The Differential the Medicine Amplifies

This is a two-part post on consent and power in psychedelic therapy. I treat the consent mechanics as their own subject in a companion piece on touch and consent in psychedelic-assisted therapy.

A supervisee told me, with complete sincerity, that the abuse cases in the field didn't worry him for his own practice. "I'd never cross a line like that," he said. I believed him. He's a careful, decent clinician. And I had to tell him that in this work, sincerity isn't a control. The reason has nothing to do with his character and everything to do with what the medicine does to the space between a therapist and a client.

There is a question I find more useful than "would I ever?": if I unintentionally steered a client in a heightened state, what in my setup would catch it? For most of us working alone, the honest answer is nothing. That gap is the real subject when we talk about power dynamics in psychedelic therapy, and it is structural, not a matter of who has good values.

Ordinary therapy already carries a power differential. Psychedelics widen it. A person under MDMA, psilocybin, or ketamine is in a state of heightened suggestibility, with altered judgment and an intensified tendency to idealize the person guiding them.

The suggestibility piece is not hand-waving. In a 2025 review in the International Journal of Clinical and Experimental Hypnosis, Stein and Terhune document how strongly suggestion shapes the psychedelic state, to the point that post-hypnotic verbal suggestion alone can reproduce the phenomenological effects of MDMA in people with prior experience of the drug. Their concern is partly methodological, that suggestion is a confound in trials, but the clinical implication is unavoidable: if a suggestion can manufacture a drug-like effect, then a clinician's words, framing, and expectations land with unusual force on a client who is pharmacologically primed to absorb them.

That force is the problem. As the bioethics literature on consent in this work has noted, when the therapist alone decides what happens and when a session ends, the differential widens further, and a client in a heightened, suggestible state is poorly positioned to advocate for themselves in real time (Lee, Rosenbaum, & Buchman, 2024, Canadian Journal of Psychiatry). The client cannot reliably tell, in the moment, the difference between their own insight and a suggestion they've absorbed. Neither, often, can the therapist.

Here is the structural observation, and I mean it as observation, not accusation: a safeguard that depends entirely on the clinician's good character is not a safeguard. It is a hope. Good people misread a room. Well-intentioned clinicians offer a suggestion that a suggestible client receives as instruction. Countertransference operates below awareness by definition. None of that requires a predator; it only requires a human being working without a backstop.

This is why I push back on "I'd never." The clinicians who cause harm in this field are not all cynical. Some are sincere people who trusted their own intentions as a substitute for structure, and the heightened state did the rest. Treating your own virtue as the control is itself a small failure of judgment about how this state works.

The structures we reach for are worth building, and I want to be careful not to oversell them. Each can make harm less likely. None makes it impossible.

• Supervision and consultation for this specific work, where decisions get examined out loud rather than rationalized privately. A case you have to describe to a group is a case you think about differently.
• A two-clinician model where feasible, so the session is witnessed rather than a closed dyad.
• Session recording, with consent, so that "what happened" is not solely the account of the person holding more power afterward.
• Explicit scope limits, written down, including a clear consent and touch practice. I treat the consent mechanics as their own subject in a companion piece on touch and consent in psychedelic-assisted therapy; the point here is that those limits have to exist on paper, not just in your judgment.
• 
  

Here is why I won't call that list a solution. In a MAPS-sponsored MDMA trial in Vancouver, a two-therapist team was in the room and the sessions were recorded on video, and serious boundary violations still occurred. According to CBC's reporting, the recordings existed precisely to confirm safety and protocol adherence, and no one watched them until roughly six years after they were filmed; MAPS later found the therapists had substantially deviated from its own treatment manual and cut ties with them. The cameras were on. The second clinician was present. The structure was there, and it did not catch the harm.

That is the part we say less often. Recording only protects a client if someone watches and acts on what they see. A co-therapy pair only helps if the second person can name what's happening and isn't held by the same deference. Supervision only works if it's honest enough to surface the case nobody wants to raise. The safeguards are necessary, and they are inert without a culture willing to use them: openness about harm, the kind Devenot argues for when she says advocates have to surface harms themselves rather than wait for prohibitionists to weaponize them. Less true-believer certainty about the medicine, more willingness to look at the tape. Structure and culture are not substitutes for each other, and the Vancouver case is what it looks like when you have the first without the second.

Now the part that should bother us. Every safeguard in that list costs something. A second clinician doubles your labor cost. Ongoing supervision is a monthly bill. Recording infrastructure and the time to use it are not free. Which means the safest configurations are most available to the best-funded settings, and the solo practitioner serving an underserved community is the one most likely to be working in exactly the closed dyad the state makes risky.

That is health capitalism distributing safety by ability to pay, and it is not a personal failing of the clinician caught in it. It is the same structure that puts a $10,000 training between competent people and this work. If safeguards are real, they should not be luxuries, and a field serious about ethics has to build shared, low-cost ways for solo and community practitioners to get supervision, consultation, and peer witness. Naming the suggestibility risk without naming who can afford to mitigate it would be only half the analysis.

So I'd retire "I'd never cross that line" as a reassurance. It answers the wrong question. The right one is what holds when your intentions aren't enough, because in a state this suggestible, sooner or later they won't be. Build the structure, build the culture honest enough to actually use it, and build the field where everyone can afford both.

Peter Addy, PhD, LPC, LMHC is a Portland-based licensed therapist and the founder of Psychedelic Affirming Education, an NBCC-approved continuing education provider for licensed mental health professionals and Oregon Psilocybin Services facilitators. His research background includes work at Yale School of Medicine on psychedelic substances.

The safeguards in this post are exactly what a licensing board expects you to have thought through. Our CE course works the ethics and documentation in detail.