IFS Therapy and the Recent Criticism: What Clients Should Know
If you've looked into Internal Family Systems (IFS) therapy in the past year, you've probably seen some high-profile media coverage of allegations of harm at a treatment center that used it.
In October 2025, New York Magazine's The Cut published a lengthy investigation into IFS and Castlewood Treatment Center, an eating-disorder facility in Missouri where IFS was used. The article described former patients’ allegations of coercive and harmful treatment at Castlewood, including intensive trauma work conducted in ways that former patients described as pressuring or suggestive. Other allegations included therapists defining clients’ experiences for them, serious professional and sexual boundary violations, and therapists leading some patients toward memories or beliefs about childhood abuse that had not previously been part of their histories. Some patients later concluded that those memories were false and recanted accusations they had made against family members.
Other clinicians and researchers have since raised broader questions about IFS, including the size of its research base and whether the model can be misapplied with some clients.
What Happened at Castlewood Deserves to Be Taken Seriously
The Castlewood story is not something I would dismiss simply because IFS is a model I use in my own practice. Accounts of coercive or suggestive therapy deserve serious attention. When a person enters treatment while already vulnerable, a therapist has considerable influence over how that person understands their own experiences. That influence creates an ethical responsibility to protect the client’s ability to question, disagree, and make their own decisions about what their experiences mean.
The Cut’s reporting includes allegations from former patients and staff describing therapists who pressured clients around trauma, engaged in sexual and other boundary violations, and, in some cases, encouraged clients to understand previously unrecognized experiences as abuse in ways that former patients later described as suggestive or coercive. Some former patients later recanted accusations they made based on those therapeutic experiences, or came to believe that memories had been constructed during treatment.
Castlewood and the people involved have disputed some of the allegations outlined in the article, and The Cut includes their responses alongside the former patients’ accounts.
What Does Any of This Have to Do With IFS Itself?
The IFS Institute responded to the article in November 2025. Its response states that the behaviors alleged at Castlewood, including bullying, hypersexualized interactions, imposing therapists’ agendas, and competition for therapists’ attention, are contrary to the principles and practices of IFS.
It also states that IFS does not treat “recovered” memories as established fact without corroboration. (To be clear, that is a point about what therapy can verify, not a judgment about the client: a therapist can take a memory seriously without deciding whether it happened exactly as remembered.)
It is important to be clear about Richard Schwartz’s relationship to Castlewood—and about which Schwartz we are talking about. Richard Schwartz, the creator of IFS, and Mark Schwartz, who operated Castlewood, are two different people and are not related to each other.
According to the IFS Institute, Richard served as a consultant and staff trainer at Castlewood and participated in some patient sessions, but was never a Castlewood therapist. The Cut describes his involvement differently: it reports that Mark, who had been one of Richard’s early trainees, hired Richard as a consultant in the early 2000s to treat Castlewood patients and train staff, and that Richard continued working with patients there over the following decade.
IFS Is One Approach to a Much Older Idea
IFS is a relatively recent formulation of an idea that has appeared in psychotherapy under many different names.
Freud's work in the late 19th and early 20th centuries described the mind in terms of interacting systems and conflicting forces, eventually leading to his structural model of the id, ego, and superego. Jung developed his theory of complexes beginning in the early 1900s, describing emotionally charged patterns of thought and feeling that could function somewhat independently.
Gestalt therapy, developed by Fritz Perls and others in the 1940s and 1950s, has long used experiential techniques such as the empty chair to explore conflicting aspects of experience. Transactional Analysis, developed by Eric Berne beginning in the 1950s and introduced to a broad audience in Games People Play (1964), described Parent, Adult, and Child ego states.
Ego-state therapy developed from the 1960s onward, drawing on earlier work by psychoanalyst Paul Federn and later clinical work by John and Helen Watkins, particularly in the treatment of trauma and dissociation. Contemporary approaches such as Schema Therapy, developed by Jeffrey Young and colleagues beginning in the 1990s, works with different “modes,” or configurations of emotion, cognition, memory, and behavior.
IFS, developed by Richard Schwartz beginning in the 1980s and formalized in Internal Family Systems Therapy (1995), is one particular way of organizing and working with this longstanding clinical idea, and it has attracted considerably more public attention in recent years. The therapeutic process operates through noticing different patterns of feeling, thinking, remembering, and responding, rather than treating parts as literal entities living inside us.
IFS makes deliberate use of imagination in that process, but using imagination therapeutically is hardly unusual in psychotherapy; Gestalt chair work, for example, explicitly asks clients to engage in an imagined dialogue with another person or an aspect of themselves. IFS is therefore neither the invention of “parts” nor the only way therapists work with them. It is one contemporary model that has developed its own language and method around a longstanding psychotherapeutic idea, and clinicians are continuing to study how useful that particular approach is for different problems and populations.
Why the Allegations Deserve Serious Consideration
I take allegations of abuse and mistreatment in treatment settings seriously, in part because of settings where I worked earlier in my career. I worked in inpatient facilities serving people with substance use disorders, severe and persistent mental illness, psychosis, active suicidality, and co-occurring conditions. I saw how vulnerable people can be when they enter a treatment facility, and how much power a facility and its staff can have over patients’ daily lives.
My social work training and professional ethics have also taught me to pay particular attention to situations in which a power differential creates opportunities for exploitation, coercion, or silencing. The history of the “troubled teen” industry is a particularly stark example of what can happen when vulnerable people are treated as unreliable reporters of their own experiences. Young people in these programs were often dealing with profound trauma or mental illness—realities that institutions could easily weaponize to make their accounts easier to dismiss.
Restrictions on contact with family and the outside world could be presented as therapeutic interventions while also making it harder for someone experiencing mistreatment to report what was happening or obtain help.
Investigations and reporting over many years have documented allegations of abuse and, in some cases, substantiated findings of serious misconduct and institutional failures in residential treatment settings. I firmly believe that allegations of abuse in a treatment setting must not be dismissed on the basis of a person's underlying distress, trauma, or psychiatric history. Allegations should be investigated carefully and seriously. That is particularly important in settings where the people making the allegations may have had very little power to begin with.
What IFS Actually Asks a Therapist to Do
IFS uses the language of “parts” to describe different patterns of feeling, thinking, reacting, remembering, and protecting ourselves. You might recognize a part of yourself that becomes highly critical when you make a mistake, a part that wants to withdraw when someone gets angry with you, or a part that goes quiet and foggy when you are overwhelmed.
The model understands these parts as having protective intentions, even when their strategies create problems. An IFS therapist is therefore supposed to approach them with curiosity rather than trying to overpower, shame, or eliminate them.
This is one of the most important things to understand about IFS if you have only encountered it through social media or through descriptions of the Castlewood controversy. The therapist is not supposed to decide what your parts are, tell you what they mean, or force them to reveal something. The work is supposed to involve helping you become curious about your own internal experience.
For example, you might say, “Every time my partner criticizes me, I get furious.” Rather than immediately deciding that the anger is evidence of some particular trauma, an IFS therapist might slow the moment down with you. What happens in your body? What does the angry response seem to be trying to accomplish? Is there another reaction underneath it? What happens when you pay attention to the part that wants to fight?
The answers belong to you. A therapist can ask questions without knowing the answer in advance. There is an important difference between helping someone explore an experience and leading them toward a conclusion about what happened to them.
Protective Parts Are Not Obstacles to Get Around
You may have protective parts that avoid certain subjects, become skeptical of therapy, intellectualize when emotions become intense, or simply say, “I do not want to do this.”
In IFS, that response is supposed to be respected. The therapist does not have to convince the protective part that it is wrong. The part’s reluctance is information about what the person needs in order to feel safe enough to continue.
This is also where I think the model can be misunderstood from the outside. If someone says, “I do not want to talk about that,” an IFS therapist should not automatically interpret the resistance as evidence that the person has a hidden trauma that needs to be uncovered. A protective response can be doing exactly what it believes is necessary.
If a client has a part that says, “I do not trust this therapist,” the therapist needs to be able to tolerate that possibility. The therapist should not simply explain the distrust as another part that needs to be worked through. That would create a closed system in which disagreement becomes evidence for the therapist’s theory.
The language of IFS gives therapists a particularly ready-made vocabulary for interpreting internal experience, but it is not unique in having concepts that can be overextended in this way. A psychodynamic therapist, for example, could interpret a client’s disagreement as resistance or a defense mechanism; a Cognitive Behavioral Therapy (CBT) practitioner could interpret disagreement with a formulation as evidence of a cognitive distortion; an attachment-oriented therapist could understand it as an attachment pattern.
A client still has to be allowed to say no.
What About False Memories?
One of the clinical questions raised by the Castlewood reporting concerns false memories.
A memory, image, bodily sensation, dream, or intense emotional experience that emerges during therapy is not automatically proof that a particular event occurred. This is true regardless of whether the therapist uses IFS, another trauma model, or no named model at all.
IFS does not require a therapist to treat an emerging memory as historical fact. The IFS Institute's response to The Cut specifically says that previously unknown or “recovered” memories should not be accepted as fact until corroborated by evidence. Richard Schwartz is also quoted in The Cut as acknowledging that there is a risk of false memories in trauma therapy in general, and saying that memories should not be assumed accurate without corroboration.
To be clear, a lack of corroboration doesn't make a memory false. It means the question of what happened stays open, and the therapeutic work can go on without settling it. Accuracy matters most outside the therapy room, when someone is weighing a confrontation, cutting off contact, an accusation, or legal action. Those decisions shouldn't rest on an uncorroborated memory alone.
Inside the session, the work looks different:
Suppose a client begins working with a part and suddenly has an image of being hurt as a child. There are several things a therapist can do with that experience without turning it into an investigation. They can help the client notice what the image brings up emotionally. Together, they can explore what the younger part seems to need now, pay attention to the beliefs and fears associated with the experience, and leave the factual question open.
A therapist can say, in effect, “This experience is important. We can explore what it means to you without deciding that we know exactly what happened.”
If what happened at Castlewood occurred as the former patients described it—including therapists treating uncorroborated abuse narratives as established facts or steering clients toward particular interpretations of their experiences—that was not following the IFS model as it is supposed to be practiced. It would also raise serious ethical concerns independent of the therapy model.
The Criticism of IFS Goes Beyond Castlewood
The Castlewood reporting is not the only reason people have questions about IFS. Researchers and clinicians have also pointed to a genuine limitation: the research base is still relatively small compared with the popularity of the model.
The authors of a 2024 article in Psychotherapy Bulletin argued that IFS had expanded beyond its evidence base. They noted that the published literature at that point included only a small number of studies examining specific concerns and raised particular questions about using IFS with people experiencing psychosis or significant reality-testing difficulties. They also expressed concern that encouraging people to separate their experience into parts could be disorganizing for some clients.
Those concerns are not established findings that IFS is harmful for those populations. The authors themselves describe the need for further research. Their argument is better understood as a caution about applying a rapidly popularizing model more broadly than the available evidence can support.
That is a reasonable question to ask of any psychotherapy approach.
The research has also continued to develop. In January 2026, the results of a randomized controlled post-traumatic stress disorder (PTSD) program based on the IFS framework were published in Psychological Trauma. The study included 60 people with PTSD, randomized to either the IFS-based program or a comparison program. Both groups had significant reductions in PTSD symptoms, and the study did not find a significant difference between the two groups in symptom change. Participants in the IFS-based group attended more group sessions and reported greater group satisfaction. The authors concluded that the program was feasible and acceptable and called for further research.
That is meaningful evidence. It is also a small study of one particular IFS-based program. It does not establish that IFS is effective for every condition, or that it is preferable to other established treatments for PTSD. The appropriate conclusion is narrower: there is emerging research supporting further study of IFS-based treatment, while substantial questions remain about where and for whom it works best.
Popularity and Evidence Are Different Questions
Part of the current controversy comes from the extraordinary popularity of IFS. The 2024 Psychotherapy Bulletin article documented tens of thousands of mentions of IFS among therapists and millions of social-media posts using IFS-related terminology. The Cut likewise describes a large and rapidly expanding public interest in the model.
Popularity can create a strange situation for a therapy modality. Clients may encounter a treatment on TikTok, Instagram, a podcast, or a book long before they encounter it in a clinical setting. By the time they meet a therapist, they may already have a detailed vocabulary for their inner life.
That vocabulary can be useful, but can become restrictive if a client feels that they are supposed to experience themselves in a particular way in order for the therapy to work.
You do not have to be good at visualizing your parts. You do not have to give them names. You do not have to experience them as separate personalities. You do not have to agree with every theoretical claim associated with IFS.
The therapist should be able to work with the person who is actually in front of them.
What IFS Can Look Like When It Is Used Carefully
In a typical session, IFS does not necessarily involve working directly with traumatic memories.
A client might describe getting disproportionately angry when a friend takes too long to respond to a text. We might slow down and notice the different reactions that appear. There may be a part that feels rejected, another that becomes angry because anger feels safer than hurt, and another that immediately criticizes the person for caring so much.
Nothing has to be uncovered.
The point is to become curious about the pattern and develop a little more freedom in relation to it. A client may discover that the angry response is trying to prevent a familiar feeling of being unimportant. They may recognize that the self-critical response is trying to prevent rejection by finding fault first. They may simply notice that several reactions are happening at once.
Sometimes deeper material does emerge. When it does, the therapist has to keep the same basic discipline: follow the client’s experience rather than supplying the answer. A part may have a memory. It may have an image. It may have a belief about what happened. Those experiences can be explored while the factual status of the memory remains open.
The therapist does not need to know what a client’s inner experience “really” means before the client does.
Questions I Would Ask an IFS Therapist
If you are interested in IFS after reading the recent criticism, you can learn a great deal by asking a prospective therapist a few direct questions.
Ask how they handle uncertainty.
What do they do if a client has an image or memory that feels real but cannot be independently verified?
Ask how they respond to disagreement.
What happens if you do not experience something as a “part,” or if you think the therapist’s interpretation is wrong?
Ask about training.
How were they trained in IFS, and what additional training do they have in trauma treatment and the particular problems they treat?
Ask what happens when a protective part says no.
A therapist who understands the model should have an answer that leaves room for the client’s autonomy.
Ask whether IFS is the whole treatment.
A thoughtful therapist should be able to explain how IFS fits into their broader clinical judgment, including assessment, stabilization, safety planning, and other interventions when those are appropriate.
What I Think the Castlewood Story Clarifies About IFS
The Castlewood reporting raises legitimate questions about therapist power, suggestibility, trauma treatment, and the consequences that can follow when a therapist becomes too certain about what a client’s experiences mean. Those questions should remain part of the conversation about IFS.
They also make it especially important to distinguish the model from the way a particular therapist may use it.
IFS is built around curiosity toward internal experience, respect for protective parts, and a stance in which the therapist is not supposed to force the client’s system to do something it is not ready to do. The model does not require treating newly emerging memories as historical facts. It does not give the therapist permission to impose a story on the client.
If what happened at Castlewood occurred as described by the former patients and other sources in The Cut’s reporting, it was not an example of IFS being practiced according to those principles. It was a serious failure of clinical boundaries and judgment, regardless of the label attached to the treatment.
That does not settle every question about IFS. The evidence base is still developing. Researchers have raised reasonable concerns about overapplication, training, and the use of the model with some populations. Those questions deserve actual research rather than either blanket dismissal or blanket endorsement.
Powell is a therapist who offers individual therapy for adults in Georgia, California, Colorado and Florida.
He offers individual therapy, but also specializes in Gender Identity, Life Transitions, and Self-Worth.
