Free Association in Psychoanalytic Training: Beyond the Institute Model
There is a settled faith that shadows most professional training, even when no one says it outright: if you accumulate the right knowledge, and if the institution that guards the knowledge certifies you, then you become the kind of person who can practice it. Study hard, learn the method, pass the examinations, demonstrate competence, and you will have earned your authority.
In Chapter 12 of Essays in Existential Psychoanalysis, Michael Guy Thompson does not merely dispute this faith, he exposes the kind of human relationship it quietly presupposes, and he asks whether that relationship can possibly yield an analyst. His provocation is not that scholarship is worthless, nor that institutes should be burned down, but that psychoanalytic education cannot be reduced to academic education without losing the very sensibility it claims to transmit. When free association is treated as the founding condition of analysis, Thompson suggests, it becomes difficult to regard it as a technique alone; it begins to look like an ethic of formation, a model of how one learns to listen without coercion (Thompson, 2024, pp. 243–246).
This is not only a question for clinicians, even if it begins there. Patients rarely ask where an analyst trained, but they do encounter, immediately and unmistakably, what the training has made possible in the analyst’s presence: whether the room is a place where speech can unfold, or a place where speech is quietly pressured to conform.
Free Association in Psychoanalytic Training: Why the Question Matters
Technique or formation: what is actually being taught?
The title of Thompson’s chapter presents itself as a technical query, almost bureaucratic in tone, as though the question were simply where to file free association in the educational syllabus. Yet the force of the chapter comes from the fact that Thompson does not accept the premise that analysis is first a body of content to be mastered and only later a relational practice to be performed. Psychoanalysis, in his telling, concerns the conditions under which one human being can meet another without turning the other into an object to be managed. If that is what psychoanalysis is for, then education cannot be a matter of training candidates to reproduce an approved discourse; it has to be a formation of character and sensibility, the slow acquisition of an ability to bear what is uncertain, unflattering, and ethically demanding (Thompson, 2024, pp. 243–244).
So the question “What is being taught?” becomes sharper than it first appears. Are institutes teaching a procedure, a method that can be applied from the outside, or are they cultivating an inward capacity, a kind of disciplined openness, without which method becomes a form of evasion? Thompson’s wager is that free association, properly understood, does not sit comfortably inside the academic model precisely because it is not a content-area. It is a way of being with experience, and it is therefore a way of being with another person.

Thompson’s Critique of the Academic Model
Why book-knowledge cannot substitute for lived authority?
Thompson’s opening declaration is deliberately unsoftened: “The academic model of education is ill-suited to train and educate people to become psychoanalysts” (Thompson, 2024, p. 243). The sentence is blunt enough to provoke defensiveness, and Thompson anticipates that response by insisting he is not speaking as someone unfamiliar with academia. His point is not that universities should have no place in psychoanalytic education, but that academic formation, by its structure, tends to treat knowledge as something possessed, verified, and then licensed.
The most striking feature of his critique is how quickly he locates the psychoanalytic task in the moral texture of ordinary life. He writes that psychoanalysts are “concerned with the way human beings treat each other,” and that they help others come into their own by treating them with “respect, compassion, and honesty” (Thompson, 2024, p. 243). This is not a sentimental aside, it is the ground of his argument, because once psychoanalysis is framed as a matter of how persons treat persons, the fantasy that authority can be granted by institutional proof begins to look naïve in the pejorative sense, a wish for certainty where uncertainty is constitutive.
Academic training, at least in its prevailing form, is built upon a tautology: you learn by reading what others claim to know, and you prove learning by repeating, with increasing sophistication, what has already been said, until the institution determines you may finally practice. Thompson names this directly, noting how psychology programs presume that students can learn to be therapists by studying books and then being evaluated on how well the material has been absorbed before they are permitted to treat (Thompson, 2024, p. 243). The problem, for him, is not the reading. The problem is that the reading is too easily mistaken for the capacity to stand in the analytic situation without substituting knowledge for contact.
Training as confrontation with one’s own suffering
Thompson deepens the critique by moving from epistemology to experience. If analysis is not primarily the application of knowledge, then what grants authority? He gives an answer that is both simple and difficult to tolerate, because it does not flatter any institutional procedure: “We learn about human misery from our own suffering, and we learn to relieve it by coming to terms with the suffering that we have experienced and continue to experience every day of our lives” (Thompson, 2024, p. 244).
This sentence is the hinge of the chapter. It shifts training from an external sequence of requirements to an inward confrontation, and it does so without romanticizing suffering, because the point is not that pain automatically ennobles, but that unworked suffering easily becomes coercion. An analyst who has not begun to come to terms with his or her own misery will be tempted, in the consulting room, to manage the patient’s misery as a way of managing his own, which is to say, to impose an agenda under the guise of care.
Thompson therefore defines psychoanalytic training as the practical task of getting in touch with the roots of one’s suffering and devoting oneself to alleviating it from a psychoanalytic perspective, so that one can eventually accompany another without pretending to stand above the human condition one is addressing (Thompson, 2024, p. 244). In this light, “authority” is not granted by the institution, it is earned in the slow work of being educated by experience, which includes, unavoidably, the experience of one’s own limitation.
Free Association as a Model for Psychoanalytic Education
Self-disclosure, openness, and the refusal of an agenda
Thompson is attentive to the way free association becomes a cliché precisely because it is so familiar. One of the benefits, he says, of attempting to define it is discovering that it resists final definition; it is recognizable, yet elusive, and any concise account tends to falsify it (Thompson, 2024, p. 245). This matters because what cannot be finalized cannot be owned, and what cannot be owned cannot be administered with the same confidence as a curriculum.
In the chapter, Thompson treats free association as inseparable from self-disclosure and openness, not in the confessional sense of saying everything for its own sake, but in the existential sense of speaking from where one actually is, rather than from where one believes one should be. Its most distinctive feature, as he emphasizes, is the absence of an externally imposed agenda: there is no plan the patient is expected to follow, no sanctioned trajectory that would reassure both patient and analyst that the “right” material is being produced (Thompson, 2024, p. 245).
If this is what free association is, then one can see why Thompson is drawn to it as an educational model. An education modeled on free association would not be organized around predetermined outcomes, because predetermined outcomes are precisely the temptation that free association resists. Instead, education would have to cultivate a capacity to remain with what is unexpected and unfinished, and to allow understanding to arise as something earned in relation, not delivered as doctrine.
The cultivation of naiveté
At the center of Thompson’s argument is a line that should not be rushed past, partly because it is easy to sentimentalize and partly because it threatens every credentialist fantasy: “The key to analytic education isn’t the acquisition of knowledge but the cultivation of naïveté” (Thompson, 2024, p. 246).
Naiveté here does not mean ignorance. Thompson defines it as an “open state of mind,” an attitude that is antithetical to skepticism in its cynical form, but aligned with the phenomenological demand that one meet what appears without immediately explaining it away (Thompson, 2024, p. 246). The analyst’s knowledge, however extensive, can become a defense, a way of neutralizing the patient’s otherness by translating it too quickly into categories. In that sense, knowledge can function as a way of not listening, because it tempts the analyst to hear only what fits.
Thompson’s claim is that analytic candidates must be educated into a disciplined openness that is capable of letting experience teach, including the experience that contradicts one’s favorite theories. In my reading, this is not a rejection of theory but an insistence that theory remain answerable to the lived encounter, which is always more singular than the conceptual net we throw over it. A training that loses this capacity may produce therapists who can speak fluently about analysis while quietly fearing the very thing analysis requires: not knowing.
When Institutes Become a “Tight Ship”
Compulsion, obsessionality, and the longing for certainty
Thompson’s critique sharpens when he turns from education in principle to institutes in practice. In his account, many institutes embody a structure that is opposed to the spirit of free association, because the institutional imperative is to control, standardize, and legitimate. He describes the “tight ship” atmosphere, the “air of military precision,” the rigidly defined roles and rules, and he notes the irony that a discipline devoted to ambiguity should so readily construct training environments that defend against ambiguity (Thompson, 2024, p. 247).
He reads this rigidity not merely as a sociological quirk but as a psychological and ethical problem, because it tends to recruit and reward an obsessional sensibility, and that sensibility can be mistaken for seriousness. Thompson observes that analytic candidates are often driven, dedicated, and studious, and that these virtues can slide into humorlessness, ambition, and a defensive posture of control when the institutional environment invites it (Thompson, 2024, p. 247). One can pass such training by becoming increasingly adept at appearing certain, and one can become increasingly frightened of what cannot be made certain.
It is here that Thompson invokes Hans Loewald, who warns that what we call reality can itself become defensive. Loewald writes that reality can take on the quality of “a hostile-defensive integration, akin to the obsessional mechanism” (Loewald, 1980, p. 30, as quoted in Thompson, 2024, p. 247). The implication is unsettling: institutions that claim to teach analysis may end up reproducing, at the collective level, the very psychic defenses that analysis exists to illuminate at the individual level.
Thompson’s point is not that all institutes are uniformly corrupt, nor that discipline and standards have no place. It is that the longing for certainty, when it becomes the organizing principle of education, is not neutral; it shapes the analyst’s temperament, and it shapes the room the analyst later builds with patients. A training that rewards control will not easily cultivate the capacity to bear the patient’s uncontrolled speech.
A Salon Instead of an Institute

Equality, conversation, and psychoanalysis as philosophy
Thompson does not end with critique, and his alternative is not a utopian fantasy, but an experiment grounded in a particular historical moment. In 1988, invited by students who wanted training but were dissatisfied with conventional institutes, he and colleagues formed a psychoanalytic salon in San Francisco. They called it Free Association, a name he treats as a double entendre: it aimed to help students learn the free association method, and it understood itself as an association of equals devoted to the free dissemination of ideas (Thompson, 2024, p. 248).
What matters here is the ethos. The group included psychoanalysts, philosophers, historians, and others with backgrounds in phenomenology or psychoanalysis, and Thompson describes them as viewing psychoanalysis as philosophical in the Socratic sense, meaning that psychoanalysis belongs to a tradition in which ethics is conceived as a therapy for suffering, not merely a set of professional rules (Thompson, 2024, p. 248). This is a profound reframing: psychoanalysis is not simply a clinical technology housed in an institution, it is a practice of inquiry into the human condition, and education should resemble inquiry rather than indoctrination.
No curriculum, and the courage to meet the unexpected
The salon model becomes concrete in Thompson’s refusal of curriculum. He states, with an almost mischievous clarity, that their curriculum did not resemble conventional curricula because they did not have one (Thompson, 2024, p. 248). The point was not to be anti-structure for its own sake, but to avoid the subtle coercion of predetermined outcomes, the way a fixed sequence of requirements can quietly teach candidates that the goal of learning is compliance.
Instead, Thompson describes an atmosphere meant to invite students to grapple with what is unexpected, unexplained, and ambiguous, by approximating, as much as possible, the experience of a psychoanalytic session (Thompson, 2024, pp. 248–249). This is where free association becomes educational: education becomes a milieu in which one is asked to tolerate not knowing, and to let thought arise from encounter rather than from the safety of advance conclusions.
Mentorship, Apprenticeship, and the Here-and-Now of Seminars
Provocation over indoctrination
Thompson’s educational model leans toward apprenticeship, not because apprenticeship is simpler, but because it keeps education tethered to the lived authority of persons rather than the impersonal authority of an institution. He describes seminars organized around the instructor’s predilections and current professional concerns, and he portrays the value of such seminars as lying in their capacity to provoke, to generate controversy, and to inspire independent study, rather than to transmit a sanctioned orthodoxy (Thompson, 2024, p. 250).
There is a quiet ethical claim here. Indoctrination offers the comfort of belonging and the relief of borrowed certainty, but it risks producing analysts who cannot think, or worse, who can only think within the approved dialect of their school. Provocation, by contrast, can feel destabilizing, even threatening, but it is closer to what analysis requires, because analysis is not the repetition of doctrine, it is the encounter with what resists doctrine.
History as a living dimension
One of the most intellectually serious dimensions of Thompson’s salon model is his insistence that psychoanalysis be situated in an historical context broad enough to include its prehistory. He describes seminars devoted to phenomenological method, intentionality, intersubjectivity, and self-disclosure, and he traces free association to antecedents in meditation, Christian mysticism, and Montaigne, while linking neutrality to ancient skeptical traditions and abstinence to German Romanticism (Thompson, 2024, p. 250).
This is not antiquarianism. It is an attempt to keep psychoanalysis from collapsing into the cult of the new guru or the latest institutional fashion. History, for Thompson, is not a museum of dead ideas; it is a living dimension that keeps education from becoming captive to a narrow present. When the analyst learns to hear psychoanalysis as part of a longer conversation about suffering and truth, the analyst may become less tempted to treat technique as a proprietary invention that must be defended, and more able to treat it as an ongoing ethical task.
Standards, Authority, and the Ambiguity of “Completion”
Self-selection and the limits of evaluation
The predictable objection to Thompson’s model is the one he himself anticipates: without curriculum, without standardized evaluation, what becomes of standards? Thompson’s reply is not to deny the need for seriousness, but to refuse the fantasy that seriousness can be guaranteed by procedure. He describes the Free Association model as self-selecting in a strong sense: anyone may join seminars, students choose supervisors, and even the practical terms of work, such as whether the patient sits or lies down and how frequently sessions occur, are left to the discretion of the analyst and patient rather than being dictated by institutional templates (Thompson, 2024, p. 250).
He acknowledges that this absence of official criteria can seem alarming, and he turns the alarm back upon the objector by naming what everyone in training already knows but rarely says: standards are often arbitrary, the evaluation process is never neutral, and the question of “completion” is inherently ambiguous, in much the same way that termination in analysis is ambiguous (Thompson, 2024, p. 250). In other words, the demand for certainty in training repeats the demand for certainty in life, and analysis exists partly to expose the impossibility of satisfying that demand without distortion.
The deeper question is what kind of authority we are seeking. If authority means institutional sanction, then Thompson’s model will always look suspect. If authority means the capacity to remain ethically present in the analytic situation, to tolerate uncertainty without collapsing into control, then standards may need to be reconceived as matters of character and judgment that cannot be fully captured by checklists.
Creating Psychoanalysis Anew
Institutions, succession, and why renewal matters
Toward the end of the chapter, Thompson places his argument inside a recurring historical pattern. Freud, Sullivan, and Laing founded their schools by gathering students around themselves, and after their deaths institutes were created in their names, which then grew into institutions (Thompson, 2024, p. 254). Thompson does not treat this as a scandal so much as an almost inevitable drift, the way living traditions harden when they are preserved rather than enacted.
His most radical claim follows from that diagnosis: it may be that the only way to preserve the psychoanalytic instrument is to periodically create it anew, forming new institutes and dismantling old ones, so that successive generations can discover its “uncanny uniqueness” for themselves rather than inheriting it as doctrine (Thompson, 2024, p. 254). Free Association, in his telling, tried to enact this principle by refusing succession, hierarchy, and the competitive promise of spoils, and by aiming to train a group and then disband when the work of that moment was complete (Thompson, 2024, p. 254).
Thompson’s closing caution is sober and recognizably true to clinical life. Psychoanalysis is isolating, alienation is not merely an intellectual theme but an occupational reality, and loneliness can tempt the analyst to compromise what he most values, not out of malice but out of an all-too-human wish to belong (Thompson, 2024, p. 254). Education, then, is not only a matter of learning the instrument. It is also a matter of remaining vigilant about the ways one’s own needs can infiltrate one’s professional life and bend it toward corruption.

Conclusion
What this approach protects in the consulting room
Thompson’s Chapter 12 is, on its surface, about psychoanalytic training, but its true subject is the ethical architecture of the consulting room. If free association is treated as a technical rule, it becomes procedural, and procedure always risks becoming a defense. If free association is treated as an educational model, it becomes formative, shaping the analyst’s capacity to remain open, to resist coercion, and to bear ambiguity without prematurely converting it into certainty (Thompson, 2024, pp. 245–250).
For patients, what is at stake is not a debate about institutes, but the quality of presence they encounter when they speak. A room governed by the analyst’s need to know, to classify, to control, can feel subtly violent even when it is polite. A room governed by an educated naiveté can feel, paradoxically, more serious, because it does not flee from what is difficult by hiding behind explanations.
Connecting Thompson’s chapter to FAC’s clinical orientation
At Free Association Clinic, our work in psychoanalytic therapy and existential therapy is grounded in this deeper ethic of listening, where technique is always answerable to encounter. For clinicians drawn to this orientation, you can learn more about our training program, or contact us to begin a conversation.
James Norwood, PsyD
Associate Director, New School for Existential Psychoanalysis (https://www.freeassociation.us)
Clinical Director, The Free Association Clinic for Existential Psychotherapy and Psychoanalysis (https://www.freeassociationclinic.com)
Founder/CEO, inpersontherapy.com (https://inpersontherapy.com)
References
Kirsner, D. (2000). Unfree Associations: Inside Psychoanalytic Institutes. London: Process Press.
Loewald, H. (1980). Papers on Psycho-Analysis. New Haven and London: Yale University Press.
Thompson, M. G. (1994). The Truth About Freud’s Technique: The Encounter With the Real. New York: New York University Press.
Thompson, M. G. (2024). Essays in Existential Psychoanalysis: On the Primacy of Authenticity. Abingdon and New York: Routledge.
Thompson, M. G., &Thompson, S. (1998). Interview with Dr. Otto Allen Will, Jr. Contemporary Psychoanalysis, 34(2).









