I am seeing that psychotherapies for depression, no matter what they are, tend to work by increasing hopefulness for improvement, ie placebo. I’m ok with this, and learning the mechanism for change for depression is placebo can lead to more refined treatments and do more informed process research to better understand and apply those mechanisms.

    I think patients/clients need to be as clear as possible about what outcomes they are looking for when seeking treatment, clinicians need to be able to support them in identifying those outcomes, and a clinician has to decide if s/he is capable of engaging in psychotherapeutic work associated with the identified outcomes, and if not, refer their client to someone who can. I have come across many people who have had treatment prolonged, experienced extreme dissatisfaction, and stopped seeking support because clinicians made the decision about if it would be better for someone to feel better to learn more truths when the client was expecting and hoping for the other experience.

  3. The episteme-techne dichotomy in therapy reminds me of the confidence-skill dichotomy in math education. My experience teaching high school math has been that many students (especially boys) consistently report high confidence (happiness) in their math abilities while displaying low to average proficiencies; conversely, many other students (especially girls) report low confidence while displaying good proficiencies.
    At the same time, some instructors teach (and grade) to improve confidence, while others teach to improve skill. The former can do immense harm if carried to dishonest extreme. Most good teachers aim for a thoughtful blend of the two. I assert that students of excellent teachers acquire proficiency in the subject and a confidence that reflects their cumulative experience of a true success.

    • Odd responding to my own earlier placeholder thought but hey!

      I did like the piece and thought, yeah I think he’s got that right.

      I’m wondering if you’re thinking talk therapy should be more focused on episteme or techne or both or doesn’t it matter. I’m struggling where to go next assuming your accuracy with the piece.

  4. Brilliant! Therapy as commodity, beware the salesperson.

    As I understand it most outcome studies on therapy show it to be effective no matter what the school is. What therapists offer is understanding, solidarity and encouragement to deal with your problems – a simple sounding message that is hard to put into practice. We can all do this for each other is we so wish. Good teachers, community workers, counsellors, self help groups and good friends all do this.

    Understanding the world, now that is more a political problem. In the UK there is a Social Materialist school whose leading light is David Smail and the Midlands Psychology Group who say, ”
    We are a group of clinical, counselling and academic psychologists who believe that psychology—particularly but not only clinical psychology—has served ideologically to detach people from the world we live in, to make us individually responsible for our own misery and to discourage us from trying to change the world rather than just ‘understanding’ our selves. What are too often seen as private predicaments are in fact best understood as arising out of the public structures of society.”

    Or am I reading you wrong? I hope not, but you never know….