Internal Family Systems has become a wildly popular form of therapy over the past few years. Some of my friends swear by it. But after trying IFS myself and studying some of its underlying assumptions, I’m skeptical. I shared some of the reasons for that skepticism in a recent installment of my Your Mileage May Vary advice column. I noted that IFS is being used to treat all kinds of psychiatric conditions, even though its scientific foundation is shaky. Yet there’s no question that this therapy, which teaches that we’re each made up of a bunch of different parts and insists that there are “no bad parts,” appeals to tons of people — including very smart people who care a lot about evidence.One of those people is Carl Erik Fisher, a Columbia University bioethicist and psychiatrist who specializes in addiction. In his book The Urge, he writes that he initially thought IFS sounded hokey — until he tried it:I feel like most of my supervisors at Columbia would turn up their noses at it — IFS does not have much of an evidence base, and it has neither the cerebral cachet of psychoanalysis nor the prestige of the more explicitly scientific therapies. But something about it works for me.That piqued my curiosity. My column, Your Mileage May Vary, is pluralistic by design; I believe in weighing multiple values and viewpoints. So I reached out to Fisher to ask what it is about IFS that he thinks works.We talked about why this therapy is trending right now, how IFS might be both helping and epistemically harming clients, and whether it’s actually misguided to require a healing modality to be backed by randomized controlled trials before we give it a try. Our conversation, edited for length and clarity, follows.Key ideas of Internal Family Systems therapy:IFS was developed in the 1980s by therapist Richard Schwartz, who was inspired by family systems therapy.Schwartz argued that just as a family is made up of members who get into conflicts and also protect each other in patterned ways — so too is your mind. You’re not a single unified self; you’re a collection of “parts.”Your parts include “exiles” (which carry pain and shame), “managers” (which try to prevent the pain and shame from surfacing — for example, through perfectionism), and “firefighters” (which may use drinking, bingeing, or numbing out to protect you when pain or shame break through). There’s also the “Self,” which is your supposed true essence, wise and undamaged by trauma.Sigal SamuelWhy do you think IFS has become so wildly popular? Carl Erik FisherI think IFS embodies a turn toward the experiential in psychotherapy these days. A lot of other modalities that are really popular right now — and not without reason — are things that are experientially based: EMDR [eye movement desensitization and reprocessing], somatic therapy, mindfulness. The thing that unites these is the shift away from rational or analytical insight to a more direct experiential practice.There’s a sort of intuitive, ecumenical, wisdom-oriented, potentially secular spirituality in it. There’s something about a deep compassion toward your own defenses and promoting a kind of metacognition that’s not just mindfulness, it’s not just being in contact with the present moment, it’s also bringing an awareness to all the different parts of yourself and identifying less with the analytic, observing, logical parts of yourself. There is a kind of lowercase-T transcendence to that practice. People are searching for that, especially in the context of having less of a properly spiritual or religious worldview.Sigal SamuelYou mean that for people who are secular but nevertheless have an ongoing yearning for spirituality, maybe the IFS framework of looking at yourself as a multiplicity of parts and moving away from the view of “I have one rational command center in my brain” helps them access this feeling of transcendence?Carl Erik FisherYeah. And I think as people get isolated — I think it’s safe to say people have gotten more isolated since Covid — there’s more of a worry that people are left to form internet-based relationships, parasocial relationships with an audiobook or a podcast or a relationship with AI, and these are all analytical, text-based, logical relationships based on explicitly verbal content as opposed to embodied content. There’s a natural inclination, maybe even a self-correcting inclination, where people want more embodied experiences. It’s a cultural reaction. When you feel like you’re just a head floating in a Zoom box for most of the day, like a lot of knowledge workers do now, you really start to feel it.Sigal SamuelThat’s really interesting because I did notice that IFS enjoyed a huge surge in popularity right around Covid.But I’m curious if you think it’s problematic that IFS has been used in the treatment of all sorts of patients, even patients with very serious conditions, even though so far we don’t have a single randomized controlled trial [RCT] done on it as a treatment for a psychiatric disorder. Carl Erik FisherIn the ’90s and the 2000s, there was a big focus on randomized controlled trials, a big focus on the evidence base. And it is important to do good-quality psychotherapy research. The field should have testable hypotheses and find ways of testing them. But the fact is that an RCT — as much as people call it the “gold standard” — is not the ultimate arbiter of truth or epistemic authority.There’s a very strong argument, to my mind, from some of the people who criticize the overinstrumentalization of psychotherapy evidence: They say that when you over-rely on all these short-term randomized controlled trials, you get a bunch of biases just built into the process. It’s short-term, so of course you’re going for the low-hanging fruit. The outcome measures are the symptom checklists, like the Beck Depression Inventory, and those are kind of diametrically opposed to the big depth psychology targets like personality change, leading a purposeful life, and all the rest. And then there are even more concrete critiques of psychotherapy evidence [when it comes to things like] control conditions and the difficulty in blinding. So I think it’s okay to be a little bit skeptical about the field of psychotherapy research overall. Whenever a therapy gets rapidly popular, more people will start to use it and the practice will go far ahead of the evidence base. And I don’t know a way around that. I mean, there’s not going to be any sort of professional organization arguing for a moratorium on the use of a psychotherapy until we get adequate data. If we did that, more than half of the psychotherapy modalities out there couldn’t be taught. Sigal SamuelOne of the good things about RCTs is that they force us to track harms to patients. And there’s been some reporting about harms done to patients in an IFS context. So who would you recommend IFS therapy to, who would you not recommend it to, and crucially, how can we tell apart people who might benefit and people who are likely to be harmed?Carl Erik FisherAnybody who does any kind of psychotherapy might experience a kind of symptom worsening; they might experience some internalization of a harmful self model. I think it’s good just in general to be conscious of the fact that not everyone will be helped by psychotherapy. Some people will be hurt. And sometimes it’s not predictable.But sometimes it is predictable! Who are the people at most risk with IFS? People who are at imminent risk of destabilization, whether it’s self-harm or substance use disorders — I don’t know that IFS is the best fit for immediately stabilizing someone seeking safety. And people who have an unstable self-concept overall, people who might be vulnerable to identity fragmentation or suggestibility. Certain personality disorders might fall into that framework.Also, someone with an eating disorder, I would put that in the category of dysregulated behavior. That’s a category of problems that traditionally we think of as needing really structured behavioral change intervention. Eating disorder treatment is a great example where there’s a highly specialized, really well-developed group of practices. One of the big problems with pseudoscience or interventions that are not necessarily proven — homeopathy is a great example of this — is that people will say, homeopathy is just diluted whatever, it’s just water, it’s placebo. Well, the problem is if you go get homeopathy for cancer, then you’re missing the opportunity to get a real treatment for cancer. Sigal SamuelA moment ago you mentioned suggestibility. There are a few things about IFS that seem to me to be relying on the power of suggestion. When you tell an IFS therapist about some anxiety or emotional pain you’re feeling, they’ll often say, “Where do you feel that in your body?” Anecdotally, my sense is that many people secretly just don’t feel it in their body. But the therapist asks that question as if it’s so profound that when you’re the client sitting there, you feel like you’ve got to come up with something. Maybe your shoulders are tense — like, just because you’ve got a desk job — so you grab onto that and say, “It’s in my shoulders.” To what extent is the power of suggestion potentially playing a big role here?Carl Erik FisherI think it’s an important point. Generally, one of the deep critiques of psychotherapy going back decades is that the therapist has a sort of causal narrative about how the world works, what counts as suffering, what counts as the good. Because they have this asymmetrical authority, they can guide a client. And some of the critiques say that invalidates the epistemic authority of the client. In other words, people have their own authority to have their own experiences, and if you undermine that and you just tell them, “No, you’ll feel the sadness in your body,” then it’s not only that you’re maybe promoting suggestibility or barking up the wrong tree, but you might actually be invalidating their own experience. So I think it’s important not to lead clients toward some sort of worldview. That’s a danger in a lot of psychotherapies, and it’s arguably a much more heightened danger in a trendy, popular psychotherapy modality where the founder and the leading popularizers are making increasingly totalistic claims about how the world works and how the mind works and how spirituality works. That’s the part where some of the rhetoric gets a little questionable to me.Sigal SamuelIn terms of totalizing claims about how people work, one of the core premises of IFS is that underneath all our parts, we each have a core wise “Self.” But that core unitary self is just…not a thing. If you dissect my brain, you will not find a homunculus in there. So I’m wondering if that emphasis on this idea of the Self bothers you?Carl Erik FisherWell, it depends on how people use it. There’s a lot of talk in similar ways in Buddhist practice. People talk about “big mind,” “wise mind,” “Buddha mind.” That might sound like you’re talking about an essential thing — like, Buddha mind is superior, it’s wiser, it’s better — but that’s balanced by the Buddhist teaching of emptiness, which says that there’s no core essential thing.In my own experience receiving IFS, there are careful practitioners who don’t essentialize the self. They don’t make it a core reified homunculus. But I am sure there are people out there who do. And I do worry that in the soundbite-y, simplified versions, it does sound like it’s a homunculus and you need to surrender your will to some sort of magical, intuitive force. That’s just a set-up for failure. There’s a very basic psychotherapy principle that says healing is possible and when people are integrated, they themselves have the agency and the capacity to get well. If you reify the self in such a way that it’s not the you that think you are, that there’s some other Self in there that’s kind of spiritual or metaphysical, then it’s kind of surrendering some autonomy. It’s a little bit like fracturing off your own capacity for change.Have a question you want me to answer in the next Your Mileage May Vary column?Feel free to email me at sigal.samuel@vox.com or fill out this anonymous form! Newsletter subscribers will get my column before anyone else does, and their questions will be prioritized for future editions. Sign up here!Sigal SamuelSometimes people report that they’ll tell their IFS therapist, “Hey, I’m skeptical about this aspect of IFS,” and the therapist will respond, “Oh, that’s your skeptical part talking.” My worry is that when any resistance on the client’s part is getting interpreted as just another fearful part acting up, then we’re in a self-reinforcing loop that makes it hard for the client to challenge how the therapist is depicting their reality — even if it seems off. Carl Erik FisherI don’t think it’s good practice in general for someone to say, “Oh, that’s just your skeptical part.” That is invalidating. I think my own response, 99 out of 100 times [as the psychiatrist] in a situation like that, would be to validate and say: “That makes sense. It’s okay to be skeptical. Psychotherapy is messy and we don’t really know how it works. We don’t know that IFS is the perfect solution. We don’t have to treat this as a final word about mental health. It’s one tool that we could try. And if it doesn’t fit, we can make an adjustment.”It’s bad practice if you just totally invalidate somebody’s reasonable skepticism about something that is really, really far from a settled science.